Go-

Circular No. I. X^ O > ^O -^ ' \

United States Department of^ncnU^iiie^

DIVISION OF AGROSTOLOGY.

A NOTE ON EXPERIMENTAL GRASS (URDENS.

There are some seven hundred distinct kinds of grass native to the
United States. This is 20 i)er cent of the whole number known
throughout the world. North America, from the southern boundary of
Mexico to the Arctic Ocean, has about one-third the total number of
known species. The grasses are the most important family of plants,
since they provide food for man and forage for cattle. Wheat, corn, oats,
rice, sugar cane, sorghum, and tiie bamboo which represents house and
furniture to so many millions of people in oriental countries — these
and many more are grasses. All of our grains the world over are grass
products. Besides what we ourselves eat, most of the fodder and
pasturage of our flocks and herds, and the pasturage for the wild game
is provided by the many species of our prairies, valleys, and moimtains.

Next in importance to the grass family is the bean family. All of
our cultivated clovers and their relatives — red clover, alfalfa, cowpeas,
and the multitude of less known forage plants — belong to the l)ean
family. There are about eight hundred members of thft bean family in
the United States, not counting the trees. These seven hundred
grasses and eight hundred legumes together form one of the richest
legacies of our country.

The amount of money invested in the cattle industry is reckoned
by the hundred million dollars, and every dollar of that value is
absolutely dependent ui)on the question of forage. Strip ^ our broad
acres of their grasses and clovers, and instead of receiving millions of
dollars for the meat products sent to foreign lands, we would have to
pay out money for them. The cattle industry and the wool industry
are dependent upon the question of grass.

Twenty years ago the whole prairie region west of the Missouri was
given over to great herds of cattle, liut the days of the cattle kings
are past, and the lands that were then cattle ranges given up to the sup-
port of a few head to the square mile, are now divided up into farms.
The native Western grasses are being rapidly driven out to make way
for the worthless weeds that civilization and scanty cultivation bring
with them. Already the buffalo grass and the mes(iuite have dis-
appeared from a large section of Kansas and Nebraska. Acres that
were once covered with these most nutritious species are now occupied
by weedy kinds, and their value as ]iasture and hay lands is constantly
diminishing Similar destruction of grasses has followed the cultiva-
tion of cotton in the South, and of tobacco and hoed crops generally in

2

all sections of the country. The American has been a grass-killer
everywhere.

Nearly all of our cultivated grasses and clovers are of foreign origin.
This is true of red clover, cowpeas, alfalfa, and scarlet clover — the
principal leguminous forage plants — and of orchard grass, red-top,
timothy, and the rye grasses. Kentucky blue-grass was cultivated in
Europe before it was in this country. Almost all of the grasses and
clovers in common cultivation came from England or Germany or
France or China, or from some place outside our borders. This is not
because we do not have just as good or better sorts. The English or
the German farmer has to earn a living from land that has been in
cultivation for hundreds of years, and whatever he takes out of his soil
must first l)e put into it. He is forced to adopt intensive rather than
extensive farming. His farm is small. His soil is poor in plant food,
though perhaps in the best physical condition ; loose, friable, and well
stirred by constant cultivation. This European farmer, with his ex-
hausted soil, must compete with the American farmer whose acres are
rich with plant food. To do this successfully, the European farmer
is prone to try anything and everything that gives promise of pro-
ducing the proverbial "two blades of grass where one grew before."
He tries the native grasses and clovers that he sees his cattle pick
from the hedge rows, to find out what effect cultivation will have
upon them ; whether they will, if planted in better soil, increase in
size and add some value to" the scant pastures and hay lands. Every
weed of the fence corners that upon chemical analysis seems adaptable
to cultivation for fodder is cultivated and perfected and improved until
it is settled beyond doubt whether it is worthy or worthless. There is
no European country whose natural resources, counted in the nurnber
of species of nutritious forage plants, equal ours. There are no finer
natural meadows and pastures in the world than are to be found in all
sections of our land. And yet these natural meadows, which form part
of our most valuable resources, are being broken up and destroyed.
Our native grasses and clovers are being driven out by foreign species,
not through any superiority of digestibility or chemical composition,
but because we have not found time to discover whether our own
species are not far better suited for our own soil and climatic conditions.
Enough money has been spent by American farmers for worthless
fodder plants from foreign sources to more than pay the expenses of the
whole Dei)artment of Agriculture for one year, with many useful lines
of investigation.

All of the cultivated grasses and clovers of which we know the his-
tory have been cultivated in a small way at first. They are mostly
species which have a large or conspicuous seed head. These were
selected and cultivated mainly because their seeds were large enough to
attract attention or because the seed could be easily shelled. Many of
the grasses sown in foreign meadows have no value with us, and yet
they are persistently advertised and offered for sale here because they are
valuable in Germany or in England, on the principle that all things
good in agriculture come from foreign lands. We have better species
and more of them right here at home, that we will not have to plant
ten or twenty years to get acclimated, or to learn whether they will
grow in our soils. There are sixty native species of trifolium, the
genus to which red clover belongs. There are sixty-five poas, all relatives
of Kentucky blue-grass. There are twenty-five grasses closely related

/

to the buffalo grass and the grama. There are ninety native lujjins,
thirty prairie clovers, forty-five vetches, forty relatives of the beggar
weed of Florida, and brome grasses, meadow grasses, pasture grasses,
and hay grasses almost numberless, suitable to every kind of soil and
climate and rock formation which we possess. Of all this wealth
of species not a dozen have been brought into cultivation.

The grasses and clovers and native forage plants of all descri{)tions
should be thoroughly tried on the different soils and in the dillercnt
sections of the United States. We want to know what plant will pro-
vide the greatest amount of the most nutritious forage in the shortest
season at the least expense to the farmer. We want to know what
forage plants remove the greatest amounts of mineral fertilizers frf)m the
soil and what improve the fields that they are i)lanted upon. In
short, we want the best, and we believe the best can be grown on
American soil from native species.

There are no finer natural meadows and pastures anywhere in the
world than in the Appalachian Mountains, and on the Western prai-
ries from the Saskatchewan to the Gulf, and in the mountain valleys of
the Western and Pacific States. Take, for an example, the State of Col-
orado as an illustration of the value of forage crops. The value of her
herds and the products of the pastures and meadows which sustain
them is equal to the output of her silver mines. Ten years ago, in
1885, there Avere nine thousand tons of alfalfa seed alone sown in Colo-
rado, and the acreage and the value of the product has increased every
year since. It is the same in the Western States and in the Southern
States and throughout the length and breadth of our land. Every year
more and more attention is being given to the raising of improved hay
and forage plants. The raising of cattle is one of the most profitable in-
dustries in which the farmer and the man of small as well as large means
can engage. There are larger returns for a farmer's labor if he feeds his
crops on the farm and sells beef, pork, and mutton, than if he sells his
hay and grain to the larger cattle feeders. Then, too, if all of the hay
and all of the grain is fed on the farm, the soil is being constantly en-
riched by the return of the manurial portions of his crops to it. The
cattle industry of the United States, instead of being on the wane, is be-
coming greater every year. The question of cultivating hay and
pasture grasses to feed our growing herds and flocks is one of vital im-
portance.

There are natural meadows and pastures in many of the newer por-
tions of the country as good as the best grown in the older-settled
Eastern States. The native grass and clover species which go to make
up these meadows should be tested under the conditions of richer and
better prepared soils.

Congress at its last session established a Division of Agrostology for
the investigation of the native and foreign fodder plants. Two grass
gardens have been started in connection with the work of the new divi-
sion. The one on the grounds of the Department in Washington
already contains four hundred sjjccies, native and foreign. This
occui)ies the strip of ground between the Seed Building and the
Agricultural Museum. It is about oOO by 60 feet. On account of the
limited space in this garden, each variety can be represented by only a
small number of plants. Special attention will be given to the culti-
vation of wild species which are known to bo of some value, and which
might be improved if planted on better tilled and richer soil. There

are now some Uvo hundred or more European, Australian, and American
species. Experiments will be made in the old Enjilish method of propa-
gating pasture grasses by inoculating land with ])ieces of turf; but the
greatest dependence must be placed upon seed, on account of the ex-
pense of the former method. The method of inoculating land, as it
is called, is probably the best one for establishing suburban and city
lawns, for this is the surest way of obtaining pure cultures, and the
value and beauty of the lawn depends as much upon the grasses of
which it is composed as Upon the care. The best lawns, and the most
pleasing to the eye, are those composed of a single variety of grass. For
establishing pastures or hay meadows inoculation is impracticable, be-
cause of the time and lal)or required.

Experiments on a larger scale have been begun at Knoxville, Tenn.,
Avhere about 8i acres have l)een secured on the grounds of the Agricul-
tural Experiment Station. About two hundred grasses and clovers are
growing there, mostly of those sorts which are sold by seedsmen, and
such as are in general cultivation in various parts of the United States.

The plats in this garden vary in size from a yard square to half an
acre. The land is rich, is in excellent condition, and may be taken as
fairly representative of a large class of farming lands in the South.
Special attention will be paid to the improvement of native forage
plants. In this garden the most promising forage plants for the South-
ern and Southwestern States will be tried on a large scale, much as
they would be if planted by a farmer or stockman. There are laree
silos on the farm, and the fodder will be tested as to its feeding value
in the form of hay and as ensilage, and also for soiling. The question
of grasses and clovers and other ibrage plants suitable to the Southern
States will be investigated from as practical a standpoint as possible.
Nearly all of the forage plants advertised or recommended by seedsmen
as good for the Southern States have been in cultivation there the pres-
ent summer, and the work will be continued through a series of years.
This kind of work can not be done successfully by the farmer or l)y
private means, because decisive results can not be obtained in less than
a term of years, and the work can better be carried on l)y those who are
trained for it and who have made this branch of agriculture a lifelong
study.

In this connection it might be well to point out the advantages to be
derived from such work. If one grass or one clover can be so im-
proved that it will thrive on the arid lands of the West, and will grow
as well as the sagebrush and creosote bush, and at the same time be of
some value as forage, many times the cost of such experiments will be
added each year to the material wealth of the nation. If a good forage
plant can be found that can be substituted for a poor one on some
sterile soil where it was thought good grasses would not grow, it means
that two hundred head of cattle can be pastured on land Avhere only
one hundred were fattened before. There is no better or more profita-
ble line of work open to the Department of Agriculture than the devel-
opment of our resources as embodied in the richest grass flora of the
world.

Approved : Jared G. Smith,

J. Sterling Morton, Acting Agros,

Secretary.

Washington, D. C, August 6, 1895.

Circular No. 2.

^ NEW YORK
BOTANICAL

United States Department of Agriculture,

DIVISION OF AGROSTOLOGY.

HAIRY VETCH, SAND VETCH, OR RUSSIAN VET( H.

{Vicia villosa.)

This annual leguminous plant is a native of western Asia. It has
been cultivated for about fifty years in some parts of Europe, especially
southern Russia, Germany, and France, and was introduced into this
country for the first time about 1847, under the name of Siberian

vetch. But its cultivation
was lost sight oi', and no
thought was given to it
until reintroduced about
ten years ago. It has been
tried in various parts of the
United States. Excellent
reports as to its drought-
resisting qualities and its
adai)tability to our climate
have been received from
Washington, Nebraska, and
Pennsylvania. It has been
grown in the past season on
the experimental grounds of
the Department of Agricul-
ture at Washington, and
has proved to be thoroughly
adapted to, and valuable
for, this region. The seed
was planted about the 25th
of April ; the i)lant com-
menced to bloom the middle
of July, and continued in
bloom until the end of

Hairy vetch {vicia villosa).

August. Hairy vetch withstands cold, heat, and drought, but it does
not do well where there is an excess of water in the soil. It is one of
the most promising fodder crops which has been brought into the
United States in recent years.

Hairy vetches may be planted in autumn, from about the middle of
August to the middle of September, or in spring, from the latter part of
April to the middle of May. Sow broadcast at the rate of a bushel and
a half of seed per acre, or plant in drills 2 to 4 feet apart. The latter
method will require a less amount of seed. The seed is as yet very
expensive in this country, about $6 per bushel of 60 pounds. When
the seed is put in broadcast, a bushel of rye, oats, or wheat, should be
sown at the same time, so as to furnish a support for the vetches, and
keep the vines up off the ground. If it is sown in drills in the latter
part of August, it should be cultivated several times between the rows.
It will furnish some forage in autumn, and where the winter is not too
severe, will start to grow again in the spring, thus producing forage in
late autumn and early sjDring, at the two periods when it is most
needed.

Prof. S. M. Tracy in " Forage Plants for the South " (Farmers'
Bulletin No. 18 of this Department) states that the seed of hairy vetch
was sown at the Mississippi Agricultural Experiment Station in October,
1888, and since that time has given heavy annual crops on the same
ground, although it has received no attention and the ground has not
been plowed since the first sowing. Its seeds germinate with the first
autumn rains and in favorable seasons cover the ground by the first of
January, and then furnish good grazing until April or May. If the
stock is taken off the field in March, the plants will mature and reseed
the ground freely for the next year.

The great desideratum in our American farming is some crop or
crops, either clovers, grasses, or other forage plants, Avhich may be
sown after the grain crop of the year is harvested, to provide green
forage and pasturage and to prevent the washing of the soils in winter
and early spring. If the land is allowed to lie bare, large amounts of
the soluble, and hence the most valuable, mineral fertilizers are washed
into the creeks and rivers. The agricultural soils in the newly settled
portions of the continent have been cultivated as though their riches
were inexhaustible. Crop after crop of grain, cotton, corn, or tobacco
is taken from the field, and no effort is made to replace the potash,
phosphoric acid, and nitrogen which they contain. The result of this
robbing of the soil is that in 15 or 20 years the yield has very
materially diminished. The loss in fertility which these lands sustain
each year is about equally divided between that portion sold from the
farm in the form of grain or other raw material and that which leaches
out of the soil through washing after the crop has been removed from
the field and the surface is no longer protected from the elements-
Now, if some crop can be planted which will cover the ground in
autumn and winter and early spring, and which will be ready to take
off the ground in time for spring planting, at least half of this loss of

fertilizing material may be saved. Furthermore, if such a winter crop
is planted, the land instead of losing organic and inorganic food ele-
ments is constantly acquiring them.

Analyses of hairy vetch made by Coudon in 1890 showed that this
forage plant contains for every 100 parts of dry matter 22.78 per cent
nitrogenous matter, or protein ; 2.61 to 3 per cent of fats ; 23.25 of
cellulose, or crude fiber; and 39 per cent of nitrogen-free extract.
Every ton of dry hay contains 45 pounds of nitrogen, 19 pounds of
phosphoric acid, and 62 pounds of potash. The yield amounts to from
6 to 10, and sometimes on very rich soils 12 to 15, tons of green forage
per acre. The nitrogen, potash, and phosphoric acid contained in a
ton of vetch hay, if bought in the form of a commercial fertilizer at
current prices would be worth about $11. The yield of dry hay taken
on this basis would amount to from 1^ to 4 tons per acre, the yield
depending much upon the fertihty of the soil and the state of cultiva-
tion. These fertilizers, as produced by the vetch, are in the best form
or combination to be used as food by the succeeding crop. A crop of
hairy vetch plowed under about the first of May, would therefore place
in the ground fertilizers that would cost from $16 to $45 per acre if
purchased in the form of commercial fertilizers. The hairy vetch has
as high a nutritive ratio as any forage plant, that has been analyzed,
excepting the soja bean.

In proportion to its cellulose and nitrogen-free extract, which repre-
sent the contained starches, sugars, and like compounds, it has a very
high per cent of protein and fat. To realize its fullest value as a soil-
ing crop it should be fed mixed with corn fodder or timothy hay or
roots, which are all rich in the carbohydrates but are deficient in
protein. The two classes of foods need to be fed together to prevent
loss of protein, which is that part of the food that becomes transformed
into blood, bone, and muscle. The carbohydrates when digested go to
produce heat and motion, and the surplus is stored up as animal fat.

Hairy vetch is eaten with relish by all kinds of farm animals. If
properly cured it makes very fine hay, though on accomit of its habit
of growth it is very difficult to cure. It has been tested in the silo in
alternate layers with green corn and alone. The former method is the
one to be used if the best ensilage is desired. It is a most excellent
forage plant for soiling purposes. On account of tbe difficulty of
curing it properly, it will give the most satisfactory results if fed green
or ensilaged.

Hairy vetch, while it gives a fair crop on sterile soil, is most profit-
able as a forage plant on rich and well-tilled land. It needs consider-
able moisture during the first six weeks of its growth, but when once
fairly established withstands drought and extremes of temperature.

SUMMARY.

Hairy vetch seed weighs 60 poirnds per bushel. Spring prices, in
New York, 1895, were $6 to $6.50 per bushel.

Sow broadcast or in drills, at the rate of 1^ bushels to the acre from
August 15 to September 15. Sow one-half bushel of winter wheat,
winter rye, or winter oats with the vetches.

Hairy vetches sown in August may be fed in October or November.
They will be ready to cut for soiling or hay from the 20th of April to
the 1st of May.

For a summer crop, sow from the 20th of April to the middle of
May. Cut when in full bloom.

Do not plant on land that is poorly drained.

To get the best results, feed with coarse fodder, hay, or root crops.
Feed with corn fodder at the rate of four tons of corn fodder to one ton
of hairy vetch.

Hairy vetch sown in autumn will cover the ground and prevent
washing during the winter.

It is one of the best crops to turn under as green manure.

Do not commence to feed hairy vetches until they have begun to
blossom. Like most of the bean and clover family they are somewhat
diuretic if feed in large quantities before mature. Use caution in feed-
ing until the animals have become accustomed to the change of food.

Approved: F. Lamson-Scribner,

Chas. W. Dabney, Jr., Agrostologist.

Acting Secretary.
Washington, D. C, October 1, 1895.

_ r

ClBCULAR No. S.

United States Department of Agriculture,

DIVISION OF AGROSTOLOGY.

SALTBUSHES.

AUSTRALIAN SALTBUSHES.

A considerable demand has arisen in the grazing regions of the West
during the last decade for seeds of the various species of Australian
ealtbush. These saltbushes are nearly all plants of the salsolaceous,
or pigweed, family, which botanists for purposes of exact knowledge
call the Chenopodiacese. There are over one hundred species of this
family in Aus-
tralia, and the
ones which
have come into
prominence as
desirable forage
plants grow
wild upon the
low plains and
in the broad
river valleys of
the interior of
that continent.
They are plants
which are spe-
cially adapted
by nature for
growth in arid
and semi arid

regions where the annual rainfall is small and torrential, and where
there are long periods of months and perhaps years when no rain Jails.
Many of them grow on alkaline soil and along the margins of brackish
or stagnant pools, and thrive where grasses and other more nutritious
but less hardy forage plants will not grow. They are annuals, with
small fleshy leaves and rigid harsh stems, and are protected by various

Fig. 1.— Australian saltbush (Atriplex semibaccata).

natural devices from the drying influence of the fierce summer heat.
The leaves and stems are covered with a coating of hairs, or are mealy
like the leaves of the common pigweed of cultivated lands, or the leaves

and stems are mucilaginous like
those of cactus. In common with
all other plants native of alkaline or
salty soils, they contain considera-
ble quantities of salt and other
soluble constituents of the soil. All
of these characteristics tend to pro-
tect the foliage and prevent the
evaporation from the leaves of the
moisture which is necessary to the
continued existence of the plant.
Again in common with plants
indigenous to semidesert regions,
each individual produces an im-
mense quantity of seed to insure
reproduction of the species under
the very unfavorable conditions
which prevail.

With the possible exception of
portions of California and Arizona,
the range of country in which the
saltbushes grow wild is relatively
much warmer than the grazing
regions of the United States. There
is never any snow and rarely frost,
the summer heat is more intense,

FIG. 2.-Australiau saltbush (Atriplex ^^^ ^^^ periods of drought longer.

leptocarpa).

AMERICAN SALTBUSHES.

The most valuable saltbushes are species of Chenopodium and Atriplex.
They are not cultivated plants and are not any better adapted to the
grazing regions of the United States than our own wild si3ecies of the
same genera. There are over thirty different kinds of American salt-
bush, only they are not so designated, in the region extending from
Montana, Colorado, and western Texas to the Pacific coast. Nearly all
of these are recognized by herders and graziers as furnishing consider-
able forage for sheep and cattle, and many of them have acquired local
importance under such names as sweet sage, white sage, or winter fat.
In dry seasons and during severe winters they are supplementary to
the native grasses. Their abundance adds value to the ranges.

NOT A CULTIVATED FORAGE PLANT.

These saltbushes are not cultivated plants. In many portions of
Australia they have become well nigh extinct through the same causes
that are leading to the extermination of our own native forage plants
and grasses — the overpasturing of the ranges and the consequent
destruction by too close feeding and by trampling. The introduction
into foreign countries of saltbush seed has been accomplished mainly
through the instrumentality of public-spirited scientists, notably through

the efforts of Baron Ferdinand von Mueller of Melbourne. Some ten
or more years ago, large quantities of seed were collected and sent by
him to South Africa, and the world now beholds the anomalous state
of affairs of Australian farmers buying Australian saltbush seed from
South Africa.

GENERAL CONSIDERATIONS.

Many, if not all, of the American Atriplex and Chenojoodiiim species, espe-
cially those that are natives of the great ranges, will become extinct along
with some of our best pasture grasses, if an effort is not made to conserve
and disseminate them. We know that they are valuable. We know that
they are eaten by all kinds of stock.
We know that they are thoroughly
acclimated. They grow on the
alkali plains, along the margins
of brackish ponds, and on sterile
lands. They are perfectly adapted
to situations where better forage
plants and grasses will not grow.
They are natives of the vast arid and
semiarid uplands, which through
absence of water will never be used
for anything but cattle and sheep
ranges. They are better fitted to
soil and climatic conditions as they
now exist with us, and they are
less liable to become weeds in culti-
vated lands. Saltbushes are valu-
able only as supplementary forage
plants. They will not take the
place of grasses for continuous
pasturage. When mixed with
native grasses of a range, they act
as a bitter tonic to increase the
appetite- and improve the general
condition of the grazing animals.
But there is sufficient nourishment
in them to sustain life when through
drought or other causes there is no
grass.

