Historic, archived document

Do not assume content reflects current
scientific knowledge, policies, or practices.

~ a ™ ate
0S DEPARIMENT OF AGRICULTURE. \ a -
BUREAU OF PLANT INDUSTRY—Circular No. 115 ee hand
| ADY 18 1944
B. T. GALLOWAY, Chief of Bureau. MAT 4
a «

>

MISCELLANEOUS PAPERS.

: La aye ,
Two Types of Proliferation in Alfalfa SAMUEL GARVER

A New System of Cotton Culture 5 ‘ : : : : : , 0. F. COOK

| R. A. OAKLEY and

Grasses for Canal Banks in Western South Dakota : , : ‘ A. C. DILLMAN

Issued March 1, 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. EK. ROCKWELL.

Chief Clerk, JAMES E. JONES.

IVOIRE SOF PROUEERATION IN-ALPALEA

By R. A. OAKLEY, Agronomist, and SAMUEL GARVER, Scientific Assistant, Office
of Forage-Crop Investigations.

INTRODUCTION.

The production of rhizomes and underground shoots in alfalfa
was observed many years ago, and brief, more or less indefinite, men-
tion made of them in literature. It may be that Jethro Tull had in
mind something of this nature in making the following statement in
his “ Horse-Hoing Husbandry ”:

Though one lucern root be much taper than another toward the upper part
of it, ‘tis sometimes seen that a single hoed plant of it has many of these per-
pendicular roots; some of them springing out from the very branches of its
crown.”

M. Lullin de Chateau-vieux may also have observed underground
stems, for in writing to M. Duhamel regarding his experience with
transplanted alfalfa plants, he says:

The stalks seem to rise out of the earth; and from the first time of cutting
them, a kind of head forms just above ground, which extends itself every

year. * * * 'These crowns have become of an oval form, having extended
themselves on the sides where they met with no resistance.®

It is very evident, however, that neither Tull nor Chateau-vieux
appreciated the value of the underground stem character, even if
they were aware of its existence, since they make no references to it
other than those above quoted. It was Thomas Le Blane who appar-
ently first called definite attention to the branching of alfalfa below
the surface of the ground, and not only should credit be given him
for calling attention to this character, but also for pointing out its
value to the plant. His statement, which follows, quoted in Young’s

1 Issued Mar. 1, 1913.

Proliferation : This word is derived from proles (shoots or offspring) and fero (1 bear),
so that the word literally means bearing offspring. Definitions.—Bearing progeny as off-
shoots.—Jackson, B. D., Glossary of Botanic Terms, ed. 2, London, 1905. The develop-
ment of an organ or a shoot from an organ which is itself normally ultimate.—Century
Dictionary and Cyclopedia. Botanically it has usually been used in reference to shoots
produced abnormally, but there seems to be no reason why the meaning should not be
extended to cover shoot production in general.

27Tull, Jethro. Horse-Hoing Husbandry, London, 1733, p. 92-102.

3 Duhamel du Moneceau, H, L. A Practical Treatise of Husbandry, London, 1759, p. 358.

[Cir. 115] : 3

CT)

4 CIRCULAR NO. 115, BUREAU OF PLANT INDUSTRY. ‘

Annais of Agriculture, leaves no doubt as to what he observed or of
his appreciation of the importance of the character:

My reasons for preferring it [variegated medick or variegated alfalfa} to
lucern are that it is hardier in bearing cold; that from its habit of branching
below the surface of the ground and the shoots being much more numerous, it is
not choked by the natural grasses, and that for the same reason it will not be
injured by being fed by sheep.*

Credit must also be given Le Blanc for advocating the value of
variegated alfalfa and for his efforts along the line of introducing this
strain into general use. The history of his experience with “ varie-
gated medick,” as written by himself, is very interesting and has a
very definite bearing on the subject of stem proliferation.

While underground stems are found in what is considered now to
be genuine JJ/edicago sativa, they are by no means as common or as
well developed as in hybrids between J/. sativa and I. falcata or in
pure J/. falcata.

There is, therefore, little wonder that Le Blanc should be among
the first to observe and appreciate them, since he was among the
first to study carefully from an agronomic standpoint J/edicago falcata
and its natural hybrids. But notwithstanding the definiteness with
which Le Blane wrote, the subterranean stem character was given
little attention by subsequent botanists and agriculturists; or, at any
rate, it was very soon lost sight of.

A search through literature fails to reveal any mention of under-
ground shoots or rhizomes from the time of the publication of Le
Blane’s work in Young’s Annals of Agriculture, 1791, up to the
publication by Blinn in 1911° of a bulletin entitled “Alfalfa: The
Relation of Type to Hardiness,” and a bulletin by Oliver in 19137 on
“Some New Alfalfa Varieties for Pastures.” These investigators
have called attention for the first time in recent literature to the
preduction of rhizomes in certain forms of alfalfa and the import-
ant part they play in rendering the plant resistant to severe condi-
tions, especially those of drought and coid.

That certain types of alfalfa have a tendency to produce high
crowns, while others have their crowns at or below the surface of the
ground, has been a matter of common observation for some time.
Moreover, this character has been associated in a rather general way
with hardiness. Brand and Waldron, in discussing qualities produc-
ing hardiness in different races of alfalfa, make the following state-
ment :

1ZLe Blane, Thomas. Experiments on the variegated medick. Annals of Agriculture,

v. 15, p. 279, 1791. (Le Blanc’s observations were made about 1783.)

2Tdem, p. 277.

3 Blinn, P. kK. Alfalfa: The relation of type to hardiness. Colorado Agricultural Ex-
periment Station, Bulletin 181, 16 p., illus., 1911.

4 Oliver, G. W. Some new alfalfa varieties for pastures. U. 8. Department of <Agri-
culture, Bureau of Plant Industry, Bulletin 258, 39 p., 11 pl., 1913.

Cire Ads, |

TWO TYPES OF PROLIFERATION IN ALFALFA. 5

The Mongolian alfalfa, which proved to be the hardiest under the severe
test at Dickinson of any of the newly imported strains, was found both there
and at Stockton, Kans., to have crowns more deeply set in the soil than other
varieties. It seems likely that this adaptation, if it may be called an adaptation,
is of importance in giving the tenderest part of the plant.the needed protection.’

It is evident that while these investigators take cognizance of the
deep-set crowns they fail to find rhizomes or underground stems pro-
duced from the plants, or at least to note their presence. In their
discussion of the root system of the Grimm variety they recognize
only the ability of this strain to recover after the taproot has been
broken off 4 to 8 inches below the surface, and state in this connec-
tion that “this condition has suggested the possibility that the long
endurance of stands of the Grimm strain may be due in some
measure to the capacity fer putting out new roots and reestablishing
itself after the taproot has been broken.” This statement is in-
tended in no wise to convey the idea of proliferation, but only of a
reestablishment of the original crown. Headden hints of this in
mentioning the old method of transplanting mature plants and the
continuation of gopher-eaten plants in some soils,? but he also did
not have in mind the production of new plants vegetatively from
stems or roots.

