*j>. V

\ V

.^

-> .J "

fV^

>

I > :*&

./::.'

-*■ r-

Issued February 10, 1913.

U. S. DEPARTMENT OF AGRICULTURE.

BUREAU OF PLANT INDUSTRY— Circular No. 105.
B. T. GALLOWAY, Chief of Bureau.

EXPERIMENTS IN THE CONTROL
OF GRAPE ANTHRACNOSE.

BY

LON A. HAWKINS,

Scientific Assistant, Fruit-Disease Investigations.

05010° 13 WASHINGTON : GOVERNMENT PRINTING OFFICE : 1918

• rf>3

BUREAU OF PLANT INDUSTRY.

Chief of Bureau, Beverly T. Galloway.
Assistant Chief of Bureau, William A. Taylor.
Editor, J. E. Rockwell.
Chief Clerk, James E. Jones.
[Cir. 105]

2

B. P. I— 790.

EXPERIMENTS IN THE CONTROL OF GRAPE

ANTHRACNOSE.

INTRODUCTION.

Anthracnose of the grape, or bird's-eye rot, caused by the fungus
Sphaceloma am/pelinum De Bary, is a disease of considerable impor-
tance on certain varieties of grapes grown in the eastern United States.
While the disease does not seem to attack the Concord, which is the
most extensively grown of American varieties of grapes, Vergennes,
Salem, Diamond (Moore's Diamond), Norton (Norton's Virginia),
Missouri Riesling, Diogenes, Champion, and other varieties of both
wine and table grapes are very susceptible to anthracnose. In
serious outbreaks of the disease the fruit in a vineyard of one of
these susceptible varieties may be almost entirely destroyed and the
vines seriously damaged. Certain exceptionally fine varieties of
wine and table grapes can not be grown successfully in some localities
because of unusual susceptibility to it.

Anthracnose affects practically all the green parts of the plant. On
the green shoots it causes cankers (PL I, fig. 3), which first appear as
small, light-brown, slightly sunken spots. These spots enlarge, sink
deeper, and become darker in color as the disease progresses. If the
canker is near the base of a shoot, the whole shoot may be killed. If
not killed directly by the disease, the shoot may be so weakened as to
be easily broken. The fungus causes cankers on the petioles and ribs
of the leaves (PI. I, fig. 2), on the stems of bunches of grapes, and on
the tendrils. On the grape berries it first appears as light-brown
spots, which rapidly become darker and are soon surrounded by a
circular bright-red zone, which, as seen against the green background
of the grape, presents the very striking appearance that gives the
disease the name of "bird's-eye rot." The berries in later stages of
the disease shrivel and dry up (PI. I, fig. 1).

The fungus is also reported to be the cause of a leaf-spot. The
writer has, however, never observed any case of leaf-spot on grape
leaves that could be attributed to it. The fungus is supposed to live
over the winter in the old wood and in mummied grapes. It repro-
duces itself by means of minute, oblong, colorless spores, which
infect the young, green parts of the plant. The disease is well known
in Europe and, according to reports, may be controlled by proper
treatment. Very little has been done in America, however, toward
working out the best method for preventing the disease, and the
recommendations for its treatment in this country are usually based

[Cir. 105] 3

EXPERIMENTS IN THE CONTROL OE GRAPE ANTHRACNOSE.

on its treatment abroad. As the climatic conditions of American
and European vineyards differ, as well as the varieties of grapes
usually grown in them, it was thought that it might be necessary to
modify the treatment of the disease as practiced in Europe, in order
to adapt it to American vineyards. The experiments described in
this circular were accordingly taken up by the writer, under the direc-
tion of Dr. C. L. Shear, Pathologist, of this Bureau, to see whether
the disease could be controlled in this country by the use of a solution
of iron sulphate and sulphuric acid in water, which is employed
extensively in Europe. It was also considered worth while to
experiment with applications of other fungicides to the dormant
vines and to try the effect of summer sprays in comparison with
the iron-sulphate and sulphuric-acid treatment mentioned.

EXPERIMENTATION.

The experiments which are here reported were conducted at
Lawton, Mich., in a vineyard of Champion grapes belonging to Mr.
John Robinson. In this vineyard in 1909 about two-thirds of the
fruit had been destroyed and the young wood seriously injured by
anthracnose. The vines selected for experimentation were a block
of about 600 where the disease had been the most severe. This block
was divided into nine plats, seven of which were to be treated and
two to remain untreated as controls. The order in which the treated
plats and controls w r ere arranged is shown in Table I.

Table I.-

-Treatment (riven plats of grapevines at Lawton, Mich., in 1910, 1911, and
1912 and the results obtained in 1911 and 1912.

No. of
plat.

Treatment in 1910.

Treatment in 1911.

Diseased

fruit in

1911.

Treatment in 1912.

Diseased

fruit in

1912.

1

(When dormant.) Iron-
sulphate mixture No. 1.

(When dormant.) Iron-
sulphate mixture No. 2.

Control, untreated

(When donnant.) Ten
per cent solution of sul-
phuric acid.

(When dormant.) Con-
centrated lime-sulphur
solution (1-10).

Same as in 1910

Per cent.

Same as in 1910

Per cent.

2

Same as in 1910. .

Same as in 1910

3
4

Control, untreated

4 per cent solution of sul-
phuric acid.

Same as in 1910

32.6

Control, untreated

Same as in 1911

10

5

Same as in 1910 .......

Control, untreated

Self-boiled lime-sulphur
solution (8-8-50); four
applications.

4-3-50 Bordeaux mix-
ture; four applications.

4-3-50 Bordeaux mix-
ture with two pounds
of rosin-fishoil soap to
50 gallons of mixture;
four applications.

12.2

Sprayed 5 times with
4-3-50 Bordeaux
mixture. Two
pounds of rosin-fish-
oil soap were added
in the last two appli-
cations.
do

7

8

..do :.

<i

..do

[Cir. 105]

Cir. 105, Bureau of Plant Industry, U. S. Dept. of Agriculture.

Plate I.

Fig 1.-A Bunch of Grapes Destroyed by Anthracnose.

Fig 2 -A Portion of a Leaf Showing Anthracnose Cankers on the Under

Side of Petioles and Ribs.
Fig. 3. -Portions of Grape Shoots Showing Various Stages in Anthracnose

Cankers.

Cir. 105, Bureau of Plant Industry, U. S. Dept. of Agriculture

Plate II.

Fig. 1.— A Portion of an Average Vine with Some Leaves Removed, Showing
the Condition of Fruit on the Treated Portion of the Vineyard in
1912.

Fig. 2.— A Portion of a Vine from the Control Plat in 1912 Seriously
Damaged by Anthracnose, About the Same Number of Bunches Being
Shown as in Figure 1.

EXPERIMENTS IN THE CONTROL OF GRAPE ANTHRACNOSE. 5

The following fungicides were tested in the experiments:

(1) Iron sulphate, 13J pounds; sulphuric acid (commercial), 7 ounces; hot water,
3} gallons.

(2) Iron sulphate, 6| pounds; sulphuric acid (commercial), 3^ ounces; hot water,
3| gallons.

(3) Sulphuric acid (commercial), 10 per cent solution.

(4) Sulphuric acid (commercial), 4 per cent solution.

(5) Commercial lime-sulphur solution (1-10).

(6) Self -boiled lime-sulphur solution (8-8-50).

(7) Bordeaux mixture (4-3-50).

(8) Bordeaux mixture (4-3-50) with 2 pounds of rosin-fishoil soap added.

The mixtures were prepared as follows:

For the iron-sulphate mixtures the required amount of crystallized
iron sulphate was placed hi a wooden tub and treated with the proper
quantity of sulphuric acid. The boiling water was then added and
the mixture stirred until the iron sulphate was all in solution. The
mixture was used immediately after preparation.

The sulphuric-acid solutions were made by adding the requisite
quantities of the acid to the water hi a wooden tub and stirring.

The concentrated lime-sulphur solution was prepared by mixing
1 gallon of commercial lime-sulphur solution, 32.8° Baume, with 9
gallons of water.

The self-boiled lime-sulphur mixture was prepared as recommended
by Scott. 1

The Bordeaux mixture and the Bordeaux mixture with soap added
were made in the usual way.

Mixtures Nos. 6, 7, and 8 were used only as summer sprays, while
the others were applied only to the dormant vines.

The vines in all the plats were pruned carefully each season during
which the experiment was in progress. All the canes on which the
disease could be readily recognized were cut away, and the primings,
together with the dead leaves and other debris, were removed from
the vineyard and burned.

Plats Nos. 1, 2, 4, and 5 were treated before the buds opened, as
shown in Table I. In applying the mixtures to these plats it was
considered inadvisable to use a sprayer, because of the effect of the
sulphuric acid in mixtures Nos. 1, 2, 3, and 4 on the brass and iron
parts of the spray pump. Several methods were tried in applying
the mixtures, the best results being obtained by using common
whitewash brushes provided with short handles. With these brushes
the vines of plats Nos. 1, 2, 4, and 5 were thoroughly washed with
the fungicide. The summer sprays were applied with a traction
sprayer.

i Scott, W. M. Lime-sulphur mixtures for the summer spraying of orchards. Circular 27, Bureau of
Plant Industry, U. S. Dept. of Agriculture. 1909.

[Cir. 105]

6 EXPERIMENTS IN THE CONTROL OF GRAPE ANTHRACNOSE.

In the spring of 1910 after the fungicide had been applied to plats
Nos. 1, 2, 4, and 5, a series of late frosts killed back the young growth
three different times. As this practically eliminated all possibility
of fruit for the season, the experiments were discontinued for the
year. The experiments were continued in 1911, as had been planned.
Plats Nos. 1, 2, 4, and 5 were treated while the vines were dormant,
as in 1910, except that a 4 per cent instead of a 10 per cent solution
of sulphuric acid was used on plat No. 4. The stronger solution
seemed to injure some of the buds. Plats Nos. 7, 8, and 9 were
thoroughly sprayed four times — once before the blossom, a second
application immediately after the blossom, and the other two appli-
cations at intervals of two weeks.

The treatment of the dormant vines on plats Nos. 1, 2, 4, and 5
was continued in 1912 as in 1911. Plats Nos. 6, 7, 8, and 9, how-
ever, all received the same treatment in 1912. They were sprayed
five times with 4-3-50 Bordeaux mixture— once before the blossom,
once immediately after the blossom, and the other applications at
intervals of about two weeks. Two pounds of rosin-fishoil soap
were added to each 50 gallons of Bordeaux mixture in the last two

applications.

RESULTS.

As there was no fruit in 1910 it was, of course, impossible to obtain ac-
curate data as to the effect of the various treatments. There seemed
to be less diseased wood in the plats which were treated than in the
others. In 1911, counts were made of the sound and diseased
berries on 1,000 average bunches of grapes from each control plat,
with the results shown in Table I. Much of the young wood in the
control plats was injured by the disease. A portion of a vine from
one of the control plats is shown in Plate II, figure 2, while a por-
tion of a vine showing the average condition in the treated plats is
reproduced in Plate II, figure 1 . There was practically no damage
from disease in plats Nos. 1, 2, 4, and 5, which were treated before
the buds swelled, while in the plats which received only the summer
spray there were a few diseased berries and some diseased wood.

The results in 1912 were much the same as in 1911. The amount
of diseased fruit on plat No. 3, the control, was not as great as in
1911, according to counts made of the sound and diseased berries
on 1,000 average bunches of grapes, owing probably to weather
conditions and the pruning out of the diseased wood the previous
seasons. Considerable diseased wood was present in the control,
as in the season of 1911, and the foliage was somewhat damaged.
There was practically no evidence of disease in plats Nos. 1, 2, 4,
and 5, which were treated while the vines were dormant. In the
plats receiving only the summer treatment with Bordeaux mixture

[Cir. 105]

EXPERIMENTS IN THE CONTROL OF GRAPE ANTHRACNOSE. (

occasional diseased berries and cankered canes could be found. In
plat No. 6, which had been treated only one year, the diseased wood
and fruit* were especially abundant.

Of the different mixtures used, the self-boiled lime-sulphur is
perhaps the least promising. It burned the foliage somewhat, and
there were more diseased fruit and wood in the plat sprayed with
this mixture than in that sprayed with Bordeaux mixture.

The fungicides used in treating the dormant vines all seemed equal
in efficiency. For commercial purposes, however, the ease with
winch the mixture may be prepared must be taken into considera-
tion. From this point of view the concentrated lime-sulphur seems
to be best, as it is easily prepared and can be applied with any
good sprayer. The iron-sulphate and sulphuric-acid mixture is
somewhat more difficult to prepare, and any of the mixtures con-
taining sulphuric acid must be applied by hand unless a sprayer
especially adapted to withstand the corrosive action of the sulphuric
acid is devised. It would seem, therefore, that the lime-sulphur
solution will be most economical for dormant application.

CONCLUSIONS.

In conclusion, it may be said that anthracnose of the grape was
controlled in these experiments bj pruning out the diseased wood
and applying any one of four different treatments to the dormant
vines. The fungicides used successfully on the dormant vines were
two concentrations of the iron-sulphate and sulphuric-acid mixture,
a 4 per cent solution of sulphuric acid and a concentrated lime-
sulphur solution (1-10). A careful pruning out of the diseased
wood and treatment with summer sprays of 4-3-50 Bordeaux mix-
ture reduced the damage from disease to practically nothing. In
contrast to the results from the treatment the loss from disease on
the control plats in 1911 was 32.6 per cent and 12.2 per cent. In
1912 the loss on the control plat was 10 per cent of the fruit. In
addition to the diseased fruit, the young shoots and foliage in the
controls were considerably injured, while the treated plats were
practically undamaged.

In combating anthracnose it would probably be advisable to com-
bine the treatment of the dormant vines with the use of the summer
spray, especially since black-rot or downy mildew of the grape is
usually present. The following treatment of anthracnose-infected
vineyards would then be recommended:

(1) Prune out and burn all diseased wood; (2) spray the dormant
vines thoroughly with a concentrated lime-sulphur solution in the
proportion of 1 gallon of 32° Baume test lime-sulphur solution to 9
gallons of water; (3) spray the vineyard thoroughly five times, as in

[Cir. 105]

8 EXPERIMENTS IN THE CONTROL OF GRAPE ANTHRACNOSE.

treating black-rot of the grape, with 4-3-50, 4-2-50, or 3-2-50
Bordeaux mixture, the applications to be as follows: (1) When the
shoots are from 8 to 12 inches in length; (2) just before the flower
buds open; (3) immediately after the blossoms fall; (4) 10 days to
2 weeks after the third application; and (5) about 10 days to 2 weeks
after the fourth application.

Trailers, consisting of lengths of hose with short extension rods
attached, so that the nozzles may be manipulated by hand, should
be used in the last two applications. The addition of 2 pounds of
rosin-fishoil soap to 50 gallons of spray mixture in the last two appli-
cations is advisable, as it greatly increases the adhesiveness of the
fungicide.

Approved :

James Wilson,

Secretary of Agriculture.

Washington, D. C, October 12, 1912.

[Cir. 105]

ADDITIONAL COPIES of this publication
-t\. may be procured from the Superintend-
ent of Documents, Government Printing
Office, Washington, D. C. , at 5 cents per copy

Issued March 10, 1913.

U. S. DEPARTMENT OF AGRICULTURE.

BUREAU OF PLANT INDUSTRY— Circular No. 106.

B. T. GALLOWAY, Chief of Bureau.

FORAGE-CROP EXPERIMENTS AT THE
SAN ANTONIO FIELD STATION.

BY

S. H. HASTINGS,

Superintendent, San Antonio Field Station.

69799° — ( ir. 106 — 13 1 Washington : government printing office : 1913

[Cir. 106]

2

Itl'REAV OF PLANT INDUSTRY.

Chie) oj Bureau, Beverly T. Galloway.
Assistant Chief of Bureau, William A. Taylor.
Editor, J. E. Rockwell.
( 1,„ i clt ik, James E. Jones.

ADDITIONAL COPIES of this publication
iA may be procured from the Superintend-
ent of Documents, Government Printing
Office, Washington, D. C. , at 5 cents per copy

B. P. I.— 796.

FORAGE-CROP EXPERIMENTS AT THE SAN
ANTONIO FIELD STATION.

INTRODUCTION. 1

The country about San Antonio, Tex., was an open prairie not
many years ago and was covered by a luxuriant growth of native
grasses which furnished excellent pasturage. The prairie fires which
periodically swept the country kept down the growth of trees and
shrubs. 2 As the country became more thickly settled the ranges
were overstocked, the native grasses were kept closely cropped, and
the mesquite (Prosopis juliflora), the native cactus (Opuntia sp.),
and other shrubby plants covered the ground. At present the
forage furnished by the native vegetation, except the cactus, plays
a decidedly insignificant part in the agriculture of the region.

Aside from some irrigation farming along the San Antonio River,
the early agriculture consisted of stock raising. Crop production on
a commercial scale is of comparatively recent date. As the stock
ranches have been divided into smaller farms the stockmen have given
place to cotton growers and truck farmers, until only a few large stock
ranches are left in the vicinity of San Antonio. Some dairy farming
has been undertaken, but the demand for dairy products resulting
from the rapidly increasing population has increased faster than tin*
supply, until the point has been reached where dairying would be
one of the most profitable industries of the region if the necessary
forage could be produced at a moderate cost. The breaking up of
the large stock ranches into cotton, truck, and dairy farms has
increased the need of forage production until it is now an acute
problem, for the success of dairying largely depends on the cheapness
with which forage can be grown or purchased.

The rainfall in the section is so unevenly distributed that native
grasses can not be relied upon to furnish pasture or hay; consequently

1 The information contained in this circular is based on cooperative experiments conducted by t he Offices
of Forage-Crop Investigations and Western Agricultural Extension of the Bureau of Plant Industry since
1907. The work during the first two years was carried on by Mr. F. B. Headley and since then by Mr. S. H.
Hastings.

2 For a discussion of this subject, see Cook, O. F., "Change of Vegetation on the South Texas Prairies,"
Circular 14, Bureau of Plant Industry, U. S. Dept. of Agriculture. 1908.

[Cir. 106] 3

4 FORAGE-CROP EXPERIMENTS AT SAN ANTONIO.

the supply of forage must depend upon the production of cultivated
crops. The increasing demand for roughage has brought the price
of forage crops to the point where their production for sale is nearly,
if not quite, as profitable as cotton growing. While it is to be expected
that cotton will continue to be the most important crop for the dry-
land farmer, yet certain forage crops might well be grown on nearly
every farm, either for sale or as the basis for live-stock production.

It should be understood, however, that while under present condi-
tions the production of forage may be more profitable than cotton
growing, yet if the area devoted to forage crops were greatly increased
prices would be depressed unless live-stock production or dairying
were also developed. Increasing local production of such a staple-
as cotton, on the other hand, would have little or no appreciable
effect on its market value.

The experiments with forage crops herein reported were conducted
on the San Antonio Experiment Farm during the years 1907 to
1912, inclusive. The objects in view were (1) to test forage crops
not generally known and select from these the ones best suited to
the region and (2) to test the best methods of planting and tillage.

The United States Department of Agriculture has introduced
from foreign countries during recent years numerous new or little-
known varieties of forage crops, many of which were deemed worthy
of trial in this region. These have been on trial in comparison with
local varieties. Some very promising new crops have been found,,
and experiments with methods of planting and tillage have shown
that there are better ways of growing forage crops than those now
generally followed by the farmers of the region.

CROP SEASONS.

In the San Antonio region the crop seasons are much less definite
than in most of the semiarid regions of the United States, because
of the mildness of the winter. By proper selection, forage crops may
be grown throughout the year, and green forage may be had at any
time except during a protracted drought. Oats, rye, and Can-
ada, iield peas are winter crops and may be planted at any time
from October to the latter part of February, with the expectation
of securing a crop. The summer crops, such as sorghum, millet,
and cowpeas, may be planted from the first of March to the first of
August, although very late planting is not recommended. Corn is
ordinarily planted soon after February 20.

An examination of the records of the United States Weather
Bureau at San Antonio for a period of 20 years shows that the last
killing frost of the spring occurs about February 24, but in 1900 it

[Cir. 106]

FORAGE-CROP EXPERIMENTS AT SAN ANTONIO. 5

was as late as April 12, and in several years it has occurred in the
second or third week of March. The first autumn frost may be ex-
pected about November 26, though it often occurs as early as Novem-
ber 12 to 14. A minimum temperature of 4° F. was recorded in the
winter of 1899, though the temperature does not usually fall below
18° F.

The weather records show that the region is one of short, mild
winters, with a long growing season, characterized, fortunately, by
a very early spring. During seasons of unusually heavy rainfall or
with irrigation it is possible to grow two different forage crops on the
same land in a single season. Where crops are grown without irriga-
tion the important point is to have the land in condition for planting
as soon as possible after the previous crop is removed, so that when
favorable rains come the next crop may be seeded promptly and the
moisture thus utilized before it is lost by evaporation.

RAINFALL.

The mean annual rainfall of the section for the past 20 years has
been about 25 inches. Much of this rain, especially during the sum-
mer months, comes in torrential storms and falls on soils that are
easily compacted, so that the proportion of run-off is large. There
are also many light showers, of one-half inch or less, that do not
moisten (he soil deeply enough to benefit the growing crops, par-
ticularly during the summer months. The potential evaporation,
as measured by the loss from a free water surface, is very high. It
has been shown by Briggs and Belz in their studies of dry farming in
relation to rainfall and evaporation in the Great Plains area ' that
the rainfall at San Antonio corresponds approximately to a 20-inch
rainfall in Minnesota or South Dakota, the average annual rainfall
at San Antonio being approximately 24| inches. These figures are
based on the differences in evaporation in the two localities. The
annual evaporation at San Antonio averaged 09.2 inches during
the four years 1908-1911.

The following table shows the rainfall records by months for each
year since 1907 and the average monthly rainfall for the past 20
years. It will be observed that while the average monthly rainfall
is fairly uniform, the actual rainfall for each month is extremely
variable and, furthermore, that there are no regular wet and dry
seasons.

1 Briggs, Lyman J., and Belz, J. '). Dry Farming in Relation lo Rainfall ami Evaporation. Bulletin
188, Bureau of Planl Industry, U. S. Dept. of Agriculture. 1910.
[Cir. km; |

6 FORAGE-CROP EXPERIMENTS AT SAN ANTONIO.

Table I. — Monthly rainfall at the San Antonio Experiment Farm for the years 1907-191 .',

i nc! iisi r<\ toe/ether with the averages of the years 1891-1911.

Month.

January

February

March . .'

April

May

Juiie

July

August

September

October

November

December

Total

Monthly n n

Monthly

average

rainfall,

1S91-1911.1

Inches.
1.35
1.47
1.41
2.84
3.01
2
2
2
2
1
1

i :>'i

24.47
2.04

Rainfall by months.

1907

Inches.
0.56
1.72
1.70
4.50
2.81

.31
2.65

. 39

.64
2. 98
6. 78

.64

25.68
2.14

1908

Inches.
0.47
2.61
1.27
2.39
5.44
.88
1.37
3. 79
2. 5(1
1.62
2.87
1.50

26.80
2.23

1909

Inches.

0.00

.41
1.12

.85
1.72

.72
2.86
1.75

.29
1.20

.27
1.95

13.14
1.09

1910

Inches.

I. lb

.68

.21

4.4H

1.77

2. 24

1.12

.45

Jin

3.7(1

L.30

1.74

20.H2
1.67

1911

Inches.
0.16
1.73
3.34
23.41
1.41'
.OS

1.14
2. lfi
1.24
3.09
2.13
1.53

21.42
1.64

1912

Inches.
0. 31
6.21
2.30
2.00
1.(4
3.42
.IIS

1 Average for the years 1891-1911 from records of the United States Weather Bureau. Rainfall from 190V
to 1912 from records at the San Antonio Experiment Farm.
-> Record for April, 1911, from the United States Weather Bureau service at San Antonio.

The rainfall for the years 1909 to 1911 w^as considerably below
normal, while that of 1909 was the lowest thus far recorded. Con-
sequently the crop yields reported in these experiments were rather
lower than may be expected under normal conditions. The year
1908, as indicated by the crop yields, w r as perhaps more favorable
than normal, although the total rainfall for the year does not indi-
cate it. The heavy rains in November of the preceding year w r ere
apparently stored in the soil and used during the season of 190S.

FACTORS INFLUENCING CROP YIELDS.

The yields of forage crops reported in the following pages are
extremely variable and often are not significant. They are pub-
lished as a matter of record, but the interpretation of the experi-
ments should be based on the printed conclusions and recommenda-
tions rather than on the figures alone. It is not always clear which
way the crop or the treatment of the land the previous season affected
the subsequent crop yields. Usually, fallowing land for several
months results in larger yields of the following crop, but this is not
always the case. The kind of crop grown the previous year also
affects the yields obtained. For instance, the small grains, such as
wheat or oats, are harvested in May or early in June, corn in July,
while cotton and sorghum occupy the land until late in October
and use any available soil moisture until harvested. A crop planted
the following March or April on land that had been devoted to cotton
or sorghum is ordinarily at a disadvantage wdien compared with a
crop on land that had been in oats or corn the previous year.

It lias seemed advisable to publish the crop yields even at the risk
of some misinterpretation, but in each case it is urged that for pur-

[Cir. 106]

FORAGE-CROP EXPERIMENTS AT SAN ANTONIO. 7

poses of practical application the notes and comments concerning
each crop or experiment be given the greater emphasis and that any
broad generalizations from the figures in the tables be avoided.

THE SOIL ON WHICH THE EXPERIMENTS WERE CONDUCTED.

The San Antonio Experiment Farm is located in the southern
extension of what is called the "Black Prairie region," known
geologically as the Upper Cretaceous formation, and is representative
of a large portion of central Texas extending from the Oklahoma line
to the Rio Grande. The surface soil, which is a black, clayey loam,
is capable of absorbing a high percentage of moisture, but tends to
puddle and bake. It is very sticky when moist, and all tillage opera-
tions are difficult because the plows and cultivators do not scour.
The subsoil is of about the same mechanical composition as the
surface soil, but it often contains more lime, which occurs as gravelly
concretions. The subsoil is also much lighter colored, probably
due in part to the lack of organic matter as well as to the larger
proportion of lime. When supplied with plenty of water, the soil is
usually very productive and has proved well adapted to growing
many of the ordinary forage crops.

CROPS ORDINARILY GROWN FOR FORAGE.

The crops that are generally grown in the San Antonio region for
forage are sorghum, oats, Johnson grass, millet, cowpeas, and corn,
while alfalfa is grown to a limited extent where irrigation water is
available. Of the crops mentioned, sorghum, oats, and Johnson
grass are the most common, and they rank in the order named in
popularity and value. The native wild grasses are used only for
pasture. Most farmers grow only sorghum for forage, and conse-
quently it has been deemed advisable to test other forage crops with
a view to suggesting some practical diversification as a basis for
systematic crop rotation. The native cactus, or Opuntia, is being
used as a forage crop and has proved to be a valuable addition to
the list, especially during unfavorable seasons. 1

SORGHUM.

Experiments designed to determine the most prolific varieties, the
best methods of planting, the proper rate of seeding, and the best
date of planting sorghum have been carried on at the San Antonio
Experiment Farm during the past five years.

i For information concerning this crop, consult the following bulletins:

■Griffiths, David. The Prickly Pear as a Farm Crop. Bulletin 121, Bureau of Plant Industry, U. S.
Dept. of Agriculture. 1908.

The "Spineless" Prickly Pears. Bulletin 140. Bureau of Plant Industry, P. S. Dept. of Agri-
culture. 1909.

The Thornlcss Prickly Pears. Farmers' Bulletin 483, P. S. Dept. of Agriculture. Pill'.

[Cir. 106]

8

FORAGE-CROP EXPERIMENTS AT SAN ANTONIO.

Sorghum may be planted any time from March 1 to August 1 with
the expectation of securing a crop, although as a rule the earlier the
planting the larger the yield. When planted early in March, from
two to four cuttings can be obtained from the earlier maturing varie-
ties, such as Red Amber, while from the later maturing varieties,
such as Sumac, two or three cuttings can be obtained under favor-
able conditions.

The varietal tests in 1908 indicated that Sumac and Red Amber
were the best varieties for this region; hence, all later tests were
carried on with these two varieties only. Table II gives the results
of a test of varieties in 1908, which was an unusually favorable year.
It has been observed that an unfavorable season often magnifies the
differences in favor of the better varieties

Table II. — Yield of sorghum varieties grown in one-tenth acre plats at (he San Antonio

Experiment Farm in 1908.

S. P.I.
No.

Variety.

Date
planted.

Number
of cut-
tings.

Yield
per acre.

17554

Sumac

Mar. 30
..do

2
3
2
2
3
2
2

Tons.
12.29

17548

Red Amber

10.17

17556

Orange ...

Mar. 18
...do

7.12

17539

Planter's Friend

6.00

20799

Minnesota Amber

...do

5.38

17569

Kafir

...do

4.83

18684

Dwarf milo

...do

4.38

The sorghum was planted in rows 4 feet apart and the plats were
one-tenth of an acre in size. As will be seen from Table II, the Sumac
is the heaviest yielder of the varieties tested, but the hay is somewhat
coarser than that from the other varieties. On the whole, the Sumac
is deservedly the most popular variety grown. (PI. I, fig. 1.)

In 1910 a comparative test of the Red Amber and Sumac sorghums
was carried out under irrigation. Table III gives the results of this
test.

Table III. — Comparative test of the Red Amber and Sumac sorghum* grown under irriga-
tion at the San Antonio experiment farm in 1910. Triplicate plantings.

Red Amber sor-
ghum.

Sumac sorghum.

Number

of
< 'in lings.

Yield
per acre.

Number

of
cuttings.

Yield
per acre.

3
3
3

Average.

Tons.
14.41
12.45
11.62

2
2

2

Tons.
14.75
14.00
12.60

12. 83

13. 78

[Cir. 106]

Cir. 106, Bureau of Plant Industry, U. S. Dept. of Agriculture.

Plate I.

Fig. 1.— Sorghum in Shock, Showing the Method of Harvesting when a Heavy
Crop is Grown. As Much as 13 Tons Per Acre of the Sumac Variety has
been Grown without Irrigation.

(Photographed June 8, 1908.)

Fig. 2. -Broadcasted Sorghum on Field B5-6. There was Practically no Hay

Yield from this Plat.

(Photographed July 27, 1910.)

Cir, 106, Bureau of Plant Industry, U. S. Dept. of Agriculture.

Plate

Fig. 1. -Drilled Sorghum on Field B5-14. The Hay Yield from this Plat was

3.76 Tons Per Acre.

(Photographed July 27, 1910.)

Fig. 2.— A Plat of the Appler Rustproof Oats. Probably the Best Variety

for the San Antonio Section.

(Photographed May 17, 1910.)

FOKAdE-CROP EXPERIMENTS AT SAX ANTONIO. 9

It will bo observed that the average yield from the three plats of
the Sumac variety was nearly a ton larger than from the three plats
of the Red Amber. Another point in favor of the Sumac variety is
that there were only two cuttings of the Sumac, while there were
three of the Red Amber, and it costs more to harvest tliree crops
than two.

METHODS OF PLANTING.

Two methods of planting are practiced by farmers, namely, plant-
ing in rows far enough apart so that cultivation is possible, and
planting closer, either in drills or by broadcasting. Practically no
sorghum is grown under irrigation in cultivated rows, but in recent
years an increasingly large area is being planted on dry land in culti-
vated rows. When sorghum is planted in rows and cultivated, the
crop may have to be harvested with a grain harvester or by hand if
there 1 is a heavy growth, but if the rows are not more than 2 feet
apart or the growth is not too heavy it may be cut with a mowing
machine and handled in the ordinary manner. Even with the rows
4 feet apart the sorghum at the field station the past four years
has been harvested in the same way as broadcasted sorghums.

Drilled compared with cultivated rows. —The yields from sorghum
grown without irrigation have been larger when planted in cultivated
rows than when the crop was sown broadcast or in 8-inch drills.
Unfortunately, this comparison was not made in 1908, which was a
very favorable season. Table IV gives the yields in this experiment
for the years 1907 to 1911.

Table IV. — Yields per acre of Sumac sorghum planted in cultivated rows 4 feet apart

and in 8-inch drills.

1907

1908

1909

1910

1911

Aver-

Method.

Num
berof
plats.

Aver-
age
yield.

Num-
ber of

plats.

Aver-
age
yield.

Num- Aver-
ber of age
plats, yield.

Num-
ber of
plats.

Aver-
age
yield.

Num-
ber of
plats.

Aver-
age
yield.

age

for

four

years.

Rows 4 feet apart

5
2

Tons.

3.20

1.68

4

Tons.
13.02

4
3

Tons.
1.68
1.07

4
3

Tons.

3.06
.74

4
3

Tons.
3.44
4.36

Tons.
2.84
1.96

8-inch drills

1 There were no plats of Sumac sorghum close-planted in 1908.

The great difference in yield between the cultivated plats and
those in 8-inch drills in 1910, the comparatively slight difference hi
1909, and the difference in favor of the 8-inch drills in 1911 are due
to the seasonal rainfall. As will be noticed by reference to the rain-
fall table (Table I), the year 1908 was a very favorable one and
much moisture that collected in the soil was carried over to 1909,
while 1909 was exceedingly dry, and 1910 was more favorable dur-
69799°— Cir. 106—13 2

10

FORAGE-CROP EXPERIMENTS AT SAN ANTONIO.

ing the early part of the season. The cultivation of the sorghum in
4-foot rows conserved a large part of the rainfall for the use of the
crop, while the plats in 8-inch drills were dependent entirely upon
the season's rain, as there was very little moisture carried over from
1909 and the rain that did fall was lost very quickly by evaporation.
Up to about May 1 the spring of 1911 was unusually favorable,
the rains coming so frequently that considerable periods passed when
it was impossible to cultivate. Thus, before May 1 cultivation did
not affect the yields as much as it would otherwise have done and
the shortage of rain in June prevented a second cutting, enabling the
drilled plat to outyield the cultivated plat. When a series of years
is considered, it is safe to conclude that for unirrigated land planting
in rows, so that cultivation is possible, will increase the yields enough
to justify this method of planting. (PI. I, fig. 2, and PI. II, fig. 1.)
The effect of the two systems of cultivation on the succeeding crops
is also important. In the rotation experiments in 1911 there was a
slight difference in the yield of corn and a very noticeable difference
in the cotton yield in favor of the land on which sorghum had been
grown in cultivated rows the previous year. Under irrigation the
advantage was still with the cultivated rows. In rows 3 feet apart
the yield was 14.41 tons per aero, while in 8-inch drills the yield was
12.6 tons. The difference in yield is, however, comparatively slight;
and when one considers the expense of cultivating the rows, the
quality of the forage, and the greater ease of handling the broad-
casted sorghum it is doubtful whether it is profitable under irrigation
to plant in rows and cultivate.

Width of rows. — A test was made in one-ten th-acre plats of sorghum
planted in rows, 2, 3, and 4 feet apart. Table V shows the yields
from these plats.

Table V. — Yields of Sumac son/hum planted in rows 2, ■ >. and 4 feet apart.

Distance between rows.

Yield per acre.

1910

1911

Average.

2 feet . .

Tons.
5.07
5.90
5.00

Tons.
3.30
2.82
3.30

Tons.

4.19

3 feet . .

4.36

4 feet .

4.15

The plats of 1910 were planted in a field summer-fallowed in 1909,
and consequently the yields are not comparable with those given in
Table IY. The land on which the experiment was conducted in 1911
was cropped the previous year. In both years the quality of the
forage was best on the plat with the rows 2 feet apart, with the forage
from the rows 3 feet feet apart second in quality. While the differ-
ences are not great, yet the quality of the forage was so much better

rr-ir ififii

FORAGE-CROP EXPERIMENTS AT SAN ANTONIO.

11

from the closer plantings and cultivation with the ordinary farm
machinery so much easier that undoubtedly a closer planting than 4
feet is advisable.

Bate of seeding in rows. — In 1910 a test of Sumac sorghum in rows
4 feet apart, with the plants 1, 2, 4, and 6 inches apart in the row,
was attempted. Owing to poor germination there was not much
dilFerence in the resulting stands of the different plats, so that the
results were of no comparative value. Table VI shows the yields of
this experiment as carried out in 1911, the rows being 4 feet apart
as before. . The figures are based on averages of actual counts of
plants at the end of the growing season.

Table VI. — Yields of Sumac sort/hum with the plants different distances apart in the row.

Distance between plants.

Yield
per acre.

Distance between plants.

Yield
per acre.

2 inches

Tons.
3.30
2.70

7 inches

Tons.
2 05

•1 inches. .

8 inches

1 97

These results are definite and conclusive, as they show very clearly
the effect of reducing the thickness of the stand. There has been
some discussion relative to the effect of thick seeding on the yields
in cultivated rows, and the foregoing results bear directly on this
question. Aside from the increased yields, the quality of the forage
is so much better in the thicker secdings that if the yield were some-
what less where thicker seeding is resorted to this method would
still be advisable.

Bate of seeding test in 8-inch drills. — There appears to be a wide
difference of opinion in this locality regarding the proper rate to
seed sorghum broadcasted or in close-drilled rows. To throw some
light on the subject, the following test was made.

On April 29, 1912, seven plats of one-tenth acre each were seeded
to Sumac sorghum in 8-inch drills, the rate of seeding varying from
25 pounds to 174 pounds per acre. Table VII presents the results
of the experiment.

Table VII. — Rate-of -seeding test of Sumac sorghum in 84nch drills.

No. of plat.

Seed
per acre.

Yield
per acre.

1

Pounds.

20

37

52

88

121

153

174

Tons.
4 54

2

4 97

3

5 12

4

5 27

5

5 nl

6

4 r >">

7

4 'JO

[Cir. 1061

12 FORAGE-CEOP EXPERIMENTS AT SAN ANTONIO.

This season the best results were obtained from sowing 88 pounds
to the acre, with 52 pounds and 121 pounds second and third, re-
spectively. With these results as a basis, to seed a much greater or
less amount does not appear to be advisable.

The quality of the hay on plat 4 was fully as good as on the plats
with a thicker seeding, and was naturally much better than on plats

1 and 2.

SORGHUM-LEGUME MIXTURES.

In 1008 a series of experiments was begun to test the growing of
legumes intermixed with sorghums for forage. Considering the
extent to which sorghum is grown and the way it reduces the pro-
ductivity of the land on which it is grown continuously, there seems
to be need of some practical way to lessen the injury to the soil, as
well as to produce a better balanced ration. This work has a direct
bearing on the production of a better quality, as well as on the main-
tenance of the quantity of forage produced.

METHODS OF PLANTING.

Different varieties of cowpeas and other legumes and several
different methods of planting were tested in 1909 on a small scale,
to determine as far as possible from one year's work the best varie-
t ies to grow and the best methods for growing them. The sorghum
varieties were the Sumac and the Red Amber and the legume varieties
the Iron, Whippoorwill, and Unknown cowpeas and a variety ol
kulthi (Dolichos Ujlorus). Two methods of planting w-ere adopted
in 1909, the row method and the band method. The band method
consisted of planting two rows of sorghum 2 feet apart, with one row
of the legume between, the bands being planted 3 feet apart and the
intervening space kept cultivated. In the row method the sorghum
and legume seeds were plant ed in the same row, the rows being 4.1
feet apart and the ground kept cultivated between the rows. The
band method was not as successful as the row method, apparently
because the legumes did not receive the benefit of cultivation. Of
the legumes used in the mixtures, the cowpeas did the best, the
Unknown cowpeas being on the whole slightly better than any of
the other varieties. Table VIII gives the yields of the mixtures.
There was one-thirtieth of an acre hi the plats planted by the band
method and one-fortieth of an acre in those planted by the row
method. The land was summer-fallowed the preceding year and
was a Johnson-grass meadow previously.

[Cir. 106]

FORAGE-CROP EXPUEIMENTS AT SAN ANTONIO.

13

Table VIII. — Comparison of the row and band methods in 1909, on land summer-
fallowed the preceding near.

Mixture.

Sumac sorghum

Iron cowpeas

Sumac sorghum

Unknown cowpeas

Sumac sorghum

Whippoorwill cowpeas. -

Sumac sorghum

Knlihi i Dolichos biflorus)

Average

Yield per acre.

Table IX. — Comparison of the roiv and band methods in 1910 on land cropped to son/hum

the preceding year.

Mixture.

Yield per acre.

Row
method.

Band
method.

Sumac sorghum

Unknown cowpeas

Sumac sorghum

Brabham cowpeas. ..."..

Sumac sorghum

Kulthi ( Dolichos biflorus I

Pearl millet

Kulthi (Dolichos biflorus)

Average

Tons.
1.76

1.18

.70

1.17

1.21

1.21

Of the cowpeas planted by the band method fully 50 per cent
died before reaching maturity, while the legumes planted by the row
method made a good growth, although the sorghum was too thick,
not allowing the legumes to develop as well as they probably would
have done had the sorghum plants been farther apart. The band
method, it will be noted, outyielded the row method in 1909, but this
was due to the predominance of the sorghum and is no indication of
the comparative success of the two methods, as the forage from the
row method contained a much greater percentage of cowpeas. The
total yield from the two methods of planting was the same in 1910,
but the value of the forage from the row method was much higher,
owing probably to the fact that the cowpeas had a better chance to
develop than when the band method was used.

A test of mixtures seeded in 8-inch drills was made under irrigation

but was entirely too thick, so that the legumes were crowded out.

The yields were about the same as would have been expected had the

sorghums been planted alone. It is very doubtful whether it is advis-

[Cir. me,]

14

FORAGE-CROP EXPERIMENTS AT SAN ANTONIO.

able to plant the sorghum-legume mixtures under irrigation in this
manner, and it is certainly not advisable to do so without irrigation.
Even with a thin rate of seeding the sorghum plants stool profusely
and have a tendency to crowd out the shorter leguminous plants.
So far, none of the tests with this method of planting have been satis-
factory.

On a field which was fallow in 1909 and planted in Johnson grass
previously, a rate-of-seeding test for sorghum-legume mixtures was
made in 1910, to determine as far as possible in one year's test the
best proportion of each seed for the mixture. The row method of
planting was adopted and the rows were 3^ feet apart, the plats being
one-tenth of an acre in size. Sumac sorghum and Unknown cowpeas
were the varieties used in this test. Table X shows the results of
this test.

Table X. — Yields of rate-of-seeding test of sorghum-cowpea mixtures, planted by the

row method in 1910.

Mixture.

Sorghum
Cowpea .
Sorghum
Cowpea .
Sorghum
Cowpea.
Sorghum
Cowpea .

Stand,

plants

in 50

feet.

95
90

74

100

70

105

30

100

Number

ol-
rul tings.

}

Yield

per acre.

Tons.

4,:

5. 15
4.02
2.35

With all points considered, the second plat showed the best rate of
seeding, although a test of one year is not conclusive. (Fig. 1.)

In the first plat the sorghum appeared to be a little too thick and
crowded the cowpeas, reducing the quantity of forage from them. In
the case of the last two plats the sorghum was entirely too thin, mak-
ing it rather coarse, but the cowpeas made an excellent growth.

While the combination of sorghum and cowpeas is recommended,
especially for the irrigation farmer, yet it should be borne in mind
that extreme care must be exercised to secure the proper mixture.
If the sorghum is too thick, the cowpeas will not properly develop and
the forage will be little better than if the sorghum were planted alone.
On the other hand, if there is too great a proportion of cowpeas the
yield will be materially reduced and will prove a disappointment to
the grower.

Sumac sorghum is undoubtedly the best variety to use when these
mixtures are planted, owing to its leafy character and slow growth at

[Cir. 106]

FORAGE-CROP EXPERIMENTS AT SAN ANTONIO.

15

the beginning of the growing season. Early-maturing sorts like the
Red Amber crowd out the cowpeas.

The yields from the plats of sorghum indicate that fully as much
fodder can be obtained in the mixture as by seeding alone; therefore,
the advantage derived from the admixture of the cowpea hay, if
not obtained at too great an expense in seeding and cultivation, is
well worth while.

Fig. 1.— Sumac sorghum and Unknown cowpeas planted in the same row, showing the excellent growth
made by both the cowpeas and the sorghum. There are about three plants of the sorghum to four of
the cowpeas. (Photographed July 27, 1910.)

SMALL GRAINS AS FORAGE CROPS. 1

While in the immediate vicinity of San Antonio the grain yields
from such cereals as oats, barley, wheat, and rye are low, yet oats,
and rye to a lesser extent, are quite extensively grown for forage.
Oats are also very useful for whiter pasturage, and are generally a

' The growing of cereals for grain production only at low altitudes in the vicinity of San Antonio is not a
profitable venture. The principal cause of low yields is the susceptibility of grains to rust. In higher alti-
tudes cereals are grown to a considerable extent. In the high land near Kerrville, and even no farther
distant than the town of Boerne (30 miles), satisfactory yields have been realized from wheat and>oats.

[Cir. 106]

16

FORAGE-CROP EXPERIMENTS AT SAN ANTONIO.

profitable crop. The oats are ordinarily seeded in October or Novem-
ber and are not pastured until the plants are well established, which
is some time in December, depending on the weather conditions. If
a hay crop is to be secured, the stock is taken off as soon as the spring
growth starts. Little is known as to the extent of the damage done
to the crop by grazing, but farmers believe that the whiter pasturage
alone hi a year of normal rainfall is worth all it costs.

Table XI shows the yields of oats grown during the years 1908 to
1911. In 1910 three local varieties were obtained to compare with
four varieties introduced by the United States Department of Agri-
culture. These varieties were the Texas Red Rustproof, Hastings'
Hundred Bushel, and the Winter Turf. The plats were one-fifth of
an acre in size, and the land had been summer-fallowed the preceding
year in each case.

Table XI.

Yields of oats hay at the San Antonio Experiment Farm, in 1908-1911,

inclusive.

Variety.

Yield of hay (for single plats, except as indicated).

1908

1909

1910

1911

Average.

Appier Rustproof

Pounds.

4,720
4,170

Pounds.

2,399

1 1,990

Pounds.
2 1,474

2 1,146
1,260

800

3 2, 000
3 1,780

3 940

Pounds.
1,855
Failed.
2,120
2,130
1,730

Pounds.
2 624

Culberson Winter

1 826

Texas Red Rustproof

1 690

Hastings' Hundred Bushel

1 465

W inter Turf

1 865

l 780

Snoma

940

1 Averages of four plats.

2 Averages of five plats.

3 The actual weight on these plats was less than the figures indicate, as the hay was not thoroughly dry
when weighed.

From the tests made thus far the variety of oats known as the
Appier Rustproof has proved to be the best for this locality, although
the 191.1 test apparently does not indicate this superiority. The low
yield of the Appier Rustproof oats hi 1911 was probably due to the
following causes: The four plats, which were planted with different
rates of seeding, were located north of four plats of Culberson winter
oats, which were so severely affected with rust that they were not
worth cutting. As the prevailing winds are from the southeast, the
rust spores were undoubtedly carried from the Culberson Whiter oats
to the Appier Rustproof oats, causing a much more severe infestation
than would otherwise have been the case. A plat of Appier Rust-
proof oats on the south side gave a hay yield of 2,260 pounds, and the
average yield of this variety in the rotation plats was 2,437 pounds
per acre. (PI. II, fig. 2.)

[Cir. 100]

FORAGE-CROP EXPERIMENTS AT SAN ANTON lo.

17

Since the Appier Rustproof oats proved to be the best, in the fall of
1909, 39 half-bushel lots of these oats were sent to different farmers in
the San Antonio region. Later a circular letter was sent out, request-
ing a report on the results obtained with the variety, and its value as
compared with the variety ordinarily grown. Twenty-six replies
were received, of which 11 were favorable, 3 , unfavorable, and the
remaining 12 equally divided — 6 did not think the test was satisfac-
tory, and found little difference between this variety and that com-
monly grown.

In 1909 a ratc-of -seeding test was made to determine the rate that
would give the highest yield. The varieties of oats planted were
Appier Rustproof and Culberson Winter. Table XII presents the
results of the test.

Table XII. — Yields <;/' Appier Rustproof and Culberson Winter oats in n rale-qf-

seeding test, 1909.

Variety.

Kale of
set 'ding.

Approxi-
mate num-
ber of seeds

per acre.

Yield per acre.

Si raw. Grain

Culberson Winter

I ><>

Do

.1)0

A ppler Kusl proof

Do.-

Do

Do

Pounds.
59
43
28
23
59
43
30
24

2. 124, 000
1,550,000
1,008,000
828, 000
950,000
700,000
480. 000
384,000

Pounds.
1,750

1,835
2,065
2,310
2,600
2,500
2,225
2,200

Bushels.
6. 87
7.85
11.76

13. 05

19. 06
is. (HI
16. 13
13.44

Apparently conflicting results were obtained hi this test. With
the Culberson Winter oats the lightest seeding nearly doubled the
grain yield and gave an increased yield of 560 pounds of hay. With
the Appier Rustproof oats, however, the heaviest seeding yielded
5.62 bushels of gram and 400 pounds of hay more than the lightest.
By taking a count of the oats it was found that there were about
two and one-half times as many seeds of the Culberson Winter oats
as there were of the Appier Rustproof oats in a single pound, so that
in the heaviest seeding of the latter the number of seeds per acre was
only slightly greater than the lightest seeding of the former.
Although it is to be expected that the seasonal variations will change
the results somewhat, yet the results from the season of 1909 indicate
that 2 bushels of seed per acre is about the right quantity to sow.
This test shows conclusively that the same rates of seeding of two
varieties are not necessarily comparable and that the size and weight
of the seed are important factors to be considered.

For further information as to the best rate of seeding this experi-
ment was repeated in 1910. Table XIII shows the yields for the year.

[Cir. 106]

18

FORAGE-CROP EXPERIMENTS AT SAN ANTONIO.

Table XIII. — Yields in a rate-of-seeding test of Appier Rustproof and Culberson

Winter oats, in 1910.

Variety.

Rate of
seeding.

Yield per acre.

Straw.

Grain.

Appier Rustproof. .

Pounds.
71
59
45
30
37
28
23
14

Pounds.
1,890
1,225
1,609
1,320
1,000
1,200
1,280
1,050

Bushels.
21.5

Do

14.0

Do

16.4

Do

15.2

Culberson Winter .

8.6

Do..

7.2

Do

9.1

Do

6.6

With Appier Rustproof oats the heaviest seeding gave the highest
yield, but with Culberson Whiter oats there was so little difference
that the deviation could be attributed to variation in the soil.

OTHER SMALL GRAINS THAT MAY BE USED AS FORAGE CROPS.

Other small grains which have been tested for use as forage crops
are wheat, barley, emmer, and rye. Table XIV shows the yields
from these cereals. These plats were one-tenth of an acre in size and
the land was summer-fallowed the previous season.

Table XIV. — Hay yields from various cereals tested for forage.

( Viral.

Hay yield.

1910

1911

Abruzzes rve

Pounds.
2,480
1,100
1,740
'800

Pounds.
3,380

Rieti wheat

Black winter emmer

Barley

2 1,951

1 Average yield from six varieties.

2 Average yield from eighl varieties.

With the exception of rye, none of the cereals mentioned in Table
XIV gives indications of being of any great value for this section,
from a forage standpoint. With the exception of rye, all those listed
are very susceptible to rust. While the variety of rye grown appears
to be somew r hat better than oats from the standpoint of yield, yet the
value of oat hay is somewhat higher. Rye comes second on the list
of cereals for hay and is recommended next to oats.

OTHER FORAGE CROPS.
JAPANESE SUGAR CANE.

A one-tenth-acre plat of Japanese sugar cane was planted at the field
station on March 22, 1910. This is a forage crop of value in Florida

[Cir. 106]

Cir. 106, Bureau of Plant Industry, U. S. Dept. of Agriculture.

Plate III.

Fig. 1.— Japanese Sugar Cane Grown Under Irrigation. A Promising Crop for

THE IRRIGATIONIST.
(Photographed November 1, 1910.)

Fig. 2.- Unknown Cowpeas Planted in Rows 3 Feet Apart. Cowpeas are a
Crop that should be More Widely Grown in the San Antonio Section.

(Photographed July 27, 1910.)

Cir. 106, Bureau of Plant Industry, U. S. Dept. of Agriculture.

Plate IV.

Fig. 1.— Canada Field Peas Grown without Irrigation. These Peas Gave a

Yield of 1.9 Tons per Acre.

(Photographed May 19, 1911.)

Fig. 2.— Guinea Grass Grown Under Irrigation. This Grass has Possibilities
as a Forage Crop when Irrigated.

fPhotoeraiiln'il November 1, I'.tlO.I

FORAGE-CEOP EXPERIMENTS AT SAN ANTONIO. 19

and the Gulf coast region. The yield from the plat, planted under
irrigation, was 13.5 tons per acre in 1910, and 11.95 tons per acre in
191 1 . (PL III, fig. 1 .) To obtain the best results late fall planting is
recommended. This cane promises to be a valuable addition to the
list of forage crops that can be grown under irrigation in this section,
as it makes a large yield of very palatable fodder.

In the plats tested the canes were planted in furrows about 6 inches
deep and 6 feet apart. Ordinary cultivation was given the crop.
Mr. John M. Scott, in a recent bulletin, 1 says that the number of whole
canes required to plant an acre is about 3,000. He suggests that the
rows be made about 8 feet apart and that the canes be cut in pieces
having three or four eyes to each piece and dropped in a double line.
There is some reason for believing that a closer planting than 8 feet
may be advisable in Texas as the 1 canes would then be less coarse.

ALFALFA.

A number of tests have been made with different varieties of alfalfa
and with different methods of seeding this crop. None of the results
so far obtained have been sufficiently satisfactory to warrant alfalfa
being grown on a commercial scale without irrigation. The principal
reason for this difficulty seems to be that in the immediate locality of
San Antonio it is not possible to store enough moisture in the soil to
carry the plants over the periods of drought that occur in the summer.
If there were no loss by surface run-off and evaporation, both of which
are important factors in the section mentioned, the crop could prob-
ably be grown at a profit without irrigation. Alfalfa is being grown
successfully in some localities in this region under irrigation, but
apparently is not adapted to this section unless it can be irrigated
occasionally.

In 1907 a series of plantings of a number of varieties of alfalfa were
made in 6-inch drills, in single rows 2 feet apart and in double rows
about S inches apart, with a cultivated space of 24 inches between.
These plats were not irrigated. Of the three methods of planting, the
double-row method, with the intervening space kept cultivated, proved
to be the best, although the yields from the plats planted in this
manner were so low that they were not profitable, even during the
favorable season of 1908. Six strains were also seeded in broadcast
plats in December, 1906. Two of the plats died out almost entirely,
the stand in many of them being reduced to 5 or 10 per cent. Several
plats seeded September 25, 1906, gave much better results, the stand
in October of the following year being about 90 per cent. Thorough
inoculation of the soil seemed to be decidedly beneficial, the plat

i Scott, John M. Japanese Sugar Cane for Forage. Bulletin 105, Florida Agricultural Experiment
Station.

[Cir. 1061

20 EORAOE-CROP EXPERIMENTS AT SAN ANTONIO.

inoculated with soil giving much the best growth. The cuttings con-
tinued so light, however, that there was not sufficient return for the
expense of culture.

Table XV .—Comparative yields of varieties of alfalfa.

S. P. I.

No.

Source of variety.

Yield per
acre.

12549.
12846.
14786.
L8627.

L8628
18664.

Argentina

Tripoli

Turkestan

Arabia (irrigated seed)

Arabia (nonirrigated seed; .
Provence, France

Pounds.
400
320
4S0
120
400
80

The best yield made by alfalfa in rod-square plats was 880 pounds
an acre at one cutting. The other cuttings were only clippings
made for the purpose of protecting the alfalfa from the drought as
much as possible.

Four varieties of alfalfa were seeded under irrigation in one-tenth
of an acre plats in September, 1909. Two of these plats were almost
completely ruined, and the other two badly injured by root-rot before
July, 18, 1910. The root-rot proved, perhaps, more destructive
than usual in these plats.

POSSIBILITIES OF ALFALFA.

Where alfalfa is grown in the vicinity of San Antonio under irri-
gation it is generally necessary to replant the fields or portions of
them every fall. Where this is clone the reported yields indicate
that alfalfa is a fairly satisfactory forage crop. On the Sewer farm,
south of San Antonio, five to eight cuttings are ordinarily obtained,
although the last cuttings are light and mixed with native grass and
weeds. Where root-rot has destroyed the plats the vacant areas
are taken up by grass and weeds.

It must be admitted that the possibilities of alfalfa in soils infested
with root-rot are limited, and the writer has yet to see a cultivated
field of any area in the black soil of this section that is not infested
with this fungus. That the light, sandy soils south of San Antonio
are in some instances less infested, if not entirely free from the
fungus, has also been observed, but whether these isolated areas are
of sufficient extent to permit the establishment of permanent alfalfa
fields is yet to be demonstrated. The best superficial method of
determining whether the root-rot fungus is in the soil is to observe
the cotton fields. If there are small areas of varying size with dead
cotton plants, the farmer will be safe in concluding that the fungus is
in his fields.