The preservation of our native
The introduction of foreign ones should be secondary. There is al)un-
dant proof for the statement that the plants native to any soil or any
climate are better suited and better adapted to that particular location
than plants from another climate or that grew on anotlier kind of soil.
The American species of sweet sage and winter fat are thoroughly accli-
mated, and the seed can be readily and cheaply procured. We do not
need to send to Australia or to South Africa or to Argentina to find
forage plants that will grow on alkaline soil.

The indiscriminate introduction of new plants into any country is
always fraught with danger, especially when such plants are of a known
weedy character. The Russian thistle belongs to the same plant family
as the Australian saltbush. This- weed in its early stage furnishes

Fig

3.—" Winter fat," or sweet sage. (Buro-
tia lanata.) Abundant from Montana to
Arizona.

forage plants is of first importance.

good forage for sheep and cattle, and while it may be a blessing on a
sheep ranch, it is an undisguised curse on the prairie hay lands and in
cultivated fields. Considerable caution should therefore be exercised
with these foreign saltbushes lest they spread to fields and places

where they are not wanted.

Jared G. Smith,
Approved : Assistant Agrostologist.

J. Sterling Morton,

Secretary.

Washington, D. C, March IS, 1896.

/

(JIKCILAH No. !.

United States Department of Agriculture,

DIVISION OF AGROSTOLOGY.

[Grass and Forage Plant Investigations.]

THE RENEWIXU OF WORN OUT NATIVE PRAIRIE PASTURES.

Throiighotit the prairie regions of the West one frequently sees native pastures
nearly flpvoid of gras« av.d ofter; j^rovrr. tip tc vreedb of va'ioii:; kintl^. Plainly
such pastures are of little value to the owners, so far as the forage obtained from
them is concerned, and the weeds are a constant eyesore, often rendering the
land even worse than worthless.

Such a condition of things is usually due
to two causes, drought and ovei'stocking.
While the farmer may have no control
over the drought itself, he can. by a little
care and foresight, put the pasture in con-
dition to withstand it in a great measure,
and he certainly can prevent the pastiire
from being overstocked. Not infrequently
the use of the pasture is almost entirely
lost for one or two seasons, when a little
rest by removing the stock for a time or
by feeding green corn, sorghum, or other
soiling crops would have kept it in good
condition.

The native grasses are hardy and are
adapted to the natural conditions which
prevail on the prairies. Some species
stand grazing much better than others,
and after a pasture has been used for sev-
eral years it will be found that the weaker
grasses are giving way to the stronger
ones.

As a rule the forage obtained from the
average prairie pasture is furnished by a
comparatively small number of species.
Tr the more thickly settled portions of the
great prairie States big blue-stem, bushy
blue-stem, western wheat-grass, switch
grass, prairie June-grass, wild rye, blue
joint, and the various species of SHpa and
Bouteloua furnish most of native pastur-
age.

WEEDS OF THE NATIVE PASTURES. Kio. 1.— Western wheat-grass {Agropyron

spicatum).

The most troublesome weeds are either
annuals or perennials. The former, because of their vigorous and rapid growth,
spring up and take possession of a pasture in a very short time. The latter spread
more slowly, but are more difficult to eradicate. Left to themselves, the native
grasses will hold their own against the weeds, but when they are pastured off and
trampled upon by the stock they are less able to cope with the more aggressive
species, and soon begin to die otit. This is the time when the farmer should give
the grasses some extra care. With very little trouble the pasture can be kept in
condition for profitable grazing, while neglect or carelessness may result in the
practical loss of the use of the pasture for one or more seasons.

Perennial weeds, sucli as golden-rod, iron-weed, and some of the sunflowers.
can usually be kept in check by mowing when in early bloom. The mower
should be riin high so as to miss as much of the grass as possible. Occasionally
it is necessary to grab out such plants as the rosinweeds.

Annual weeds, like marestail or fireweed. sunflower, and ragweed, may be
destroyed in the same manner. The mowing should always be done before
the seeds ripen. This not only gets rid of the weeds, but gives the grass a chance
to take advantage of the fall rains. A thorough harrowing in the spring while
the weeds are very young will destroy many of them and will also give new life
to the grass. Very often weeds of both classes may be very effectually checked
by burning over the pasture after they are well started in the spring.

EFFECT OF CULTIVATION ON NATIVE GRASSES,

The various native grasses are very differently affected by cultivation. Some
do not do well at all and soon die oiit, others are but little affected either way,
while still others respond very quickly and improve almost at once. This last
class includes the most valuable of the native species, such as big blue-stem,
western wheat-grass, wild-rye, and prairie June-grass.

The effect of loosening up the soil is very ap-
parent in a field which has "gone back" and
seeded itself to wheat-grass or blue-stem. In
many parts of Nebraska and the Dakotas three
tons or more of hay is often cut from such
fields. The fine growth which most grasses
make along the edges of cultivated fields is a
sight familiar to all who have traveled over
the western prairies, and ought to be an object
lesson to those to whom these same grasses are
of so much importance.

The fact that cultivation improves the more
desirable native grasses has been demonstrated
by nearly every experiment station in the
West and by a great many private parties as
well.

An experiment made at the Kansas Station
in 1892 shows what a thorough stirring up of
the soil will do for an upland prairie pasture.
The experiment was made on a pasture in
which the grasses had been dying out for some
time and the weeds were beginning to appear
in abundance. It had been reduced to this
condition by drought and overpasturing. The
surface was thoroughly loosened up by driving
a weighted disc harrow over the field in sev-
eral directions. The pasture was sown to a
mixture of orchard grass, meadow fescue, blue
grass, timothy, red top, clover, and alfalfa,
which was harrowed in and a roller was driven
over the field to level the surface and firm the
ground. The seed germinated quickly and the
tame grasses made an excellent start, but by
September the wild grasses had crowded them
out and held complete possession of the field.
In this case the stirring of the soil and the season's rest not only enabled the
prairie grasses to recover and to overcome the weeds, but to crowd oiit a good
stand of tame grasses as well.

This has been the experience in Nebraska and South Dakota where like attempts
have been made to renew worn out pastures. The tame grasses are undoubtedly
valuable aids, since the hardier of them will retain at least partial possession
of portions of the pasture and add considerably to the forage obtained. Many
of them, though they do not as a rule stand drought so well as the native species,
start earlier in the spring or make a better growth in the fall, and thus lengthen
the season during which the pasture may be used.

The continual trampling of the stock can not help but pack the soil more or
less, and consequently prevent its proper aeration. This packed condition also
keeps the water from gaining ready entrance, and it runs off and is lost. This,

Fig. 2.-

-Slender wheat-grass (Agro-
pyron tenerum).

too, when lack of moisture is i)erhaps the principal reason for the failure of the
pasturage. The old grass roots become crowded and die out or are weakened
through lack of available food and suitable soil in which to develope.

It is ver J' readily seen then why the treatment which was given to the pasture
at the Kansas Station produced such excellent results. The tearing up of the
soil gave ready access to air and moisture, putting new life into the roots of the
grasses which were cut up and separated by the disc harrow, so that thousands
of new shoots sprang up immediately. The rest for one season gave these new
l)lants time to get well established and form a new sod.

If this treatment is given before the pasture is too badly damaged, there is
iisually no need of sowing so much tame grass seed. There is little doubt that
an occasional tearing up of this kind and a little care given to the time and
manner of pasturing will get as much pasturage from the native grasses as can
be obtained from tame varieties under the same conditions.

MANURING NATIVE PASTURES.

There is quite a diversity of opinion among farmers and stock raisers on the
question of manuring native pasture lands. Some have obtained excellent re-
sults b}' manuring, while others seem to have had quite the opposite experience.

The soil of the western prairies is very rich and under
ordinary circumstances will give fair returns without the
application of fertilizers of any kind. Nevertheless, it is
certain that better returns may be had if more available
food is placed within reach of the grasses. Any one who
has observed a piece of grass land so situated as to receive
the wash from a barnyard, will have found that near
the yard where the supply of fertilizer has been great
the grasses have become thinned out to a few species,
while where the supply has been moderate the grasses are
much more evenly developed and the yield decidedly
better than upon the unfertilized prairie. The lesson is
plain. A too plenteoiis application of fertilizer will thin
out the grasses and reduce the yield of forage at least
for the first season or two. since many species will not
stand such treatment. On the other hand, a proper
amount of fertilizer will increase the yield. It is quite
possible to use too much fertilizer for any crop, and the
native grasses seem to be more sensitive in this respect
than the ordinary cultivated species.

Any pasture which has been grazed closely for some
time will be benefited by an application of a thin top-
dressing of well rotted stable manure, followed by a
thorough harrowing. It is doubtful if much is gained
by putting coarse unrotted manure on the pasture; it can
be used to better advantage on cultivated land. Ashes
usually have a beneficial effect upon grasses on soils not
too plentifully supplied with alkali.

SOWIN(i TAME GRASSES ON NATIVE PASTURE LANDS.

While it is hardly possible, and not always desirable, to
make a native pasture over into a tame one, yet, as stated
before, the pasturage may be materially increased by the
addition of some of the cultivated species. A pa.sture
which has had the thin places seeded to hardj' tame
grasses is certainly more valuable than it would be were
these same places grown up to weeds.

In dry upland pastures such grasses as Kentucky blue-grass, sheep's fescue,
red fescue, and Canadian blue-grass may be used to advantage. The fescues are
especially valuable if the soil is very sandy.

Lowland pastures, particularly those in which the grass has been killed out by
overflowing, may be reseeded with timothy, fowl meadow grass, red top, meadow
fescue, and alsike. Kentucky blue-grass will do well if the soil is not too wet.
It is likely that smooth brome-grass will prove useful on pastures that are to be
kept for long periods of time.

The practice of collecting the seeds of such native species as western wheat
grass, slender wheat-grass, wild rye, prairie June-grass, and the blue-stems and

Fn;. ;].— Bifr blue-siciii

(Andropogon provin-

cialis).

sowing them on the pasture is to be recommended. The writer recalls an in-
stance where a farmer in South Dakota obtained an excellent pasture by collect-
ing western wheat-grass and filling in the bare places
with it.

Though timothy as a general thing is a poor pasture
grass for upland soils, it may sometimes be profitably
employed in old or worn pastures. The farmer very often
has a greater or less quantity of seed which has shattered
out in the hay mow. It has cost practically nothing and
would probably hardly pay for the cleaning if he were
to sell it. If this be scattered about over the pasture,
either in the fall or spring, it will pay very well indeed.
The timothy may not live in the pasture more than two
or three years, but it will yield considerable forage in
the mean time and help the native grasses keep down
the weeds. In eastern Nebraska, Kentucky blue-grass is
one of the best grasses that can be used for reseeding the
native pastures. The seed may be sown just as the last
snow is melting in the early spring. The grass when
once started keeps slowly spreading, and after a time
forms an excellent sod. It begins its growth early in the
spring and, thoiigh often dry and short during midsum-
mer, makes good grazing after the fall rains, and hence
gives a longer season, during which the stock can be kept
on the pasture.

SUMMARY.

1. Keep from overstocking.

2. When the soil begins to get baked and packed, stir
it up with a harrow.

3. Give occasional light top-dressings of well rotted
stable manure.

4. Fill in thin spots with hardy tame or wild grasses
before the weeds get a start.

5. Keep the weeds mowed off so that the grasses may
get the benefit of all the plant food there is in the soil.

Fig. 4 — Bushy blue-stem
(Aiidropogoii nutans).

Approved :

Chas. W. Dabney, Jr.,

Assistant Secretary.

Washington, D. C, November 2, 1896.

Thomas A. Williams,

Assistant Agrostologist.

V-

J^ O T ^ ii^l » '^ *• ^

CiRct i.AK Xi>. 4.— (Revised.)

United States Department of A.^

DIVISION OF AGROSTOLOGY.

[Cirass and Forage Plant Investigations.]

THE RENEWIXi OF AVORN-OLT NATIVE PRAIRIE PASTURES.

Throtighoiit the prairie regions of the West one frequently sees native i^astures
nearly devoid of grass and often grown np to weeds of various kinds. Plainly
such pastures are of little value to the owners, so far as the forage obtained from
them is concerned, and the weeds are a constant eyesore, often rendenng the
land even worse than worthless.

Such a condition of things is usually due
to two causes, drought and overstocking.
While the farmer may have no control
over the drought itself, he can, by a little
care and foresight, put the pasture in con-
dition to withstand it in a great measure,
and he certainly can prevent the pasture
from being overstocked . Not infrequently
the use of the pasture is almost entirely
lost for one or two seasons, when a little
rest by removing the stock for a time or
by feeding green corn, sf)rghum, or other
soiling crops would have kept it in good
condition.

The native grasses are hardy and are
adapted to the natiiral conditions which
prevail on the prairies. Some species
stand grazing much better than others,
and after a pasture has been used for sev-
eral years it will be found that the weaker
grasses are giving way to the stronger
ones.

As a rule, the forage obtained from the
average prairie pasture is furnished by a
comparatively small number of species.
In the more thickly settled portions of the
great prairie States big blue-stem, bushy
blue-stem, the wheat-grasses, switch grass,
prairie June-grass, wild-rye, blue-joint,
and the various species of Stipa, Poa, and
Boutelona ftmiish most of the native pas-
turage.

WEEDS OF THE NATIVE PASTURES. FiG. 1. -Western wheat-grass (Jaropyron

spicatum).
The most troublesome weeds are either
annuals or perennials. The former, because of their vigorous and rapid gi-owth
spring up and take possession of a pasture in a very short time. The latter spread
more slowly, but are more difficult to eradicate. Left to themselves, the native
grasses will hold their owni against the weeds, but when they are pastured oflf and
trampled upon by the stock they are less able to cope with the more aggressive
species, and soon begin to die out. This is the time when the farmer sliould give
the grasses some extra care. With very little trouble the pasture can be kept in
condition for profitable grazing, while neglect or carelessness may result in the
practical loss of the use of the pasture for one or more seasons.

3

Perennial weeds, such as golden -rod, iron-weed, and some of tlie sunflowers,
can usually be kept in check by mo^^ang when in early bloom. The mower
should be run high so as to miss as much of the grass as possible. Occasionally
it is necessary to grub out such plants as the rosinweeds.

Annual weeds, like mare's-tail or fireweed, simflower, and ragweed may be
destroyed in the same manner. The mowing should always be done before
the seeds ripen. This not only gets rid of the weeds, but gives the grass a chance
to take advantage of the fall rams. A thorough harrowing in the spring while
the weeds are very voung will destroy many of them and will also give new life
to the grass. Very'often weeds of both classes may be very effectually checked
by burning over the pasture after they are well started in the spring.

EFFECT OF CULTIVATION ON NATIVE GRASSES.

The various native grasses are very differently affected by cultivation. Some
do not do well at all and soon die out, others are but little affected either way,
while still others respond very quickly and improve almost at once. This last
class includes the most valuable of the native species, such as big blue-stem,
western wheat-grass, ^v'ild-rye, and prairie Jiine-grass.

The effect of loosening up the soil is very ap-
parent in a field which has "gone back" and
seeded itself to wheat-grass or blue-stem. In
many parts of Nebraska and the Dakotas three
or more tons of hay are often cut from such
fields. The fine growth which most grasses
make along the edges of cultivated fields is a
sight familiar to all who have traveled over
the western in-airies, and ought to be an object
lesson to those to whom these same grasses are
of so much importance.

The fact that cultivation improves the more
desirable native grasses has been demonstrated
by nearly every experiment station in the
West and by a great many private parties as
well.

An experiment made at the Kansas Station
in 1892 shows what a thorough stirring up of
the soil will do for an upland prairie pasture.
The experiment was made on a pastiire in
wiiich the grasses had been dying out for some
time and the weeds were beginning to appear
in abundance. It had been reduced to this
condition by drought and overpasturing. The
surface was thoroughly loosened up by driving
a weighted disk harrow over the field in sev-
eral directions. The pasture was sowai to a
mixture of orchard grass, meadow fescue, blue
grass, timothy, red top, clover, and alfalfa,
which was harrowed in and a roller was driven
over the field to level the surface and firm the
ground. The seed germinated quickly and the
tame grasses made an excellent start, but by
September the wild grasses had crowded them
out and held complete possession of the field.
In this case the stirring of the soil and the season's rest not only enabled the
prairie grasses to recover and to overcome the weeds, but to crowd out a good
stand of tame grasses as well.

This has been the experience in Nebraska and South Dakota where like attempts
have been made to renew worn-out pastures. The tame grasses are imdoubtedly
valuable aids, since the hardier of them will retain at least partial possession of
portions of the pasture and add considerably to the forage obtained. Many of
them, though they do not as a rule stand drought so well as the native species,
start earlier in the spring or make a better growth in the fall, and thus lengthen
the season during which the pasture may be iised.

The continual trampling of the stock can not help biit pack the soil inore or
less, and consequently prevent its proper aeration. This packed condition also
keeps the water from gaining ready entrance, and it runs off and is lost. This,

Fig. 2.— Slender wheat-grass (Auro-
pyroH tcnerum).

3

too, when lack of uioistiire is perhaps the i)rinc;ii>al reason for the failure of the
pasturage. The old grass roots become (-rowded and die out or are weakened
through lack of available food and suitable soil in which to develop.

It is very readily seen then why the treatment which was given to the pasture
at the Kansas Station produced such excellent results. The tearing n\t of the
soil gave ready access to air and moisture, i)utting new life into the roots of the
gi'a.sses whi(-h were cut up and separated by the disk han*ow, so that thousands
of new shoots sprang up immediately. The rest for one season gave these new
plants time to get well estal)lished and form a new sod.

If this treatment is given before the pasture is too badly damaged, there is
usually no need of sowing so much tame grass seed. There is little doubt that
an occasional tearing up of this kind and a little care given to the time and man-
ner of pasturing will get as much pasturage from the native gi'asses as can be
obtained from tame varieties under the same conditicms.

MANURING NATIVE PASTURES.

There is cpiite a diversity of opinion among farmers and stock raisers on the
cpiestion of manuring native pasture lands Some have obtained excellent re-
sults by manuring, while others seem to have had quite the opposite experience.

The soil of the western prairies is very rich, and under
ordinary circumstances will give fair returns without the
application of fertilizers of any kind. Nevertheless, it is
certain that better returns may be had if more available
food is placed wdthin reach of the grasses. Anyone who
has observed a i)iece of grass land so situated as to receive
the wash trom a barnyard, will have found that near the
yard where the supply of fertilizer has been great the
gi-asses have become thinned out to a few species, while
where the supply has been moderate the grasses are much
more evenly developed and the yield decidedly better
than upon the unfertilized prairie. The lesson is plain.
A too plenteous application of fertilizer will thin out the
grasses and reduce the yield of forage at least for the first
season or two, since many species will not stand such
treatment. On the other hand, a proper amount of fer-
tilizer will increase the yield. It is quite possible to use
too much fertilizer for any crop, and the native grasses
seem to be more sensitive in this respect than the ordinary
cultivated species.

Any pasture which has been grazed closely for some
time will be benefited by an application of a thin top-
dressing of well rotted stable manure, followed by a
thorough harrowing. It is doubtful if much is gained
by putting coarse imrotted manure on the pasture ; it can
be used to better advantage on the cultivated land. Ashes
usually have a beneficial effect upon grasses on soils not
too plentifully sux)plied with alkali.

SOWING TAME (BRASSES ON NATIVE PASTURE LANDS.

While it is hardly possible, and not always desirable, to
make a native pasture over into a tame one, yet, as stated
before, the pasturage may be materially increased by the
addition of some of the cultivated species. A pasture
which has had the thin places seeded to hardy tame
grasses is certainly more valuable than it would be were
these same places grown up to weeds.

In dry upland pastures such grasses as Kentucky blue-grass, sheep's fescue, red
fescue, and Canadian l)lue-grass may be used to advantage. The fescues are
especially valual)le if tlie soil is very sandy. It is likely that smooth brome-
grass will prove useful on pastures that are to be kept for long periods of time.

Lowland pastures, particularly those in which the grass has been killed out by
overflowing, may be reseeded with timothy, fowl meadow grass, red top, meadow
fescue, and alsike. Kentucky blue-grass will do well if the soil is not too wet.

The practice of collecting the seeds of such native species as westeni wheat-
grass, slender wheat-grass, wild-rye, prairie June-grass, and the blue-stems and

Fig. 3.— Big blue-stem

{Anrlr(>))im(»i, jjro-

viiici(tlis).

sowing them on the pasture is to be recommended. The writer recalls an
instance where a farmer in South Dakota obtained an excellent pasture by col
lecting western wheat-grass and filling in the bare places with it. Where the
wheat grasses are naturally abundant excellent results have been obtained by

plowing uj) the pasture and keeping the stock off of it
until these grasses take possession, which occurs in a very
short time.

Though timothy as a general thing is a poor pasture
grass for upland soils, it may sometimes be profitably
employed in old or worn pastures. The farmer very
often has a greater or less quantity of seed which has
shattered out in the hay mow. It has cost practically
nothing and would probatsly hardly pay for the cleaning if
he were to sell it. If this be scattered aboiit over the
pasture, either in the fall or spring, it will pay very well
indeed. The timothy may not live in the pasture more
than two or three years, but it will yield considerable
forage in the meantime and help the native gi"asses keep
down the weeds. In eastern Nebraska, Kentucky blue-
grass is one of the best grasses that can be used for reseed-
ing the native pastures. The seed may be sown just as the
last snow is melting in the early spring. The grass when
once started keeps slowly spreading, and after a time forms
an excellent sod. It begins its growth early in the spring
and, though often dry and short during midsummer, makes
good grazing after the fall rains, and hence gives a longer
season, during which the stock can be kept on the pasture.
Another plant which can be profitably used for this pur-
pose is white clover ; it persists in the soil for a long time
and, though small, adds considerably, both in quantity
and quality, to the pasturage and exerts a beneficial
influence upon the soil.

SUMMARY.

1. Keep from overstocking.

'2. When the soil begins to get baked and packed, stir it
up with a harrow.
^(findVoSou'^iutan^"' •^- ^^^^^ occasional light top-dressings of well rotted

stable manure.

4. Fill in thin spots with hardy tame or wild grasses before the weeds get a.
start.

5. Keep the weeds mowed off so that the grasses may get the benefit of all the
plant food there is in the soil.

Thomas A. Williams,
Approved : Assista ii f Agrostologist.