Just why the production of rhizomes and underground stems in
alfalfa was lost sight of until the last few vears is hard to explain,
but the fact that they were observed at about the same date by sev-
eral investigators working independently is very largely due to the
comparatively recent increased interest in breeding and selection
and the introduction of Medicago falcata and other alfalfas. The
erowing of plants in rows and hills in connection with breeding
furnishes ideal conditions for rhizome development, and the intro-
duction of new alfalfas furnishes forms in which rhizomes are very
commonly produced. The practice of growing alfalfa in rows and
hills, strongly advocated in Europe in the time of Tull and Harte,’
afforded Le Blanc and others a condition for study such as exists at
the present time.

STEM PROLIFERATION, CHARACTERS, AND FUNCTIONS.

Conclusions of other investigators.—In his study of the relation of
type to the hardiness of alfalfa Blinn found some very important
differences between the type of crown and the stooling habits in
certain hardy and nonhardy strains.* He states that certain of the

1 Brand, C. J. and Waldron, L. R. Cold resistance of alfalfa and some factors influenc-
ing it. U. S. Department of Agriculture, Bureau of Plant Industry, Bulletin 185, p.
68-69, 1910.

2 Headden, W. P. Alfalfa. Colorado Agricultural Experiment Station, Bulletin 35, p. 6,
1896.

3 Harte, Walter. An account of Some experiments tending to improve the culture of
lucern by transplantation. Essay II of his Essays on Husbandry, ed. 2, London, 1770,
illus.

4Blinn, P. K. Op. cit.

[Cir. 115]

6 CIRCULAR NO. 115, BUREAU OF PLANT INDUSTRY.

nonhardy types under his observation have “compact. upright-
growing crowns, with comparatively few buds or shoots below the
surface of the soil”; while certain of the hardy types are “ char-
acterized by a more spreading crown, with buds and shoots springing
from the crown below the surface of the soil.” He has not only
observed these underground buds, or shoots, as he calls them, but cor-
relates them with drought resistance and hardiness. Furthermore,
he notes the tendency of the rhizomes to take root at some distance
trom the old crown and produce new plants. No detailed descrip-
tion, however, is given by him of these underground growths or the
manner or time of their development.

Oliver, in his work with numerous alfalfas, including natural and
artificial hybrids of J/edicago sativa and MW. falcata, developed some

Fic. 1.—A plant of hybrid alfalfa six months from the seedling stage. The prominent
growths appearing below the crown are rhizomes; in other words, true stems. (I*rom
Bulletin 258 of the Bureau of Plant Industry, by George W. Oliver.)

exceedingly interesting material. He finds that there are two gen-

eral types of rhizomes produced—rooting and nonrooting—and that

during the early stage of their development they appear to be of the
same character. He also finds that both types of rhizomes are pro-
duced most abundantly during the autumn months and that they
originate at or near the crown of the plant, one case only coming
under his observation where they were produced below the surface.’
in this case the rhizome was at least 3 inches below the crown.

Oliver gives a rather detailed description of the gross morphology |

of the rhizomes, their life history, and functions. His descriptions

and photographs show positively that the underground growths
which he describes are true and modified rhizomes, according to the
accepted definition of the term. In other words, they are of stem

ovens (Hic 13)

1 Oliver, G. W. Op. cit. 2 Oliver: G: We “Op. cit. ps 22.) Neenioes
[Cir. 115]

TWO TYPES OF PROLIFERATION IN ALFALFA. | m

In summarizing the observations and conclusions of the various
investigators it will be found that certain types of alfalfa, chiefly
those which are associated with hardiness and drought resistance,
produce underground stems and rhizomes; that these growths are
true stems having nodes, scales, and buds; that although sometimes
found several inches beneath the surface, they usually originate at
or near the crown, seldom below; that the rooting and nonrooting
rhizomes may result in new shoots and in some cases ultimately in
independent plants; that by virtue of their ability to store up water
and produce protected buds they constitute a drought-resistant and
hardy character of much value to the plant; and that the rhizomes
in their various forms are more abundantly produced in J/edicago
falcata and its natural and artificial hybrids with J/. sativa than in
pure J/. sativa.

The rediscovery, if such it may be called, of stem proliferation in
alfalfa is of vast importance, especially to plant breeders, as it gives
them a tangible character with which to work in developing hardy
and drought-resistant strains.

Field studies and correlation of characters—In May, 1912, the
writers took up a general study of proliferation in J/edicago falcata
with a view to determining the extent of its prevalence and variation
in the different types of that species. Although late spring is prob-
ably not the best season of the year in which to study this character,
the abundant material in the nurseries at Brookings and Highmore,
S. Dak., afforded an excellent opportunity for observation, especially
since the plants were growing in cultivated hills and most of them
entering into their fourth season of growth.*

A careful examination of the numerous forms of J/edicago falcata
showed underground stems or rhizomes to be more prevalent and
better developed in the low-spreading, fine-leaved, fine-stemmed types
than in those more closely resembling J/. sativa. However, not all
of the forms included in the low-spreading type exhibit tendencies
to produce new plants vegetatively. Some spread only over the
surface of the ground, the procumbent stems proceeding from fairly
compact and high-set crowns without becoming attached to the soil
at any point. It was extremely difficult in the case of certain pro-
cumbent types to determine whether proliferation was largely the
result of an inherent tendency of the plant or was induced to a
considerable degree by cultivation. Cultivation unquestionably is
im a measure responsible for rooting stems in the erect as well as
in the procumbent forms. This point was quite definitely demon-

1The plants were grown in the nursery at Brookings and transplanted in hills at
Highmore in April, 1910. It is a well-known fact that the pruning incident to trans-
planting large alfalfa plants affects the development of the taproot. The above-mentioned
plants, however, were so small at the time of transplanting that little change in this
respect is believed to have been produced.

[Cir. 115]

8 CIRCULAR NO. 115, BUREAU OF PLANT INDUSTRY.

strated in experiments conducted at the Arlington Experimental
Farm during the past season.

In a great many cases where stem proliferation is developed to
2 considerable degree the center of the plant dies out, leaving that
portion of the crown quite devoid of shoots. A similar condition
is found in the true grasses, especially in forms of Festuca rubra and
other fine-leaved fescues. Many explanations of this condition have
been based on erroneous conclusions; in other words, it has been
argued that underground stems and rhizomes by virtue of their
ability to produce new growth keep these old plants alive, even after
the original crown has died. As a matter of fact, the barrenness
of the old crown is probably due in most cases to unfavorable con-
ditions for the development of buds or new growth in the center of
the clump. Where the original plant is grown in a crowded condi-
tion or in a condition not conductive to spreading, the center remains
productive for an indefinite period. The fescues already referred
to behave almost exactly in this manner. Barren centers are found
very frequently in procumbent types, while in the semierect or up-
right plants the tendency is not so common.

ROOT PROLIFERATION.