[Cir. 106]

FORAGE-CROP EXPERIMENTS AT SAN ANTONIO.

21

CLOVER.

So far, only one variety of clover appears to be adapted to this
section. This is called shaft al clover {Trifolium suaveolens).
Only one test was made with this crop, but the plants made an ex-
cellent growth, and this clover may have possibilities as a winter
forage, although it is rather susceptible to low temperatures when
not well established. Five other clovers have been tested, including
two varieties of crimson clover (Trifolium incamatum) , and one
variety of sweet clover (Melilotus alba). The best plat of crimson
clover, which was planted in November, 1907, gave a yield of 1,240
pounds per, acre the following spring. The plat of sweet clover
gave a yield of 1,640 pounds per acre. These plats were one-twen-
tieth of an acre in size. Neither of the last two clovers was at all
promising.

COWPEAS.

The cowpea is apparently as well adapted to the section of San
Antonio as any other leguminous plant that has been on trial. It
is of value as a forage crop, but is particularly useful as a green
manure, and in a well-planned rotation there is no more valuable
plant. One great difficulty with the local soil is its compactness and
tendency to run together and bake. Plowing under a crop of cow-
peas every three or four years will do much to make the soil more
friable, and if the rotation is well planned the yields should eventu-
ally be increased.

The best results have been obtained from growing cowpeas in
cultivated rows about .3 feet apart. No complete failure of this
crop has been recorded in the five years it has been grown, and good
yields have been secured under favorable conditions. Ordinarily from
1J to 2^ tons of hay per acre are obtained, although under favorable
conditions the yields may be greater. Several varieties have been
grown, but those that have proved best adapted to local conditions
are the Whippoorwill and the Unknown. (PI. Ill, fig. 2.)

Table XVI.- Yields of cowpeas without irrigation.

s. P. i.
No.

24919
25078

24414
17330

L'tiSKI

Variety.

Red Ripper

droit ,

Brabham

Chinese Whippoorwill.
Unknown

Yield per acre.

1909

Tans.

0. 40
.68
.72
.92

1910

Tons.

1.38
"i.93

[Cir. 106]

22 FORAGE-CROP EXPERIMENTS AT SAN ANTONIO.

CANADA FIELD PEAS.

For a winter legume the Canada field pea does fairly well. In the
winter of 1908-9, which was extremely unfavorable, a fair crop was
produced, and an excellent crop was grown during the whiter of
1907-8. This pea does not successfully withstand the hot summer
weather, and consequently it can be grown profitably only as a winter
crop. The peas should be planted about the first of November or
earlier, in order to become well established before a severe freeze
occurs. Later planting sometimes proves successful, but it is not
generally advisable. It is necessary to plant early enough to have
the crop mature before the hot months of the summer. In this
respect Canada field peas are much like oats.

A somewhat larger test was made during the winter of 1910-11,
when three varieties wore grown. Owing to the severe freeze of Jan-
uary all of the varieties were killed nearly to the ground and some
of them were killed outright. S. P. I. No. 22042 (McKay) proved to
be the hardiest of those grown. Three plats one-tenth of an acre in
size were devoted to as many different methods of planting.

Table XVII. — Yields in 1911 of ( 'anada field peas planted alone and with oats in 1910.

Method of planting.

Peas and oats mixed, proportion 1 to 5, 12-inch drills.

Peas alone, 12-inch drills

Peas and oats mixed, proportion 1 to 5, 24-inch drills.

Rate of
seeding.

Pounds.

112

120

56

Yield

of hay

per acre.

Tons.
1.23
1.94
1.16

It is doubtful whether any combination with oats will give as high
a yield as when the peas are grown alone, as will be seen from the
yields shown in the table. It will be observed by consulting Table
XI that the peas gave a much larger yield than did any of the varie-
ties of oats (PI. IV, fig. 1), and being a leguminous crop they also
improved the soil on which they were grown. The hay is extremely
palatable, and being rich in protein it is of value to mix with oat or
sorghum hay. As a winter green-manure crop it is of particular
value.

A rather more extensive test of this crop was made in the winter
of 1911-12. About 50 varieties bearing S. P. I. numbers were plant ed
in 1-rod rows, but owing to the ravages of birds it was necessary to
replant, making the resulting crop so late that the test was of little
practical value.

Plantings in the orchards on the farm resulted more satisfactorily.
S. P. I. Nos. 18806 (Erfurt), 30307 (Canadian Beauty), and 30134

[Cir. 100]

FoliAGE-CROP EXPERIMENTS AT SAN AXTOXIO.

23

(Golden Vine) were seeded with a grain drill. Nos. 18806 and 30307
were completely destroyed by the unusually severe winter, but No.
30134 came through with practically a perfect stand. The hay yield
from this planting was at the rate of 2,783 pounds per acre and the
grain yield 14 bushels. The hay yield in this instance does not
represent what it would have been had the vines not been allowed
to mature. The "vines were so ripe when cut that the greater part of
the leaves dropped oil'.

OTHEE LEGUMES.

Table XVIII gives the yields of a number of leguminous crops
tested in 1909.

Table XVIII. — Yields of legumes in 1909.

s. r. I.
No.

Variel v.

Yield

per acre.

212S6
22025
17534
21600

Kulthi (Dolichns biflorus)

Bonavist beans (Dolichos lablab). . .
do

Moth beans (Phaseolus aconitifolius)

Pounds.

320
4G0
520
900

Four species of Stizolobium bearing S. P. I. numbers were tested in
2-rod rows in 1910 and gave promise of being desirable legumes for
this section. No. 24870 (the Bengal bean, Stizolobium aterrimum)
and No. 25715 (the Florida velvet bean, S. deeringianum) were the
most promising. The only objection to this legume which has been
suggested is its slow growth. From the time of planting to the
flowering of S. deeringianum 164 days intervened. The quantity of
forage was much greater than that produced by any other legume
that has been on trial.

MILLET.

As was noted in a previous publication, 1 millet may be used to
advantage as a catch crop when an early crop of hay is desired or
when the seasonal rains are very late. A crop may be matured in
40 to 60 days, and a yield as high as 1 to H tons of hay per acre has
been obtained from the better varieties of early-maturing millet.
The slower growing varieties give, as a whole, a somewhat larger yield
per acre under favorable moisture conditions, but they are not so sure
of maturing as the early ones. On the whole, millet is not as desir-
able, as many other forage plants, because of its lack of drought
resistance. Table XIX shows the yield-; of the varieties tested in
1908.

1 S. H. Hastings. The Work of the San Antonio Experiment Farm in 1908. Circular 34, Bureau of riant
Industry, U. S. Dept. of Agriculture. 1909.

[Clr. 106]

u

FORAGE-CROP EXPERIMENTS AT SAN ANTONIO.

Table XIX. — Number of days necessary to mature several varieties of millets grown in

1908, with yield of hay.

s. P.I.
No.

21533
21287
22340
22425
21601
22423
22420
22426
22427
22424
21074
22422

Valid \ .

Kutki (Panicum psilopodium)

Shama (Panicum colonuvfi)

German ( Chaetochloa italica)

do

Sanwa (Panicum frumentaceum)

( 'ommon ( Chaetochloa italica)

Kursk ( Chaetochloa italica)

Hungarian ( ( 'haetochloa italica )

Japanese (Panicum frumentaceum)

Siberian ( Chaetochloa italica)

Broom-corn millet (Panicum miliaci um i
do

Yield
per acre.

Tons.
2.50
1.93
1.70
1 65
1 60
1.57
1.56
1.50
1.34
1.25
.25
.25

Pearl millet (Pennisetum americanum) was grown in 1910 under
irrigation. It gave two cuttings, totaling 10.3 tons per acre, but the
hay was of poor quality and the yield less than that of Sumac sorghum
under like conditions.

In 1 009 and 1910 tests were made, but the season was very unfavor-
able and practically no hay yields were obtained.

On the whole, millets are not to be recommended for use in the
cropping system in the locality of San Antonio, since they are not
dependable as a crop without irrigation, and under irrigation certain
grasses are much more productive.

GRASSES.

Up to the present time no perennial grass except Johnson grass
has survived for any great length of time and given a satisfactory
yield in the locality of San Antonio. There is need of a good grass,
especially by the irrigation farmer, owing to the fact that alfalfa,
clovers, etc., are attacked by root-rot and that none of the grass
family are subject to this disease. Johnson grass is grown com-
mercially in a few cases, but on the whole it is rightfully considered
more of a pest than a profitable crop. Owing to the nature of its
rooting system it is capable of withstanding long periods of drought
without dying out, but its culture should be discouraged, as there are
much more profitable grasses without the objectionable features of
Johnson grass which can be grown in its place.

Para grass (Panicum barbinode) has been tried, and although it
made a luxuriant growth under irrigation at the field station, when
it was tried without irrigation it made practically no growth and
completely died out during the season of 1909. The plat that was
irrigated proved somewhat difficult to handle on account of the
thick mass of roots which had formed during the two years the grass
was being tested. The ground on which it was grown was very

[Cir. 106]

FORAGE-CROP EXPERIMENTS AT SAX ANTONIO. 25

difficult to plow, requiring four heavy mules hitched to a sulky
plow to break the laud in the spring. This may be an objection to
its being grown on a large area, unless the roots are allowed to decay
before plowing. In the coast country it is being grown quite exten-
sively and the reports show it to be a valuable pasture and hay grass.
If the field is not irrigated for a season it is said that there i^ no
difficulty in plowing the land.

The plat of Para grass just mentioned was transplanted in field
D3 in February, 1909, but was not irrigated that year. In 1910
this plat received eight irrigations and a hay yield of 4.65 tons was
obtained.

Guinea grass (Panicum maximum) has also been tried and appears
to be somewhat more promising than Para grass. The plat was
treated the same way as the plat of Para grass and in 1910 gave a
yield of 5.29 tons per acre in two cuttings. This grass sets seed
freely, while Para grass has not been known to seed hi this country.
On the plat of guinea grass the stand was very thin in 1909, but in
1910 there was an excellent stand brought about by the grass reseed-
ing itself. (PI. IV, fig. 2.)

In July, 1910, a one-eleventh-acre plat was planted to Rhodes-gra>s
(CJdoris gayana), which made a very satisfactory growth. From two
cuttings 4.5 tons per acre of very excellent hay was obtained.

Owing to the somewhat severe winter of 1910-11 the spring stand
of grass from all these plats was very poor, as they had been badly
winterkilled. As this has not happened before, there is no reason
for making an adverse report on any of the varieties until more
extensive trials have been made.

MINOR FORAGE CROPS.

LENTIL.

Six varieties of lentil (Ervum lens) were planted in February, 1908.
None of the varieties tested gave promise of sufficient value to war-
rant extensive experimentation.

MUNG BEANS.

In the spring of 1908 two varieties of urd (Phaseolus max), S.P.I.
Nos. 16129 and 17309, and one variety of mung bean {Pliaseohis
radiatus), S. P. I. No. 16793, were tested. While these beans proved
somewhat more promising than some other leguminous plants tested,
they did not show indications of equaling either cowpeas or stizolo-
biums, and so were discarded.

CHICK-PEAS.

In March, 1908, seven varieties of chick-peas (Cicer arietinum) were
tested. These varieties seem not to be adapted to this district.

[Cir. 106]

26 FORAGE-CROP EXPERIMENTS AT SAN ANTONIO.

GUAR.

In 1906 and 1907 a plant introduced from India called guar (Cyamo-
pis tetragonoloba) was tested. While this made an excellent growth
and proved to be a most drought-resistant plant, the only live stock
that would eat it were sheep and goats. The stalk is rather inclined
to be woody, the foliage scant, and the seeds small, although it is
very prolific. The fact that stock did not eat the guar may have
been due in part to the age which the plants were allowed to reach
before being harvested.

In India the pods are said to be used as a vegetable, while in other
cases it is used as a stock food or as a green-manure crop. Accord-
ing to Sir George Watt, 1 when the plant is cultivated as a vegetable
it is grown on highly manured land and apparently is well supplied
with moisture, as it is generally planted in May and irrigated until
it rains. The unsatisfactory showing made by the plant here may
have been caused partly by a lack of appreciation of the requirements
of the plants in order to secure the best results, as the tests were made
without irrigation and during comparatively dry years. It is prob-
able that the greatest value of guar for this section will be as a green-
manure crop.

VETCHES.

About 30 different lots of vetches were tested on irrigated land
during the winters of 1906-7 and 1907-8, but none of them showed
indications of being a reliable winter crop. Under the most favor-
able conditions the better varieties made a fair growth, but when the
conditions were at all adverse, especially as to moisture supply, the
growth was very unsatisfactory. Of the varieties tested, hairy vetch
(Vicia villosa) and scarlet vetch (V.fulgens) made the best showing.
Further trials of the vetches under irrigation are needed.

SOY BEANS.

In 1907 and again in 1910 several varieties of soy beans were
planted and were entirely destroyed by rabbits, as the areas planted
were small. Undoubtedly either large areas will have to be grown
or the farm inclosed in a rabbit-proof fence if this crop is to be grown
successfully.

RAPE.

A plat of Dwarf Essex rape was grown during the winter of 1906-7.
While the plants made a strong early growth, they were later badly
affected by the drought, and in February were attacked by the har-
lequin cabbage bug, which entirely destroyed the crop.

PEANUTS.

A plat of Spanish peanuts was grown in 1908. While the plants
made a fair growth, yet then' behavior in this heavy black soil and

i Watt, George. The Commercial Products of India, pp. 449-450.
I fir. 106]

FORAGE-CROP EXPERIMENTS AT SAN ANTONIO. 27

the unevenness of the distribution of rainfall make the growing of
the crop in this section inadvisable.

SUMMARY.

Though cotton lias heretofore been the principal crop of the San
Antonio region, the increasing demand for roughage has made forage
crops as profitable as cotton.

( !rop seasons are such that green forage may be available during the
entire year under favorable conditions.

In' the selection of crops drought resistance is of primary impor-
tance; the consideration of earliness is probably second in impor-
tance.

Sorghum is the most widely planted forage crop of the section, and
the Sumac variety has given the highest yields.

Sorghum planted in rows so that cultivation is possible has given
better yields on nonirrigated land than that sown in 8-inch drills.
Rows 3 or 3^ feet apart are preferable to those of either a greater or
less distance, yield and quality considered.

Oats and rye are valuable as whiter forage. The Appier Rustproof
oats have given the highest yields of seven varieties grown.

Japanese sugar cane is a new forage crop well adapted to this section.

Alfalfa can not be successfully produced without irrigation. On
irrigated land frequent reseeding is necessary to maintain the stand
on account of root-rot.

Cowpeas are the best summer leguminous crop for this section.
The Whippoorwill and the Unknown have proved to be the best
varieties of those tested.

Canada field peas have proved particularly well adapted to the
local conditions in the vicinity of San Antonio as a winter crop.

Millets are not as well adapted to these conditions as many other
forage crops because of their lack of drought resistance.

Xo perennial grasses are grown without irrigation except Johnson
grass, and the growth of this is not recommended.

Para grass (Panicvm l>arbinode) and guinea grass (/'. maximum)
have given fair yields when grown under irrigation.

Rhodes-grass (CMoris gayana) appears to be somewhat more prom-
ising than either Para grass or guinea grass.

Approved:

James Wilson,

Secretary of Agriculture.

Washington, I). (\, November 9, 1912.

I Cir. 106]

o

Issued February 7. 1913.

U. S. DEPARTMENT OF AGRICULTURE.

BUREAU OF PLANT INDUSTRY— Circular No. 107.
B. T. GALLOWAY, Chief of Bureau.

A VARIETY OF MAIZE WITH SILKS MATURING
BEFORE THE TASSELS.

tfeJMny

Ne * vo* K

BY

G. N. COLLINS,

Botanist, Crop Acclimatization and Adaptation
Investigations.

70277° fir 1(17 13

' 10 WASHINGTON : GOVERNMENT PRINTING OFFICE : 1913

[Cir. 107]
2

BUREAU OF PLANT INDUSTRY.

Chief of Bureau, Beverly T. Galloway.
Assistant Chief of Bureau, William A. Taylor.
Editor, J. E. Rockwell.
Chief Clerk, James E. Jones.

ADDITIONAL COPIES o f this publication
A may be procured from the Superintend-
ent of Documents, Government Printing
Office, Washington , P. C. , at 5 cents per copy

B. P. I.— 799.

A VARIETY OF MAIZE WITH SILKS MATURING
BEFORE THE TASSELS.

INTRODUCTION.

In the varieties of maize cultivated in the United States the anthers
usually open before the silks on the same plant appear, and pollen is
still falling when the silks are in a receptive condition. Except where
high and constant winds prevail at flowering time, this results in a
large proportion of self -pollinated seed.

In maize self-pollination so reduces the vigor of the plants that any
means by which it can be avoided or reduced is worthy of careful
consideration. In nature wind-pollinated species, such as maize,
usually avoid self-pollination by producing pollen at a time when the
stigmas of the same plant are not in a condition to be pollinated. To
have the pollen produced before the stigmas are receptive (proteran-
dry) or the stigmas receptive before the pollen falls (proterogyny)
will prevent self-pollination.

Of the two methods proterogyny would appear to be the more
advantageous, and is in fact more widespread, though Kerner's
statement x that "As far as we can tell at present all monoecious
plants are protogynous" has been materially modified. According to
Hackel, 2 proterandry is more general than proterogyny among the
grasses. Of the primitive varieties of maize from tropical America
that have been under observation in this country, nearly all are more
decidedly proterandrous than the improved varieties grown in the
United States.

In spite of the fact that primitive varieties of maize are more often
proterandrous than improved varieties, there are reasons for believing
that the more remote ancestors of maize were proterogynous.
Whether teosinte (Evchlaena mexicana Schrad.), the nearest wild
relative of maize, is naturally proterandrous or proterogynous, seems
not to have been recorded. In our experiments one type of Euch-
laena secured from Durango, Mexico, was proterandrous to about
the same extent as the unimproved varieties of maize. Samples of

i Kerner, A. J., and Oliver, F. W. Natural History of Plants, vol. 2, 1895, p. 313.
2 Hackel, E. " Gramineap," in Engler and Prantl, Naturliehen l'flanzenfamilien, Th. 2, Abt. 2, p. 2.
[Cir. 107 J 3

4 MAIZE WITH SILKS MATURING BEFORE THE TASSELS.

Euclilaena seed from Florida, Costa Rica, and one of unknown origin
imported from London produced plants that were all proterogynous.
The season was too short, however, for any but the Durango plants
to mature seed, and the unusual conditions may have caused the
plants to behave abnormally.

It is known that hybrids between maize and teosinte are common
in the region about Durango. In fact, seeds that were obviously
hybrids between the two species were found in the lot from which our
plants were grown. It is therefore possible that although the plants
showed nearly all the characters peculiar to Euclilaena, except that
they were less branched, they may still be very dilute hybrids with
maize, thus accounting for their proterandrous habit. The pro-
terandrous habit in the Durango variety of teosinte is evidently asso-
ciated with the development of a well-defined mam stalk, or culm,
a character which it shares with maize and by which it differs from
the other types of teosinte.

Aside from the question whether teosinte is naturally proterogy-
nous or not, there is evidence for the belief that the proterandrous
habit of maize is the result of the separation of the sexes into different
parts of the plant and that the perfect-flowered ancestors of maize
were proterogynous.

It is a very common occurrence in maize for pistillate flowers to
be developed in the staminate inflorescence as perfect flowers or
as scattered pistillate spikelets, or, more often still, for a portion of
the inflorescences to be completely pistillate. An example of the
frequency with which pistillate flowers occur in the terminal inflores-
cence of suckers may be seen in the classification of suckers given
on page 10. Another abnormality in maize of even more frequent
occurrence than pistillate flowers in the staminate inflorescence is
for the pistillate inflorescence, or ear, to terminate in a staminate
spike.

Whenever male and female flowers occur in the same inflores-
cence of a maize plant as a result of either of these common abnor-
malities it has always been observed that the silks are developed
some days before the male flowers open. If these abnormalities are
looked upon as reversions to a perfect-flowered ancestor it seems
reasonable to assume that this primitive ancestor was proterogynous.

ADVANTAGES OF PEOTEROGYNY.

Selection for earliness and the crusade against the so-called barren
stalks have without doubt operated to change the condition of pro-
terandry that obtains in the unimproved varieties of maize to one of
approximate synacme, where pollen and silks appear simultaneously.
Synacme is certainly less desirable than proterandry, but now that

[Cir. 107]

MAIZE WITH SILKS MATURING BEFORE THE TASSELS. 5

we have unwittingly progressed thus far may it not be possible to
proceed to the still more desirable condition of proterogyny %

Self-pollination, while less desirable, is at the same time less
precarious than cross-pollination. It seems to be an inefficient
arrangement to have the opportunity for the deleterious (but rela-
tively more certain) self-pollination precede the opportunity for
the advantageous (but more precarious) cross-pollination. If the
conditions could be reversed by having the plants moderately pro-
terogynous, the proportion of cross-pollinated seed would be greatly
increased, while if by any chance the silks failed to receive foreign
pollen, pollen from the same plant would be forthcoming as a last
resort, and the danger of poorly filled ears would be avoided.

The advantages of proterogyny in maize appear so important
that the effort has been made to develop this habit in some of the
improved varieties. Occasional proterogynous plants are to be
found in most varieties, and wherever such plants have been observed
in our experiments seed has been saved and planted the following year.

In many instances it was found, however, that the proterogyny
was only apparent, resulting from an abnormal development of the
staminate flowers, and many of the plants noted as proterogynous
proved to be sterile as far as the production of pollen was concerned.
As with other characters that result from injury or abnormal condi-
tions, proterogynous variations of this nature have been found
to be inherited very imperfectly, if at all, and little progress has
been made by this method. The discovery of a variety that appears
to be regularly proterogynous makes the outlook for improvement
in this direction much more hopeful.

Two ears of a red pop corn, purchased at Granada, Spain, by Mr.
Waiter T. Swingle, of this Bureau, in January, 1912, and planted at
Lanham, Md., the following spring, produced plants, nearly every one
of which was proterogynous.

The ears of this variety are so small that the variety will probably
be of no economic importance except perhaps as a pop corn. The
subsequent behavior of the variety can not be safely inferred from
its behavior the first year, but that the proterogynous tendency exists
to a marked degree seems certain, and to attempt its transfer to our
improved varieties would seem well worth the effort. The present
account is published in the belief that those interested in developing
new types of maize will find valuable breeding material in this strain.
Only a small quantity is on hand, but a few seeds can be supplied to
those wishing to make a study of this variation and to assist in deter-
mining its economic importance.

The improvement of maize varieties should be looked upon as a local
problem. The acclimatization of an improved variety is likely to
entail nearly as much time and labor as the original work of improve-

[Clr. 107]

MAIZE WITH SILKS MATURING BEFORE THE TASSELS.

ment. Before the value of a variety possessing novel characteristics
can be determined it should be tried in a great variety of different
conditions and in combination with different varieties.

DESCRIPTION OF THE VARIETY.

The two original ears secured by Mr. Swingle, though apparently
belonging to the same general type, showed obvious differences.

Ear No. 1 (fig. 1) was 12 cm. long, 13
cm. in circumference, rather abruptly
tapering, and with 20 somewhat ir-
regular rows of small, rounded grains.
The pericarp was wine colored or gar-
net. A few of the seeds of this ear
had a colored aleurone visible beneath
the colored pericarp. The stigmas were
persistent after the manner of the
"rice" varieties, but the seeds were
not beaked. The cob was a dirty red.
Ear No. 2 (fig. 2) was 16 cm. long,
10.5 cm. in circumference, more nearly
cylindrical, and with 16 regular rows.
The pericarp was much lighter than in
ear No. 1, almost a brick red. The
stigmatic point was apparent, but no
portion of the stigma remained at-
tached to the seed. The cob was col-
ored, but was of a lighter shade than
that of ear No. 1.

Ears Nos. 1 and 2 have been num-
bered 34426 and 34427, respectively, in
the seed and plant introduction series of
the Department of Agriculture.

Seeds from each of the ears were
planted in separate rows and 62 plants
grew to maturity. The plants were
strong, vigorous, and unusually uniform
(fig. 3). Aside from the proterogynous
habit, the most striking characteristic
of the type was the dense clustering of
the spikelets toward the end of the
branches of the tassel. To make pos-
sible a quantitative statement of this
tendency, the uppermost branch of the
tassel of each plant was divided at the middle and the spikelets
of the two halves counted. It was found that on the average

[Cir. 107]

5B*ffi«!^«a

Fig. 1.— Proterogynous variety of maize.
Ear No. 1. (Natural size.)

MAIZE V/ITH SILKS MATURING BEFORE THE TASSELS.

the distal half of the branch bore 2.13
times as many spikelets as the prox-
imal half.

It was in connection with this va-
riety that tins character was first
noted, and consequently the data for
making comparisons with other varie-
ties are meager. Measurements of 50
plants taken at random from a neigh-
boring field of Boone County White
corn showed the distal portion to
have only 1.33 times as many spike-
lets as the proximal portion. In other
varieties the distribution of the spike-
lets is still more nearly uniform.

The increase in the number of the
spikelets toward the end of the branches
is brought about by a change in the
arrangement. In the basal portion of
the branches the arrangement is normal,
that is, the spikelets are arranged alter-
nately in pairs, each of which consists
of a stalked and a sessile spikelet.
Toward the end of the branches all the
spikelets are sessile and the number in
each group increases from two to three
or more.

Another peculiarity, perhaps asso-
ciated with the proterogynous habit,
is that the silks in most of the plants,
instead of appearing first on the upper-
most ear, appear almost simultaneously
from three ears, and in a number
of plants the silks of the second
ear showed before those of the first
ear.

Daily notes were taken on the indi-
vidual plants, and the flowering time
of both male and female flowers was
noted. Of 59 plants, 8 had sterile
tassels and failed to produce pollen.
Of the remaining 51 plants, in 1 the
anthers opened two days before the

[Cir. 107]

*J*

1 ft v^ % s V

Qc%

Or?

4* Ft

Fig. 2.— Proterogynous variety of maize.
Ear No. 2. (Natural size.)

8

MAIZE WITH SILKS MATURING BEFORE THE TASSELS.

silks emerged, in 9 the silks appeared on the day the first
pollen was shed, and in 41 silks were produced from 1 to 7

Fig. 3. — Plant of a proterogynous variety of maize.

days before pollen fell. The average number of days that elapsed
after the silks emerged before pollen was shed was 2.1. The rela-
tor. 107]

MAIZE WITH SILKS MATURING BEFORE THE TASSELS.

9

tive frequency of the different degrees of proterogyny is shown
below :

Number of days after the silks ap-
peared before pollen was shed — 2 —1

Number of plants 1 9

12 3 4 5 6 7
7 13 13 5 2 1

Of the 29 plants grown from ear No. 1, the pericarps of 9 were
white, the remaining 20 being colored as in the parent ear. Of the
33 plants grown from ear No. 2, the pericarps of 11 were white, the
remaining 22 being a red of the same shade as the parent ear and
consistently lighter than the progeny of ear No. 1.

The measurements shown in Table I are given with the idea of
providing data for determining the effects of acclimatization. The
size of a plant and of its various organs is so directly dependent on
the nature of the environment that measurements are of little value
for descriptive purposes except as they are made comparative by
including familiar varieties that may be used as a standard. If,
however, the expression of characters has been seriously disturbed
by the change of locality, the number of the different parts as well
as the actual dimensions may be expected to change in subsequent
generations, and the data here presented may serve as a rough indi-
cation of such comparative relations. Since in a number of char-
acters the differences appear to be significant and not the result of
chance variation, the average measurements of the plants grown
from each of the two ears as well as the combined averages are
recorded separately.

Table I. — Measurements of plants grown, from, two ears of red pop com from Spain.

Character.

Height centimeters .

Number of nodes above the ground

Number of nodes alio ve the ear

Numl >er of green leaves at silking

Length of longest leaf centimeters.

Width of longest leaf do.

Number of nodes above longest leaf

Distance from top of last leaf sheath to the lowest tassel

branch centimel srs - .

Distance from first to last tassel branch do. . . .

Distance from last tassel I iranch to tip of tassel do

Number of tassel branches

Number of secondary 1 irancb.es in tassel

Length of uppermost tassel branch centimeters. .

Number of spikelets on uppermost tassel branch

Length of staminate spikelets millimeters. .

Number of ears per plant

Number of rows in upper ear

Number of rows in second ear

Length of upper ear centimeters. .

Number of suckers per plant

Ear No. 1.

249 ±3.55

15.4 ± .12
4.93± .08

14.5 ± .16
99.8 ±1.14
10.1 ± .13
10 ± .16

3.55± .38

13.8 ± .29

29.9 ± .48

21.6 ± .03
6 ± .39

9. (if

2.5

18.3

19.4

19.3

1.0

.1(1
.08
.19
.42
.46
.19

Ear No. 2.

226.5 ±2.57

14.3 ±
4.26±

13.8 ±

94.6 ±

9.6 ±

9.2 ±

.12

.1

,09

.88
09
13

3.2
15

27.9
26.6
• 7.4
18.2
82.8

±

±
±
±
±
±
±

9.78± .16

2.8 ± .14

15.8 ±

17.3 ±

18 ±

2.1 ±

.31
.31
. 30
.76
.48
.17
. 32

.10

.27
.37
.18

Combined

average.

238.5 ±2.34

14.8

4.6

14. 1

97.4

9.8

9.6

9. 73 ±

2.7 ±
16.2 ±
18.8 ±
18. 6 ±

1.8 ±

.07
.07
.09
.75

.07
.11

3.4 ± .25
14.2 ± .22
29 ± .30
23.9 ± .59

6.5 ± .30

07
13
13
24
23
10

[Cir. 107]

10 MAIZE WITH SILKS MATURING BEFORE THE TASSELS.

ABNORMALITIES.

The nature and relationships of a variety are often best determined
by a study of the abnormalities which the plants exhibit. Though
usually more pronounced the first year that a variety is grown in a
new locality, the relative frequency of the different abnormalities
is often significant.

The following abnormalities were observed in the plants grown
from both ears: Husks coalesced, 2; fascicled spikelets in tassel, 3;
lobed husks, 5; ears with staminate flowers at tip, 4; fasciated (or
"bear's-foot") ears, 9; ears exserted beyond husks, 4.

In addition to these common abnormalities that occur with varying
frequency in nearly all varieties, single examples of the following were
discovered: A leaf like bract in the tassel; a small ear inclosed in
husks at the base of the tassel; perfect flowers in the ear (in this plant
three well-defined stamens were found at the base of nearly eveiy
seed — an abnormality which may have been overlooked in other
plants, since the presence of less well-developed stamens could have
been determined only by carefully dissecting each ear); two seeds
grown together in the ear; 1 and a secondary ear borne directly in the
axil of the prophyllum instead of the axil of a husk, a phenomenon
of rare occurrence in maize, except in the hairy Mexican type (Zea
hirta), in which it is normal.

The above abnormalities were confined to 21 plants. In the re-
maining 38 plants no abnormalities were observed.

Great diversity was exhibited in the suckers, a characteristic of

unimproved types, especially when grown in a new locality. The

series was so extensive that an attempt was made to classify the

different types, the result of which is shown in Table II, together

with the relative frequency of the different types. These variations

might be classed as abnormalities, but they all represent intermediate

stages between the two normal types of branches, viz, the pistillate

inflorescence (or ears) and normal suckers that are replicas of the

main stalk.

Table II. — Classification of suckers.

No.

Description of sucker.

Number
of plants.

Resembling main stalk, ear and tassel normal

Similar to No. 1, with no ear developed

Terminal infloresence staminate, with tips of central spike and branches aborted; no ear. .

Similar to No. 3, with ear in normal position

Terminal inflorescence staminate, with scattered pistillate flowers on both central spike
and branches; no ear

Similar to No. 5, with ear in normal position

Terminal inflorescence staminate and normal except for one pistillate branch; no ear

Terminal inflorescence staminate and normal except for a pistillate portion at tip of cen-
tral spike; no ear

Similar to No. 8, but with ear in normal position

Terminal inflorescence staminate except for a pistillate portion at the base of both central
spike and branches; no ear

17

20

4

1

1
1
1

11
1

1 This abnormality has been described and its significance discussed by Mr. J. H. Kempton in " Two-
Flowered Female Spikelets in Maize," Bulletin 278 of the Bureau of Plant Industry.

[Cir. 107]

MAIZE WITH SILKS MATURING BEFORE THE TASSELS. 11

Table II. — Classification of suckers — Continued.

No.

n

12
13

14

15
16

18
19

20

Description of sucker.

Terminal inflorescence staminate except for a pistillate portion at base of central spike and

tips of branches; no ear - - ;---.;---:- / - - ■

Terminal inflorescence staminate, with pist Mate portion at base of central spike; branches

wanting; no ear - .•--.----; :----.;--■

Terminal inflorescence with pistillate portions at the base and tip of central spike;

branches staminate; no ear

Terminal inflorescence with central spike completely pistillate; branches stammate, or

nearly so; no ear -

Similar" to No. 14 except for a staminate portion at tip of central spike

Terminal inflorescence with central spike completely pistillate; branches staminate, with

pistillate portion at tip; no ear

Terminal inflorescence with central spike completely pistillate: branches pistillate, or

nearly so; no ear

Terminal inflorescence pistillate; no branches; no ear

Terminal inflorescence a normal staminate tassel, with an ear at its base; no ear in normal

position

No terminal inflorescence; no ear

Number
of plants.

The suckers which resembled main stalks, producing; both ears and
tassels, were also for the most part proterogynous, though less uni-
formly so than the main stalk. Notes were secured on the suckers of
15 such plants. Of these, two were proterandrous, one was synac-
mic, two had one sucker proterogynous and one proterandrous, four
had sterile tassels, and eight were proterogynous. In every plant
with sterile tassels on the sucker the tassel of the main stalk was
also sterile. There were apparently no intermediate stages between
completely sterile tassels and those producing an abundance of pol-
len. Sterile tassels are not uncommon in many varieties of maize
which show no proterogynous tendencies.

SUMMARY.

The present paper records the discovery of the proterogynous
habit in a variety of maize introduced from Granada, Spain. In
plants of this type the silks are exserted and receptive before the
pollen begins to fall. This proterogynous characteristic insures a
much larger percentage of cross-pollinated seed than is obtained in
the ordinary varieties, in which the falling of the pollen is simulta-
neous with or precedes the exsertion of the silks. If this character can
be combined with the good qualities of improved American types it
will obviate the necessity of detasseling to secure cross-pollinated seed.

A small quantity of pure seed of this variety was produced during
the past season and is available for those wishing to undertake the
introduction of this character into local varieties.

Approved :

James Wilson,

Secretary of Agriculture.

Washington, D. C, November 21 , 1912.

[Cir. 107]

o

Issued April 5, 1913.

U. S. DEPARTMENT OF AGRICULTURE.

BUREAU OF PLANT INDUSTRY— Circular No. 108.
B. T. GALLOWAY, Chief of Bureau.

THE CHINESE WOOD-OIL TREE.

BY

DAVID FAIRCHILD.

Agricultural Explorer in Charge of Foreign Seed
and Plant Introduction.

WASHINGTON : GOVERNMENT PRINTING OFFICE : 1913

[Cir. 108

2

BUREAU OF PLANT INDUSTRY.

Chief of Bureau, Beverly T. Galloway.
Assistant Chief of Bureau, William A. Tayl
Editor, J. E. Rockwell.
Chief Clerk, James E. Jones.

Cir. 108, Bureau of Plant Industry, U. S. Dept. of Agriculture.

Plate I.

■ l .!

An Old Wood-Oil Tree 30 Feet High in Full Flower in Western Szechwan,
China, on the Banks of the Yangtze River.

(Photographed by Mr. E. II. Wilson, Arnold Arboretum Expedition, 1908.)

B. P. I.— 802.

THE CHINESE WOOD-OIL TREE.

INTRODUCTION.

The Chinese wood-oil or tung-oil tree (Aleurites fordii) is grown on
the heavy-clay hillsides and waste places along the Yangtze River
above Hankow, where the rainfall is heavy and the climate is similar
to that of some of our Southern States, although the temperature
does not go so low in winter. The tree is probably not very long-
lived and would be comparable in this respect to the silver maple. It
drops its leaves in winter and does not wake up early in the spring,
like many trees, and therefore is not likely to be severely injured by
late frosts.

The flowers come out before the leaves. They are fully as large as
catalpa flowers, and the tree in bloom is a very pretty sight. As an
ornamental the wood-oil tree (Pis. I and II) is likely to prove about
as desirable as the catalpa, but the soft wood is of little value and,
like many other soft-wooded trees, the branches break off easily in
heavy winds.

The Chinese wood-oil tree commences to bear fruit when 4 or 5
years old. The fruits are the size of small apples — from 2 to 3 inches in
diameter. (PI. III.) They contain from two to eight (most commonly
five) large, oily seeds, that are reported to be poisonous and should
not be eaten. They at least have a purgative effect, similar to that
of the castor bean, to which the wood-oil tree is botanically distantly
related. The yield of these nuts in China is reported to be from 30
to 75 pounds per tree.

ECONOMIC VALUE OF WOOD OR TUNG OIL.

The value of this Chinese tree lies in the fact that the nuts contain
one of the best drying oils known, called wood or tung oil. In recent
years this oil has produced, it is reported, a revolutionary effect on
the varnish industry of the United States. It has largely taken the
place of kauri gum and has made possible the manufacture of a
quicker drying varnish, which is less liable to crack than that made
from kauri gum, and has been found of special value in waterproof
priming for cement.

72160°— Cir. 108—13 3

4 THE CHINESE WOOD-OIL TREE.

This valuable oil constitutes about 23.9 per cent of the substance
of the nuts, and its market price, which is normally 6 to 7 cents per
pound, has risen recently to 12 cents. If figured out on the basis
of 1 pound of the applelike fruits yielding 0.58 pound of nuts and
0.138 pound of oil, a pound of fresh fruits would be worth from 0.82
cent to 1.6 cents, and a bushel, which would weigh approximately
27 pounds, would be worth from 22 to 43 cents, depending on the
market price of the oil. For comparison, 65 cents a bushel is
considered a fair price for apples.

Not more than 108 trees (i. e., 20 by 20 feet apart) should be
planted to the acre, and this would mean a possible gross yearly
return of from $23 to $46, depending on the market price of the oil
figured on the above basis. One 8-year-old tree belonging to Mr.
W. H. Raynes, of Tallahassee, Fla., bore this year 852 fruits, prac-
tically 2 bushels, which would make the gross returns from $46 to
$92 an acre, provided that all the trees in an orchard did as well as

this one.

CULTIVATION.

In the neighborhood of Tallahassee, Fla., land suitable for the
cultivation of the Chinese wood-oil tree is selling for $10 to $15 an
acre, and land on which it will grow can probably be had for much
less. The cost of planting, cultivation, and marketing would prob-
ably not exceed $15 an acre yearly, so that as a tree crop this wood-oil
tree is worthy of consideration by the owners of cheap lands in the
South.

Experiments might be made in blasting holes for the reception of
the trees where soil conditions render digging difficult in the ordinary
way. Since the trees are rapid growers, it is also probable that it
will be satisfactory if the soil is kept free of weeds for a space of 4
feet on each side of the row and the center space between the rows
left in sod. After an orchard is well established it may not require
any cultivation, although possibly the increased crops would more
than pay for the cost of clean culture.

PRODUCTION OF WOOD OIL.

The work of gathering these large fruits will not, in all probability,
approach the cost of gathering apples, which is commonly 5 cents a
bushel. The nuts are removed from the husks very readily after the
fruits have been stacked in heaps, and the process of crushing and
extracting the oil is not likely to be more expensive than the ex-
traction of cottonseed oil. It is also probable that the same or
similar machinery can be used for the expression of wood oil.

In China, according to reports, after the fruits have been gathered
and the husks removed, the nuts are put into a large iron pan about

[Cir. 108]

Cir. 108, Bureau of Plant Industry, U. S. Dept. of Agriculture.

Plate II.

Wood-Oil Tree in Full Bloom, near Tallahassee, Fla., on the Estate of Mr.

William H. Raynes.

This tree was imported as a seed from Hankow, China, by Mr. L. S. Wilcox in 1905. It bore
410 fruits in mil and 852 fruits in 1912, approximating 1 and '2 bushels, respectively. It lias
not been injured by a temperature of 14° F.

Cir. 108, Bureau of Plant Industry, U. S. Dept. of Agriculture.

Plate III.

CO

+3

CO

ta

2

'53

-

•"*■

X

>,

O

rQ

_i

t- _;

o ~

DO

t- .—

co

~ "

z

7*r-

*:

i-S

a.

s ^

O

O at

I

^.^2

1*

CO

C°

<

t. CH

ci~*

cr

;. '"■

2

K a

u_

S-

O

Cy CD

J= >

LU

-w ^

r-

_, CD

<

X ~

r-

co

LU

M -X3

io

UJ

I

i'd

1-

z

*=■ si

o

o

z

f &

5

%w

o

^H

rr

£fl

a

0) 0)

„

/— _£;

UJ

+j +-»

u

u~t *y

rr

r-^ O

H

o

"t^ £

z

o o

3

-s p.

1-

.1^

tr

Bo

O

~ *-*

GO

_J

m"

o

o3

■u o

Q

mJ-

o

„&*

o

c

£

o •
"T'o

LU

"'SO

co

UJ

03 S

z

S -^

s-. .

I

2 S

o

^ •—

<

0/ C

u_

.■z —

o

C3 CD

43 x

r- C

CO

o &

h

Sp

z>

»: O

£E

.ti °

u.

p

z

f— H

UJ

CD

Ul

X

rr

O

p

THE CHINESE WOOD-OIL TEEE. 5

18 inches in diameter and are stirred over a fire until parched. The
seeds are then ground into a fine meal, which is heated or steamed
before it is put in the press, supposedly for the purpose of assisting in
the extraction of the oil.

POSSIBILITIES OF THE WOOD-OIL INDUSTRY.

In starting an oriental industry in America the most important
factor to be considered is the amount of hand labor involved. There
does not appear to be much involved in this industry, as the gathering
and husking of the fruits seem to be the only hand work required.

The freight haul from China to America is a water haul practically
all of the way, but instead of its acting as a tariff wall to protect the
grower in this country it apparently acts as a handicap, at least at
the beginning of such an industry. According to recent freight
rates a grower in Montgomery, Ala., routing his product to Mobile
by rail and thence by water to New York, would have to pay 86 cents
per 100 pounds, whereas his Chinese competitor at Hankow pays
only 61.5 cents per 100 pounds to ship around the world to New
York. However, it is not to be expected that these rates would be
maintained in case the supply of oil became an important article of
commerce. There enters into this discussion, however, the factor of
a home supply of the material, and undoubtedly this is very desirable.

The American farmer has the advantage over the Chinaman of
cheap, accessible lands and team labor. Since the hand labor
involved in a well-planned orchard is not great, it would seem to be
entirely possible by the systematizing of such an industry on large
plantations to produce this wood oil more cheaply than it is now
produced by the wayside plantings in China, which must be very
wasteful of human labor. This labor factor in China is now becom-
ing an important one, as the cost of labor is rapidly rising. It seems
reasonable, therefore, to suppose that the American extensive method
of handling such a tree crop would in time overtake and outstrip the
back-yard and wayside methods of the Chinese.

The prospects are that there will be a continual and growing demand
for wood oil. Five million gallons were imported from China last
year. The growing use of soy-bean oil, it is reported, will tend to
increase rather than decrease the consumption of wood oil, as soy-
bean oil dries too slowly and requires an addition of wood oil to help
it dry. The home demand in China is likely to increase and the
opinion of importers seems to be that the American-grown oil could
capture the market. If it does, 40,000 acres of trees would be required
to supply the present demand.

There now remains to be considered the very important question
whether the American-grown trees will produce as good a product as

|Cir. 1081

6 THE CHINESE WOOD-OIL TREE.

the Chinese trees. This problem still remains to be determined, but
will doubtless be settled in the course of a few months. There seems
to be no basis for the theory, however, that the oil produced here
by the same species of tree will be materially different from that
produced in China. Whatever differences have already been observed
in the small samples which have been expressed from the nuts pro-
duced in Florida are attributed by Dr. Rodney H. True, of the Office
of Drug-Plant, Poisonous-Plant, Physiological, and Fermentation
Investigations, to the method of extraction of the oil rather than to
any specific difference in its composition.

SUMMARY.

To sum up the whole wood-oil situation as it stands to-day in the
minds of investigators of the Department of Agriculture:

An important plant industry, involving a present area of 40,000
acres and a possible one of several times this acreage, is here repre-
sented.

Chinese wood-oil trees have grown and fruited well in South Caro-
lina, Florida, Alabama, Louisiana, Mississippi, Georgia, Texas, and
California. These trees are growing on cheap land and do not require
very careful attention. The tree has stood a temperature as low as
4° F., at Clemson College, S. C, without injury, except the loss of a
few of the small lateral branches, and is slow to start into growth
even when subjected to a temperature of 80° F. It is therefore not
so liable to be injured when this temperature is immediately fol-
lowed by a drop to 18° F. There are large regions in the South where
the temperature for decades at a time does not go below zero Fahren-
heit, and it is in these that it should be tried. Until it is known just
how low a temperature the wood-oil tree will stand without injury
it is not safe to predict the northern limit of its probable cultivation.

The distribution of several thousand wood-oil trees through the
South in 1906 and 1907 has brought in a considerable amount of infor-
mation as to the behavior of the tree in this country. From these
data it appears that the tree has done best in the more moist parts
of the Gulf coast region, on deep loam soils which are underlain with
stiff clay. The sticky gumbo soils of eastern Texas seem unfavorable
to its growth, and it has not done well on the almost pure sand soils
of Florida.

The wood-oil tree quickly recovers from frost injury, sprouting
up again from the stump. Being of rapid growth, it should recover
more quickly than an orange tree. The indications are that the
blooms may be occasionally caught by late frosts, as they open in
March in the latitude of Tallahassee, Fla., but in this respect they
seem to be less liable to injury than pear and peach trees.

[Cir. 108]

THE CHINESE WOOD-OIL TEEE. 7

The Department of Agriculture will have on hand one year from
now for distribution to bona fide experimenters a limited number of
1-year-old trees, and these will be ready to send out in February and
March. Applicants for them must have the necessary conditions of
soil, temperature, and rainfall and the actual intention to take up
the serious study of this industry before their requests for plants can
be granted. Experiments with single trees have been made, and
what is now desired is the creation of acre plantations in the hands
of private individuals.

Approved :

James Wilson,

Secretary of Agriculture.

Washington, D. C, November 29, 1912.

Cir. 108J

ADDITIONAL COPIES of this publication
-£*- may be procured from the Superintend-
ent of Documents, Government Printing
Office, Washington, D. C, at 5 cents per copy

U. S. DEPARTMENT OF AGRICULTURE.

BUREAU OF PLANT INDUSTRY— Circular No. 109.
B. T. GALLOWAY, Chief of Bureau.

MISCELLANE( )US PAPERS.

United States Official Cotton Grades N. A. COBB

Potato Leaf-Roll • . W. A. ORTON

Morphology of Cotton Branches 0. F. COOK

The Wilting Coefficient for Plants in Alkali Soils T. H. KEARNEY

Interpreting the Variation of Plat Yields F. D. FARRELL

RK

Issued January 4, 1913.

WASHINGTON:

GOVERNMENT PRINTING OFFICE.

1913.

BUREAU OF PLANT INDUSTRY.

Chief of Bureau, Beverly T. Galloway.
Assistant Chief of Bureau, William A.Taylor.
Editor. J. E. Rockwell.
Chief Clerk, James E. Jones.

[Cir. 109]
2

UNITED STATES OFFICIAL COTTON GRADES. 1

By N. A. Cobb, Agricultural Technologist in Charge of Agricultural Technology and
Cotton Standardization and Grading Investigations.

AUTHORIZATION AND SALE OF STANDARD COTTON GRADES.

The appropriation bill for the Department of Agriculture for the
year ended June 30, 1909, authorized the fixing of a standard of
middling cotton and, using the same as a basis, the establishment
of nine grades of cotton, to be designated —

Middling fair. Strict good middling. Low middling.

Strict middling. Middling. Strict good ordinary.

Good middling. Strict low middling. Good ordinary.

The Secretary of Agriculture was authorized to call to his assist-
ance expert cotton classers, and he was directed to prepare copies
of the official standard in practical form for distribution at cost.

The Secretary convened in Washington a committee of prominent
cotton growers, dealers, manufacturers, and experts in February,
1909, and after due deliberation the committee submitted a unani-
mous report, together with a set of grades which they had prepared
in accordance with the wording of the appropriation act. This
report was approved by the Secretary.

To familiarize the public with the official grades, copies of them
were temporarily placed with a number of associations, exchanges, col-
leges, and other organizations for examination. Within six months
after their promulgation sets had been purchased by cotton exchanges,
cotton mills, and textile schools in 21 States, as well as in England,
Germany, and Mexico. The sales of the grades have doubled in
each succeeding year, so that at the present time they are distrib-
uted in 40 States and 10 foreign countries, and have been officially
adopted by the following cotton exchanges and associations:

New Orleans. Little Rock. New England Buyers.

Memphis. Galveston. Arkwright Club.

St. Louis. Macon. Southern Cotton Buyers.

Charleston. Mobile. Fall River Cotton Buyers.

Natchez. Oklahoma.

i Issued Jan. 4, 1913.
[Cir. 109] 3

CIRCULAR NO. 109, BUREAU OF PLANT INDUSTRY.

The grades are in use in a number of other associations that have not
gone through the formality of adopting them by vote. The accom-
panying chart indicates the distribution of the grades up to the 1st
of October, 1912.

The original set of standards was prepared by experts under the
direction of the committee referred to, and from this set the copies
for sale have been made by experts from time to time, as required.
The original cost of preparing a complete set of the grades was
$35, but within the past year it has been reduced to $25. A set

DISTRIBUTION

Official Cotton Gmdes

of the

US'Depdriment
ofAcriculfure

The United States
Official Cotton Grades have
been on sale for two
years and one month.
The present price is §25
per set of nine grades.

The chart does not shon the distribution abroad.
The Grades have been sold in Great Britain, Ger-
many. France, Italv. Belgium, Japan. India, and Mexico.
The Official Grades are prepared and issued by the
Secretary of Agriculture in accordance with law.
The Department has no power to dispose of the
Official Grades except by sale.

of the grades consists of 9 boxes, each containing 12 samples, sepa-
rately packed, which show the range of diversity within the grade.
In the top of each box is a photograph showing the appearance of
the cotton when certified by the Secretary. Fractional sets are
sold practically pro rata.

PERMANENCY AND VALUE OF STANDARDIZATION.

One of the problems which confronted the department at the
beginning was the preservation of the grades as originally prepared
by the committee, and numerous experiments were made to deter-
mine the best plan to be followed. The experiments indicated that

LCir. 109]

UNITED STATES OFFICIAL COTTON GRADES. 5

by the use of large vacuum tubes the cotton types could be kept
without appreciable change for an indefinite period of time. Fifty
sets of the grades have been put up in vacuum tubes and stored for
future use. Each separate sample in each grade was placed in a
vacuum tube, making 12 tubes for each grade, or 108 for a complete
set. The cotton is first wrapped in chemically pure wliite paper and
then in pure black paper, before being placed in the tube, and in the
top of the tube, after the insertion of the wrapped cotton, a 2-inch
layer of specially prepared asbestos is placed. The air is then
exhausted and the tube sealed by fusing off. No such system has
ever before been put into effect in connection with cotton grades or
any other graded agricultural product. This is probably one of the
largest undertakings of the kind ever attempted.

From time to time one of these vacuum sets will be broken open
and replaced in grade boxes, each sample having been specially pre-
pared and labeled with that end in view. In this way absolutely
exact copies of the original types will be available from time to time
for the next 100 years or more, and cotton grading will, for the first
time, be freed from the drawback of depending solely on the memory
of experts.

A campaign to educate the producers and local buyers in the use
of the grades has been carried on through the agents in charge of
the Farmers' Cooperative Demonstration Work. Several of the
agricultural colleges are now teaching cotton grading at their summer
schools, as well as during the regular term, and are using the official
grades as the basis of instruction. With the return of the students
to the farm a demand for the grades for use on the farm is bound to
grow. With the increasing quantity of cotton which is now sold by
the grower directly to the mill, the value of the grades is increasing,
all branches of the industry, producer, merchant, and manufacturer
alike, using a uniform standard whose correctness and constancy are
guaranteed by the Government.

When the original set of types was prepared by the committee
appointed for the purpose, the aim was to produce a set which would
be representative of American white cotton as a whole. No atten-
tion was paid to the source from which the cotton came, and this
policy has been continued in the preparation and sale of copies of
the original types. The committee had before it an exceedingly
great variety of American cotton. Not only did the department
secure, through the officials of about 30 American cotton exchanges,
grade types in accordance with the ideas of those exchanges and place
these entirely at the disposal of the committee, but members of the
committee were invited to bring with them cotton winch they con-
sidered would be suitable for the preparation of the official grades.

[Cir. 109]

6 CIRCULAR NO. 109, BUREAU OF PLANT INDUSTRY.

A number of members provided cotton in accordance with tins invi-
tation. It is safe to say that the committee's attempt to make a set
of grades that would apply to the whole of the American cotton belt
was so sincere and so adequately backed up by suitable material
from which to prepare the grades that the result was all that could
possibly have been expected.

During the past few months an effort has been made to ascertain
what proportion of the 1912 crop is classifiable on the basis of the
present official grades. From a collection of many hundreds of sam-
ples taken at random from various parts of the cotton belt, very
little cotton has been secured which was not readily classifiable on
the basis of the present official grades.

There is a tendency toward the increased use of cleaning machines
as adjuncts to ginning machinery. . These cleaning machines remove
trash and dirt and thus raise the grade of the cotton, so that in the
future we may expect more and more cotton to fall into the higher
grades, where there is less difference between the cottons from various
parts of the country. This tendency toward cleaner cotton will
doubtless facilitate the use of the official grades.

It is important to keep in mind that many people, in the practice
of grading cotton, are influenced, consciously or unconsciously, by
other factors than the color and the amount of trash. Where such a
practice prevails, a sample of cotton that on the basis of color and
trash would pass as official middling, may pass as of a higher grade,
if the staple is unusually good. It is believed that such grading can
result only in confusion and that length of staple should be entirely
dissociated from the official grades.

FURTHER WORK IN CONTEMPLATION.

The officers of the department are carefully investigating the
various problems connected with the question of cotton standardiza-
tion, and whenever the time seems ripe for further steps toward the
establishment of universal cotton grades the department will be
equipped and ready to give all the assistance possible.

If the plan of establishing standards for the grading of American
cotton is to be brought to a successful issue, it is of the utmost
importance that the foundations upon which these standards rest
shall be secure. The firmest foundation possible will be that which
is in thorough accord with present practices, and the wish of the
officers of the department is that no mistakes be made in this funda-
mental work. If a foundation can be laid which meets with univer-
sal approval there is not the slightest doubt that there can be built
upon it a system of cotton standardization which will be of great
service throughout the cotton industry.

[Cir. 109]

POTATO LEAF-ROLL. 1

By W. A. Orton, Pathologist in Charge of Cotton and Truck Disease and Sugar-Plant

Investigations.

INTRODUCTION.

The years 1911 and 1912 have been marked by the prevalence of
potato troubles in some of our Western States more serious than any
hitherto experienced. Such heavy losses have resulted, particularly
in eastern Colorado and western Nebraska, that the production of
potatoes, generally one of the most profitable crops for these irrigated
districts, has been rendered so uncertain that the growers have been
compelled to greatly curtail their acreage.

The Bureau of Plant Industry has nearly ready for publication a
bulletin on these new diseases, but as this will necessarily be some-
what technical in character and not available for distribution until
later in the season, this paper is issued for the information of the
potato growers in the districts affected.

The identification of the disease responsible for the decrease in
potato production in the country east of the Rocky Mountains and
west of the one hundredth meridian has been in doubt until recently.
It was at first thought to be an aggravated form of diseases already
known, Rhizoctonia, stem-blight, and Fusarium wilt, but it is now
considered a new and distinct malady called "leaf- roll." This has
been common in Europe since 1905, but has not hitherto been iden-
tified in the United States.

DESCRIPTION OF LEAF-ROLL.

As indicated by the name, leaf-roll is marked by an upward rolling
of the leaves on then- midrib. There is usually a pronounced change
in the color of the foliage to a yellow, unhealthy shade, often tinged
reddish or purplish.