James Wilson,

Secretary.

Washington, D. C, Norembcr 15, is'.rr.

r^.

Circular No. 5.

United States Department of Agriculture,

DIVISION OF AGROSTOLOGY.

COWPEAS. "^s^ B O T * i« J C *. t

( f'ifjna catjarifi.)

(Reprinted from the Yearbook of the U. S. Department of Agriculture for 1S96.)

ORIGIN AND GENERAL CONSIDERATIONS.

The cowpea is to the South what alfalfa is to the West and red
clover to the North — a forage plant perfectly adapted to the needs of
the region where it grows. The cultivation of this crop in America
dates back to the early part of the eighteenth century. A Soutli Car-
olina planter received a quantity of seed from a foreign source, which,
according to certain authorities, was an English acclimatization society
or the captain of a trading vessel from far-off India or China. From
this small and obscure beginning cowpeas spread throughout the
South, and their cultivation has been essayed as far north as Connec-
ticut, New York, and South Dakota, and westward to California,

Cowpeas grow wild in far eastern tropical lands, including India,
China, Siam, the Malay Archipelago, and portions of Central Africa,
and have become an escape from cultivation in the southern United
States and tropical America. From the South the plant has been car-
ried in recent years to South Africa and Australia, so that it is now
grown as a forage plant or for human food throughout all the warmer
quarters of the globe. Cowpeas are in their relationship and habit of
growth really beans, and not, as the name would indicate, peas. They
belong to the genus Vigna, the members of which are largely repre-
sented in South Africa, and are closely related to the lablab, lima, and
haricot beans of our gardens, as well as to numerous cultivated or half-
wild garden sorts common in tropical Asia and America, but little
. known to us.

There are a very large number of named forms or varieties of this
forage plant. New forms are constantly arising, due to variations
in habit of growth, color of leaf, stem, and pod, and the shape and
color of the seed. Variations from any chosen type are constantly
appearing, and as one or another of these sports or forms gains suffi-
cient local reputation a new name is applied and sooner or later the
supposed new variety is placed upon the market. In this way one
variety of cowpea may be cultivated in a dozen different localities
under as many names, oi- a dozen different peas may bear the same
name. The whole subject of the nomenclature of varieties is in a
chaotic state and can be straightened out only after years of careful

study have been given to it bj' botanists and the experimental agri-
cultui-ists. No valid conclusions can be drawn from the brief study
of a subject so complex. Cowpeas pass through every gradation of
form, from a short, stocky, upright bush having single stems a foot
high with very short lateral branches to those w^th trailing runners
growing as flat upon the ground as sweet-potato and melon vines, the
prostrate stems 15 or 20 feet in length. The pods vary from 4 to 16
inches in length, and the peas are of everj^ imaginable shade through
white, yellow, green, pink, gray, brown, red, purple, and black, of
solid colors or variously mottled and speckled, and of varying sizes
and forms from large kidney-shaped to little round ones smaller than
the garden pea. There is a like variation in the length of time the
different forms require to ripen seed, some requiring eight or nine
months, a few ripening ill sixty days from the time of i)lanting.

There seems to be a somewhat constant relation between the time
required for attaining maturitj'^ and the habit of growtli. The bush
varieties ripen in a shorter season than the trailers, but a bush vari-
ety taken from the North will, in the course of a few seasons, assume
the trailing habit and lengthen out its period of growtli in any of the
Southern States. Also, a runner or creeper requiring six to eight
months for reaching maturity in Louisiana will, if planted each year
a hundred miles farther north, gradually accommodate itself to the
shorter season and at the ^ame time shorten its runners, approaching
more and more to the upright or "bush " habit of growth. There can
be no hard and fast line of separation between bush j)eas, trailers,
and runners. The best varietal character is probably the color of
the seed. It is quite probable that more than one species is in culti-
vation. The ' ' red " and ' ' black " varieties are closely allied ; the round
"lady" peas form a separate group; the large " black-ej^ed " and
"purple-eyed" are typical of another, and the variously mottled and
speckled "whip-poor-wills" are only a degree removed from the solid-
colored yellow, pink, and light-brown ones, and together would nat-
urally be taken to constitute one species or variety. The black peas
jjass through various shades of red before maturity. The red varieties
sometimes carry their change of color in ripening so far that they can
not be distinguished from the black. The ' ' black-eyed " and ' ' i)urple-
eyed " are of the same ground color, differing only in the color of the
ring surrounding the eye. The various " crowders," yellow and white,
the whip-poor-will, clay, and "yellow-eyed" forms have numerous
crosses and so-called hybrids in which the fundamental yellows and
browns form varying mixtures.

, COWPEAS AND SOIL RENOVATION.

A field of cowpeas has been very happilj" designated ' ' the poor man's
bank," for in common with all its leguminous congeners, the field
pea, clovers, alfalfa, and a score of others, this crop has the power of

Increasing the fertility of the soil upon which it gro-ws. This fact has
long l)een accepted by farmers and students of agriculture, but until
recent discoveries in Germany and America it was believed that the
chief function of these plants was to pumj) up nitrogen from the sub-
soil reservoir to the surface by means of their long roots for the use
and benefit of succeeding crops.

But experiments in the field and laboratory for the purpose of deter-
mining the causes of natural phenomena have taken the place of class-
room philosophy and speculative reasoning. . Within the last twenty
years scientific workers have discovered that minute micro-organisms,
or bacteria, which live within the tissues of the roots of leguminous
plants take up free nitrogen froui the gases in the soil, just as the
higher i:)lants and animals utilize the oxygen of the air. This nitrogen
enters into combination to form nitric acid, Avhicli unites with the min-
eral elements of the soil to form nitrates, a kind of plant food exceed-
ingly vahmble to the growing crop. Nitrogen, when in combination
with other elements, is an indispensable form of i^lant and animal
food, but the free element can not be utilized, uncombined, by any of
the higher organisms. Small amounts of nitrous acid are formed
as a result of lightning discharges and are washed out of the air by the
rains, to be in part absorbed by the soil, and in part carried by rivers
and drainage waters into the sea. Free nitrogen exists onl}^ in the
air and in the gases of the soil, but as ammonia, nitrous and nitric
acid, nitrites and nitrates, it is present in varying quantities in
the soil, the unbroken rocks, and the waters of continents and
oceans.

The most available j^urchasable nitrogen is obtained either as salt-
peter or nitrate of soda from the extensive deposits in the Peruvian
deserts, or from some form of animal wastes, such as freshly ground
bone, dried blood, guano, tankage, and fish scrap, and from cotton-seed
meal and other like bj'-products of the oil mills. These fertilizers are
all exf)ensive, so much so that they can be i)rofitably employed b\' the
farmer only in intensive farming with specialized crops. The gain in
yield with low-jDriced crops, such as corn, cotton, tobacco, cowpeas,
and the grasses, using high-grade and costly fertilizers, is not com-
mensurate with the additional expense. But everj^ farmer, rich and
poor, has over three thousand tons of atmospheric nitrogen resting on
every acre of his farm, a certain quantity of which can be transformed
into available plant food every time that he grows a crop of cowpeas,
red clover, or alfalfa.

There are a great many acres of farming hind in the South in need
of renovation. The red uplands and yellow-clay soils were undoubtedly
less fertile originally than the alluvial and black prairie soils, and the
methods of cultivation which formerly prevailed have still further
diminished their productiveness. In the days when every plantation

nuiubered its acres by the tliousand and labor was cheap, the planter
could afford to clear off tlie native forest groAvth and l)ring fresh fields
into cultivation whenever the yields of cotton and tobacco fell below
what was considered a profitable figure. The old field, stripped in a
few years of its accumulated store of humus, was abandoned and
allowed to grow up to weeds and underbrush. The forest again spread
across it, and gradually, in the slow course of half a lifetime, the nat-
ural enrichment of its surface soil by the growth of the woodland
grasses made it ready for another robbery.

But with the breaking up of the large estates and tlie abrupt change
in the labor conditions this method of farming became no longer profit-
able or even possible. A planter with fewer acres could no longer
afford to await nature's slow process of rejuvenating the soil. A new
system of farming was necessary. The land must not be allowed to
"go back. " It must be kept up to the highest state of productiveness
by a rotation of crops, a judicious use of commercial fertilizers, the
growth of nitrogen-fixing leguminous crops, and good and thorough
cultivation. To maintain the fertility of any soil the amount of humus
or decaying organic matter in it must be kept up. Take two soils of
as nearly as possible the same pliysical and geological formation, but
the one rich in humus and the other lacking it, and fertilize them with
equal quantities of commercial manures; the one which has the most
organic matter in its composition will yield the largest crop. The soil
on\hat field will stand drought better, will wash less under torrential
rains, and be more friable and of better tilth. The average soils of
the South need more humus. It can be best supplied by sowing more
grass, more permanent pasture lands, more leguminous crops. In a
word, plant cowpeas.

COWPEAS FOR FORAGE.

There is no forage plant better adapted to the needs and conditions
of Southern agriculture than this rank, free-growing annual. It will
thrive luxuriantly upon the rich, swampy, cane lands of Louisiana.
On the driest and most sterile worn-out uplands it serves the admir-
able purpose of supplying a larger quantity and better quality of for-
age than any other bean or clover. And whenever a crop of cowpeas
has been taken off a field the surface soil is left richer by a good
many pounds of that most costly of all plant foods, nitrogen. The
roots of the co\\^ea enter deeply into the soil, opening and loosening
it far down for the benefit of the roots of the succeeding crops of corn,
cotton, and tobacco. It has been found by experiment that the ferti-
lizins: value of the roots and stubble of the cowpea are very consider-
able, but not as great as that of tlie hay removed from the field. The
best and most economical use of this forage crop is, then, to cut for hay,
feed to stock, and return the stable manure to the soil. Plowing the
whole crop under is less remunerative because there is much needless

5

waste of the muscio-iuaking and fat -form in. i>- constituents of tiie i)lant
which would bring more j^rofit if turned into beef, pork, wool, cheese,
•or butter.

As regards the disposal of the crop, there is a wide variation in
practice. The feeding value of vines and peas much exceeds their
fertilizing value. But as between the practice of turning the vines
under green in autumn and that of allowing them to lie on the
ground during the winter, the latter is undoubtedly sometimes to be
preferred, though theoretically wrong. Theoretically, to plow the
vines under in autumn will be to save all the available nitrogen and
convert the whole plant into humus. Practicall}', the turning under
of so large an amount of watery green herbage is highly injurious,
causing a too rapid decay and consequent "burning" or souring of
the soil. The upper soil layers, freshly stirred and mellowed in
autumn, lose more b}^ leaching and washing than they do in an
unplowed field covered by its winter mulch of decaying herbage,
though in both cases there is a decided loss of fertility over what
would result by following the peas with a crop of rye, ^vinter Avheat,
the turf-forming winter oats, winter vetch, or crimson clover. The
yields of forage are better on rich soils than on poor ones, but the
beneficial effects upon the succeeding crop due to the growth of this
one are not so marked in the former case as in the latter.

METHODS OF CULTIVATION AND HARVESTING.

Cowpeas are planted broadcast or in drills, vevy commonlj^ between
the corn rows after the crop is laid by. The amount of seed used
varies from 4 quarts to 2 bushels per acre, the average amount being,
perhaps, about 3 pecks. If sown in drills, 18 to 30 inches apart, less
seed is required than when sown broadcast. The seed w'ill stand being
covered to the depth of 2 or 3 inches, but care must be taken to plant
when the ground is neither too wet nor too cold, as the peas rot very
rapidly under such circumstances. In regard to excess of moisture
cowpeas behave like beans, and in the earl}' stages delight in a warm,
mellow seed bed. Much of the failure that has attended the attempted
introduction of cowpeas into the Northern States is due to planting
before the ground is warm enough. It must be remembered that this
X^lant originated in the Tropics and that when transplanted to higher
altitudes it makes its best growth in the hottest weather. It is even
more susceptible to cold and wet than is Indian coi-n. Hence, proper
delay in planting will permit economy in the use of seed. A\1iere the
vines are grown for hay, the yield will be larger if the seed is planted
in drills and cultivated a time or two. The yield of peas is also larger
Mhen only a moderate amount of seed is sown and the vines have more
space and light and air between them. It is also heavier from late-
planted vines than from the very early ones. In tests to determine
the relative value of different named varieties it has been found that,

6

as a rule, those which make the heaviest j'ields of vines also bear larj^e
crops of peas.

The vines should be mowed for hay when the peas are well formed
and the leaves are first beginning to turn j'ellow. After wilting on
the grovmd or in windrows from twentj-four to forty-eight hours, the
hay is placed in small, thin piles, or cocks, and allowed to cure tor
several daj's, when it may be carted to the barn or stacked under
sheds. The haymaking process is a difficult one, requiring more care
and attention than in the case of red clover, because the broad leaves
and thick stems contain a larger amount of water. The hay must be
placed in cocks before the leaves become brittle, and the piles must
be small enough to allow free circulation of air to the center of each.
Bright cowpea hay, clean and well cured, is worth as much as the
best red-clover hay, and there is no good reason why the Southern
farmers and planters should buy the jSTorthern-grown article for their
working stock or for fattening their cattle. Every ton of hay used on
the estate should be grown there. Another method of caring hay is
to stack the vines in a pen or rack of rails or j^oles so arranged as to
allow the air to enter every part of the pile. This stacking over poles
is the best where the vines are pulled, or where the trailing and creep-
ing sorts are used. The bush varieties are the best for hay, because of
the greater ease with which they may be mowed and handled. They
also hold their leaves better than the ranker trailing sorts. The yield
of hay varies according to the fertility of the soil upon which it is
raised, whether it is grown on rich lowlands or on the drier and more
sterile uj)lands. In the Gulf States cowpeas will probably give an
average yield of 2 to 3 tons per acre, while 4 to 6 tons are not uncom-
mon. Farther north the average Avill range from 14- tons in Ohio to
2i tons in Arkansas, Missouri, and Tennessee. As with other crops,
the time of planting, the character of the soil and of the cultivation,
and the amount of rainfall have much to do with the yield. Along
the Gulf it is one of the best hay crops. North of the latitude of the
Ohio River it is chiefly valuable as an addition to the list of drought-
resistant, summer-soiling crops and as a crop that will yield a con-
siderable amount of forage on soil too sterile to grow red clover. The
commercial value runs from $G to 120 per ton, being governed by the
relative abundance of other grades of hay and fodder. Its feeding
value is equal to that of the best red clover, and the hay ranks high
in palatability and digestibility.

roWPEAS FOR SWIXE AXD CATTLE,

When cowpeas 'are planted for green manure, it is an excellent prac-
tice to turn hogs into the field about the time that the first peas are
ripening. Young i^igs thrive amazingly on the succulent foliage and
well-filled pods, and the (quality of the pork raised on such a healthful

and nutritions diet is very fine. This is a very profitable metliod of
fattening hogs or of preparing them foi* topping off witli corn or
sorghum for market. An acre of i-ipening cowpeas will pasture from
fifteen to twenty hogs for several weeks, and the gain in fertility
from the droppings of the animals during that period will more than
counterbalance the fertilizing value of the forage eaten. The rapid
increase in weight will thus represent so much clear profit, and the
farmer is richer by half a ton or more of prime pork for every aci'o
planted. Chickens and turkej^s also eat the ripe peas and do well
upon them. Cattle and horses are sometimes pastured on them, but
the safer and more economical way of feeding cowpea vines to such
stock is to cut or pull and feed partially wilted. There will be less
waste and destruction from trampling, and if each animal is given
only so much as it can eat clean, the greatest economy as well as
greatest i)rofit will result. Furthermore, cattle and sheep are liable
to bloat if allowed to eat too ravenously of cowpea vines or any other
rich and succulent forage, and by using it as a soiling crop the danger
maj' be more readily controlled and the loss prevented. The report has
been sent out from some of the Northern experiment stations, where
this forage plant is not ordinarily cultivated, that cattle will not eat
the green vines except after having been starved to it, and then only
sparingly. We have seen Western horses and ponies that would not
touch red clover or a grain ration of oats, and have heard of Eastern
stock that would not eat alfalfa hay. But these few adverse cases
do not prove that red clover, alfalfa, and oats are not good forage.
With the cowpea the case is similar. It is very rarely that any South-
ern planter reports that this forage is refused by any kind of stock.

COWPEAS FOR ENSILAGE.

Reports are very conflicting in regard to the value of this crop for
ensilage. There is much positive testimony both for and against, some
authorities stating that the quality is excellent and others that the
vines contain, too much water, the product of the fermentation being
a slimy, foul-smelling mass, unfit for food for any kind of animals.
From reports on the subject it is to be believed that the attempt to
convert cowpea vines into good ensilage can not be made with such
uniform success as in the case of red clover. 'J'he percentage of water
in the tissues is too high, and the mechanical difficulties in the way
of running a mass of tangled herbage through the feed cutter are too
great. Special niachinery would have to be constructed for the pur-
pose. Indian corn will probably remain for many years the best all-
round forage plant for this purpose. The consensus of opinion among
agricultural workers seems to be that ensilage made from any legume,
whether it be cowpeas, vetches, soja beans, alfalfa, oi- the clovers, does
not equal in feeding value good hay made from the same. Under

8

certain conditions that arise in the silo the crude protein is converted
into indigestible or insoluble nitrogenous compounds. The cowpea
or clover ensilage is then valuable only for the carbohydrates that it
contains, and either corn or sorghum is far superior to it.

HARVESTING THE SEED.

The majority of farmers harvest only enough seed of cowpeas to
plant again the next season. The ripe pods are picked by hand and
are stored in barrels until needed or are thrashed out by machine or
with flails on the barn floor during the winter. Sometimes, if the
crop is heavj^ enough to render it profitable, the vines are run through
an ordinary thrashing machine from which the concaves and alternate
teeth of the cylinder have been removed. But a machine breaks and
bruises more of the seed than when the pods are first picked off by
hand. Fully 95 per cent of the seed jplaced upon the market is
hand picked. The yield per acre varies according to the varieties and
the method of cultivation. Eight to twelve bushels is a fair average
of the amount that can be obtained when the peas are planted in the
corn rows. Sown alone, broadcast or in drills, yields of from twenty
to thirty-five and even, in rare cases, fifty l>ushels are obtained. The
Black, Unknown, Red Ripper, Clay, and Calico varieties are all heavj^
seed bearers. Lady and White Crowder are good for table use and also
yield well. The Black-eyed,- Red Crowder, and AVhip-poor-will or
Speckled are very widely cultivated and find ready sale. Those which
make the largest growth of vines for green manure, as a winter soil
mulch, for hay or soiling are the Unknown, Red Ripper, Southdown,
and Claj-. Whip-ijoor-will, Black-eyed, White, and Red Crowder
ripen in from twelve to fourteen weeks, and hence are adaj)ted to
cultivation farther noi-th than the very late, but ranker growing,
Unknown, Wonderful, Red Ripi^er, Black, and Gourd varieties. The
New Era and Lee ripen seed in from six to seven weeks, and hence are
the ones to recommend for summer-soiling crox)S in the upper prairie
region of the Mississippi Valley or anywhere else that an earlj'- matur-
ing cowpea is required. This is one of the species of cultivated plants
which is very readily modified by change of habitat. Early and late
maturing forms may be found of every strain that has been in culti-
vation for any considerable time.

THE FEEDING VALUE OF COWPEAS.

The feeding value of cowpea vines is very high, as shown by both
feeding tests and chemical analyses. As hay the vines are more val-
uable than fed green for soiling purposes. A comparison with red
clover and alfalfa is made in the table on the next page, a compila-
tion ^ of the averages of a number of analyses from various sources.

1 Handbook of Experiment Station Work, Appendix, 1893.

9

Feeding ixilue of cowpeas compared with red clover and alfalfa.

Article.

Cowpeas:

Green .

Hay ...
Red clover

Green .

Hay...
Alfalfa:

Green .

Hay...

Fresh or air-dry material.

Number

of
analyses.

10

8

43

38

23
21

Water.

8;). ti

10.7

70.8
15.3

71.8
8.4

Ash.

1.7
7. 5

2.1

G.2

Protein.

2.4
II). ti

4.4

12. 3

4.8
14.3

'Nitrogen-
Fiber. | free
extract.

Fat.

4.8
20.1

8.1
24.5

7.4
25.0

7.1
42.2

13. 5
.■!3. (1

12.3
42.7

O.Oi
2.9

1.1
3.9

1.0
2.3

Cowpea.s:

Green . .

Hay...
Red clover:

Green . .

Hay....
Alfalfa:

Green . .

Hay....

Article.

Calculated to water-free substance.

Ash. Protein. I Fiber.

1U.5
8..")

Nitrogen-'

free ; Fat.
extract.

7.3

9.4

8.1

11.3

29.0

4.3. t;

18. ti

22.5

47.2

!.->.:;

27.8

45. 8

13. 5

31.3

43.0

17.1

2(i.3

43.9

15.15

27.3

46.6

2.6
3.2

3.9
4.9

3.4
2.4

A study of the percentages here given will show that the green
vines contain more water, less protein or nitrogenous, muscle-making
food, and less of the fat-forming crude fibers, fats, and nitrogen-free
extracts than either the green alfalfa or red clover. The air-dry hay,
however, contains more protein than either of the others, less fiber,
more nitrogen-free extracts than the red clover, and more fat than tlie
alfalfa. As is the case with leguminous forage plants in general, a
ration of cowpeas, to be well balanced, requires the addition of some
coarse fodder, such as corn stover, sorghum, timothy, Bermuda, or
prairie hay, otherwise a portion of the protein will be wasted.

FERTILIZERS.