The brief studies at Brookings and Highmore developed nothing
particularly new regarding the type of proliferation heretofore
described, but tended only to confirm the conclusions of other investi-
gators. The studies disclosed, however, what is thought to be a type
of proliferation heretofore undescribed as occurring in alfalfa—a
type wherein new plants are produced from true lateral roots 4 to
12 inches, rarely more, beneath the surface and running almost
directly parallel with it.1 These proliferous roots apparently are
given off from any portion of the branching taproot within the limi-
tation of the depths above indicated. At irregular intervals along
the lateral roots, swellings occur, usually about twice the diameter
of the roots and three-fourths of an inch to 14 inches in length.
(See figs. 2and3.) Upon these swellings buds appear, some of which
give rise to new plants, while many remain undeveloped.? Fibrous
roots are rather sparingly produced from the lateral proliferous
roots, and in many instances the swellings are devoid of them; in

1 Hitchcock, A. S., and Clothier, G. L. Fifth report on Kansas weeds—Vegetative prop-
agation of perennial weeds. Kansas Agricultural Experiment Station, Bulletin 76, 23 p.,
12 pl., 1898. Clothier, G.L. Root propagation of Ipomoea leptophylla. Botanical Gazette,
vy. 25, p. 52-53, illus.,. 1898. These articles discuss vegetative reproduction by true roots
in Rhus glabra, Ambrosia psilostachya, Cnicus undulatus, Convolvulus arvensis, and
Tpomoea leptophylla.

2In this type of proliferation it will be noted that definite provision is apparently
made by the lateral proliferous roots for the production of new plants. There is a mate-

rial difference in this respect from the case in stem proliferation noted by Oliver, where
new shoots arise from the taproot appreciably below the crown.

Riri]

WO VEYPHS OH RPROLIBRRATION IN ALKALHA. 9

fact, new aerial shoots reach a considerable degree of development
before roots of any importance are sent out from the swellings.
There appears to be quite a difference in the plants that arise from

Fig. 2.—A plant of Medicago falcata (S. P. I. No. 28071) found at Highmore, 8. Dak.,
May, 1912. The root proliferation is remarkably developed in this plant.

the proliferous roots with respect to root development; some develop
a crown and root system of their own at an early date, while others
rely for an indefinite period on the proliferating roots for their

&

Fic. 5.—Sections of proliferating roots of alfalfa, showing swellings upon which the buds
that develop new plants appear. The buds are also sometimes produced from other por-
tions of these roots, but so far no shoots have been observed to arise from thein.

support. (See figs. 4, 5, 6, and 7.) This point, however, needs a
more thorough investigation, as the number of cases thus far found is
insufficient for definite conclusions.

79505°— Cir. 115—13 2

10 CIRCULAR NO. 115, BUREAU OF PLANT INDUSTRY.

The behavior of the proliferous roots with respect to growth in
diameter is quite peculiar. In comparatively few cases do they ex-
ceed one-fourth of an inch, except at the swellings, the average di-—

coos A — a |

Fic. 4.—Sections of proliferating roots of alfalfa. showing buds of the new shoots in
different stages of development. ;

Fic. 5.—A portion of a plant of Medicago falcata (S. P. I. No. 28071), showing the origi-
nal crown, with new plants that have resulted from proliferating roots. Some of the
new plants, it will be noted, have already developed crown and taproot.

ameter being about three-sixteenths of an inch, but it is by no means

uniform. At the point of origin the roots may be much smaller than

at some distance from this point. At 2 or more feet from the main

root they may become reduced in size to one-sixteenth of an inch, en-
[Cir, 115] =

TWO TYPES OF PROLIFERATION IN ALFALFA. 11

larging again to a diameter equal to or greater than any previously
attained. The greatest length of any proliferating lateral root so

Fic. 6.—Plant of Medicago falcata (S. P. I. No. 28071), showing new plants produced from
proliferating roots. The character of growth in the diameter of these roots is typified
in this illustration. Plants of considerable size are produced in this case without a
definite root system of tbeir own.

Fic. 7.—Shoots of alfalfa starting from proliferating roots. These shoots had their origin
from 4 to 6 inches beneath the surface and when photographed were depending entirely
on the proliferous root for support.

- far found is a little over 4 feet, but indications are that their growth
in length is limited largely by soil conditions—a loose, loamy soil
[Cir. 115]

12 CIRCULAR NO. 115, BUREAU OF PLANT INDUSTRY.

being conducive to their development (fig. 8). Many of the plants
produced from these proliferous roots in time form a new crown and
a more or less branching taproot, which in turn is capable of sending |
out proliferous roots, as does the original plant. No cases have so
far been observed of the new plants becoming separated from the
mother plant, even after they have become well equipped with a
root system of their own, and while no evidence has been noted of
the lateral roots disintegrating or becoming atrophied at any point.
it 1s reasonable to assume that at least some of the new plants in
time become detached, either through natural process of growth or
through accidental causes.

Fic. 8.—Proliferating root of alfalfa, showing branching. While this character is quite
common, a single root does not appear to branch more than three or four times at most.

The type of proliferation described in this paper was found at
Highmore, S. Dak., in a fairly upright plant of Medicago falcata,
seed of which was secured from Orenburg Government, Russia, and
introduced under 8. P. I. No. 28071. The mother plant was per-
fectly normal in the appearance of its growth above ground, and
attention was attracted to it only by the fact that smaller plants of
the same general appearance surrounded it, coming up even from
the portions of the row that had been cultivated frequently the sea-
son before. The plants were so large that they could be readily dis-
tinguished from seedlings and were arranged in such manner as

[Cir. 115]

TWO TYPES OF PROLIFERATION IN ALFALFA. 13

to indicate vegetative propagation. Later in the season other plants
were found at Highmore having root proliferation remarkably de-
veloped, but they were all from seed bearing the S. P. I. number
mentioned, and all conform in a general way to the same type with,
respect to growth aboveground. There is no definite evidence re-
garding the period in the plant’s life in which proliferous roots are
first developed. There are some data, however, to indicate that these
roots are not produced until the plant is at least 2 years old.

The production of proliferating roots in alfalfa seemed so ab-
normal as to indicate a possible pathogenic condition such as might
be caused by crown-gall. The external symptoms of this disease,
however, are not visible, and a careful microscopic examination of
live material failed to disclose the presence of pathogenic organisms
of any kind.!

Critical studies are now under way to determine as fully as pos-
sible the conditions which induce the roots to produce new plants.
In this connection the point of origin of the later roots, the swell-
‘ings on the roots, and the buds on the swellings and other portions
of the roots are being given careful consideration. Field studies are
also being made to determine the hfe history of these shoot-bearing
roots, the conditions favorable to their development, and the possibil-
ity of utilizing the character which they represent in the development
of strains of alfalfa particularly adapted to severe conditions. No
prediction is ventured regarding the agronomic value of the pro-
liferous root character, especially since it is not known to what
extent this character is transmitted or whether cultural conditions
influence its development. Close and open fertilized seed has been
secured from plants possessing proliferous roots and will be sown in
the spring of 1913. New plants have already been produced from
stem and root cuttings, so that abundant material should soon be
available for use in determining many of the points that are now
unsettled.

1 Dr. Erwin F. Smith, of the Bureau of Plant Industry, examined the material for crown-
gall and other disease-producing organisms. He found nothing to indicate that the
production of proliferous roots in the specimens which he examined was in any way
pathological.

Cab, AUalay]|

aa

a

A NEW SYSTEM OF COTTON CULTURE?

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

INTRODUCTION.

The way to secure an early short-season crop of cotton is to thin
the plants later and leave them closer together in the rows than is
now customary. Neither of these policies is advisable if used alone,
but they give a real advantage when properly combined. Keeping
the plants closer together during the early stages of growth restricts
the formation of vegetative branches and induces an earlier develop-
ment of fruiting branches. The new system is based on the princi-
ple of controlling the formation of the branches, as explained in
previous reports and publications of the Bureau of Plant Industry.?