The disease may start early in summer and be far advanced by
the end of July. The plants do not die quickly, as they do when
attacked by Fusarium wilt, but may live nearly as long as healthy
oil es. The growth is checked and the formation of tubers prevented.
Often no potatoes are set, or only small ones clustered around the

i Issued Jan. 4, 1913.
[Cir. 109] 7

S CIRCULAR NO. 109, BUREAU OF PLANT INDUSTRY.

base of the stem, while numerous rudimentary tubers are formed on
the stolons.

The browning of the woody part of the potato stem and the pres-
ence of a brown discolored ring at the stem end of the tubers is not
so much a character of the leaf-roll, but is rather to be taken as an
indication of the presence of another disease, the Fusarium wilt.

The formation of aerial tubers is sometimes a feature of leaf -roll,
but in other cases is the result of stem cankers caused by the fungus
Rhizoctonia.

Leaf-roll is considered to be hereditary through the seed potatoes;
that is, if potatoes borne on plants affected by leaf-roll are planted
the resulting crop will be diseased and usually much worse than the
first crop.

The cause of leaf-roll remains unknown, though it has been pre-
valent in Europe smce 1905 and has been given much study there.
It is now believed to be a physiological disorder rather than one
caused by a parasite. Many consider it due to some unfavorable
soil or climatic condition, but no one has been able to show what
conditions produce it or how it may be controlled by any cultural

practices.

LEAF-ROLL AN UNSOLVED PROBLEM.

No fully satisfactory remedy for leaf-roll has been discovered. It
presents one of our most serious problems for investigation and one
which it is hoped to push actively as soon as means are provided.
We have, however, the benefit of seven years of German experience
with the same trouble. The anxiety caused by its appearance in
Germany has been somewhat allayed with the passage of time, and
the best authority on potato diseases there even states that through
the awakening of interest in better culture and in improvement of
seed the leaf-roll will prove in the end a benefit to German agricul-
turists and their potato production will be permanently increased.

CONTROL MEASURES.

While we can not recommend any preventive treatment with the
confident assurance that it will be successful, there are some points
of attack that are strongly to be recommended as having given the
best results elsewhere and as being common-sense measures whether
disease is present or not. Most important of these are good seed,
crop rotation, and improved culture.

The seed problem takes first place in any movement for the better-
ment of our potato industry and particularly in these western dis-
tricts, where diseases are extensively carried on seed. While there
may be occasional apparent exceptions, it is the general rule that
seed from fields affected by leaf -roll will give a diseased crop. It is

[Cir. 109]

POTATO LEAP-ROLL. 9

therefore strongly to be recommended that no seed be planted
except that known to come from healthy fields. If there are none
in the neighborhood, seed should be brought in from outside. Leaf-
roll is not known to occur in Minnesota, Wisconsin, or Michigan.

The expense of. bringing seed from distant points and the uncer-
tainty of getting a vigorous stock of the variety desired emphasize
the great need of a better organization of the potato growers for
seed selection and inspection. It should be possible to buy seed
potatoes accompanied by a certificate from a reliable authority
that they are free from disease and of the variety claimed. Such
certification should be based on a field inspection made in early
autumn, when the foliage is still alive. Leaf-roll can not be detected
by an inspection of the tubers.

In purchasing seed potatoes, those infected with Fusarium wilt
should also be avoided. Any lot where many tubers show a brown f
discolored ring when cut across the stem end should be discarded.

Crop rotation is absolutely essential to permanent potato culture.
It is a common practice to grow several successive crops in new
western land, but this always has one inevitable result — diseases are
introduced and spread until it is no longer possible to grow potatoes
with profit. Must every community and every farmer learn this
lesson separately, or will the experience of the many profit the remain-
ing few ?

There is abundant land in every irrigated district for a long rota-
tion, and particularly at this time when so few potatoes are being
planted it is possible to put them on land that has not been in this
crop for four or five years.

Cultural methods may reduce the injury from leaf -roll and other
diseases. No pains should be spared to give the potato crop the
care and attention needed for its most favorable development. Some
of the most essential points are given by Prof. L. C. Corbett in Circu-
lar No. 90 of the Bureau of Plant Industry, as follows:

PREPARATION OP LAND.

Since the potato is a deep-rooted crop and forms its tubers beneath the soil, it standa
to reason that it requires a much deeper seed bed than will be necessary for cereal
crops. In fact, the preparation for potatoes should be as deep and as thorough as for
sugar beets, whether in the irrigated or in the humid region. If a preparatory crop,
such as alfalfa or clover, is to be turned under for potatoes, it is advisable to plow this
crop under in the fall and to compact the soil sufficiently to make it a good retainer for
water, but not so smooth that it will blow. Before planting in the spring the land
should be made fine to a depth sufficient to admit of planting and cultivation. If
unusually dry the land should be irrigated before the crop is planted. If the normal
precipitation has occurred during the winter and spring months, the crop may be
planted without irrigation. It is not advisable, however, to plant the crop in dry,
hot earth and to immediately irrigate it. Irrigation should precede rather than follow
[Cir. 109]

10 CIRCULAR NO. 109, BUREAU OP PLANT INDUSTRY.

the planting. If the crop does not grow rapidly after planting, irrigation should be
provided from time to time as the appearance of the crop and the condition of the soil
would indicate ; a dark-green or blackish color shows a lack of moisture on the part of
the plants, while light-green or yellowish tints indicate the presence of too much
moisture. The plant should be kept growing at the maximum rate from the time it
appears above the ground until it has completed its season's work, and cultivation
supplemented by irrigation must be relied upon to keep the plant working.

METHOD OF IRRIGATION.

The water for irrigating potatoes is best applied in every other furrow, the furrows
being narrow and deep and the water so applied that the ground will not be saturated
above the point where the tubers are formed. This will induce the formation of a
deep instead of a superficial root system. In order to accomplish this the rows must
be sufficiently wide apart to admit of throwing up broad high ridges, with narrow deep
furrows between, in which the water can be led in a small stream for a long period
rather than by means of a large stream flowing only for a short time. The successive
irrigations should be carried on hi alternate rows; the second irrigation should be in
the rows not used by the first, and the third in the rows used during the first. Culti-
vation should follow irrigation as quickly as the condition of the soil will permit, but
as soon as the tubers have made then growth, usually about September 1, water should
be withheld, so that the soil will dry and the crop ripen in proper condition for
harvesting.

[Cir. 109]

MORPHOLOGY OF COTTON BRANCHES. 1

By O. F. Cook, Bionomist in Charge of Crop Acclimatization and Adaptation

Investigation*.

INTRODUCTION.

The cotton plant has two distinct kinds of branches, differing in
position, structure, and function. The upright "limbs," or vege-
tative branches, behave like the main stalk of the plant and do not
produce flowers or bolls. The floral buds appear on horizontal fruit-
ing branches. The specialized habits of growth are of interest from
the general standpoint of morphology, as well as in their applications
to cultural and breeding problems. The practical considerations
have been- treated in several publications of the Bureau of Plant
Industry. 2

POSITIONS AND HOMOLOGIES OF BUDS.

The positional relations of the two types of branches have seemed
to throw the clearest light upon their structural homologies. It is
easier to determine the positions of the buds from which the branches
develop than to understand the relations of the mature branches. In
addition to the bud that serves to continue the growth of the shoot,
each internode of the cotton plant produces two other buds, one in
the axil of the leaf and another to the right or left of the axil. Some
stalks are right handed and others left handed with respect to the
position of the extra-axillary buds and the branches produced by
these buds.

The axillary buds usually remain dormant, but may be developed
into vegetative branches when the conditions are favorable to luxu-
riant growth of the plants. The fruiting branch es arise from the extra-
axillary buds of the main stalk and the vegetative branches. The
floral buds are also developed in the extra-axillary position on the
fruiting branches. Thus the floral buds of the fruiting branches
have been considered homologous with the extra-axillary buds that
give rise to the fruiting branches. The axillary buds of the fruiting
branches are able to produce vegetative branches, like the axillary

1 Issued Jan. 4, 1913.

i Weevil-resisting adaptations of the cotton plant, Bulletin 88. A study of diversity in Egyptian
cotton, Bulletin 156. Dimorphic branches in tropical crop plants: Cotton, coffee, cacao, the Central Amer-
ican rubber tree, and the banana, Bulletin 198. Arrangement of parts in the cotton plant, Bulletin 222.
Dimorphic leaves of cotton and allied plants in relation to heredity, Bulletin 221. The branching habits
of Egyptian cotton, Bulletin 249.

[Clr. 109] 11

12 CIECULAR NO. 109, BUREAU OF PLANT INDUSTRY.

buds of the main stalk. The tunctional specialization, of the branches
oi the cotton plant is much less complete than in coffee and cacao,
where one form of branches is unable to regenerate the other form.

A SYMPODIAL THEORY OF THE FRUITING BRANCHES.

Another interpretation, advanced by Mr. H. M. Leake, of Cawn-
pore, India, makes the two kinds of branches appear much more
fundamentally different. According to Leake the vegetative
branches are monopodial and the fruiting branches sympodial.
This means that a vegetative branch is looked upon as a single con-
tinuous shoot, while a fruiting branch is supposed to be composed of
a succession of independent shoots, each represented by a single
internode and terminated by a flower. Thus, the floral buds of the
joints of the fruiting branches would correspond to the terminal buds
of the vegetative branches or of the main stalk. 1

The structure of the fruiting branches and the nature of the dif-
ferences between the two kinds of branches are matters of special
interest, because Leake has reported Mendelian reactions in crosses
between varieties with different habits of branching. Varieties with
many vegetative branches on the main stalk are described as mono-
podial, while the varieties with only a few vegetative branches are
called sympodial, and the two characters are supposed to reappear
in definite proportions of the hybrid progenies. These results have
been accepted by Bateson and used as one of the chief illustrations
of the practical importance of the Mendelian methods of breeding
hybrid varieties. 2

INFLUENCE OF ENVIRONMENT ON HABITS OF BRANCHING.

The same characters that have appeared to Mendelize in India
have shown in the United States a wide range of environmental
modification. Varieties that would be considered sympodial under
some conditions become altogether monopodial under other condi-
tions. When the conditions keep the plants small, only a few of the
vegetative branches are produced or none at all; but when the con-
ditions favor luxuriant growth the fruiting branches may be replaced
or transformed into vegetative branches. Even under the same con-
ditions in other respects it is possible to control the development of
the branches by different methods of thinning. Plants that are
thinned early, so as to stand far apart during the earlier stages of
growth, are likely to show a much greater development of the vege-
tative branches than plants that are left closer together. In some
varieties of cotton the number of vegetative branches may be in-

i Leake, H. M. Studies in Indian cotton. Journal of Genetics, v. 1, p. 205, 1911.
2 Bateson, W. Genetics. Popular Science Monthly, v. 79, p. 319, 1911.

[Cir. 109]

MORPHOLOGY OF COTTON BRANCHES. 13

creased by the growth of more of the axillary buds into branches,
but in most cases the additional vegetative branches are produced
from extra-axillary buds that under other conditions would have
developed into fruiting branches.

The nature of the differences between the two kinds of branches
must be understood before either the Mendelian reactions or the
environmental transformations can be appreciated. Though there is
nothing inherently improbable in the idea that the same character
may show both kinds of reactions, such cases would not be expected
under the Mendelian theory of heredity as a process of alternative
transmission of "unit characters." It would not be expected that a
character which behaved as a pure recessive in a Mendelian experi-
ment could be called forth by a change of external conditions. Yet
this possibility seems to be indicated by the variations that have
been observed in the brandling habits of cotton.

RELATIONS OF FLORAL BUDS TO STIPULES.

Though a casual examination of the cotton plant may appear to
support the theory that the fruiting branches are sympodial, there are
several facts that seem to forbid the acceptance of this view. The
relations of the floral buds to the adjacent parts of the fruiting
branches do not warrant the conclusion that the floral buds of the
fruiting branches correspond to the terminal buds of the vegetative
branches or of the main stalk.

If the floral buds of the horizontal branches were homologous with
the terminal buds of the upright shoots, they should ave the same
symmetrical position in relation to the stipules. But instead of
standing midway between the stipules the pedicel of the floral bud is
always nearer to one of the stipules than to the other. The leaves of
the main stalk and vegetative branches have equal stipules, but on
the fruiting branches the stipules are generally unequal. The
inequality is especially pronounced in Egyptian cotton, where one of
the stipules is often more than twice as wide as the other. In all cases
the more enlarged stipule subtends the floral bud and frequently
shows a slight attachment to the base of the pedicel.

A STIPULAR RIM ON FRUITING BRANCHES.

Another peculiarity of the fruiting branches, shown especially in the
Egyptian cotton, is a narrow rim, or collar, that connects the bases
of the stipules, so that the base of the leaf may be said to encircle the
branch. This is not true of the leaves of the main stalk, where the
stipules become much more widely separated, with no indication of a
connecting rim. These differences may be ascribed to the fact that
the thickening of the joints of the fruiting branches is not symmetrical

[Cir. 109]

14 CIRCULAR NO. 109, BUREAU OF PLANT INDUSTRY.

in relation to the base of the leaf. The faet that no stipular rim
appears on the main stalk or vegetative branches doubtless results
from the fact that the principal thickening of the vegetative shoots
takes place in the segment of the stalk that is more directly between
the stipules, whereas on the fruiting branches the first thickening is
chiefly on the side that has the enlarged stipule.

POSITIONS OF FLORAL BUDS.

The chief reason for the sympodial theory of the fruiting branches
of cotton is doubtless to be found in the fact that the flowers and
fruits appear to stand opposite to the leaf of the same node, as in other
plants that have sympodial habits of growth. But closer attention to
the cotton plant will make it apparent that the floral bud is not
directly opposite to the base of the petiole of the leaf. When younger
joints of the fruiting branches are examined, the floral buds are found
closer to the axillary buds — that is, more nearly in the position of the
extra-axillary buds of the main stalk and the vegetative branches.

The floral bud appears to have the terminal position because the
true terminal bud remains relatively undeveloped until the growth
of the floral bud is well advanced. The fruiting branches might be
said to grow a joint at a time by successive enlargements of basal
internodes of a small terminal shoot. The accelerated development
of the floral bud leaves the terminal bud at one side with the axillary
bud. But the apparently terminal position of the floral bud is lost
when the base of the next joint thickens to its full size. This pushes
the pedicel farther away from the original insertion next to the axil-
lary bud and also brings it into a more erect position at an angle with
the branch. When the terminal bud is aborted, as often happens,
the pedicel of the last flower or boll remains in fine with the last
internode of the branch. 1

The idea of sympodial branching might be strengthened by cases
where the pedicel is abnormally short and thick or where the devel-
opment of the next internode occurs too late to push the pedicel out
of line. A still closer approximation to the sympodial condition
would be reached if an aborted terminal bud of a fruiting branch
were replaced by the development of an axillary bud. This probably
occurs in some cases, though examples have not been observed thus
far. The axillary buds are usually aborted with the terminal buds.

1 The relative rates of growth of the floral buds and the adjacent terminal buds or young shoots will
probably be found to differ in the various species and varieties of cotton, like other characters. It was
noticed by Mr. G. N. Collins, of the Bureau of Plant Industry, in a variety of Sea Island cotton growing
at Tuxtla, Mexico , in January , 1907, that all of the involucres had one of the bracts with a deep channel or
reentrant angle on the side next to the terminal shoot. It has also been observed by Mr. Rowland M.
Meade, of the Bureau of Plant Industry, that the bract on the side toward the terminal shoot is usually
smaller than the other bracts of the same involucre. See "Arrangement of parts in the cotton plant,'
Bulletin 222, Bureau of Plant Industry, U. S. Department of Agriculture, 1911.

ICir. 109)

MORPHOLOGY OF COTTON BRANCHES. 15

and this is especially likely to occur on the shortened fruiting branches
of small plants or at the end of the season when growth has nearly
ceased. In connection with the bractlike leaves of abnormally
shortened fruiting branches both the axillary and the terminal buds
are sometimes completely suppressed, as in the axils of the bracts of
normal involucres.

TRANSFORMATION OF FRUITING BRANCHES INTO VEGETATIVE

BRANCHES.

Even on normal fruiting branches the pedicels of the floral buds
are not turned aside far enough to bring the successive joints of the
branches entirely into line, which accounts for the characteristic zig-
zag form of the fruiting branches, but when the floral buds are aborted
at a very early stage the branches are able to grow straight, like the
vegetative branches. Such transformed branches may show the scars
of aborted floral buds on a few of the lower joints, while the upper
joints are without scars and are as straight and upright as normal
vegetative branches. This is one of the ways in which fruiting
branches are transformed into vegetative branches, but such cases of
transformation during the growth of the branch are seldom found.
The transformations usually occur at earlier stages of growth, before
the formation of any floral buds, even on the basal joints. On nor-
mal fruiting branches the basal internode is longer than the others,
whereas vegetative branches often have short internodes at the base;
but some branches are intermediate in this respect, as well as in
other characters.

Individual plants with branches of intermediate form have been
found in unacclimatized stocks of Kekchi and other imported types
of cotton. Though behaving as vegetative branches in other re-
spects, an abortive floral bud appears at each node throughout the
season. On some plants all the buds are shed while still very minute,
while on other plants the buds reach more advanced stages of devel-
opment. The closer the approach of the plants to normal fertility
the more zigzag the branches become. The occurrence of s,uch
transformations and in such various degrees does not indicate that
the two forms of branches are so fundamentally different as the
sympodial theory would imply.

SUCCESSIVE GROWTH OF INTERNODES OF VEGETATIVE SHOOTS.

That the zigzag form of the fruiting branches may be due to the
accelerated growth of the floral buds is also indicated by a similar
tendency in rapidly developing terminal shoots of main stalks or
vegetative branches, especially on large luxuriant plants. Here the
joints also appear somewhat zigzag at first, apparently because the

[Cir. 109]

16 CIRCULAR NO. 109, BUREAU OF PLANT INDUSTRY.

young leaf and the fruiting branch of each internode have a period
of very rapid growth when they take nearly upright positions and
throw the delicate terminal shoot out of the perpendicular. The
petiole of a young leaf often becomes thicker, for a time, than the
next joint of the stalk. Subsequent thickening of the joints pushes
the leaves and branches over to the side, so that the zigzag form of
the young terminal shoots is gradually lost and the mature stalk
becomes straight and upright. In cluster varieties the mature
stalks may still show the zigzag form, which is doubtless connected
with the fact that such varieties have an accelerated development of •
short, thick fruiting branches. Thus, if the zigzag form were to be
accepted as a sufficient indication, the main stalk might be consid-
ered sympodial as well as the fruiting branches, but there is no occa-
sion for such a theory, in view of the fact that the mam stalk has no
floral buds or branches opposed to the leaves.

CONCLUSIONS.

There is no apparent necessity to consider the normal development
of the cotton plant as resulting in any truly sympodial structure.
The pseudosympodial form of the fruiting branches arises from the
fact that a flower, instead of a branch, is developed from the extra-
axillary bud, and from the further peculiarity that the fruiting
branches have a stronger tendency to develop one joint at a time.
But notwithstanding these differences the two kinds of branches
may be considered as homologous in other respects. The floral buds
have an extra-axillary position on the internodes of the fruiting
branches, corresponding to the extra- axillary position of the buds
that produce the fruiting branches on the mam stalk or the vegeta-
tive limbs. Transformations of branches from the fruiting to the
vegetative form afford additional evidence that the two types of
branches are not so fundamentally different as the sympodial theory
Would imply. Changes from one form of branches to the other are
readily induced by differences of external conditions.

[Cir. 109]

THE WILTING COEFFICIENT FOR PLANTS IN ALKALI SOILS. 1

By Thomas H. Kearney, 2 Physiologist in Charge of Alkali and Drought Resistant

Plant Investigations.

INTRODUCTION.

In a previous publication of this bureau 3 the wilting coefficient was
defined as " the moisture content of the soil (expressed as a percentage
of the dry weight) at the tune when the leaves of the plant growing
in that soil first undergo a permanent reduction in their moisture
content as the result of a deficiency in the soil-moisture supply. By
a permanent reduction is meant a condition from which the leaves
can not recover in an approximately saturated atmosphere without
the addition of water to the soil." It follows that when the moisture
content of a given soil falls below the percentage representing the
wilting coefficient, the plants are unable to continue their growth.
In the present publication, therefore, the soil water in excess of the
wilting coefficient will be referred to as "moisture available for
growth."

The term "alkali" is here used in the popular sense to designate
those alkali and alkali-earth metals (Na, K, Ca, Mg) which are most
frequently present in excess in the soils of arid regions, together with
the acid radicles (CI, S0 4 , C0 3 , HC0 3 ) with which these bases are
generally associated hi alkali soils.

The experiment described in the following pages was designed to
ascertain whether an alkali content of the soil not sufficiently high to
cause evident injury to the plants would have the effect of raising
the wilting coefficient; in other words, whether the plants under
such conditions would reach a permanently wilted condition before
they had exhausted the soil moisture to a point corresponding to the
wilting coefficient as calculated from the moisture equivalent for the
type of soil used. It was assumed that if this proved to be the case
it would be possible to determine the minimum concentration at
which alkali of a given composition becomes the limiting factor for
plant growth from the inability of the plants to reduce the soil
moisture to the calculated wilting coefficient.

For this purpose it was decided to grow seedling plants of wheat
in a series of soil mixtures having a gradually increasing salt con-
tent. Each soil was to contain at the outset the same amount of
water available for growth — i. e., the same percentage of moisture
.above the calculated wilting coefficient — and no additional water

1 Issued Jan. 4, 1913.

2 Dr. L. J. Briggs, Physicist in Charge of Biophysical Investigations, Bureau of Plant Industry, cooper-
ated by having the moisture equivalents determined and the soil extracts prepared in his laboratory. The
chemical analyses of the extracts were made by Mr. J. F. Breazeale, of the Bureau of Chemistry.

3 Briggs, L. J., and Shantz, H. L. "The wilting coefficient for different plants and its indirect determi-
nation," U. S. Department of Agriculture, Bureau of Plant Industry, Bulletin 230, 1912.

71357— Cir. 109—13 2 17

18

CIRCULAR NO. 109, BUREAU OF PLANT INDUSTRY.

was to be supplied. It was believed that this method of experi-
ment would eliminate all varying factors except the salt content of
the soil solution. In order to determine whether the plants reduced
the moisture content to the wilting coefficient, it was necessary to
provide against direct evaporation from the soil during the course
of the experiment. This provision also served to reduce to a mini-
mum variations in the concentration of the solution in different
portions of the soil mass.

STOCK SOILS USED.

Two natural soils from North Platte, Nebr., one of which had a
much higher salt content than the other, were employed. Each
was moistened sufficiently to make it easy to handle and was then
thoroughly mixed. That a high degree of uniformity was thus
obtained is shown by the close agreement of the determinations of
electrical resistance, moisture content, and moisture equivalent *
made on different portions of each stock soil.

The two stock soils had a mean moisture equivalent of 25.5 and
24.1 per cent, respectively, indicating wilting coefficients of 13.8
and 13.1 per cent. They gave, when saturated, a mean electrical
resistance of 205 and 38 ohms, respectively. 2 The total water-
soluble material was 0.17 and 1.03, respectively, in percentages of
the dry weight of the soil.

The chemical composition of the alkali salts in each of these stock
soils is given in Table I, which follows. The data were obtained by
analyses of extracts prepared in the ordinary manner for alkali
determinations — agitation for 30 minutes with 500 c. c. of water
per 100 grams of soil.

Table I. — Chemical composition of the extracts from soils used in the experiment, in
percentages of the total solids in the extracts.

Soil

Total soluble

(in percentage

of dry weight

of soil).

Percentages of the total soluble solids.

No.

Ca.

Mg. K.

Na.

CI.

SO,.

C0 3 .

HCO3.

78

0.17
1.03

12.2
1.6

5.1

11.5
6.9

14.3

29.2

2.0
8.8

8.7
44.2

25.7

79

8.4

The composition of the soluble matter was markedly different in
these two soils, that which had a low salt content (No. 78) having

1 The moisture equivalent has been shown by Briggs and McLane (Bulletin 45, Bureau of Soils, 1907)
to be an accurate indicator of the moisture-holding capacity of the soil and to be more readily determinable
than the moisture capacity itself. As defined by Briggs and Shantz (Bulletin 230, Bureau of Plant Indus-
try, p. 56), it is "the percentage of water which it [the soil] can retain in opposition to a centrifugal force
1,000 times that of gravity." The method followed in determining moisture equivalents is described on
pp. 56 and 57 of Bulletin 230. From the moisture equivalent the wilting coefficient can be calculated
by means of a formula given on p. 58 of the same publication.

2 In the standard container (capacity about 50 c. c. ) used in determinations of the salt content of the
soil by means of the electrolytic bridge. The apparatus is described by Dr. L. J. Briggs in Bulletin 15,
Bureau of Soils, 1899, pp. 32 to 35. See also "The electrical bridge for the determination of soluble salts
in soils," by R. O. E. Davis and H. Bryan, Bulletin 61, Bureau of Soils, 1910, wherein tables are given
for temperature correction and for computing from the resistance at 60° F. the percentage of salts present.

[Cir. 109]

THE WILTING COEFFICIENT FOli PLANTS IN ALKALI SOILS.

19

been absolutely, as well as relatively, richer in calcium and magne-
sium, but both absolutely and relatively very much poorer in sodium,
in chlorids, and especially in sulphates. Soil No. 78 is also relatively
richer, although absolutely poorer, in potassium and bicarbonates.

In order to insure as nearly as possible the same percentage of water
available for growth in all soil mixtures used in the experiment, suf-
ficient water was added to each stock soil to give them practically the
same "available" moisture content. The total water contents at-
tained were 24.4 per cent in soil Xo. 78 and 23.9 per cent in soil No.
79, which gave available-water contents of 10.6 and 10.8, respectively.

PREPARING THE SOIL MIXTURES.

It was decided to use in the experiment stock soil No. 78 and 14
mixtures of the two stock soils. A graduated salt content in these
mixtures was obtained by varying the proportions of the two stock
soils. Precautions were taken to reduce evaporation to a minimum
during the mixing and subsequent handling of the soils. Determina-
tions of the electrical resistance (in ohms, at 60° F.) and of the total
water-soluble material, moisture equivalent, and moisture content (in
percentages of dry weight of soil) were made upon each mixture as
prepared for planting. These values, together with the initial con-
centration of the soil solution (calculated from the total water-soluble
material and the initial water content) , the wilting coefficient (calcu-
lated from the moisture equivalent), and the percentage content of
water available for growth (by subtraction of the wilting coefficient
from the percentage of total water present) are given for each of these
mixtures in Table II.

Table II. — Electrical resistance, total soluble material, concentration of solution, mois-
ture equivalent, uniting coefficient, moisture content, and moisture available for growth
in the soil mixtures used (means of all determinations) .

Mixing pro-
portions of
stock soils.

Soil-
mix-
ture

Electri-
cal re-
sistance.

Total

soluble

material.

Initial
concen-
tration
of solu-
tion. 1

Moisture
equiva-
lent.

Wilting
coeffi-
cient
(calcu-
lated).

Moisture
content.

Moisture

available

for

78

79

No.

growth.

Ohms.

Per cent.

20

1

200

0.183

0.75

25.5

13.8

24.4

10.6

19

1

2

105

.222

.94

25.3

13.7

23.6

10.1

18

2

3

136

.236

1.00

25.5

13.8

23.6

9.8

17

3

4

122

.305

1.24

24.5

13.3

23.6

10.3

16

4

5

111

.336

1.42

24.5

13.3

23.7

10.4

15

5

6

99

.365

1.54

24.4

13.3

23.7

10.4

14

6

7

89

.405

1.71

25.0

13.6

23.7

10.1

13

7

8

81

.418

1.76

25.5

13.8

23.7

9.9

12

8

9

76

.488

2.06

25.2

13.7

23.7

10.0

11

9

10

72

2.409

(?)

25.0

13.6

23.6

10.0

10

10

11

65

.509

2.14

23.7

12.9

23.8

10.9

9

11

12

62

.027

2.63

24.9

13.5

23.8

10.3

8

12

13

59

.675

2.85

24.5

13.3

23.7

10.4

7

13

14

55

.740

3.15

24.1

13.1

23.5

10.4

6

14

15

50

.759

3.17

23.8

12.9

23.3

10.4

1 Since the percentage of moisture corresponding to the wilting coefficient for the 15 soils averaged 56 per
cent of the percentage representing the initial moisture content and since there was little change in the
total salt content of the soils (as shown by determinations of electrical resistance) during the period of the
experiment, it follows that the concentration of the solution in each soil had very nearly doubled when
the wilting coefficient was reached.

2 There was evidently an error here, either in taking the sample or in the determination.

[Cir. 109]

20 CIRCULAR NO. 109, BUREAU OP PLANT INDUSTRY.

The figures for moisture equivalent in Table II show some irregu-
larities, doubtless attributable to errors in the method of determina-
tion. Such errors could be eliminated only by making a larger
number of determinations for each soil. The contents of water
available for growth in the whole series of mixtures ranged from
9.8 to 10.9 per cent, but these differences were hardly of sufficient
magnitude to affect the results of the experiment.

Chemical analyses of extracts from each of these soil mixtures
showed, for most components, an increase from mixture to mixture
corresponding to the increase of total soluble matter. A note-
worthy exception was HC0 3 , of which practically the same quan-
tity was found in all. The mixtures having the lowest total salt
content had the highest content of calcium and magnesium, as would
be expected from the fact that larger amounts of these bases existed
in stock soil No. 78 than in stock soil No. 79.

PLANTS USED AND CONDITIONS OF GROWTH.

Kubanka wheat (Grain Investigations No. 1440) was used, the
stock having been the same as that employed by Messrs. Briggs
and Shantz in their investigations of the wilting coefficient.

The seeds were placed in moist blotters on April 3; three days
later they were in proper condition for planting. While no special
effort was made to select seedlings of the same size, none were planted
winch did not have the pluinule and the rootlets extruded. All
seedlings planted appeared to be perfectly healthy and had no roots
much longer than 1 inch. The seedlings were planted in straight-
walled, cylindrical drinking glasses, each holding about 250 grams
of soil. Five seeds were planted in each glass and there were six
glasses of each soil mixture, making a total of 30 seedlings planted
in each soil mixture used in the experiment.

In planting, most of the soil required was placed in the glass
and was shaken down by tapping the bottom of the glass on the
table. The seedlings, taken directly from the moist blotters, were
laid on the loose surface of the soil; the glass was then filled evenly
to the brim with the remaining soil, so as to completely cover the
seedlings, and the soil was pressed down firmly, first around the
edges and then in the center.

The surface was then sealed over with- a melted mixture of 80
per cent of paraffin and 20 per cent of petrolatum, as described in a
previous publication of this bureau. 1

In order to equalize as far as possible the temperature of the
soil in the glasses and to prevent condensation, which would result
in an unequal distribution of the water and salts in different portions

i Bulletin 230, pp. 10-14,
[Cir. 109]

THE WILTTNO COEFFICIENT FOR PLANTS IN ALKALI SOILS.

21

of the soil mass, the glasses were kept in a partitioned metal tank,
through which a stream of water was kept flowing during the
whole period of the experiment. 1 A light tent of bolting cloth was
stretched over the tank during hours of bright sunlight, in order to
prevent, if possible, the heating and melting of the wax seals. 2

EMERGENCE OF THE PLANTS.

A record was kept of the rate at which the plants in each glass
appeared above the surface of the soil, this record having been based
on daily observations during the first 12 days (with the exception of
the eighth day after planting) and a final observation on the four-
teenth day, after which no more plants emerged. In mixture No. 15
no plants appeared.

Table III shows the total number of plants (out of the 30 inserted)
which finally appeared above the seal and the number of days from
the date of planting to that when the last plant emerged in each soil
mixture.

Table III

.—Rate oj

emergence of the plants in the different soils.

Soil-mixture No.

Number of

plants iliai

emerged.

Number of

days until

emergence

ceased.

Soil-mixture No.

Number of

plants that

emerged.

Number of

days until

emergence

ceased.

1

30
30
30
30
30
30
30
30

4
4
4
4

4
4
4
4

9

30
30
24
15
23
10

6

2

10

7

3

11

10

4

12

14

5

13

10

6

14

14

7

15

8

The behavior of the plants in the several glasses of the more con-
centrated soils (Nos. 11 to 14) was decidedly puzzling. In mixture
No. 11 all plants emerged in five of the glasses and none in the sixth.
In mixture No. 12 all plants emerged in three of the glasses and none
in the other three. In mixture No. 13 all plants emerged in three of
the glasses, all but one of the plants in two others, and none in the
sixth. In mixture No. 14 all plants emerged in two of the glasses
and none in the other four. No explanation could be found for these
irregularities. Electrolytic bridge readings at the close of the experi-
ment showed that they could not be attributed to differences in salt
content of the soil as between the glasses in which growth was made
and those in which no plants emerged.

1 This method is described and illustrated in Bulletin 230, Bureau of Plant Industry, p. 13.

2 In spite of these precautions, owing to the rather high temperatures that prevailed during the period
of this experiment , some softening and buckling took place in several of the glasses, which made it neces-
sary to break the seals in order to release plants caught beneath them. The seals were repaired so quickly
that no appreciable loss of water by direct evaporation could take place. This could have been obviated
by using for the seals a composition having a higher melting point.

rrtr mm

22

CIRCULAR NO. 109, BUREAU OF PLANT INDUSTRY.

Nevertheless, the figures given in Table III show clearly a retarding
effect of the more strongly saline soil mixtures. Observations at more
frequent intervals than every 24 hours might have shown some retar-
dation in mixtures less concentrated than No. 9, although it was noted
that emergence was at least as prompt in No. 5 as in Nos. 1 to 4.

CHARACTER OF GROWTH.

On April 13, one week after planting, the longest leaf in each glass
was measured and the mean of these values for the six glasses of each
soil was taken as expressing the growth made in that soil while there
was still abundant "available" moisture. Table IV shows the values
thus obtained.

Table IV. — Mean length of the longest leaf in each soil one week after planting.

Soil-mixture No.

Mean
electrical
resistance.

Mean
length of
the long-
est leaf.

Soil-mixture No.

Mean
electrical
resistance.

Mean
length of
the long-
est leaf.

1

Ohms.

200

165

136

122

111

99

89

81

Millimeters.
153.5
156.0
138.0
134.0
128.0
121.0
113.0
96.7

9

Ohms.
76
72
65
62
59
55
50

Millimeters.
81.0

2

10

66.0

3

11

42.9

4

12

29.0

5

13

21.0

6...

14

29.0

7

8

These data show a much more marked correlation between the
rate of growth of the plants and the salt content of the respective
soils than the figures relating to rate of emergence given in Table III.

In soils Nos. 1 to 8 the seedlings made a good, healthy growth in all
the glasses. In Nos. 9 to 1 1 their growth was good in the main, but
the plants were smaller. In Nos. 12 to 14 the growth was irregular
in the glasses in which any plants emerged, some plants having been
small and sickly, while others had made a fair growth and appeared
fairly normal. For some reason the average growth was better in
No. 13 than in No. 12.

DETERMINATION OF THE WILTING COEFFICIENT.

As soon as the plants appeared to be in a wilted condition the
soils were collected from the glasses for moisture determination.

Table V gives the final water content of the soil in each glass in
which sufficient growth was made to reduce the water content to
anywhere near the wilting coefficient, as well as the mean of these
determinations for each soil mixture. For comparison, the mean
wilting coefficient is given for each mixture, as calculated from the
moisture equivalents at the outset of the experiment.

[Cir. 109]

THE WILTING COEFFICIENT FOR PLANTS IN ALKALI SOILS. 23

Table V. — Water contents of the soils when plants first appeared wilted, with calculated
wilting coefficients of the corresponding mixtures.

Soil-mix-
ture No.

Glass
No.

Water content
when plants first
appeared wilted.

For each
glass.

Mean for
each soil
mixture.

13.4
13.3
13.2

12. S i

13. 3
13.4

15.4
13.7
13.5
14.6
13.9
13.8

14.2
14.0
14.1
14.8
14.5
14.5

14.7
14.8
13.8
14.1
14.9
15.1

14.7
14.0
14.4
14.1
15.0
14.1

14.0
13.2
14.0
13. 6
15.2
14.5

13.2

14.1

14.3

14.6

Wilting
coefficient
(calcu-
lated
from the
moisture
equiva-
lents).

14.4

14.1

13.8

13.7

13. S

13.3

13.3

13.3

Soil-mix-
ture No.

Glass
No.

10.

11.

12.
13.

I I

Water content
when plants first
appeared wilted.

For each
glass.

13.5
13.8
13.4
13.9
13.6
13.2

13.0
12.8
13.4
13.4
13.3
13.5

15.2
14.5
14.1
14.4
14.0
13.6

13.6
14.0
14.1
13.4
14.1
14.6

14.0
14.2
13.8
13.9
14.7
15.2
13.4
12.8
15.5
13.3
13.3
13.5
14.8

Mean for
each soil
mixture.

Wilting
coefficient
(calcu-
lated
from the
moisture
equiva-
lents).

13.6

13.2

14.3

14.0

14.1

13.6
13.8
13.7

13.6

12.9

13.5
13.3
13.1

The figures in Table V show the mean water content at the time
of apparent wilting to have been higher in most of the soil mixtures
than the calculated wilting coefficient. 1 Nevertheless, even in the
most strongly saline mixture (No. 14) in which any emergence took
place, the plants in at least one of the glasses were able to reduce the
moisture content to a percentage agreeing, within the limits of
experimental error, with the calculated wilting coefficient.

On the other hand, in glasses where the plants remained very
small and sickly this was not possible. Thus, in one of the glasses
of mixture No. 13, in which the plants began to die at about the date

i Probably because the high temperatures that prevailed toward the close of the experiment made it
rather difficult to distinguish between temporary and permanent wilting. More accurate determina-
tions of the permanent wilting point could doubtless have been made if the plants, as soon as they ap-
peared to be wilted, had been transferred to a saturated atmosphere in order to determine whether
they could recover their turgidity.

[Cir. 109]

24

CIRCULAR NO. 109, BUREAU OP PLANT INDUSTRY.

when the wilting coefficient had been reached in the least saline mix-
tures, the soil-moisture content (determined when it became evident
that the plants would make no further growth) was found to be 22.8
per cent. Since the initial water content of this soil was 23.7 per
cent and the initial moisture available for growth was 10.4 per cent
(Table II), the loss of water by transpiration in this glass amounted
to only 3.8 per cent of the total and 8.6 per cent of the available
moisture present at the outset.

While the moisture content of all the soils was finally reduced in at
least one of the glasses to the wilting coefficient, tins happened much
more rapidly in the soils winch had a high electrical resistance (low
salt content) than in those which had a low electrical resistance (high
salt content). There was, on the whole, a remarkably close correla-
tion between the salt content of the soil and the length of time
required for the plants to reach a wilted condition. The relation is
expressed hi Table VI.

Table VI. — Time required to exhaust in each soil mixture the water available for growth.

Number
of glasses
in which
wilting
occurred.

Soil mixture No.

Mean
electrical
resist-
ance.

Mean

number

of days

from

date of

plant ing

until

wilting

began.

Number
of glasses
in which
wilting
occurred.

1

200
165 •
136
122
111

99

89

18.0
18.0
19.3
22.0
25.0
23.5
24.3

6
6
6
6
6
6
6

2

3

4

5

6

7

Soil mixture No.

Mean
electrical
resist-
ance.

Mean
number
of days

from

date of

planting

until
wilting
began.

8

9

81
76
72
65
62
59
55

27.0
27.6
31.5
33.4
36.0
38.7
43.5

10

11

12

13

14

CONCLUSIONS.

The results of this experiment make possible the following conclu-
sions :

(1) The presence of an excess of soluble salts in a soil did not affect
the ability of young wheat plants to reduce ultimately the water con-
tent of that soil to the wilting coefficient, unless the quantity of salts
was sufficient to induce marked pathological symptoms in the plants.

(2) The time required for the exhaustion of the water available for
growth increased steadily with increasing concentration of the soil
solution (Table VI).

(3) The amount of growth made, so. long as the moisture content
of the soil remained well above the wilting coefficient, was deter-
mined by the concentration of the soil solution (Table IV).

(4) The presence of alkali increased the water requirement of the
plants ; in other words, increased the quantity of water transpired in

LCir. 109]

THE WILTING COEFFICIENT FOR PLANTS IN ALKALI SOILS. 25

producing a unit weight of dry matter. The conditions of the experi-
ment here described did not allow of accurate determinations of
water requirement, but this conclusion seems unavoidable in view
of the following facts: (a) Some of the plants in each soil mixture
were able to reduce the moisture content to the wilting coefficient.
Hence, practically the same total quantity of water was finally lost by
transpiration from each soil, (b) The plants in the more strongly
saline soils were much smaller when the wilting coefficient was reached
than were those in the soils having a lesser salt content.

Experiments are now in progress to measure the effects of different
quantities of alkali upon the water requirement of plants.

[Cir. 109]

INTERPRETING THE VARIATION OF PLAT YIELDS. 1

By F. D. Farrell, Agronomist, Office of Western Irrigation Agriculture.

FACTORS OF ERROR IN FIELD EXPERIMENTS.

All results obtained in field experiments are subject to errors the
extent of winch can not always be accurately determined by physical
measurements. These errors may be due to a large number of causes,
such as incorrect weighing of crop products, faulty determinations of
plat areas, variations in the quantities of products recovered and
wasted, unobserved variations in field treatment, etc. These are
termed experimental errors, and the influence of some of them can be
determined and controlled. Other important causes are individual
plant variations, soil irregularities due to natural conditions or to
nonuniform previous treatment, uneven distribution of soil moisture,
temperature variations, etc. The influence of these factors can sel-
dom be accurately determined or controlled. Frequently the total
effect of all these causes is so great as seriously to influence the conclu-
sions drawn from observed results, and if it is not taken into con-
sideration the results may be badly misinterpreted. Where a large
number of observations are made of results obtained under condi-
tions winch are uniform, except as they are affected by accidental
error, the extent of that error can be calculated by a mathematical
formula, and a part of the calculation results in what is known as the
"probable error."

EXPERIMENTAL DETERMINATION OF THE PROBABLE ERROR.
FIELD TESTS IN NEBRASKA AND MONTANA.

In order to determine as nearly as possible to what extent these
factors would influence the crop yields of plats to be used in the field
experiments under irrigation at the Scottsbluff (Nebr.) and Huntley
(Mont.) experiment farms, certain of the fields to be used were
planted to uniform crops and given uniform treatment. In 1911
field K, at Scottsbluff, containing 66 quarter-acre plats, and field K,
at Huntley, containing 68 quarter-acre plats, were planted to oats.
At Huntley in the same year fields B II and B III, which are adjacent,
and which contain 46 quarter-acre plats, were planted to sugar beets.
In 1912 fields B II and B III, at Huntley, were planted to alfalfa.
Ten adjacent quarter-acre plats were planted to alfalfa in 1912 in

1 Issued Jan. 4, 1913.
[Cir. 109] 27

28 CIRCULAR NO. 109, BUREAU OP PLANT 'INDUSTRY.

field A II at Huntley. In each case the plats were harvested sep-
arately, and the yields obtained form the basis of the material pre-
sented in this paper.

The results obtained with 66 plats of oats on field K at Scottsbluff
in 1911 will serve as an illustration of the method pursued. These
plats received uniform treatment throughout the season, so that
without a consideration of accidental errors it would be expected that
the plat yields would be equal ; or, in other words, it would be expected
that the yield of any plat taken at random would fairly represent the
yield of any other plat in the field. Such an expectation is, of course,
never realized. It was found in the case of the oat plats at Scotts-
bluff that the yields varied from 372 pounds to 212 pounds — an
extreme range of 160 pounds per plat. This variation is equivalent
to 20 bushels per acre, or 55.7 per cent of the mean yield. Where a
difference of 20, or even 10 or 5, bushels per acre is obtained from two
plats receiving different treatments the difference is commonly con-
sidered significant and attributable to the treatments. But it is
obvious that a difference of 5 bushels per acre would not have been
significant on field K in 1911 if the oat plats had received different
treatments.

It is commonly found that approximately one-half the results
observed in a series will be greater than the mean of all the results
and that approximately one-half the results will be less than the mean,
and also that the number of results differing from the mean will
decrease as the magnitude of the difference increases. For example,
there are 29 plats in field K which yielded within 24 pounds of the
mean yield, while only 10 plats differed from the mean yield by more
than 48 pounds. Because of this tendency for the majority of the
results to lie relatively close to the mean, the mean is taken as a basis
upon which to calculate the probable error.

In a "perfect" series " The probable error is such that, talcing any
single result at random, the chances are even for or against that result
differing from the average by the amount of the probable error. In other
words, half the results should differ from the mean by less than the
probable error, the other half by more." 1

The probable error is calculated in the following manner: The
mean of all the results is determined. This mean is subtracted from
each result to obtain the difference, each difference is squared, and
the sum of these squares, disregarding the sign, is found. Then the
following formula is employed:

The probable error of a single result=

P fi7 .r /The sum of the squares of the differences.
yThe number of results less 1.

i Wood and Stratton. Journal of Agricultural Science (Cambridge), v. 3, pt. 4, p. 417. December, 1910.
[Cir. 109]

INTERPRETING THE VARIATION OF PLAT YIELDS. 29

The 66 plats at Scottsbluff are not a sufficient number to form a
"perfect" series, where the variations are so great, but they afford an
opportunity of calculating the probable error with sufficient accuracy
for practical purposes. The mean yield of the 66 plats is 287 pounds
and the probable error of any single result is found to be ±23.9
pounds, or 8.3 per cent of the mean yield. This is equivalent to 3
bushels per acre. According to the above definition of probable, error,
33 plats should yield within 23.9 pounds of the mean, and the yield of
the other 33 plats should vary from the mean by more than 23.9
pounds. It was actually found that 29 plats varied from the mean by
less than the probable error and 37 varied by more. While those
numbers 29 and 37 are not equal, they are as close as can be expected
in a series of only 66 observations and close, enough for practical
purposes.

The specific application of these determinations to field experiments
is this: In a series of plats planted to a single crop and given uniform
treatment marked variations may occur and from these variations
the probable error may be determined. If, then, in the same series of
plats later planted to a single crop and given different treatments, the
variations in yield do not exceed the probable error as previously
determined, these variations can not properly be attributed to the
effects of the treatments and they should not be considered as sig-
nificant. In other words, the probable error may be employed as a
measure of the confidence which can safely be placed in experimental
results. The more a variation, supposedly due to a treatment, ex-
ceeds the probable error the greater the. confidence which can properly
be placed in the supposition that the treatment is responsible for the
variation. If a variation is greater than twice the probable error, its
significanee is very materially increased, antl if it is so large as to
exceed three, four, or five times the probable error, its significance
approaches a practical certainty.

In the series of plats at Scottsbluff only ten plat yields Varied from
the mean by more than twice the probable, error, and only four varied
by more than three times the probable error. There were no vari-
ations from the mean equal to four times the probable error. In
other words, to have obtained a variation equal to four times the
probable error it would have been necessary to introduce, a variation
in the treatment. If under this different treatment a variation in
yield greater than four times the probable, error were obtained, it
would be a practical certainty that the variation was due to the
treatment. But any variation less than this would be subject to the
suspicion that it might be due to accidental errors, such as soil irregu-
larity, etc., rather than to treatment, and the suspicion would increase
as the size of the variation decreased.

[Cir. 109]

30

CIRCULAR NO. 109, BUREAU OF PLANT INDUSTRY.

Table I contains the probable error determinations made on five
groups of plats, one group at Scotts bluff, Nebr., and four at Huntley,
Mont., together with some other data associated with the probable
error in each case. Explanations of the various items are given after
the table under the respective numbers of the items.

Table I. — Probable error determinations made on five groups of field plats at the Scotts-

bluff and Huntley experiment farms .

Items.

1. Mean yield (groupsA,B,D,andEln pounds: group

C in tons per acre)

„ /Number of plats yielding above the mean

\ Number of plats yielding below the mean

3. Extreme range of plat yields, expressed in percent-

age of mean yield of each group

4. Mean deviation of plat yields, expressed in per-

centage of mean yield of each group

5. Probable error of asingle result expressed in terms

of yield

6. Probable error, expressed in percentage of mean

yield of each group:

a Of any single result

6 Of the average of 10 results

c Of the average of 20 results

d Of the average of 30 results

e Of the average of 40 results

/Of the average of 46 results

a Of the average of 66 results

ft Of the average of 68 results

7. Number of plat yields differing from mean:

a By less than the probable error —

Actual

Theoretical

6 By more than the probable error-
Actual

Theoretical

c By less than twice the probable error

d By more than twice the probable error

« By more than three times the probable error.

/ By more than four times the probable error. .

Groups.

A.

66 plats

oats,

Scotts-

blufffleld

K, 1911.

287
35
31

55.7

9.8

±23.9

8.3

2.6
1.9
1.5
1.3
1.2
1.0

29
33

37
33

56

10

4

B.

68 plats

oats,

Huntley

field K,

1911.

554
37
31

106.0

22.5

±105.95

19.1

6.0

4.3

3.5

3.0

2.8

2.34

2.31

26
34

42
34

63
5

C

46 plats

beets,

Huntley

fields B II

and Bill,

1911.

D.

46 plats

alfalfa,

Huntley

fields B II

and B III,

1912.

11.71

29

17

67.9

12.9

± 1.29

±11.0

3.5
2.4
2.0
1.7
1.6

25
23

21
23

38

8
4

356
23
23

52.5

12.5
±38.18

±10.7

3.4
2.4
1.9
1.7
1.6

19
23

27
23

38
8

E.

10 plats

alfalfa,

Huntley

field A II,

1912.

572
5
5

51.5

9.6

±49.27

± 8.6

±2.7

6
5

4
.">
7
3
1

(1) The mean yield of each group of plats is expressed in pounds per plat, except
in the case of group C, where, for convenience, it is expressed as tons per acre.

(2) This item is given to show how closely the plat yields in the different groups
agree with the assumption that approximately one-half the results will exceed the
mean. It is shown that in groups D and E the division is exactly equal; in A it is
very close, and in B and C it is rather noticeably unequal. It should be stated that
in group C, 7 of the 46 plats were spring plowed and the others fall plowed in prepara-
tion for beets. To this extent the treatment of the plats in group C was not uniform.
The mean yield of the 7 spring-plowed plats was 0.83 ton per acre less than the mean
yield of all 46 plats. The discrepancy introduces an error into the calculations,
but it has only a very slight effect on the probable error.

(3) The extreme range in the yields obtained in each group is stated as a percentage
of the mean yield of each group so as to facilitate comparisons among the groups.
The extreme range in the plat yields at Scottsbluff was from 372 to 212 pounds per

[Cir. 109]

INTERPRETING THE VARIATION OF PLAT YIELDS. 31

plat, a total variation of 160 pounds, or 55.7 per cent of the mean yield. It will be
noted that a high extreme range is closely associated with a high probable error.

(4) This item includes the mean of the deviations from the mean yield of each group
expressed as a percentage of the mean yield. It will be noted that this is closely
associated with and slightly greater than the probable error of a single result in each
case.

(5) The probable error of a single result is first given in terms of yield. The prob-
able error of ±23.9 pounds in the case of group A means that if the yield of any plat
in the group is taken at random the chances are even that the yield of such plat will
be within 23.9 pounds of the mean yield of the group. It should be noted that the
figure expressing the probable error is always plus or minus. Hence, while a result
may vary from the mean by less than the probable error, it may be either greater
or less than the mean.

(6) In order to facilitate comparisons the probable error is here stated as a percentage
of the mean yield. The probable error in group A is ±23.9 pounds of oats per plat,
or 8.3 per cent of the mean yield. The probable error of ±49 pounds of alfalfa per plat
in group E is 8.6 per cent of the mean yield of that group. When reduced to a per-
centage basis the probable error of group A and that of group E are seen to be very
nearly the same. The probable error of group C and that of group D are also extremely
close. This fact is of special interest, because the fields used were the same in both
cases — fields B II and B III, Huntley — while the crops were different and were,
of course, produced in different seasons.

So far, the probable error of a single result, the yield of a single plat, has been con-
sidered. When the average of several results is taken, the probable error of that
average is, of course, less than the probable error of one result. In other words, the
average yield of 10 plats is likely to represent the mean yield of the group more nearly
than does the yield of any single plat taken at random. The average yield of any 10
plats taken at random in group A will have a probable error of ±7.59 pounds per plat,
or 2.6 per cent of the mean yield, as compared with 8.3 per cent of the mean yield,
the probable error of a single result. If, then, the number of results averaged be
increased to 20, the probable error is still smaller — 1.9 per cent of the mean yield in
group A ; and the probable error of the average grows continually smaller as the num-
ber of results averaged is increased, as is shown in the table. The fact that the prob-
able error of the average of several results is less than that of a single result empha-
sizes the desirability of conducting field experiments in multiplicate. The probable
error of an average can be obtained by dividing the probable error of a single result
by the square root of the number of results averaged.

(7) Under this heading are given the number of plat yields differing from the mean
by more or by less than the probable error or one of its multiples. Under a and b is
given the number of results which, theoretically, should differ from the mean by
more or by less than the probable error, as well as the actual number found. Theo-
retically, half the results should differ from the mean by less and half by more than
the probable error. The table shows that the actual results are in the main reasona-
bly close to the theoretical requirements. They are as close as can be expected
where such small numbers of results are considered. Under d, e, and fit is seen that
comparatively few plat yields differ from the mean by more than twice the probable
error, still fewer by more than three times, and none by more than four times the
probable error. Hence, it may be expected that on these fields accidental variations
as great as four times the probable error are not likely to occur; so that when different
treatments are applied and variations greater than four times the probable error are
obtained, such variations can be attributed to the treatments with a high degree of
certainty.

[Cir. 109]

32 CIRCULAR NO. 109, BUREAU OF PLANT INDUSTRY.

FIELD TESTS IN ENGLAND.

Wood and Stratton, of Cambridge, England, have published some
results relative to the probable error of field experiments. Their in-
vestigations covered numerous observations on wheat, barley, oats,
mangels, potatoes, and other crops. They observed that there was
no significant difference between the probable error of one of these
crops and that of any other. This is in a way supported by the close
agreement found at Huntley with sugar beets and alfalfa, groups C
and D, where the same field was used for loth trials. But Wood and
Stratton take 5 per cent as the probable error of a single plat in
carefully conducted field experiments. 1 The facts that at Huntley
and Scottsbluff widely different probable errors were found for the
same crop (oats) and that the probable errors determined at the two
farms are considerably greater than 5 per cent indicate that to take
5 per cent as " the probable error of field experiments " is not generally
justified.

CONCLUSIONS.

The material here presented may be used to emphasize the follow-
ing points:

(1) Probable error determinations are valuable as a measure of
the reliability of results of field experiments and of the confidence
which may properly be placed in conclusions drawn from such results.

(2) The value of experiments would be greatly increased if the
probable error applicable to the field were determined in advance;
and where the experiments contain several plats receiving the same
treatment it is desirable to determine each year the probable error
of the results obtained on such plats, as in this way the effect of
seasonal factors on the probable error may be estimated.

(3) In conducting field experiments all possible precautions should
be taken in order to minimize accidental errors.

(4) It is always desirable to conduct tests in multiplicate, so as to
be able to compare averages rather than single results.

(5) In interpreting and publishing experimental results the prob-
able error should be given consideration; otherwise, insignificant
differences may be unduly emphasized.

1 " The probable error of field experiments is investigated by two independent methods, and found to
be about 5 per cent of the crop."— Journal of Agricultural Science (Cambridge), v. 3, pt. 4, p. 440. De-
cember, 1910.

[Cir. 109]

ADDITIONAL COPIES of this publication
-ii- may be procured from the Superintend-
ent of Documents, Government Printing
Office, Washington, P. C. , at 5 cents per copy

U. S. DEPARTMENT OF AGRICULTURE.

BUREAU OF PLANT INDUSTRY— Circular No. 110.
B. T. GALLOWAY, Chief of Bureau.

MISCELLANEOUS PAPERS.

Grass Demonstrations in the South.

Some Asiatic Actinidias DAVID FAIRCHILD

Powdery Dry-Rot of the Potato W. A. ORTON

Preparation of Land for Egyptian Cotton in the Salt River Valley, Arizona . E. W. HUDSON
Agriculture on the Truckee-Carson Project:

Vegetables for the Home Garden . . F. B HEADLEY and VINCENT FULKERSON

Fungous Staining of Cotton Fibers ALBERT MANN

The Jack Bean and the Sword Bean C. V. PIPER

Fiber from Different Pickings of Egyptian Cotton .... T. H. KEARNEY

LIBRARY

NE W YORK

EN

Issued January 18, 1913.

WASHINGTON:

GOVERNMENT PRINTING OFFICE.

1913.

[Cir. 110]
2

BUREAU OF PLANT INDUSTRY.