It has been found that, as a rule, it does not pay to use high-grade
commercial fertilizers on cowpeas; this, however, depends a good
deal on the soil and on what crop is to follow this green manurial one.
It is usually unprofitable to fertilize with expensive nitrogen, in the
form either of nitrate of soda or of guano, and even the organic
nitrogen of cotton-seed meal does not act upon this crop as rapidly
as upon cotton and the cereals. The nitrogen of the fertilizers seems
not to influence the percentage of protein in the crop, and the general
opinion of agriculturists in the South is that it does not cause a

10

sufficient increase in yield of vines to pay the cost. At the Delaware
Station IGO pounds of muriate of potash per acre doubled the yield of
vines, and superphosphate produced no effect. At the Georgia Station
combinations of superphosphate and potash gave the best results, but
later experiments there indicated that large amounts of potash are
unprofitable, and that superphosphate at the rate of from 200 to 400
pounds per acre gave better results. Superphosphates are very much
preferable to untreated rock phosphate. The latter can be sold at
much lower rates, and it remains to be seen whether it would not be
a profitable method to apply the soft phosphate to the cowpeas for the
benefit of the succeeding crop in the rotation, for it has been found
that the insoluble phosphoric acid of the untreated rock becomes
changed to forms available as plant food in the presence of large
amounts of decaying vegetable matter in the soil. If it is found that
this process can be relied upon, then the cowpea will have another
valuable quality added to it, namely, that of being able to change
into high-grade and more costly superphosphate the low-grade and
cheap but unavailable phosphoric acid of the untreated rock.

The chief functions of this crop, then, are to furnish large amounts
of nitrogen abstracted from the air and fixed in the roots and stubble
in a conveniently available form for the use of succeeding crops; sec-
ond, to produce a large yield of vines and peas rich in digestible pro-
tein, which, either as haj^ or for soiling purposes, will take the place of
concentrated nitrogenous foods; and, third, to supply humus, which
acts directly and indirectly to produce fertility by breaking down and
rendering available the basic minerals of the soil. The fertilizing
value of the nitrogen in the vines is entirely dissipated or greatly di-
minished by weathering when they are left on the surface of the field
during the winter. Hence, to secure the full value, the cowpeas should
be fed and the stable manure returned to the field. If the vines are
plowed under in autumn, a winter forage crop, such as winter oats,
-crimson clover, rye, or vetches, should be planted to prevent the
leaching and washing action of the winter rains.

Jared G. Smith,

Assistant A{/rostolo</ist.
Approved :

James Wilson, Secretary.

Washington, D. C., November 26, 1897.

Circular No. 5.— (Revised edition.)

United States Department of Agriculture,

DIVISION OF AGROSTOLOGY.

COWPEAS.

{Vigna catjang.)

(Reprinted from the Yearbook of the U. S. Department of Agriculture for 1896.)

ORIGIN AND GENERAL CONSIDERATIONS.

The cowpea is to the South what alfalfa is to the West and red
clover to the North — a forage plant perfectly adapted to the needs of
the region where it grows. The cultivation of this crop in America
dates back to the early part of the eighteenth century. A South
Carolina planter received a quantity of seed from a foreign source,
which, according to certain authorities, was an English acclimatiza-
tion society or the captain of a trading vessel from far off India or
China. From this small and obscure beginning cowpeas spread
throughout the South, and their cultivation has been essayed .as far
north as Connecticut, New York, and South Dakota, and westward
to California.

Cowpeas grow wild in far eastern tropical lands, including India,
China, Siam, the Malay Archipelago, and portions of Central Africa,
and have become an escape from cultivation in the southern United
States and tropical America. From the South the plant has been
carried in recent years to South Africa and Australia, so that it is
now grown as a forage plant or for human food throughout all the
warmer quarters of the globe. Cowpeas are in their relationship and
habit of growth really beans, and not, as the name would indicate,
peas. They belong to the genus Vigna, the members of which are
largely represented in South Africa, and are closely related to the
lablab, lima, and haricot beans of our gardens, as well as to numer-
ous cultivated or half- wild garden sorts common in tropical Asia and
America, but little known to us.

VARIETIES.

There are a very large number of named forms or varieties of this
forage plant. New forms are constantly arising, due to variations
in habit of growth, color of leaf, stem, and pod, and the shape and

2

color of the seed. Variations from any chosen type are constantly
appearing, and as one or another of these sports or forms gains suffi-
cient local reputation a new name is applied and sooner or later the
supposed new variety is placed upon the market. In this way one
variety of cowpea may be cultivated in a dozen different locaHties
under as many names, or a dozen different peas may bear the same
name. The whole subject of the nomenclature of varieties is in a
chaotic state and can be straightened out only after years of careful
study have been given to it by botanists and the experimental agri-
culturists. No vahd conclusions can be drawn from the brief study
of a subject so complex. Cowpeas pass through every gradation of
form, from a short, stocky, upright bush having single stems a foot
high with very short lateral branches to those with trailing runners
growing as flat upon the ground as sweet-potato and melon vines, the
prostrate stems 15 or 20 feet in length. The pods vary from 4 to 16
inches in length, and the peas are of every imaginable shade through
white, yellow, green, pink, gray, brown, red, purple, and black, of
sohd colors or variously mottled and speckled, and of varying sizes
and forms from large kidney-shaped to httle round ones smaller than
the garden pea. There is a hke variation in the length of time the
different forms require to ripen seed, some requiring eight or nine
months, a few ripening in sixty days from the time of planting.

There seems to be a somewhat constant relation between the time
required for attaining maturity and the habit of growth. The bush
varieties ripen in a shorter season than the trailers, but a bush variety
taken from the North will, in the course of a few seasons, assume the
trailing habit and lengthen out its period of growth in any of the
Southern States. Also, a runner or creeper requiring six to eight
months for reaching maturity in Louisiana will, if planted each year
a hundred miles farther north, gradually accommodate itself to the
shorter season and at the same time shorten its runners, approaching
more and more to the upright or "bush" habit of growth. There
can be no hard and fast line of separation between bush peas, trailers,
and runners. The best varietal character is probably the color of
the seed. It is quite probable that more than one species is in culti-
vation. The ' ' red "and ' ' black " varieties are closely allied ; the round
"lady" peas form a separate group; the large "black-eyed" and
"purple-eyed" are typical of another, and the variously mottled and
speckled "whip-poor-wills" are only a degree removed from the
solid-colored j^ellow, pink, and light-brown ones, and together would
naturally be taken to constitute one species or variety. The black
peas pass through various shades of red before maturity. The red
varieties sometimes carry their change of color in ripening so far that
they can not be distinguished from the black. The "black-eyed"
and "purple-eyed" are of the same ground color, differing only in

the color of the ring surrounding the eye. The various "crowders,"
yellow and white, the whip-poor-will, clay, and "yellow-eyed" forms
have numerous crosses and so-called hybrids in which the funda-
mental yellows and browns form varying mixtures.

METHODS OF CULTIVATION.

Cowpeas are planted broadcast or in drills, very commonly be-
tween the corn rows after the crop is laid by. The amount of seed
used varies from 4 quarts to 2 bushels per acre, the average amount
being, perhaps, about 3 pecks. If sown in drills, 18 to 30 inches
apart, less seed is required than when sown broadcast. The seed
will stand being covered to the depth of 2 or 3 inches, but care must
be taken to plant when the ground is neither too wet nor too cold, as
the peas rot very rapidly under such circumstances. In regard to
excess of moisture cowpeas behave like beans, and in the early stages
delight in a warm, mellow seed bed. Much of the failure that has
attended the attempted introduction of cowpeas into the Northern
States is due to planting before the ground is warm enough. It
must be remembered that this plant originated in the Tropics and that
when transplanted to higher latitudes it makes its best growth in the
hottest weather. It is even more susceptible to cold and wet than is
Indian corn. Hence, proper delay in planting will permit economy
in the use of seed. Where the vines are grown for hay, the yield
will be larger if the seed is planted in drills and cultivated a time or
two. The yield of peas is also larger when only a moderate amount
of seed is sown and the vines have more space and light and air be-
tween them. It is also heavier from late-planted vines than from the
very early ones. In tests to determine the relative value of different
named varieties it has been found that, as a rule, those which make
the heaviest yields of vines also bear large crops of peas.

HARVESTING.

The vines should be mowed for hay when the peas are well formed
and the leaves are first beginning to turn yellow. After wilting on
the ground or in the windrows from twenty-four to forty-eight hours,
the hay is placed in small thin piles, or cocks, and allowed to cure
for several days, when it may be carted to the barn or stacked under
sheds. The haymaking process is a difficult one, requiring more care
and attention than in the case of red clover, because the broad leaves
and thick stems contain a larger amount of water. The hay must
be placed in cocks before the leaves become brittle, and the piles
must be small enough to allow free circulation of air to the center of
each. Bright cowpea hay, clean and well cured, is worth as much

as the best red-clover hay, and there is no good reason why the South-
ern farmers and planters should buy the Northern-grown article for
their working stock or for fattening their cattle. Every ton of hay
used on the estate should be grown there. Another method of cur-
ing hay is to stack the vines in a pen or rack of rails or poles so
arranged as to allow the air to enter every part of the pile. This
stacking over poles is the best where the vines are pulled, or where
the trailing and creeping sorts are used. The bush varieties are the
best for hay, because of the greater ease with which they may be
mowed and handled. They also hold their leaves better than the
ranker trailing sorts.

The majority of farmers harvest only enough seed of cowpeas to
plant again the next season. The ripe pods are picked by hand and
are stored in barrels until needed or are thrashed out by machine or
with flails on the barn floor during the winter. Sometimes, if the
crop is heavy enough to render it profitable, the vines are run through
an ordinary thrashing machine from which the concave and alter-
nate teeth of the cylinder have been removed. But a machine breaks
and bruises more of the seed than when the pods are first picked off
by hand. Fully 95 per cent of the seed placed upon the market is
hand picked. The yield per acre varies according to the varieties and
the method of cultivation. Eight to twelve bushels is a fair average
of the amount that can be obtained when the peas are planted in the
corn rows. Sown alone, broadcast or in drills, yields of from twenty
to thirty-five and even, in rare cases, fifty bushels are obtained. The
Black, Unknown, Red Ripper, Clay, and Calico varieties are all heavy
seed bearers. Lady and White Crowder are good for table use and
also yield well. The Black-eyed, Red Crowder, and Whip-poor-will
or Speckled are very widely cultivated and find ready sale. Those
which make the largest growth of vines for green manure, as a win-
ter soil mulch, for hay or soiling are the Unknown, Red Ripper,
Southdown, and Clay. Whip-poor-will, Black-eyed, White, and Red
Crowder ripen in from twelve to fourteen weeks, and hence are
adapted to cultivation farther north than the very late, but ranker
growing. Unknown, Wonderful, Red Ripper, Black, and Gourd vari-
eties. The New England and Lee ripen seed in from six to seven
weeks, and hence are the ones to recommend for summer-soiling
crops in the upper prairie region of the Mississippi A'^alley or anywhere
else that an early maturing cowpea is required. This is one of the
species of cultivated plants which is verj" readily modified by change
of habitat. Early and late maturing forms may be found of every
strain that has been in cultivation for any considerable time.

YIELD AND VALUE.

The yield of hay varies according to the fertihty of the soil upon
which it is raised, whether it is grown on rich lowlands or on the
drier and more sterile uplands. In the Gulf States cowpeas will
probably give an average yield of 2 to 3 tons per acre, while 4 to 6
tons are not uncommon. Farther north the average will range from
li tons in Ohio to 2^ tons in Arkansas, Missouri, and Tennessee. As
with other crops, the time of planting, the character of the soil and
of the cultivation, and the amount of rainfall have much to do with
the yield. Along the Gulf it is one of the best hay crops. North of
the latitude of the Ohio River it is chiefly valuable as an addition to
the list of drought-resistant, summer-soiling crops and as a crop that
will yield a considerable amount of forage on soil too sterile to grow
red clover. The commercial value runs from $G to $20 per ton, being-
governed by the relative abundance of other grades of hay and fod-
der. Its feeding value is equal to that of the best red clover, and
the hay ranks high in palatability and digestibility,

FERTILIZERS.

It has been found that, as a rule, it does not pay to use high-grade
commercial fertilizers on cowpeas; this, however, depends largely
on the soil and on what crop is to follow this green manurial one.
It is usually unprofitable to fertilize with expensive nitrogen, in the
form either of nitrate of soda or of guano, and even the organic
nitrogen of cotton-seed meal does not act upon this crop as rapidly
as upon cotton and the cereals. The nitrogen of the fertilizers seems
not to influence the percentage of protein in the crop, and the general
opinion of agriculturists in the South is that it does not cause a
sufficient increase in yield of vines to pay the cost. At the Delaware
Station 160 pounds of muriate of potash per acre doubled the yield of
vines, and superphosphate produced no effect. At the Georgia Sta-
tion combinations of superphosphate and potash gave the best results,
but later exjDeriments there indicated that large amounts of potash
are unprofitable, and that superphosphate at the rate of from 200 to
400 pounds per acre gave better results. Superphosphates are very
much preferable to untreated rock phosphate. The latter can be sold
at very much lower rates, and it remains to be seen whether it would
not be a profitable method to apply the soft phosphate to the cowpeas
for the benefit of the succeeding crop in the rotation, for it has been
found that the insoluble phosphoric acid of the untreated rock becomes
changed to forms available as plant food in the presence of large
amounts of decaying vegetable matter in the soil. If it is found that
this process can be relied upon, then the cowpea will have another

valuable quality added to it, namely, that of being able to change
into high-grade and more costly superphosphate the low-grade and
cheap but unavailable phosphoric acid of the untreated rock.

THE FEEDING VALUE OF COWPEAS.

The feeding value of cowpea vines is very high, as shown by both
feeding tests and chemical analyses. As hay the vines are more val-
uable than fed green for soiling purposes. A comparison with red
clover and alfalfa is made in the following table, a compilation' of
the averages of a number of analyses from various sources.

Feeding value of coicpeas compared with red clover and alfalfa.

Fresh or air-dry material.

Calculated to water-fi
substance.

ee

Article.

Num-
ber of
analy-
ses.

Wa-
ter.

Ash.

Pro-
tein.

Fi-
ber.

Nitro-
gen-
free ex-
tract.

Fat.

Ash.

Pro-
tein.

Fi-
ber.

Nitro-
gen-
free ex-
tract.

Fat.

Cowpeas :

Green

Hay

10

8

4.3
38

23

21

83.6
10.7

70.8
15.3

71.8
8.4

1.7

7.5

2.1
6.2

2.7

7.4

2.4
16.6

4.4
12.3

4.8
14.3

4.8
20.1

8.1
24.5

7.4
25.0

7.1
42.2

13.5
a3.6

12.3

42.7

0.04
2.9

1.1
3.9

1.0
2.2

10.5
8.5

7.2
7.3

9.4

8.1

14.3
18.6

15.3
13.5

17.1
15.6

29.0
22.5

27.8
31.3

26.2
27.3

43.6

47.2

45.8
43.0

43.9
46.6

2.6
3.2

Red clover :

Green

Hay

Alfalfa :

Green

Hay

3.9

4.9

3.4

2.4

A study of the percentages here given will show that the green
vines contain more water, less protein or nitrogenous, muscle-making
food and less of the fat-forming crude fibers, fats, and nitrogen-free
extracts than either the green alfalfa or red clover. The air-dry hay,
however, contains more protein than either of the others, less fiber,
more nitrogen-free extracts than the red clover, and more fat than
the alfalfa. As is the case with leguminous forage plants in general,
a ration of cowpeas, to be well balanced, requires the addition of
some other fodders, such as corn stover, sorghum, timothy, Bermuda,
or prairie hay, otherwise a portion of the protein will be wasted.

COWPEAS FOR FORAGE.

There is no forage plant better adapted to the needs and conditions
of Southern agriculture than this rank, free-growing annual. It will
thrive luxuriantly upon the rich, swampy cane lands of Louisiana.
On the driest and most sterile worn-out uplands it serves the admir-
able purpose of supplying a larger quantity and better quality of for-
age than any other bean or clover. And whenever a crop of cowpeas
has been taken off a field the surface soil is left richer by a good

^Handbook of Experiment Station Work, Appendix, 1893.

many pounds of that most costly of all plant foods, nitrogen. The
roots of the cowpea enter deeply into the soil, opening and loosening
it far down for the benefit of the roots of the succeeding crops of corn,
cotton, and tobacco. It has been found by experiment that the fer-
tilizing value of the roots and stubble of the cowpea are very con-
siderable, but not as great as that of the hay removed from the field.
The best and most economical use of this forage crop is, then, to cut
for hay, feed to stock, and return the stable manure to the soil.
Plowing the whole crop under is less remunerative because there is
much needless waste of the muscle-making and fat-forming con-
stituents of the plant which would bring more profit if turned into
beef, pork, wool, cheese, or butter.

As regards the disposal of the crop, there is a wide variation in
practice. The feeding value of vines and peas much exceeds their
fertilizing value. But as between the practice of turning the vines
under green in autumn and that of allowing them to lie on the
ground during the winter, the latter is undoubtedly sometimes to be
preferred, though theoretically wrong. Theoretically, to plow the
vines under in autumn will be to save all the available nitrogen and
convert the whole plant into himius. Practically, the turning under
of so large an amount of watery green herbage is highly injurious,
causing a too rapid decay and consequent "burning" or souring of
the soil. The upper soil layers, freshly stirred and mellowed in
autumn, lose more by leaching and washing than they do in an
unplowed field covered by its winter mulch of decaying herbage,
though in both cases there is a decided loss of fertility over what
would result by following the peas with a crop of rye, winter
wheat, the turf -forming winter oats, winter vetch, or crimson clover.
The yields of forage are better on rich soils than on poor ones, but
the beneficial effects upon the succeeding crop due to the growth of
cowpeas are not so marked in the former case as in the latter.

COWPEAS FOR SWINE AND CATTLE.

When cowpeas are planted for green manure, it is an excellent prac-
tice to turn hogs into the field about the time that the first peas are
ripening. Young pigs thrive amazingly on the succulent foliage and
well-filled pods, and the quality of the pork raised on such a healthful
and nutritious diet is very fine. This is a very profitable method of
fattening hogs or of preparing them for topping off with corn or
sorghum for market. An acre of ripening cowpeas will pasture from
fifteen to twenty hogs for several weeks, and the gain in fertility
from the droppings of the animals during that period will more than
counterbalance the fertilizing value of the forage eaten. The rapid

increase in weight will thus represent so much clear profit, and the
farmer is richer by half a ton or more of prime pork for every acre
planted. Chickens and turkeys also eat the ripe peas and do well
upon them. Cattle and horses are sometimes pastured on them, but
the safer and more economical way of feeding cowpea vines to such
stock is to cut or pull and feed partially wilted. There will be less
waste and destruction from trampling, and if each animal is given
only so much as it can eat clean the greatest economy as well as
greatest profit will result. Furthermore, cattle and sheep are liable
to bloat if allowed to eat too ravenously of cowpea vines or any other
rich and succulent forage, and by using it as a soiling crop the danger
may be more readily controlled and the loss prevented. The report has
been sent out from some of the Northern experiment stations, where
this forage plant is not ordinarily cultivated, that cattle will not eat
the green vines except after having been starved to it, and then only
sparingly. We have seen Western horses and ponies that would not
touch red clover or a grain ration of oats, and have heard of Eastern
stock that would not eat alfalfa hay. But these few adverse cases
do not prove that red clover, alfalfa, and oats are not good forage.
With the cowpea the case is similar. It is very rarely that any South-
ern planter reports that this forage is refused by any kind of stock.

COWPEAS FOR ENSILAGE.

Reports are very conflicting in regard to the value of this crop for
ensilage. There is much positive testimony both for and against, some
authorities stating that the quality is excellent and others that the
vines contain too much water, the product of the fermentation being
a slimy, foul-smelling mass, unfit for food for any kind of animals.
From reports on the subject it is to be believed that the attempt to
convert cowpea vines into good ensilage can not be made with such
uniform success as in the case of red clover. The percentage of water
in the tissues is too high, and the mechanical difficulties in the way
of running a mass of tangled herbage through the feed cutter are too
great. Special machinery would have to be constructed for the pur-
pose. Indian corn will probably remain for many years the best all-
around forage plant for this purpose. The consensus of opinion among
agricultural workers seems to be that ensilage made from any legume,
whether it be cowpeas, vetches, soy beans, alfalfa, or the clovers, does
not equal in feeding value good hay made from the same. Under
certain conditions that arise in the silo the crude protein is converted
into indigestible or insoluble nitrogenous compounds. The cowpea
or clover ensilage is then valuable only for the carbohydrates that it
contains, and either corn or sorghum is far superior to it.

COWPEAS AND SOIL RENOVATION,

A field of cowpeas has been very happily designated ' ' the poor
man's bank," for in common with all its leguminous congeners, the
field pea, clovers, alfalfa, and a score of others, this crop has the
power of increasing the fertility of the soil upon which it grows. This
fact has long been accepted by farmers and students of agriculture,
but until recent discoveries in Germany and America it was believed
that the chief function of these plants was to pump up nitrogen from
the subsoil reservoir to the surface by means of their long roots for
the use and benefit of succeeding crops.

But experiments in the field and laboratory for the purpose of
determining the causes of natural phenomena have taken the place
of class-room philosophy and speculative reasoning. Within the last
twenty years scientific workers have discovered that minute micro-
organisms, or bacteria, which live within the tissues of the roots of
leguminous plants take up free nitrogen from the gases in the soil,
just as the higher plants and animals utilize the oxygen of the air.
This nitrogen enters into combination to form nitric acid, which unites
with the mineral elements of the soil to form nitrates, a kind of plant
food exceedingly valuable to the growing crop. Nitrogen, when in
combination with other elements, is an indispensable form of plant
and animal food, but the free element can not be utilized by any
of the higher organisms. Small amounts of nitrous acid are formed
as a result of lightning discharges and are washed out of the air by
the rains, to be in part absorbed by the soil, and in part carried by
rivers and drainage waters into the sea. Free nitrogen exists only
in the air and in the gases of the soil, but as ammonia, nitrous and
nitric acid, nitrites and nitrates, it is present in varying quantities in
the soil, the unbroken rocks, and the waters of continents and oceans.

The most available purchasable nitrogen is obtained either as salt-
peter or nitrate of soda from the extensive deposits in the Peruvian
deserts, or from some form of animal wastes, such as freshly ground
bone, dried blood, guano, tankage, and fish scrap, and from cotton-
seed meal and other like by-products of the oil mills. These fertil-
izers are all expensive, so much so that they can be profitably employed
by the farmer only in extensive farming with specialized crops. The
gain in yield with low-priced crops, such as corn, cotton, tobacco,
cowpeas, and the grasses, using high-grade and costly fertilizers, is
not commensurate with the additional expense. But every farmer,
rich and poor, has over three thousand tons of atmospheric nitrogen
resting on every acre of his farm, a certain quantity of which can be
transformed into available plant food every time he grows a crop of
cowpeas, red clover, or alfalfa.