CONTROLLING THE FORMATION OF BRANCHES.

The principle of branch control is more likely to be understood
if studied as the basis of a new cultural system. Application of
the principle will involve a reconsideration of all current opinions
regarding such questions as distances between rows, times of plant-
ing, methods of cultivation, and the values of different varieties.
The spacing of the plants and the stages at which thinning should
be done will depend upon the local conditions and the habits of the
varieties that are being grown, so that it will not be possible to give

1Tssued Mar. 1, 1913.

2 Cook, O. F. Weevil-resisting adaptations of the cotton plant. U. S. Department of
Agriculture, Bureau of Plant Industry, Bulletin 88, 87 p., 10 pl., 1906.

Cotton selection on the farm by the characters of the stalks, leaves, and

bolls. U. S. Department of Agriculture, Bureau of Plant Industry, Circular 66, 28 p.,
1910.

Dimorphic branches in tropical crop plants: Cotton, coffee, cacao, the Central
American rubber tree, and the banana. U. S. Department of Agriculture, Bureau of
Plant Industry, Bulletin 198, 64 p., 9 fig., 7 pl., 1911.

and Meade, R. M. Arrangement of parts in the cotton plant. U. 8S. Depart-
ment of Agriculture, Bureau of Plant Industry, Bulletin 222, 26 p., 9 fig., 1911.

Cotton improvement under weevil conditions. U. S. Department of Agricul-
ture, Farmers’ Bulletin 501, 22 p., 1912.

Results of cotton experiments in 1911. U. S. Department of Agriculture,
Bureau of Plant Industry, Circular 96, 21 p., 1912.

Morphology of cotton branches. U. S. Department of Agriculture, Bureau
of Plant Industry, Bulletin 256, 113 p., 19 fig., 6 pl., 1913.

Heredity and cotton breeding. U. S. Department of Agriculture, Bureau of
Plant Industry, Bulletin 256, 113 p., 19 fig., 6 pl., 1913.

[Cir. 115} weilley

16 CIRCULAR NO. 115, BUREAU OF PLANT INDUSTRY.

specific directions that can be used everywhere without discrimina-
tion. In agriculture, as in other arts, every new application of a
scientific principle makes an additional demand for intelligence and
insight into the problems of production. The contrasts with preva-
lent theories and practices of cotton culture are so great that careful
consideration of the habits of the cotton plant is needed before the
full possibilities of cultural improvement can be appreciated.

The first step toward permanent progress in the new direction
is to secure the attention of the intelligent farmer to the principle
itself, so that he can begin to observe and experiment on his own
account with rows of cotton thinned to different distances and at
different stages of growth and thus see for himself the relation
of the habits of the plants to the cultural problems. In this case
no special equipment of books or instruments is necessary to enable
the farmer to study the plant and learn what he needs to know
regarding its habits of branching. It is true that these habits are
somewhat peculiar from the botanical and biological standpoint, as
already pointed out in preceding papers on the subject, but there
are no technicalities that interfere in any way with direct observa-
tions of the behavior of the plants under the usual farm conditions.

APPLICATION OF IMPROVED METHODS.

Unless the farmer can understand the underlying reasons he is
not likely to adapt a new method or to apply it properly, any more
than he can use a new machine to advantage without knowing how it
works. This requirement of inteligence may hmit the application
of an improved method, just as it restricts the use of high-grade
machines to those who have the ability to handle them properly
and understand their construction. But it is generally agreed that
larger rewards for more intelligent and skillful farming are in the
interest of agricultural progress, and this is especially true in rela-
tion to the cotton industry. To make it seem worth while for intelli-
gent men to remain on the farm would soon counteract the urban
tendencies now so much deplored.

That cotton has been considered a “sure crop” even with the
most careless farming is one of the chief reasons for the backward
state of the industry. But the need of improvement is now recog-
nized as never before, as a result of the many changes that are being
enforced by the invasion of the boll weevil. With the continued
advance of the boll weevil the period of sure-crop cotton is drawing
to a close, and the rapid expansion of cotton culture in foreign coun-
tries shows that a new test of competition in the production of this
crop must be met in a few years. In the mean time any Improvement
that promises increased efficiency of production is worthy of careful

consideration.
(Cir. 115]

A NEW SYSTEM OF COTTON CULTURE. ALG
IMPORTANCE OF STIMULATING EARLINESS.

The chief advantage of the new system of cotton culture, based on
the improved method of thinning, is the increased earliness of the
crop; or, in other words, the production of more cotton in a shorter
period of time. The need of shortening the growing season of cot-
ton has been recognized as the best solution of the problem of secur-
ing protection against injury from the boll weevil, but is no less
important in regions where the crop is limited by drought or by
short seasons, as in the northern districts of the cotton belt.t

CONFLICTING OPINIONS ON SPACING COTTON ROWS.

Many intelligent farmers are aware of the fact that rows of cotton
accidentally left without thinning are sometimes much more productive
than rows that were thinned in the usual manner and have reflected
on the possibility of securing larger crops by closer planting, but the
underlying biological principle has not been understood. The be-
havior of cotton in different seasons or under different conditions
is so extremely variable that any intelligent farmer might well
hesitate to adopt a method of culture suggested by an occasional
occurrence like the production of a larger crop on an unthinned row.

In each cotton-growing community there are usually some farmers
who believe that cotton should be left closer together in the rows,
but the tendency in recent years has been toward wider spacing,
owing to a general recognition of the evil effects of having the
plants too close together, especially under conditions that favor
luxuriant growth. Those who use narrow spacing may boast of
phenomenal yields in some seasons, but in other years they appear
at a disadvantage with their neighbors. The possibility of making
a safe combination of the two conflicting methods seems not to have
been suggested. The same conflict is shown in the results of formal
experiments to determine the best planting distances as in the popu-
lar opinions on the subject. Wide spacing in the rows seemed better
in some cases and narrow spacing in others, so that no definite con-
clusions could be reached.’

LARGE PLANTS PRODUCE LATE CROPS.

When the habits of the cotton plant are taken into account it
becomes apparent that the theory of wider planting has its limita-
tions, as well as the theory of closer planting. To reduce the num-
ber of plants by wider spacing in the rows means that a !onger period

1Cook, O. F. Relation of drought to weevil resistance in cotton. U. S. Department of
Agriculture, Bureau of Plant Industry, Bulletin 220, 30 p., 1911.
2 Redding, R. J. Essential steps in securing an early crop of cotton. U.S. Department
of Agriculture, Farmers’ Bulletin 217, 16 p., 1905.
[Cir. 115]

18 CIRCULAR NO. 115, BUREAU OF PLANT INDUSTRY.

of time is required to produce a crop, for the reason that large luxuri-
ant plants do not begin to produce flowers and bolls as early as
plants of more restricted growth. This is not in accord with what
might be considered as the most logical view of the subject. Most
people are ready to argue that the plants making the most rapid
growth must produce the earliest and largest crop, but the actual
behavior of the cotton plant is otherwise. In such cases the bio-
iogical facts have to be taken into account instead of relying upon
the logical deductions.

The biological fact in the present case is that the large luxuriant
plants are later in setting and maturing a crop. This is because the
young plants in a condition of luxuriant growth develop vegetative
limbs at the expense of the lower fruiting branches that are neces-
sary to the production of an early crop. The cotton plant has two
different kinds of branches—vegetative branches, sometimes called
“wood limbs.” which correspond te the main stalk of the plant, and
fruiting branches, which produce the flowers and bolls.