Chief of Bureau, Beverly T. Galloway.
Assistant chief of Bureau, William A.Taylor.
Editor, J. E. Rockwell.
Chief Clerk, James E. Jones.

GRASS DEMONSTRATIONS IN THE SOUTH.

The accompanying correspondence, dealing with the subject of
demonstrations in grass culture, is submitted as showing the interest
in this kind of work in South Carolina. It is published by direction
of the Honorable Secretary, who for years has advocated the pro-
duction of grass and legumes as a step toward greater diversification
of crops and the increase of animal husbandry in the Southern
States.

MEMORANDUM FOR THE SECRETARY.

Dear Mr. Secretary: You may be interested in the efforts being
made to encourage the production of grass as a feature of stock raising
in South Carolina. Last spring Mr. W. W. Long, of our Farmers'
Cooperative Demonstration Work, inaugurated a series of demon-
strations in portions of the Piedmont region to show the possibilities
of grass and hay production. He took up the work with 100 different
farmers, inducing each of them to make an acre demonstration with
grass. The predominant industry of this section is cotton raising and
not a great deal of interest is taken in live stock. Each farmer put
out an acre of grass under the direction of Mr. Long. The land was
thoroughly plowed, a ton of lime to the acre was applied, and, in
addition, 400 pounds of fertilizer used. The fertilizer consisted of 200
pounds of acid phosphate and 200 pounds of ground bone. The acid
phosphate cost $12 a ton and the ground bone $29 a ton, which would
make the cost of the 400 pounds of fertilizer $4.10. The lime cost
about $5 a ton. The farmers furnished all the work, the lime, and the
fertilizer. The grass seed was furnished from cooperative funds.
The most successful mixture for hay in this section of the South is
found to be one made of orchard grass, tall meadow oat-grass, Italian
rye-grass, and red clover. A half bushel each of orchard grass, tall
meadow oat-grass, and Italian rye-grass, and 10 pounds of red-clover
seed were sown to the acre. This very heavy seeding was believed to
be necessary in order to insure a good stand. We have been using
this mixture for some time on the Arlington Experimental Farm with
good success, cutting as high as 2J and 3 tons of hay to the acre.

The work in South Carolina has been very satisfactory, practically
all of the 100 farmers having a fine crop of hay this season. From
all indications, some of these plats will cut 2h to 4 tons per acre. So
much interest has been aroused in the matter that farmers are now
preparing to put out 2 additional acres, making 3 acres in all. They
have found by this demonstration that they can grow enough hay

1 Issued Jan. IS, 1913.
[Cir. 110] 3

4 CIRCULAR NO. 110, BUREAU OF PLANT INDUSTRY.

to take care of the two or three head of horses used in their general
farm work. The success with the hay crop has encouraged farmers,
furthermore, to look into the live-stock proposition to the end of
introducing greater diversification and devoting less acreage to cotton.
Very respectfully,

B. T. Galloway,

Chief of Bureau.
April 28, 1912.

memorandum for the secretary.

Dear Mr. Secretary: You will probably recall that' last spring
we sent you a memorandum concerning some important demonstra-
tion work in South Carolina by Mr. Long, who is connected with this
Bureau under Mr. Knapp. Mr. Long started out with the plan of
demonstrating to the farmers of South Carolina the practicability
of growing grass. About 100 good farmers were selected and each
agreed to put out an acre of grass under Mr. Long's direction. In
the memorandum which I forwarded you the success of the work at
that time was indicated. As a result of further work a great deal of
interest has been aroused in the subject. Mr. Long informs me that
300 demonstration plats have been put out this fall.

Recently Hon. A. F. Lever made a trip over the State in company
with Mr. Long, and I inclose herewith a copy of a letter winch Mr.
Lever has forwarded; also a copy of a letter which he wrote to Prof.
English, of Clemson College, both of which are self-explanatory.
Very respectfully,

November 25, 1912.

B. T. Galloway,

Chief of Bureau.

[Inclosure.]
Prof. B. T. Galloway,

Washington, D. C.

Dear Dr. Galloway: I am inclosing herewith a copy of a letter which I have
written to-day to Prof. English, of Clemson College, which explains itself. I have
expressed to him what I desire you to know as my feeling in regard to this grass-
demonstration work now being conducted in this State. I do not believe that I shall
have any hesitation in saying that the Department is undertaking no greater work
than this is, and I sincerely trust that the funds set aside for this work for this season
will be sufficient to enable it to develop as rapidly as possible. I desire to talk with
you about it when I come on to Washington.

I regret that you could not be with us on the trip, as 1 know that you would have
enjoyed it.

With personal regards, I am, very truly,

A. F. Lever,
Member of Congress from 'South Carolina.
Lexington, S. C, November 22, 1912.

[Inclosure.]
Prof. W. L. English,

Clemson College, S. C.

My Dear Prof. English: You will recall seeing Mr. Long, Mr. Dorrick, and me
starting on our trip to look at the grass demonstrations being conducted by the Depart-
ment of Agriculture in 20 of the counties of this State. I have become so much inter-
ested in the work and feel that it is of so much importance to the future of the State
that I felt that I should call your attention to it that you in turn might bring the
work to the notice of those in authority at Clemson.

[Cir. 110]

GRASS DEMONSTRATIONS IN THE SOUTH. 5

The demonstration plats in the upper part of Richland County have been growing
for one year, and not only do they strike a layman, such as I, as having been entirely
successful, but the demonstrators themselves are enthusiastic over the results ob-
tained and are proving it by the fact that they are increasing their acreage, and their
neighbors are following their example.

The plat of Mr. Sam C. Cathcart, a few miles above Winnsboro, was particularly
successful, for that gentleman told me that he had gathered 3 tons of first-class hay
from a little less than an acre of ground and that he valued it at about $33 per ton,
that being the price of good timothy hay at Winnsboro. Mr. Cathcart is not only
a large farmer, but runs a considerable-sized dairy, and he tells me that in this grass
work of the Department he sees the beginning of the solution of the agricultural
problems of this State as far as forage crops are concerned and as far as soil building
goes.

From our Fairfield trip we came back into the Dutch Fork of Lexington County
and saw there the plat of Dr. J. L. Shuler, one of the most substantial citizens 'of the
State; and not only could we see with our own eyes that it had been entirely suc-
cessful, but Dr. Shuler assured us that it was successful beyond his expectations, and
as a result he has doubled his acreage this year. Mr. James W. Shealy, who is in charge
of this section of the county, informs me that his 3 acres of experimental work for
this past season has been increased to 17 for the coming season, which, in my mind,
is evidence not only of the success of the demonstrations, but of the interest of the
people in it, who are beginning to realize that they must grow more in connection
with cotton. We came back into the territory covered by our mutual friend,
Judge Derrick, and saw two of his demonstrations, one on the farm of Mr. G. W.
Caughman and one on that of Mr. D. J. Caughman. The former has his demonstra-
tion plat upon about as poor a piece of red Spanish oak land as you ever saw, and
notwithstanding this he got from it 2 tons of fine hay. Mr. D. J. Caughman has his
plat on better land, and yet not the best type of land by any means, and he tells me
that he cut his grass three times and gathered 3 tons to the acre. He is now grazing
his hogs upon it, and the clover is 6 or 8 inches high and practically green. The
cured hay is as fine as I ever saw.

What I have seen on this trip convinced me that my own idea of the feasibility of
growing grasses in this country, formed some 10 years ago, was entirely correct. We
can grow grasses without any doubt, and it is up to the Department of Agriculture
and Clemson College workers to lock arms and demonstrate that fact to the farmers
of the State. To my mind there can be no line of work undertaken that will mean
so much ultimately to the wealth of the State as to induce our farmers to grow their
own forage crops; and this, as you know, is unequaled as a soil builder and con-
servator. I think every student of the agricultural future of this State is agreed
upon the proposition that we must make ourselves, to as large an extent as possible,
a live-stock State. We must grow cattle; we must have some substantial crop with
which to reenforce our cotton crop; we must have some way of reducing the enor-
mous fertilizer bills of the State. We must find some means of producing enough
beef and dairy products in this State to save us the enormous drain because of our
deficiency hi this respect; and, to my mind, this grass work, if we can make it suc-
cessful, and I am sure it can be made successful with the proper efforts behind it,
will bring about the very condition of affairs which, I think you will agree, is so
necessary to the agriculture of the State.

I have taken the liberty of writing you at this length because of my very deep
interest in this work and because of the impression I formed of you personally of
your great desire to be of the greatest benefit to the people of the State.
With personal regards, I am, very truly,

A. F. Lever.

Lexinoton, S. C, November 22, 1912.

[Cir. 110]

SOME ASIATIC ACTINIDIAS. 1

By David Fairchild. Agricultural Explorer in Charge of Foreign Seed and Plant

Introduction.

ACTINIDIA ARGUTA.

At least five species of Actinidia are now growing in this country,
and they deserve much more consideration than has been given them.
There is no finer climbing shrub for porches in this latitude than
Actinidia arguta Miq. Its foliage seems to be practically free from
diseases, is of a beautiful dark-green color with reddish midribs, and
for situations where a mass of uncontrolled irregular foliage is
wanted and when a trellis can be provided, it is unusually successful.
It is a very vigorous grower and will cover a trellis 20 feet long and
10 feet high in two or three years.

Several years ago the writer had the pleasure of tasting a few
fruits picked from a vine of this species which covered over a hundred
feet of trellis on the house of Mr. Charles N. Parker, of Marblehead,
Mass. This vine had fruited more or less regularly since it was 9
years old, and it was at that time 20 years of age. The flavor of the
fruits was very sweet and pleasant, reminding one of figs. They were
about the size of damson plums, with very thin skins and filled with
extremely small seeds. A woodcut of the fruit of this species has
been published in Bailey's Cyclopedia of American Horticulture.

The value of this plant as a fruiting vine seems to have been little
emphasized in America. The long time required for it to come into
bearing and its dioecious nature, which has no doubt resulted in many
males being planted, has probably disheartened Americans from
growing it more extensively. From Asia, however, there have come
various reports about its usefulness and productiveness which
make it seem probable that there are better fruiting strains there
than we have in America. In northern Chosen (Korea) Mr. J. D>.
Hubbard, of the Oriental Consolidated Mining Co., of Unsa, informed
the writer in 1 909 that this species was known there as the tara, or
wild fig, and that it climbed 30 feet high and fruited profusely.

ACTINIDIA CALLOSA.

The species Actinidia callosa Lindl. is allied to A. arguta and
probably has not been grown outside of arboretums and botanic
gardens in America.

'Issued Jan. 18, 1913.
[Cir. 110] 7

8 CIRCULAR NO. 110, BUREAU OF PLANT INDUSTRY.

ACTINIDIA POLYGAMA.

In the Arnold Arboretum, vines of Actinidia polygama Miq. have
been growing successfully for years, and from one point of view, at
least, it is one of the most interesting plants in the arboretum. Its
leaves and twigs are so relished by cats that a wire cage has been
constructed about it, and evidently they still molest it, for wherever
the twigs or leaves come near enough to the wire netting they are
scratched and mutilated by the cats and cats' hairs are sticking to
the wires, which are evidences of efforts to get at the plants. Mr.
Jackson Dawson's observations indicate that the Boston cats learned
that this was good to eat a few months after it was first introduced.

ACTINIDIA KALOMIKTA.

Actinidia Icalomikta Ruprecht, which has also proved hardy in
Boston, although so far as known it has not fruited, is reported to
fruit well in the mountains near Merkoechofka, Siberia. The fruits
are dried by the Russian settlers and put aside for winter use in the
making of confectionery and to put in their bread. Mr. Frank N.
Meyer has described a variety of this species from Tungying, China,
which is shorter and shrubbier than the ordinary one. He found
the same species indigenous in the mountains of Okyansky in eastern
Siberia. In the mountains of northern Chosen (Korea), where he
saw it in August, 1906, he reported it to be a scant bearer.

ACTINIDIA CHINENSIS.

The most striking species of this interesting genus yet brought into
America is Actinidia cJiinensis Planch., known in China as the yang-
taw, and it is of very recent introduction. The attention of English
horticulturists was first called to it by Mr. Robert Fortune, who found
it when traveling hi behalf of the Royal Horticultural Society. Mr.
E. H. Wilson was the first to send seeds of it to Veitch & Sons, in
London, from his first expedition to western China. As early as
1900 seeds were sent to the Office of Foreign Seed and Plant Introduc-
tion from Ichang, but they failed to grow. In 1904, however, the
office received through the initiative of Consul General L. S. Wilcox,
of Hankow, a shipment of plants which had been carefully packed at
Chungking by Mr. E. H. Wilson, at that time engaged in an explora-
tion of Szechwan Province. The shipment arrived in excellent con-
dition, and as it is from these plants that the distributions through
the Southern States have been made, sufficient historical interest
attaches to it to warrant publishing Mr. Wilcox's letter in full.

[Cir. 110]

SOME ASIATIC ACTINIDIAS. 9

[No. 115.]

Hon. Francis B. Loomis,

Assistant Secretary of State, Washington, D. C.

Sir: I have the honor to inform you that last fall I obtained a sample of fruit called
by the Chinese "yang-taw." On investigation I learned that the original plants were
brought by Mr. Wilson, a botanist of Kew Gardens, London, from near the borders of
Yunnan in the foothills in the southern part of the province of Szechwan (a climate
similar to southern California). He sent some of the plants to England, where they
endeavored to introduce them, but found the climate unsuitable, being too cold,
except those that were planted in the Kew Gardens. The latter have proved a success.
The botanical name given them at Kew Gardens is Actinidia chinensis. The fruit is
about the size of a hen's egg,. has a thin, leathery, hairy skin covering it, and is full of
meat; seeds very similar to a gooseberry or fig. It is sometimes called the hill goose-
berry. In bloom it has a beautiful flower, and in my opinion it belongs to the same
family as the passion flower. When the fruits are picked and left for a few days until
soft they are very fine eating. They have the flavor of the gooseberry, fig, and citron.
They make delicious jam, pies, and sauce.

I asked Capt. Lovett, the Chinese Imperial Maritime Customs' Harbormaster here, if
it was possible to procure a few plants, as I would like to send them to the United States
Department of Agriculture. We supposed they had come from Ichang and would be
like currant shoots. He wrote to the party that sent the fruit (Mr. Goodhart, of Ichang)
who said he would try. After giving up all idea of receiving them, a box came two
days ago, weighing three or four hundred pounds, with the information that they had
been secured at Chungking (1.000 miles up river), from plants formerly obtained on
the borders of Yunnan by Mr. Wilson, under whose advice they have been packed in
moss and sand, warranted to keep for months. I felt I had a white elephant on my
hands; the bill for them has not yet been presented. I have inquired of several
southerners and none of them are acquainted with the fruit in the South, and it cer-
tainly would be a valuable fruit to introduce there. I think it worth the cost and
trouble of sending it. I will forward it to Shanghai and request Consul General
Goodnow to have it transferred to a Japanese steamer to Consul Harris, of Nagasaki,
and have requested him to put it on a transport for San Francisco, in care of Mr. William
A. Cooper, U. S. Despatch Agent, to hold it pending instructions from the State
Department as to where he is to ship it. I will also write Mr. Cooper to inform the
State Department of its arrival. I have packed in the box two bottles of the fruit that
I put up in alcohol last fall in order that the Agricultural Department may get some
idea of the two varieties that I received. The long variety seems to be the finer
flavored.

As the plants are carefully packed and strongly boxed, I trust that they may arrive
in good condition and that the fruit may be more valuable for family use than the
navel orange. At least I trust it will not be the case of a "mountain bringing forth a
mouse."

I have the honor to be, sir, your obedient servant,

L. S. Wilcox,
Consul General.

Hankow. China, March 21, 1904.

At the Plant Introduction Field Station at Chico, Cal., these plants
have flowered (PI. I) and have been propagated. Since 1904 the Office
of Foreign Seed and Plant Introduction has sent out 1 ,340 young plants
of this remarkable climber, and one of these flowered in 1910
at Durham, N. C. Unfortunately, however, the speeimens which
flowered at Chico turned out to be males, and therefore the plants
72780°— Cir. 110—13—2

10 CIRCULAR NO. 110, BUREAU OF PLANT INDUSTRY.

propagated and distributed are valuable only as ornamentals. Our
subsequent introductions of seedling plants have not as yet given any
indication of flowering. Doubtless there are females among them.
Repeated efforts to get authentic female plants of good varieties from
China have been unsuccessful, owing to the difficulty of getting living
plants properly packed and shipped from the interior. In the Gar-
deners' Chronicle for July 31, 1909, is an account of the first flowering
of this species in Europe. The flowers borne in 1909 at Nice, France,
were also all males. It is therefore too early to predict anything as
to the probable behavior and value of the species from the fruit-
culture point of view.

Reports have reached us that a specimen of this species has fruited
at Veitch & Sons' nurseries, near London, but no details have yet been
published, so far as the writer is aware, regarding the quality of
this fruit borne in England.

These facts are much to be regretted, since from Mr. Wilson's
descriptions the fruit must be excellent. In his report of 1898 he
describes the plant as "fruiting abundantly, bearing fruits 1 to 2\
inches long and 1 to If inches across (PI. II). Epicarp membranous,
russet brown, more or less clothed with villous hairs. Flesh green, of
most excellent flavor, to my palate akin to that of the common goose-
berry but tempered with a flavor peculiarly its own." 1

Dr. Samuel Cochran, of Hope Hospital, Hwaiyuan, Anhwei, China,
in a letter dated September 26, 1911, writes :

We find them delicious; they have something of the tart, rich flavor of the goose-
berry or strawberry. Not that the flavor is the same, but it has the same attractiveness,
being entirely free from mawkishness or insipidity. They make excellent preserves.
The flavor scarcely needs improvement. The fruit eats well raw after the skin is re-
moved, needing about as much sugar as strawberries.

Rev. William F. Beaman, of Kiatingfu, near Yachow, has reported
orally to the writer that the flavor of this fruit varies remarkably,
some tasting like a pineapple and others like a strawberry, confirm-
ing in this respect the experience of Mr. Leigh Hunt respecting the
Korean species (probably Actinidia arguta) . According to Mr. Hunt,
who spent many years in northern Chosen (Korea), one could find on
the same vine fruits of a flavor to suit everyone's taste.

Others have written enthusiastically of this fruit, and photographs
and alcoholic specimens show it to be one worthy of consideration.
The behavior of the plant solely as a climber warrants its wide distri-
bution, however, and it has already been taken up by at least one
American nursery firm. This species is not so hardy as the others
mentioned, being killed to the ground by severe freezes, but the
rapidity of its growth in the spring and the interesting character

1 Wilson, E. H. In Seeds and plants imported during the period from January 1 to March 31, "1908:
Inventory No. 14, U. S. Department of Agriculture, Bureau of Plant Industry, Bulletin 137, p. 14, No.
21781, 1909.

[Cir. 110]

Cir. 1 1 0, Bureau of Plant Industry, U. S. Dept. of Agriculture.

Plate I.

TO r*

03

a-?
■5 .

o o
°p

32

m o3

a a

oj O

. Ss

■" o
^ P

5
o
o
-I
CD

<

I-
i

o
z
<
>•

UJ

X
1-

u

z

11

i- - TO

Oft
ai-a^j

a — u

ssa

,^ St o

a° s

- .a
-o .^>

'OBI'
in 03 >

.5 Xl "3

t* o 5
03 -S 03

S-ci-a

•HpO
> *- fc-l

2 ftbC
2

Cir. 1 10, Bureau of Plant Industry, U. 5. Dept. of Agriculture.

Plate II

a

o

w

I ft)

— Soo

A J!

< -3°

> M

o %

ui "3 -

< 3S g

o

0>

B

O

SOME ASIATIC ACTIXIDIAS.

11

of its rapidly growing shoots will make it a desirable plant for trellises
and porches far north of the region where it can pass the winters unin-
jured (fig. 1). The remarkable plushlike texture of the leaves, their
unusual dark-green color, the bright-pink hairiness of the young
shoots, and the regular spacing of the leaves on the stem all contrib-

Fig: 1.— The yang-law on a wall trellis. Shoots produced before the middle of Julv bv a
vine of Ac/inidia chinensis Planch, which was killed to the ground by a temperature
of —17° F. the previous winter. Before autumn the trellis was almost concealed by
the foliage. The trellis is on the north side of the house, and the spaces of the trellis
are 10 inches square. Photographed at "In the Woods," Chevy Chase, Md., July

ute to make this climber especially suited to locations where broad
masses of green foliage are wanted. An old wall or outhouse, a
broad porch or pergola, where the plant is not confined, would be a
suitable place for this tit tractive vine. The Chico specimens
have flowered profusely. The male flowers, which are borne in
masses, are single white blossoms as large as a wild rose and with a
[Cir. 110]

12 CIRCULAR NO. 110, BUREAU OF PLANT INDUSTRY.

great quantity of yellow stamens (fig. 2). The petals fade to a pink
as they wither. The flowers have a delicate but peculiar fragrance.

Considering the wide range of climate to which the different spe-
cies of this genus of Asiatic plants is adapted, from the climate of
California to that of Maine, the thought has occurred to the writer

Fig. 2.— Male flowors of the yang-taw. The flowers are pure white when they open but turn pink
as they fade. The stamens are bright yellow. Natural-size photograph taken of flowers borne by
tlir original introduction of the species (Actinidia chinensis Planch.) into America (S. P. I. No.
1 L629) at the Plant Inl reduction Field Station, Chico, Cal., by Dr. Walter Van Fleet, April 18, 1910.

that the genus ought to be a promising one for the hybridizer to work
with. Whether the pollen of male plants of Actinidia chinensis in
California can be kept long enough to pollinate the female flowers of
A. arguta in Massachusetts can be easily determined, and. if not, the
cultivation of the different species side by side in regions where both
will flower is a possibility.

[Cir. 110]

POWDERY DRY-ROT OF THE POTATO. 1

By W. A. Orton, Pathologist in Charge of Cotton and Truck Disease and Sugar-Plant

Investigations.

INTRODUCTION.

In recent years a new potato disease, which has been named the
"powdery dry-rot/' has come to the front. It has caused heavy
losses in several Western States from Minnesota to Washington and
is a special menace to those irrigated districts where the potato is
one of the main money crops and where the product must be shipped
many hundreds of miles to reach a market. Several instances have
recently occurred where carloads of potatoes were shipped from the
Northwest to Texas points and to Chicago. Leaving their point
of origin in apparent good order they arrived at their destination
badly decayed, were rejected by the purchasers, and had to be con-
signed to the dump. The cause of this rapid deterioration was the
powdery dry-rot. Such experiences are exceedingly harmful to the
reputation of a new potato district. Buyers will not erect ware-
houses or provide shipping facilities for a permanent trade, nor will
they purchase for distant shipment save at the producers' risk. It
is therefore imperative that the growers take every possible means to
prevent the spread of this disease.

DESCRIPTION OF POWDERY DRY-ROT.

This disease is an external dry-rot. It may start at any point on
the outside of the tuber or gain entrance at the stem end. It starts
most readily in wounded potatoes, but may spread to uninjured ones.
The spots are wrinkled, discolored, and somewhat sunken, externally
darker brown than the normal epidermis, internally sepia brown, with
a dark, discolored layer next the sound flesh. In the later stages
the decayed portions become dry and powdery, with internal cavities
filled with fungous mycelium.

The cause of powdery dry-rot is a newly described fungus, Fusa-
rium tricJiothecioides Wollenw. 2

1 Issued Jan. 18, 1913.

J Jamieson, C. O., and Wollenweber, H. W. An external dry-rot of potato tubers caused by Fusarium
trichothecioides Wollenw. Journal, Washington Academy of Sciences, v. 2, no. 6, p. 14(>-152, Mar. 19, 1912.

[Clr. no] 13

14 CIRCULAR NO. 110, BUREAU OF PLANT INDUSTRY.

The disease is still too new and our experience too limited to permit
positive statements relative to means of control, but the following
advice is based on what is known of the life history of the causal
fungus.

POSSIBLE MEANS OF CONTROL.

Clean seed. — The powdery dry-rot is a storage trouble and appears
not to begin work till after the harvest, yet there is evidence that
land becomes infected through planting diseased seed potatoes and
that this contagion is carried over until fall and communicated to the
new crop. Consequently the most rigid inspection of the seed planted
is advised. No potatoes with a trace of decay should go into the
ground. All the seed stock should be disinfected by soaking for two
hours in a solution of 1 pint of formaldehyde in 30 gallons of water.
This will also kill the germs of potato scab.

Rotation of crops. — Rotation is necessary for the permanent success
of potato culture in any country, and particularly in irrigated dis-
tricts. While it is possible to produce two or three successive crops
on new land, it is unwise to attempt it on account of the disease factor.
One potato crop in three years has been proved too short a rotation in
many districts. One in five may serve the purpose in our Western
States. If a field has produced potatoes affected by powdery dry-rot,
it is especially desirable to rotate before planting potatoes there
again.

The storage problem. — Most important of all for the control of dry-
rot is the method of handling the potatoes after digging and during
storage, as it is probably here that the main fault lies.

Well-built storage cellars are a necessity from an economic stand-
point, to enable the grower to await better prices and to properly
assort and pack his crop. In such cellars, if properly built, venti-
lated, and wtached, the greatest immunity from decuy may be
insured.

These storage houses should be thoroughly cleaned at the end of
the season, all old potatoes and debris carried out, and the walls and
floors washed with a disinfectant like copper sulphate (blue vitriol),
using a 1 per cent solution, or corrosive sublimate, using 2 ounces to
16 gallons of water.

The greatest losses that have been brought to our attention have
occurred in that portion of the crop which was not put in storage or
shipped direct from the field to the market, but held for some weeks
after having been dug and sacked. Such potatoes, stored in sacks
in large piles and subjected to variable and occasionally rather high
temperatures, offer most favorable conditions for the development
of the dry-rot fungus.
[Cir. 110]

POWDERY DRY-ROT OF THE POTATO. 15

It is suggested that, when it is necessary to hold potatoes for a
time before shipping and it is not feasible to put them into a good,
cool storage cellar, the experiment be tried of leaving them unsacked
in the field in long, low piles or ricks, covered with sufficient earth to
protect them from sun and frost, and that they be sent from these
piles to market with as little intervening delay as practicable.

It is perhaps hardly necessary to warn growers that the dry-rot
parasite attacks most readily bruised or wounded potatoes and that
care in digging and handling, to leave them with nature's protecting
skin intact, is a great insurance against loss.
[CIr. 110)

PREPARATION OF LAND FOR EGYPTIAN COTTON IN THE
SALT RIVER VALLEY, ARIZONA. 1

By E.W. Hudson, Assistant in Arboriculture, Crop Physiology and Breeding

Investigations.

INTRODUCTION.

The Department of Agriculture has from time to time during the
past year sent out circular letters to the Egyptian cotton growers in
the Salt River Valley, calling their attention to various phases of
cotton culture and offering suggestions in handling the crop.

The harvest of the Egyptian cotton crop of 1912 grown in the Salt

River Valley is so nearly completed that the growers who have taken

good care of their crop are confident of a yield of from 600 to 700

pounds of lint per acre. Owing to the success of the present crop,

the growers are planning to increase their acreage and a number of the

farmers throughout the valley are planning to plant on a large scale.

From the present indications it is safe to estimate that from two to

three thousand acres of Egyptian cotton will be planted in the valley

in 1913.

SELECTION OF LAND.

Since it is very important that the land be prepared as early as
possible in the winter, it is well for those who are planning to plant
to give some thought at this time to the selection and preparation of
their fields. To secure the best crop of Egyptian cotton, it is of
very great importance that the grower select uniform land with very
slight grade on winch alfalfa has grown for at least three years.

The amount of fall or grade in the land is of great importance to the
cotton crop. To produce the best crop the land should be almost
level. In many instances where alfalfa is plowed up it is advisable
to irrigate in a different direction to secure a lighter grade. Land
with a slight grade will require less water for the crop and will irrigate
more evenly and produce a more uniform cotton. If there is much
grade to the land it will be found that the fields dry out hi spots dur-
ing August and September, and in order to avoid injury to the quality
of the cotton it will be necessary to irrigate these spots separately,
thus causing much extra work.

i Issued Jan. 18, 1913.
72780°— Cir. 110—13 3 17

18 CIRCULAK NO. 110, BUREAU OF PLANT INDUSTRY.

The question of the soil best adapted for cotton has been raised
many times. Any land that will grow good crops of alfalfa and grain
will also grow good cotton. The heavier class of soil will as a rule
grow a smaller and more fruitful plant with shorter nodes, and hence
more numerous fruiting branches. While some raw land will make
good cotton, it has been clearly demonstrated that land previously
in alfalfa will produce a higher grade of cotton and the crop can be
produced more economically, for new land as a rule requires more
frequent irrigations and it will be found necessary to irrigate hi spots
owing to the uneven way it will take water.

A great deal of the best land in the Salt River Valley is overrun with
Bermuda and Johnson grasses. The badly infested area comprises
much of the land that was under cultivation prior to the completion
of the Roosevelt Dam. There are, conservatively speaking, fully
10,000 acres of land on the south side of the river alone that would be
greatly benefited by cotton. This is the best land for cotton in the
valley, since most of it is level and rich from previous crops of alfalfa.

PREPARATION OF THE LAND.

In preparing this land for cotton the best plan would be to plow
about 2 inches deep during August, allowing the soil to dry out
thoroughly. This should be followed by thorough disking and har-
rowing, using a long-tooth harrow and dragging as many of the roots
to the surface as possible. A spring-tooth harrow might be used to
advantage hi this work. During November or December the land
should again be given a shallow plowing and pulverized by disking
and harrowing.

At this time it would be well to plow the land both ways with an
orchard cultivator or similar tool with long teeth. This work will
bring a great many of the roots to the surface, and if hi sufficient
quantity they should be raked up and burned or hauled off the field.
The land may then lie fallow until the latter part of February, when
borders should be thrown up about 2 rods apart. Just before planting
time, which is between March 10 and April 1, the land should be
flooded and then disked and harrowed until in perfect tilth. At this
time, if the soil is a very heavy clay, in some instances it may be
advisable to throw up beds from 4 to 4^ feet wide and 8 inches high,
with the idea of dragging off fully half of this height before planting.
A drag can be easily constructed of 2 by 4 inch or 2 by 6 inch scantling
that will take two or possibly three beds at a time. It should be
weighted down until it drags off enough of the surface clods to get
down to the moist soil. If a lighter soil is used it will not be necessary
to throw up beds. It is advisable to bed some classes of heavy soil
because it may be necessary to irrigate in order to germinate the seed,

[Cir. 110]

PREPARATION OF LAND FOR EGYPTIAN COTTON. 19

whereas a lighter soil may be pulverized sufficiently after a thorough
irrigation to hold enough moisture to germinate the seed without
further irrigation. Only as much land should be irrigated at one time
as can be prepared and planted before becoming too dry to germinate

the seed.

Seed must always be planted in moist soil. The seed may be
drilled in with a cotton planter, which will plant either one or two
rows. This may be either a one or two horse machine. If a single-
row walking planter is used it may be necessary to mark out the rows
on the bed before planting in order to keep the planter on the bed.

PREPARING BERMUDA AND JOHNSON GRASS LANDS.

While it may cost the grower from $8 to $10 an acre to put Bermuda
and Johnson grass land in shape for cotton, by intensive cultivation
during the first season the grass can be kept down, although it may be
found necessary to chop it out in the rows at the time the cotton is
being thinned. Bermuda grass alone is not so hard to eradicate as
Johnson grass or a mixture of Johnson and Bermuda grasses.

It is possible by shallow plowing during November or December,
followed by thorough disking and harrowing, to put Bermuda-grass
land in shape for cotton. If the land is kept thoroughly disked and
harrowed during the winter the freezing will greatly aid in killing the
roots. Two or three weeks before planting the land should be plowed
from 4 to 6 inches deep and thoroughly pulverized. If regular culti-
vation be kept up during the early part of the growing season or until
the cotton plants become large enough to shade the ground, the
Bermuda grass will not have a chance to establish itself enough to
become a nuisance. By such culture it is believed to be possible for
the growers to eradicate the grass entirely within two or three years,
growing a remunerative crop on the land.

PREPARING ALFALFA LAND.

In preparing for cotton alfalfa land which is not overrun with Ber-
muda or Johnson grass the same general plan could be followed.
However, it will not be necessary to do as much disking and harrowing
or cross plowing as in the case of land infested with Bermuda and
Johnson grass. If alfalfa land is plowed 2 inches deep and turned up
to the sun until thoroughly dried and then later in the season plowed
4 to 6 inches deep there will be very little trouble caused by alfalfa
during the following season. Alfalfa land may be prepared any time
prior to the planting season, but the best results will be obtained if the
land is plowed first in October or November, followed by a second
plowing in January.

[Cir. 110]

20 CIECULAE NO. 110, BUREAU OF PLANT INDUSTRY.

PREPARING COTTON OR GRAIN LAND.

In preparing cotton land for again planting to cotton a stalk cutter
should be used to chop the stems into small pieces; then the land
should be plowed, disked, and harrowed until in perfect tilth, when
it may be left until planting time. In the absence of a stalk cutter
the plants can be dragged down with a heavy drag after a hard
freeze. A great many of the stalks will be pulled out, and those
remaining in the ground must be dug up with a mattock. This
operation is inexpensive, costing only $1 or $1.25 per acre. After
all the plants are pulled out of the ground the field should be raked
crosswise with a hayrake and the stalks put into windrows, where
they can be easily burned.

Land previously in cotton or grain if irrigated before plowing can
be put in perfect condition with one plowing. Land previously in
alfalfa should be plowed twice. Last year one farmer plowed his
land and leveled it veiy poorly, and instead of disking several times
after the irrigation just before planting the seed made small furrows
and planted the seed with the idea of pulverizing the land when cul-
tivating. This piece of cotton had to be hoed twice and cultivated
several times more than a near-by field which was double disked
and harrowed until it was in perfect condition before the seed was
planted. A farmer planting cotton will make a very great mistake
if he does not thoroughly prepare his land before planting the seed.
It depends upon the kind of soil and the condition it is in whether a
double plowing is necessary to put the land in good tilth or whether
this can be done with a single plowing and double disking and double
harrowing. No amount of labor within reason expended during the
whiter in preparing the seed bed will be regretted, since if this is
thoroughly done a great deal less work during the summer will be
required to grow the crop, and the yield will be correspondingly
better.

[Cir. 110]

AGRICULTURE ON THE TRUCKEE-CARSON PROJECT. 1

VEGETABLES FOR THE HOME GARDEN.

By F. B. Headley, Superintendent, and Vincent Fulkerson, Scientific Assistant,

Western Irrigation Agriculture.

INTRODUCTION.

The notes on vegetables presented in this circular embody the
results of trials for three years on the Truckee-Carson Experiment
Farm. Cultural directions have been given only in part. Most
farmers have some knowledge of the growing of vegetables, and it is
therefore only the special or little-known methods that are suggested
here.

Many of the common vegetables are so easily grown on the Truckee-
Carson Project that every farmer should set aside enough land for a
vegetable garden, where sufficient produce can be raised to supply
the family throughout the season.

Among the most easily grown vegetables are asparagus, beans,
beets, carrots, cucumbers, eggplants, kohl-rabi, lettuce, muskmelons,
onions, peas, peppers, potatoes, pumpkins, radishes, squashes,
tomatoes, turnips, and watermelons.

A well-kept vegetable garden on the farm is one means of lessening
the cost of living, and during the summer it provides a large variety
of foods that a farmer could not otherwise afford, and even during
the winter, if proper care be taken to preserve them.

Asparagus, beans, cucumbers, peas, pumpkins, squashes, beets,
and tomatoes may be put up in glass jars and preserved satisfactorily. 2
Winter muskmelons, pumpkins, squashes, and many of the root
crops can be kept by storing them in pits or cellars.

CULTURAL DIRECTIONS.

Asparagus. — This is a perennial crop and should therefore be
planted in some portion of the garden where it will not interfere
with the easy cultivation and growing of other crops. The best
method of starting an asparagus bed is by planting 1-year-old roots,

i Issued Jan. 18, 1913.

2 See Farmers' Bulletin 359, entitled "Canning vegetables in the home."

[Cir. 110] 21

22 CIRCULAE NO. 110, BUREAU OF PLANT INDUSTRY.

but the plants may also be grown from seed. A very fertile soil is
essential and the crop should therefore be manured heavily each
year. It is better not to harvest the asparagus the first two seasons
after planting, for it is necessary that a strong root system be devel-
oped. Commencing with the third year, harvesting may begin.
Asparagus will grow in soils too alkaline for most other crops.

Beans. — As beans are tender annuals, they must be planted after
danger of frost is past, which on the Truckee-Carson Project is about
May 15. Over 20 varieties have been tried at the experiment farm
in comparative tests. As the result of these tests the following
varieties are recommended for home use:

Bush varieties of string beans, green Pole varieties of string beans:

podded: Kentucky Wonder Pole.

Early Yellow Six Weeks. Field varieties for dry shelled beans;

Burpee's Stringless. Colorado Mexican.

Bush varieties of string beans, wax Navy,

podded: Mexican Pinto.
Improved Golden Wax.
Dwarf Horticultural.
Davis White Wax.

Beets.— Beets are of easy culture and will grow in soil containing
considerable salt. The subsoil should be loose and mellow, as a hard-
pan tends to cause the roots to grow short and misshapen, with many
side roots. A late planting may be made between June 20 and July
15 for autumn and winter use. For fall planting a turnip-shaped
variety rather than a long-shaped, slow-growing variety should be
selected.

Carrots. — Carrots are somewhat more difficult to grow than beets.
The seeds are small and slow to germinate, so that it is important to
have a mellow, well-tilled soil — one that does not "bake" or crust
badly. Sometimes seeds of quick-maturing radishes are mixed with
the carrot seeds. The radish seeds, being relatively large and quick
germinating, break the crust and make it easy for the tender carrot
seedlings to come through. The soil for the carrots should be loose
and mellow, as a hardpan subsoil tends to make them grow short,
misshapen, and much branched. A late planting may be made dur-
ing the last half of June for autumn and winter use.

Corn. — Sweet corn does not usually produce well on desert soils,
but it grows well on old alfalfa land. The greatest difficulty in con-
nection with the growing of sweet corn is due to the prevalence of the
corn earworm, for which no effective remedy is known. It is essen-
tial that early varieties be selected. Peep o' Day, Golden Bantham,
and Early Minnesota are good varieties to use.

Cucumbers. — As cucumbers are sensitive to frost they must be
planted after danger of frost is past, unless arrangements are made

[Cir. 110]

AGRICULTURE ON THE TRUCKEE-CARSON PROJECT. 23

for covering and protecting them. They transplant with difficulty,
and it is therefore usual to plant the seed in the hills where they are
to grow. This crop is easily grown and does well on most soils of the
Truckee-Carson Project. The heaviest yielding and most satisfac-
tory varieties tried so far at the experiment farm are the Early Frame,
Long Green, Early Russian, and White Spine. The Gherkin is a
small, productive, prickly variety, useful only for pickling purposes.

Kohl-rabi. — This vegetable has been on trial at the experiment farm
for three years and has always given good results. The flavor of
kohl-rabi is between that of a turnip and a eabbagc. It is necessary
to use this vegetable while it is growing and tender, as when it is fully
matured it is likely to be tough and stringy. The culture is the same
as for turnips. Unharvested plants need not be destroyed in the
summer when they have become too tough for use, for if they are left
in the ground until fall a second growth takes place, new bulbs grow-
ing out of the old ones. This growth is as sweet and tender as the
spring growth.

Melons. — Watermelons are easily grown and are very productive
in most of the soils of the project. As the growing season is short,
only the earliest maturing varieties should be planted. Those which
have given best satisfaction at the experiment farm are the Kleck-
ley Sweet, Kentucky Wonder, Chilean, and Rocky Ford.

Muskmelons, like watermelons, are easily grown, especially on the
lighter soils of the project. Winter varieties, such as the Khiva,
Winter Pineapple, and Kriss Kringle, may be successfully grown.
These varieties, when stored in a cool place, will keep well into the
winter, so the farmer may have muskmelons on his table for Christ-
mas and New Year's Day. Desirable early-maturing summer varie-
ties are the Rocky Ford, Fordhook, and Extra-Early Hackensack.
Both watermelons and muskmelons may be planted in the field
before the danger of frost is past if boxes, guimy sacks, straw, or
some other suitable means of protection are thrown over them on cold
nights.

Onions. — On some of the better soils of the Truckee-Carson Project
onions are grown commercially and a yield of 10 to 20 or more tons
per acre obtained. It would not be advisable to attempt to grow
them commercially on new land. Local onion growers have found it
advisable to plant the seed very early in the spring, before April 1,
so as to allow the plants to become well established before the dry,
hot weather begins. Good varieties to plant for the home garden are
the Yellow Globe Danvers, Red Wethersfield, and Prize Taker.

Peanuts. — Peanuts have been grown for several years, but satis-
factory yields have not been obtained. The small Spanish peanuts
matured earlier and yielded better than the Mammoth Virginia pea-
nuts. Peanuts can not be recommended for general planting.

[Cir. 110]

24 CIRCULAR NO. 110, BUREAU OF PLANT INDUSTRY.

Peas. — Peas usually produce a good crop in the gardens over the
project, but sometimes fail in raw or alkaline soils. The Champion
of England and the Alaska were the best-yielding varieties tried at
the experiment farm in 1912. The seed should be planted very early
in the spring.

Potatoes. — Potato growing 1 is one of the recognized agricultural
industries of Nevada. Potatoes are best grown on river-bottom
land, tule land, or old alfalfa land. They do not always grow well
on the raw desert soils, which are deficient in humus and likely to
contain harmful quantities of alkali salts. The varieties which have
given best results at the experiment farm are the White Beauty,
Burbank, Triumph, Mammoth Pearl, Peachblow, and Early Ohio.
A variety test was conducted in 1911 by Wallace Ferguson in coop-
eration with the experiment farm. Eight varieties were included in
the test. The three heaviest yielding varieties were the White
Beauty, Mammoth Pearl, and Rose Seedling.

Pumpkins and squashes. — Pumpkins and squashes should be grown
in every farmer's garden, as they are prolific and are valuable for
table use. Any overproduction can be profitably fed to dairy cows
or hogs. Good varieties of pumpkins for table use are the Japanese
Pie, Small Sugar, and Cushaw. They are sweet and fine grained.
The large varieties, such as the Mammoth King and Connecticut
Field, are heavy producers, but they do not have the fine flavor and
quality for cooking of those mentioned above.

The most desirable summer squashes tried are the Yellow Summer
Crookneck 2 anjd White Bush Scallop, and the most desirable winter
varieties tried are the Warted Hubbard and Delicious. The Mam-
moth Chile was the heaviest yielding variety, but its quality is not
good for cooking purposes.

Tomatoes. — Tomato seed should re sown about April 15 in boxes
or hotbeds, so as to have good-sized plants by the time danger of
frost is past. The growing season is so short that an early start is
essential. Tomatoes usually produce satisfactory crops, but they
are subject to a disease known as the wilt (Fusarium sp.). This
disease attacks the individual plants in the tomato patch. The first
indication is a wilting of the leaves of the affected plants. The wilt
becomes more noticeable from day to day, until it finally causes the
death of the plants. No remedy is known, but damage from the
disease can be reduced by growing the plants on soil that has not
recently produced tomatoes. As the wilt is infectious, all diseased
plants should be pulled and burned as soon as they become affected.

i For the culture of the potato, see Farmers' Bulletin 3S6, entitled " Potato culture on irrigated farms of
the West;" or "Nevada potatoes," by C. A. Norcross, Nevada Bureau of Industry, Agriculture, and Irri-
gation, Bulletin 6, 1912.

2 Also known as the Golden Summer Crookneck.

[Cir. 110]

AGRICULTURE ON T1TE TRUCKEE-CARSON PROJECT. 25

Good varieties are Early Jewel, Dwarf Champion, New Stone,
New Coreless, New Globe, and Golden Qneen. After frost has
killed the vines the larger green tomatoes may be picked, each
wrapped in a piece of newspaper, packed in a shallow box and left
to ripen by storing in a closet or other suitable place where there
is no danger of freezing. By doing this, farmers may have ripe
tomatoes up to Thanksgiving time.

Turnips and rutabagas. — Turnips seeded in April are usable late
in June or early in July, but soon become woody and bitter. Ruta-
bagas are ready for use by the middle of July. For winter use ruta-
bagas should be seeded in June and turnips in early July. If seeded
earlier than this they become woody in the late fall. When the
greater part of the growth takes place in the cool weather of autumn
the quality is better and the size larger.

Wonderberry and garden Imckleberry . — The fruit of the wonderberry
is fairly agreeable when eaten raw, but is used chiefly in the making
of pies and jams. The vines are usually heavily loaded, but the
fruit is so small that it is somewhat difficult to gather. The plant
is an annual, but reseeds itself, so that volunteer plants come up in
the garden each year.

The garden huckleberry is advertised by some seedsmen as being
the same as the wonderberry, but the two plants are very different.
The garden huckleberry is an upright-growing, bushy plant, while
the wonderberry is decumbent. The fruit of the garden huckleberry
is two or three times as large as that of the wonderberry and when
raw is disagreeable to the taste. It must be cooked to be palatable.
After cooking there is practically no difference in flavor between the
two fruits. These two fruits do not deserve to be extensively grown,
but under conditions on the Truckee-Carson Project they are suffi-
ciently valuable to deserve a place in the family garden, at least
until the more' desirable small fruits have come into bearing.
[Cir. 110]

FUNGOUS STAINING OF COTTON FIBERS. 1

By Albert Mann, Plant Morphologist, Agricultural Technology and Cotton
Standardization and Grading Investigations.

OCCURRENCE OF THE DEFECT.

The manufacturers of certain grades of cotton fabrics are occa-
sionally annoyed by the appearance of bright-red or deep purple-
blue threads appearing in white cotton cloth, thereby greatly lower-
ing its commercial value. The number of such threads in a square
yard is extremely small, but the brilliant colors make these threads
very prominent.

The American Cotton Yarn Co. has sent to the Office of Agricul-
tural Technology and Cotton Standardization and Grading Investi-
gations two sets of manufactured samples showing these defects,
with a request for information as to the cause. Inquiry makes it
practically certain that the fibers are not colored in the process of
manufacture. Careful microscopical examination of the colored
fibers gave no indication of bacteria or of fungous growth. The
colors were very resistant to alkalis of moderate strength, as well as
to bleaching agents.

PROBABLE DISCOVERY OF THE CAUSE.

Recently a boll of cotton was obtained from Diamond Spring, Va.,
containing a quantity of lint that was brilliant carmine red. Exam-
ination showed that this was due to the presence of a fungus, which
was identified by Dr. H. W. Wollenweber, of the Office of Cotton
and Truck Disease and Sugar-Plant Investigations, as belonging to
the genus Fusarium. The fungus was very evident, completely
investing the cotton fibers and even penetrating their internal
cavities. It was in fruit, and the sickle-shaped compound spores of
this genus were abundant.

A Fusarium grows upon various host plants in the United States,
including the cotton plant, which under alkaline conditions may have
either a strong carmine-red or an intense violet-purple pigmentation.
It is Fusarium metaclwoum App. and Wr. Unlike some blue and red
pigments in plants, the two colors are not convertible into each

i Issued Jan. 18, 1913.
[Cir. 110] 27

28 CIRCULAR NO. 110, BUREAU OF PLANT INDUSTRY.

other by the alternate action of alkaline and acid reagents, though
both result in a yellow color under acid modification.

This agrees so closely with the facts observed in the colored fibers
of the cotton cloth that it seems probable these colored fibers are
caused by growths of fungi, and probably of the species mentioned,
occurring upon cotton exposed to excessive dampness in the field or
picked in an unripe condition. The reason the fungus was not
discovered on the manufactured fibers is probably due to the mechani-
cal and chemical processes through which it passed. The samples of
cloth submitted had been mercerized. The external hyphae of the
fungus would probably be removed in the various processes of
ginning, spinning, • and subsequent treatment. The absence of
hyphae within the fibers may have been because of no penetration
having taken place in the specimens examined, or possibly because
of some obliterating effect of the process of mercerizing the cloth.

The manufacturers are now preparing to adopt some means of
eliminating these undesirable samples of cotton from the raw material.

U'ir. 110]

Cir. 1 10, Bureau of Plant Industry, U. S. Dept. of Agriculture.

Plate III.

Pods and Seeds of the Jack Bean and the Sword Bean.

Central figure, pod of a jack bean (S. P. I. No. 21609) from Brazil. Right-hand figure, a pod of a
sword bean (S. P. I. No. 27676) from India. Left-hand figure, pod of a sword bean (S. P. I No
27704) from China.

THE JACK BEAN AND THE SWORD BEAN. 1

By C. V. Piper, Agrostologist in Charge of Forage- Crop Investigations.

THE JACK BEAN.

During the past few years the jaek bean (Canavali ensiformis) has
attracted a good deal of attention in Texas and other Southern
States. Many inquiries have been received concerning its agricul-
tural value, as under ordinary conditions it grows vigorously and
produces large yields of beans.

DESCRIPTION.

The jack bean is a bushy, semierect annual plant, growing to a
height of 2 to 4 feet. Its stems are rather coarse and become woody
toward the base. The rather thickish leaves have a decidedly bitter
taste. The flowers are purple, at least in all varieties so far intro-
duced. The first blossoms are borne near the base of the stem, so
that many of the pods hang low. When mature the pods are hard
and firm, 9 to 14 inches long, each containing 10 to 14 seeds. These
are pure white, with a brown hilum. Ordinarily the roots are well
tubercled, and the plant will withstand much drought. It is re-
markably free from insects and fungous diseases and but slightly
affected by root-knot. (PI. III.)

HISTORY.

The jack bean is a native of the West Indies and the adjacent
mainland. In Jamaica, whence it first became well known, it is
called the horse bean or the overlook bean. The horse bean of
Europe is a very different plant. In Antigua it has been called the
Babricou bean, and in this country has been designated the Pearson
bean, and recently the wonder bean.

Owing to confusion with the similar species cultivated in Japan,
China, and India, it has also been called the sword bean and the
knife bean, but those names properly belong to the Asiatic species.

The first published description of the plant is by Clusius 2 in 1605,
whose figures of the pods and seeds received from Brazil are unmis-

« Issued Jan. 18, 1913. 2 Clusius, Carolus. Exoticorum, p. 09, 1005.

[Cir. 110] 29

30 CIRCULAR NO. 110, BUREAU OF PLANT INDUSTRY.

takable. Clusius, however, describes the seeds as brownish, but all
subsequent authors state that the seeds are white, which is the case
in all that have been grown in this country. Sloane x gives a good
figure and description of the plant in 1707 under the name "horse
bean," as he found it in Jamaica. He writes further:

They are eaten as other Phaseoli by some and counted good food, though their
greatest use is to fatten hogs.

Fifty years later Browne wrote of the horse bean as follows :

This plant grows in many gardens in Jamaica, where it is cultivated chiefly out of
curiosity. It seems to keep a main between the upright and the climbing species of
Phaseolus, for the stem seldom rises above 3 or 4 feet, though it emits some slender
delicate shoots that run much further. The pods are commonly between 10 and 14
inches in length and generally contain 10 or 11 seeds, but the pulse is very seldom
used, being generally thought more or less of a deleterious nature. 2

Macfadyen, writing in the Flora of Jamaica in 1837, records as
follows:

Sloane considers this species to be indigenous to the island and says the seeds were
in his time used by some as food and given to fatten hogs. I do not find, however,
on inquiry that any use is made at present of them, except that they are commonly
planted by the negroes along the margin of their provision grounds from a supersti-
tious notion, probably of African origin, that the overlook fulfills the part of a watch-
man and, from some dreaded power ascribed to it, protects the provisions from plunder.
Even the better informed adopt the practice, though they themselves may not place
confidence in any particular influence this humble plant can exercise, either in pre-
venting theft or in punishing it when committed. 3

The same idea prevails at the present day in Panama.

BOTANICAL RELATIONSHIPS.

The j ack bean was briefly described and named DolieJios ensiformis
by Linnaeus. 4 His description is based primarily on the previous
account and illustration of Sloane. In later descriptions, however,
Linnaeus also included under his species various plants described by
other authors. In 1759 he redescribes his Dolichos ensiformis, 5 citing
as its basis the plate and description of an Amboyna plant. 6 Linnseus
describes the plant as twining, which accords with the descriptions
of both Sloane and of Rumphius. In 1763 Linnaeus 7 again describes
the plant under the same name, but states that the plant is erect.
Besides the plants of Sloane and of Rumphius he includes the Indian
plant described and illustrated by Rheede. 8

1 Sloane, Sir Hans, bart. A Voyage to the Islands Madera, Barbados, Nieves, S. Christophers, and
Jamaica, with the Natural History, v. 1, p. 177, 1707.

2 Browne, Patrick. Civil and Natural History of Jamaica, p. 291, 1879.

3 Macfadyen, James. Flora of Jamaica, p. 292, 1N37.

* Linnaeus, Carolus. Species Plantarum, t. 2, p. 725, 1753.

6 Linnams, Carolus. Systema Naturae, ed. 10. 1. 2, pt. 2, p. 1102, 1759.

6 Rumphius, G. E. Herbarium Amboinense, pars 5, pi. 142, 1747.

7 Linnseus, Carolus. Species Plantarum, ed. 2, t. 2, p. 1022, 1763.

s Rheede tot Draakenstem, H. A. van. Hortus Malabaricus, pars 8, pi. 44, 1688.

[Cir. 110]

THE JACK BEAN AND THE SWOBD BEAN. 31

Finally in 1767 Linnaeus J gives two separate descriptions of Doli-
chos ensiformis, in one of which the plant is said to be erect, in the
other twining. Whether this was an editorial blunder or whether
Linnaeus considered the two distinct and inadvertently used the same
specific name twice is open to question. Jacquin evidently thought
that the same name covered two species, as the twining one agreed
with a plant growing m his greenhouse at Vienna which he had
obtained from the West Indies. He redescrlbed and figured it as
Dolichos acinaciformis. 2 There can be scarcely a question, however,
that these two names include but a single species. It is true Sloane
records the plant as twining. Brown says it is suberect, while Mac-
fadyen writes "at first suberect, afterward twining." Two varieties
were grown at Biloxi, Miss., and Gainesville, Fla., in 1911 and 1912,
one early, the other later and larger. The former is suberect and
bushy, while the latter has longer, viny branches. Under favorable
conditions it would not be surprising for these plants to assume a
vining habit, as a similar phenomenon is well known in the case of
bush cowpeas and soy beans.

The plant of Rheede is a distinct species, Canavali gladiata, and
that of Rumphius is probably the same, but with a faulty figure.
Later authors have also confused this East Indian plant with the
West Indian, so that it is often difficult to determine which species
is referred to.

The genus Canavali founded by Adanson 3 has received general
recognition, and the jack bean is therefore properly known as Canavali
ensiformis (L.) DC. Later authors have usually modified the generic
name to Canavalia.

ECONOMIC VALUE.

In the last 20 years the jack bean has several times attracted atten-
tion on account of its vigorous growth and large yield of pods and
seeds. It was extensively tested at the Mississippi Agricultural
Experiment Station during the years 1890 to ,1895. Under field con-
ditions yields of 30 to 40 bushels of beans per acre were obtained, even
when grown on thin soil. Attempts were made to utilize these beans
as feed for both beef and dairy cattle, 4 but the beans were found to
be both unpalatable and indigestible.

Three of the lots fed with bean meal made a smaller gain than did many of the lots
receiving cottonseed meal, and the best of the bean-meal lots, No. 4a, gained only 3.4
pounds more than did the poorest of the lots receiving cottonseed meal. It should be
noted that the lots receiving bean meal were also fed cottonseed meal amounting to

1 Einna-us, Carolus. Systems Naturae, ed. 12, t. 2, p. im' 483, 1707.

2 Jacquin, N. J. von. Collectanea, t. 1, p. 114, 1786.

3 Adanson, Michel. Families des Plantes, pt. 2, p. 325, 1763.

4 Lloyd, E. R. , and Moore, J. S. Feeding for beef. Mississippi Agricultural Experiment Station, Bul-
letin 39, p. 162-163, 1896.

[Cir. 110]

32 CIRCULAR NO. 110, BUREAU OF PLANT INDUSTRY.

about one-third of their grain ration, and a large part of their small gain should doubt-
less be credited to the cottonseed meal and not to the bean meal.

The complete failure to secure profitable results from the use of the bean meal was a
surprise and disappointment. Various methods of feeding were tried, both coarse and
fine meal being used, and during a portion of the time the meal was cooked until it
was thoroughly softened. At first very few of the steers would eat any of the meal,
but were finally induced to do so by mixing it with salt and cottonseed meal, so that
when the trial feedings began all ate it fairly well, though not with much apparent
relish. The meal which was eaten appeared to be very indigestible for all the ani-
mals, and the same was found to be the case when it was fed to milch cows, as will be
shown in another bulletin, soon to be published.