10

There are a great many acres of farming land in the South in need
of renovation. The red upland and j^ellow-clay soils were undoubtedly
less fertile originally than the alluvial and black prairie soils, and the
methods of cultivation which formerly prevailed have still further
diminished their productiveness. In the days when every plantation
numbered its acres hj the thousand and labor was cheap, the planter
could afford to clear off the native forest growth and bring fresh fields
into cultivation whenever the yields of cotton and tobacco fell below
what was considered a profitable figure. The old field, stripped in a
few years of its accumulated store of humus, was abandoned and
allowed to grow up to weeds and underbrush. The forest again spread
across it, and gradually, in the slow course of half a lifetime, the nat-
ural enrichment of its surface soil by the growth of the woodland
grasses made it ready for another robbery.

But with the breaking up of the large estates and the abrupt change
in the labor conditions this method of farming became no longer profit-
able or even possible. A planter with fewer acres could no longer
afford to await nature's slow process of rejuvenating the soil. A new
system of farming was necessary. The land must not be allowed to
"go back." It must be kept up to the highest state of productiveness
by a rotation of crops, a judicious use of commercial fertilizers, the
growth of nitrogen-fixing leguminous crops, and good and thorough
cultivation. To maintain the fertility of any soil the amount of humus
or decaying organic matter in it must be kept up. Take two soils of
as nearly as possible the same physical and geological formation, but
the one rich in humus and the other lacking it, and fertilize them with
equal quantities of commercial manures ; the one which has the most
organic matter in its composition will yield the largest crop. The
soil on that field will stand drought better, will wash less under tor-
rential rains, and be more friable and of better tilth. The average
soils of the South need more humus. It can be best supplied by
sowing more grass, more permanent pasture lands, more leguminous
crops. In a word, plant cowpeas.

SUMMARY.

The chief functions of this crop, then, are to furnish large amounts
of nitrogen abstracted from the air and fixed in the roots and stubble
in a conveniently available form for the use of succeeding crops ;
second, to produce a large yield of vines and peas rich in digestible
protein, which, either as hay or for soiling purposes, will take the
place of concentrated nitrogenous foods ; and, third, to supply humus,
which acts directly and indirectly to produce fertility by breaking
down and rendering available the basic minerals of the soil. The
fertilizing value of the nitrogen in the vines is entirely dissipated or

11

greatly diminished by weathering when they are left on the surface

of the field during the winter. Hence, to secure the full value, the

cowpeas should be fed and the stable manure returned to the field.

If the vines are plowed under in autumn, a winter forage crop, such

as winter oats, crimson clover, rye, or vetches, should be planted to

prevent the leaching and washing action of the winter rains.

Jaked G. Smith,

Assistant Agrostologist.
Approved :

James Wilson,

Secretary.

Washington, D. C, March 25, 1898.

O

^^^

Circular No. 6.— (Agros. 25.)

United States Departnfetrf of Agriculture,

DIVISION OF AGROSTOLOGY.

[(irass and Forage Plant Investigations.]

THE C'ULTIVATEI) VETCHES.

The demand for early spring forage plants is increasing in almost
every section of the United States. This demand arises from a
variety of causes, chief among which is the rapid increase of the
dairying industry. Soiling crops and pastures supply the desired
succulent forage from early summer until the first hard autumn
freeze, and ensilage and root crops tide over the early winter. With
good management such sul:)stitutes for green forage may be made to
last until the grass starts, but on too many American farms there is
a period of shortage of succulent feed in late winter and spring. To
bridge over this critical period, annual leguminous crops, such as
crimson clover and the vetches, are each year coming deservedly into
greater prominence. The vetches are nitrogen gatherers. Like the
clovers they have the property of absorbing through their roots the
free gaseous nitrogen of the air, which is present in all well cultivated
soils, and fixing a portion of it in a form which may become either a
fertilizer if left in the soil or a muscle-making element in the forage.
Nearly all leguminous forage crops have this property and hence
their great importance as green manure and soiling crops.

The vetches are also useful because they form a living mulch in
spring and early winter, shading the ground and preventing the
growth of weeds, thus retarding the constant loss of soluble plant
foods that is going on wherever a soil is left bare and unprotected
from the direct action of the elements. A mulch of weeds would
serve the same purpose, but all can see why it would be better to
have the winter cover consist of vetches or clovers rather than of a
tangle of noxious weeds.

In the Southern States vetches should be sown in the late summer
or early fall, so that they may be out of the way in time to plant the
next season's field crops. In the North or wherever the winters are
severe they may be sown in early spring at the same time as the
spring wheat or other small grain. They are all crops which delight
in cool growing weather like that of the northern spring or fall and
the southern winter.

HAIRY VETCH, OR SAND VETCH.

( Vicia villosa. )

This annual leguminous plant is a native of western Asia. It has
been cultivated for about fifty years in some parts of Europe, especially
southern Russia, Germany, and France, and was introduced into
this country for the first time about 1847, under the name of Siberian

2

vetch. Its cultivation was then neglected until its reintroduction
about twelve years ago by this Department. It has since been tried
in various parts of the United States. Excellent reports as to its
drought-resisting qualities and its adaptability to our climate have
been received from Washington, Nebraska, Georgia, New Mexico,
South Dakota, Minnesota, Montana, and Pennsylvania. It has been
grown on the experimental grounds of the Department of Agricul-
ture at Washington, D. C, and has proved to be thoroughly adapted

Fig. 1.— Hairy vetch or sand vetch {Vicia villom): o, cross-section of stem ; b, flower;

c, stamens ; d, pod.

to and valuable for this locality. The seed was sown about the 25th
of April, and the plants commenced to bloom the middle of July, con-
tinuing in flower until the end of November. Sown in August it grew
well until the first hard frost and continued an intermittent growth all
winter during periods of open weather. By the middle of March it
had formed a thick mat of vines over the soil. It blossomed by the

first of May and was then ready to be cut. Hairy vetch withstands
winter cold and summer drought, but it does not do well where there
is an excess of water in the soil. It is one of the most promising
fodder crops which has been brought into the United States in recent
years and by some is considered especially valuable for light sandy
soils.

CULTIVATION.

Hairy vetch (fig. 1) may be sown in autumn, from about the middle
of August to the middle of September ; or in spring, from the latter
part of April to the middle of May. It should be sown broadcast or
with a grain drill at the rate of 1 to 1^ bushels of seed per acre.
The latter method will require a less amount of seed. When the
seed is put in broadcast, a bushel of rye, oats or wheat should be
sown at the same time, so as to furnish a support to keep the vines
up off the ground. If it is sown in drills in the latter part of August
the crop should be cultivated several times. It will furnish some
forage in autumn, and where the winter is not too severe will start
to grow again in the spring, thus producing forage in late autumn
and early spring, at the two periods when it is most needed.

While it gives a fair crop on poor soil it is most profitable as a
forage plant on rich and well- tilled land. It needs considerable
moisture during the first six weeks of its growth, but when once
fairly established withstands drought and extremes of temperature.
The seeds germinate poorly when they are more than two years old.
Most of the seed used in this country is imported from Europe, so
that particular care ought to be taken by importers and dealers to
handle none but such as can be sold under guarantee as good, fresh
seed.

At the Mississippi Agricultural Experiment Station seed of this
vetch was sown in October, 1888, and since that time has given
heavy annual crops on the same ground, although receiving no
attention. Its seeds germinated with the first autumn rains, and
covered the ground ])y the first of January, furnishing good grazing
until April or May. If the stock is taken oft' the field in March the
plants will mature and reseed the ground freely for the next year.

FORAGE VALUE.

Hairy vetch is eaten with relish by all kinds of stock. If properly
cured it makes good hay, though on account of its habit of growth
the process is difficult. It has been tested in the silo in alternate
layers with green corn and also alone. The former method is the
one to be used if the best ensilage is desired. It is a most excellent
forage plant for soihng purposes. On account of the difficulty of
cutting it properly, it will give the most satisfactory results fed in
this way.

The seed is as yet very expensive, being about $4 per bushel of 70
pounds. As soon as its cultivation has increased to such an extent
that the seed may be obtained at prices less prohibitive, this vetch
will undoubtedly occupy a permanent place in American agriculture.

SPRING VETCH, OR TARES.
( Vicia sativa. )

Spring vetch is a leafy, annual, trailing herb, 1 to 2 feet high, with 4-
to 5-angled stems, simple or branched from the base. The leaves are
compound and are terminated with 3 or 4 tendrils. The 5 to 7 pairs
of leaflets are broadest above the middle, blunt or notched at the end,
and tipped with an abrupt point (fig. 2) . The flowers are rather large,
deep purple, one or two together in the axil of the leaf on a very short
stalk. The plant is soft and hairy all over.

HISTORY.

This old-world forage plant has
been cultivated in Europe for
nearly twenty centuries. It is a
native of western Asia and of all
Europe except Lapland. It was
cultivated by the Romans, and was
esteemed by them a valuable fod-
der crop. In Italy it has been
grown continuously up to the pres-
ent day. It is one of the many
soiling crops in use in northern
Europe and the British Isles.
Spring vetches were introduced
into the United States in a casual
way nearly a hundred years ago.
They have been tried in nearly all
the States and have proved very
unsatisfactory, except for certain
districts in New England, New
York, northern Michigan, Wiscon-
sin , and lower Canada. They were
very largely grown throughout the
New England and Northern States
during the period from 1865 to
1885, but their cultivation there
has now almost ceased, it having
been found that the yield of hay or of green fodder is not a profitable
one compared with that of the red and crimson clovers and field peas.
The principal drawback to their more extensive cultivation is the high
price of the seed ($2.50 to $3 per bushel of 70 pounds) and the fact
that they can not withstand even temporary drought or hot weather.
In England, where they are extensively used, the growing season is
much cooler, with more rain and an equable temperature.

Fig. 2.-

-Spring vetch or tares (Vina aativa):
a, pod.

CULTIVATION.

Spring vetches seem to be adapted more particularly to northern
countries, where the season is short and the rainfall abundant.
The seed should be sown at the rate of 5 to 8 pecks per acre, with
one bushel of rye or oats as a nurse crop. As high a seeding as 3
to 3i bushels per acre is sometimes recommended, but the product

per acre will not warrant the use of so much seed at the present high
prices. Vetches should be sown in April or May. They will be
ready to cut by the middle of June or the first of July, from full
bloom until the pods are half formed. When sown alone the vines
lodge and make a dense mat, and the object of the nurse crop is to
furnish a support to lift the vines up off the leaves. The vines are
very difficult to harvest when sown alone, on account of the tangled
mass of stems, but may be easily cut with a pea harvester. An
acre of vetch and oats yields ordinarily from 6 to 8 tons of green
forage. Where it can be grown,, its chief value arises from the fact
that it is ready to cut between the first and second crops of red clover,
thus filling a gap in the series of early summer soiling crops. Spring
vetches are also used for hay. To make hay, more care is required
than with red clover. Two crops are sometimes cut in one season,
and where this is possible the second is the one to be saved for seed.
The first crop ripens very irregularly, and some of the pods will be
shelled before the rest are ripe. Where they can be grown they are
a very good summer feed for horses, but must not be fed earlier
than full bloom, on account of their diuretic action. They are good
for soiling sheep and milch cows, and are said to very materially
increase the flow of milk.

FEEDING VALUE.

The percentage of digestibility of spring vetch forage has not been
determined in this country, but analyses show a high food content
comparable with alfalfa rather than the clovers. The average sample
of vetch hay contains 11.3 per cent water, 7.9 per cent ash, 17 per
cent crude protein, 25.4 per cent fiber, 3G.1 per cent nitrogen-free
extract, and 2.3 per cent fat. The flat pea and the soy bean are
the only leguminous fodders which exceed this in the crude protein
content.

FERTILIZING VALUE.

At the time when ready to cut for hay the vetch contains about
20 percent of dry matter,^ and in this 20 per cent there are contained
3.10 per cent nitrogen, 0.72 per cent phosphoric acid, and 3.30 per
cent potash. Calculating on this basis the fertilizing ingredients con-
tained in a crop of twelve tons of green forage produced from one
acre there would be 153 pounds of nitrogen, 37 pounds of phosphoric
acid, and 103 pounds of potash. In addition to this the stubble and
roots to the depth of 22 inches would contain 27.2 pounds of nitrogen,
7.2 pounds of phosphoric acid, and 21.8 pounds of potash, making a
total of 180.2 pounds of nitrogen produced in a single season by one
crop of spring vetches, or as much as is contained in 1 8 tons of barn-
yard manure.

Spring vetches are not recommended as a forage crop for general
cultivation. They have value for some few northern localities, but
have proved a signal failure elsewhere in this country. The plants
come into flower very unevenly, so that sometimes the seed does not
ripen in sufficient quanties at one time to pay for harvesting. The
crop is liable to injury by drought and excessive heat,

'Third Ann. Rept. Conn. (Storrs) Agr. Expt. Sta. (1890).

WINTER VETCH.

(Latliyms Mrsutus. )

Winter vetches (fig. 3) were introduced into the United States from
Italy, where they are grown quite extensively as a winter soiling
crop. They are of value only in the Southern States and have not
heen found hardy anywhere north of the latitude of Washington, D. C.
Winter vetches are very similar in hahit and manner of growth to
the spring vetches or tares. They are trailing, vine-like plants that
grow in dense masses. The stems are narrowly wing-margined ; the
narrow leaflets are in pairs with a tendril arising between them ; the

inch-long pods are quite hairy; and the
rounded, dark-brownish seeds appear warty
under a lens.

WINTER VETCHES FOR THE SOUTH.

Winter vetches should be sown broadcast
in August or September at the rate of two
bushels of vetch and one bushel of winter
rye or winter oats per acre. If sown in the
latter part of August they furnish a bite of
green forage in November and December,
at a time when it is particularly desirable,
and can be cut for hay in the early spring.
Winter vetches sown in February in the
Gulf States provide a supply of green for-
age in April or May. The plant deserves
to be more widely cultivated as it is valuable
both as a soiling and a hay crop. It makes
its best growth in the spring and autumn,
when the weather is cool. Winter vetches
thrive on any soil which will grow cowpeas,
provided that it is not too wet. Its cultiva-
tion has been very successful in all portions
of the South where it has been tried, and
particularly so in central Georgia and Ala-
bama. The winter vetch is desirable as an
addition to our list of forage plants, because
-wintCT jetch (Lathiinis j^ lengthens out the soiling season, and fur-
nishes green foliage late in autumn and very
early in spring, during two periods of scanty vegetation. Winter
vetch should be cut for hay when in full bloom. Considerable care
is required to get it into the stack or barn without its heating. Any-
one who can make good cowpea or alfalfa hay can successfully han-
dle winter vetch.

Fig. 3.

KIDNEY VETCH.

{Anthyllis vulneraria. )

The kidney vetch (fig. 4) is a perennial leguminous plant which is
found wild over a large part of Europe. It grows naturally along
roadsides, wherever the soil is dry and thin and the subsoil calcareous.

'^

It was first introduced into cultivation by a German peasant about
40 years ago. This farmer noticed that the vetch grew on the dry
calcareous soils of hillsides, in places too poor to support even white
clover. He gathered a few seeds, sowed them the next year, and
kept on saving and sowing the seed
until he had enough to plant quite a
large field. From this small begin-
ning the cultivation of the kidney
vetch has spread throughout northern
Germany and many foreign coun-
tries, and to the United States.

CULTIVATION.

In Germany the custom is to sow
the seed in autumn at the rate of 18
to 22 pounds per acre, with oats, bar-
ley, or other small grain as a nurse
crop. Sometimes it is sown alone in
the spring. The product of the first
year is very small, so that it is only
a profitable crop when it is sown with
grain, in order that some income may
be derived from the land during that
time. The second year the vetch
throws up large stems that often
make a growth 3 or 4 feet high.

The yield of hay is quite small, gen-
erally not more than one cutting per
season, and perhaps a ton or a ton
and a half of hay per cutting. It is
cut in full bloom, and cured in about
the same way as red clover. Two
crops may l)e secured in one season
by cutting the first before the plant
blossoms, but usually the aftermath,
consisting entirely of root leaves, is
depastured and no attempt is made
to get more than one crop of hay.

Kidney vetch is not recommended
for sowing in the United States, ex-
cept on poor, thin, calcareous or very
sandy soils, which are too sterile to
support the red or crimson clovers, or any of the better forage crops.
It has been tried at a number of tlie experiment stations throughout
the United States, but has been reported as of small value.

Jared G. Smith,

Fig. 4.

-Kidney vetch {Anthyllis vuUier-
aria): a, flower.

Assistant Agrostologist.

Approved :

James Wilson,

Secretary of Agriculture.

Washington, D. C, April 5, 1898.

Circular No. 6.— Revised. (Agros.

United States Department of Agricultnre,

DIVISION OF AGROSTOLOGY.

[Grass and Forage Plant Investigations.]

THE €ULTIVATT:D VETCHES.

The demand for early spring forage plants is increasing in almost
every section of the United States. This demand arises from a
variety of causes, chief among which is the rapid increase of the
dairying industry. Soiling crops and pastures supply the desired
succulent forage from early summer until the first hard autumn

Fig. 1.— Hairy vetcli: Plot in (irass Garden, U. S. Department of Agriculture.

freeze, and ensilage and root crops tide over the early winter. With
good management such substitutes for green forage may be made to
last until the grass starts, but on too many American farms there is
a period of shortage of succndent feed in late winter and si)ring. To
bridge over this critical period, annual leguminous crops, such as
crimson clover and the vetches, are each year coming deservedly into
greater prominence. The vetches are nitrogen gatherers. Like the
clovers they have the property of absorbing through their roots the
free gaseous nitrogen of the air, which is present in all well cultivated

2

soils, and fixing a portion of it in a form which may become either a
fertihzer if left in the soil or a muscle-making element in the forage.
Nearly all leguminous forage crops have this property and hence
their great importance as green manure and soiling crops.

The vetches are also useful because they form a living mulch in
spring and early winter, shading the ground and preventing the
growth of weeds, thus retarding the constant loss of soluble plant
foods that is going on wherever a soil is left bare and unprotected

Fig.

-Hairy vetch or sand vetch {Vici<i riUom): a, cross-section of stem ; /), flower ;

c, stamens ; d, pod.

from the direct action of the elements. A mulch of weeds would
serve the same purpose, but all can see wh}" it would be better to
have the winter cover consist of vetches or clovers rather than of a
tangle of noxious weeds.

In the Southern States vetches should be sown in the late summer
or early fall, so that they may be out of the way in time to plant the

next season's field crops. In the North or wherever the winters are
severe they may be sown in early spring at the same time as the
spring wheat or other small grain. They are all crops which delight
in cool growing weather like that of the northern spring or fall and
the southern wnnter.

HAIRY VETCH, OR SAND VETCH.
( Vicia villosa. )

This annual leguminous plant is a native of western Asia. It has
been cultivated for about fifty years in some parts of Europe, especially
southern Russia, Germany, and France, and was introduced into
this country for the first time about 1847, under the name of Siberian
vetch. Its cultivation was then neglected until its reintroduction
about twelve years ago by this Department. It has since been tried
in various parts of the United States. Excellent reports as to its
drought-resisting qualities and its adaptability to our climate have
been received from Washington, Nebraska, Georgia, New Mexico,
South Dakota, Minnesota, Montana, and Pennsylvania. It has been
grown on the experimental grounds of the Department of Agricul-
ture at Washington, D. C, and has proved to be thoroughly adapted
to and valuable for this locality. The seed was sown about the 25th
of April, and the plants commenced to bloom the middle of July, con-
tinuing in flower until the end of November. Sown in August it grew
well until the first hard frost and continued an intermittent growth all
winter during periods of open weather. By the middle of March it
had formed a thick mat of vines over the soil. It blossomed by the
first of May and was then ready to be cut. Hairy vetch withstands
winter cold and summer drought, but it does not do well where there
is an excess of water in the soil. It is one of the most promising
fodder crops which has been brought into the United States in recent
years and by some is considered especially valuable for light sandy
soils.

CULTIVATION.

Hairy vetch (fig. 2) may be sown in autumn, from about the middle
of August to the middle of September ; or in spring, from the latter
part of April to the middle of May. It should be sown broadcast or
with a grain drill at the rate of 1 to li bushels of seed per acre.
The latter method will require a less amount of seed. When the
seed is put in broadcast, a bushel of lye, oats or wheat should be
sown at the same time, so as to furnish a support to keep the vines
up off the ground. If it is sown in drills in the latter part of August
the crop should be cultivated several times. It will furnish some
forage in autumn, and where the winter is not too severe will start
to grow again in the spring, thus producing forage in late autumn
and early spring, at the two periods when it is most needed.

While it gives a fair crop on poor soil it is most profitable as a
forage plant on rich and well-tilled land. It needs considerable
moisture during the first six weeks of its growth, but when once
fairly established withstands drought and extremes of temperature.
The seeds germinate poorly when they are more than two years old.
Most of the seed used in this country is imported from Europe, so that
particular care ought to be taken by importers and dealers to handle
none but such as can be sold under guarantee as good, fresh seed.

At the Mississippi Agricultural Experiment Station seed of this
vetch was sown in October, 1888, and since that time has given
heavy annual crops on the same ground, although receiving no
attention. Its seeds germinated with the first autumn rains, and
covered the ground bj- the first of January, furnishing good grazing
until April or May. If the stock is taken off the field in March the
plants will mature and reseed the ground freely for the next year.

FORAGE VALUE.

Hairy vetch is eaten with relish
by all kinds of stock. If properly
cured it makes good hay, though
on account of its habit of growth
the process is difficult. It has been
tested in the silo in alternate layers
with green corn and also alone.
The former method is the one to be
used if the best ensilage is desired.
It is a most excellent forage plant
for soiling purposes. On account
of the difficulty of cutting it prop-
erly, it will give the most satisfac-
tory results fed in this way.

The seed is as yet very expensive,
being about $4 per bushel of 70
pounds. As soon as its cultivation
has increased to such an extent that
the seed may be obtained at prices
less prohibitive, this vetch will un-
doubtedly occupy a permanent
place in American agriculture.

SPRING VETCH, OR TARES.