When the habits of branching are understood it becomes apparent
that the idea of the largest plants producing the earliest and largest
crops does not apply to cotton. Spreading, treelike plants, with
numerous vegetative branches, do not represent a favorable condi-
tion for earliness or for large vields in short seasons. In the interest
of correct thinking on cultural problems the row rather than the
individual plant should be considered as the unit. The advantages
of the new method are gamed by improving the form of the rows.
More plants are left in the rows, and yet injurious crowding is
avoided. Plants that have numerous vegetative branches are more
crowded at 2 or 3 feet than plants with single stalks at 8 or 10 inches.
With the vegetative branches controlled, the spacing is no longer a
question of feet, but of inches. Rows spaced at 6 inches have usually
given better results than those at 12 inches or any greater distance.

EXPOSURE OF FRUITING BRANCHES TO LIGHT.

By avoiding the development of the large wood limbs the rows are
kept narrower and more hedgelike, so that the fruiting branches
receive sunlight throughout the season. This provides much more
favorable conditions for the ripening of the crop. When the vegeta-
tive branches are allowed to shut off the light by growing up be-
tween the rows. most of the bolls on the lower fruiting branches fail
to reach normal maturity. Fields of large luxuriant plants often
produce very small crops because only the upright growing ends of
the stalks and vegetative branches have access to the light. This
undesirable condition is avoided by restricting the development of
the vegetative branches in the earlier stages of growth.

[Cir. 115]

A NEW SYSTEM OF COTTON CULTURE. 19

Numerous experiments have demonstrated the fact that the usual
custom of giving the seedlings full exposure by thinning them to wide
distances in the rows is a means of inducing the development of large
numbers of vegetative limbs. Too much exposure for the young
plants results in too little exposure for the adults by increasing the
number of vegetative branches. The effect of exposure at wide dis-
tances is influenced, of course, by temperature and fertility of soil,
larger numbers of vegetative limbs being produced under conditions
that favor the luxuriant growth of the plants. But it does not ap-
pear that the production of vegetative branches is desirable under
any condition. The improved method of thinning restricts the
development of vegetative limbs or avoids their formation alto-
gether. This permits a better development of the fruiting branches
of the lower part of the main stalk. The plants are induced to
fruit earlier and the crop is made larger because more of the early
plants can be grown on the same area.

COMPETITION BETWEEN TWO KINDS OF BRANCHES.

The reason this possibility of cultural improvement has not re-
ceived adequate consideration in the past is doubtless to be found in
the fact that the distinctness of the two kinds of branches has not
been recognized, nor the relation of this specialization to the method
of thinning. When the plants are thinned too young, so as to stand
more than 6 inches apart, they put forth a full quota of vegetative
limbs, and the subsequent competition and crowding of these limbs
with each other and with the main stalks interfere with the devel-
opment of normal fruiting branches. As it is the low joints of the
stalk that produce the undesirable vegetative branches, the plants
must be allowed to grow beyond these joints before thinning. Ex-
posure of the stalk to the hght in the early stages of growth is one
of the factors that lead to the putting forth of the vegetative
branches.

EFFECTS OF EXTERNAL CONDITIONS ON BRANCH FORMATION.

The number of vegetative branches is also influenced by tempera-
ture and soil conditions. If the weather remains cool, or if the soil
is very dry, not many vegetative branches will develop, even when
the young plants are widely separated. But if the conditions favor
a luxuriant development of the young plants, early thinning will
result in the development of a large number of vegetative branches,
and the subsequent crowding will be great. Even in the absence of
any disease or insect pests the crop may be ruined by crowding alone.
Thus, the extent of the injury from crowding depends very largely
on the conditions that obtain during the early development of the

plant when the formation of vegetative branches is determined.
[Cir. 115]

20 CIRCULAR NO. 115, BUREAU OF PLANT INDUSTRY.

Until the habits of branching are taken into account it seems
impossible to explain the widely different results that are often
secured when the same experiments are repeated in different places
or in the same place in different seasons. From the present point
of view it is easy to understand that merely statistical experiments
made without recognizing the effects of different methods of thin-
ning upon the formation of branches would be likely to reach
only ambiguous results. The development of the branches, though
very easily influenced in the early stages of growth, completely alters
the subsequent behavior of the plants. The effect seems out of all
proportion to the exciting cause, hke touching off a charge of powder
or pulling the trigger of a gun.

Wider spacing appears as the only alternative as long as the young
plants are led to put forth a full equipment of vegetative limbs by
too much exposure in the early stages of growth. That the develop-
ment of these limbs may be avoided by a later and more gradual
thinning of the young plants must be recognized before it is possible
to understand the advantages of the new system. When good crops
are produced on rows that are not thinned at all, it is because the
plants remain so close together that no vegetative limbs are devel-
oped. The new system provides for a more regular and effective ap-
plication of the same principle of suppression of vegetable branches.

THINNING EXPERIMENTS WITH DURANGO COTTON.

The behavior of the Durango cotton at Norfolk, Va., in the season
of 1912 affords an excellent illustration of the application of the
principle of controlling the formation of branches as a means of
securing earlier and larger crops of cotton under short-season condi-
tions. Planted in a row test with other varieties and thinned in the
usual manner to the ordinary distances, the Durango cotton, which
is unusually productive for a long-staple variety, yielded at the rate
of about 1,175 pounds of seed cotton per acre, though considerably
exceeded by the Trice, an extra-early short-staple variety, which
produced at the rate of 1,756 pounds per acre. In a field planting
on lighter and less fertile soil the rows of the Durango cotton that
were thinned in the usual manner to ordinary distances yielded at
an average rate of 909 pounds of seed cotton per acre, while alter-
nate rows that were thinned late and left with the piants closer
together yielded at a rate of 1,391 pounds, or about 53 per cent
higher than the others. Table I shows some of the figures obtained
from the Norfolk experiment by Mr. G. S. Meloy.

[Cir. 115]

A NEW SYSTEM OF COTTON CULTURE. Dil

Taste I.—Results of a thinning experiment with Durango cotton at Norfolk,
Va., in 19712.

|
| Yield of seed Yield of seed
Average | cotton ee, Average ee nae
Num-| “ly ‘Gross eee Num- ay: Gross Tete,
Row | ber of now ha yield of Row | ber of wees yield of
‘ ‘ x ant 5 x +
No. |Pl@nts| based on eee COV Ordi= plants | ,ased on ae a Ordi-
oa counts on, a nary | Hedge- Bane counts aig nary | Hedge-
BOW foro grams. | wide- | formed TOM for 10 grams. | wide- | formed
days. spaced | rows. days. spaced | rows.
rows. rows.
i eae 176 Wo Oull Oni oe(llow asc aes ILS) Mees oe 60 68. 1 4,054 i) Ie oa sa ae
Oe ae ike 65 79.3 4, 601 i OUZS | eee ee ae 1b eee 159 75 BW Ey Iaoaqouse olen
Secu eee 203 Silas | 20923 0a lseeesee iS ifileel|| eli eereere 64 70.8 4, 282 942) Posse
Bice 66 79: 9e\ 45.81645)| e500. ck eon Te ee 166 6250 OK670 | aa 1, 249
Be cees 229 STS SAIC Oa7 rhe shoe see 14 OTe alates 61 50.8 | 3,659.5 805) Sone
Que: 61 68.6 | 3,384 AANA reese.