When tried on the table of one of our station stall', the beans were of fair quality,
though rather coarse, and no one of the six persons who ate them experienced any ill
effect from them.

If the different feeds are considered with reference to only single ingredients of the
several rations, the average gains per steer were as follows:

Pounds.

Shredded corn and silage 35. 1

Crab-grass hay 51.5

Pea- vine hay 00. 2

Pounds.

Red-clover hay 71.7

Cottonseed meal 02.

Bean meal 28. 4

Seeds of the bean were distributed by Mr. P. Pearson, of Starkville,
Miss., from which fact it became known as the Pearson bean. At the
Texas Agricultural Experiment Station it produced 35 bushels per
acre. 1 At the North Carolina Agricultural Experiment Station it
produced an estimated yield of 40 bushels per acre. 2 It was also
tested at the Louisiana Experiment Station. 3 None of these sta-
tions regarded the bean as promising, but, so far as recorded, no
attempt was made to utilize either the herbage or the seeds as forage.

More recently the plant has been tested in Hawaii, and favorable
reports as to its forage value have been published. 4

While grown to some extent in the Southern States, the plant does not appear to
thrive as well there as here, and no extensive feeding experiments are reported. The
bean meal is said not to be very palatable or digestible for cattle, but this may be due
to a too-limited experience in its use. The early feeding experiments with the green
fodder in Hawaii gave similar results to those reported above, but as feeders gained
in experience the fodder was found to be both palatable and nutritious for dairy cows
as well as swine. As with most new feeds, it is important to use in the beginning
only a small proportion of the new feed in the accustomed ration and then increase the
proportion gradually. The Dowsett and Pond dairies have fed green jack beans and
sorghum in equal proportions to dairy cows with excellent results.

i Connell, J. H., and Clayton, James. Field experiments at MeKirmey, Wichita Falls, and College sta-
tion with wheat, corn, cotton, grasses, and manures. Texas Agricultural Experiment Station, Bulletin
34, p. 584, 1895.

2 McCarthy, Gerald. Some new forage, fiber, and other useful plants. North Carolina Agricultural
Experiment Station, Bulletin 133, p. 343, 1896.

3 Dodson, W. R., and Stubbs, W. C. Grasses, clovers, forage, and economic crops. Louisiana Agri-
cultural Experiment Station, Bulletin 53, p. 42, 1898.

4 Krauss, F. G. Leguminous crops for Hawaii. Hawaii Agricultural Experiment Station, Bulletin 23,
p. 19-21, 1911.

[Cir. 110]

THE JACK BEAN AND THE SWORD BEAN. 33

The crop requires about a month longer to mature than do cowpeas, but the yield
is proportionately greater. Yields of 16 to over 20 tons of green fodder per acre have
been reported from various sources. The best yield of seed reported is 1,200 pounds
per acre. While a single crop is usually grown from each sowing, the station has occa-
sionally grown a good rattoon crop. Such crops, however, are subject to a leaf blight
common to the bean family. Otherwise the crop is exceptionally free from diseases
and insect pests, a point greatly in its favor over the cowpea. Another possible advan-
tage possessed by the jack bean over the rambling legumes is the absence of trailing
stems, which might interfere in some forms of intercropping.

While the crop is quite drought resisting, as was shown in the excellent yield pro-
duced at Kunia during the dry season of 1909, it responds well to irrigation and makes
a good growth during the wet season if the weather is not too cold. The jack bean
develops a strong root system . The roots are nearly always well supplied with the
nodules produced by the nitrogen-fixing bacteria, so that the stubble remaining after
the crop is harvested should prove beneficial to the soil. The best method for the
culture of the crop, whether it is to be used for green fodder or seed, is to plant in
rows and cultivate freely throughout the growing season. .For the largest amount of
green matter, plant the seed 3 to 6 inches apart in rows 2 to 3 feet apart. If seed is
the object, especially if wanted for planting, the rows should be at least a foot farther
apart and the seed planted 6 to 10 inches apart in the row. Forty to sixty pounds of
seed will be required to plant an acre. The optimum amount of moisture for seed
growing is about two-thirds that required for maximum fodder production.

In Porto Rico the jack bean has been found very useful as a green-
manure and cover crop in citrus groves. Judging from the behavior
of the plant in experimental trials in Florida, it should prove equally
valuable there. Its bushy habit makes it especially desirable, as it
does not interfere by climbing the trees, while its dense, vigorous
growth shades the ground during the heat of summer and provides
abundant vegetable matter to add to the soil.

The value of the plant as forage is yet problematical. Its successful
utilization as green feed in Hawaii encourages the belief that it may
be found equally valuable in this country, especially in Texas and
Oklahoma, where its great drought resistance gives it particular
promise.

The large yield of seed per acre justifies further experiments to
determine whether any means can be devised to utilize the seeds
profitably as feed, which the work referred to of the Mississippi Agri-
cultural Experiment Station indicates is a difficult problem. The
jack bean has recently been introduced into Java, where on account
of the large yield of seed the agricultural authorities were endeavoring
to find a market for the product in Europe.

There is also the probability that the jack bean may prove to be
valuable as ensilage. Its coarse habit and heavy tonnage should
adapt it well to this purpose.

[Cir. 110]

34

CIRCULAR NO. 110, BUREAU OF PLANT INDUSTRY.
Table I. — Analyses of the jack bean.

Material.

Water.

Protein.

Fat.

Nitrogen-
free
extract.

Crude

fiber.

Ash.

Publication.

Seeds

Per cent.

11.48

76.81

Per cent.
26.85

2.44
17.76

23. 75
5.21

Per cent.
2.99

.52
3.06

2.65
.48

Per cent.
56.90

48.51

65. OS

50.37
8.44

Per cent.
39. 90

43.89
19.31

8.75
6.36

Per a nt.
3.38

4.64
3.79

3.00
2.70

Miss. Exp. Sta.

Hulls

Ann. Rep. 8, p.
43, 1895.
Do.

Whole pods with

seeds.
Bean meal

Do.
Miss. Exp. Sta.

Green plant

Bui. 39, p. 159,
1896.
Hawaii Exp. Sta.

Bui. 23, p. 31,
1911.

THE SWORD BEAN.

The sword bean (Canavali gladiata), also known as the knife bean
and the saber bean, is closely related to the jack bean and the two
have been much confused. The sword bean is found cultivated
through much of southern Asia and also in Africa. At various
times it has been introduced into America, but is still cultivated almost
entirely as a curiosity or as an arbor vine.

HISTORY.

The sword bean was briefly described and well illustrated by
Rheede x under the name "Bara-mareca," the plants being found at
Angiecamal and other places on the southwest coast of India as a
cultivated vegetable. The color of the beans and seeds is not
stated. What is quite certainly the same species is described and
figured by Rumphius 2 under the name Lobus machaeroides, the
native name in Amboyna being Cacara parrang. The figure of the
pods is, however, faulty. Rumphius describes the seeds as either
intense red or else fuscous and states it is rarely cultivated in Amboyna
but abundantly in Java and Baleya. Linnaeus erroneously supposed
the plants of Rheede and of Rumphius to be the same as his Dolichos
ensiformis, the jack bean. Jacquin, however, grew the sword bean
in the greenhouse of the botanical garden at Vienna and in 1788
described it fully under the name Dolichos gladiatus, 3 later publish-
ing a beautiful colored plate. 4 The variety he grew had dark-red
seeds and white flowers, which later became suffused with pink.
The source of the plant is not stated.

Later authors have greatly confused Canavali gladiata and Cana-
vali ensiformis, but both the pods and seeds of the two are very

i Rheede tot Draakenstein, H. A. van. Hortus Malabaricus, pars 8, p. 85, pi. 44, 1688.

2 Rumphius, G. E. Herbarium Amboinense, pars 5, p. 376, 1747.

a Jacquin, N. J. von. Collectanea, t. 2, p. 276, 1788.

4 Jacquin, N. J. von. Icones Plantarum Rariorum, v. 3, pi. 560, 1786-1793.

[Cir. 110]

THE JACK BEAN AND THE SWORD BEAN. 35

different (PL III). According to Roxburgh four varieties of the
sword bean are found in India, one having red seeds and red flowers,
a second with red seeds and white flowers, a third with white seeds
and white flowers, and a fourth with light-gray seeds and red flowers.
Three of these varieties are now introduced into the United States.
All of them are supposed by most Indian botanists to be derived
from a brown-seeded wild plant, Canavali virosa, the seeds of which
are reputed poisonous. Macfadyen reports the variety with red
flowers and red seeds as cultivated in Jamaica in 1837. This variety
does not fully mature as far north as Washington, D. C, but does
produce an abundance of green pods in late September and early
October. It seems well worthy of attention as a vegetable through-
out the Southern States.

ECONOMIC VALUE.

The sword bean is commonly cultivated as a vegetable in Japan,
India, Burma, Ceylon, Java, Mauritius, and apparently in Africa.
In India it is eaten both by the natives and by Europeans, the
variety with white seeds being .most esteemed. The young pods are
prepared after the manner of snap beans and are well flavored and
wholesome. Firminger considers it "about the nicest of all the
native vegetables" in India. The very young pods have but little
flavor, but when half grown their taste suggests mushrooms. They
are best when about half grown, as the full-sized green pods are
rather fibrous. The mature seeds do not seem to be much used as
food, though they lack the strong odor of those of the jack bean.

An extended account of the sword bean as grown in Mauritius has
been published by Boname, 1 but under the erroneous name Canavali
ensiformis. This author highly recommends the green pods as a
vegetable and gives analyses showing the composition not only of
the entire plant but of the beans and pods, both green and ripe.

In China and Japan occur two varieties of the sword bean which
differ from those of India in that the pods are thicker and the seeds
much plumper and distinctly keeled on the back (PL III). This form
was described by Thunberg 2 as Dolichos incurvus and by Petit-
Thouars as Canavali incurva. The Japanese name is Nata mame
and the Chinese Tau tou. One variety has the seeds cream colored,
another pink. In Japan "the young pods of the former are pre-
served in salt, and the latter is eaten fresh and boiled." 3 Rev. E. R.
Miller, of Morioka, Japan, in sending a sample of seed (S. P. I. No.
6132) writes "This as a string bean eaten when very young is one of

1 Boname, P. The sword bean. L'Agriculture Pratique des Pays Chauds, arm. 10, sem. 2, p. 371, 1910.

2 Thunberg, K. P. Flora Japonica, p. 280, 1784.

3 Agricultural Society of Japan. Useful Plants of Japan, [v. 1], p. 9, 1S95.

[Cir. 110]

36 CIRCULAR NO. 110, BUREAU OF PLANT INDUSTRY.

the finest I ever tasted. * * * The Japanese use them generally
for pickling when young and they are very fine for this purpose."
In China the pink-seeded form (S. P. I. No. 23216) is said by Mr.
Frank N. Meyer to be "a very rare bean used mainly as a stomach-
strengthening food, and for this reason to be had only in medicine
shops." Brill, under S. P. I. No. 6570, also records that these "beans
are very good but expensive." As the plant yields heavily, it is
difficult to understand why the beans should be expensive.

VALUE FOR FORAGE AND AS A COVER CROP.

All the varieties of the sword bean tested are rambling vines, none
of them being bushy like the jack bean. As forage they are not as
desirable, as the foliage is just as bitter and the habit inferior.

As a cover crop the Indian variety with red seeds and red flowers
has proved very satisfactory in Porto Rico. Cattle are said to graze
on the plant there to a limited extent.

The sword bean is not infrequently seen in the South, grown as
an arbor plant, but little seems to be known of the value of the green
pods as a vegetable. Indeed, the impression prevails that the seeds
are deleterious, an idea doubtless obtained by association with the
very similar jack bean, as the sword bean is everywhere utilized as
a vegetable in the Orient. The plant will develop full-grown green
pods as far north as Washington, D. C, but ordinarily the season is
not long enough for the seeds to ripen.

[Cir. 110]

FIBER FROM DIFFERENT PICKINGS OF EGYPTIAN COTTON. 1

By Thomas H. Kearney, Physiologist in Charge of Alkali and Drought Resistant Plant

Investigations.

INTRODUCTION.

Two pickings were made from the individual selections of the
Yuma variety of cotton at Sacaton, Ariz., in 1911, the first early in
October, the second about a month later. The first picking consisted
largely of the first bolls which open on the plants, these being borne
principally on the lower fruiting branches of the main stem. The
development of these lower bolls is frequently checked by the heavy
shading to which the lower part of the plant is subject in this variety.
Practically all of the bolls included in the first picking had matured
and opened during the hot weather in September. The second, or
November, picking took in the cotton from bolls which were more
favorably situated on the plants (nearer the tops) and which had
matured during the cooler weather of October.

At Bard, Cat., in 1911, the first cotton in the progeny rows of the
Yuma variety ripened so early that these rows were picked in bulk
about the middle of September, before individual selections had been
made, thus eliminating the cotton which was included in the first
individual picking at Sacaton — that from the early-opening bolls
near the bases of the plants. For this reason, it would be expected
that the cotton of the two pickings made from the individual
selections would show less difference at Bard 2 than at Sacaton. A
comparison of the fiber from the two pickings of each individual plant
at Sacaton and at Bard, respectively, proved that this was the case.
No differences in length, strength, and abundance (lint index) of fiber,
or in weight of seeds, could be detected in samples from the two pick-
ings at Bard. The conclusions herein reached are therefore based
solely upon the material collected at Sacaton.

i Issued Jan. 18, 1913.

2 At Bard the first picking of individual selections was made about October 15, the second at the end of
November.

[Cir. 1101 37

38 CIRCULAR NO. 110, BUREAU OF PLANT INDUSTRY.

COMPARISON OF THE DIFFERENT PICKINGS.
LENGTH OF STAPLE.

In 33 out of the total number of 60 individual selections made at
Sacaton, careful measurement and comparison of the length of staple
in the first and in the second picking, respectively, were made. 1 The
fiber from the second picking of these 33 plants averaged one-sixteenth
of an inch longer than that from the first picking. The uniformity
of length was also greater hi the second picking. In 42 selections
in which the variation in length was recorded, it averaged three-
sixteenths of an inch hi the first picking and two-sixteenths of an inch
in the second picking. In 30 of these plants the variation was greater
in the first picking, hi 6 plants it was greater in the second picking,
and in the remaining plants it was the same in both pickings.

STRENGTH OF FIBER,

In 44 individual selections hi which the first and second pickings
were compared in respect to strength, 27 showed a greater strength
of fiber in the second picking than hi the first, 13 showed no marked
difference between the two pickings, and only 4 showed stronger fiber
in the first picking.

LINT INDEX.

The fiber was decidedly more abundant in the second picking than
in the first, the average lint index (weight in grams of fiber per 100
seeds) in the 60 individual selections having been 4.90 for the first
picking and 5.17 for the second picking.

FINENESS OF FIBER.

It has been observed every year during which breeding work with
Egyptian cotton has been carried on in the Southwest that the fiber
from the bolls which open first is often coarser and harsher to the
touch than the fiber produced in bolls higher on the plant, which
ripen later under the influence of more equable temperatures.

WEIGHT OF SEEDS.

The average weight of 100 seeds in the 44 individual selections in
which the two pickings were compared in this respect was 12.9 grams
for the first picking and 13.4 grams for the second picking, indicating
that the seeds produced in the bolls which ripen earliest are decidedly
lighter than those hi later ripening bolls borne higher on the plant.

1 The measurement of length in each of the two pickings was made as follows: From the sample of seed
cotton representing each picking from a given individual plant, 10 seeds were drawn at random from different
portions of the mass. A cluster of fibers from near the middle of each seed was pulled out and the average
length of these fibers was estimated. The mean of the average lengths of the fiber from the 10 different
seeds was taken as representing the length of staple for the picking in question.

[Cir. 110]

FIBER FROM DIFFERENT PICKINGS OF EGYPTIAN COTTON. 39

FUZZINESS OF SEEDS.

There was some indication that the seeds from the first picking had
a tendency to develop a rather larger quantity of fuzz on the seed
coat than was the case in the second picking, but it is by no means
certain that any noteworthy or constant difference in this respect
exists.

CONCLUSIONS.

It is clear from the foregoing that the earliest ripening bolls (chiefly
those near the base of the plant), which open while extremely high
temperatures still prevail, are likely to contain less abundant, shorter,
weaker, coarser, and less uniform fiber than bolls which ripen later.
The cotton from the earliest picking is also likely to contain more
dust and "trash," since the bolls which produce it are closer to
the ground and since the leaves and bracts around them have often
dried up and become very brittle before the picking is made. For
these reasons it would seem advisable, in order to maintain the high-
est possible standard for Egyptian cotton grown in the Southwest, to
make the first picking as early as the number of open bolls will war-
rant the expense of the operation and to sell the fiber from this pick-
ing as a separate grade, not allowing it to become mixed with the bulk
of the crop, which may be expected to show decided superiority in
grade and quality. It is also probable that the latest cotton to ripen,
especially that contained in bolls which open after a severe frost,
should also be graded separately. The investigation here described
does not, however, throw any light upon this latter point. 1

Since the bolls which ripen first have often cracked open prema-
turely under the influence of the high temperatures and extremely
dry winds which usually characterize the month of September in the
Southwest, the seeds contained in them are frequently not thoroughly
mature. This is indicated by their smaller average weight and
in fact is evident to the eye from their frequently lighter color.
Although no experiments have yet been made to test the point, it is
not unlikely that these seeds will be found to give a lower percentage
of germination and to produce less vigorous plants than seeds from
the second picking. In such cases it would be advisable to use the
latter as much as possible for planting.

1 It is interesting to note in this connection that spinners and buyers to whom samples of the 1910 crop
of Arizona-grown Egyptian cotton were submitted found the fiber to be weaker in the sample which repre-
sented the fourth picking than in samples which represented earlier pickings from the same fields.
LOir. 110]

o

U. S. DEPARTMENT OF AGRICULTURE.

BUREAU OF PLANT INDUSTRY— Circular No. 111.
B. T. GALLOWAY, Chief of Bureau.

MISCELLANEOUS PAPERS.

Preliminary Report on Sugar Production from Maize

Durango Cotton in the Imperial Valley

Improved Apparatus for Detecting Sulphured Grain

Supernumerary Carpels in Cotton Bolls

Keeping Soft Cuttings Alive for Long Periods .

C. F. CLARK

0. F. COOK

G. H. BASTON

R. M. MEADE

G. W. OLIVER

Issued February 1, 1913.

WASHINGTON:

GOVERNMENT PRINTING OFFICE.

1913.

[Cir. Ill]
2

BUREAU OF PLANT INDUSTRI,

Chief oj Bureau, Beverly T. Galloway.
Assistant Chief of Bureau, William A. Taylor.
Editor, J. E. Rockwell.
Chief Clerk, James E. Jones.

PRELIMINARY REPORT ON SUGAR PRODUCTION FROM

MAIZE. 1

By C. F. Clark, Assistant Agronomist, Office of Cotton and Truck Disease and Sugar-
Plant Investigations.

INTRODUCTION.

The experiments described in the following pages were conducted
under the direction of Mr. W. A. Orton, of the Bureau of Plant
Industry, in cooperation with the Sugar Laboratory of the Bureau of
Chemistry. The work was carried on during the summer of 1912 at
Garden City, Kans., and Washington, D. C. At Garden City the
field work was in charge of the writer and the analytical work under
the direction of Mr. W. B. Clark, assisted by Messrs. R. J. Hamon
and C. A. Hauser. At Washington the experimental plats, which
were located on the Arlington Farm, were under the supervision of
Mr. J. I. Milstead, of the Office of Horticultural Investigations, while
the analytical work was performed under the direction of JVIr. A. Hugh
Bryan, of the Bureau of Chemistry. These preliminary studies have
been carried out with only two varieties of corn. Other varieties
might give more or less favorable results. It is well known that the
maize plant responds to selection for specific purposes, and it is
possible that results achieved in the selection of the beet to increase
its sugar content might be paralleled with the proper selection of
maize.

EXPERIMENTS AT GARDEN CITY.

The variety used was Stowell's Evergreen sweet corn. It was
planted on May 15 and the ears were removed on August 9, when in
the milk stage. For comparison alternate rows were left with the
ears on. These stalks were analyzed at the same time as those from
which the ears had been removed.

Analyses were made at frequent intervals extending from August 28
to September 25. The samples, which consisted of 10 stalks each,
with the exception of the one analyzed September 7 which contained
30 stalks, were taken on the day in which the analyses were made.
As soon as cut the stalks were taken to the laboratory and the juice
extracted by running twice through a 3-roller mill. The solids were
determined by the Brix spindle and the apparent sucrose by direct
polarization, from which data the apparent purity was calculated.
The results are given in Table I.

i Issued Feb. 1, 1913.
[Cir. Ill] 3

4 CIECULAR NO. Ill, BUREAU OF PLANT INDUSTRY.

Table I. — Composition of cornstalks grown and analyzed at Garden City, Kans. 1

Date of
analysis.

Field treatment.

Solids.

Differ-
ence.

Sucrose,
direct

polariza-
tion.

Differ-
ence.

Appar-
ent
purity.

Differ-
ence.

Aug.

28

Sept.

3

Sept.

5

Sept.

7

Sept.

9

Sept.

18

Sept.

25

/Ears removed

\Ears not removed.

/Ears removed

(Ears not removed .

fEars removed

\Ears not removed.

Ears removed

/Ears removed

(Ears not removed.

/Ears removed

\Ears not removed.

(Ears removed

\Ears not removed .

Per cent.
15. 45
11.70

15.80
8.90

16.80
11.55

16.60

16.25
10.95

16.17
13.22

17.54
12.87

Per cent.
| 3.75

}■ 6.90

!■ 5.25

Per cent.

9.76
5.77

5.30

2.95

4.67

Average increase following removal
of ears

4.80

3.65

11.33

5.96

10.85

10.65
4.60

10.03

7.41

s 10. 50
2 6.50

Per cent.
\ 3.99

\ 6.33

\ 5.37

Per cent.
63.20
49. 30

/ 63.20 \
\ 49.30 J

Per cent.
13.90

63.20
41.00

/ 67.40 \
\ 51.60 /

22. 20

15.80

| 6.05
| 2.62
} 4. 00

65.40

65.60
42.00

62.00
56.15

2 59. 86
2 50. 50

4.73

23. 60

5.85

9.36

15.12

1 Analyses made at the Garden City Field Laboratory of the Office of Sugar-Plant Investigations.

2 These results are from Hauser's analyses; they seem to have been calculated from Clerget polarizations,
in which case the values for sucrose and purity would be true instead of apparent, which would make both
sets of figures higher, relatively, than those of the preceding analyses.

The data presented in Table I show that throughout the entire
period covered by the analytical work there was a marked increase
in the solids, in sucrose, and in purity as a result of removing the
ears, the average increase in each case being 4.80, 4.73, and 15.12 per
cent, respectively. In no instance was there a decrease. It will be
seen that in the stalks from which the ears had been removed there
was an increase in the percentage of sugar from the first analysis
until September 5, when the maximum was reached. After this
there was a small, though fairly uniform, decrease until September 25,
when there was a slight increase.

EXPERIMENTS AT WASHINGTON.
FIELD DATA.

For the experiments at Washington, D. C, a dent corn was used,
which was a selection of the Boone County White. It was planted
about May 30 and the ears were removed on August 22. The follow-
ing data were obtained regarding the weight of ears and the cost of
removing the same:

Weight of ears from one-third acre 2, 754 pounds.

Time required to remove ears from one-third acre 5 hours.

Time required to pick up ears and carry them from field 3 hours.

Total cost of removal of ears from one-third acre (8 hours, at
15 cents) $1. 20.

ANALYTICAL METHODS AND RESULTS.

Analyses of the stalks from which the ears had been removed, as
well as from those on which the ears remained, were made on Tuesday
and Thursday of each week from August 27 to October 29. The

REPORT ON SUOAR PRODUCTION FROM MAIZE.

samples consisted of 24 stalks each and were taken the morning of
the day the analyses were made. As soon as cut they were sent to
the Bureau of Chemistry, where the juice was immediately extracted
by running once through a 3-roller mill. The following analytical
determinations were made of this juice: (1) Solids by Brix spindle,
(2) direct and invert polarizations, (3) sucrose by Clerget, (4) invert
sugar before reduction, and (5) percentage of moisture in the sample
of cane as originally received. The purity of the juice — that is, the
percentage of sucrose divided by the solids — was calculated from the
data thus obtained. The results are presented in Table II, the sum-
mary of which shows the changes in composition of the juice follow-
ing the removal of the ears.

Table II. — Composition of cornstalks grown and analyzed at Washington, D. C l

EARS REMOVED.

Serial
No.

Juice.

Cane.

Date.

Solids.

Su-
crose.

Invert
sugar.

Non-
sugar
solids.

Purity.

Water.

Su-
crose.

Invert

sugar.

1912.
Aug. 27

Percent.

9728
9736
9740
9758
9769
9773
9791
■ISOI)
9817
9823
9837
9840
9854
9871
9939
9956
10005
10025
10072

Percent.
12.78
12.46
14.11
14.50
15.32
16.81
16.25
15. 64
14.68
9.50
la 07
15. 30
15.50
15.52
11.22
10.45
15.83
13. 87
15.22

Percent.

7.58

6.94

8.58

9.11

9.54

11.27

10.70

10.18

9.42

4.86

9.42

9.72

10. 02

10.10

6.40

5.56

9.46

7.61

9.01

Percent.
2.88
2.90
3.00
2.24
2.33
1.98
1.74
1.69
1.51
1.67
1.85
1.84
1.70
1.76
1.62
1.92
2.26
2.55
2.38

Percent.
2.32
2.62
2.53
3.15
3.55
3.56
3.81
3.77
3.75
1.97
3.80
3.74
3.78
3.66
3.20
2.97
3.11
3.71
3.83

Percent.
59.3
55. 7
60.8
62.8
62.3
67.0
65.9
65.1
64.2
51.1
62.5
63.5
64.6
65.1
57.0
52.2
59.8
54.9
59.2

Percent.

Percent.

Percent.

29

Sept. 3..

73.94
74.56
71.86
73.91
70.58
74.64
66.40
75.97
70.79
70.54
69.88
72. 74
78. 70
71.03
69.35
67.36
67.01

7.38
7.94
8.09
10.01
9.02
9.01
7.33
4.08
7.85
8.09
8.28
8.70
5.67
4.40
7.80
5.95
7.12

2.58

5

1.95

10

1.98

12

1.76

17

1.46

19

1.49

24...

1.17

26...

1.40

Oct. 1

3

1.54
1.53

8

1.40

10

1.51

15

1 43

17

1.52

22

1.86

24...

1.99

29...

1.88

Average

14.21

8.71

2.09

3.31

60.68

71.73

7.45

1.67

EARS NOT REMOVED.

Aug. 27. .

29.

Sept. 3.

5.

10.

12.

17.

19.

24.

26.
Oct, 1 .

8.
10.
15.
17.
22.
24.
29.

Average .

9729
9737
9741
9759
9770
9774
9792
9801
9818
9824
9838
9841
9855
9872
9940
9957
lOOOo
10026
10073

10. 08
8.41
7.50
8.70
6.45
0.30
8.38
6. 99
7.48
5.97
6.09
6.87
9. 03
8.95
8.07
10. ON
10.21
9.05
9.35

8.13

4.08
3.13
2.64
3.53
1.81
1.88
3.39
2.53
2. 53
1.63
1.55
2.34
4.24
4.24
3.18
5.18
6. 86
4.45
3.73

3.31

3.49
2. 7( '.
2.51
2.92
1.90
1.80
2.32
1.80
1.83
1. 01
1.42
1.54
1.93
1.91
1.92
2.35
1.96
1.96
2.57

2.13

2.51
2.52

3.12
2.99
2.80

2. SO
2.97
2.55
1.39
3.24

3. 05

2. 69

24.7
37. 2
35.2
40.7
29.0
30.0
40.3
36.2
33. 9
27.3
25.4
34.0
46. 9
47.4
39.4
51.3
67.1
41 i. 1

38.55

80.35
79.30
82.94
80.42

75. 8G
82.56
72.46
80.60
81.42
81.94
70.90
76. 52
77.10
77.66
69. 75

76. 41
73.52

77.98

2.29
3. 06
1.60
1.61
2.80
2.24
1.98
1.39
1.34
2.06
3.58
3.56

2. 66
4.47
5. 32

3. 76
3. 03

2.75

2.18
2.53
1.73
I . .". 1
1.92
1.59
1.43
1.38
1.23
1.35
1.63
1.60
1.61
2.03
1.52
1.66
2.08

1.71

1 Analyses made by Sugar Laboratory, Bureau of Chemistry.

CIRCULAR NO. Ill, BUREAU OP PLANT INDUSTRY.

Table II. — Composition of cornstalks grown and analyzed at Washington, D. C. — Con.

SUMMARY SHOWING CHANGES IN COMPOSITION OF JUICE FOLLOWING REMOVAL

OF EARS.

Date of
analysis.

Aug. 27
29

Field treatment.

/Ears removed

\Ears not removed

(Ears removed

\Ears not removed

c . o /Ears removed

aepi. 6 -i Ears not removed

. |/Ears removed

\Ears not removed

10
12
17
19
24

/Ears removed

\Ears not removed

(Ears removed.

\Ears not removed

(Ears removed

(Ears not removed

/Ears removed

\Ears not removed

/Ears removed. .
\Ears not removed

„„ /Ears removed. . .
i\Ears not removed

Oct.

1
3
8
10
15
17
22
24
29

/Ears removed

\Ears not removed

/Ears removed.. .
\Ears not removed

/Ears removed. . .
\Ears not removed

/Ears removed.. .
\Ears not removed

/Ears removed . . .
\Ears not removed

Ears removed. . .
t Ears not removed

/Ears removed...
\Ears not removed

/Ears removed. . .
\Ears not removed

/Ears removed . . .
(Ears not removed

Average increase or de
crease (— ) following
removal of ears

Solids.

Per

cent.

12.78

10.08

12.46
8.41

14.11

7.50

14.50
8.70

15.32
6.45

16.81
6.30

16/25
8.38

15.64
6.99

14.68
7.48

9.50
5.97

15.07
6.09

15.30
6.87

15.50
9.03

15.52
8.95

11.22
8.07

10.45
10.08

15.83
10.21

13.87
9.65

15.22
9.35

}5.80{

Differ-
ence.

Per
cent.

\ 2.70
4.05
| 6.61
5.80
}• 8.87
J-10. 51
| 7.87
} 8.65
| 7.20

Su- I Differ-jlnvert
crose. ence. sugar.

Per

cent.
7.58
08

{1

6.94

{ 3.13

/ 8.58
\ 2.64

9.11
3.53

/ 9.54
\ 1.81

ft

/10. 18
\ 2.53

i 9.42
\ 2.53

},
},

}«■

| 6.47

),

)
I
)

}<■

},

4.86
1.63

/ 9.42
\ 1.55

57

3.15

.37

5.62
22

s7

6.08

'-'7

10.70
39

9.72
2.34

10.02
4.24

110.10
\ 4.24

/ 6.40
\ 3.18

{I

{I:

{

5.56
18

9.46
86

61
45

9.01
3.73

Per
cent.

| 3.50

3.81

5.94

5.58

| 7.73

} 9.39

} 7.31

7.65

| 6.89

| 3.23

| 7.87

7.38

5.78

\ 5.86

| 3.22

} .38

2.60

3.16

} 5.28

/ 1.74
\ 2.32

5.40

Per

cent.

2.88

3.49

2.90
2.76

3.00
2.51

2.24
2.92

2.33
1.96

1.98
1.80

Per
cent.

}•»

}■
h
}

{?•"

1.69
1.80

1.51
1.83

1.67
1.61

1.85
1.42

1.84
1.54

1.70
1.93

1.76
1.91

1.62
1.92

1.92
2.35

2.26
1.96

96

2.38
2.57

Differ-
ence.

68

.37

18

}■

\— .58

h"

}-.32

}

.06

.43

.30

(
>

|— .23

}-'=

| — .30

} .30

}•

}-

.7.1

L9

-.04

Non-
su?ar
solids.

Per

cent.
2.32
2.51

2.62
2.52

2.53
2.35

3.15
2.25

3.55
2.68

3.56
2.62

3.81
2.67

3.77
2.66

3.75
3.12

1.97
2.73

3.80
3.12

3.74
2.99

3.78

2.80

3.66
2.80

3.20
2.97

2.97
2.55

3.11
1.39

3.71
3.24

3.83
3.05

Differ-
ence.

Per
cent.

}...
}.»

| .90

j .»

| .94
} 1.14

},.u

I .03

}-,«

| .68
(.,5
92

{
{

/ 60.
\ 35.

/ 62.
\ 40.

{
{

/ 65.
\ 40.

!■
>'■
>■
}■

>'•
}-
}-

.02

Pur-
ity.

Per
cent.
59.3
24.7

55.7

37.2

62.8

7

62.3
29.6

67.0
30.0

65.9
3

65.1
36.2

64.2
9

51.1
3

02.5
4

/ 04.
\ 33.

I 51.
\ 27.

/ 02.
\25.

/ 63.5
\ 34.0

/ 64.6
\ 46.9

/ 65.1
\ 47.4

/ 57.0
\ 39.4

/ 52.2
( 51.3

59.8
1

/ 59.
\ 67.

/ 54.

\ 46.

/ 59.
\ 39.

54.9
1

59.2

Differ-
ence.

Per
cent.

34.6

18.5

25.6

22.1

• 32.7

■ 37.0

25.6

28.9

30.3

23.8

37.1

29.5

| 17.7

} 17.7
} 17.6

} ■•

| -7.3
| 8.8

}

19.4

22. 13

The summary of Table II shows that very marked changes were
produced in the composition of the juice as a result of the removal
of the ears. Each analysis showed an increase in solids, in sucrose,
in nonsugar solids (with two exceptions), and in purity (with one
exception). Of the 19 determinations for invert sugar 9 showed an
increase and 10 a decrease, the average of which ( — 0.04 per cent) is
so small as to be considered negligible.

[Cir. Ill]

REPORT ON SUGAR PRODUCTION FROM MAIZE. 7

If we compare the averages of all determinations which are
brought together in the table, it will be seen that the greatest effect
resulting from the removal of the ears is on the sucrose content.
While there was a substantial gain in nonsugar solids, the increase
was not nearly so great as in the case of the sucrose. The percentage
of invert sugar, as has already been pointed out, was scarcely affected.

The sugar content where the ears had been removed (Table II)
increased at a very uniform rate from the date of the first analysis
until September 12, when the maximum was reached. After this
there was a decrease, which, however, was subject to considerable
fluctuation. It will be noted that the last sucrose determination
(9.01 per cent) is somewhat higher than the average for the entire
season (8.71 percent).

The percentage of invert sugar was highest during the early part
of the season, after which there was a decrease, followed by an in-
crease toward the end of the season.

That the accumulation of sugar is affected very quickly by the
removal of the ears is indicated by the fact that the analysis of
August 27, which was only five days after the ears were removed,
showed an increase in sugar content of 3.5 per cent.

PRESSING, CLARIFICATION, AND EVAPORATION.

The experiments included under this heading were conducted by
Mr. Bryan, who reports as follows:

EFFECT OF SEPARATE PRESSINGS.

In order to study the effect of extra pressing, a sample of cornstalks was run through
the mill and the juice collected. The stalks were again run through the mill and
this juice collected. The pressed cane coming from this second pressing was mois-
tened with water and then run through the mill. This juice was collected. The juice
coming from the three pressings was analyzed separately; then all mixed together
and the mixed juice analyzed. The results of this work are shown in Table III.

Table III. — Composition of juice from separate pressings.

Juices analyzed.

Juice from first pressing. . .
Juice from second pressing

Juice from maceration

Mixed juice

Solids.

Percent.

14.76

15.36

9.81

13. 41

Sucrose.

Per cent.

10.02

10.17

6.56

8.97

Invert
sugar.

Per cent.

1.82

1.63

.93

1.56

Ash.

Per cent.

0.66

.68

.45

Nonsugar
solids.

Per cent.
2.26
2.88
1.87
2.30

Purity.

Percent.
67.9
66.2
66.9

66.8

It will be noted that the juice from the second pressing contains slightly more solids
and also sucrose than the juice of the first pressing. At the same time it is much
higher in nonsugar solids, which gives it a lower purity. This fact is always noted
in cane-sugar manufacture, viz, that the purer juice comes from the first pressing
and as pressing increases the juice becomes more impure.
[Cir. ill]

CIRCULAR NO. Ill, BUREAU OF PLANT INDUSTRY.

CLARIFICATION AND EVAPORATION.

To the mixed juice, milk of lime was added to a slight alkalinity, as shown by litmus
paper. The lime used contained about 40 per cent magnesium. The whole was
then heated nearly to boiling or to a point where the scum easily separated from the
clear juice. The heat was then turned off and the juice allowed to settle. The clear
juice was then filtered off, brought to a boil, and skimmed.

In order to test the effect of concentration in acid solution, water saturated with
sulphur dioxid was then added to the boiling juice until it was decidedly acid to
litmus paper. The whole was then evaporated to a sirupy consistency. The analy-
sis of the juice after sulphuring but before evaporation showed the following com-
position:

Table IV. — Analysis of juice after sulphuring.

Juice.

Solids.

Sucrose.

Invert
sugar.

Ash.

Nonsugar

solids.

Purity.

After sulphuring but before evaporation

Percent.
14.96

Per cent.
10.10

Per cent.
1.79

Per cent.
0.71

Per cent.
2.36

Percent.
67.5

By the use of lime and sulphur dioxid it is seen that there had been a rise in purity,
which indicates a clarification by removal of some of the organic material. When this
concentrated material had reached a proper density, as told by appearances, it was
allowed to cool slowly. Crystals of sucrose did not appear immediately, but after a
week well-defined crystals of sucrose were noted. These continued to grow.

COMPARISON OF CORN WITH SORGHUM AND SUGAR CANE.

In order to show the relative value of corn as a source of sugar in
comparison with other sugar-producing plants belonging to the grass
family, Table V has been prepared.

Table V. — Comparison of corn with sorghum and sugar cane as a source of sugar.

Plant source.

Solids.

Sucrose.

Invert
sugar.

Nonsugar
solids.

Purity.

Corn:

Grown at Garden City, Kans.—

Ears removed

Per cent.
16.37
11.53

14.21

8.13

14.11

15.30
17.86

Per cent.

10.44

5.65

8.71
3.31
9.11

12.15
15.59

Per cent.

Per cent.

Per cent.
63 81

Ears not removed

48.43

Grown at Washington, D. C—

Ears removed

2.09

2.13

.82

1.35
.43

3.31

2.69
" 4.18

1.78
1.84

60.68

Ears not removed

38. 55

Sorghum '

64.48

Sugar cane:

Louisiana 2

79.19

Hawaii 3

87.28

i Average of analyses of 20-stalk samples of four varieties (Early Amber, Coleman, Orange, and McLean)
grown at Garden City, Kans., in 1912 in the same field as the corn.

2 Average of analyses of D. 74, D. 95 and home canes for nine years, 1895 to 1903. See Stubbs, W. C,
and Blouin, R. E., Comparative results of seedling sugar canes, D. 74 and D. 95, with our home sugar
canes (Louisiana Striped and Louisiana Purple), Louisiana Agricultural Experiment Station Bulletin,
s. 2, no. 78, p. 7-11, [1904].

3 Average of analyses of 17 varieties of plant cane harvested in 1904 and ratoon crop from the same plant-
ing harvested in 1906. See Eckart, C. F., Comparative tests with varieties of sugar cane, Hawaiian Sugar
Planters' Association, Division of Agriculture and Chemistry, Bulletin 17, p. 6.

The corn grown at Garden City from which the ears had been
removed was higher in solids than either sorghum or Louisiana
[Cir. in j

REPORT ON SUGAR PRODUCTION FROM MAIZE. 9

sugar cane, but lower than Hawaiian cane. The corn grown at
Washington was slightly higher than the sorghum in solids, but
lower than either Louisiana or Hawaiian sugar cane. At Garden
City the sucrose content was higher in the corn than in sorghum,
while at Washington the opposite condition was found. In both
instances it was lower than in sugar cane. The percentage of invert
sugar was higher in corn than in either of the other plants. The
nonsugar solids were lower in corn than in sorghum, but higher
than in sugar cane. The purity was lower in corn from both locali-
ties than in sorghum or sugar cane.

SUMMARY.

The results of the present season's work show that by the removal
of the immature ears from cornstalks the sucrose content of the
juice is greatly increased. There is also a small increase in non-
sugar solids and a very pronounced increase in purity. The per-
centage of invert sugar is not materially affected.

The purity coefficients of the juices are relatively low, the highest
being 67. 4 per cent.

As a sugar-producing plant corn compares very favorably with
sorghum in the chemical composition of its juice, but it is much
lower than sugar cane in sucrose and purity.
74943°— Cir. 111—13 2

DURANGO COTTON IN THE IMPERIAL VALLEY. 1

By 0. F. Cook. Bionomist in Charge of Crop Acclimatization and Adaptation

Investigations.

INTRODUCTION.

Duraiigo cotton is a new type of long-staple Upland cotton recently
introduced and acclimatized by the United States Department of
Agriculture. The behavior of the Durango cotton in numerous
experimental and field plantings leaves no doubt that the variety
is well suited to the irrigated lands of the Imperial Valley of southern
California. Other long-staple varieties have given excellent results
when the conditions happened to be favorable, but the Durango
has shown a wider range of adaptation and has produced good crops
in places where other long-staple cottons were comparative failures.
The success of the new variety has aroused popular interest and is
stimulating the development of a long-staple cotton industry in
southern California.

There is every reason to expect that long-staple cotton will
become one of the chief products of irrigation farming in the South-
west. In addition to very favorable natural conditions and the
absence of boll weevils, the cotton growers of the irrigated districts
are more ready to avail themselves of the many advantages that are
to be secured through the organization of local cotton-growing asso-
ciations for the production of a single superior type of cotton in the
community. 2

DEVELOPMENT OF DURANGO COTTON.

Durango cotton is one of several varieties developed by the De-
partment of Agriculture with a view to mitigating the injuries of
the boll weevil in Texas and adjacent States. The probability that
the weevil would seriously interfere with the cultivation of the late-
maturing types of long-staple Upland cotton formerly grown in the
Red River Valley and in the Yazoo Delta region of Mississippi and
Louisiana was foreseen several years before the long-staple districts

1 Issued Fob. 1, 1913.

2 These advantages have been considered in detail in an article published in the Yearbook of the Depart-
ment of Agriculture for 1911 under the title "Cotton improvement on a community basis." Separate
reprints of this article can be obtained by application to the Superintendent of Documents, Government
Print ing Oilier, Washington. D. C, upon the payment of a fee of 5 cents each.

[Cir. HI] H

12 CIRCULAR NO. Ill, BUREAU OF PLANT INDUSTRY.

were actually invaded by the weevils. It was feared that the weevil
invasion might result in the complete destruction of the long-staple
industry unless earlier and more prolific varieties could be developed.

The Columbia, Foster, and Durango varieties may serve to illus-
trate three different methods that were employed by the Department
of Agriculture in the effort to develop earlier long-staple sorts. The
Columbia cotton was obtained by straight selection from a short-
staple variety, the Foster by crossing long and short staple varieties,
and the Durango by acclimatization and selection of an imported
stock.

Several other promising types of Upland cotton have been secured
by the Department of Agriculture from weevil-infested regions in
Mexico and Central America and acclimatized in the United States
on account of special characters or habits that render them less sus-
ceptible to weevil injury. The Durango is the first of these to attain
sufficient uniformity to justify general distribution. The fact that
the Durango cotton has weevil-resisting characters is not important,
of course, in the Imperial Valley, where there are no boll weevils; but
the variety has other good qualities, quite apart from weevil resist-
ance.

That valuable new varieties should be found in Mexico and other
parts of tropical America should not appear surprising in view of the
fact that most of the types of cotton now cultivated in the United
States appear to have been introduced from tropical America. The
so-called Texas big-boll varieties are supposed to have been derived
from a Mexican stock introduced only a few decades ago. The intro-
duction of the Sea Island cotton from the West Indies is also a matter
of history.

The acclimatization of Durango cotton was begun seven years ago,
in 1905, when the original stock of seed from the State of Durango,
Mexico, was planted for the first time in Texas. The superior type
represented by the present select strain was first recognized and iso-
lated at Del Rio, Tex., in October, 1907. The introduction of the
variety into the Imperial Valley occurred in 1911, when a field of
about 3 acres was grown on the farm of Mr. W. E. Wilsie, near El
Centre About 200 acres were planted in 1912 in the vicinity of El
Centro and Holtville. Recent advices state that a carload of the
Durango cotton has been sold at El Centro at 17^ cents per pound,
which is equivalent to about 19 cents in eastern markets.

CULTURAL CHARACTERS OF DURANGO COTTON.

Durango cotton is valuable because it represents an unusual com-
bination of desirable cultural characters with length and high quality
of staple. From the agricultural point of view all of the well-known

[Car. -Ill]

DURANGO COTTON IN THE IMPERIAL VALLEY. 13

»

long-staple Upland varieties have appeared at a serious disadvantage
in comparison with the short-staple varieties in heing later and less
productive and more difficult to pick because of the smaller size of
the bolls. But Durango cotton is as early, as prolific, and as easy to
pick as most of the short-staple varieties that are now being grown
in the cotton belt. In the Imperial Valley the Durango cotton has
outyielded the short -staple varieties, in addition to producing a much
more valuable fiber.

The only varieties of short-staple cotton that can be said to com-
pete with the Durango cotton in the Southwest belong to the Texas
big-boll type, such as the Rowden, Triumph, and Lone Star. These
varieties have larger bolls than the Durango cotton, but are distinctly
inferior in habits of growth. When the conditions favor a luxuriant
development of the plants, as they usually do in the Imperial Valley,
the Texas big-boll varieties appear at a special disadvantage on
account of the weaker stalks and heavier foliage.

The Triumph cotton, which has been the principal short-staple
variety in the Imperial Valley, often becomes entirely prostrate
before the picking season arrives, or hides the bolls under a dense
mass of foliage. It is often necessary to hold up the plants with one
hand in order to pick the cotton with the other. The heavy shade
also tends to keep the bolls moist, and many of them are rotted or
mildewed. The Durango variety has stronger stalks. The plants
maintain a more erect position, with the bolls much more readily
accessible to the picker. (Fig. 1.)

The lighter and more open foliage renders the Durango cotton less
susceptible to rotting of the bolls or to discoloration of the fiber by con-
tact with the soil or by molding under heavy shade. The contrast in
foliage characters and habits of growth between the Durango and Tri-
umph is so striking that some of the Imperial Valley growers were able
to pick the two varieties separately in fields that had been planted
with mixed seed. Even the most unskillful picker could see the
difference, between the small, low, or prostrate Triumph, densely cov-
ered with large, leaves, and the large, upright Durango plants, with
their more open foliage and larger crops of bolls.

DURANGO COTTON COMPARED WITH THE COLUMBIA.

Several varieties of Upland long-staple cotton have, been tested in
the Imperial Valley in experimental plantings, so as to permit close
comparisons to be made between adjacent rows of the different sorts.
In this way it is possible to secure a much clearer idea of differences
of behavior than can be gamed from plantings hi separate fields.

Under the Imperial Valley conditions Durango cotton has shown
marked superiority over the Columbia both in yield and in quality of
fiber. In 1911, 4 bales of Durango cotton were grown on less than

[Cir. Ill]

14

ClRCULAB NO. Ill, BUREAU OF PLANT INDUSTRY.

3 acres, while an equal area of Columbia produced less than 1 baJe. In
L912 both fields were planted to Durango cotton, which produced an
excellent crop on the same land where the Columbia had failed the year
before. The behavior of the Columbia hi a near-by experimental
planting was much the same in 1912 as in 1911. Even where the
plants had ample exposure instead of field conditions the yield was
altogether inferior to that of adjacent plants of the Durango cotton.
(Fig" 2.)

Fig. 1.— Durango cotton plant at El Centro, Cal., with leaves removed to show habits of branching.

Comparisons between some of the best plants in the two rows showed
over three times as many open bolls on the Durango plants and a
general yield of about twice that of the Columbia. These contrasts
appeared the more remarkable because the Columbia is a very pro-
lific variety in the South Atlantic States.

The differences of behavior can be understood when the habits of the
two varieties are compared in detail. The failure of the Columbia
cotton to develop a large crop of bolls under the Imperial Valley con-
ditions is evidently connected with the tendency to produce a dense
canopy of leaves. This is quite different from the behavior of the

DURANGO CO!' TON l.\ 111

MI'KUIAI. VALLEY

. \t

variety in the Atlantic States, whore the plants are smaller and more
prolific and have much more open foliage.

When the foliage becomes too dense only the growing ends of the
stalks and vegetative 1 branches have adequate exposure to the sun-
light, while the fruiting branches are heavily shaded. Many of the
buds are blasted and many of the bolls fail to reach normal maturity.
Premature opening of the bolls and mildewing under the continuous
shade also contribute to the inferiority of the fiber. There are also
many plants with short lint, showing that degenerate variations have

Fin. 2.— Columbia cotton plant (left) and Durango cotton plant (right), showing different habits of growth,
at El Centro, Cal. The ( olumbia plant had 22 open bolls, the Durango plant 62.

boon induced by the new conditions. Such degeneracy was especially
noticeable among volunteer plants of the Columbia variety in the
season of 1912.

If the tendency to luxuriant growth and heavy foliage could be
restrained by withholding water or planting on less fertile soils,
somewhat more favorable results could doubtless be secured with
Columbia cotton. But the Durango is much better adapted to the
general conditions in the Imperial Valley, and there is no reason for
making special efforts to grow the Columbia. The fact that the
I oh-, mi

16 CIRCULAR NO. Ill, BUREAU OF PLANT INDUSTRY.

Columbia variety is not well adapted to southwestern conditions
does not affect its value in South Carolina and adjacent States,
where it is now attaining a well-merited popularity.

DTJRANG-O COTTON COMPARED WITH THE FOSTER.

The Foster is a variety of long-staple Upland cotton bred by Dr.
D. A. Saunders, of this department. It was developed from a
hybrid between the Triumph variety of Texas big-boll cotton and
Sunflower, a well-known long-staple Upland variety of the Peeler
type. - Foster cotton is adapted to the Red River Valley of Louisiana
and northeastern Texas, where it originated. It is an early, prolific
variety, producing 1^-inch lint under favorable conditions, but shows
a persistent tendency to occasional reversions toward the characters
of the Triumph parent. It is on account of this tendency to variation
that the Foster cotton has not been distributed more widely in the
long-staple districts of Louisiana and Mississippi, though excellent
results have been secured with many plantings.

The behavior of Foster cotton in the Imperial Valley does not
present the same difficulties as in the case of Columbia. The plants
do not become very luxuriant or densely leafy, but show rather an
opposite tendency to early maturity and dwarfing, which limit the
yield. Thus, in an experimental-row planting at El Centro in 1912
even the largest of the Foster plants were scarcely half the size of the
Columbia and Allen that stood on either side of the Foster row.
(Fig. 3.) Some of the Foster plants were very prolific for their size,
but the best of them produced less than a third as many bolls as the
Durango. The yield was only a little more than half that of the
Columbia or a quarter that of the Durango. (Fig. 4.)

A few fields of Foster cotton were grown in the Imperial Valley in
1912 from seed secured from Mississippi under the name " Unknown."
The seed seems to have been brought to the valley under the impres-
sion that it represented the regular long-staple or Peeler cotton, in
order to test the possibilities of this type of cotton in the Imperial
Valley. But irr reality the behavior of the Foster cotton is radically
different from that of a true Peeler cotton, such as the Allen, which
stood next to the Foster in the experimental plantings made by the
Department of Agriculture. In the fields of Foster, or so-called
"Unknown," diversity was very apparent in the habits of growth of
the plants and the sizes and shapes of the leaves and bolls. Lint
samples showed the usual tendency to Triumphlike reversions, ren-
dering the staple distinctly less uniform than in the Durango, though
still of good, marketable quality. The yield also did not approach
that of many of the Durango fields.

[Cir. Ill]

DURANGO COTTON IN THE IMPERIAL VALLEY. 1 <

DURANGO COTTON COMPARED WITH THE ALLEN.

The Allen variety has been included in experimental plantings as
the best representative of the Peeler varieties which formerly fur-
nished the long-staple crop of Mississippi and Louisiana. The results
leave no doubt that in the ImperiaL Valley the Allen is a very un-
promising variety, on account of the usually low yields of seed cotton,
the low percentage of lint, and the small bolls. Even when the plants
are large and vigorous they may yield very little (fig. 3). In all
these particulars the Allen falls far below any of the other varieties
included in the test. In the El Centro planting of 1912 the lint

Fig. 3— Allen cotton plant (left) and Foster plant (right), showing different habits of growth at El
Centro, Cal. The Allen plant had 12 open bolls, the Foster plant 14, in rows adjacent to those
illustrated in figure 2.

percentage was only 20, whereas none of the other varieties fell below
27 and the Durango gave over 30. With respect to the size of the
bolls the Allen cotton has little or no advantage over the Egyptian
varieties and is also less productive. If the Peeler cottons were
the only Upland long-staple type available there would be no pros-
pect of establishing a long-staple industry in the Imperial Valley,
for planters would continue with short-staple cottons or go over
to the Egyptian. Either of these alternatives would be more promising
than the growling of the Peeler varieties.

[Cir. Ill]

18

CIRCULAR NO. Ill, BUREAU OF PLANT INDUSTRY,

DURANGO COTTON COMPARED WITH THE EGYPTIAN.

On account of the present scarcity of Upland long-staple cotton
the Durango cotton enjoys a somewhat artificial and perhaps tem-
porary market advantage over the Egyptian. The scarcity of long-
staple Upland cotton is due largely to the fact that the presence
of the boll weevil interferes with the growing of the Peeler varieties
in the former centers of production in Mississippi and Louisiana.
Short-staple varieties have replaced the long staples in many dis-
tricts, but efforts are now being made to go back to the cultivation
of long staples on the basis of the new early varieties like the Colum-
bia, Foster, and Durango. A part of the deficiency is being made
good by the planting of Columbia and other early long-staple varie-

t ies in the Piedmont region of the South
Atlantic States, but it may be several
years before an adequate supply of long
staples can be developed in this region.
Although the relative desirability of
the Durango or of Egyptian cotton for
the Imperial Valley and other irrigated
regions of the Southwest is likely to
fluctuate with the state of the market,
it will be very poor policy for cotton-
growing communities to make frequent
changes from one type of cotton to the
other. To learn the methods of grow-
ing, selecting, handling, and marketing
either type to the best advantage is
likely to require several years, so that
a change from one to the other must be
a very expensive operation. Scarcity of
labor and the higher cost of picking are
local factors that now stand seriously in
the way of developing community pro-
duction of the Egyptian cotton in the Imperial Valley and make it
appear preferable to adopt the Durango. Even though the individual
farmer may not agree with his neighbors regarding the variety or
type of cotton that should be grown, he is likely to gain more by
growing the community variety than by producing a small quantity
of some other kind of cotton that he is unable to sell to advantage.

SPECIAL CHARACTERS OF DURANGO COTTON.

In addition to the features to which reference has been made in the
preceding paragraphs, several other characteristics of Durango cotton
are w'orthy of mention.

[Civ. 11 1 I

Fig. 4.— Boll of Durango cotton, with
involucre, showing relatively small
bracts. (Natural size.)

DURANGO COTTON IN THE IMPERIAL VALLEY. 19

One of the most peculiar characters is shown in the seedlings. In
most other kinds of cotton the seedlings produce at least two or three
simple oval leaves before beginning to develop leaves of the usual
lobed form of the adult plant. But in Durango cotton a large pro-
portion of the seedlings have lobes on all the leaves, including the first
leaf above the seed leaves. This character might be of use in deter-
mining early in the season whether any planting of Durango cotton
is pure. The oidy other variety that has been found to have any
considerable proportion of the first leaves lobed is the Alien.

In Durango cotton the tendency to produce vegetative branches is
less than in most other varieties, and it can be reduced still further by
leaving the plants close together in the rows. Earlier crops and
larger yields have been obtained in this way than by the usual method
of thinning the plants early to 2 feet or upward. Some of the best
yields have been obtained from plants only 6 or 8 inches apart in the
rows.

Another peculiarity of Durango cotton is the unusually small size
of the involucral bracts. The difference in this respect between the
Durango and the Triumph or other Texas big-boll varieties is very
striking. The smaller bracts of the Durango cotton are a distinct
advantage from the standpoint of clean picking, for the ends of the
dry bracts are not so likely to be broken off and gathered with the
fiber (fig. 4). Fragments of the involucral bracts furnish a large part
of the "trash" that goes to the gin with the cotton and lowers the
grade of the bale.