( Vicia sativa. )

Spring vetch is a leafy, annual,
trailing herb, 1 to 2 feet high, with 4- to 5-angled stems, simple or
branched from the base. The leaves are compound and are termi-
nated with :> or 4 tendrils. The 5 to 7 pairs of leaflets are broadest
above the middle, blunt or notched at the end, and tipped with an
abrupt point (fig. 3). The flowers are rather large, deep purple, one
or two together in the axil of the leaf on a very short stalk. The
plant is soft and hairy all over.

HISTORY.

This old-world forage plant has been cultivated in Europe for
nearly twenty centuries. It is a native of western Asia and of all
Europe except Lapland. It was cultivated by the Romans, and was
esteemed by them a valuable fodder crop. In Italy it has been grown
continuously up to the present day. It is one of the many soiling
crops in use in northern Europe and the British Isles. Spring vetches
were introduced into the United States in a casual way nearly a
hundred years ago. They have been tried in nearly all the States
and have proved very unsatisfactory, except for certain districts in

Fig. 3.

-Spring vetch or tares {Vkia satira):
rt, pod.

New England, New York, northern Michigan, Wisconsin, and lower
Canada. They were very largely grown throughout the New Eng-
land and Northern States during the period from 1805 to 1S85, but
their cultivation there has now almost ceased, it having been found
that the yield of hay or of green fodder is not a profitable one com-
pared with that of the red and crimson clovers and field peas. The
principal drawback to their more extensive cultivation is the high
price of the seed ($2.50 to $3 per bushel of 70 pounds) and the fact
that they can not withstand even temporary drought or hot weather.
In England, where they are extensively used, the growing season is
much cooler, with more rain and an equable temperature.

CULTIVATION.

Spring vetches seem to be adapted more particularly to northern
countries, where the season is short and the rainfall abundant.
The seed should be sown at the rate of 5 to 8 pecks per acre, with
one bushel of rye or oats as a nurse crop. As high a seeding as 3
to 3i bushels per acre is sometimes recommended, but the product
per acre will not warrant the use of so much seed at the present high
prices. Vetches should be sown in April or May. They will be
ready to cut by the middle of June or the first of July, from full
bloom until the pods are half formed. When sown alone the vines
lodge and make a dense mat, and the object of the nurse crop is to
furnish a support to lift the vines up off the leaves. The vines are
very difficult to harvest when sown alone, on account of the tangled
mass of stems, but may be easily cut with a pea harvester. An
acre of vetch and oats yields ordinarily from G to 8 tons of green
forage. Where it can be grown, its chief value arises from the fact
that it is ready to cut between the first and second crops of red clover,
thus filling a gap in the series of early summer soiling crops. Spring
vetches are also used for hay. To make hay, more care is required
than with red clover. Two crops are sometimes cut in one season,
and where this is possible the second is the one to be saved for seed.
The first crop ripens very irregularly, and some of the pods will be
shelled before the rest are ripe. Where they can be grown they are
a very good summer feed for horses, but must not be fed earlier
than full bloom, on account of their diuretic action. They are good
for soiling sheep and milch cows, and are said to very materially
increase the flow of milk.

FEEDING VALUE.

The percentage of digestibility of spring vetch forage has not been
determined in this country, but analyses show a high food content
comparable with alfalfa rather than the clovers. The average sample
of vetch hay contains 11.3 per cent water, 7.9 per cent ash, 17 per
cent crude protein, 25.4 per cent fiber, 30.1 per cent nitrogen-free
extract, and 2.3 per cent fat. The flat pea and the soy bean are
the only leguminous fodders which exceed this in the crude protein
content.

FERTILIZING VALUE.

At the time when ready to cut for hay the vetch contains about
20 percent of dry matter,^ and in this 20 per cent there are contained

'Third Ann. Kept. Conn. (Storrs) Agr. Expt. Sta. (1890).

3.16 per cent nitrogen, 0.72 per cent phosphoric acid, and 3.36 per
cent potash. Calculating on this basis the fertilizing ingredients con-
tained in a crop of twelve tons of green forage produced from one
acre there would be 153 pounds of nitrogen, 37 pounds of phosphoric
acid, and 103 pounds of potash. In addition to this the stubble and
roots to the depth of 22 inches would contain 27.2 pounds of nitrogen,
7.2 pounds of phosphoric acid, and 21.8 pounds of potash, making a
total of 180.2 pounds of nitrogen produced in a single season by one
crop of spring vetches, or as much as is contained in 18 tons of barn-
3"ard manure.

Spring vetches are not recommended as a forage crop for genera
cultivation. They have value for some few northern localities, bu

have proved a signal failure elsewhere ii
this countrj^. The plants come into flower
very unevenly, so that sometimes the seed
does not ripen in sufficient quanties at one
time to pay for harvesting. The crop is liable
to injury by drought and excessive heat.

W^INTER VETCH.
{Lathyrus liii'sidus. )

Winter vetches (fig. 4) were introduced
into the United States from Italy, where
they are grown quite extensively as a win-
ter soiling crop. They are of value only in
the Southern States and have not been found
hardy anywhere north of the latitude of
Washington, D. C.

Winter vetches are very similar in habit
and manner of growth to the spring vetches
or tares. They are trailing, vine-like plants
that grow in dense masses. The stems are
narrowly wing-margined ; the narrow leaf-
lets are in pairs with a tendril arising be-
tween them; the inch-long pods are quite
hairy ; and the rounded, dark - brownish
seeds appear warty under a lens.

WINTER VETCHES FOR THE SOUTH.
Fig. 4.— Wintei' vetch (Lathyrus

Winter vetches should be sown broadcast
in August or September at the rate of two bushels of vetch and one
bushel of winter rye or winter oats per acre. If sown in the latter
part of August they furnish a bite of green forage in November and
December, at a time when it is particularly desirable, and can be
cut for hay in the early spring. Winter vetches sown in February
in the Gulf States provide a supply of green forage in April or May.
The plant deserves to be more widely cultivated as it is valuable both
as a soiling and a hay crop. It makes its best growth in the spring
and autumn, when the weather is cool. Winter vetches thrive on
any soil which will grow cowpeas, provided that it is not too wet.
Its cultivation has been very successful in all portions of the South
where it has been tried, and particularly so in central Georgia and
Alabama. The winter vetch is desirable as an addition to our list of

forage plants, because it lengthens out the soiHng season, and fur-
nishes green foliage late in autumn and very early in spring, during
two periods of scanty vegetation. Winter vetch should be cut for
hay when in full bloom. Considerable care is required to get it into
the stack or barn without its heating. Anyone who can make good
cowpea or alfalfa hay can successfully handle winter vetch.

KIDNEY VETCH.
(Anthyllis vulneraria. )

The kidney vetch (fig. 5) is a perennial leguminous plant which is
found wild over a large part of Europe. It grows naturally along
^badsides, wherever the soil is dry and
thin and the subsoil calcareous. It
was first introduced into cultivation
by a German peasant about 40 years
ago. This farmer noticed that the
vetch grew on the dry calcareous soils
of hillsides, in places too poor to sup-
port even white clover. He gathered
a few seeds, sowed them the next year,
and kept on saving and sowing the
seed until he had enough to plant quite
a large field. From this small begin-
ning the cultivation of the kidney
vetch has spread throughout northern
Germany and many foreign coun-
tries, and to the United States.

CULTIVATION.

In Germany the custom is to sow
the seed in autumn at the rate of 18
to 22 pounds per acre, with oats, bar-
ley, or other small grain as a nurse
crop. Sometimes it is sown alone in
the spring. The product of the first
year is very small, so that it is only
a profitable crop when it is sown with
grain, in order that some income may
be derived from the land during that .
time. The second year the vetch
throws up large stems that often •
make a growth 3 or 4 feet high.

The yield of hay is quite small, gen-
erally not more than one cutting per
season, and perhaps a ton or a ton
and a half of hay per cutting. It is
cut in full bloom, and cured in about
the same way as red clover. Two fig..5.
crops may be secured in one season
by cutting the first before the plant blossoms, but usually the after-
math, consisting entirely of root leaves, is depastured and no attempt
is made to get more than one crop of hay.

-Kidney vetch (Aiitlnjllis mdncr-
aria): n, flower.

8^

Kidnej^ vetch is not recommended for sowing in the United States,
except on jjoor, thin, calcareous or verj" sandy soils, which are too ster-
ile to support the red or crimson clovers, or any of the better forage
crops. It has been tried at a number of the experiment stations
throughout the United States, but has been reported as of small value.

DAKOTA VETCH.

{Lotus americanus, )

One of the most valuable leguminous plants of the vetch family
is Dakota vetch {Lotus americanus). It is a bushy annual found

throughout the Rocky Mountain
region of the Northwest, being
most abundant upon the sandy
river bottoms. Stock of all kinds
are fond of it and it is used either
as pasturage or hay. In the upper
Missouri region it is one of the
most highly prized native forage
plants. Being an annual it must
be permitted to mature at least a
portion of its seed and consequently
should not be grazed too closely or
cut too early. The blooming sea-
son covers a considerable period, so
that buds, flowers and both mature
and green pods are often found
upon a plant at the same time. As
a rule many of the seeds are rip-
ened before the haying season ar-
rives and it is a common practice
among ranchers to use hay racks
with tight bottoms in order to save
the shattered seed, which later is
scattered or sown over the meadows.
The plant reseeds itself freely and
when once established voluntary
Fig. c— Dakota vetch (Littm aincricaims). crops are assured irom year to year.

BIRD VETCH.

{Vicia cracca).

Bird vetch is a perennial species from Europe, where it is culti-
vated, and is used both for soiling and for hay. It is recommended
for cultivation in low meadows and especially in open woodlands. In
Germany it is often found in barren sheep pastures and is there
deemed especiallj^ valuable on account of its nutritive value, the fod-
der being especially rich in protein.

Jared G. Smith,
Approved: Assistant Agrostologist.

James Wilson,

Secretary of Agriculture.

Washington, D. C, April 15, 1898.

V

y^3^ ^ f^ y

CiRcrLAR No. 7.-(Agros,X) • ■ •^^\P'J Y O ti Vi

United ^ates

C A

t:\V

Ae^riculture,

DIVIsTo^Cdl^^ApR'OSTOLOGY.

[Grass and Forage Plant Investigations.]

GRAM, CHICK-PEA, OR IDAHO PEA.*

HISTORY.

Gram, Idaho Pea, or Chick- Pea {Cicer arietinum), has been in
cultivation in eastern countries longer than any other leguminous
crop. Its native country is unknown, but it is supposed to have
originated in Asia Minor. It was one of the plants of the Greeks at
the time of Homer, and occupied a prominent place among the food
plants of the lower classes during the time of the Roman Empire.

From Europe it was intro-
duced in comparatively re-
cent times into India, where
it is estimated that there are
now over 5,000,000 acres de-
voted to its cultivation either
alone or as a by-crop with
wheat. Next to the cereals,
gram forms the largest part
of the food used in India
and in portions of northern
Africa, Spain, and other
countries bordering on the
Mediterranean.

DESCRIPTION.

This plant is a branching
annual, with many upright
stems from the same root.
The leaves resemble those of
the vetch, having seven pairs
of small leaflets. These are
oblong, soft-hairy all over,
one-half inch long or less, and
sharply toothed on the margins
the axils of the leaves, on short stalks about one-half inch long.

*An article has been going the rounds of some of the agricultural papers and
the newspapers concerning a new forage plant, the Idaho pea, which is said to
have been a native of the northern Rocky Mountains, first inti-oduced into cul-
tivation less than twenty-five years ago. Its cultivation has been highly recom-
mended in the West, on account of the value of its seeds both as horse feed and
as a coffee substitute.

Fig. 1.— Gram (Clcer arietinum).

The flowers are borne singly in

2

The pods are bladdery, inflated, from one-half to three-fourths of
an inch long, and finely pubescent with glandular hairs. Each pod
contains one, or very rarely two, large seeds, which are wrinkled
and bear a fanciful resemblance to a ram's horn, whence the Latin
name arietinum. The seeds are a Httle larger than those of the
common garden pea, to which they are quite similar.

USE AS AN ADULTERANT.

This crop is cultivated in Mexico, where it is known by the Spanish
name " Garabanza." It is there used to some extent as food, but is
considered inferior to corn. Gram has been introduced into this
country at various times since 1864 as a substitute for coffee, but it
should be remembered that there is no such thing as a coffee substi-
tute. Various peas and beans, chickory, or even rye flour or bread
crumbs, may be roasted and prepared in the same manner as coffee, but
the beverage has none of the stimulating qualities of that drink and
only resembles it in color and to some extent in taste. It can be
used as an adulterant of coffee. The Idaho pea was cultivated at
the Colorado Experiment Station in 1895 and 1896. Professor Cooke
states that it "has demonstrated its ability to make a large growth
with plenty of water and a fair growth with a very limited supply.
It belongs to the pea family and is grown in rows, 30 inches apart,
and the plants 6 to 12 inches apart in the rows. Its growth indi-
cates that it can be raised for about 1 cent per pound."

SEED PER ACRE— GENERAL CLIMATIC CONDITIONS NECESSARY.

About 30 to 50 pounds of seed are used per acre, depending upon
whether it is sown in drills or broadcast. In India the largest acre-
age is in the Northwest Provinces, where the soils are. similar to
those west of the one hundredth meridian, and the climate is much
like that of New Mexico and Arizona. All authorities agree that it
is better suited to arid and semiarid regions than to humid ones, the
crop apparently requiring a great many sunny days during its season
of growth. Better results are secured in growing it with irrigation
than without, although it makes a fair yield on comparatively dry
soils. If continued experiments with this plant in the West prove
that its average yield is as high as has been claimed, it will undoubt-
edly prove a valuable addition to the list of forage plants suitable to

semiarid regions.

TIME TO SOW^ SEED.

There are a number of varieties, which differ fropi one another in
the color of the seed and length of season required for maturity.
The forms which have been cultivated in Spain, Mexico and the
Northwest Provinces of India are liable to prove more adaptable to
American conditions than those from subtropical India. The gram
plant is very sensitive to cold. The seed should be sown not earlier

than May 15, or at the higher altitudes about the Ist of June, and,
if some of the short-season varieties are procured, there will be
less danger of their being caught by early frosts. Gram is grown
in India as a winter crop. The seed is sown there in October or
November and the crop ripens in February, March, or April, accord-
ing to the portion of the country in which it is grown. It is said to be
adapted to almost any soil, from Ught sandy to heavy clays or loams,
apparently preferring the latter. It might prove of some value in
parts of the Southern States as a winter crop and soil cover, on lands
which are unsuited to the vetches and the crimson clover. It requires
only moderate amounts of moisture and is said to be injured by pro-
longed cloudy weather or abundant rains, which cause it to flower
prematurely and thus materially affect the yield of seed.

AS A SOIL RENOVATOR.

Gram is one of the leguminous forage crops which has the power
of absorbing gaseous nitrogen from the air, thus adding to the stores
of nitrogen in the soil, and though the fact of nitrogen-absorption by
leguminous plants has not been well understood until recent years,
this has long been considered one of the best Indian crops for soil
renovation and improvement. Watt states that over a large portion
of India gram is grown either to check weeds or as green manure,
and where this is the main object, only enough seed is saved to sow
again the coming season.

FOR FEEDING PURPOSES.

Gram is a staple article of horse feed in India. The seed is also
highly valued for fattening sheep and cattle. There is a considerable
trade with England and other foreign countries where the peas are
used for the same purposes. The total exports amounted in 1887 to
over 15,000 tons. The average analyses of the seeds show that they
contain about' 20. 5 per cent crude protein, 3.9 per cent fat, and 50.4
carbohydrates, having approximately the composition of the seeds of
the field pea commonly grown in the Northern States. Digestion
experiments have not been made with them but their fattening quali-
ties in use show them to be fully as valuable as the seeds of many
of the other legumes.

Besides serving as a fattening ration for cattle and sheep the seeds
and different parts of the plant find many uses among the natives of
India. The green peas are eaten as a vegetable. The meal is used
for porridge ; and the parched peas are used either in the prepara-
tion of a beverage or in various confections and candies. The young
plants are eaten as a salad and sometimes cooked like spinach.

OTHER USES — POISONOUS QUALITIES.

The leaves of the gram are viscid with a secretion which contains
oxalic, acetic, and mahc acids, the first of these predominating. In

India the secretion is collected by means of cloths spread over t'le
plant at night and wrung out in the morning when wet with d^ ;V,
The solution thus obtained is used in the preparation of cooling drinks
and also finds sale as a vinegar. The forage is said to be actually
poisonous to horses on account of the excess of oxalic acid in the
leaves. Cattle eat it, but it often proves injurious to them, although
to a less extent than to horses. However, this crop is not ordinarily
grown as a forage crop, but for the seeds, and the seeds alone are
used in India for feeding purposes.

PRODUCTION PER ACRE.

Gram has been grown experimentally at the grass garden in "Wash-
ington, D. C, and seed has also been distributed to a limited number
of farmers in various parts of the country. The reports concerning it
were not very favorable except from some parts of Colorado, Wyo-
ming, Utah, Idaho, and Montana. In the grass garden of the
Department of Agriculture it did not grow higher than from 8 to 10
inches and the seed production was very moderate. The newspaper
reports of this pea grown in the Rocky Mountain regions mention
yields estimated at the rate of 90 bushels to the acre, but this is very
unusual. The average crop in India is about 10 bushels to the acre
and the highest yields do not exceed 25 bushels, the latter only when
grown on the best soils under the most favorable conditions.

In common with other leguminous crops gram is dependent on the
presence of certain organisms in the soil for its ability to absorb nitro-
gen from the air and it is possible that the lack of success with this
crop in the Eastern States has been due to the absence of the partic-
ular tubercle-forming bacteria which are parasitic upon this plant.
To determine finally upon this point importations could be made of
soil upon which this crop has been grown, or pure cultures could be
secured of the organism itself and the soil or seed inoculated before
planting, as has been so successfully done in the case of the hairy
vetch at the Alabama Experiment Station. If this crop will yield
even 25 bushels per acre during average seasons, it will prove a val-
uable addition to Western forage crops on account of the high feeding
value of its seeds, but until fuller and more extended trials have been
made we would advise Western farmers to be cautious of investing
too much money or devoting too much land to the cultivation of the
Idaho pea, chick-pea, or gram.

Jared G. Smith,

Approved : Assistant Agrostologist.

James Wilson,

Secretary of Agriculture.
Washington, D. C, November 18, 1898.

Cll OLAR No. 8. (Agros. 37.)

United States Department of Agriculture,

DIVISION OF AGROSTOLOGY.

[Grass and Forage Plant Investigations.]

EXPERIMENTS IN RANGE IMPROVEMENT.

The chief problem in the cattle regions of the Southwest is, How
shall we restore or bring back the grasses on lands where they have
been destroyed by overstocking? An estimate based on such statis-
tics as we have been able to obtain from correspondents indicates
that the carrying capacity of the southwestern ranges was 40 per
cent less at the beginning of 1897 than it had been in 1880. The
money value of this loss has been variously estimated at from ten
million to forty million dollars in the State of Texas alone, and on
the other ranges in the Western States and Territories the aggregate
loss from this cause (overstocking) is not less than a hundred million
dollars. In other words, if the natural pastures in the country west
of the ninety-eighth meridian were now covered with as luxuriant a
growth of grass as they were twenty years ago, the additional num-
ber of live stock which could be carried would be worth probably
upwards of a hundred million dollars.

The regrassing of overstocked lands is to the interest both of the
individual stockowner and the commonwealth. The smaller losses
sustained by each owner become in their aggregate a sum which
materially affects the welfare of the State. It is the common testi-
mony of stockmen that there are vast areas where the abundance
and quality of the natural herbage has been decreased. The better
grasses have been run out by overstocking during years of drought.
Weedy annuals of less value, because less palatable to stock and less
nutritious, have taken their places. If these fail the ground becomes
entirely bare of vegetation. In other sections the amount of natural
pasturage has been decreased by the encroachment of perennial weeds
and thorny shrubs and by the cactus thickets, or the grasses have
been destroyed by rabbits and prairie dogs. Overstocked lands are
not only unproductive, but they rapidly deteriorate in productive
capacity. They require rest and treatment to again restore them.
The soil soon becomes hard and compacted by the trampling of cattle.
Less of the annual rainfall is absorbed by the soil, and more each
year is lost in the flood waters. Moreover the finer and hence richer
portions of the surface soils are washed into the streams, because
there is no protecting mat of grass roots to retain them.

2

EXPERIMENTS BY STOCKMEN,

Many experiments have been undertaken by individual farmers
and stockmen to determine methods of again bringing up the value
of the prairie pastures and to increase the grazing capacity. Such
work, undertaken as it often is by men who have a \^dde influence
among stockmen and stock-owners, is of great value. But the results
of such experiments, however valuable, do not obtain that extensive
and rapid circulation which they merit. There is probably no class
of American producers who are more ready to undertake work which
will tend toward the betterment of the marketable product than the
cattlemen and sheepmen of the West. All who are familiar with the
history of the cattle industry will admit that there is room for
improvement in methods all along the line, from the breeding sec-
tions to the feeding pens, but nowhere is it more important than in
the pasture^ — to provide more and better grasses. An abundant
supply of nutritious forage in the pastures means more continuous
and rapid growth of the animal, and in the end a finer quality of
beef and mutton, a better yield and grade of wool. There is immedi-
ate need of work along this most important line — range improvement.

EXPERIMENTS BY THE DIVISION OF AGROSTOLOGY.

Experiments have been undertaken by this Division at two points
in Texas. An effort is being made to determine the most practicable
and at the same time the most economical way of treating the nat-
ural pastures in order to again cover them ^vith the native grasses,
or with other better species from similar regions in other countries.
One experiment is being made at Channing, in Hartley County,
which will in a large measure represent the conditions that prevail
on the high plains of the Panhandle of Texas, western Kansas, and
Oklahoma, and parts of Colorado and New Mexico ; and one at Abi-
lene, to serve for the central and western prairies of Texas, up to the
border of the Staked Plains.

The work was commenced at both of these places in March, 18U8.
An idea of its scope may be obtained from the accompanying plan
(fig. 1) of the range station, near Abilene, Tex. The use of an
irregular body of land, containing about 640 acres, was donated by
Mr. C. W. Merchant, for a period of three years. The fences and
two tanks to supply water were provided by the citizens of Ablilene.
The work was inaugurated by Prof. C. C. Georgeson, formerly of
Kansas, at that time in the employ of this Division. Since its
establishment it has been in charge of a special field agent, Mr. H.
L. Bentley.