Costar 215 OB ATO O24. (5) |= ees 1, 435 AV CLAIR Chaya ps Hae eines game 909 1,391

SCR aos 63 75.9 | 4,191.5 O70 | ree Percentage of increase in |
Orie: 211 HOM BE GSC as ay eee ote 1, 495 hedge-formed rows......- Neeenete heyee 53

The rate of flowering, which also serves as a measure of earliness,
showed an advantage of 42 per cent in favor of the hedge-formed
rows. Though the proportion of 5-locked bolls was less in the
hedge-formed rows, a census on the basis of locks showed a difference
of 48 per cent in favor of the hedge-formed rows.

Many of the plants in the hedge-formed rows had no vegetative
branches—29 out of 52 in a series that was counted. An equal dis-
tance in the next wide-spaced row was occupied by 16 plants, all
but one of which produced vegetative branches, most of them three
or four. The average amount of vegetative branch development on
the widely spaced plants was 38.6 inches; on the hedgerow plants only
6.9 inches.

BRANCHING HABITS OF UPLAND AND EGYPTIAN COTTON.

The results obtained at Norfolk are in general agreement with
those obtained from other experiments with the Durango and other
varieties of Upland and Egyptian cotton in Texas and southern
California. The idea of controlling the formation of the branches
through cultural methods was first developed through a study of
the behavior of the Egyptian cotton in Arizona and southern Cali-
fornia. But it has now become evident that the principle has a
vastly greater practical importance in connection with Upland
cotton. Though the tendency to overdevelopment of vegetative
branches is not so strong in the Upland cotton as in the Egyptian,
the results are often worse, on account of the heavier foliage of the
Upland cotton and the greater tendency to shedding the buds and
young bolls.

3 [Cir. 115]

22 CIRCULAR NO. 115, BUREAU OF PLANT INDUSTRY.
NORTHWARD EXTENSION OF LONG-STAPLE COTTON.

The Norfolk experiment is significant, not only as an example of
the beneficial effects of controlling the formation of the vegetative
branches, but also as an indication of the possibility of extending the
cultivation of the Durango cotton and perhaps of other long-stapie
varieties along the northern border of the cotton belt and even beyond
the present limits of cotton cultivation. At any such rate of pro-
duction as that shown at Norfolk, the Durango cotton is a very
profitable crop, for the fiber attains a length of 14 inches under
favorable conditions, is of high quahty, and sells, when properly
handled, at a premium of 5 cents or more a pound above correspond-
ing grades of short-staple cotton. Such a premium means an addition
of about $25 per bale to the farmer’s profits. Though it has long
been known that fiber of good quality could be produced in the
northern rim of the cotton belt, the late-maturing habits of the older
long-staple varieties rendered the crop small and precarious. But
with the new varieties and new cultural methods that have now
become available the outlook is entirely changed.

CONCLUSIONS.

The new system of cotton culture is based on the application of a
principle not hitherto recognized in cultural experiments—the con-
trol of the vegetative branches by improved methods of thinning.
The formation of vegetative branches can be controlled by leaving the
plants closer together during the early stages, until the stalks have
grown beyond the stage where vegetative branches are produced.

The essential feature of the new system is later or more gradual
thinning. This makes it possible to leave more plants in the rows
than is now customary, and yet injurious crowding is avoided
through suppression of the vegetative branches.

The control or suppressicn of the vegetative branches also permits
an earlier development of fruiting branches and leads to the produc-
tion of an earlier crop. In regions where the pericd of crop produc-
tion is limited, either by short seasons or by the presence of the boll
weevil, increased earliness is a means of securing larger yields.

[Cir. 115]

GRASSES FOR CANAL BANKS IN WESTERN SOUTH
DAKOTA?

By ArtHUR C. DILLMAN,’ Assistant Physiologist, Office of Alkali and Drought
Resistant Plant Investigations.

INTRODUCTION.

In the construction of canals and ditches for the purpose of irri-
gation, the area of soil exposed and left bare of vegetation is very
large in the aggregate. Experiments have been conducted on the
canal banks of the Bellefourche Irrigation Project in western South
Dakota to determine what grass or combination of grasses will estab-
lish a satisfactory sod on such banks. This circular describes the
experimental work and gives such recommendations for seeding
canal banks on this project as are justified by the results thus far
obtained.

THE NEED OF A GRASS COVERING.

The advantages of a grass sod on the canal and ditch banks are,
chiefly, to prevent erosion of the soil, to prevent the growth and
spread of weeds, and to furnish pasture for live stock.

The erosion of the loose soil on the steep slopes of the ditch
banks is often serious. Most of the soils of this region are composed
of fine particles and therefore wash more readily than soils of a
coarser character. The amount of erosion on any short section of
ditch bank might not be considered serious, but the quantity of
soil washed down into the canals and ditches from over 300
miles of canal banks must be enormous.? Any covering of sod
which will prevent or lessen this erosion will mean a real economy
in the maintenance of the project.

One of the chief uses of a grass covering is to exclude the growth
of annual weeds which usually infest ditch banks. Unless a perma-

1JTssued Mar. 1, 1913.

2 The investigations here reported have been conducted by the writer, representing the
Office of Alkali and Drought Resistant Plant Investigations, in cooperation with Prof. C. V.
Piper, of the Office of Forage-Crop Investigations, who has furnished useful suggestions,
as well as most of the seed required.

3Mr. F. C. Magruder, project engineer, writes that there are 50 miles of main canal,

85 miles of main laterals, and 314 miles of small laterals on the Bellefourche Project.
These canal banks have a total area of about 1,200 acres.

ECir lst] 23

24 CIRCULAR NO. 115, BUREAU OF PLANT INDUSTRY.

nent sod is established, the banks become covered with Russian
thistles, buffalo bur (Solanum vostratum), wild oats, sweet clover,
and other troublesome weeds. The seeds of these weeds are freely
carried by the canal water and are scattered far and wide over the
fields. The resulting growth of weeds greatly increases the expense
and labor of cultivation. It is practically impossible to produce
pure seed crops of grain and alfalfa if weeds are allowed to multiply
along the banks of canals and ditches which cut through the fields.
The best check to the growth of weeds is a good grass sod.

Another advantage of the grass covering is to furnish pasturage
for live stock. The area occupied by ditch banks can not be util-
ized for the production of cultivated crops, but can be made to
vield some return when used for grazing. On many of the larger
farms the aggregate area that may be made use of in this way is
considerable. A half mile of canal bank of medium size has an
area of 1 to 2 acres. Moisture is usually abundant on the lower
slope of the banks, and a good growth of grass can be secured, which
will furnish a fair return for the trouble and expense of seeding.

! NATURE OF THE SOIL.

The soils of the Bellefourche Project are of two principal types.
In the northern part of the area the soils are mostly of the heavy
clay loam, known locally as “ gumbo.” These soils are formed by the
weathering of the Pierre shale which underlies the whole area. In
the southern part of the project the soils are somewhat lighter on
account of the silt and sand which they contain. The unweathered
shale which is turned up in many places in building the canal banks
is very unfavorable for the growth of grasses, but after a year or
two of weathering it is possible to get grass started even in the partly
decomposed shale.