The following formal description of the Durango variety has been
supplied in connection with the congressional seed distribution:

Plan! <if uprighl habit with a strong central stalk and rather stiff ascending vegeta-
tive branches. Fruiting branches of moderate length or rather short , under sonic con-
ditions becoming semicluster. Foliage rather deep green, reddening early in the
season. Leaves of medium size, usually with five or seven rather narrow tapering lobes,
leaves with three lobes being less frequent than in most other varieties of upland
cotton. Involucral bracts with rather small, triangular cordate bracts, margined
with rather short teeth. Calyx lobes irregular in length, sometimes very long and
slender. Bolls of medium or rather large size; under favorable conditions about GO
to the pound. Shape of bolls conic oval with nearly smooth surface, the oil glands
deeply buried. (Fio-. 5.) The proportion of five locked bolls varie: usually from 40
to 50 per cent. Seeds of medium size, covered with white fuzz and uniform abundant
lint about 1] inches long under favorable conditions.

THE PURE-SEED PROBLEM.

As uniformity is a prime requisite with long-staple cotton, it is idle
to expect that a prosperous long-staple industry can be built up in
the Imperial Valley or elsewhere unless provision is made for main-
taining the purity of the variety on which the industry is based.
With short-staple cotton uniformity in the plants is important as a
[Cir. Ill]

20

CIRCULAR NO. Ill, BUREAU OF PLANT INDUSTRY.

means of securing high yields, though the question of uniformity of
fiber is seldom raised in the short-staple market. But uniformity is
an important factor in the market value of long-staple cotton. Loss
of uniformity is one of the chief causes for the failure of many
attempts to introduce long-staple varieties into short-staple districts
of the Southern States. The usual inference in such cases is that the
conditions are unfavorable for the long-staple varieties, whereas the
injury has come usually through admixture of short -staple seed and
neglect of selection. This will soon destroy the uniformity and com-
mercial value of any long-staple variety, no
matter how well bred at the beginning.

EDUCATION OF COTTON BREEDERS.

Selection is not difficult for those who
have sufficient familiarity with the character-
istics of a variety, but most farmers are not
accustomed to doing work of this kind and
prefer to buy their seed instead of growing it
for themselves. Nevertheless, each commu-
nity should have enough local breeders or
seed growers to maintain the necessary supply
of select seed. The work of selection has
been rendered easier and more effective by the
development of improved methods, as de-
scribed in other publications of the Bureau of
Plant Industry. 1

A certain amount of practice in observation
is required to develop skill in recognizing
differences between individual plants, so as to
preserve those that are true to type and reject
all that are abnormal or aberrant. Ability to
do this is not rare, though seldom developed
or applied. The problem of educating more
cotton breeders is now receiving attention,
and it is hoped that opportunities of learning this art may be
provided in the Imperial Valley in the coming season.

ERADICATION OF WEED COTTON.

An unusual difficulty arises from the fact that in the Imperial
Valley cotton so frequently survives the winter in the fields or escapes
in waste places, wherever the seed happens to be scattered. Though

" Local adjustment of cotton varieties, U. S. Department of Agriculture, Bureau of Plant Industry,
Bulletin 159, and Cotton selection on the farm by the characters of the stalks, leaves, and bolls, U. S. De-
partment of Agriculture, Bureau of Plant Industry. Circular 60.
[Cir. Ill]

Fig. 5.— Boll of Durango cotton,
showing typical form. (Natural
size.)

DURANGO COTTON IN THE IMPERIAL VALLEY. 21

the plants are usually killed to the ground in the fall, the roots remain
alive and send up new shoots in the spring from subterranean buds.
If a determined effort is to be made to place the growing of cotton
in the Imperial Valley on a Durango basis one of the first steps should
be to eradicate all the plants of other varieties from the fields as well
as those that are now scattered about like weeds in many neighbor-
hoods. An escaped cotton plant is much more dangerous than any
ordinary weed. The contamination of a select variety is not a merely
temporary injury, but may take years of careful selection to eliminate.
In view of the obvious danger of contamination from the over-
wintered plants or from volunteers that spring up from scattered
seed of the previous season, it is certainly inadvisable to plant Du-
rango cotton on any land where cotton was grown the year before.
Neglect of this precaution and that of separate ginning of the seed
planted in the season of 1912 has resulted in a contamination of
nearly all of the Durango seed in the Imperial Valley, which is now
recognized as a serious loss to the community. The present stock
of seed will need to be discarded and replaced by new supplies of
select seed that are being brought in from Texas. But any pure
stock will soon become contaminated if other kinds of cotton are
allowed to grow in adjacent fields or along borders or ditch banks,
where volunteer plants are often found.

CONCLUSIONS.

The Durango cotton is a new type introduced and acclimatized by
the Department of Agriculture and recommended in 1911 as the
Upland long-staple variety best suited to irrigated conditions in the
Southwestern States. Further experimental and field plantings in
1912 have rendered the superiority of the new type still more ap-
parent.

Excellent results have been secured in numerous plantings of the
Durango variet}' in the Imperial Valley in 1912. The superiority of
the new type has become still more apparent and is recognized by the
local cotton-growing community, which is now making an organized
effort to place the valley exclusively on a Durango basis.

The superiority of the Durango variety lies in the fact that it com-
bines the desirable cultural qualities of short-staple varieties with
length and strength of lint formerly obtained only from the so-called
Peeler varieties of the Delta region of Mississippi and Louisiana. All
the experiments indicate that the Durango cotton is much better
adapted to the Imperial Valley conditions than any other long-
staple Upland variety.

The idea of transferring the Delta varieties to the Imperial Valley
is impracticable. Experiments have shown that these varieties are

[Cir. Ill]

22 CIRCULAR NO. Ill, BUREAU OF PLANT INDUSTRY.

not suited to conditions in the Imperial Valley on account of later
maturity, low yield, small percentage, and uneven length of lint.
With respect to these cultural characters, the Peeler cotton is inferior
to the Egyptian cotton as well as to the short-staple varieties. The
defects of the Peeler cotton are not found in the Durango, which is
early and productive and has abundant, uniform fiber.

Though the soil and climate of the Imperial Valley are favorable
to the production of an excellent quality of Egyptian cotton, the
lack of an adequate supply of labor for picking interferes with the
development of an Egyptian cotton industry in that community.
The introduction of the Durango cotton meets the demand for a long-
staple variety adapted to the local conditions and promises the
largest profits for the farmers of the valley.
[Cir. Ill]

IMPROVED APPARATUS FOR DETECTING SULPHURED GRAIN. 1

By George H. Baston, Assistant in Grain Standardization.

The apparatus - commonly in use for detecting sulphured grain
consists of an Erlenmeyer flask of 500 c. c. capacity, fitted with a
cork stopper and a delivery tube.

The chief objection to this apparatus is that it is almost impossible
to make a cork connection perfectly tight, and it is also very hard
to keep clean. The escape of the gas around the stopper was the
first difficulty that the writer tried to overcome. Rubber stoppers
are not practicable for this purpose on account of the sulphur which
they contain. A hollow, ground-glass stopper with a glass delivery
tube attached was substituted for the cork connections. The hollow
in the center of the stopper should be filled with cotton, which acts
as a filter, thereby preventing the dirt and dust from being carried
over with the gas. The dimensions of the apparatus are shown in
figure 1.

The method used for detecting sulphured grain is as follows :

Place 100 grams of the grain to be examined, together with 10 grams
of zinc, mossy or granular, chemically pure and free from sulphur,
in a 500 c. c. flask. Pour into the flask enough diluted hydrochloric
acid (1 part of acid to 4 parts of distilled water, by volume) to just
cover the grain. Close the flask with the ground-glass connection
and place the end of the glass delivery tube in a test tube containing
a 2 per cent solution of lead acetate (2 grams of chemically pure lead
acetate in 98 c. c, of distilled water) which has just been filtered.
The test tube should be only about three-fourths full of the lead-
acetate solution, in order to prevent spilling when the gas begins to
pass over. The delivery tube should extend to within about 1 inch
of the bottom of the test tube. Figure 1 shows how the apparatus
should be adjusted.

In the case of unbleached grain, the gas liberated from the zinc
and hydrochloric acid is hydrogen; with grain which has been
bleached with sulphur, the gas liberated is hydrogen sulphid, the

i Issued Feb. 1,1913.

2 The method for the detection of sulphured gram is described in detail in Circular 40, Bureau of Plant
Industry, U. S. Department of Agriculture, entitled "A simple method of detecting sulphured barley
and oats," by W. P. Carroll, Assistant in Grain Standardization.

[Cir. Ill] 23

24

CIRCULAR NO. Ill, BUREAU OF PLANT INDUSTRY.

presence of sulphur being indicated by a black precipitate of lead
sulphid which forms in the test tube.

To avoid any mistake in determining whether or not grain has been
bleached it is advisable before making the test to make several
experiments with both bleached and unbleached grain and also

HEAVY NECK
INSIDE DIAMETER
A PR OX I MAT ELY 1 in.

GROUND GLASS
CONNECTION

Fut. 1. -Improved apparatus for detecting sulphured grain, showing its dimensions and the manner of

adjustment.

with mixtures of the two. For example, samples containing 2 per
cent, 5 per cent, 10 per cent, and 50 per cent of bleached grain should
be tested, in order to become familiar with the appearance of the
precipitate when there is only a small quantity of sulphur present.

[Cir. Ill]

SUPERNUMERARY CARPELS IN COTTON BOLLS. 1

By Rowland M. Meade, Scientific Assistant, Crop Acclimatization and Adaptation

Investigations.

DISCOVERY OF THE ABNORMALITY.

An examination of a large number of bolls from several varieties of
cotton grown at Lanham, Md., in 190S, revealed an interesting abnor-
mality. An early frost had destroyed the leaves, but the stalks and
bolls were in the normal green condition. Many bolls had reached
maturity, but none had opened naturally, so it was necessary to
make all observations in artificially
opened bolls.

In its simplest form the abnor-
mality consisted of a solid, elon-
gated, whitish body, developed in
the center of the boll between the
placentae. (See fig. 1.)

In many instances, however, this
body was divided into from two to
five longitudinal compartments, re-
sembling miniature locks, which
inclosed rudimentary ovules. In
some cases it extended nearly the
entire length of the boll, but was
always unattached except at the
base. Although the abnormalities
observed at Lanham were poorly
developed, the longitudinal com-
partments with the inclosed rudi-
mentary ovules warranted the
belief that they might be regarded
as supernumerary carpels.

In some respects this abnormal-
ity resembles that found hi the navel orange. There is a difference,
however, in that the supernumerary carpels of the cotton boll develop
at the base of the placentae, while in the orange they are produced
at the opposite end of the fruit.

Fig. 1. — Supernumerary carpels between the pla-
centae oj a green cotton boll. From Glendale,
Cal. (Natural size.)

» Issued Feb. 1, 1913.

[Cir. Ill]

25

26 CIRCULAR NO. Ill, BUREAU OF PLANT INDUSTRY.

At Lanham the abnormality occurred particularly in the bolls of
a new type of Upland cotton introduced from the State of Durango,
Mexico, but it was also found in the Triumph, a Texas variety. Since
1908 the same abnormality has been observed in other varieties at
different points in the United States and has been given more thorough
examination.

OCCURRENCE AT GLENDALE, CAL.

During the seasons of 1910 and 1911 more perfect examples of this
abnormality were found at Glendale, Cal., both in green and open
bolls. Here the resemblance to miniature carpels was very plain.
None of the Upland varieties grown at Glendale were exempt from
this abnormality, while only one or two cases were found in several
rows of different varieties of Egyptian cotton. Although the vege-
tative parts of the plants showed no excessive growth, the bolls com-
monly grew to an unusual size. These were frequently abnormal.
A large number of bolls had the carpels separated at the apex and in
some cases the carpels were distinct to the base, with the exception
of the parts along the placenta?.

At Glendale supernumerary carpels sometimes developed in other-
wise normal bolls, but they were most frequently associated with
the bolls having separated carpels. The true carpels of such bolls
were loosely connected, and the inner faces of the placentae were
flattened or hollowed to provide space for the central carpels. Al-
though they completely filled the space between the placenta?, the
abnormal carpels were in no way attached to the true carpels or to
the placenta?. Supernumerary carpels measuring more than an
inch in length by three-eighths of an inch in breadth were frequently
noted. Ovules, an eighth of an inch or less in length, producing
extremely weak fiber, which in some cases measured half an inch,
developed within the compartments of the supernumerary carpels,
apparently without fertilization. In open bolls the supernumerary
carpels became dry and woody and the compartments split along
the back. Through the narrow slits the fiber emerged, and the
resemblance to a miniature boll was striking.

RELATION OF CLIMATIC CONDITIONS TO ABNORMALITY.

There appear to be similarities in the climatic conditions of the
localities where these supernumerary carpels have been observed
to develop most frequently. At Lanham, Md., and Glendale, Cal.,
the climatic conditions are not favorable for the production of
cotton on a commercial scale. At Lanham the growing season is
generally too short to mature a crop before frost, and the bolls
seldom ripen normally. At Glendale the nights are cool for most of
[Cir. in j

SUPERNUMERARY CARPELS TN COTTON BOLLS.

27

the season, though the days are hot and dry. As the normal period
of growth and development of the plant is at night, it seems not
unreasonable to suppose that the low temperatures might induce
such abnormalities. This abnormality has also been observed at
other points where the growing seasons are limited or the nights
cool, such as Kerrville, Tex., and Chico, Cal.

OCCURRENCE AT CLARKSVILLE, TEX.

During the autumn of 1912 many unusually good examples of
these supernumerary carpels were observed at Clarksville, Tex.,
occurring commonly in otherwise normal bolls. They were detected
only in the bolls at the base of the plant — those developed earliest
in the spring. The spring of that year was rather late and somewhat
cooler than usual, and it may

be on this account that the !

abnormalities were produced
in such abundance.

Nearly allspecimens exam-
ined were so well developed
that they contained rudi-
mentary lint-covered seeds.
Specimens consisting of from
two to five separated or con-
nected leaf -like members were
examined and there was
no mistaking their nature 1 .
When flattened out they re-
sembled miniature elliptical
or slightly lobed leaves with
inturned thickened margins.
The rudimentary ovules in
some cases appeared to be
borne on the thickened mar-
gins, wliile in others they
seemed to be in a line extending from the sinus between the small
lobes and their point of attachment to the peduncle.

An exceptional specimen in which apparently normal seeds had
developed in one of these supernumerary carpels was found at Clarks-
ville, Tex., by Mr. C. H. Clark. (See fig. 2.)

The two seeds inclosed in one of these carpels, though somewhat
smaller than normal, produced abundant fiber of good length and
strength. This specimen was not observed until after the locks had
been removed from the normal carpels and the supernumerary carpels
had been somewhat mutilated, but the indications from the broken

[Cir. Ill]

Fig. l>. Well-developed supernumerary carpels inclosing
apparently normal seeds. From Clarksville, Tex. ( Nat-
ural size.)

28 CIRCULAR NO. Ill, BUREAU OF PLANT INDUSTRY.

tip of one of the carpels were that a threadlike projection extended
beyond the normal carpels. Fertilization may have taken place
through this extended tissue. It remains to be seen whether these
seeds are capable of producing plants, but their presence in such a
mature form proves most emphatically the real nature of these
abnormalities. Mr. Clark's own description of the specimen found
at Clarksville follows:

A boll of an Upland variety of cotton found at Clarksville, Tex., in the fall of 1912
had developed between the placenta? two large supernumerary carpels which con-
tained two apparently normal seeds. One of these abnormal carpels was thick and
hard, resembling a normal carpel except that it was narrower (8 millimeters) and
attached along its back to the placenta of a normal carpel for half the length of the
normal boll. From here it projected independently. Although this projection had
be'en broken off before the specimen was found, it must have extended to the apex
of the boll and have furnished means of fertilization for the seeds. The other carpel
had the appearance of a thin obovate leaf, measuring 1 centimeter in width and about
2 centimeters in length. This was attached only at the base. Three veins could be
easily traced in this leaf-like carpel and the surface was dotted with oil glands, but it
is doubtful whether placentge had been developed on the margins. Two locks were
contained within the cavity formed by these two abnormal growths. One lock con-
sisted of two apparently normal seeds and three rudimentary ovules. The other con-
sisted of rudimentary ovules only. The two seeds, though a little undersized, pro-
duced abundant lint, that of one measuring 33 millimeters and of the other, 25 milli-
meters in length. The surrounding boll wa" in other respects normal, with the excep-
tion that the five placentae of the carpels were flat and widened to about 3 millimeters.

OCCURRENCE ON OTHER PLANTS.

At Washington, D. C, on the grounds of the Department of Agri-
culture, are several plants of althaea (Hibiscus syriacus) which in 1911
were abnormal in this same respect. Supernumerary carpels, two to
five in number, develop between the placentae, although the normal
series of carpels are not distorted. The central carpels in some cases
contain rudimentary ovules.

[Cir. Ill]

KEEPING SOFT CUTTINGS ALIVE FOR LONG PERIODS. 1

By George W. Oliver, Plant Breeder and Propagator, Office of Foreign Seed and

Plant Introduction.

It has often been found desirable to bring soft or herbaceous plant
cuttings from long distances, but the difficulty heretofore attending
their transportation has been that the cuttings do not remain in
good condition longer than a day or two. This difficulty has been
removed by an exceedingly simple contrivance.

Dormant hard-wooded cuttings and scions can be sent long dis-
tances by mail, as was demonstrated a few years ago in a collection
of scions and bud sticks forwarded to Mr. William S. Lyon, at that
time in the service of the Government of the Philippine Islands.
Not only did the material reach its destination in good condition,
but some of it was repacked according to instructions and returned
to Washington, where it was successfully grafted in the greenhouses
of the Department of Agriculture.

Soft or herbaceous cuttings, on the other hand, such as those of
alfalfa, clover, and many other plants, can not be sent long dis-
tances by mail or express, but they will survive a journey of six
weeks in perfect condition if kept where they can be given light
occasionally and attention is paid to supplying the water lost
through evaporation. This treatment in the case of alfalfa and
many other plants induces healthy root action during a journey of
several weeks' duration.

The apparatus for successfully bringing cuttings of herbaceous
plants from distant places is of the simplest nature. The necessary
articles are a small quantity of living sphagnum moss, two sheets of
strong glass, 5 by 7 inches or larger, and some string. The cuttings
should be prepared in much the same way as though intended to be
placed in a propagating bed.

Arrange the first layer of cuttings without too much crowding and
with the upper surfaces of the leaves on the first piece of glass and
on top of the cuttings, and place about 2 or 3 inches of living sphag-
num moss evenly distributed over the cuttings. Place another layer
of cuttings on top of this moss, with the under surfaces of the leaves
next to the moss, so that all the available space will be covered, and

i Issued Feb. l, 1913.
[Cir. Ill | 29

30

CIRCULAR NO. Ill, BUREAU OF PLANT INDUSTRY.

on top of this second layer of cuttings place the second piece of glass.
Press down firmly, remove the moss which protrudes beyond the
edges of the glass, and tie together with stout twine (fig. 1).

Fig. 1.— Alfalfa cuttings arranged on dampened sphagnum moss and covered with glass.

The package now consists of two pieces of glass, 2 inches of pressed
sphagnum moss, and two layers of cuttings — one between each piece
of glass and the moss. By keeping the moss moist and giving all
the light possible (direct sunlight is best, and it does not raise the
temperature of the moss to 'an appreciable extent beyond that of

[Oir. Ill |

KEEPING SOFT CUTTINGS ALIVE EOR LONG PERIODS. 31

the surrounding atmosphere), the cuttings are not in the least
injured, provided the material is free from fungous troubles. If
the journey is long enough — say, of four to six weeks' duration —
cuttings such as those of clover, a^alfa, Dorycnium, lotus, and
many other plants will have rooted freely while closely pressed
against the glass. During the time of rooting no attention is re-
quired beyond keeping the moss wet and exposing the cuttings to
the light for a few hours each day.

With the moss only slightly dampened, scions and bud sticks of
rare plants keep a very long tune in good condition under the same
treatment.
[Cir. Ill J

o

U. S. DEPARTMENT OF AGRICULTURE.

BUREAU OF PLANT INDUSTRY— Circular No. 112.

B. T. GALLOWAY, Chief of Bureau.

MISCELLANEOUS PAPERS.

Opportunities in Pecan Culture C. A. REED

The Jonathan Fruit-Spot W. M. SCOTT and J. W. ROBERTS

Egyptian Cotton as Affected by Soil Variations T. H. KEARNEY

Relation of Stand to Yield in Hops . W. W. STOCKBERGER and JAMES THOMPSON

Issued February 8, 1913.

LIBRARY

DRK
ANICAL

'GARDEN

WASHINGTON:

GOVERNMENT PRINTING OFFICE.

1913.

BUREAU OF PLANT INDUSTRY.

Chief of Bureau, Beverly T. Galloway.
Assistant Chief of Bureau , "William A. Taylor.
Editor, J. E. Rockwell.
Chief Clerk, James E. Jones.
[Cir. 112]
2

OPPORTUNITIES IN PECAN CULTURE. 1

By C. A. Reed, Scientific Assistant in Pomology.

INTRODUCTION.

The demand for authentic information pertaining to pecan culture
was never so great as at present. Much has been said in a speculative
way regarding the commercial future of the industry, the suitability
of various localities, the probable bearing age, the size of crops, their
immunity or susceptibility to disease and insect pests, then* relation
to the weather, and many other matters, but with the industry still in
an experimental stage such statements have been necessarily based
largely upon limited experience and are therefore subject to revision
on short notice.

The feature of the industry concerning which there is the greatest
interest is its commercial future. This information can be gained
only through a full knowledge of what it has cost to establish and
maintain pecan orchards properly, what the crops of nuts have been,
what prices have been realized, the influence of increased production
upon future prices, and the possibility of increasing the market
demand for the nuts.

DIFFICULTY OF OBTAINING ACCURATE DATA.

Although the planting of pecan trees was probably begun in the
eighteenth century, or perhaps even earlier, their planting in orchard
form is comparatively recent. Some pecan orchards are known to
have been planted prior to 1880, but until about 1905 practically all
consisted of seedling trees. In the planting of these early orchards
little attention was paid to such matters as seed selection, the adapta-
bility of locality and soil to the species, proper distances between
trees, their cultivation, etc., the importance of which is now well
understood. A few such seedling orchards when kept under culti-
vation have begun to bear crops of more or less importance at 15 to
25 years of age, but few orchards have made records as to bearing
which would be of interest even if obtainable. The chief value of
such orchards is in the opportunity they afford of quick transforma-
tion by top-working to named varieties.

The growing of named varieties — trees propagated by budding and
grafting with scions from individual parent trees — began to be active
about 1890. At that time comparatively few varieties were known,

1 Issued Feb. 8, 1913.
[C3r. 112.] 3

4 CIRCULAR 112, BUREAU OF PLANT INDUSTRY.

and during the next decade less than a dozen were widely dissemi-
nated. Of these several proved unsatisfactory and have since been
superseded. During the second decade of active interest in pecan
orcharding a great many varieties were introduced, most of winch
are already disappearing. The dissemination of new varieties, many
of which may later prove to be of little importance, is still going on
and may be expected to continue. Many of the disseminated varieties
have been disappointing in that they have not fruited as was expected,
especially in localities other than those where they originated, while
some have proved highly susceptible to fungous diseases.

By far the greater number of promising orchards of budded or
grafted trees now existing are still too young to bear commercial
crops. It is only here and there that orchards more than five or six
years of age, of good varieties, well adapted to local conditions, and
under a high state of cultivation are to be found. A number of
orchards which might otherwise have been in bearing have been so
heavily cut for bud wood that the chances of fruiting have been im-
paired for the time being.

Occasional individual tree records made under highly favorable
conditions are frequently taken as a basis for estimates of what may
be expected from orchards of the same varieties of the same age.
The fallacy of taking these records as a basis for orchard estimates
is apparent when it is realized that it is entirely impracticable to
maintain in large orchards the garden conditions under which such
records have usually been made. This has recently been emphasized
by the record of an orchard of 200 trees in southwestern Georgia,
the exact weight of the nuts from each tree for the last two seasons
having been personally recorded by the writer during the harvest.
In 1911, the seventh season from planting, the crop from these 200
trees, all of which are of one variety, of the same age, and under the
same degree of cultivation, amounted to 1,137 pounds and ranged
from a few nuts per tree, and occasionally none at all, to 17^ pounds
in the case of one tree. The average for the entire orchard was 5.08
pounds per tree. In 1912 the total crop fell to 639 pounds, or an
average of 3.19 pounds per tree, the range of individual trees being
from no crop to 13^ pounds. If estimates of the probable yield of
the entire orchard were to be based upon records of single trees in
this orchard the figures of the total crop would show a range from
no crop to 3,500 pounds during 1911 and 2,700 pounds during 1912,
none of which would be correct.

THE COST OF ESTABLISHING AND MAINTAINING PECAN

ORCHARDS.

A letter of inquiry regarding the most important items in the cost
of establishing and maintaining pecan orchards was recently sent
out from the Bureau of Plant Industry to a number of persons in the

[Cir. 112]

OPPORTUNITIES IN PECAN CULTURE. O

growing districts who have had experience in orchard culture. Re-
plies were received from 17 growers in the States of Georgia, Florida,
Mississippi, Louisiana, and Texas. The questions asked and the
replies made were both limited in number and brief, but as they brought
out the opinions of some of the most prominent and succsssful pecan
growers, the questions and a summary of the replies are here presented.

(1) At about what price per acre can cleared land in your section suitable for pecan
planting be purchased?

The estimates made were mostly from $20 to $40 per acre; one
was $40 to $100, one $75, and one $200.

(2) At what price can uncleared land be obtained?

One estimate was $5 to $15 per acre. Most were from $10 to $30.
One was from $20 to $50, and one $125.

(3) About what ought it to cost to put uncleared land into shape for planting?

In some instances the estimates made included the cost of removing
stumps, stating that frequently the value of the standing timber w T as
sufficient to pay for clearing. The estimate for clearing, removal of
the stumps included, ranged from $18.21 to $28.21 per acre.

(4) About how much per acre has it annually cost you thus far for fertilizer in your
pecan orchard?

The kind and quantity of fertilizer pecan trees should receive vary
greatly, depending upon local conditions. Some orchardists use no
commercial fertilizer, relying entirely upon leguminous crops and
stable manure. Others feed the trees by fertilizing the crops grown
between the rows, while some rent the land between the rows, reserv-
ing a strip along the row which they (the owners) cultivate and
fertilize independent of the rest of the land. Estimates as to the
actual cost of fertilizing the trees alone are very difficult to obtain.
Some replies indicated that from $10 to $25 per acre was paid annu-
ally for fertilizer for the land, including both that given the trees and
the intercrops. Others showed that to apply 2 pounds of fertilizer
to each tree during the season of its planting and to increase this by
1 pound per tree each year thereafter had cost thus far from $1.50
to $2 per acre.

(5) Have you found the growing of other crops between the trees to be practicable;
and, if so, about how much per acre have been the annual gross returns from such
crops?

(6) What crops have you found to be most practicable for growing between pecan
trees?

The invariable reply to these inquiries was in favor of growing
intercrops. Cotton, corn, and leguminous crops were most com-
monly recommended, although a few from truck-growing districts
reported much better returns from vegetable production. One or

[Cir. 112J

6 CIECULAR 112, BUREAU OF PLANT INDUSTRY.

two reported very favorably on growing nursery stock. Some said
that during the first five to seven years the land would be quite as
valuable for cultivated crops as open land. It is obvious, however,
that after the trees reach bearing age the value of the land between
the rows for intercropping must depreciate rapidly.

(7) Have you kept any record of the cost of cultivation? If so, about how much
has it been per acre each year, including both the cultivation of the trees and the
crops between the rows?

With but one or two exceptions the replies to this inquiry were to
the effect that no record had been kept. Such records as have been
kept include the cost of cultivating the intercrops, and it was there-
fore impossible to determine the separate cost of cultivating the
trees. One letter stated that the annual cost of cultivating the trees
would not be more than $1 per acre; another, that this cost would
not exceed $5 per acre.

The questions asked omitted a discussion of the price of the trees,
the cost of planting, the salary of a supervisor, the cost of replant-
ing dead trees, pruning, spraying, and harvesting the crop, and
many other items which must be taken into account when a com-
plete record of the cost of establishing and maintaining pecan orchards
is undertaken. As it is customary to plant from 17 to 20 trees per
acre, the cost of the trees can be quickly computed by consulting the
nurserymen's price lists. The other factors are all either very vari-
able or else few data are available.

YIELDS THAT MAY BE REASONABLY EXPECTED.

The gross returns which may be realized from an orchard at any
given age depend upon the size of the crop and the price received.
Considerable light upon the former may be obtained from a summary
of yield records already made, and such records as it has been pos-
sible to obtain are here included. The price which has been realized
in the past will, of course, have some bearing upon what may be
expected in the future, but it can not be taken as a safe criterion.
In the past pecans of named varieties have been grown in very
small quantities and have been largely utilized by nurserymen as
samples and by fancy confectioners, tourists, and land sellers. In
this way a high price has been maintained. Ordinarily the pro-
ducers have received from 30 to 50 or 75 cents a pound for nuts of
the best varieties. The demand for the best pecans can not become
general until prices settle to a uniform level within reach of the con-
sumer of average means. The present extensive planting insures
abundant production, and the need of a wide market is therefore
self-evident. With the disappearance of fancy prices, the general
demand will undoubtedly materially increase. Further, the prices

[Cir. 112]

OPPORTUNITIES IN PECAN CULTURE.

which the producer will realize in the future will depend more largely
than in the past upon the prices of wild nuts, with which the culti-
vated product must soon enter into competition. The present pre-
vailing retail price for pecans from native trees is from 20 to 30 cents
per pound, uncracked, and from 60 to 90 cents for half kernels.

Estimates of yields that may be expected must be based upon
records of orchards of the same varieties and age when under sub-
stantially the same conditions of climate, soil fertility, moisture sup-
ply, and cultivation. The difficulty in obtaining records upon which
to base such estimates has already been shown. The figures pre-
sented in Table I will serve for general guidance until more and better
records are available.

Table I.—

Yield records of pecan orchards.

W inter

planted.

Num-
ber of
trees.

Average number of pounds per tree each season.

Location and variety.

4th.

5th.

6th.

7th. 1

8th.

9th.

10th.

11th.

12th.

Southern Georgia:

Jewett, Rome, Stuart,

[■1899-1900.

1901-2....
1902-3....

•1904

250
[2 321

0.40

0.90

0.68

and Van Deman.
Southern Mississippi : '

Delmas

'0.44

0.81

5.23

12.15

Stuarl

(*)

m

( 4 )
( 4 )

0)
.78

( 4 )
.21

Pahst

70

70
12
70
70

292

1.43

3.21

2. H

3.97

5 1.55

3.28
3.28

7.71

5.28

.50

16. 42

6 3.21

7.84

Total

.50

2.53

3.2S

\

Mixed, Stuart, Schley,

?209
200

Delmas, and others.
Southwestern Georgia :

J

1904-5....

1904-5....

0.92

1.05

5.08

270

2

1

1

26

300

3.23
.20

. 12

Clark

.00

.81

Total

.08

.12

.66

3.13

ll905-6. . . .

f 1,265
68

• 413

34S

I 330

.23
.73
.09
.14

.06

Schley

Stuart

Total

2,424

1 1904-5....

.45

Northern Florida:

Mixed, Frotscher, Egg-
shell, Schley, Money-
maker, and others.

120

.48

1 In southern Mississippi two Stuart trees planted in the winter of 1889-90 bore 250 pounds each in 1911;
in 1912 the crop from these two trees fell materially.

2 This orchard was reported as having 300 young trees set in the spring seasons of 1902 and 1903 and about
7 old trees top-worked in 1902 and 1904. The old trees annually bear about three times as much as the
same number of young trees. In order to get an average per tree of the trees of one age, the top-worked
trees are counted as being equivalent to three times as many, or 21. young trees. The date of planting
the 300 trees is taken as having been 1903.

' Owing to a very hard storm in the fall of 1906 there was no crop in 1907 or 1908, the fifth and sixth
seasons from planting.
* A few nuts.

5 Schley and Stuart not included,
s Schley, Stuart, and Van Deman not included.
» Mature trees 25 to 30 years old; averaged 38. 49 pounds in 1911 and 9.57 pounds in 1912.

[Cir. 112]

8 CIRCULAR 112, BUREAU OF PLANT INDUSTRY.

The figures shown in Table I, with the exception of those regarding
the orchard of 200 trees previously mentioned, the yields of which
were recorded by the writer, have been made by careful orchard
owners, who furnished the data to the Department of Agriculture.

Another record, which has been submitted in different form from
middle Georgia, is as follows:

Stuart Few nuts fourth year; increase yearly to

ninth year 2£ to 3 pounds.
Mobile Few nuts third year; increase yearly to

ninth year 5 to 15 pounds.
Teche and Frotscher Few nuts fifth year; increase yearly to

ninth year 2 to 5 pounds.

Rome Few nuts eighth and ninth years.

Capitol Few nuts fiflh year; increase yearly to

ninth year 2 to 3 pounds.

Senator

Atlanta \ "No good.'

Centennial

CONCLUSIONS.

Pecan orchards demand the same intelligent management as other
orchards. It is sometimes held that being a native of the forest the
tree not only needs no cultivation but will do better without it.
This theory might be applied with equal reason to any fruit tree, and
in answer it is only necessary to suggest a comparison between wild
and cultivated apples, pears, oranges, or other fruits. The impres-
sion that the pecan tree has no enemies in the way of insect pests or
fungous diseases, that it is not subject to damage from drought, wet
weather, or freezing temperatures, and that the nuts are in unlimited
demand at a dollar or more a pound is at variance with the facts.
The pecan is often subject to serious injury by numerous insects and
diseases, and it is also much affected by unfavorable weather. Warm
spells in winter followed by sharp freezes not infrequently result in
the death of young trees; rains occasionally interfere with pollina-
tion; prolonged dry weather causes the nuts to be small and perhaps
to drop prematurely; and warm, wet weather may cause the nuts to
become moldy or to germinate while still in the hulls. Although in
the past choice pecans have frequently brought two, three, or more
cents a nut when sold by the pound, it is evident that when sold on
their merit in competition with wild pecans and when the orchards
now being planted reach bearing age the prices will fall materially
below the figures now often cited by enthusiastic exploiters.

The prospective pecan grower should, of course, bear in mind that
no horticultural product is free from its troubles, and while the pecan
has its full share its culture probably has no more drawbacks than

[Cir. 112]

OPPORTUNITIES IN PECAN CULTURE. 9

other similar industries. Ordinarily, commercial returns are not to
be expected until the trees are 10 to 12 years old. The length of time
trees will continue to bear is a matter of conjecture. Commonly, old
seedling trees are moderate bearers, the heavy bearers being fairly
young or of middle age. The indications are that the most produc-
tive trees of the forest are not long lived, and as the varieties selected
and propagated for planting in orchards are usually heavy bearers it
is not improbable that the shorter lived varieties are unconsciously
selected at the same time. It is therefore not unlikely that the life
of bearing pecan trees in orchard form will be much shorter than that
of the average pecan tree of the forest.
76060°— Cir. 112—13 2

THE JONATHAN FRUIT-SPOT. 1

By W. M. Scott, Formerly Pathologist, and John W. Roberts, Assistant Pathologist,

Fruit-Disease Invest igat ions .

INTRODUCTION.

In February, 1911, the senior writer 2 published a preliminary
report on "A new fruit-spot of apple," 3 in which he stated that the
cause of the disease was unknown, but that there was a strong
suspicion of injury produced by arsenate of lead used in spraying.
It was also stated that the fungi Cylindrosporium pomi Brooks and
Alternaria sp. were isolated from a few of the spots, indicating a
possible connection of one or both of these organisms with the
disease.

The results of spraying experiments and further laboratory studies
conducted by the writers show that the spots are not due to arsenate
of lead injury and probably are not caused by any vegetable organism.

THE NATURE AND IMPORTANCE OF THE DISEASE.

The spots, though seldom more than skin deep, detract greatly
from the appearance of the apple and afford a place of entrance for
decay fungi. They are dark brown in color, more or less circular
in outline, at first scarcely depressed, later becoming considerably
sunken, and vary from one-eighth to three-fourths of an inch in
diameter. (Figs. 1 and 2.) They resemble very young bitter-rot
spots and are not easily distinguished from the advanced stage of
the New Hampsliire fruit-spot (Cylindrosporium pomi Brooks). As
many as 25 spots often occur on one apple, and a lenticel usually
forms the center of each spot. Since the spots are entirely super-
ficial, the intrinsic value of the fruit is not seriously affected, but
its market value is greatly reduced.

The disease occurs only on fully matured fruit and usually develops
after the crop is picked. If left on the trees long after maturing,
the fruit of susceptible varieties may become affected before being
picked. This was observed on the Jonathan variety in Virginia

1 Issued Feb. 8, 1913.

^ The work covered by this paper was done previous to the resignation of Mr. Scott, which occurred in
February, 1912.
3 Phytopathology, vol. 1, no. 1, p. 32-34.

[Cir. 112] 11

12 CIRCULAR 112, BUREAU OF PLANT INDUSTRY.

and West Virginia during the fall of 1911. According to numerous
observations made by the writers, fruit picked at the proper time,
or rather early, and rushed into cold storage with only two or three
days' delay, and consumed within a few days after removal from
storage, will not develop the disease to any serious extent. Fruit
of susceptible varieties kept in common storage or delayed in reaching
cold storage usually becomes affected. The disease has been par-
ticularly annoying to fruit growers who have attempted to keep prize
specimens of the Jonathan in cellar storage for exhibition purposes.
The growers of Esopus (S pitzenberg) in Oregon and Washington

Fig. 1.— Esopus (Spitzeriberg) apple showing early stages of the Jonathan fruit-spot.

have perhaps suffered most from this trouble, the spots often develop-
ing on the fruit en route to the eastern markets. The writers have
observed large quantities of affected fruit from the Northwest in the
markets of Washington and New York.

The Jonathan is the most susceptible variety grown in the east,
and its commercial standing is greatly impaired on account of this
weakness. The disease is now rather commonly known among
apple growers as the "Jonathan spot," and for that reason the writers
have adopted the name "Jonathan fruit-spot." The Esopus is
almost, if not quite, as susceptible to the disease as the Jonathan,
and the Yellow Newton apparently ranks third in degree of suscepti-

[Cir. 112]

JONATHAN FRUIT-SPOT.

13

bility. It has also been observed to a very slight extent on the
Grimes, Arkansas Black, and a few other varieties of less importance.
Dry weather during the summer is apparently favorable to the de-
velopment of the Jonathan fruit-spot. It was very bad in 1910 and
1911, both of which were dry seasons, while in 1912, a comparatively
wet season, it was not common on eastern-grown fruit. In the fall
of 1911 the spotting was particularly serious on the Jonathan, speci-
mens having been received from practically every section of the
country where that variety is grown.

Fig. ?..— Esopus (S pitzenberg) apple showing older stages of the Jonathan fruit-spot.

SPRAYING AND STORING EXPERIMENTS.

In order to test the supposition that the Jonathan fruit-spot
might be due to arsenical injury, spraying experiments were con-
ducted in the orchard of Air. S. II. Derby, at Woodside, Del., during
1911. A block of Jonathan apple trees about 15 years old was
divided into 5 plats of 6 trees each and treated as follows:

Commercial lime-sulphur solution at the rate of \\ gallons to each 50 gallons of

water was used in connection With arsenate of lead on all of lite sprayed plats. The

amourjt of arsenate of lead was varied from one-half pound to 5 pounds in each 50

gallons of spray. Plat I was sprayed with one-half pound, Plat II with 1 pound,

[Cir. 112]

14 CIRCULAR 112, BUREAU OF PLANT INDUSTRY.

Plat III with 2 pounds, and Plat IV with 5 pounds of arsenate of lead to each 50
gallons of the diluted lime-sulphur solution. Three applications were made in
accordance with the usual directions for the control of the codling moth, i. e., (1) as
soon as the petals fell, (2) three weeks later, and (3) ten weeks after the petals fell.
The trees were thoroughly sprayed each time, so that the fruit remained coated with
the arsenate of lead well on toward picking time. Plat V was left unsprayed as a
check.

The crop was picked on September 12 and found to be practically
free from insects and diseases. No spotting was discernible at this
time. Two boxes of fruit from each plat were immediately sliipped
to Washington, reaching the laboratory on September 15, three days
after picking. On this date a careful examination revealed no
indication of the disease on any of the fruit, sprayed or unsprayed.
One box from each of the five plats was then placed in cold storage,
while the remaining five boxes were stored in a moderately cool
basement.

The basement-stored apples were examined on September 30 with
the following results: The fruit from Plat I showed 41 per cent
affected with the Jonathan fruit-spot, Plat II 52 per cent, Plat III
36 per cent, Plat IV 36 per cent, and Plat V (check) 46 per cent. A
reexamination of the same apples on October 23 showed Plat I to
have 56 per cent of the fruit affected, Plat II 70 per cent, Plat III 52
per cent, Plat IV 42 per cent, and Plat V (check) 64 per cent. Many
of these apples were seriously injured, being literally covered with
spots measuring from 5 mm. to 1 cm. in diameter. These results
show that unsprayed fruit may become quite as badly affected with
the Jonathan fruit-spot as fruit sprayed with arsenate of lead. An
unusually heavy dose of the poison, as shown in the results from
Plat IV, which was sprayed with 5 pounds of arsenate of lead to 50
grallons of water, did not increase the amount of affected fruit.

The fruit which was placed in cold storage was examined on
November 10 and all of it found to be free from the disease. Finally,
on December 18 these apples were removed from cold storage and
examined with the following results: Plat I had 5 per cent of its
fruit spotted, Plat II 10 per cent, Plat III 20 per cent, Plat IV 14
per cent, and Plat V (check) 33 per cent. In most cases the spots
were small, inconspicuous, and few to an apple, being in these respects
in great contrast to those appearing on the basement-stored fruit.
The cold storage prevented the spotting for at least two months,
and at the end of nearly three months tins fruit was not nearly so
much affected as the cellar-stored fruit was at the end of six weeks.

On September 25, 1911, one bushel of unsprayed and one bushel of
sprayed Jonathan apples were received from Watervliet, Mich. These
were sent in by Mr. E. W. Scott, of the Bureau of Entomology; they
were taken from an orchard in which that bureau was conducting spray-

[Cir. 112]

JONATHAN FRUIT-SPOT. 15

ing experiments. The plat from which the sprayed fruit was taken had
received the tliree usual codling-moth applications, arsenate of lead
at the rate of 2 pounds to each 50 gallons of water having been used.
Upon arrival an examination of this fruit failed to disclose any of the
spot disease in either lot. Both lots were covered over in baskets
and left in the laboratory at room temperature and reexamined on
September 29. At this time characteristic spots averaging 5 mm.
in diameter and from 1 to 25 to each apple were found on 9 per cent
of the unsprayed and on 18 per cent of the sprayed fruit. On October
23, 65 per cent of the unsprayed fruit was found to be spotted and
66 per cent of the sprayed fruit was similarly affected.

One can only conclude from the results of these experiments that
spraying with arsenate of lead is not in any way responsible for the
Jonathan fruit-spot. The spots develop on unsprayed fruit as readily
as on that which has been thoroughly sprayed with arsenate of lead.
It is evident that tins poison neither favors nor retards the develop-
ment of the disease.

LABORATORY STUDIES.

Nearly 400 cultures of the diseased spots have been made in various
ways and on various media, but no organism has been isolated with
any degree of consistency. A species of Alternaria often occurred
in cultures from fruit grown in the eastern part of the country, but
cultures from northwestern-grown fruit were almost entirely barren.
A few apparently successful inoculations were made by spraying
Alternaria spores on Jonathans kept in moist chambers and the
fungus reisolated, but both the Jonathan and Esopus (Spitzenberg)
are so susceptible to the disease that they are apt to become spotted
under any conditions outside of cold storage. In some cases both
the inoculated fruit and the controls contracted the disease at about
the same time. Spores of this fungus inserted through needle
punctures failed to produce the disease. As Alternaria is very
commonly associated with the rotting of apples, especially when the
fruit is placed in cellar or basement storage, the possibility of its being
the cause of this disease becomes very remote.

The fungus Cylindrosporium pomi Brooks occurred in a few of the
cultures, but this was probably accidental. It is not unlikely that
the Brooks spot and the Jonathan fruit-spot occurred together on
some of the apples from which cultures were made, and for this
reason the fungus causing the former might easily have found its
way into a few of the cultures, particularly since the two spots are
somewhat similar in appearance. Cultures from the true Brooks
spot produced the fungus readily, while those from the Jonathan
fruit-spot were, with few exceptions, barren. Moreover, spraying
[Cir. 112]

16 CIRCULAR 112, BUREAU OF PLANT INDUSTRY.

with a fungicide prevents the former disease, but has no effect upon
the latter. It seems evident, therefore, that these two diseases are
distinct.

Microscopic examinations of the affected tissues failed to reveal
the presence of any organism to which the disease could be attributed.
The cells involved resemble similarly located cells in cases of "bitter-
pit," or "Baldwin spot," in which that disease extends to the
surface of the apple. Bitter-pit differs from this disease, how-
ever, in that it is essentially a disease of the fleshy portion of the
fruit, often reaching to the core without affecting the skin, while the
"Jonathan spot" is usually little more than skin deep. The writers
consider the disease a physiological one, but as in the case of the
bitter-pit the cause is at present obscure.

SUMMARY OF CONCLUSIONS.

The investigations conducted by the writers seem to warrant the
following conclusions :

(1) The Jonathan fruit-spot of the apple is due neither to spraying
with arsenate of lead nor to a specific organism.

(2) It is probably a physiological trouble, falling in the same
category as the bitter-pit or Baldwin spot.

(3) Early picking, prompt cold storage, and immediate consump-
tion of the fruit after removal from storage, will largely obviate
losses from the disease.

[Cir. 112]

EGYPTIAN COTTON AS AFFECTED BY SOIL VARIATIONS. 1

By Thomas H. Kearney, Physiologist in Change of Alkali and Drought Resistant

Plant Investigations.

INTRODUCTION.

Observation of the growth of Egyptian cotton in irrigated soils of
the southwestern United States for several years past has shown that
this plant is decidedly sensitive to variations in its physical environ-
ment, Differences in the texture, and consequently in the moisture-
holding capacity of the soil, are easily detected from the accompany-
ing differences in the size, appearance, and f ruitfulness of the plants, in
the size of the bolls, and in the quality of the fiber. The presence of
alkali salts in the soil also induces marked differences in the growth
and behavior of the cotton plants. It is evident that in order to
obtain the largest yields and, what is of the utmost importance, the
greatest possible uniformity in the staple, strength, and other quali-
ties of the fiber, Egyptian cotton must be grown in soils that do not
vary greatly in texture and salt content.

MOISTURE CAPACITY OF THE SOIL.

In many plantings of Egyptian cotton which have been made on
recently cleared land the fields have appeared more or less spotted,
the plants in some places being smaller, more erect, and lighter col-
ored, with fewer and smaller bolls and shorter, often weaker, fiber
than in other places. Marked differences of this kind frequently
appear within distances of a few feet. A field of the Yuma variety
on the United States Experiment Farm at Bard, Cal., in 1911 showed
conspicuous local differences in the growth and appearance of the
plants. Soil samples were therefore collected at a number of differ-
ent points corresponding to various stages in the size and condition
of the plants.

Upon making the borings it was at once evident that the variations
in growth of the plants were closely correlated with variations in the
depth of the blanket of silt loam which overlaid a subsoil of coarse,
light-colored sand. The depth of the silty layer varied from 5 to 18
inches in different parts of the field. Where it was thinnest the plants
were poorest, and vice versa.

The moisture-holding capacity of these two soils was widely dif-
ferent, that of the silt loam being high, as indicated by a moisture

i Issued Feb. 8, 1913.

17

18

CIRCULAR 112, BUREAU OF PLANT INDUSTRY.

equivalent, 1 in samples from different parts of the field of 24.4 to 31.4
per cent, corresponding to a wilting coefficient of 13.3 to 17 per cent.
The coarse sand had a moisture equivalent of only 2.8 per cent
(wilting coefficient, 1.5 per cent). The salt content of the soil in the
different samples was investigated by means of the electrolytic bridge
and was found to be nowhere sufficiently high to indicate that alkali
was a factor in bringing about these differences of growth. The case
was clearly one of the presence or absence of a sufficient depth of
soil having a high enough water-holding capacity to prevent the
plants from suffering as a result of drought between irrigations. It
is rather surprising, in view of the very low water capacity of the
underlying sand, that a depth of the silty layer of only 14 or 15
inches should have been sufficient to enable the plants to make a
strong growth and produce numerous large bolls.

The results of this investigation are summed up in Table I.

Table I. — Relations between the depth of silt and the growth of Egyptian cotton at

Bard, Cat., in 1911.

Boring
No.

Depth of

silt loam

overlying

coarse

sand. 1

Condition of the plants.

6

1

5

2

4

3

7

8

Inches.

2 18

17

14 to 15

12

11

9

3 7 to 8

5

Rank growing, dark green, very fruitful, bolls large; growth here is heaviest in field.
Good sized, dark green, fruitful.
Similar to those at boring No. G.

Smaller, lighter green, and less fruitful than at boring No. 1.
Small, erect, light colored, with few small bolls and inferior fiber.
Similar to those at boring No. 4.
Similar to those at borings Nos. 3 and 4.

Poorest in the field, only 3J feet high, erect, light colored; bolls very few and very
small, fiber very inferior.

1 Unless otherwise specified, the silt rested directly upon the rather coarse, light-colored sand. This
sand extended to a depth of at least 4 feet at every boring which was carried to that depth (Nos. 1, 3, 5,
and G).

2 Here the silt was underlain by 15 inches of a fine, reddish-colored sand (moisture equivalent, 7.6 per
cent), which in turn rested upon the above-described coarse sand.

3 Here the surface silt was of lighter texture than elsewhere in the field (moisture equivalent, 24.4 per
cent). It rested upon G to 7 inches of very fine sand, which was in turn underlain by the coarser sand
above described.

In 1912 similar observations were made on the same farm. A
series of soil samples were taken midway between two rows of
Egyptian cotton, in one of which the plants had been thinned to a
distance of 6 inches and in the other to a distance of 18 inches. 2
Borings were made at three points: (1) Where the plants were tall,
luxuriant, and dark green in color; (2) where the plants were smaller
and of a light yellowish green color; and (3) where the plants were

1 The term "moisture equivalent" is defined and the value of this factor as a measure of the moisture-
holding capacity of the soil is pointed out by Briggs and Me Lane in Bulletin 45 of the Bureau of Soils (1907).
The moisture equivalent of a given soil being known, the wilting coefficient for plants growing in that soil
can be calculated by means of the formula given by Briggs and Shantz in Bulletin 230 of the Bureau of
Plant Industry, p. 58 (1912).

All determinations of moisture equivalent referred to in this paper were made by Mr. J. W. McLane,
of the Biophysical Laboratory, Bureau of Plant Industry.

2 The field had been planted under the direction of Mr. O. F. Cook in order to study the effect of dif-
ferent thicknesses of stands upon the development of the vegetative branches.

[Cir. 112]

EGYPTIAN COTTON AS AFFECTED BY SOIL VARIATIONS.

19

small and yellowish green. Each of the three samples consisted of
three cores taken about 1 foot apart and to a depth of 4 feet. The
soil from all three cores at each successive 1-foot depth was thoroughly
mixed together to represent that depth of the boring in question,
and upon each of the 12 samples as thus prepared four determina-
tions of moisture equivalent were made.

The electrical resistance of saturated soil from each sample was
measured in order to ascertain whether there were significant differ-
ences in the salt content of the soil. The high resistances observed
in every case made it evident that the effects noted could not be
attributed to alkali. On the other hand, the fact that the resistances
were much lower at all depths of boring No. 1 than of borings Nos.
2 and 3 indicated that a deficiency of nutrient salts at the two latter
borings may have been a factor in the poor growth of the plants.

On several plants in the neighborhood of each boring counts were
made of the number of the node on the axis at which the first fruiting
branch was retained and of the number of set bolls to the plant. 1

Table II gives for each of the three borings the wilting coefficient
(calculated from the moisture equivalent) and the electrical resist-
ance of the soil at successive depths, as well as the average height of
the plants, the mean of the numbers of the node bearing the first
fruiting branch, and the mean number of bolls per plant.

Table II. — Relations between the wilting coefficient of the soil and the growth and
fruitfulness of Egyptian cotton plants at Bard. Cal., in 1912.

Soil.

Plants.

Boring
No.

Depth.

Willing
coefficient
calculated
from the
moisture
equiva-
lent. 2

Electrical
resistance

of the
saturated

soil.

Spacing.

Number

of plants

on which

counts

were

made.

Color of foliage.

Height.

Mean
number
of node
of first
fruiting
branch.

Mean

number

of bolls

per plant.

1

Feet.

1 !
1 i
1 §

Per cent.
11.7
8.2
4.5
1.6
2.8
2.0
1.5
1.2
1.8
1.2
1.3
1.2

Ohms.
427
271
460

968
1,370
1,318

2,108

Inches.
6

7

Dark green

Feet.
8

15.1

65.4

18

4

•Dark green

8

17.2

78.0

2

6

11

Light green . . .

4.5

18.5

21.2

18

5

Light green . . .

4.5

17.0

38.0

3

1,353
2,164
2.043
2.164

6

7

Light green . . .

2.5

15.4

10.6

18

5

Light green...

2.5

15.2

15.0

< The counting was done by Mr. Rowland M. Meade. The importance of retention of the fruiting
branches at low nodes on the axis as an indication of fruitfulness is pointed out by Mr. Argyle McLachlan
in Bulletin 249, Bureau of Plant Industry, 1912, entitled "The branching habits of Egyptian cotton."

2 Based upon thoroughly mixed samples of each 1-foot depth, regardless of variations of texture within
that depth. The variations are indicated by the following notes:

Boring No. 1. Soil silty from the surface to a depth of 14 to 20 inches (varying in the several cores), then
fine sand, with some admixture of silt at a depth of 24 to 30 inches, then coarse sand from the depth of 30
to 48 inches.

Boring No. 2. First 12 inches much sandier than at boring No. 1. Coarse sand began at a depth of 20
to 24 inches and continued to the bottom of the boring (48 inches).

Boring No. 3. Soil sandy to the surface. Coarse sand began at a depth of about 18 inches and continued
to the bottom of the boring.

[Cir. 112]

20 CIRCULAR 112, BUREAU OF PLANT INDUSTRY.

These results confirm those of the previous year as to the effect of
differences of soil texture upon the size and vigor of the plants, and
also give a more definite expression of the effect upon the fruitfulness
(number of bolls). There was apparently no consistent effect upon
the height of the first fruiting branch.

It is evident, therefore, that in the absence of appreciable quantities
of alkali salts, the size, vigor, and fruitfulness of Egyptian cotton
plants in alluvial soils along the Colorado River is largely determined
by the depth of the layer of silt with its relatively high moisture
capacity and its doubtless greater supply of plant food.

The very sandy soils have so low a moisture capacity that they
hold very little water even immediately after an irrigation and with
the customary intervals between irrigations they soon become so dry
that much of the time the plants are without available water. This
frequent condition of virtual drought seriously impairs the yield and
quantity of fiber produced in such soils.

ALKALI CONTENT OF THE SOIL.

In a previous publication l the results of observations on the
growth of Egyptian cotton plants hi alkali soil at Sacaton, on the
Gila River, hi southern Arizona, were summed up as follows:

No plants grew at Sacaton in places where the average amount of alkali in the first
3 feet of soil was as high as 1.7 per cent. While resistant individual plants can pro-
duce a small amount of fairly good fiber in the presence of from one-half to 1 per cent
of alkali, it is probable that land containing considerably less than one-half of 1 per
cent must be selected in order to obtain anything like a full stand and the best quality
of fiber. The actual limit of safety remains to be determined.

Since this statement was written, further observations upon the
alkali resistance of this plant were made at Sacaton, Ariz., hi 1910,
and at Bard, Cal., hi 1911. Even with these additional data the
quantity of alkali of a given composition which limits the successful
growing of Egyptian cotton can be stated only approximately. 2

In a general way, however, the more recent observations confirm
the conclusion previously readied that Egyptian cotton is superior to
many other crop plants in its ability to endure an excess of salts hi
the soil.

OBSERVATIONS AT SACATON, ARIZ. 3

A portion of one of the fields planted to the Yuma variety of
Egyptian cotton was located where the soil contained more or less
salts and where as a consequence the stand was very irregular. The

i Circular 29, Bureau of Plant Industry, entitled "Experiments with Egyptian cotton in 1908," 1909,

p. 18.