The section was divided into six pastures of 80 acres each, two of
40 acres, and one of 70 acres, and the remaining 10 acres is being
devoted to the cultivation of grasses. The work as planned is as
follows :

Pastnre No. 1. — No treatment except to keep stock off until June
1, pasturing the balance of the season.

Pasture No. 2. — Cut with a disk harrow and kept stock off until
June 1, pasturing the balance of the season.

Fig. 1.— Plan of 040-acre tract used for the range experiments at Abilene, Tex. (Scale 3 inches

to the mile. )

Pastures Nos. "' and 4 (of 40 acres each). — Grazed alternately,
the stock being changed from one pasture to another every two
weeks, thus allowing the grasses a short peiiod for recovery after
each grazing.

Pasture No. 5. — No treatment except pasturing until June 1 and
keeping stock off the balance of the season.

Pasture No. 6. — Left as a check, without any treatment whatever
except to keep stock off" during the first season.

Pasture No. 7. — Dragged with an ordinary straight-toothed har-
row and stock kept off during the first season.

Pasture No. 8. — Disked and stock kept oft' during the first season.

The 70-acre pasture. No. 9, was not grazed. Seeds of a number
of wild and cultivated varieties were sown directly upon the sod.

An effort has been made to get a stand of Texas Blue Grass and
Curly Mesquite by transplanting fragments of sod to the bare spots.
East and west furrows were also broken in order to arrest the grass
seeds Avhich are blown over the ground by the prevailing north and
south winds, the idea being to form seed beds, from which the most
valuable sorts should spread in every direction.

To insure uniformity the section of land was inspected before the
commencement of work by a committee of stockmen, who made an
estimate of the carrying capacity of the land at that time (April 1,
1808). The committee decided that the land would carry 40 head of
stock cattle to the square mile, in the proportion of ten cows with
calves, 15 yearlings, and 15 two-year-olds, which proportion will be
maintained as closely as possible until the end of the experiment.
The land will be judged again at intervals during the experiment,
in order to determine as exactly as possible the percentage and rate
of improvement in the different portions under the various methods
of treatment.

Mr. H. L. Bentley, the special agent in charge, estimates that
there has already been a gain of 25 per cent in the amount of grass
on the land, in the case of those pastures which have been disked
and harrowed. His report, dated November 24, 1898, concerning
the experiments, is herewith appended :

REPORT OF MR. BENTLEY.

The drought that has been on here for several months has not yet been broken.
Since April 1 the rainfall in the immediate section embracing the Station has
been distressingly small. There are two large tanks on the property, but there is
water in only one of them, and at present only a small amount. There are also
15 or 20 shallow water-holes that in ordinary years catch and hold water most of
the season, but they have been only partially full twice in seven months and
now all are dry. As a result there has been no end of difBculty in keeping cattle
in the pastures on account of the water scarcity. Just now I have no cattle in
pasture at all, not having sufficient water for even the 13 head called for by the
experiment. I hope, however, that we will soon have another good season here.
If once the tanks are filled there will be sufficient water to carry the cattle
through the winter in accordance with the plan. Notwithstanding the protracted
drought the grass is good — astonishingly good. In the pastures which Were har-
rowed I believe that there is fully twice as much grass as at this time last year.
Had we had normal seasons since April 1 it is plain to me that the harrowing
experiments would have shown splendid results.

The 10 acres of garden land are in first class condition to catch and hold all
the rain that shall fall during the winter. The woven wire fence around the
10-acre tract was placed in position by experienced fence builders, and I feel
confident now that the grasses and legumes vdll not be troubled seriously in the
spring by the prairie dogs and rabbits. The prairie dogs moved en masse on the
garden as soon as the crops appeared. Before the fence (woven wire) was placed
in position they had nearly destroyed the roots of the alfalfa and cowpeas, but
I anticipate no further trouble from them.

Enough seed of the cowpeas was saved for the next year's trials. The velvet
bean vines bloomed too late to matnre the beans before frost. An abundant
crop of pods formed, but no matured seed was secured. The teosinte grew to
be from 18 to 33 inches tall, then the drought set in and it never recovered stiflEi-
cient to do much good. The roots lived and it made some growth, but did not
mature any seed.

I am certain that with normal amount of rain next year we will be able to
demonstrate :

(i) That the ctdturc of the native sod tvith disk and tooth harrotvs will pay well
in this section.

(2) That alfalfa of all kinds can be grown successfully without inigation.

(3) That the teosinte is a splendid forage plant for this section, superior to any
of the sorghums so far tested.

(4) That the velvet bean will do astonishingly well and prove a crop of much
value.

(5) That all of the many varieties of cowpeas which were experimented with
this year are available crops for forage purposes.

There liave been i^ractically no results from any of the grass .seeds that were
sown. A number of varieties germinated, but they did not develop satisfactorily
because the rains did not come in time.

The data thus far secured at the close of eight months' work give
sufficient promise that definite, tangible results will accrue from
these experiments for the benefit of stockmen. It is too soon to draw
conclusions, but the outlook for rapid increase in the quantity of
grass on these overstocked pastures is encouraging. Moreover, the
methods in use are such as are well within the reach of any stock
owner, should he wish to avail himself of the results.

FUTURE WORK.

During the succeeding seasons experiments will be made as to the
practicability of sowing alfilaree, bur clover, Bokhara clover, alfalfa,
sorghum, and other wild and cultivated grasses and forage plants
directly on the sod without further treatment than to keep stock oft^'
during at least the first year. On an examination of the plans it
will be seen that a number of methods are being undertaken which
may be adopted at but little expense by stock owners, should they
prove to result profitably.

These experiments ought to be carried on for at least three years.

At the end of that time sufficiently tangible results will undoubtedly

be secured to enable stockmen to decide what is the best method of

bringing back the grasses.

Jared G. Smith,

Approved : Assistant Agrostologist.

James Wilson,

Secretary of Agriculture.

Washington, D. C, December 27, 1898.

O

I83-JED F33Y. 24. ]899.

Circular No. 9.— (Agros. 39.) at »

United States Department of Agrfcuitu?e? .

DIVISION OF AGROSTOLOGY O > ' '

[Grass and Forage Plant Investigation.] ^•

*• '^ ■' '•' A L

, » . ...

NEW SPECIES OF NORTH AMERICAN GRASSES.

In the present circular are brought together the descriptions of a
number of new species of grasses, mostl}^ belonging to the genus
Poa, from the Avestern part of the United States. This form of issu-
ing descriptions is adopted to expedite publication and to allow the
citation of names in non-technical publications where new descrip-
tions can not well be given.

1. POA CAPILLARIFOLIA Scribn. & Williams, sp. nov.

A smooth, csespitose peremiial, 3-4.5 dm. high, with long, thread-like leaves,
marcescent basal sheaths and narrow greenish panicle, about 4 or 5 cm. long.
Culm with the short, basal internodes usiially much exceeded by the sheaths.
Leaves soft, 1-2 dm. in length, abxmdantly produced from numerous inti'a-
vaginal innovations. Sheaths smooth, striate, the basal ones swollen, mem-
branous, and persistent. Panicle narrowly ovate, or oblong in outline,
closely flowered, lower branches solitary or in twos, or more rarely in threes,
the lower ones from 1 l-'i cm. in length ; spikelets 8-4-flowered, 5-6 mm. long ;
empty glumes unecpial, scabrous on the back, 3-nerved, scarious-margined,
the lower one Ijroadly lanceolate, aciite, 2i mm. long, the upper obtuse,
3 mm. long, both often erose-dentate above ; flowering glumes thin, 5-nerved,
broadly ovate, obtixse or acutish, minutely hispid on the back above and on
the keel and nerves, with broad hyaline margins and usually erose-dentate
apex, about '^ mm. in length. The palet nearly or quite equaling the
flowering glume, hyaline, with distinct grt^enish cili ate -hispid keels and
erose-dentate apex.

Type specimen No. 2614, Geo. Hansen, California.

Allied to Poa cusickii Vasey, but with smaller flowers, longer and more filiform
leaves and smaller, closer panicle. In some ways this species is intermediate
between P. cu.siekii Vasey and P. idahoensls Beal.

Poa vaseyochloa Scribn., new name. P. pulchella Vasey (1882), not Salisb.
nor Pari.

PoA NUDATA Scribn., new name. Poa capillaris Scribn. (1898), not Linn.

2. POA SAXATILIS Scribn. & Williams, sp. nov.

A slender, closely csespitose perennial, 2^3^ dm., tall, with soft, smooth leaves
and stems and lax, variegated panicles fr<mi 5-7 cm. long. Culms erect or
somewhat geniculate below, very smooth, with conspicuous purplish nodes.
Leaves flat, rather obtusely pointed, those of the stem from 1^2 cm. or more
in length ; those of the sterile shoots aboi;t twice as long. Sheaths smooth,
or basal ones sparsely pubescent, striate, shorter than the internodes, the
lower ones swollen and often split open by the growth of the intravaginal
shoots. Ligule 2-3 mm. long, acute. Panicle branches ascending, filiform
and somewhat flexuous, usually in twos, the lower ones about 4 cm. long,
flower- bearing above the middle. Spikelets compressed, lanceolate, 3-5-
flowered, 6-9 mm. long. Empty glumes somewhat unequal, obtuse, or sub-
acute, 3-nerved, smooth, about 4 mm. long. Flowering ghmie obscurely

2

5-nerved, oblong, obtuse or subacute, minutely scabrous on the back, sparsely
pubescent at the base, more densely pubescent on the keel and marginal
nerves, green below, purplish above with rather broad, scarious, erose-den-
tate margins, 4-4^ mm. long. Palet nearly equaling the flowering glume,
ciliate on the keels, apex bidentate.

Type specimen No. 1964, C. V. Piper, dry rocky places, Mt. Rainier, Washington,
August 1895. Altitude 2100 meters.

This species has been most frequently referred to Poa gracilHma Vasey, but dif-
fers in its more densely csespitose habit, shorter, flatter leaves, more closely-
flowered panicle, larger spikelets and flowers, and in the firmer texture of
the flowering glumes.

3. POA LECKENBYI Scribner, sp. nov.

A pale green, tufted perennial, about 7^ dm. high, with long, slender leaves,
somewhat inflated, persistent basal sheaths and pale rather narrow panicle
about U dm. long. Culms smooth, frequently rooted at the basal nodes.
Leaves glaucous green, smooth below, rough above and on the margins, •
acute ; sheaths smooth, all except the uppermost exceeding the internodes,
the lower ones membranous ; ligule about 6 mm. long, acute, or often fim-
briately divided, decurrent. Panicle branches erect, flower-bearing nearly
or quite to the base, the longer ones about 5 cm. in length. Spikelets com-
pressed, oblong-lanceolate, 5-6-flowered, 9-11 mm. long. Empty gliimes sub-
equal, 3-nerved, rough-hispid on the nerves, rather narrowly lanceolate, acute
or acuminate, 7 mm long, slightly exceeding the lower flowering glumes.
Flowering glumes ovate-oblong, 5-nerved, 5-6 mm. long, scarious-margined
and often erose-dentate above, hispidulous on the upper and pubescent on
the lower half, pubescence most abundant on the keel and marginal nerves.
Palet shorter than the flowering glume, 3-keeled, conspicuously ciliate-pubes-
cent on the keels and minutely pubescent below between the keels, narrowed
above to an erose-ciliate apex.

Type specimen from Scott, Klickitat County, "Washington, June 5, 1898, collected
by A. B. Leckenby for whom the species is named. The grass grows in very
sandy soil and is an excellent sand binder. It has something the appearance
of Poa nevaclensis Vasey, but may be easily distinguished by its inflated
membranous sheaths, larger flowers, and pubescent flowering glumes, which
are much less firm in texture.

4. POA BREVIPANICULATA Scribn. & Williams, sp. nov.

An erect, rather rigid, ctespitose grass, 1^-3 dm. high, from short, stout rhi-
zomes, with a short, broad, closely-flowered, greenish or purplish panicle.
Culms smoothish or somewhat roughened below the panicle; lower inter-
nodes short, exceeded by the sheaths. Leaves flat or more often conduplicate,
smooth below, rough above and on the margins ; cauline leaves 2-8, short,
the upper usually reduced to a mucro, the ligule short-truncate, often
reduced to a narrow band; leaves of the sterile shoots longer, 1 dm. or more
in length, acute with rigid points, ligule obsolete ; basal sheaths persistent.
Spikelets compressed, 4-6-flowered, 4-6 mm. long, ovate-lanceolate, rather
acute. Empty ghimes unequal, hispid on the keel and more or less scabrous
on the back; the lower narrowly lanceolate, acuminate, 1 -nerved; the upper
broader, acute, 3-nerved, 3i mm. long. Flowering glumes broadly ovate,
obtuse or acutish, strongly pubescent on the lower half of the keel and mar-
ginal nerves, minutely scabrous above on the back or sometimes hispidulous,
intermediate nerves obsolete, 3^-4 mm. long; palet hyaline with green,
hispid-pubescent keels, apex bidentate.

3

Type specimen No. 554, G. F. Breninger, Table Rock, Colorado, May 25, 1891.
Meadows. Altitude 2200 meters.

Tall forms of this species approach Poa longepedvnoulata Scribn., but maybe
distinf,niished ])y the smaller size, stout rootstocks, very .short lower inter-
nodes, broader panicle, denser pubescence on the flowering glumes and more
rigid appearance. It also approaches foi-ms of Poa eatoni S. Wats., but dif-
fers in its shorter, broader panicle, shorter basal intemodes and fewer-flow-
ered spikelets. Like the following species it has usiially been refeiTed to
PoafendJeriana. Its range extends from Colorado to New Mexico and Ari-
zona.

5. POA LONGILIGULA Scribn. & Williams, .sp. nov.

A caespitose glaucous perennial. 4 or 5 dm. high, with rather harsh culms and
leaves abundantly produced from short rhizomes, persistent basal sheaths
and rather crowded panicles of large, shining, green or purplish spikelets.
Leaves of the sterile shoots erect or spreading, usually U to 2+ dm. long,
flat or conduplicate, with a conspicuous, decurrent ligule; cauline leaves
shorter, with longer ligules (5-7 mm.), the upper one sometimes reduced to
a mere mucro; leaves all rather rigid and miacronate-pointed. Panicle
ovate in outline, erect or rarely somewhat nodding, f to \i dm. long, lower
branches seldom exceeding 3^ cm. Spikelets compressed, ovate-lanceolate,
rather acute, 4-6-flowered, 6-10 mm. long. Empty glumes unequal, more
or less rough-hispid on the back ; the lower lanceolate, acute or acuminate,
1 -nerved, 44 mm. long, the upper much broader, 3-nerved, 5 mm. long.
Flowering glumes ovate-oblong, rather obtuse, 5-nerved, the intermediate
nerves inconspicuous , more or less scabrous above on the back, woolly pubes-
cent on the lower half of the keel and marginal nerves, with broad hyaline
margins, usually erose-dentate above, 4-5^ mm. long; palet much shorter
than the flowering glume, hyaline, with green, hispid -pubescent keels.

Type specimen No. 5149, Marcus E. Jones, Silver Reef, Utah, May 3, 1894.
Altitude, 1,200 meters.

This species has been referred to Poa fendleriana (Steud.) Vasey, from which it
is easily distinguished by its coarser growth, harsh stems and leaves, much
longer decurrent ligule, conspicuous on the leaves of both sterile shoots and
culms, larger spikelets and flowers, and denser pubescence of the flowering
glumes.

Sderochloa calif arnica Mnnro, in Bentham's PI. Hartweg. (without description),
to which specimens of both this and the preceding species have been referred, .
is a very different plant and is also distinct from Poa fendleriana. It is
represented in the National Herbarium by Hartweg's No. 2035 and Brewer's
No. 1122, both from California, which agree exactly with the type of Poa
secunda Presl. in the herbarium of the Missouri Botanical Garden.

Poa longiligula ranges from southwestern South Dakota to Southern Colorado,
Arizona, and Oregon. «

0. POA PLANIFOLIA Scribn. & William.s, sp. nov. / ^- lA^j^i^ti pr^^

An erect, loosely spreading, glaucous perennial, about 6 dm. high, from short
rhizomes, with rather long, flat, abruptly pointed leaves and an open, sub-
pyramidal panifle 1-2 dm. long. Culm smooth, except just below the nodes,
where it is minutely roughened. Sheaths striate, somt-what keeled, more or
less roitgh on the keel, otherwise smooth ; leaves flat, glaucous, smooth below,
roughened above, especially on the margins, U to 2 dm. long; ligule obtuse
or truncate, about 2 mm. long. Panicle branches in 3's or 5's, scabrous,
spreading, the lower 5-10 cm. long, flower-bearing along the upper half.

Spikelets compressed, ovate-lanceolate, acute, 8-4 -flowered. Empty glumes
ovate, obtuse or acutish, 3-nerved, scabrous on the back, subequal, about
4 mm. long ; flowering glumes oblong-ovate, obtuse, 5-nerved, puberulent on
the back, conspicuously piibescent on the lower half of the keel and marginal
nerves, and more or less pubescent on the intermediate nerves, scarious-mar-
gined above, 3-4 mm. long; palet nearly or quite equaling the flowering
glume, rather narrow, tapering above to a narrow truncate, more or less
irregularly fringed apex, the prominent green keels hispid-pubescent. Grain
pointed below, narrowed above, and bearing three minute, whitish tubercles
on the summit.

Type specimen No. 2814, Thos. A. Williams, moist banks. Spring Creek, in the
Big Horn Basin, Wyoming, August 4, 1897. Altitiide, 1,500 meters.

This species has been confiTsed with Poa lucida Vasey, from which it differs more
particularly in haljit of growth and in the longer, flat, glaucous leaves, stems
rough below the nodes, and conspicuously piibescent nerves of the flowering
glumes, which are glabrous between the nerves.

Its range as indicated by specimens in the U. S. National Herbarium extends
from southwestern South Dakota to central Wyoming and north to central
Montana,

7. POA ACUTIGLUMIS Scribner, sp. nov.

A loosely tufted perennial about 6 dm. high, with rather short radical leaves and
pale green or purplish panicle about 1 dm. in length. Culm smooth or some-
what roughened at the nodes and just below the panicle. Cauline leaves
2-3, minutely roughened on both the upper and lower surfaces. Sheaths
smooth or somewhat roiTghened above, striate, shorter than the internodes.
Ligule aboiit 5 mm. long, acute, often more or less fimbriately divided,
rough-hispid on the back. Panicle ovate in outline, acute at both ends,
branches in twos or threes, rough-hispid, ascending, the lower longer ones
about 5 cm. in length. Spikelets lanceolate, acute, remotely 4-6-flowered,
8-11 mm. long; empty glumes unequal, rough-hispid on the back, the lower
long-acuminate, 3 nerved, the upper lanceolate, acute or aciiminate, more
or less distinctly 5-nerved, 5-6 mm. long; flowering glumes rather conspicu-
ously 5-nerved, narrowly oblong-lanceolate, acute, rough-hispid, on the
back, with a short, crisp pubescence at the base and on the lower half of the
nerves, about 5 mm. long. Palet a little shorter than the flowering glume,
oblong-linear, acute, conspicuously ciliate on the keels and sparsely pubes-
cent near the base between them.

Type specimen collected by Thos. Howell at Grave Creek, Oregon, May 21, 1884.
A very distinct species, apparently without any very near relatives.

8. POA TENERRIMA Scribn., sp. nov.

A very slender, csespitose perennial, 3-5 dm. high with short, filiform leaves and
spreading capillary panicles, 7-9 cm. long. Leaves of the intravaginal inno-
vations 4-8 cm. long, thread-like ; those of the culm 3-5 cm. long, the upper-
most usually about 4 mm., involute, filiform, at least when dry, smooth.
Panicle branches in twos and threes, or solitary above, minutely scabrous,
flower-bearing only near the tips ; spikelets 3-4-flowered, aboiit 7 mm. long,
florets rather remote on a slender rachilla ; empty glumes lanceolate, acute,
the first about 2.5 mm. long, the second a little longer and broader than the
first. Flowering glumes about 4 mm. long, oblong, obtuse, 5-nerved, per-
fectly smooth excepting for a short, crisp pubescence near the base. Palet
nearly as long as the glume, 2-keeled, 2-tootlied at the apex, keels ciliate,
excepting near the base.

Type specimen from California without special locality or collectors named. A
striking species, remarkable for the short, filifonn leaves and rather few-
flowered capillary panicles. The spikelets in the specimen in hand are
purplish, as is the culm, and the plants appear to be staminate.

9. POA LIMOSA Scribn. & Williams, sp. nov.

An erect, robust, glaucous perennial, 7 or 8 dm. high, from a rhizomatous base,
with broad, flat leaves and narrow, closely flowered panicle 1-U dm. long.
Culm smooth. Culm-leaves 2 or 3, 4-10 cm. long, those of the sterile shoots
10 to 20 cm. or more long, all flat (rarely conduplicate), smooth except on
the margins, and abruptly pointed : sheaths smooth, conspicuously striate,
all except the lowermost much shorter than the internodes ; ligule about
5 mm. long, acute or acuminate, often fimbriate, rough-hispid on the back.
Panicle-branches scabrous, erect (or sometimes spreading during anthesis),
the lower ones about 4 or '•> cm. in length. Spikelets compressed, lanceolate,
4-r)-flowered, about G mm. long ; empty glumes unequal, smooth or slightly
rough on the back, the lower lanceolate, acute, indistinctly 3-nerved, a little
less than 3 mm. long, the upper ovate-lanceolate, rather obtuse or subacute,
strongly 3-nerved, .slightly exceeding 3 mm. in length ; flowering glume con-
spicuously 5-nerved, oblong-ovate, obtuse, minutely scabrous on the back
and rough-hispid on the nerves, ?A mm. long; palet about equaling the
flowering glume, sometimes slightly exceeding it, ciliate on the keels, biden-
tate at the apex.

Type specimen collected by H. N. Bolander, at Mono Lake, California, evidently
growning in wet boggy soil, but no notes as to habitat accompany the speci-
men. Specimens collected by Bolander, No. 6114, and with no locality other
than "California," belong to this species.

The species is related to Poa nevadensis Vasey and is apparently confined to the
Pacific Slope.