OBSTACLES TO SECURING A STAND.

Alkali salts are present, although usually in small quantity, in
many of the soils of the project. Over the greater part of the area,
especially north of the Belle Fourche River, the alkali content ranges
from 0.2 to 0.6 of 1 per cent in the first 6 feet of soil. South of
the river there is generally less than 0.2 of 1 per cent. The alkali
is of a relatively harmless type, consisting mostly of sulphates
of calcium, magnesium, and sodium. Fortunately no black alkali
(sodium carbonate) is present. Alkali is especially hkely to accumu-
inate along ditch banks because of the high water content of the
soil and the high rate of evaporation from the joose surface. This
accumulation is especially noticeable in large fills, for in such places
some water often seeps through the canal banks. Wherever more

[Cir. 115]

GRASSES FOR CANAL BANKS IN WESTERN SOUTH DAKOTA. 25

than 1 per cent of alkali is present it will be difficult to start even
the most resistant grasses.

Drought is an even more sericus obstacle in starting grasses on
the ditch banks and is the general cause of failure to secure a good
sod. The upper part of the canal banks is often very dry, since the
steep slope is unfavorable to the penetration of rain water and since
the seepage or capillary water does not rise to this height. After
the more drought-resistant grasses are once well started they will be
able to survive any ordinary period of drought. On the smaller
ditch banks and on the lower slopes of the larger canal banks mois-
ture is generally so abundant that there should be no difficulty in
establishing a good growth of grass.

The growth of annual weeds, especially Russian thistle, hinders
the growth of grass during the first season. If, however, the canal
bank is harrowed at frequent intervals during one summer, espe-
cially after each rain, this treatment will destroy the growth of
weeds and conserve moisture, which will put the soil in good condi-
tion for seeding early the following spring.

Live stock should not be allowed to graze on the newly seeded
canal banks until the grass is well established. The trampling of
stock, cattle especially, kills out many of the young plants on the
steep slopes, and close grazing also is injurious to the new growth of
grass until a sod is obtained.

It is not probable that all of these adverse conditions—alkah,
drought, the competition of weeds, and close grazing—will be en-
countered in every part of the area, but a combination of grasses
should be planted which will endure any of these conditions as well
as possible. Some of the experiments with the different grasses de-
scribed in this paper were made under very adverse conditions, while
others were made under more favorable conditions.

OUTLINE OF THE EXPERIMENTS.

The first planting was made in 1907 on the south canal near the
storage dam. Three plantings were made in 1908: One on the sur-
face of the storage dam, one on the south canal near the present
town of Nisland, and one on the Indian Creek lateral on the farm of
Mr. George Maass. Duplicate plantings were made again in 1909
and 1910 on the farm of Mr. Maass. Each species was seeded on a
section of the canal bank about 20 feet wide, the plat extending
from the water line on the inner surface over the bank to the ground
level on the outside. The time of the different seedings ranged from
the first of April to the middle of June.

GRASSES AND OTHER PLANTS TESTED.

Several of the common hay and pasture grasses have been tested
for the purpose of soil binding. The list includes brome-grass, west-
| [Cir. 115]

26 CIRCULAR NO. 115, BUREAU OF PLANT INDUSTRY.

ern wheat-grass, slender wheat-grass, meadow fescue, tall fescue,
ved fescue, redtop, orchard grass, Canadian bluegrass, and alfaifa,
both singly and in various mixtures. A few species have given good —
results in pure stands, though the best results have come from certain
mixtures. The value of the different species is discussed in the fol-
lowing pages.?
BRO ME-GRASS.

Brome-grass (Bromus inermis) has been used with good success,
whether planted alone or in mixture with-other species. This grass
is well adapted to the climatic conditions of the Northwest, and on

Fic. i.—A canal bank, showing western wheat-grass in the foreground and brome-grass
and western wheat-grass in the middle of the picture.

account of its habits of growth it is well adapted to the particular
purpose of soil binding. The initial growth is fairly rapid, so that
the grass gets well established the first season. It endures drought
well and is quite resistant to alkali.

The growth of brome-grass is unhindered where the alkali is of the medium
grade (0.5 per cent). In fact, this grass can make excellent growth and seed
production where the alkali is medium strong (0.7 per cent), and it has been
observed to ripen some seed in the presence of very strong alkali (1.5 per cent) .*

1Jt should be understood, however, that the results here reperted apply only to the con-
ditions under which the experiments were made and do not necessarily indicate that the
same species would behave similarly in other parts of the country.
2 Kearney, T. H. The choice of crops for alkali land. U. S. Department of Agriculture,
Farmers’ Bulletin 446, p. 19, 1911.
Cre alatsy]

GRASSES FOR CANAL BANKS IN WESTERN SOUTH DAKOTA. 27

Brome-grass will not endure as severe drought as will western
wheat-grass, but 1s more drought resistant than any of the other
grasses tested. In certain plats on the ditch banks it maintained
a good stand during the extremely dry seasons of 1910 and 1911.
Tis habit of spreading by means of underground rootstocks enables
it to form a good sod and makes it one of the best grasses for soil
binding. (See fig. 1.)

WESTERN WHEAT-GRASS.

Western wheat-grass (Agropyron smithii) is native to western
South Dakota and is the most valuable pasture and hay grass of the
region. On the drv plains it forms, with buffalo grass, the chief
forage for the range stock, and in the creek valleys, where it 1s some-
times watered by overflow from the streams, it makes a thick growth
and is the chief native hay grass. It is extremely drought resistant
and yet will endure a great amount of flooding. It is the most re-
sistant to alkali of the grasses mentioned in this paper. In the
field experiments of Mr. T. H. Kearney? it proved to be the
most resistant to alkali of any of the grasses tested. On the
dry ditch banks it is slow in starting growth, but when once
established it makes a tough, permanent sod. The seed can not
be obtained commercially, but can usually be harvested in favor-
ably situated places along the creek bottoms. In 1909 the writer
thrashed 100 pounds of seed from a few bundles which were cut
in. an hour’s time with a common grain binder. The germi-
nation of the seed is often rather poor, so that a large quantity, about
30 or 40 pounds per acre, should be used. The chief objection to
western wheat-grass is its slow early growth which makes it long
in establishing itself. Its valuable characters—drought endurance,
alkali resistance, ability to endure flooding, and good sod-producing
habit—make western wheat-grass one of the best of soil binders.
In places where a native sod is cut through in making the ditches,
the surviving plants of western wheat-grass at the edge of the cut
will extend their growth up the canal bank, so that in a few years
a geod covering is naturally produced. This, however, will only
occur where the grass is already present at the edge of the bank
and not in cultivated fields where the native sod has been destroyed.

SLENDER WHEAT-GRASS.

Slender wheat-grass (Agropyron tenerum) makes a comparatively
rapid growth and is valuable to give quick results in a mixture. This
grass, unlike western wheat-grass, is a bunch grass and does not
form a continuous sod unless planted very thickly. It is not nearly
as drought resistant as either brome-grass or western wheat-grass.

B 1 Toc. cit.
[Cir. 115]

28 CIRCULAR NO, 115, BUREAU OF PLANT INDUSTRY.

Its chief value for the purpose here desired is its early, vigorous
growth. In a pure stand, however, it has maintained a fair growth
since it was planted in 1909.