2 A discussion of the varying factors which make it difficult to determine in the field the limits for the
growth of a particular plant in the presence of alkali will be found in Farmers' Bulletin 446, entitled " The
choice of crops for alkali land," 1911, pp. 8 to 12.

3 These were made in October, 1910, at a time when the cotton had begun to ripen in most parte of the
field.

[Cir. 112]

EGYPTIAN COTTON AS AFFECTED BY SOIL VARIATIONS. 21

cotton rows in this field began in good soil where the growth of the
plants was normal and where the yield and character of the fiber
were as good as could be expected considering the late date of planting
(April 12). Toward the ends of the rows where the soil contained
gradually increasing quantities of alkali salts the plants were small,
produced few bolls, and ripened very late. Finally no plants
remained where the quantity of alkali was excessive.

In two of these rows the soil was sampled to the depth of 3 feet
at the base of every tenth plant from the good end of the row until the
region was reached where the stand was much interrupted. There
samples were taken at more frequent intervals and finally alongside
each remaining plant.

The soil from each 3-foot boring was thoroughly mixed and its
electrical resistance when saturated was determined by means of
the electrical bridge, the standard container of which has a capacity
of about 50 cubic centimeters. 1 From the resistances, the percentage
of total salts to dry weight of soil were computed by means of a special
correlation curve for the Sacaton type of alkali. This curve was
based upon determinations of the electrical resistance and of the
total water-soluble salts hi 48 samples of soil which had been col-
lected two years previously on the experiment farm at Sacaton.
Analysis of these samples hi the chemical laboratory of the Bureau
of Soils showed the average composition of the Sacaton alkali to be as
follows: '-'

Per cent. I'er cent.

Ca 1.0

Mg 3

K 1.0

Na 31.9

CI 24.4

S0 4 16:4

HC0 3 22.1

C0 3 5.0

Notes were made on the condition of each plant where a soil sample
was taken and the seed cotton was collected for examination in the
laboratory. Mr. Argyle McLachlan made a series of diagrams for
each of these plants, showing graphically the number and position
of the fruiting branches and the number of developed and aborted
bolls on each branch. Comparison of these diagrams indicated that

1 The bridge is described and figured and directions are given for its use in determining the salt content of
soils in Bureau of Soils Bulletins 15 (by L. J. Briggs) and 01 (by R. O. E. Davis and H. Bryan). In the
latter publication tables are given for temperature correction (pp. 22 to 24) and for computing from the
resistance the percentage of total salts to dry weight of the soil (pp. 14 to 16).

2 This composition of the alkali, and especially the presence of carbonates and of large quantities of
bicarbonates, probably explains the fact that the Sacaton curve does not agree with that on which is based
Table III, p. 14, in Bulletin 61 of the Bureau of Soils. The latter applies to alkali consisting of one-half
chlorids and one-half sulphates. For resistances below 4() ohms it indicates higher percentages of total
salts and for resistances above 40 ohms it indicates lower percentages than does the Sacaton curve. The
disparity increases until a resistance of 170 ohms indicates twice as much total salts on the Sacaton curve
as on the chlorid-sulphate curve. On the other hand, the curve for carbonates (" black alkali") on which
is based Table VI, on p. 10, Bulletin 61 of the Bureau of Soils, indicates much higher percentages of total
salts for given resistances than on either of the other curves until a resistance of 170 ohms is reached, from
which point the Sacaton curve agrees very closely with the carbonate curve.

[Cir. 112]

22

CIKCULAE 112, BUREAU OF PLANT INDUSTRY.

there was a tendency for the plants growing in soil relatively free
from alkali to retain the first fruiting branch at a lower node on the
axis than in the case of plants in the stronger alkali soils, although
the correlation between the height of the first fruiting branch and
the alkali content of the soil was not a close one.

At most of the borings the resistance of the saturated soil ranged
from 200 to 400 ohms. Where it exceeded 200 ohms no differences
which were not well within the limits of individual fluctuation
could be detected in the plants. Table III summarizes the notes
made chiefly upon plants growing where the soil gave resistances
lower than 200 ohms.

Table III.— Electrical resistance of saturated soil, indicated percentage of total salts, and
character of the plants ami fiber of Egyptian cotton at Sacaton, Ariz., in 1910.

Soil (averages for
the 4-foot column).

Plants.

Electri-
cal resist-
ance in
ohms.

Indicated
percent-
age of to-
tal salts
(Sacaton
curve).

Size, earliness, and fruit fulness.

Number of
the node
on axis
where the
first re-
tained
fruiting
branch
occurs.

Bolls.

Fiber.

+200
200
185

155

140

110

100

90
70

-0.34
.34
.36

.41

.43

.48

.51

.54
.63

10 to 16

12 to 17

13 to 15

16 to 17
12

18

16

17
18

Normal

Normal.

do

do

Do.

Ripening somewhat retarded,

otherwise normal.
Smaller and later ripening

do

Do.

Normal or rather

small.
Small or medium. .

Small

Rather short, other-
wise normal.
Good in length,

Late ripening, fairly produc-
tive.

Small, late ripening, with few
bolls.

do

strength, and
quality .
Normal.

..do

Short, but strong

Very small

Fair sized

and fine.
Normal.
None matured.

The data given in Table III indicate that with alkali of the Sacaton
type, where the salt content of the soil exceeds 0.4 per cent of its
dry weight (electrical resistance 150 ohms or lower), the fruitfulness
of the plants is likely to be impaired and the ripening of the bolls
seriously retarded. This would seem to be about the limit for profit-
able production of this crop in the presence of alkali of the Sacaton
type, although it is evident that the quality of the fiber does not
necessarily suffer in the presence of 0.55 per cent of salts (electrical
resistance of 90 ohms).

OBSERVATIONS AT BARD, CAL.

A small field of Egyptian cotton grown in the vicinity of Bard
in 1911 was located on a sandy soil containing in spots so much
alkali that the resulting stand of cotton was very uneven. There

[Cir. 112]

EGYPTIAN COTTON AS AFFECTED BY SOIL VARIATIONS.

23

were several areas of greater or less size where the plants either had
failed to appear or had subsequently died. The electrical resistance
of the soil was determined in different parts of this field on October
23, and notes were made upon the character of the plants where the
respective soil samples were taken. The results of these observa-
tions are summarized in Table IV. The percentages of total salts
indicated by the electrical resistances as given in this table are
computed from Table III, page 14, Bulletin 61 of the Bureau of Soils,
which applies to a type of alkali consisting of one-half chlorids and
one-half sulphates.

Table IV. — Electrical resistance of the saturated soil, indicated percentage of total salts,
and condition of Egyptian cotton plants, at Bard, Cal., in 1911.

Bor-
ing
No.

Depth.

Character of
soil.

Elec-
trical
resist-
ance.

Indicated

total
salts in
percent-
age of dry
weight
soil.

Condition of plants.

1

Fat.

i 1

2

3

4

Average.

Sandy loam...

do

...'..do

Fine sandy
loam.

Ohms.

45

(53

113

117

0.75
. 53
.29

.28

No plants; boring in center of a small bare spot.

.46

{ I

i 1
2
3

4

Average.

Sandy loam...
do

Sandv loam...
do

Fine sandy

loam.
do

33
63

63
91
62

59

.'

1.12

.53

1 Between two good -sized, healthy, fruitful plants bear-
> ing abundant fiber of excellent quality, strong, 1|
1 inches long.

3

.53
.37
.53

.56

Among small scattered plants (less than half normal
height) mostly dead. Plants shallow-rooted here.

.50

1

2
3

4

A verage.

Sandj' loam. . .

do

Sand

55
50
69

85

.60
.67
.48
.39

Between two good-sized, healthy, fruitful plants bear-

4

do

> ing strong, abundant, silky fiber. Root system
comparatively shallow.

.53

.60
.41
.45
.34

5

1
2
3
4

Average

Sandv loam. . .

do

do

Sand

55

81
75
97

Beside a dying plant at edge of spot in center of
which boring No. 3 was made. This plant bore a
few open bolls, the fiber in which was coarse and

weak.

.45

Inspection of this table shows no close relation between the salt
content of the soil and the growth of the cotton plants. Thus, at
boring No. 2, located midway between two plants which were about
as good as any in the field, the first foot of the soil contained con-
siderably more soluble salts than the first foot of borings Nos. 1, 3,
and 5, where there were either no plants at all or the plants were evi-
dently suffering. 1

1 Much of this alkali had doubt less accumulated in the upper soil after the cotton was planted, the groimd
water table in this field having reached the surface of the soil during the high-water stage of the Colorado
River in June. It had lowered by the date when these borings were made, saturated soil having been
encountered at a depth of about 4i feet in the neighborhood of boring No. 3.

[Cir. 112]

24 CIRCULAR 112, BUREAU OF PLANT INDUSTRY.

The only conclusion that may safely be drawn from the foregoing
data is that when other conditions are favorable plants of Egyptian
cotton can remain hi good condition and produce numerous bolls
and strong, abundant fiber of good length and quality where the soil
to a depth of 4 feet contains as much as one-half of 1 per cent of
water-soluble salts of this composition.

The apparent ability of Egyptian cotton plants to withstand more
alkali at Bard than at Sacaton is perhaps partly to be ascribed to
accumulation of the alkali in the upper soil at the former locality
after the plants had reached an advanced stage of growth. The
different composition of the salts at the two localities is doubtless
also partly responsible for the difference.

The average composition of the alkali on the experiment farm at
Sacaton is given on page 2 1 . The alkali in the field at Bard where the
above-described observations were made had the following average
composition:

Per cent.

Ca 8.4

Mg 4.

K (included with Na).

Na 37.1

Per cent.

CI 33.0

S0 4 21.0

HC0 3 •---. 15-6

C0 3 None.

At Bard, none of the very injurious free carbonates (" black alkali")
was detected, while considerable quantities were present in some of the
Sacaton samples. Moreover, there was a much higher proportion
of lime (Ca) in the Bard samples, and this substance, as is well known,
is very effective in neutralizing the poisonous effects of the sodium
salts which form the bulk of the alkali at both localities.

CONCLUSIONS.

The moisture capacity of the soil is an important factor hi deter-
mining the size, vigor, and fruitfulness of Egyptian cotton plants.
A larger supply of nutrient salts in the heavier soils is probably also a
factor. With irrigation as ordinarily practiced hi the Southwest,
very sandy soils, having a low moisture capacity, are unsuited to this
crop, since the plants are exposed to virtual drought during much of
the period between irrigations. Recurring deficiencies of available
water hi the soil are very unfavorable to the yield and quality of the
fiber. New land as a rule should be avoided hi growing Egyptian
cotton, as the soil commonly varies greatly in moisture capacity
and the crop produced will be correspondingly lacking hi uniformity.

The alkali resistance of Egyptian cotton is relatively high when
other conditions are favorable. It would appear that fair yields of
fiber of good commercial quality can be obtained where nearly one-
half of 1 per cent of the total dry weight of the soil consists of readily
soluble alkali salts, provided that carbonates ("black alkali") are
absent or form only an inconsiderable proportion of the total alkali.
[Cir. 112]

RELATION OF STAND TO YIELD IN HOPS. 1

By W. W. Stockberger, Physiologist, and James Thompson, Scientific Assistant,
Office of Drug-Plant, Poisonous-Plant, Physiological, and Fermentation Inves-
tigations.

INTRODUCTION.

Among many hop growers the impression prevails that the average
yield of hops per acre is annually growing less and that the produc-
tivity of a large proportion of the hop soils is decreasing. The
statistical data on this point, however, are so meager that it seems
unwise to draw from this source very definite conclusions regarding
the increase or decrease in yield per acre. From the records of the
United States Census the average yield per acre of hops can be
determined only at 10-year intervals throughout a period of 30
years, and since the figures for any given year will vary widely,
depending on whether a light or a heavy crop is produced, it is mani-
festly unsafe to assume that the averages for the census years neces-
sarily represent actual conditions for the intervening years. If
records of the average yield were available for each one of the 30
years the evidence of the figures might be accepted as a fair indica-
tion of the general trend of the yield of this crop.

The average yield per acre is materially influenced by a number
of factors, prominent among which are seasonal conditions, soil type,
and cultural methods. In case large areas are under consideration,
such as a county or State, extensive changes in acreage or a shifting
of the area of production may also materially affect the average
yield. When such modifications take place, changes in the average
yield reported for the given area have little bearing on the question
of diminishing crop yields. Nevertheless, the statistical data on the
average yield per acre in the several hop-growing States and in the
chief hop-producing counties therein are worthy of careful considera-
tion by every grower of hops, but it is of far greater importance that
he should be fully informed as to the successive yields of his own
fields.

On certain types of soil not so well adapted to hops as the richer
alluvial soils there is ample evidence of a declining yield, due funda-
mentally to soil conditions. This decline is most noticeable in hop-
yards located on uplands where beneath the shallow surface soil

i Issued Feb. 8, 1913.
[Cir. 112] 25

26 CIRCULAR 112, BUREAU OF PLANT INDUSTRY.

there is a layer of hardpan and clay. On the other hand, it is far
from clear that the diminished output per acre reported for some of
the rich, deep alluvial soils is due to their decreased productiveness,
especially if such soils are overflowed in winter and thereby receive
a deposit of sediment. The fact that the application of commercial
fertilizers to some of these soils has as a rule yielded negative results
seems to indicate that they are not lacking in available plant food,
and a study of the other factors concerned will probably reveal the
most important causes of the decline in yield, if such is actually
taking place.

It is the purpose of the present publication to direct attention to
the often unappreciated extent of the losses due to imperfect stand
and to offer certain suggestions which, if followed, should result in
an increased yield without materially increasing the cost of pro-
duction.

CAUSES OF IMPERFECT STANDS.

In newly planted yards a small percentage of missing hills may
normally be expected, owing to the failure of some of the cuttings to
strike root. In most cases, after a yard has come into full bearing
the stand tends to become poorer and poorer through the dying out
of the plants from causes at present imperfectly understood. This
dying out occurs in all the hop-growing sections of the United States,
but it is far more prevalent in some districts than in others. Many
ingenious explanations have been offered to account for this trouble,
but a satisfactory one yet remains to be found. From extensive
observations made in the hop fields of the United States and of
Europe the writers have reached the tentative conclusion that a
primary cause lies in too severe or faulty pruning, in the bruising of
the roots in plowing, and in the crushing of the crown of the plant by
the feet of horses and the wheels of wagons when teams are driven
over the fields.

Hills often die out because of weakness or disease induced either
by the rough treatment received when they are uncovered at pruning
time or by the injuries inflicted by the plow or other implements used
in cultivation. When the roots are bruised or torn they heal slowly
and imperfectly and are almost certain to become infected by some
of the destructive organisms widely distributed in the soil.

In most hopyards some attention is given each year to replanting
the missing hills, but, since the trouble is rarely taken to make
certain that the cuttings are sound and vigorous and that they come
from hills selected for their thriftiness and high yield, many replants
either die outright the first year or maintain a struggling and un-
profitable existence. The vigor of the cuttings is often impaired
through the lack of precaution to keep them from drying out before

[Cir. 112]

RELATION OF STAND TO YIELD IN HOPS. 27

they are set, or the replanting is deferred until the soil has become so
dry that it does not afford the conditions essential to proper growth.
Replants usually make a poor growth unless the site of the missing
hill which they are intended to replace is excavated, the dead crown
and roots removed, and the soil replaced by fresh earth taken from
midway between the rows.

The stand may become imperfect through numerous other causes,
but the ones here described should receive first consideration, since
it is within the power of the hop grower to minimize in great measure
their effect.

VARIATION IN THE PERCENTAGE OF PERFECT STAND.

The percentage of perfect stand varies widely and is to a large
extent dependent upon the local conditions affecting a given hop
field and upon the knowledge, skill, and industry of the hop grower.
In some yards which have come under the writers' observation a
careful count of the missing hills showed the stand to be 99.3 per cent,
while in other yards the stand was found to be as low as 75 per cent.
These, of course, represent extreme cases and are far less numerous
than those in which the stand ranges from 90 to 95 per cent for indi-
vidual fields. The percentage of stand for any given yard will be
found to fluctuate from year to year, according to the rate at which
the hills are dying out and the care and attention given to replanting.

The estimate by inspection of the number of missing hills and the
percentage of stand have been found to be very misleading. In every
case in which a grower's estimate of the percentage of stand has been
verified by an actual count of the missing hills, his estimate has
proved too high, and it is believed that growers often deceive them-
selves as to the extent of the loss suffered through an imperfect stand.
An estimate of the percentage of stand that is based on a count of
the missing hills in every fifth or tenth row, although less accurate
than a full count, is much to be preferred to one based on inspection
alone.

VARIATION IN STAND ON A SINGLE ACRE.

An exact record of the stand on an acre for 4 consecutive years
shows some striking variations which are believed to be fairly repre-
sentative of the conditions existing in many hopyards. This acre
was laid off at one side of a large field which had been under hops
continuously for 10 years, and during the 4 years it was under obser-
vation it received the same attention and culture treatment as the
remainder of the field of which it forms a part. At harvest time each
year a record was made of the condition of each hill, and the position
of each hill that was missing or which had vines bearing no hops was
noted on a chart. From this chart the data in Table I were compiled.

[Cir. 112]

28

CIRCULAR 112, BUREAU OF PLANT INDUSTRY.

Table I. — Comparison of the stand of hop plants on 1 acre for the years 1909 to 1912,

inclusive.

Factors of variation.

Productive hills

Missing hills

Hills having vines with no hops

Hills having " bastard " vines

Hills having male vines

Total

Stand per eenl

Productive stand do. .

1909

1910

1911

1912

853

56

43

5

10

865

66

21

10

5

887
24
50

790
113

58

6

6

967

967

967

967

94.2
89.1

93.1
89.9

97.5
92.2

88.3
82. 2

With a perfect stand, under the system of planting followed on
this acre, there would be living plants in each of the 967 hills. Owing
to the prevalence of missing hills, however, the stand has been more
or less imperfect each year, as shown by the percentages given in
Table I. Aside from the missing hills the crop is further influenced
by the constant occurrence of unproductive plants. Of these, there
are three classes : The male plants, of which a small number is con-
sidered essential by American hop growers; the "bastard," or mon-
grel, plants, the frequent occurrence of which is restricted to certain
localities; and the normal female plants which from one of several
causes are nonproductive. When yield is considered, the non-
productive as well as the missing hills must be taken into account,
since the yield per acre is directly proportional to the number of
productive hills. The percentage of productive stand, by which is
meant the percentage of bearing hills, is an important inaex of produc-
tiveness, and on the acre in question tins figure shows, as may be seen
from Table I, that each year about one-tenth of the hills are wholly
unproductive.

The records of the individual hills show some of the important
reasons for the variation from year to year in the number of missing
hills. The two chief causes of this variation are the more or less
successful yearly replanting and the annual occurrence of new
missing hills. The variation in these factors is numerically expressed
in Table II.

Table II. — Annual variation in the number of replanted and missing hills of hops on

1 acre.

Factors of variation.

1909

1910

1911

1912

No record.
...do

12

57

21

r

44
22

9

15

3

...do

110

56

66

24

113

■

[Cir. 112]

EELATION OF STAND TO YIELD IN HOPS. 29

The importance of replanting and the success with which it has
been carried out on the acre under observation may readily be
judged from Table II. In 1910 less attention was bestowed upon
the replanting than in the two succeeding years, with very obvious
results. Were it not for the continuous dying out of the hills an
almost perfect stand could readily be attained. It is important to
note that each year there was added to the list of missing hills a
number that previously had been productive. In fact, it frequently
occurs that a hill which has been producing heavily for several years
suddenly becomes "missing." Of the 110 new missing lulls recorded
in 1912 the average yield for the previous year was 10.2 pounds
green weight, and 56 of these hills had each given a liigh yield in the
years 1909 to 1911, inclusive.

Out of the entire number of hills on this acre only 1 has been
missing for the whole period of four years, 4 have been missing for
three consecutive years, and 45 for two years in succession. Of the
56 lulls missing in 1909 only 6 were still missing in 1912. Altogether,
193 different lulls have been missing on this acre during the past
four years, which would have necessitated the replanting of more
than 20 per cent of the entire number of hills if a perfect stand were
to be main tamed. The fact that new missing Mils occur each year,
many of which have previously been highly productive for several
3-ears, strongly suggests that the average length of life of the culti-
vated hop plant may be less than is popularly supposed. Cases are
known where it is claimed that individual plants have given a fair
yield each year for 30 years, but many growers agree that, with a
newly planted yard, after three or four crops have been harvested
the hills begin to die out to a greater or less extent. Positive con-
clusions on this point, however, can not be drawn from the data in
hand, since the period covered by the observations is entirely too
short to be more than suggestive.

VARIATION IN PRODUCTIVE STAND.

A large part of the variation in productive stand is caused by the
occurrence of lulls having vines producing no hops. Such hills pre-
sent a greater problem than those which are missing, since many of
them if left undisturbed produce a good crop the following year and
digging them out and setting new plants might result in loss rather
than gain. Each year a few of the replants fail to bear hops; others
of the hills are probably unproductive through loss of vigor, since a
number are dead the following year, and some vigorous and normally
productive hills through some accident fail to yield a crop. How
these various classes among the hills having vines but no hops are
distributed from }^ear to year is shown in Table III.

[Cir. 112]

30

CIKCULAK 112, BUREAU OF PLANT INDUSTRY.

Table III.— Record of the hills on 1 acre having vines but jrroducing no hops, for the years

1909 to 1912, inclusive.

Total
hills.

Re-
planted
hills.

Productive hills.

Nonproductive hills.

Hills
dead the

Year.

Previous
year.

Follow-
ing year.

Previous
year.

Follow-
ing year.

following
year.

1909

43
21

= 50

5S

2

19

3

C)
17
25
51

36
18
37

0)

2

6
4

1
1

5

0)

6

1910

2

1911

S

1912

(')

1 No record.

2 The crop on 24 of these hills was lost through defective supports, which allowed the vines to fall to the

ground.

This table shows that the relation existing between the newly
replanted hills and those having vines but no hops is less close than is
generally supposed, since the number of the latter which were replants
is small both in comparison with the total number of hills having vines
producing no hops and with the number of hills successfully replanted,
as shown in Table II. The figures in columns 4 and 5 of Table III
indicate that of the hills having vines but no hops in any given year
the greater number were productive in the previous year as well as
in the one immediately following. Similarly, the figures in columns
6 and 7 show that very few of these hills were nonproductive in either
the previous or the following year. Finally, from the last column
it appears that relatively few of the hills having vines but producing
no hops are numbered with the dead the following year.

In view of the facts here presented there seems no escape from the
conclusion that a large number of the cases of lulls having vines but
no hops arise through neglect or carelessness in cultivating or caring
for the plants up to harvest time.

LOSS IN YIELD DUE TO DEFECTIVE STAND.

Everyone recognizes that, as a rule, a poor stand means a diminished
yield, but it frequently happens that the extent of tins loss is not
fully appreciated. This is particularly true when the number of
missing or nonproductive hills is small, for then the grower often feels
that the saving would not be large enough to warrant his giving the
time and attention necessary to maintain a full productive stand.
This impression is likely to persist unless some relative numerical
expression is found that will approximately represent the extent of
the loss. A fairly satisfactory method of estimating loss is to deter-
mine the percentage of productive stand and the actual yield, say on
1 acre, and from these figures to calculate what the yield would be
on the basis of a productive stand of 100 per cent. The difference
between the estimated yield and the actual yield will then represent

[Cir. 112]

RELATION OF STAND TO YIELD IN HOPS.

31

the logs. Applying this method to the records of the acre discussed
in the previous paragraphs the results set forth in Table IV were
obtained.

Table IV. — Estimated loss and comparison of actual with estimated yield on 1 acre of

hops.

Year.

Pro-
ductive
stand.

Actual

yield, dry

weight.

Estimated
yield with
full pro-
ductive
stand.

Estimated loss due
to lack of stand. 1

Quantity.

Value.

1909

Per cent.
89.1

S9.9
92.2
82.2

Pounds.

1,487
1,443
2,353
1,828

Pounds.
1,668
1,605
2,552
2,223

Pounds.
181

If'..'
199
395

$29. 32

1910

15.87

191 1

i.-i. 27

1912

50. 56

i The estimate of value is on the basis of the farm value of hops in cents per pound, less 6 cents per
pound for harvest costs. These farm values are officially estimated by the Bureau of Statistics, U. S.
Dept. of Agriculture, as follows: 1909, 22.2 cents; 1910, 15.8 cents; 1911, 38.3 cents; 1912, 18.8 cents.

When the effect on yield of missing and unproductive hills is thus
translated into terms of dollars and cents per acre, the results of
inattention to proper cultural methods become very clear. The aver-
age loss on this acre for the four years 1909 to 1912 was $40, a
sum certainly well in excess of that required to pay for the labor and
supervision necessary to maintain a maximum percentage of pro-
ductive stand.

SUGGESTED PROCEDURE FOR MAINTAINING A GOOD STAND.

Although some growers succeed in maintaining a practically per-
fect stand, others may fail to do so owing to causes clearly beyond
their control. However, strict attention to the suggestions which
follow will eliminate nearly all of those cases of missing or nonpro-
ductive hills which are due to carelessness or neglect. Such cases, as
is shown on previous pages, are responsible for the greater part of the
loss due to defective stand.

PRACTICAL SUGGESTIONS.

(1) Just before harvest time mark by means of stakes driven well
into the ground all missing, "bastard," and excess male hills. After
harvest dig out these hills and leave an open excavation at least 3
feet across and 2 feet deep.

(2) At pruning time dig out all hills that have died during the
winter; then, before replanting, fill the site of all excavated hills with
fresh soil mixed with manure.

(3) If possible, replant early while the soil contains an abundance
of moisture to support the growth of the cuttings; cuttings planted
in dry soil or sand should be well watered when they are set out.

[Cir. 112]

32 CIRCULAR 112, BUREAU OF PLANT INDUSTRY.

(4) In replanting use only sound, vigorous cuttings taken from high-
yielding hills and see that the cuttings are not allowed to dry out
before planting.

(5) After the plants are well started inspect the hills carefully
and replace all weak or dead plants with vigorous reserve plants from
the nursery.

(6) Personally supervise the work of replanting, especially when
it is done under contract or when immigrant labor is employed.

(7) In pruning, carefully distinguish (a) normal, well-developed
stocks, which may be cut back either quite close to the crown or so
as to leave only the first set of eyes on the stumps of the vines of the
previous year, and (b) small, weak stocks, which should be so cut
that two or even three sets of eyes will be left on the stumps of the
vines.

(8) See that the vines are properly tied up, so that they will not be
caught and broken or torn down by the implements used in cultivat-
ing or spraying.

(9) Keep a constant oversight of the fields and whenever a vine
is torn down or falls to the ground see that it is immediately replaced
on its proper support.

[Cir. 112]

ADDITIONAL COPIES of this publication
-Ti- may be procured from the Superintend-
ent of Documents, Government Printing
Office, Washington, D. C. , at 5 cents per copy

U. S. DEPARTMENT OF AGRICULTURE.

BUREAU OF PLANT INDUSTRY— Circular No. 113.
B. T. GALLOWAY, Chief of Bureau.

MISCELLANEOUS PAPERS.

Soil Bacteriology as a Factor in Crop Production
A New Ornamental Palmetto in Southern Texas .

Commercial Truck Crops on the Truckee-Carson Project

K. F. KELLERMAN

0. F. COOK

|F. B. HEADLEYand
VINCENT FULKERSON

A Purple-Leaved Mutation in Hemp

The Tuber-Unit Method of Seed-Potato Improvement

. L. H. DEWEY
WILLIAM STUART

Issued February 15, 1913.

WASHINGTON:

GOVERNMENT PRINTING OFFICE.

1913.

[Cir. 113]

2

BUREAU OF PLANT INDUSTRY.

rhief of Bureau, Beverly T. Galloway.
Assistant Chief of Bureau, William A.Taylor.
Editor, J . E. Rockwell.
chief Clerk, James E. Jones.

ADDITIONAL COPIES of this publication
11 may be procured from the Superintend-
ent of Documents, Government Printing
Office, Washington, P. C, at 5 cents per copy

SOIL BACTERIOLOGY AS A FACTOR IN CROP PRODUCTION. 1

By Karl F. Kellerman, Physiologist in Charge of Soil-Bacteriology and /'laid Nutri-
tion Investigations.

IMPORTANCE OF SOIL BACTERIOLOGY.

This day and generation are in the grasp of the invisible, the things
infinitesimal, the things minute. The physician and the engineer

have learned that they can battle with the complex conditions of
to-day only by understanding a wealth of microscopic detail that
formerly would have been considered ridiculous. These two pro-
fessions have now joined with biology in recognizing the microscope
as a necessity. It is not surprising, therefore, that the most impor-
tant profession in the w r orld, that of farming, should awaken to the
desire of comprehending and controlling the tremendous although
imperceptible forces which may enrich or ruin fields and crops.

It no longer thrills an audience to say that soil is a living tiling; it
is either a trite remark, a commonplace relegated to the shelves as
uninteresting, or it is a statement quoted impressively to explain
everything in soil fertility. Both attitudes are unfair to this new
science. Real progress is slow, and there are good reasons for asking
the farmer to maintain a deep interest in the unseen population of
the fields and to beware of dogmatic explanations unsupported by
properly marshaled facts. The average mind is more than willing to
believe a fairy tale, but unquestioning belief in hasty conclusions is
more injurious to future progress than temporary lack of interest or
appreciation.

Most readers of modern agricultural literature are familiar with the
generalization that but for the various agents of decay the world
would in a very short time become uninhabitable. It is through the
constant disintegration and decay of dead plants and animals that
fertile soils are regenerated and that the processes of growth of plants
and animals can continue. Oidy during the last half century has the
science of bacteriology shown how remarkable is this transmutation
of the dead plants and animals back to soil. Bacteria themselves are
plants. They form the simplest group of the fungi, or plants that are
lacking in chlorophyll. They are exceedingly minute; the majority

1 Issued Feb. 15, 1913.

This paper was presented Jan. 7, 1913, as one of a series of lectures given before the scientific staff of the
Bureau of riant Industry.

[Cir.113] 3

4 CIECULAK NO. 113, BUREAU OF PLANT INDUSTRY.

are not more than one one-hundred-thousandth of an inch in diameter,
and it is believed that some bacteria exist which are too small to be
seen even with the aid of the most powerful microscope. In spite of
then small size, however, they are concerned with every phase of our
daily life and by their incredible numbers and ceaseless activity over-
come their apparent insignificance. Bacteria cause diseases, make
milk sour, and in many ways are most troublesome. In spite of the
evil that some species of this group of plants cause, however, other
species, and even some of the troublesome species under different
conditions, are beneficial.

CLASSES OF SOIL BACTERIA.

The bacteria of the soil are chiefly of the beneficial types. They
occur in almost infinite numbers, a fertile soil having from 15, 000, 000
to 300,000,000 to the ounce. Their functions and value are variable,
both because the kinds of bacteria differ in soils and because any
given species may vary physiologically within certain limits according
to environmental conditions. The moisture, the temperature, the
degree of pulverization, the rock formation, or the geological history
of the soil, the aeration, the drainage, etc., are all factors which
partly determine the action of soil bacteria; and perhaps more
important than any of these limiting conditions is the effect of one
kind of organism upon those with which it is closely associated, or,
more broadly speaking, the effect of the associative or competitive
action of the various groups of micro-organisms which act and react
upon each other.

In outlining briefly the relation of soil bacteriology to crop pro-
duction it is simpler to consider the subject from the standpoint of the
bacteriologist and refer to groups of organisms more or less well known
that perform different functions in maintaining the fertility of farm
soils. These groups taken in their entirety include practically all of
the known types of decomposition and synthesis which take place in
the soil. They include organisms which produce hydrogen-sulphid
gas, indol, mercaptan, and other substances of this character whose
importance in agriculture has yet to be demonstrated and need not be
considered at this time. The chief groups and the order in which they
will be discussed are the carbohydrate fermenters, which change
starch, sugar, cellulose, etc.; the ammonifiers, which decompose com-
plex nitrogenous compounds and form ammonia; the nitrifiers, which
oxidize ammonia to nitrite and nitrite to nitrate; the denitrifiers,
which reduce nitrate to nitrite and sometimes to free nitrogen gas;
and the nitrogen fixers, which "fix" or combine the free nitrogen gas
of the air with other substances to form proteids.

(Cir. 113]

SOIL BACTERIOLOGY AS A FACTOR TX CROP PRODUCTION. 5

CARBOHYDRATE-FERMENTING BACTERIA.

The group of organisms fermenting carbohydrates, such as sugar,
starch, and especially cellulose, which is the substance forming the
firm or fibrous portions of plants, has many representatives, but their
functions and their relations to crop plants are in most cases not well
understood. It is known, of course, that many species of bacteria
will ferment the different sugars, forming some organic acid and fre-
quently carbon dioxid and hydrogen or carbon dioxid and methane.
It is highly probable that the constant production of these gases, in
minute quantities it is true, in the chemically powerful condition of
the nascent or freshly liberated gas exerts a much greater influence in
soil weathering and the solution of plant food from small particles of
minerals than any of the much-discussed agencies of the plant roots
themselves. Both the carbon dioxid and the organic acids produced
by these organisms are presumed to be of importance in controlling
the availability of phosphates. No satisfactory methods of con-
trolling this action have been developed, however, and no accurate
estimate of this function has been possible.

The most resistant carbohydrate found in the soil and the type of
carbohydrate which is most important because of the enormous
quantities added to soils annually is plant fiber or cellulose. This is
a substance so resistant to even moderately strong acids and alkalis
that it appears at first glance almost indestructible. Obviously,
however, active agents for cellulose destruction must be constantly
operating; the decay of wheat stubble, of cornstalks, and of green
manure turned under, as well as the rapid rotting of dead trees, are
perhaps the best illustrations of this. It should be noted that humus
has long been recognized as an essential constituent of a fertile soil,
and more recently it has become apparent that many of the processes
constantly going on in the soil which result in the formation of avail-
able plant food are controlled in a large measure by the quantity and
quality of the humus compounds. In the formation of these com-
pounds the cellulose, which is decomposed in the soil in such large
quantities, must play a prominent role.

Extensive studies in our laboratory during the past two years
have shown that there are numerous organisms endowed with the
power of actively fermenting cellulose. The cellulose is made soluble
and becomes available as carbohydrate food for other organisms,
especially the nitrogen fixers, which will be referred to later. There
is another phase of cellulose destruction which appears to be of
great economic importance from the soil-fertility standpoint, and
that is the action of filamentous fungi. Several investigators from
time to time have found one or more species of molds to which they

[Cir. 113]

6 CIRCULAR NO. L13, BUREAU OF PLANT INDUSTRY.

have attributed the power to dissolve cellulose, but no systematic
attempt has been made to determine the extent of this phase of
cellulose destruction. It appears, however, that cellulose destruc-
tion by molds is as important, if not more important, than its destruc-
tion by bacteria. There are at least 75 species of cellulose-destroy-
ing molds, representing a large number of genera, including such
common forms as Aspergillus, Fusarium, and Penicillium, which is
the green mold so often seen on fruit and moist hay. The enor-
mous quantity of cellulose available in the soil and the tremendous
energy that can be developed by bacteria in breaking up the cellulose
into simpler ferments make it seem at least possible that the mainte-
nance of a high degree of fertility depends more upon the presence in
the soil of the proper flora for decomposing the cellulose than upon
the existence of any other group. It should be remembered that this
accumulation of carbohydrate food, the source of energy for so many
soil processes, is due to the fact that chlorophyll, the green coloring
matter of the leaves of plants, in some manner as yet unexplained,
enables the plants to absorb the carbon dioxid of the atmosphere
and to utilize the energy of the sunlight to transform this carbon
dioxid and the water of the plant sap into carbohydrates. The
synthesis of carbohydrates is logically the most fundamental ques-
tion of plant physiology, for this great quantity of potential energy
is chiefly responsible for the continuance of plant and animal life.

AMMONIA-FORMING BACTERIA.

There are many species of bacteria very widely distributed in
nature which break down proteid compounds, forming ammonia
together with other decomposition products. These ammonifieis,
indeed, comprise what is numerically perhaps the largest group of
soil organisms. It was formerly supposed that crop plants were 1
unable to assimilate nitrogen in the form of ammonia. The pro-
duction of ammonia by bacteria was therefore regarded as merely
a preliminary step in the preparation of suitable plant food. More
recent investigations show clearly that many crop plants, especially
cereals, are able to assimilate ammonia nitrogen. The ammonifieis
therefore do prepare a suitable food for crop plants, and because
of the rapidity of their action and because of the enormous number
of them that are found in all soils we shall probably have to con-
sider this the most important group of bacteria. The fact that
they are widely distributed, however, makes them somewhat less
interesting from the standpoint of controlling fertility, since it is
doubtful whether conditions obtain in any agricultural region where
these bacteria are not present in sufficient numbers to transform
rapidly into ammonia whatever proteid nitrogen may be turned

[Cir. 113]

soil, BACTERIOLOGY AS A FACTOR IN CROP PRODUCTION. 7

into the soil, provided conditions for bacterial growth are favorable.
In this connection it is interesting to note that the Subtilis group, a
subdivision of the group of ammonifiers comprising four or five
distinct species, occurs probably in every cultivated country. It
has been found in every country where any attempt has been made
to isolate these species. The products formed by the various species
of bacteria in the ammonifying group, in addition to the ammonia
formed from the proteid nitrogen, are not well known. It is there-
fore a possibility that some of these bacteria may produce substances
injurious to the growth of crop plants. The so-called soil toxins,
which may or may not be important factors in limiting the develop-
ment or continued cultivation of certain crops, may be produced
by these bacteria; there is as yet, however, no satisfactory evidence
that such is the case.

NITRIFYING BACTERIA.

The nitrifying bacteria are in reality two groups of organisms.
The first group when supplied with mineral plant food and with
ammonia nitrogen transforms the ammonia into nitrite. This group,
which formerly was considered to consist of a few very highly special-
ized organisms, is now known to number many species, most of which,
however, have the power of making the transformation of ammonia
to nitrite only to a moderate degree. The large number of organisms
having this power makes it seem probable that usually they may be
more important than the few species that are more active and also are
more seldom found. The second group of nitrifying bacteria is able
to further oxidize the nitrogen in the form of nitrites to nitrate nitro-
gen. Nitrate nitrogen is the form of nitrogen usually considered to
be of the highest value as plant food, while nitrite nitrogen, except
in very minute quantities, is an actual poison to crops. This special-
ized group of bacteria, therefore, which is able to feed on nitrate
nitrogen and as fast as it is produced transform it into desirable food,
is of extreme importance in all cultivated soils. Like the ammoni-
fying bacteria, however, the bacteria of this group also are widely
distributed and usually are present in sufficient numbers to perform
their proper functions. They develop best in soils that are kept in
good tilth, and usually grow very sparingly, if at all, in water-logged
acid soils, and but sparingly in the fine-grained clays and gumbos that
are almost impervious to the air.

DENITRIFYING BACTERIA.

The denitrifying bacteria, many of which are also ammonifying bac-
teria, are those which have the power almost diametrically opposed
to that of the two groups just described. These organisms trans-

[Cir. 113]

8 CIRCULAR NO. 113, BUREAU OF PLANT INDUSTRY.

form nitrate into nitrite, perhaps also into ammonia, and certainly
many of them are capable of breaking up the molecule entirely and
giving off free nitrogen gas. This group of organisms if functioning
in this manner constantly would soon destroy the most fertile land,
but although these organisms are found generally, they are not usu-
ally very plentiful and they do not become active denitrifiers unless
afforded certain definite conditions for growth. The requirements of
different species of the denitrifying group vary considerably; most of
them will transform nitrate nitrogen into nitrite if organic material,
such as beef broth or plant extracts, or even manure, be supplied them
together with nitrate nitrogen. This fact illustrates the danger of
applying nitrate fertilizers and manure simultaneously to a crop.
Especially if the soil is moist or water-logged at the time there ensues
a rapid development of nitrites with consequent injury to the crop.
If nitrates and manure are to be applied to a given field, the manure
should be applied weeks or, preferably, months in advance, in order
to allow the preliminary decomposition of the proteid compounds in
the manure to proceed to a point where they will no longer be food
for denitrifying bacteria.

NITROGEN-FIXING BACTERIA.

The nitrogen-fixing bacteria are those which have the power of
combining the atmospheric nitrogen with their other food materials
and forming proteid nitrogen. Their activity can be enhanced by the
improvement of the soil tilth, and sometimes also by the addition of
carbohydrates, such as sugar and molasses. It is probable, however,
that cellulose which has been decomposed to some soluble form is as
good a food as any of the sugars produced commercially. Since the
development of these free-living, nitrogen-fixing bacteria is presumed
to be largely responsible for the continued fertility of agricultural land
and since it is probable that these bacteria are the chief cause of the
undiminished fertility of the lands of the intermountain district in the
Northwest, where continuous grain cropping has produced no apparent
diminution hi soil nitrogen, both the importance of this group of
bacteria and the probable correlation of the development of nitrogen-
fixing bacteria with cellulose-dissolving bacteria are obvious.

NODULE-FORMING BACTERIA.

An important subdivision of nitrogen-fixing bacteria and one
which is perhaps better known than any other group of soil bacteria
is the nodule-forming bacteria of the legumes. These bacteria when
associated with the roots of the leguminous plants and a few non-
leguminous plants form nodules upon the roots which are capable of
using as food large quantities of atmospheric nitrogen. The recipro-

[Cir. 113]

SOIL BACTERIOLOGY AS A FACTOR IN CROP PRODUCTION.

9

fating or complementary action of the different groups of soil bacteria
which transform nitrogen compounds is more clearly expressed by a
diagram. Figure 1, which is self-explanatory, shows the relation of
bacteria to some of the most essential changes. Since the study of
the nitrogen-accumulating power of legumes bearing root nodules

BAC TERIA

CHANGE

NITROGEN GAS INTO

PFtOTED NITROGEN

BACTERIA OF LEGUME

AND NON-LEGUME

ROOT NODULES

CHANGE

NITROGEN GAS INTO

PROTEID NITROGEN

DENITRIFYING

BACTERIA

<-««<C/-/A/VGC-<— m

AMMONIA TO
NITROGEN GAS

//I

<&*>

o

o

y

#

Fig. 1.— Diagram showing the nitrogen changes produced in the soil by the action Of bacteria. The arrows
indicate the course of the changes which various groups of bacteria may produce in the n it r< men compounds
of the soil.

deals with a definite and comparatively simple relation of a single
type of organism to a single type of plant, it has been possible to
develop these investigations upon a pure-culture basis; in other
words, it has been possible to isolate the nitrogen-fixing organism
from nodules of leguminous plants and with these cultures produce
77106°— Cir. 113—13 2

10 CIRCULAK NO. 113, BUREAU OF PLANT INDUSTRY.

nodules upon other plants of the same species in different localities.
Owing to the fact that different legumes are constantly being intro-
duced into agricultural regions, the importance of being able to dis-
seminate the nodule-forming bacteria is obvious. For many types of
legumes this is desirable, not only from the standpoint of making the
crop a better nitrogen fixer and better soil renovator, but because for
most legumes the crop is actually larger when properly inoculated.
Pure-culture inoculation is less certain than inoculation by means of
soil from old, well-inoculated fields, though of course it is free from
the danger of introducing troublesome weed seeds or plant diseases.

WORK OF THE BUREAU OF PLANT INDUSTRY.

The Bureau of Plant Industry is carrying on field experiments to

determine, if possible, what soil conditions are most favorable for the

successful inoculation of leguminous crops by the use of pure cultures

and also to determine under what conditions it is useless to attempt

to inoculate certain of the legumes without some radical change in

the method of fertilizing or cultivating these fields. To extend this

experimental work as far as possible, the Department of Agriculture

is willing to supply cultures in any reasonable quantity, requiring

only the filling in of blank reports which are occasionally forwarded

for this purpose.

CONCLUSION.

A casual review of the present status of soil bacteriology shows that
it is a subject of almost bewildering complexity, but very intimately
associated with the normal physiology of all crop plants. By learning
what functions must be performed in the soil and by studying each
group of organisms that has a measurable function, we believe that
we can learn how to enhance the desirable activities of thaliving soil
and to check the undesirable ones. We may never find another rela-
tionship so simple as that existing between the legume crop and the
nodule bacillus or one which we can control so simply from a central
laboratory; but surely a thorough comprehension of what happens
in the soil will teach us how to maintain the fertility of the soil more
surely than blind and rather purposeless experimentation. The great
majority of plat tests of fertilizers illustrate the impossibility of satis-
factory deductions from experiments entirely upon an empirical
basis. We must deal in irons and not in generations if agricultural
science is to be advanced by the cut-and-try method.

[Cir. 113]

A NEW ORNAMENTAL PALMETTO IN SOUTHERN TEXAS. 1

By O. K. i'ook, Bionomist in Charge of Crop Acclimatization and Adaptation

Investigations.

Apart from the date palms, which are of Old World origin, only two
kinds of palms have qualified for general planting in southern Texas.
These are the Washingtonia palm, a native of the desert region of south-
ern California, and the palmetto. Several times in recent years the
Washingtonia palms in the city parks of San Antonio have had tneir
leaves partially killed by cold weather, while adjacent palmettos re-
mained entirely uninjured. Thus the palmettos may be reckoned as
the hardiest of all of the native North American palms.

The minimum temperature recorded at San Antonio in the winter of
1910-11 was 16° F. and in the following year 18° F. At Victoria,
Tex., the cultivated palmettos have passed without any damage to
the leaves through freezes that killed many of the wild huisaches
(Acacia farnesiana) . Though certain other palms are able to survive
such temperatures and are worthy of being planted for special pur-
poses, the mutilation of the leaves means a loss of decorative value
for several months. Frost-proof foliage is especially desirable in an
ornamental species.

There is a native Texas palmetto (Inodes texana), which is not
known to exist in a wild state except in the vicinity of Brownsville;-
but it seems to have extended much farther northward only a few
decades ago, and specimens may still be found about Indianola or at
other points along the Gulf coast. No doubt the same species extends
into adjacent regions of Mexico, but it differs from the palmettos of
other parts of that country. Sargent and other writers have referred
the Texan palmetto to Sabal mexicana Martius, but this species was
based on a small t runkless palm from a remote locality on the southern
coast of the State of Oaxaca, not far from the Isthmus of Tehuantepec

1 Issued Feb. 1".. L913.

2 This is in addition to the low, creeping, scrub palmettos which are widely distributed in swamps and
wet river bottoms of the Coastal Plain. The scrub palmettos belong to the genus Sabal, which was formerly
supposed to include both groups of palmettos. The genus Inodes differs from Sabal in the formation of a
thick, upright trunk by secondary thickening of the fibrovascular system below the terminal bud. As a
consequence of this habit of growth the leaf bases are split down the middle, as in i he Washingtonia palms,
but not in Sabal, Erythea, or Brahea. Another difference is that leaves of Sabal are few in number and
deeply divided in the middle, because of the slight development of the midrib. The strongly specialized
recurved midrib is a peculiarity of the genus Inodes.

[Cir. 113] 11

12 CIECULAE NO. 113, BUREAU OF PLANT INDUSTRY.

The palmettos planted for ornament in San Antonio and other
plaees in southern Texas were at first supposed to represent the
native Texan palmetto. They have broad leaf segments like Inodes
texana, instead of narrow segments like the well-known cabbage pal-
metto of Florida {Inodes palmetto), of which only a few specimens
have been seen in Texas. The Texan palmetto is a handsome palm
and is used for ornamental purposes in a few places in the region of
Brownsville, but it now appears that other species are represented
among the cultivated palmettos. This was learned recently at
Victoria, Tex., where one of the cultivated palms was found with an
abundant crop of fruit. The thickened terminal branchlets of the
inflorescence showed at once that it could not be the Texan palmetto,
which has slender branchlets.

Some of the Victoria palmettos are really magnificent, with their
stately crowns of large vivid-green leaves firmly supported on massive
petioles, also of living green. Even the trunk appears green, for the
sheathing leaf bases retain their color. The crown is more ample
than in most palms because of the firm texture and persistent vitality
of the leaves. This lends an impression of extreme vigor and luxu-
riance and adds greatly to the decorative effect. In short, it seems
not unlikely that the Victoria palmetto may find a place in the front
rank of ornamental species.

Inquiries regarding the origin of the palmettos at Victoria revealed
a local opinion that they had been brought from California, and this
may indicate that the stock came originally from Mexico. There are
no native palmettos in California. Many introduced species are
grown in collections, though they do not seem to thrive very well in
places near the coast, probably because of insufficient heat.

At first it seemed likely that the Victoria palmetto might be
referred to a species with thickened branchlets, Inodes uresana
Trelease, from the vicinity of the town of Ures in Sonora. 1 But
there are several features that seem to forbid such an identification.
The leaves of Inodes uresana are described as extremely glaucous, to
such an extent that Prof. Trelease compared it with the so-called
"blue palm" (Erythea armata), noted for its very light grayish or
bluish color. This characteristic does not appear in the Victoria
palm, the foliage of which is of a deep, vivid green, as far from
glaucous as could well be imagined.

To judge from the photographic illustration of Inodes uresana, the
leaves have a more strongly decurved rachis and more deeply sepa-
rated, narrower segments. The measurements given for the Sonora n
species likewise indicate a smaller palm. The trunk is said to be 30
centimeters thick, the petiole 2 centimeters wide, and the blade 1

i Trelease, W. A Pacific-slope palmetto. Missouri Botanical Garden Keport, v. 2, p. 79, pi. 35-37, 1901.
[Cfr. 113]

A NEW OENAMENTAL PALMETTO IN SOUTHEKN TEXAS. L3

meter wide, or about only half the size of the corresponding parts in
the palms at Victoria, where the individual leaf segments arc nearly
a meter long and 5 centimeters wide.

The Victoria palmetto agrees with Inodes uresana in the size of the
fruit and seed, but the seed is not coarsely wrinkled above as in
Inodes uresana nor hollowed below. The absence of any distinct exca-
vation of the ventral surface renders the seeds of the Victoria pal-
metto quite distinct from those of any of the large-seeded palmettos in
the National Herbarium or in the seed collection of the United States
Department of Agriculture. Instead of being hollowed out, the
lower side of the seed is flat or slightly convex and usually shows a
small, smooth prominence in the middle, surrounded by coarse, irreg-
ular wrinkles. The hilum is eccentric, but close to the central promi-
nence and surrounded by stronger radiating wrinkles. The remainder
of the surface, outside the flattened areas of the lower side, is not
wrinkled but only finely coriaceous or leathery under a lens. The
micropyle, marking the position of the embryo, is represented by a
very distinct conical elevation surrounded by a flat rim or a slight
circular depression. The embryo is lateral, just above the transverse
middle of the seed. The seeds are about 7 millimeters thick, with a
diameter of from 10 to 12 millimeters, about two-thirds of them
being 11 millimeters and less than 10 per cent attaining 12 millimeters.
One of the seeds showed no micropyle, embryo, or cavity in the
endosperm.

The seeds are surrounded in the mature fruit by a thin membrane of
rather loose corky texture, with the outer surface torn into numerous
spinelike points by the shrinkage of the pulp which dries down
against the outer wall into a brownish layer, usually less than 1
millimeter thick. This leaves an empty space, often 2 millimeters
across, between the membranous coating of the seed and the similar
coating of the inner surface of the pulp. The total diameter of well-
developed fruits is about 18 millimeters and the length nearly the
same, owing to the presence of a somewhat oblique narrowed base
bearing the conical persistent style as a spinelike basal projection
nearly 2 millimeters hi length. Only one fruit is developed from a
flower. There are none of the twin and triplet fruits that are of
frequent occurrence in Inodes texana.

Another related species from the Pacific slope of Mexico is Inodes
rosei Cook, described from specimens collected by Dr. J. N. Rose,
at Acaponeta, Tepic, but this palm has the ultimate branchlets veiy
slender and the seeds like those of Inodes uresana, strongly rugose above,
and deeply umbilicate below. The fruits of Inodes rosei are usually
smaller than those of the Victoria palmetto, with thinner membranes
and less pulp. The lobes of the calyx are shorter and the persistent

[Cir. 113]

14 CIRCULAR NO. 113, BUREAU OF PLANT INDUSTRY.

style small, shriveled, and appressed instead of forming a prominent
spur.

On account of the divergencies from the related forms it seems
necessary to recognize the Victoria palmetto as a new species, for
which the name Inodes exul is proposed, in allusion to the fact that
the original habitat is unknown, though there is every probability
that it will be found in some part of northern Mexico. The diagnostic
characters at present available are the large size, the deep-green foli-
age, the thickened branchlets of the inflorescence, the solitary fruits,
and the large seed not wrinkled above nor hollowed out below.
Type specimens and photographs have been deposited in the United
States National Herbarium (No. 691415) and a sample of the seeds
in the Economic Herbarium of the United States Department of
Agriculture. Other characters of the species remain to be learned
by further study, but it seems desirable to publish the present state-
ment as a means of bringing so promising an ornamental to public
attention.

The type individual of the new species graces the lawn of Mrs.
Martin O'Connor, of Victoria, Tex., who has kindly presented the
seed to the United States Department of Agriculture through Mr.
John H. Kinsler, of this Department. The seed lias been planted at
the United States Experiment Farm at San Antonio with a view to
wider distribution in southern Texas, where congenial conditions
seem to be assured; and in view of the demonstrated hardiness of
the species it may be expected to thrive in other parts of the coun-
try, probably throughout the Gulf coast "and South Atlantic regions,
as well as in the warmer parts of the Southwest.

[Cir. 113]

COMMERCIAL TRUCK CROPS ON THE TRUCKEE-CARSON

PROJECT. 1

By F. F>. Headley, Superintendent, and Vincent Fulkerson, Scientific Assistant,

Office of Western Irrigation Agriculture.

INTRODUCTION.

The growing of truck crops on the Truckee-Carson project is assum-
ing increasing importance from year to year. This class of farming
is largely followed by men who rent land on the older alfalfa ranches.
Suitable land for this purpose has been rented for $12 to $25 per
acre and generally for a term of years, since subduing and leveling
old alfalfa land is often expensive.

Potatoes, onions, celery, and cantaloupes are grown extensively for
marketing outside the State, but the growing of other truck crops is
limited to the demands of the local non-agricultural population resid-
ing on the project and in adjacent mining camps. The future devel-
opment of the truck-crop industry is therefore limited, except for
those crops that will bear shipment and can compete with the outside

market.

POTATOES.

About 480 acres of potatoes were grown on the Truckee-Carson
Project in 1912. The Burbank is about the only variety grown for
shipment. The yield per acre is exceedingly variable, owing to a
number of causes, such as variation in soil, irrigation, and seed. The
average yield on good alfalfa land is probably about 10 tons of market-
able potatoes per acre, while the average yield for the entire project
in 1912 was 4.1 tons per acre. Market prices in the last few years
have ranged between $12 and $25 per ton.

The future of the potato industry is promising. The local demand
is large 1 , and the completion of the Panama Canal may so reduce the
transportation rates that Nevada-grow T n potatoes can be placed in
eastern markets. The quality of Nevada potatoes is excellent and
by careful selection of seed and type, with due care in growing, sort-
ing, and marketing, there should be a large development in this
industry.

i Issuerl Feb. 15, 1913.
[Cir. 113] 15

16 CIRCULAR NO. 113, BUREAU OP PLANT INDUSTRY.

Cost of 'production . — The following statement gives an estimate of
the cost per acre of potato production in the Truckee-Carson Project:

Rental cost of land $15

Plowing land 3

Leveling and harrowing 1

Seed 15

Planting 3

Cultivating 3

Irrigating, nine times 3

Digging 20

Sacks 15

Marketing 6

Total cost per acre 84

Numerous publications have treated of the growing of potatoes
under irrigation, and the general principles of successful practice are
as applicable to this region as to other Western States, yet ignorance
of local conditions invites failure. Some of the special precautions
necessary to observe are given below.

Selection of seed. — No seed containing eelworms, produced either
locally or imported, should be planted. Misshapen tubers and even
good specimens from poor hills should not be used for seed. Selec-
tion of good hills should be made in the field. A man can usually
follow a digger and select such hills as yield abundantly and produce
fair-sized, well-shaped tubers. Imported Oregon seed stock has been
largely used and is considered most desirable by the more successful
growers, but careful hill selection of suitable varieties promises better
results in increasing the yield, as well as the percentage of smooth,
marketable potatoes.

Selection of land. — New land seldom gives good crops of potatoes,
although some of the black tule lands are fairly productive. Gener-
ally, only alfalfa land should be planted to potatoes, and the growers
should not be satisfied with less than 9 tons (300 bushels) per acre.
Either sandy or heavy loams are suitable if properly handled. A
little white alkali (sodium sulphate) when present is not noticeably
injurious, but the slightest traces of black alkali will injure the growth
of the crop.