ALFALFA.

Alfalfa has proved very satisfactory as a covering for canal banks,
especially when planted in mixtures with some species of grass. The
growth during the first season is fairly rapid, the roots strike deep
and soon reach the seepage water present in the lower depths of the
soil, and the forage produced is of first-rate value. Alfalfa does not

Bic. 2.—Alfalfa on a canal bank, with western wheat-grass crowding in near the water’s
edge.

stand close grazing and excessive trampling so well as do some of the
grasses, but when planted with a vigorous grass, such as brome-
grass, it makes a good covering. Pocket gophers have given some
trouble by eating off the roots of alfalfa plants, but this is not likely
to prove a serious drawback to the use of alfalfa on the ditch banks.
Where seed of a hardy strain can be obtained, alfalfa may be recom-
mended for use in a mixture with grasses (fig. 2).

REDTOP.

Redtop (Agrostis alba) makes a good growth where an abundance
of moisture is available, but is not at all drought resistant. It may
[Cir. 115]

GRASSES FOR CANAL BANKS IN WESTERN SOUTH DAKOTA. De)

be seeded near the water’s edge and where seepage water occurs, or, if
included in a mixture with other grasses, it will make a good growth
on all moist parts of the bank.

THE FESCUES.

The three species of Festuca tested, meadow fescue, tall fescue, and
red fescue, have not shown any particular value as soil binders in
this region. None of them are able toendure the dry conditions of
the higher part of the banks, nor do they make a permanent sod,
even under more favorable conditions.

OTHER GRASSES.

The other grasses tested, namely, orchard grass, Italian rye-grass,
creeping bent-grass, and Canadian bluegrass, have not shown any
value for soil binding in this region. They are unable to endure
the trying conditions, nor can they compete with the more resistant
grasses when used in mixtures.

MIXTURES TESTED.

Several mixtures of grasses have been tested to determine if any
combination could be found which is better adapted to the purpose
than a single species.

The following mixtures (by weight) have been used:

» Mixture No. 1: Mixture, No. 3:
Brome-grass _____- 4 parts. Western wheat-
Western wheat- PRASS ae ee ee 4 parts.
ER SESS Hes i pee 4 parts. Red fescue________ 2 parts.
Meadow fescue___~- 1 part. ~ Al Talia ce es ste i part.
Italian rye-grass___ 1 part. Italian rye-grass___ 1 part.
Mixture No. 2: Mixture No. 4:
Brome-grass _____-~ 4 parts. Slender wheat-
Canadian bluegrass_ 2 parts. SLASS Ee ee ees 2 parts.
Italian rye-grass___ 1 part. Mallee SCuee i ea ds 2 parts.
White clover ______ 1 part: Redtopew ae 2s 1 part.
Orchard grass_____ 2 parts.

Mixture No. 1 combines two hardy and permanent but rather slow-
growing grasses with others which quickly form a cover. The brome-
grass and western wheat-grass have survived since planting in 1909
and now form a good cover. The other species made a fair growth
during the first season and may be desirable to use in a lawn where
quick results are desired, but they have no permanent place on a
canal bank.

In mixture No. 2 brome-grass is the only species that has survived.
The Canadian bluegrass did not do well at any time in this mixture.

[Cir. 115]

30 CIRCULAR NO. 115, BUREAU OF PLANT INDUSTRY.

As it was not used in a pure seeding, it is uncertain whether its
failure was due to the nature of the soil or to the competition with
the more vigorously growing brome-grass.

In mixture No. 3 western wheat-grass and alfalfa have done well
and show a good covering three years after planting. No doubt
brome-grass would do as well as western wheat-grass in mixtures
with alfalfa under most conditions; and it can be used more exten-
sively, since the seed is more easily obtained.

Mixture No. + made a good showing during the first and second
years, but has not proved permanent. The redtop, however, still
survives hear the water’s edge.

PREPARATION OF SOIL AND SEEDING.

In order to insure a successful growth of grass it is necessary to
have the soil in as favorable a condition as possible. There must be
a well-prepared seed bed—the soil should be fine, firm, and moist—
in order to insure a quick and continuous growth of the grass. A
good preparation is to harrow the canal bank at frequent intervals
during one summer, so as to prevent the growth of weeds and con-
serve as much moisture ‘as possible. This will put the soil in good
condition for seeding early the following spring.

The seed may be sown broadcast by hand, or, in case a large area
is to be seeded, a hand seeder may be used to advantage. The soil
should be harrowed before seeding and the seed covered by harrow-
ing again after seeding. In covering the seed, the harrow teeth
should be set at a shght angle, so as not to dig too deeply. A com-
mon 2-horse steel harrow is convenient for this “work

From 15 to 30 pounds of seed per acre, depending upon the kind
of seed and the condition of the seed bed, should be enough to secure
a good stand of grass. Any heavier rate of seeding will hardly com-
pensate for poor soil conditions. The smaller seeds, such as redtop,
need not be sown as heavily as the larger seeded grasses. A heavy
erowth of grass can not be expected on the drier parts of the canal
banks. and yet on account of the less favorable conditions there a
liberal quantity of seed should be used.

It is very important that the planting should be done early, in
order .to give the young grass the benefit of the cooler growing
weather and the early summer rains. Competition from weeds,
especially Russian thistle, is also less during the early summer. The
best results have been obtained by seeding in April rather than later.
Live stock should not be allowed on the banks after seeding until a
sod is well established. Even then, close grazing should not be al-
lowed if a permanent sod is to be maintained.

[Cir. 115]

GRASSES FOR CANAL BANKS IN WESTERN SOUTH DAKOTA. 381
RECOMMENDATIONS.

From the results thus far obtained it seems safe to recommend
brome-grass, western wheat-grass, and alfalfa as the best plants for
covering the canal banks of the Bellefourche Project. Redtop may
be seeded in a mixture with these or seeded alone where an abun-
dance of moisture is available, as is the case near the water line on
the inner slope of the canals. Western wheat-grass should not he
seeded alone, on account of its slow early growth, but if seed can be
obtained it should be included in a mixture with brome-grass or
alfalfa. A mixture which can be recommended for all types of soil
on the project is the following:

Brome-grass iS) aap lap a 7 EO 6 pounds.
Western wheat-grass____ =e ie = Teteed _ 6 pounds.
PMeleteeaietreqae sees SS ae Tae EEE PIC R i ag ne hls CE SOUL CS:
edto pes ee a sida lo ner eae agate 2 pounds.

This should be enough seed, under ordinary conditions, for plant-
ing 1 acre. If all of these species are not available a mixture of
equal parts of brome-grass and alfalfa will in most cases give a
satisfactory covering. Care should be used to secure a hardy strain
of alfalfa seed, preferably one that has been grown in the neighbor-
hood for a number of years.

The seeding should be done early in the spring, during a rainy
period if possible, and care should be taken to obtain a firm, moist
seed bed, in order to insure rapid early growth.

Grazing on the banks should not be allowed for a vear or two after
seeding, and close grazing should never be permitted if a good sod
is to be maintained.

[Cir. 115]

ONE COPIES of this publication
may be procured from the SUPERINTEND-
ENT OF DOCUMENTS, Government Printing
Office, Washington, D. C., at 5 cents per copy