Preparing land. — When alfalfa land is to be used for potatoes, the
alfalfa should be ''crowned" in late summer or early autumn pre-
vious to the year that potatoes are to be grown. By "crowning" is
meant cutting off the alfalfa plants about 2 to 3 inches below the
surface of the ground. To do this well a sharp plowshare is essential,
and the plow must be strong, so that it can not be twisted or broken.

After plowing, the field should be harrowed, so as to bring the
crowns to the surface. If they are buried deeply in the soil they will

[Cir. 113]

{

COMMEBCIAL TEUCK CROPS ON TRICK KIX 'ARSON PROJECT. 17

take now root and give trouble in cultivating the potatoes. In the
spring the land should be again plowed, (> to 8 inches deep, and
smoothed and graded to a uniform slope. A uniform grade is of
great importance, as the water should at no time rise high enough to
cover the tubers. A grade of 0.1 foot per 100 feet will irrigate, but
a little more slope is rather better. It is usually advantageous to
irrigate before the potatoes are planted, particularly if cut seed is
used, as the dry soil absorbs the moisture from the cut potato.

Varieties. — The early varieties are not recommended for shipping.
Long-keeping and high-yielding sorts are better if the erop is to be
shipped.

The Early Rose is grown to some extent, but is generally not satis-
factory, having too much waste from deep eyes.

The Burbank, or some similar white potato, is the favorite, being
more sought after in the California markets than other sorts. It is
by no means certain that the Burbank is the most desirable variety
from the growers' standpoint, as the tubers too frequently become
knotty, rough, and pointed.

A variety known variously as Netted Burbank, Netted Gem, and
White Beauty has been tried in a small way. In shape it is better
than the Burbank and it is productive.

The Rural New Yorker and Colorado Pearl have done well where
tried and seem to have less waste and fewer unmarketable tubers
than the Burbank variety. It is probable that some such varieties
will be found more profitable than the Burbank if no marketing
difficulties are encountered.

Planting. — The rows should not be more than 3 feet apart, for
when farther than this there is the disadvantage of the waste of the
land and the greater length of time required for the water to seep
through under the hills. Most potato growers drop the seed 12 to
15 inches apart in the rows and cover to a depth of about 4 inches.

Planting with a potato planter is more economical and satisfactory
than furrowing with a plow and dropping by hand. A, man and
boy with a team and planter will plant an acre quicker and easier
than three men with a team and plow.

Irrigating. — Potato soil should be kept as uniformly moist as
possible. If the field is well irrigated before planting and thoroughly
harrowed after planting it will not usually be necessary to water
again until the sprouts are out of the ground. After the potatoes
are up they will need irrigating in most soils about every 10 days
until the crop begins to mature, when water should be held off
entirely. There can be no set rules as to the frequency of irrigation,
since some lands of heavy loam or with a high water table will need
less water than others. The fact that the evaporation from an open

[Cir. 113]

18 CIRCULAR NO. 113, BUREAU OF PLANT INDUSTRY.

water surface during the summer months is about 10 inches per
month explains the necessity for irrigating oftener than in the less
arid regions.

Provision should be made to drain away the water that runs out
at the ends of the furrows, so that it can not back up over the hills.
The water must be kept in the furrows below the tubers.

Irrigating in alternate furrows is not practiced here and probably
is not advisable. Great care must be exercised in keeping the growth
of the potatoes continuous and unchecked or the result will be knotty,
badly shaped tubers.

Cultivating. — It is a distinct advantage to cultivate after each
irrigation until the vines interfere. The tool most commonly used
is the winged-shovel plow. This keeps the furrow open and gives
the land only partial cultivation. A spike-tooth cultivator with a
large rear shovel to open the furrows gives better results.

Diseases. — The most serious disease affecting the potato is the
potato scab (Oospora scabies). No seed affected with this disease
should be planted without thorough disinfection,' but it is preferable
to use seed that has never been infected. Potatoes may be disin-
fected by immersing the sacks containing them in a formalin solu-
t ion for two hours. The solution is made by mixing 1 pint of formalin
with 30 gallons of water.

Another disease that results in considerable damage is that caused
by the potato eelworm. Seed potatoes containing the eelworm
should never be planted, as they not only injure the crop but per-
manently infest the soil. 1

ONIONS.

A large acreage of onions was planted on the Truckee-Carson
Project in 1912, the plantings varying in size from 1 to 7 acres.
One of the best fields was surveyed and found to have yielded slightly
over 20 tons per acre.

Cost of production. — An estimate of the cost of growing onions on

this same field was also made and is given herewith, figured on a

1-acre basis. -

( lost of seed $8. 00

3£ days, hauling manure 7. 90

:U days, planting and irrigating 7. 90

J day, repairing ditches 1. 70

7 days, first cultivating and weeding 15. 75

<; days, second cultivating and weeding 13. 50

30 days, pulling, topping, and hauling 67. 50

Total cost 122. 25

i For more complete information in regard to potato culture, write for the following bulletins and circulars,
which may be secured by sending requests to the Secretary of Agriculture, Washington, D. C: Farmers'
Bulletins 407, The potato as a truck crop, and 3SG, Potato culture on irrigated farms of the West; Bureau
of Plant Industry Circulars 90, Suggestions to potato growers on irrigated lands, and 91, The nematode
gallworm on potatoes and other crop plants in Nevada.

- All man labor is charged at the rate of .$2.'2. r > per day.

[Cir. 113]

COMMERCIAL TRUCK CROPS ON TRUCKEE-CARSON PROJECT. 19

If this farmer secures the normal market price of $20 per ton,
making a total of $-100 per acre, he should net a handsome profit on
this crop.

Another grower having 7 acres of onions kept a careful account of
expenses, which is given herewith, figured on a 1-acre basis: 1

Grading and leveling land $24. 70

Hauling manure 10. 30

5} pounds of seed 7. 20

225 pounds of commercial fertilizer 5. 30

Planting, tending, irrigating, and harvesting 104. 75

Total cost 152. 25

It is probable that with increased experience the total cost of pro-
duction can be reduced below the sums mentioned. The cost of
$24.70 per acre for preparing the land is more than will usually be
found necessary. The yield of this field was not ascertained, so the
profits per acre can not be calculated.

The future of onion growing.— The price received during the last
few years has ranged between $20 and $30 per ton. The character
of the Nevada climate makes it comparatively easy to keep well-
matured bulbs in inexpensive storage houses until the market
becomes favorable, which usually occurs after January 1, when the
bulk of the California crop has been marketed.

Selecting and preparing hind. — None but the best soil should be
planted to onions, as much land which produces ordinary farm crops
at a profit fails utterly to grow marketable onions. Newly plowed
alfalfa land may do well without manure, but generally a heavy
application of manure should be plowed under in the fall or winter
and a top dressing of fine, rotted manure applied in the spring before
planting. After plowing, the soil must be thoroughly pulverized,
smoothed, and graded, so that irrigation will be uniform.

Varieties. — Only early-maturing sorts should be planted. As the
Yellow Globe Danvers is productive, sells well, and is a good keeper,
it is' to be recommended. If a white onion of reliable keeping
quality can be found or developed, it should command a good market.
The globe type is to be preferred, as the yield will be larger where all
conditions are favorable.

Planting and irrigating. — Planting should be done before April 1.
With rows 12 to 1 .") inches apart, from 4 to 5 pounds of good seed will
be required per acre. Onion seedlings are delicate and can not push
up through a great depth of soil. The seed is usually planted at an
average depth of three-fourths of an inch. The soil will require
irrigation before planting, if it is not already moist enough to bring
up the seed. The furrows should be made before irrigating or plant-
ing. A suitable furrower can be made by using (> by fi inch timbers

1 All man labor is charged at the rate of $3 per day.
[Cir. 113]

20 CIRCULAE NO. 113, BUREAU OF PLANT INDUSTRY.

as runners. These should be set about 27 inches apart. Two rows
can then be planted with a garden drill between the furrows. Irri-
gation by flooding is also a common practice on this project, as well
as in other onion-growing sections. If this method of irrigation is
decided upon, the rows may all be the same distance apart.

If the weather remains dry after planting, irrigation may be nec-
essary to germinate the seed. A tendency of the soil to bake over
the seeds indicates that irrigation is needed to soften the crust

When furrow irrigation is used on soils that do not subirrigate well,
much slope to the land is to be avoided; otherwise the water may run
past without thoroughly wetting the soil about the plants. A slope
of 0.2 foot per 100 feet should give satisfactory results, and if the soil
is well supplied with humus more slope may be allowed, as such soil
takes water easily and washes less. The irrigation furrows shoidd
not exceed 300 feet in length. Often a field can be so arranged that
the water which runs from the ends of the furrows may be utilized on
a lower series. It is well to follow each irrigation as soon as practi-
cable with a wheel hoe, so that the growth of the onions will not be
checked by loss of soil moisture. Usually it is advisable to irrigate
every week or 10 days until August.

The market requires thoroughly matured onions which will keep
well. To insure such, water should be withheld after the early part
of August and the soil allowed to dry out. The bulbs should mature
by September 15. A field full of scallions and immature bulbs
indicates either poor soil, poor seed, late planting, or too much
watering.

Harvesting. — If the onions are properly matured, pulling is easy,
and two or three days' curing in the field will put them in good condi-
tion for the storehouse. It is entirely possible that it will be found
the best practice where onions are properly matured to top them and
put them at once into crates and store them in open sheds, where the
curing will take place without injury from night frosts or the hot
midday sun.

Storing. — The important points in storage are dryness and low

temperature. While a thin-board building, not frost proof, is better

than a place too warm, a cheap adobe structure is especially adapted

to this purpose, as any temperature in the fall or winter may be

secured by regulation of the night ventilation, and the nonconductive

adobe walls will prevent freezing. The bulbs can then be marketed

at anv time. 1

CELERY.

Celery growing is not yet an important industry. This crop has
been very successful, and the produce is of excellent texture and

! For further information in regard to onion culture, see Farmers' Bulletins 354, Onion culture, and 434,
The home production of onion seed and sets.
|( ir. 113]

COMMERCIAL TRUCK CROPS ON TRUCKEE-CARSON PROJECT. 21

flavor. There scons little prospect of overproduction, as much labor
and care arc required to secure a good product. The crop is so valu-
able that it has been shipped to distant towns in Nevada, as well as
into California, where it sells at a higher price than the local product.
Giant Pascal and Golden Self-Blanching arc good varieties.

Only very rich soil should be used for this crop. The seed should
be sown in April. A light mulch of old hay will prevent the baking of
the soil and protect the small plants from the hot sun when they first
come through the ground. The plants should be strong and stocky
and about b inches high by the first week in July, when they are set
in the field. The growth must be steady and rapid to secure good
quality, upon which any extension of the market depends. Blanch-
ing will require from 2 to 4 weeks. That part of the crop to be
marketed in winter must be placed in pits to prevent freezing. 1

MELONS.

Both watermelons and cantaloupes succeed well on the Truckee-
Carson Project with comparatively little care. For home use any
type of soil suited to ordinary crops will produce melons if well
manured. For best results a retentive soil is required, one that is
well subdued and well supplied with plant food. Much fresh manure
and too much water tend to produce an inferior quality, though water
needs to be applied regularly to keep the plants in thrifty condition
and to insure steady growth.

One gardener is growing both cantaloupes and watermelons for
shipment, supplying the Southern Pacific dining cars and the whole-
sale market in Reno. The varieties of watermelons grown are
Kleekley Sweets, Halbert Honey, and Florida Favorite, and of can-
taloupes, Emerald Gem, New Fordhook, and Rocky Ford. The
Emerald Gem is the best quality, but skill is required in selecting
properly ripened fruit and they do not stand shipment very well.
The New Fordhook is very early and a much better shipper. The
Rocky Ford is good, but later than the New Fordhook.

Planting. — On rich, old alfalfa land rows 8 feet apart are none
too far, as the vines will cover the entire surface. The hills can be 4
feet apart in the rows for cantaloupes and 6 for watermelons. Plant-
ing can best be done before May 1, using old hay or straw to cover
the young plants if severe frost threatens. After the litter has been
distributed in a convenient place a man with a fork can cover a large
number of plants in a short time on cold nights, and the advantage
of an early start for the plants will well repay the trouble.

Cultivating and irrigating. — There are no special requirements as to
cultivation and irrigation, and the cost of bringing the crop to matur-

1 For further information in regard to celery culture, sec Farmers' Bulletin 2N2, Celery.
[Cir. 113]

22 CIRCULAR NO. 113, BUREAU OF PLANT INDUSTRY.

ity is not large. Water should be applied often enough to keep a
steady growth, which can be hastened by frequent cultivation.

Too much water in the ripening period tends to retard ripening
and reduces the sweetness of the melons.

Harvesting and marketing. — Development of melon growing depends
as much on proper care in picking and packing as upon growing.
Only the best product carefully selected and neatly packed will justify
shipment. The California product conies on much earlier and only
an extra quality will insure sale late in the season when the Nevada
crop matures. Watermelons should be ripening freely by August 15,
for the late crop sells at low prices or can not be sold at all. Musk-
melons can be sold successfully until October if the weather is such
that the crop holds up in quality. They should be handled carefully
and wrapped in paper in the packing shed, being then packed in
crates 1 foot by 1 foot by 2 feet, outside measure. A crate of this
size holds about 60 pounds. Watermelons for short shipment are
packed in larger crates, but if the market will justify they are shipped
loose in car lots.

[Cir. 113]

A PURPLE-LEAVED MUTATION IN HEMP. 1

By Lyster H. Dewey, Botanist In ( 'ho rye of Fiber-Plant Investigations.

INTRODUCTION.

Practically all of the hemp {Cannabis sativa) cultivated for fiber
production in Kentucky and elsewhere in the United States is from
seed of Chinese origin. The seed is nearly all grown in a limited
area on the bottom lands along the Kentucky River. Hemp is
dioecious, insuring cross-pollination, and since no attempt has been
made to select or breed pure strains there has been a general mixing
of all the strains introduced, resulting in a fairly uniform homo-
genous type. The foliage of this hemp, so far as observed, is always
a rich, (lark-green color hi healthy, normal plants.

In 1912 two pistillate plants were found with foliage of a bright-
purple color, in marked contrast to the normal green color. These
grew in a small plat of hemp cultivated for seed selection on the
Potomac Flats, Washington, D. C. The history of the seed from
which this mutation sprang is definitely known for nine generations.

SELECTION FOR IMPROVED HEMP.

In the spring of 1903 about one-half of a pound of seed of the
crop of 1002 grown in Kentucky was planted at the experiment sta-
tion at St. Anthony Park, Minn. The seed plat was carefully cul-
tivated, and oidy the heaviest seed from the one best pistillate
plant was saved for planting. Cultivation and selection were con-
tinued seven generations, including the crop of 1000, at the Min-
nesota Agricultural Experiment Station. The plats were small each
year and no other hemp was grown near enough to permit cross-
pollination. It was therefore more closely inbred than often hap-
pens in dioecious crops.

Seed from the crop of 1000 was planted at the experiment station
at Lexington, Ky.. in 1010, and again in 1911. A small plat, sown
broadcast for fiber production at the Kentucky station, was more
uniform than other hemp and produced fiber of better quality, but
no marked difference in the foliage was observed in any of the plats,
either in Minnesota or Kentucky.

i Issued Feb. IS, L913.

[Or. H3] 23

24 CIRCULAR NO. 113, BUREAU OF PLANT INDUSTRY.

FIRST APPEARANCE OF PURPLE FOLIAGE.

Seed of the crop of 1911 in Kentucky was planted at Washing-
ton, D. C, in 1912 and 60 well-developed pistillate plants and 46
staminate plants were produced. Two of the pistillate plants had
purple foliage, strikingly different from the normal dark-green color
of all others. One of these was the third best plant in the plat, and
the other was above the average. The seed of the best purple plant
was saved by itself. It is of a brownish color, easily distinguished
from the normal gray of the other seeds, and it is larger in size than
the others.

Sometimes the wild hemp from bird seed has foliage with a copper-
colored hue, and in 1901 a small plat of hemp from imported Hun-
garian seed had several plants with copper-colored foliage, but none
of these had the deep, rich purple color of the two marked plants
in this seed plat of 1912. This is a singular mutation, showing
strongly marked color variation hi foliage and seed, arising from a
closely inbred strain of a uniform group of plants.

ICir. 113]

THE "TUBER-UNIT" METHOD OF SEED-POTATO

IMPROVEMENT. 1

By William Stuart, Horticulturist, Horticultural Investigations and Arlington Farm.

INTRODUCTION.

The ever-increasing importance of the potato crop makes it highly
desirable that more attention should be given to the subject of seed
improvement. As yet comparatively little thought has been given
to this phase of the potato industry by the American grower. The
European grower, on the other hand, recognizes the futility of at-
tempting to grow profitable crops of potatoes without giving the
most careful attention to the source of his seed supply. To this end
he either buys his seed from specialists in seed-potato production or
becomes a specialist himself. The result of this attention to the use
of good seed is well attested by the average yields secured. For the
years 1901 to 1910, inclusive, the average per-acre yield in bushels
was as follows: Germany 200.8, Great Britain 200.7, and the United
States 92.7. It would be absurd, of course, to claim that all of this
increase was attributable to good seed, but it is safe to assume that
good seed does play a very important role and should not be over-
looked by the American grower.

DESCRIPTION OF THE "TUBER-UNIT" METHOD.

The "tuber-unit" method was first recommended by Webber. 2 It
consists for the first year in the selection from the seed bin before
planting time of as many of the most perfectly shaped tubers as may
be desired by the grower. These tubers should be selected with
regard to type, size, and uniformity. Tubers ranging from 6 to 8
ounces in weight are preferred. Before planting, the selected tubers
should be treated by immersing them two hours in a formalin solu-
tion consisting of 1 pint of formalin to 30 gallons of water. In plant-
ing, each tuber is quartered and dropped as cut. This is done by
splitting the bud-eye cluster in four parts from seed to stem end
of the tuber. In other words, the tuber should be cut through its
longitudinal axis. The four pieces of each tuber are dropped con-

1 Issued Feb. 15, 1913.

2 Webber, If. J. Plant-breeding for farmers. New York Cornell Agricultural Experiment Station,
Bulletin 251, Feb., 1908, pp. 322-331.

[Cir. 113] 25

26 CIRCULAR NO. 113, BUREAU OF PLANT INDUSTRY.

secutively in the row at a distance of from 10 to 14 inches apart,
according to the character and moisture content of the soil. By
allowing an additional spacing between each set of fours the four
plants from each tuber are definitely isolated from adjoining ones
and the grower can readily observe any variation in vigor and uni-
formity between the various units planted. It also enables him to
detect any mixtures which may be present in the variety. If a
potato planter with a fertilizer-distributing attachment is available,
it may be used to open the furrows and drop the fertilizer by simply
removing the covering disks and setting the plow a little deeper.

The seed pieces may be covered with a plow if care is exercised
not to displace them. During the growing season such observations
as are indicated in the accompanying form of Note Blank I should
be recorded (Table I). All units which do not satisfy the require-
ments of a good strain, such as uniformity of type in habit of
growth and character of foliage and size of plant, should be marked
as undesirable units. This constitutes the first step in both the
selection and elimination processes. At digging time the product
of each unit is separately harvested by hand and a further selection
made from the marked or surviving units. Only such units should
be retained for further trial as most nearly approach the desired
ideal, which in this case is uniformity in size, shape, and productive-
ness. The tubers from each selected unit are collectively placed in
separate sacks, preferably in cotton or burlap. Heavy manila paper
bags may be used, but there is always danger of their being torn.
Before being placed in storage, each sack should be numbered with
the varietal and tuber-unit number, in order to preserve its identity
in the field notes. The final examination should consist of notes,
recorded on the note blank, of the number and weight of merchant-
able and unmerchantable tubers. From each of the units not re-
jected at this time select 10 of the best tubers for the next season's
planting. It is desirable, though not absolutely essential, to main-
tain the study of each selection on the tuber-unit basis, because it
permits a more accurate comparison of the behavior of each. The
10 selected tubers from each original unit will give 40 plants for
study the second year. As in the previous season, notes should be
taken on the growing plants, recording them on Note Blank II
(Table I) . All selections which do not produce a reasonably uni-
form lot of plants should be marked for rejection. Each selection
should be harvested separately, as in the previous year, and such as
do not meet the requirements imposed should be rejected. The fur-
ther conduct of the work will consist in the multiplication of the
selected strains for field planting and the elimination of any weak
plants that may occur during this process.

[Cir. 113]

" TUBER-UNIT " METHOD OF SEED-POTATO IMPROVEMENT. 27

VALUE OF THE "TUBER-UNIT" METHOD.

The tuber-unit method was principally intended to serve as a
ready means by which high-yielding or uniformly shaped strains
could be isolated from the general mass. The writer conceives that
its most valuable feature lies not in the isolation of highly produc-
tive strains, but in the elimination of the unproductive and diseased
plants. Extensive experiments conducted by this office show con-
clusively that the simple elimination of the unfit tends to materially
increase the per-acre yields. A careful study of strong and weak

FIG. 1.— Strong (1) and weak (2) tuber units of Rural Blush potatoes. (Photograph of plants made about

August 20, 1911.)

tuber units of 12 varieties during the past two seasons, 1911 and
1912, shows in every instance a very close relationship between the
vigor of the plant and the tuber yield. The average production of
merchantable tubers from the strong plants was over 16 times
greater than from the weak ones, while in the case of the culls the
yield was only a little over twice as great. The data further show
that the average weight of merchantable tubers from the strong
plants was 5.3 ounces, as against 3.8 ounces from the weak ones.
The small tubers or culls from the strong plants averaged 1.7 ounces,

[Cir. 113]

28

CIRCULAR NO. 113, BUREAU OF PLANT INDUSTRY,

while the weak plants averaged but 1.1 ounces. The significance of

Jp8 po W)8 S°

(\|0 loo NO 50
000 ^o

■a
o

these data is fully realized only when considered
in its relation to the use of large or small tubers
for seed purposes. It is readily seen that but
a very small proportion, if indeed any, would
be included in the use of tubers weighing 5
ounces or over, whereas in the use of small
tubers the proportion of small tubers from the
weak plants would be, relatively speaking, much
greater. Herein lies the danger in the use of
small potatoes from unselected seed stock.

The accompanying reproductions of photo-
graphs shown in the top row of figure 1 repre-
sent strong and weak plants of two tuber units
of the Rural Blush potato in 1911 . These plants
were produced from apparently identical tubers,
were grown in the same row, and were subjected
in every particular to identical cultural condi-
tions. The reproductions in the middle row
represent the tubers produced by these plants,
while the lower set represent tne tubers pro-
duced in 1912 from five tuber units of the 1911
crop. The 1912 crop is divided into two classes,
according to size. The tubers on the left in
each illustration consist of those weighing 3
ounces or over. These are designated as mer-
chantable or prime tubers, while those on the
right are the culls.

SEED-SELECTION PLAT.

The accompanying diagram (fig. 2) of a plant-
ing plan is presented more as a suggestion for
use by boys' and girls' potato clubs than for
the average potato grower. A hundred tubers
make a convenient unit to handle, which at the
same time is capable of unlimited expansion in
the hands of those whose enthusiasm makes
them willing to undertake larger operations.

Each tuber unit in the diagram is numbered
in order to facilitate note taking, as well as the
keeping of a record on the behavior of each
individual tuber unit throughout the season.
By affixing their numbers to the units selected
for another season's trial it is possible to pre-
serve the identity of each selection throughout whatever period
of years it may be studied.

[Cir. 113)

\h§ ^O ^-0 0)0

(VtO SfrO NO tfvO

'0 '0

/v.O ~.o _0
(*)0 H>0 ^O <DO
C\|0 H.O NO <T)0

^0 * '0

.O O.O O

C\Jo No t\JO No

C\J° SJ-o NO *b°
"0 N

•*§ $ 8 ^° £2

<\Jg *§ N8 °>8

/^O ,0 O i_0

£>o 90 00 *>o
(\jo TSO £0 vTvO

"0^0 ' '0

~ ° W.O «. ° n. °

<J)0 JO 0)0 ^0

\o tyo v©o <j\o
'0

—O ^0 .0

<ho «*>o «0o t*5o

\0 V>0 v©o 0\O

'0

« . O f.,0
NO C\lo No ^O

^o sso voo 0)0

,00 .„o

voo ^-0 ^0 ^0

\O\hOV0OJ)O

-.
loo Oo 'oo Oo
<o tt-o voo <j\o

^0 '0

,0 «,o w _

\J»o <^o ^0 (Do

■s.0 coo voo <no

00 00

0-

_o _

(*0 H)o 0r\O °0O

vo <S° voo vx)0
'o

*!

V

// /2

oooo oooo

36 37

oooo oooo

61 62

oooo oooo

66 87

oooo oooo

0-

VJ

_o 1 _. .

Oo 'Oo Oo "Oo

^0 ho voo 050

'0 00

«o ..
rt-0 \f-o ovo >o
°>0 (V\0 1X0 V^jO

J o '0

->o ^0

(r>0 ^O QJO Do

<0g p,o Eg Qdo

. §cvj8n§j\8

No (nO InO VTQO
O 'O 'O O

O O.O O

(p,o "^0 V)o \o

^O OrvO l(\0 ttJO

'0 'o

O rsO 1 _ O

inO Oo >oo Oo

^0 Ooo loo vxo

J o

n\° «° mO
*lo J)o S-o Jio
^0 (\jo >oo No

oooo

C*8$8fc8$o8

^0 CVjo "oo No

oooo

k° ^ „

^0 CVjo loo No
000

1-.0 O lr.O

Cvjo 'OO N
OOOO

a

a

5
I

oi

6

£

"TUBER-UNIT" METHOD OF SEED-POTATO IMPROVEMENT. 29

The seed-selection plat need not be divorced from the potato field.
It can quite as readily be a portion of it, provided it is so located as
to make it easily accessible for note taking.

NOTE BLANKS.

The note blanks shown in Table 1 are those which have been
prepared for use in the Office of Horticultural Investigations and
Arlington Farm. Note Blank I is prepared with especial reference
to the first season's work.

Note Blank II is designed for notes to be taken on the selected
tuber units hi the second and subsequent season in which they are-
studied. The only difference between the two blanks is with respect
to the notes on the upper part of the sheet in which provision has
been made in Note Blank II for recording the previous year's
data relative to the time and place of selection of the tuber unit,
together with yield of tubers secured. With such data recorded,
the observer can very readily compare the behavior of the selection
in any given season with that of the preceding one. Each note b'ank
has provision for notes on 10 tuber units.

SUMMARY.

(1) Good seed is of vital importance in the production of a maxi-
mum crop of potatoes.

(2) Good seed may be obtained by the tuber-unit method through
the elimination of unproductive and weak plants.

(3) Like produces like. Plant tubers from unproductive or weak
plants and you will reap a similar harvest.

[Cir. 113]

30

CIRCULAR NO. 113, BUREAU OF PLANT INDUSTRY,

O

s

5.

£

w

pq

<

c

P

u

m

u

03

W

a

o
o

(*

u*

t*-.

BS

a

©

o

fc.

60
O

pq
w

H

o

4
£

3

-

I

o

a.

T3

C3
4>

e

o

>

o

a
o

4>

5

a

3 «

'r

St;

41 o
>.£.

en J-

i

Total
weight.

.

o

Ml

°3

N

V

Num-
ber
small.

! 2

to

■/'.

-;

<a .

X! 41

: BE?
: X

: 1

4

3

~— ■

os

1
w

: "S

S

•

0/ 4'

3 , ^
gag

o> £
Q •«

Li

-.2.5

•

•-7c
a a;

~ CO

a a)
4) 22

so

O 41

Num-
ber
plants.

■r

E

Date
sprouts
appear.

31 SOll

Weight
of tuber
(ounces).

S

o

O

41 ■

.a o

c

z

C3

X!

O

7 fc

J; a

a a

rt H

a p.

>. o

.tj 4

2 4

c

41 41

— .5

•- -C

O 4

3 S

O 0J

s a

ft -a

o .

4 C

S fi

•c 3

itf^

■a e

— ' 03

4i

S *

0J «s

(H

- >

a, ..

a a

a 4>

4) S
o 2

i

S

T3 i"

•3 £

rt

o ■»

>

** T)

l —

X) O

o

4) o

g 6£

s

•r I*

o

05

4>

e -3
-SP a

CO r$

;-

4>

-4^»

EH +

a

a

• *.

•rt

■g 1

5

2 nf

t-i

o —

> -a

R

^ C3

<B S

•" P

.=

B?"o*

J,

a ®
a »

a

be <u

Pt

S 3

53 .."

41

— ii

4 CU

O fc!

•—

B n

o

. X!

0>

4 ©

o

V. a.

4>

5JCD

o »

3 ^

« s

a -i

60 •

a
a

41

2.2

^ § S

41 a. 41
3 E "

4 S

?7

[Cir. 113]

"TUBER-UNIT" METHOD OF SEED-POTATO IMPROVEMENT. 31

a
:

H

y.
u
&
a
w
w

H
CO

-

a
s.

c
o
H
X

K

a
u

E;

c

o

O C6

3

a

■a
3
3
o

&

it

s

o

p.

6C

■a

03

O

a.

- r-

>

.C

o

3

X!

,0 pH

-a

■d

I

c

o

5

3

p *> a) cj
o 75 70 75
< /. M CO

Unit
saved or
rejected.

:

+3

^1

O

to

6f

to"

Num-
ber

small.

60

1^*

;-

QJ .
— I

a bo
a u.

-

1

■r.
«

3
H

a 1 !

O 03

s'S

1

-—
C3

CO

■*->

c

n

Desirable,

Un-
desirable.

$4

3 c6

svri

=5

9 *

Num-
ber
plants.

1

3
o

i-

p.

• ■

Date
sprouts
appear.

Weight
of tuber
( ounces 1.

5 .

^

6

! &

> &

^ — ■

f

o

o

[Cir. 113]

U. S. DEPARTMENT OF AGRICULTURE.

BUREAU OF PLANT INDUSTRY— Circular No. 114.
B. T. GALLOWAY, Chief of Bureau.

MISCELLANEOUS PAPERS.

Sowing Flax on Winterkilled Wheat Fields
Experiments in Subsoiling at San Antonio

M. W. EVANS

Control of the Black-Rot and Stem-Rot of the Sweet Potato

Bartlett Pear Precooling and Storage Investigations in the Rogue
River Valley

Climatic Conditions on the Truckee-Carson Project .

I S. H. HASTINGS
1 and C. R. LETTEER

L. L. HARTER

A. V. STUBENRAUCH
and H.J. RAMSEY

. F. B. HEADLEY

Issued February 24, 1913.

WASHINGTON:

GOVERNMENT PRINTING OFFICE.

1913.

BUREAU OF PLANT INDUSTRY.

Chief of Bureau, Beverly T. Galloway.
Assistant Chief of Bureau, William A. Taylor,
Editor, J. E. Rockwell.
Chief Clerk, James E. Jones.
[Cir. 114]

9

SOWING FLAX ON WINTERKILLED WHEAT FIELDS. 1

By M. W. Evans, Scientific Assistant, Forage-Crop Investigations.

INTRODUCTION.

Wherever wheat is sown in the fall and the climatic conditions
during the winter are unfavorable, it is more or less subject to injury
by winterkilling. Great losses are sometimes sustained in the winter-
wheat areas of the United States from this cause.

The farmers in the vicinity of New London, in northern Ohio,
have found a way to avoid much of this loss. AVhen the stand of

Pig. 1. — A Hold of fall-sown wheat and spring-sown flax at New London, Ohio.

wheat in the spring is very thin, flax is frequently sown on the field.
In the spring of 1912, when the wheat fields were in very bad condi-
tion as a result of winterkilling, a large proportion of the wheat
acreage in that locality was seeded with flax.

In this vicinity a 5-year rotation, consisting of corn, oats, wheat,
timothy and clover, and timothy, is generally followed. Occasionallv

[Clr, 114]

1 Issued Feb. 22, 1913.

4 CIRCULAR NO. 114, BUREAU OF PLANT INDUSTRY.

flax is sown instead of oats. Sometimes the meadow is plowed up
after one year and sometimes it produces hay for three years or more,
but the rotation just described is the customary one.

How or when the practice of sowing flax in wheat originated has not
been definitely determined. It has been followed around New Lon-
don, however, for a period of approximately twenty years. Although
careful inquiry has been made, no information has been obtained that
flax is sown on winter-wheat fields in any other locality in the
United States. There is apparently no reason why the practice would
not be applicable in other parts of the country where flax thrives and
where wheat is subject to winterkilling.

CONDITIONS UNDER WHICH FLAX IS SOWN.

Ordinarily when wheat has been injured by winterkilling, the
farmer either harvests a light crop that does not pay for the cost of
producing it or else he prepares the soil for a crop of spring grain,
which not only involves additional expense, but also often seriously
interferes with his regular rotation of crops. It is under such condi-
tions that the practice of sowing flax in the wheat field has been
found to be a means of avoiding loss and interference with the regu-
lar system of crop rotation.

It is a common practice of the farmers in the vicinity of New
London to sow timothy seed at the same time as the wheat. The
weather conditions in the fall of 15)11 were very favorable for the
growth of the timothy seedlings. In the spring of 1912 there was
such a well-developed stand of timothy in some of the wheat fields
that it was evident that the fields would produce a good crop of mixed
wheat and timothy hay or, if left to mature, a light crop of wheat
and a fair crop of timothy seed. Flax is not sown, as a rule, in fields
where the indications are that the timothy will produce a fair crop.

SEEDING.

The flax is generally sown about April 1, or as soon as the weather
has become warm enough to be favorable for plant growth and dan-
ger of freezing weather is past. Light frosts that sometimes occur
after the flax seedlings are up do not seem to harm them. The seed
is sown as early as it is safe to do so; otherwise, the flax does not
mature until somewhat later than the wheat. When the flax is sown
as early as April 1 it matures in time, so that there is practically no
loss of wheat through shattering, as is the case when it is sown after
the middle of April.

The flax is ordinarily sown broadcast with one of the various types
of hand seeders. A few farmers mix the flax with timothy and
clover and sow them together. As a rule the flax is sown separately
and sometimes at a later date than the clover.

[Cir. 114]

SOWING FLAX ON WINTERKILLED WHEAT FIELDS. 5

Flax germinates very readily. The farmers sow the seed without
harrowing- the soil either before or after sowing. When the first
rain occurs, the seed germinates and usually makes a good stand. A
light harrowing, however, would no doubt be beneficial to both the
flax and the wheat.

When a flax crop is grown alone, the farmers in this vicinity gen-
erally sow 1 bushel of seed to the acre. When the flax seed is sown
on winter-wheat fields, different farmers sow amounts of seed varying
from '2 to 4 pecks per acre. The quantity of seed sown also depends
somewhat upon the extent to which the wheat has been winterkilled.

HARVESTING.

As the wheat ripens somewhat earlier than the flax, the crop is
harvested as soon as possible after the flax is mature. As has already
been stated, the date at which the flax is ready to harvest depends
largely upon the date on which the seed was sown.

As flax matures, its stems become woody, so that it is compara-
tively difficult to cut. The knives of the machine used must therefore
be sharp. Flax is harvested in various ways. Farmers who own a
drop-rake reaper usually use that machine. Where there is a large
proportion of wheat growing in all parts of the field in mixture with
the flax, a binder does excellent work. In fields where the flax pre-
dominates, the binder does not work so well. Where the flax grows
alone, particularly if it is short and fine, it tends to get under the
canvas and become wound around different parts of the machine, so
that it sometimes becomes necessary to stop the machine and remove
the flax.

Most of the flax grown in this locality is cut with a mowing
machine, with an attachment behind the cutting bar which leaves it
in light windrows, out of the way of the horses' feet and the wheels
of the mower when the next swath is cut. By means of a 1-horse
wheel rake, two windrows may be gathered at once into small
bunches, where the flax lies until dry. The flax may be loaded, pref-
erably with large wooden forks, on hayracks and hauled directly
to the thrashing machine or stored in a barn or stack until it can be
thrashed.

THRASHING AND SEPARATING THE SEED.

The wheat and flax mixture is thrashed in the same way that
wheat alone is thrashed. They go through the same sieves and come
out through the grain spouts together. The two kinds of seed are
later separated with a fanning mill. The work of separating is
usually done at the grain warehouse by the dealer who purchases
the seed.

[Cir. 114]

6

CIRCULAR NO. 114, BUREAU OF PLANT INDUSTRY.

When a mixture of flax, wheat, and timothy is thrashed the timo-
thy seed is separated by the sieves from the flax and the wheat and
conies out beneath the machine.

The cost of thrashing is determined by the proportion of flax and
wheat after the seed has been separated. In this locality the
thrashers charge 12^ cents per bushel for thrashing flax and 4 cents
per bushel for wheat.

YIELDS.

The yield of flax and wheat depends upon several factors. Soil
and climatic conditions, as well as the relative thickness of the stand
of flax and wheat, are important.

In Table I are given the yields harvested from several typical
fields in 1912. It should be understood that neither the area of the
fields nor the yields were measured exactly: the figures are quoted
from the statements of the farmers growing the crop and are no
doubt approximately correct.

Table I. — Yields of flax, wheat, and timothy <m several farms near New London,

Ohio, in 1912.

Area.

Yield.

Field No.

Area.

Yield.

Field No.

Flax.

Wheat.

Timothy.

Flax.

Wheat.

Timothy.

1

A cres.
25
10

IS

61

Bushels.

250
93
39
40

Bushels.

137

IG

37

25

Bushels.

5

A cres.
6

9

Bushih.
39
46
GO
G3

Bushels.

4

13

8

40

Bushels.

S

2

6

G

3

32
3

7

4

8

The average yield of flax on the eight fields was 7.2 bushels per
acre, and of wheat 3.2 bushels. From four of the fields an average
yield of 1.3 bushels of timothy seed per acre was also obtained. At
the time when most of the crop was sold flaxseed was worth $1.48,
wheat $1, and timothy seed $1.25 per bushel. The average total
income per acre, therefore, from the eight fields was $14.64. The
average value of the flax on the eight fields was $10.65 per acre, and
of the wheat $3.20.

AA 7 hile $14.64 per acre is not a large income, it is much better than
would have been obtained from any one of these wheat fields in 1912
if the flaxseed had not been sown.

The yields of both wheat and flax were smaller in 1012 than have
been obtained on several farms in this vicinity in former years. The
price of flaxseed and timothy seed w 7 as also lower than it has ordi-
narily been in recent years. Several farmers have harvested yields
of flax and wheat the value of which has been equal to that obtained
from a good crop of wheat growing alone. One farmer produced

[Cir. 114]

SOWING FLAX ON WINTERKILLED WHEAT FIELDS. 7

approximately 11 bushels of flax and 13 bushels of wheat to the
acre. Another farmer harvested from a 9-acre held 11 bushels of
flax and 6| bushels of wheat per acre. A third farmer produced 12
bushels of flax and !> bushels of wheat to the acre.

UTILIZATION OF SEED AND STRAW.

The flax seed is usually sold to a local grain dealer. It is then
shipped to mills which manufacture linseed oil and linseed meal.

Large quantities of thrashed flax straw produced on fields where
the flax has been grown alone are sold at a local tow mill. When the
flax is grown in mixture with wheat, however, it can not be used for
making tow unless it contains only a very small proportion of the
wheat straw. The price of clear flax straw is at present $8 per ton.
Flax straw containing a small amount of wheat or timothy was sold
from a few farms for $7 per ton. As a rule, the mixed flax and
wheat straw is retained on the farm, to be used for bedding or to be
i'vd in the barnyard to stock during the winter.

EFFECT OF THE FLAX ON THE WHEAT, TIMOTHY, AND CLOVER.

It is the general opinion of those farmers who have sown flax in
their wheat fields that the amount of wheat produced is not appre-
ciably decreased by the presence of the flax. In those spots in the
field where there may be a good stand of wheat plants the flax seed-
lings make only a small growth. In those portions of the field where
the wheat has been winterkilled w T eeds would grow if the flax were
not present.

Timothy and clover are sown in a field where flax is growing in
mixture with the wheat in the same way that they are sown with
wheat growing alone. Many good meadows have been obtained where
timothy and clover have been seeded with wheat and flax.

CONCLUSION.

This method of sowing flax in wheat fields is not recommended as
a regular farm practice. It is merely a plan that may be resorted to
in years when wheat has been injured by winterkilling. After being
given a trial of 15 or 20 years in the locality where the method was
originated it is looked upon as one of the best ways to make use of a
wheat field that has been injured by winterkilling. There is appar-
ently no reason why it could not be adapted to a large proportion of
the winter-wheat area throughout the Eastern and Central States,
where it could be made a means of decreasing the loss that frequently
occurs in seasons following a winter unfavorable to the wheat crop.
[Clr. 114]

EXPERIMENTS IN SUBSOILING AT SAN ANTONIO. 1

P.y S. II. Hastings, Farm Superintendent, and C. u. Letteer, Assistant, Office

of Western Irrigation Agriculture.

INTRODUCTION.

San Antonio, Tex., is in what is classified as a semiarid region,
the normal annual rainfall, as shown by the records of the United
States Weather Bureau for the years 1891 to 1910, inclusive, being
24.77 inches. Because of the irregularity of the rainfall, the com-
pact character of the soil, which during torrential rainstorms causes
a large loss in surface run-off, and the high evaporation as compared
with more northern sections, soil moisture is one of the most im-
portant limiting factors in crop production. Any farm practice,
therefore, which will result in increased moisture content of the soil
will be of much importance to the agriculture of the region. It is
popularly thought that subsoiling will effect this and by so doing will
increase crop production.

Subsoiling consists in loosening the soil to a depth greater than it
can be loosened with an ordinary plow. This is accomplished by a
subsoil plow, which operates in the bottom of the furrow left by a
breaking plow, loosening the soil to the additional depth of about
12 inches. It requires a 2-horse or 3-horse team and a man to per-
form the operation, in addition to the man and team that do the
plowing, so that the cost of plowing and subsoiling is practically
double what is required for ordinary breaking. It is readily seen
that the practice is an expensive one, and adds so much to the cost
of preparation for a crop that unless it results in materially increased
yields it can not be profitably followed as a regular farm practice.
The cost of subsoiling at San Antonio has been found to be practi-
cally the same as plowing, or about $2 per acre.

SOIL CONDITIONS OF THE REGION. 2

San Antonio lies in the southern extension of what is known as
the Black Prairie region, or the Black Lands of Texas, and near the
northern edge of an area known geologically as the Rio Grande

i Issued Feb. 22, 1913.

"See Circular 34, Bureau of riant Industry, U. S. Department of Agriculture, entitled
" The work of the San Antonio Experiment Farm in 1908."

78241°— Cir. 114—13 2 9

10

CIRCULAR NO. 114, BUREAU OF PLANT INDUSTRY.

Plain. The soil .is mostly the result of weathering of limestone
rocks of the Upper Cretaceous period. Recent alluvial deposits have.
been washed down from the higher lands northwest of the city, re-
sulting in local modifications through the addition of coarser mate-
rial. The typical soil is a heavy black or brownish clay or clay loam.
The lime content of the soil is unusually high, the proportion of
lime in the upper 12 inches of soil varying from 7 to 23 per cent,
This lime occurs in the soil both as finely divided material and as
gravelly concretions. In the former condition it is generally dark
colored through staining by decomposed organic matter, while in
the latter condition it is usually white.

TEST OF SUBSOILING IN ROTATION AND TILLAGE EXPERIMENTS.

In order to determine the effect of subsoiling upon crop yields and
upon the moisture content of the soil at San Antonio, subsoiling
tests as a part of the rotation and tillage experiments were begun on
the San Antonio Experiment Farm in" the fall of 1009. This rota-
tion and tillage work is conducted on 82 plats, each one-fourth acre
in size. Three years' results have been obtained with subsoiling, and
it is believed that these results are sufficiently conclusive to justify
their publication. Corn, cotton, oats, grain sorghums, and sac-
charine sorghums are the crops grown in the rotations. The va-
rieties of the various crops referred to in this paper are as follows:
Corn (Laguna), three years; oats (Appier's Tviistproof), three
years; cotton (Triumph)'. 1010 and 1011, (Acala). 1012.

Those rotations in which subsoiling has been tested are shown in
Table I. The lists are so arranged that the rotations opposite each
other are directly comparable, the only difference in treatment being
that of subsoiling or its omission.

Table I. Lists of rotations in which subsoiling has been tested at San Antonio

Experiment Farm.

Yr.

Not subsoiled.

Rotation A5-D:

i late for grain, cowpeas; plow under cowpeas
in fall.

Cotton; plow in November.

Com, cowpeas, manure; plow under cowpeas
in fall.

Cotton; plow in November.
Rotation A 6- A:

Corn, plow in July.

Oats for hay; plow in May.
Rotation 15«V A:

Corn: plow in July.

Cotton; plow in November,
isolation B6-D:

Corn, cowpeas, manure: plow cowpeas in fall.

Cotton; plow in November.
Rotation B6-I i :

Corn: plow in February.

Cotton; plow in February.

Subsoiled.

Rotation A5-E:

' >ats for grain, cowpeas; plow and subsoil in
fall.

Cotton; plow in November.

Com, cowpeas, manure; plow cowpeas and
subsoil in fall.

Cotton; plow in November.
Rotation A6-C:

Corn; plow in July.

< »ats for hay; plow and subsoil in May.
Rotation B6-B':

Corn: plow and subsoil in July.

Coll on: plow and subsoil in November.
notation B6-E:

Corn, cowpeas, manure; plow cowpeas in fall.

Cotton; plow and subsoil in November.
Rotation lie. II:

Com; plow and subsoil in February.

Cotton; plow and subsoil in February.

[Cir. 114]

EXPERIMENTS IX SUBSOILING AT SAX ANTONIO.

11

EFFECT OF SUBSOILING ON CROP YIELDS.
Tabic II shows the effects of subsoiling on the yields of corn, cot-
ton, and oats.

Table II. — Average yield of crops in rotations subsoiled and rotations not sub-
soiled at the San Antonio Experiment Farm, WW to 191.2, inclusive.

Rotation Nos.

Not sub-
soiled.

Sub-
soiled.

(iain( + )

or
loss ( — ).

Rotation Nos.

Not sub-
soiled.

Sub-
soiled.

Gain( + )

or
loss ( — ).

COKN (Bushels
teh Acre i.

A6-A and A6-C
B(V-A and B6-B....

B6-Gand B6-H....

13.0
24. S
21.1
21.2

20.9

1 1.0
■rl 2

Is! 7
20.6

17.9

+ 1.0
—2.6
-2.4

- .0
-3.0

Cotton ( Pounds of
Ski: n Cotton per
Acre).

A.5 I) and A5-E

B6 A and B0-B

B6-G and B6-H

Average

Average of 10 rota-
tions:

1910

4S7.3
5119.3
501.7
614.0
574.7

436.7

520.0

5iii.:i
643.3

502.0

- 50.0
+ 10.7

- 00.4
+ 29.3

- 72.7

Average

20.2

18.7

-1.5

549.4

521.9

- 27.5

Average of LO rota-
tions:
1910

10.8
11.9
37.9

7.2
11.8

37.0

-3.6

- .1

- .9

420.0
507.2
714.4

397.0
190. 1
077.0

- 29.0

I'll i

1911

- 16.8

1912

1912

- 30.8

3-year average.

Oats (Pounds of
Hay per Acre).

A0- A and A6-C:

3-ycar average...
Average of 2 rota-
tions:

1910

3-year average.

20.2

IS. 7

-1.5

549. 4

521.9

- 27.5

Oats (Bushels
per Acre).

A5-I) and A5-E:

3-year average. .
Average of 2 rota-
tions:
1910. .

11.5

5.1

5.4

24.1

11.7

3.8

0. 2

25.0

+ .2

-1.3
+ .8
+ .9

3,557

2,050
3,056
5,560

3,121

2,344

1,9111

5,056

- 430
+ 2SS

1911

1911

-1,092

1912..

1912

- 504

Summary of Averages.

Crop.

Corn bushels. .

Cotton pounds..

Oats, grain bushels. .

Oats, hay pounds. .

A verage

Yield per acre.

Not sub-
soiled.

20.2
549. 4
11.5
3,557

Sub-
soiled.

18.7

521.9

11.7

3,121

Gain(+)

or
loss (— ).

- 1.5

- 27.5
+ .2
-430

Yield expressed in percent-
ages of untreated plats.

Not sub-
soiled.

100
100
100
100

100

Sub-
soiled.

92.6
95

101.7
87.7

94. 25

Gain(-f)

or
loss (— ).

-7.4
- 5
+ 1-7
-12.3

From the foregoing table it is seen that in four rotations snl (soil-
ing has decreased the yields of corn and in one rotation it has
slightly increased the yield. The cotton yields have been increased
in two rotations and decreased in three rotations. In those portions
of the table showing the average yield by years it is seen that subsoil-
ing gave decreased average }delds of corn and cotton in each of the
three years. The summary shows that when the three-year average
for all crops concerned is taken into consideration, subsoiling has
decreased the yield of all crops except oats for grain. Since the tests
with oats have not been so extensive as with other crops, it is doubt-
ful whether this result can be taken as conclusive.
[Cir. 114]

12

CIRCULAR NO.

114, BUREAU OF PLANT INDUSTRY.

RESIDUAL EFFECT OF SUBSOILING ON THE YIELDS OF CORN AND

COTTON.

In order to call attention to the possible residual effect of subsoil-
ing on succeeding crops for several years after the operation, Table
III has been compiled to show the effect of subsoiling on the yields
of corn and cotton where the length of time intervening between the
time of subsoiling and that of planting varies from 1 to 16 months.

Table III.— Effect of different lengths of time between subsoiling and planting
on the average yield of corn and vOlton at the Han Antonio Experiment
Fa rm.

Crop.

Cain or loss, per acre, due to sub-
soiling.

15 lo lti
months.

3toS
months.

1 to 2
months.

bushels. .

+ 1.3

+ 7

- 1.5
-31.4

- 3.1

pounds. .

-72.7

There is a marked decrease in the yields of both corn and cotton
where the subsoiling was done from 1 to 2 months before planting,
and a decrease where the subsoiling was done from 3 to 8 months
before planting, while when the subsoiling was done from 15 to 10
months before planting there was a very slight increase. While
these results are not at all conclusive, owing to the fact that the
experiments cover only two years and the number of plats is limited,
yet they are suggestive. The figures in this table indicate that loosen-
ing the subsoil is detrimental the first few months and also at the
end of over a year, when under ordinary conditions if there was to
be any resulting benefit it should be apparent. During the interven-
ing period between subsoiling and planting, a crop was grown and
the land replowed, and it would be supposed that any unfavorable
soil conditions brought about by subsoiling would have time to read-
just themselves during the period. Yet the increase in average yields
as shown by the table is so slight as to indicate very little result-
ing benefit from the operation. In no case was the increase sufficient
to cover the expense of subsoiling.
EFFECT OF SUBSOILING ON THE MOISTURE CONTENT OF THE SOIL.

Determinations of soil moisture were made at regular intervals of
about one month on rotations B6-A and BG-B for a period of two
years. Table IV shows the percentage of moisture in plat B6-1 as
compared with B6-3, and BG-2 as compared with B6-4; also the
crops grown on each plat each year and the time of plowing or
subsoiling. The soil samples for moisture determinations were taken
by means of the standard soil tube, two cores being taken on each
plat and composited to a single sample. This composite sample
was then weighed, dried, and weighed again, and the percentage of
moisture computed on the basis of dry soil.

[C'ir. 114]

EXPERIMENTS IN SUE-SOILING AT SAN ANTONIO.

13

Table IV. — Average moisture content in land subsoiled and land not subsoiled

at the San Antonio Experiment Farm.'

Crop and year.

Plat B6-1.

Plat B 6-3.

Sampled.

Plowed.

Moisture
content.

Plowed and
subsoiled.

Moisture
content.

1910.
Nov. 21

November
do

10.9
1.3.3

12.4
11.9
12.4
13.5
14.1
8.7

9. 5

9.8
10.1
17.3*
16.2*

14.0

16.6*

16.6

20.2*

17.9*

16.

13.5*

10.5

10.6*

14.7*

November

do

do

do

do

do

do

do

do

1911.

do

do

do

do

do

do

do

do

do

do

do

do

do

11.7

Dec. 27

11.6

1911.
Jan. 12

do

11.0

Feb. 13

do

11.7

Mar. 11

Corn, 1911

do

10.3

do

do

12.4

Ma v 9

..do

..do

13.7

June 15

..do

do

8.9

July 17

..do

..do

9.1

Aug. 17

..do

1911.
do

9.5

Sept. 16

do

9.4

Nov. 16

..do

.do....

is. 3*

Dec. 13

..do

do

16. 7*

1912.
Jan. 9

do...

.do .

12.0

Feb. 9

...do

...do

16.4*

Mar. 11

..do

..do

16. 2

Apr. 17

Cotton, 1912

do

do

L9.6*

17. S*

June 12

..do

..do...

14. S

July 9

..do...

.do...

12.6*

Aug. 22

do

...do

9.2

do

do

Hi :;i :

Oct. 24

..do...

.do...

14.8*

Mean

13.0

12.4

Crop and year.

Plat B 6-2.

Plat 15 6-4.

Sampled.

Plowed.

Moisture
content.

Plowed and
subsoiled.

Moisture
content.

1910.
Aug. 17-18

Corn, 1910
do

Aug. 1 . . . .
do

10.2
9.1
11.0
11.3-
11.7

11.1
11.5
10.8
12.5
13.7
10.5

9.7
lu.1

9.4
16.6*

15.0*

11.5
16.9*
15.3
20.5*
17.1*
11.5
1 1 . 3*
9.8
10.9*
15.2*

Aug. 2....

do

do

do

do

do

do

do

do

do

do

do

do

do

do....

1911.
Nov. 29...

do

do....

do

do

do

do

do

do

do

do

10.8

Sept. 19

9.2

Oct. 14

..do...

....do

10.5

Nov. 21

..do...

..do...

9.6

Dec. 27

..do...

do...

11.9

1911.
Jan. 12

..do

do

10.0

do

11 6

Mar. 11

.do...

..do...

10.6

Cotton, 1911
do..

do

.do..

13

May 9

13.7

June 1")

...do

..do

8.6

July 17

.do

.do...

9.3

Aug. 17

.do...

.do...

8.3

Sept. 16

.do...

.do .

S. 3

Nov. 16

.do...

..do...

1.")..")*

1911.
Dec. 11....

. do

16. 2*

1912.
Jan. 11

do ..

12.2

Feb. 9

do

.do

15. 2*

Mar. 11

Corn, 1912. .

.do ..

16.7

Apr. 17

.do...

do

18.9*

May 18

.do...

do .

16.7*

June 12

.do...

..do...

1 2. 6

July 9

.do...

..do...

10.8*

Aug. 22

.do

.do...

9.9

do

9.2*

Oct. 24

.do

..do...

15.1*

Mean

11.9

11.6

1 Each moisture determination marked with a star (*) is an average of 3 feet; all others are averages of 6
feet.

rCir. 114]

14 CIRCULAR NO. 114, BUREAU OF PLANT INDUSTRY.

Table IV shows that in only a very few instances has the percent-
age of moisture in subsoiled land been greater than that in land not
subsoiled, while in most instances it has been somewhat less. The
means were obtained, not from the monthly averages, but by con-
sidering the entire number of samples taken during the sampling
period. In both pairs of plats the averages indicate a larger per-
centage of soil moisture in land not subsoiled than in the subsoiled
land.

CONCLUSIONS.

(1) Subsoiling is an expensive practice and so adds to the cost of
preparation for a crop that unless materially increased yields result
it can not be profitably adopted as a regular farm practice.

(2) Subsoiling has been tested at the San Antonio Experiment
Farm for three years in rotation experiments with corn, cotton, and
oats for hay and for grain.

(3) The yields of corn, cotton, and oats for hay and for grain have
been either slightly increased or slightly decreased on subsoiled land.
In no instance has the difference been significant.

(4) The depressing residual effect of subsoiling on the yields of
corn and cotton was most marked when the crop was planted from
1 to 8 months after subsoiling; 15 months after subsoiling but little
depressing effect was noted.

(5) In the soil-moisture studies so far made at San Antonio it has
been found that subsoiling has not increased the moisture content of
the soil.

((>) The results of these tests indicate that since neither the
moisture content of the soil nor the yields of corn, cotton, and oats
are increased by subsoiling, the practice is not advisable in connec-
tion with the crops mentioned in the San Antonio region of Texas.
[Cir. 114]

CONTROL OF THE BLACK-ROT AND STEM-ROT OF THE SWEET

POTATO. 1

By L. I.. Harter, Pathologist, Cotton and Truck Disease and Sugar-Plant

Investigations.

INTRODUCTION.

This article outlines some of the precautions necessary to reduce
the widespread loss of the sweet-potato crop by the black-rot 2 and
stem-rot 3 organisms. Several papers on sweet-pot