10. POA E PI LIS Scribn., sp. nov.

A rather slender, erect, closely caespitose perennial, 4-6 dm. high, with numer-
ous basal leaves from intravaginal shoots, and contracted, closely flowered,
green, bronze-brown or purplish, usually somewhat nodding panicles about
5 cm. long. Culm smooth, composed of about three internodes, the low-
est short, the other two long and much exceeding their sheaths. Leaves
smooth, slender, acute, flat, or convolute when dry, those of the sterile shoots
much the longest, often 2 dm. or more in length. Sheaths smooth, striate.
Ligules of the cauline leaves acute, about 3 mm. long, those of the sterile
shoots obtuse, often reduced to a narrow band. Panicle branches erect or
ascending, smooth, the longer ones more or less flexuous, simple and flower-
bearing near the top or sometimes branched and flower-bearing nearly to
the base, seldom over 3 cm. long. Spikelets somewhat compressed, ovate-
lanceolate, 3-4-flowered. about 5 mm. long. Empty glumes smooth, unequal,
the lower lanceolate or ovate-lanceolate, acute or acimiinate, 1 -nerved, 2^ mm.
long, the upper broadly ovate, acute, 3-nerved, about 3 mm. long ; flowering
glumes 5- nerved, rough-hispid on the back, the roughness most conspicu-
ous on the keel and marginal nerves, oblong-elliptical, obtuse, about 4 mm.
long ; palet a little shorter than the flowering glume, cilate on the keels,
apex bidendate. Grain pale, ellipsoid, rather acute at both ends, with a
conspicuous white tubercle at the apex. "3

Type specimen No, 14r)7, Shear and Bessey, open places, thin, moist timberland, '
Buffalo Pass, Colorado, August 13, 1898. Altitude, 3,000 to 3,oU(j meters.

6

Plants belonging to this species have nsually been referred to Poa cuspidata
Vasey, but the type of that species belongs to Poa tvheeleri Vasey, from which
Poa epilis differ in its smooth sheaths, contracted panicles and abundantly
produced intravaginal shoots.

Poa epilis ranges from central Colorado to northeast Wyoming and central
Montana, and occurs in the mountains at rather high altitudes.

11. pOA INCURVA Scribn. & Williams, sp. nov.

A slender, closely caespitose perennial about 3 dm. high, from strong fibrous roots,
with purplish fiexuous stems, rather rigid flexuously recurved leaves and
small, purplish panicles 2-4 cm. long. Culm smooth. Culm-leaves about 2,
smooth below, minutely hispidulous above and at the acute apex, convolute,
1-3 cm. long ; sheaths smooth, striate, shorter than the internodes ; ligule acute,
often irregularly fimbriate, 1-2 mm. long. Leaves of the sterile shoots much
longer than those of the culm, recurved-flexuous, otherwise similar; basal
sheaths persistent. Panicle few-flowered, fiexuous, the branches mostly soli-
tary, minutely scabrous, flower-bearing at the outer extremities. Spikelets
compressed, purplish, loosely 2-8-flowered, about 5 mm. long: empty glumes
unequal, the lower oblong-lanceolate, subacute, indistinctly 3-nerved, slightly
more than 2 mm. long, the upper broadly ovate, obtuse or subacute, dis-
tinctly 3-nerved, usually a little more than 3 mm. long; flowering glume
rounded on the back, indistinctly 5-nerved, ovate, obtuse or subacute, smooth-
ish above, densely pubescent below, the pubescence strongest on the middle
and marginal nerves, purple on the back, rather broadly scarious-margined
above, about 4 mm. long ; palet nearly or quite eqxialing the flowering glumes,
strongly 2-keeled, ciliate on the keels, apex bidentate.

Type specimen No. 1989, C. V. Piper, Morain of Duckaloose Glacier, Olympic
Mountains, Washington, August, 1895. Altitude, 2,100 meters.

Allied to Poa. sandbergii Vasey but distinguished by its flexuous stems, smaller
spikelets, and smoother, convolute, recurved-flexuous leaves.

12. POA INVAGINATA Scribn. & Williams, sp. nov.

A smooth, caespitose erect or somewhat decumbent perennial, aboiit 4 dm. high,
wdth numerous intravaginal shoots, marcescent sheaths and pale, open pan-
icles 10-12 cm. long. Leaves mostly flat, rather narrow, acute, smooth or
the margins minutely hispidulous near the base; sheaths smooth, striate;
ligule acute, decurrent, 4-5 mm. long, alike on leaves of both culms and
sterile shoots. Panicle subpyramidal to oblong, the branches usually in
threes or fives, filiform, flextious, scabrous, flower-bearing along the upper
half, the longer ones seldom exceeding 5 cm. in length; spikelets compressed,
lanceolate, rather loosely 2-4-flowered, 6-7 mm. long ; empty glumes 3-nerved,
narrowly oblong to oblong-ovate, obtuse or subacute, the lower 3-4 mm. ,
the upper 4-5 mm. long; flowering glumes keeled, oblong ovate, obtuse or
subacute, 5-nerved, minutely scabroiis above, pubescent below on the nerves
and sometimes sparsely pubescent between them, 4-5 mm. long; palet
shorter than the flowering glume, ciliate on the keels, apex erose-dentate.

Type specimen in Herb. Scribner, No. 20, Calif. Acad. Sciences, collected at
"Summit Camp, Sierra Nevada," July 10, 1870.

It has been collected by C. F. Sonne at Dormer Lake, Calif., July 4, 1892 (No. 17)
and at Mt. Stanford, Calif., August, 1888 (No. 24), and also by W. N. Suks-
dorf at Mitchells Point, Wasco Co., Oregon, May, 1884 (No. 107a).

This species has been confused with Poa gracillima Vasey and Poa cusickii Vasey
but may be easily distinguished from the former by its coarser habit, flat
leaves, larger and more abundantly flowered panicle, larger spikelets and
longer, more conspicuously nerved flowering glumes, and from the latter by
its broader, flat leaves and pubescent flowering glumes.

I

7

13. ERAGROSTIS LUTESCENS Scribn., sp. nov.

A low, much branclied and densely ciespitose annual, 1-2 dm. high, with rather
short, flat leaves, and narrow, pale green or straw-colored panicles, 4-7 cm.
long. Cvilms glabrous. Sheaths loose or somewhat inflated, striate and
glabrous, or with a few short hairs at the throat ; ligule very short, ciliate ;
leaf-blade 2-5 cm. long, 2-8 mm wide, minutely scabrous above, very acute.
Panicle branches ascending or appressed, jiaked below. Spikelets narrowly
oblong, 2-3 mm. long, about \.~) mm. broad, 3 to many- (iTsually 10-12)
flowered. The empty glumes unequal, the first about 1 mm. long, the second
one-third longer and larger; flowering glumes obtuse, about 2 mm. long,
distinctly 3-nerved.

This species is allied to Eragrostis purshii, but is more densely caespitose and is
at once distinguished by its narrow, yellowish-green panicles.

Type specimens from sandy banks of Snake River, Almota, Washington, col-
lected by C. V. Piper, July 7, 1897, No. 2624. Number 4729 Prof. L. F.
Henderson found growing in sandy, moderately dry ground along the Platte
River, Platte, Southern Idaho, August 1, 1897, belongs to this species.

14. ELYMUS ARENICOLUS Scribn. & Smith, sp. nov.

A stout, erect, glaucous perennial, about 1 m. high, with numerous, tough, creep-
ing rootstocks. Culms terete, glabrous, 8-15 dm. high: nodes glabrous.
Sheaths striate, smooth, or the upper ones minutely strigose-pubescent,
nearly as long as the nodes and closely enveloping them; ligule coarsely
fimbriate, about 1 mm. long. Blades rigid, deeply furrowed on both sides,
smooth or very minutely strigose-pubescent on the back, strongly scabrous-
pubescent on the nerves, scabrous along the margins, 2-4 mm. wide, 2-3
dm. long, strongly involute and pungently pointed. Spike rather slender,
subflexuous, 1.5-2.5 dm. long, interrupted below ; rachis compressed, sparsely
hirsute along the margins and densely hirsute at the nodes. Spikelets single
or in pairs, lanceolate, comijressed, about ()-flowered; empty glumes subulate,
or nan-owly lanceolate, rigid, scabrous or sparsely hispid above, slightly
unequal, 7-10 mm. long, about half or less than half the length of the spikelet.
Flowering glumes rounded on the back, obliquely lanceolate, acute, sparsely
hirsute, especially toward the base, 5-nerved, mucronate- or sometimes awn-
pointed, the lowest about 10 mm. long. Palea a little shorter than its glume,
scarious along the margins, hispid scabrous on the keels, obtuse or emargi-
nate at the apex.

Tyiie collected by A. B. Leckenby, at Suferts, Oregon, November 26, 1898. Grow-
ing in the sand dunes along the Columbia River.

Other specimens examined are Nos. 466 and 468 Sandberg and Leiberg, Wash-
ington, 1893; No. 1176 Suksdorf, Rockland, Klickitat Co., Washington,
June 3, 1890.

It differs from Elymus flavescens, with which it has been previously confounded,
in the narrowly lanceolate flowering glumes, which are less densely jmbes-
cent. Like Elymus flavescens it grows on the dryest sand dunes and is an
excellent sand binder. The bases of the culms are clothed with dead leaf
sheaths.

F. LaMSON-ScR1J5NER,
Approved : Agvostoloijist.

James Wilson,

Secretary of AgricnUnre.

Washington, D. C, February 15, 1809.

O

BOTANICAI HAPnr^j

GIVEN 6r i\. l. BRJTTON.

Circular No. 10.— (Agros. 42.)

Issued April 1.5, 1H99.

United States Department of Agriculture,

DIVISION OF AGROSTOLOGY.
[Grass and Forage Plant Investigation.]

rOA FENDLERIANA AND ITS ALLIES.

Poa feudler^icma and its allies form an interesting and difficult
group of species and are found chiefly in the southwestern part of
the United States. The species are
much confused in herbaria and have
been very differently understood by
the various botanists who have studied
them. The first name published for
any of the species was Eraqrostis
fendleriana Steudel, based on Fend-
ler's No. 932 from New Mexico.
Munro confused this as well as allied
species with Hartweg's No. 2035, to
which he gave the manuscript name
Sclerochloa californica and much
of the difficulty that later botanists
have experienced has arisen from the
fact that in his determinations of ma-
terial he gave this name to such a
great variety of plants, many of them
differing widely in botanical charac-
ters. Thurber followed Munro in
making the Atropis californica of
Watson's Botany of California in-
clude very diverse forms. The Poa
californica of Coulter's Manual of
Rocky Mountain Botany and of
which Munro is cited as the authority,
is, according to both synonymy and
range, even more puzzling than Thur-
hev^'& Atropis californica. As stated
in a recent publication of this Division
(Circ. 9, Div. Agros., p. 3) Hartweg's No. 2035, to which Munro first
gave his manuscript name Sclerochloa californica, is a verj'^ different

Fio. 1.— Poa fendleriana.

2

plant from the type of SteudeFs Eragrostis fendleriana. It is a
member of the " tenuifolia " group and agrees with the type of Poa
secunda Presl, at least so far as the specimens in the National and
Gray Herbaria are concerned. Dr. Vasey in " Illustrations of North
American Grasses," Vol. 2, took up Steudel's specific name but still
confused that species with Munro's Sclerochloa californica and
figured a plant which seemingly does not belong to either and which
is certainly not typical of Poa fendleriana.

Fendler's No. 932 seems to have been a mixture. There are two
sheets under this name in the Gray Herbarium. The larger speci-
men on one of the sheets and a smaller plant in an envelope agree
well with the specimen in the National Herbarium and with Steudel's
description. A smaller undeveloped plant on this sheet is apparently
the same as the larger specimen on the other sheet, which differs
from what is here taken to be the type of Poa fendleriana, in having
longer, narrower flowering glumes, which are rough-hispid on the back
between the nerves, and relatively broader flat leaves. The panicle
is also different. The smaller plant on this second sheet has a pani-
cle much like that of Poa bi'evipaniculata S. & W., but the leaves
and glumes are those of typical Poa fendleriana. A specimen in the
Herbarium of Academy of Natural Sciences, Philadelphia, to which
one of Fendler's labels is attached, is still different, belonging to an
entirely different section of the genus.

The grasses of the fendleriana group are all conspicuous for their
tufted habit of growth and are more or less completely dioecious.
This latter character can not be satisfactorily studied in herbarium
material, and careful field studies are necessary before all the mem-
bers of the group can be properly understood.

SYNOPSIS OF THE SPECIIES.

1. Ligules usually 5-7 mm. long, acute or acuminate, decurrent, those of the
leaves of the sterile shoots conspicuous. P. longilignla.

1. Ligules much shorter, rounded or truncate at the apex; those of the leaves of

the sterile shoots reduced to a narrow band or fringe, or obsolete. 2.

2. Panicle narrow and contracted. P. longipeduncidata.

2. Panicle open. 3.

3. Panicle short, subpyramidal ; lower internodes very short, exceeded by the

sheaths; plants seldom reaching 3 dm. in height. 4.

3. Panicle longer, usually 8 cm. or more in length; internodes longer, seldom

exceeded by the sheaths ; plants usually 4-6 dm. in height. 5.

4. Flowering ghimes oblong, leaves slender, rough. P. scahriuscula .

4. Flowering glumes broadly ovate, leaves smoother, broader and more rigid.

P. b re 1 nj mnicu la ta .

5. Spikelets rather acute,/ ovate, 4-7-flowered ; leaf -blades hispi^-fnibesceul on

the upper surface. -^'' '^^ - "' « ' P. fendleriana.

5. Spikelets more obtuse, oblong to somewlL^t ovate, .j-!)flowered ; leaf -blades

rough-hispid above, but not pubescent. -^ --..^^o >«^->-»*,Xw2_ P- eatoni.

POA LONGILIGULA Scribner & Williams, Circ. 9, Div. Agros., p. 8 (Feb.

24, 1899).
This is one of the most clearly defined species of the fendleriana group, at once
distinguished by its larger proportions, rough leaves, and long, decurrent
ligules, well developed on leaves of both culm and sterile shoots.
Specimens have been examined as follows :

Soutli Dakota: Black Hills (Rydberg 1144).

Wyoming: Evanston (Nelson 2991); Slough Creek, Yellowstone Park

(Tweedy, 631).
Colordclo : Navajo Canyon (Eastwood 10|) ; "Colorado Ter." (Thomas).
Utah: Silver Reef (Jones 5176, 5149, type); Echo (Jones); Springdale
(Jones 5249) ; Santa Clara Valley (Jones 5139) ; Copper Mine, 18 miles
west'of St. George (Jones 5006) : -'S. Utah" (A. L. Siler 85).
Neui Mexico: Fort Wingate (Dr. W. Matthews 61).

Arizona : Pagiimpa (Jones 5089) ; Congress (Orcutt 2534) ; Red Creek
(Palmer 474+) ; Peach Spring (Lemmon 3172) ; Kingman (Orcutt
2474) ; Oak Creek (Lemmon 3155) ; Fort Mohave (Lemmon 3170).
California : Lemmon 18, without locality ; San Francisco (Bolander 2286) ;

Sierra County (Lemmon).
Nevada : West Humboldt Mountains (Watson 1312) ; Mica Spring (Jones

5057) : Virginia City (Bloomer 2266) ; "Nevada" (Wheeler 1872).
Oregon: Steins Mountains, opposite Devine Ranch (Leiberg 2480).
POA LONGILIGULA WYOMINGENSIS var. nov.
Leaves more slender than in the species, usually strongly convolute, and the

ligules less conspicuous : culm less leafy ; jjanicle more contracted.
Type sjjecimen No. 4799a, Aven Nelson, "In draws, Tipton, Sweetwater County,

Wyoming, June 17, 1898."
Specimens have been received from the following localities, all in the south-
western part of Wyoming:

Leroy (Nelson 4589) ; Point of Rocks (Nelson 4758) ; Evanston (Nelson 4526
4527, 4562a and b) ; Tipton (Nelson 4799, 4799a, type).
Occurs in dry soil of broken prairies and on hillsides at an altitude of about
2250 meters. It approaches P. longipeduncidata in habit of growth, but is
readily distinguished by the character of the leaves, ligules, and glumes.
POA LONGIPEDUNCULATA Scribn., Bull. 11, Div. Agros. 54 (July 20,
1898), Poa andina var. spicata Vasey, Bot., Wheeler Exped. 290 (1878), riot
P. spicata Linn.
The original description of P. longipedunculata was drawn from a specimen
with short leaves, much contracted panicle, and glumes nearly or quite
smooth })etween the nerves, but does not seem specifically distinct from
Vasey "s plant, which has longer leaves, somewhat looser panicle, and glumes
minutely hispid on the back. However, Vasey's name is preoccupied and
the species must bear the name above given. The species is abundantly
distinct from Poa fendleriau a in its more densely csespitose habit of growth,
generally smoother leaves and culms, narrow, contracted panicle, and smaller
spikelets and flowers.
Specimens of this species have been examined from the following localities :

Wyoming: Sheep Mountain (Nelson 3292, type of Poa longipeduncidata;
Williams 2296) ; Quarry Canyon (Nelson 3180) ; Pole Creek (Elias
Nelson 3194).
Colorado: Silver Plume (Shear 666; Rydberg 2418); Grays Peak (Shear
685; Lettorman 49; Patterson 5); Graymont (Rydberg 2444; Letter-
man 64, 65); Stove Prairie (Osterhout 16) ; Lake Ranch (French) ; Bear

Creek Divide (Tracy, Earle and Baker 42G3) ; Beaver Creek (Pammel) ;
Marshall Pass (Shear 981); Manitou (Shear 765); "Colorado "(Wolfe
1135, 1136, 1137).
Neiv) Mexico : Santa Fe Canyon (Heller 3685).

POA LONGIPEDUNCULATA VIRIDESCENS var. nov.

Differs from the species in its greener color, generally longer, softer leaves, less
contracted, often interrupted jjanicle, the more delicate texture of the glumes
and the usually more conspicuous intermediate nerves of the flowering
glumes.

Type specimen No. 2302, Thomas A. Williams ; open places, dry moimtain sides,
Sheep Mountain, Wyoming, July 2, 1897. Altitude 2,400 meters.

Specimens have been examined as follows :

Wrjoming: Sheep Mountain (Williams 2302, 2302a, 2276; Nelson 3304).
New Mexico : Barranca (Heller 3588); Santa Fe (Heller 3611); Santa Mag-
dalena Mountains (Vasey).

The general aspect of this variety is quite different from the species and a study
of a more extended series of specimens -v^all possibly prove it to be deserving
of specific rank.

POA SCABRIUSCULA sp. nov.

A pale green, closely cfespitose perennial, 2A-3 dm. high, with slender, lax
leaves, abundantly produced from intravaginal shoots and rather closely
flowered subpyramidal panicles 5-7 cm. long by 2-3.V cm. broad. Culm
erect, striate, smooth except for a short distance immediately below the
panicle ; lower internodes short, usually exceeded by the sheaths, upper one
much exserted. Leaf -blades flat or convolute, those of the sterile shoots
8-12 cm. long by 1-1^ mm. wide, those of the culm short, the upper one
seldom over 1 cm. long, often reduced to a mucro ; leaf-blades all striate,
more or less roughened throughout; ligule obtuse, about 1+ mm. long;
sheaths striate, rough, the lower ones persistent. Panicle erect, the branches
scabrous, seldom exceeding 3 cm. in length. Spikelets compressed, usually
4-flowered, 5-7 mm. long, rachilla pubescent; empty glumes somewhat
unequal, lanceolate, acute or subacuminate, smooth or very sparsely
rough-hispid on the keels, shining, hyaline with green keel and nerves, lower
1-, ujiper 3-nerved, about 3 mm. long. Flowering glumes oblong, obtuse,
. keeled, smooth with keel and marginal nerves pubescent below and sparsely
hispid above, intermediate nerves subobsolete, pale green or purplish with
hyaline margins, about 3A mm. long ; palet much shorter then the flowering
glume, hyaline with green ciliate-hispid keels, apex bidentate.

Type specimen No. 136, L. F. Ward, moimtains south of Glenwood, Utah, May
29, 1895. Altitude 2,550 meters. The species was also collected by J. Wolf
(No. 184) in 1875 in South Park, Colorado.

The species is in some respects intermediate between Poa fendleriana (Steud.)
Vasey and P. brevipaniculata Scribn. & Williams, but it is easily distin-
guished from the former by its shorter panicle, smaller spikelets and flowers,
oblong flowering glume and rougher leaves, and from the latter by the
character of the spikelets, texture and shape of the flowering glumes, slender,
lax, rough leaves and lack of rhizomes.

POA BREVIPANICULATA Scribn. & Williams, Circular 9. Div. Agros., p. 2
(Feb. 24, 1899).

Specimens referable to this species have been examined from the following-
localities :

Colorado: Table Rock (G. F. Breninger 554, type); Mt. Hesperus (Tracy,
Earle and Baker 262) ; West Mancos Creek (Tracy, Earle and Baker
160) ; Bob Creek (Tracy, Earle and Baker 205) ; Cripple Creek( Tweedy 279).

The grass occurs iu dry soil of meadows and on mountain sides at an altitude of
from 2,300 to 8,500 m.

POA BREVIPANICULATA SUBPALLIDA var. nov. Differs from the
species in its paler green color, slender, soft leaves, which are less often con-
duplicate, and usually more contracted panicles.

Type specimen Hall & Harbour (674 in part). Rocky Mountains, Colorado, 18(52.
A specimen belonging to this variety and possibly of the same collection as
the preceding is in the National Herbarium ticketed as Poa alpina var. ,
Rocky Moimtains, 18(i2, E. Hall.

POA FENDLERIANA (Steud.) Vasey, 111., N. Am. Grasses 2. No. 74 (1898),
in part. Eragrostis fendleriana Steud., Gram., 278 (1855). Atropis calif or-
nica Tlmrber, Wats. Bot. Calif. 2 ; 809, in part. Poa andma Vasey, Bot.
Wheeler Exped. 289, not of Nuttall nor of Linn. Poa calif ornica, in part,
Coulter, Man. Rocky Mountain, Bot., 420 (1885). (Fig. 1.)

An erect, caespitose perennial, 3^-6 dm. high, with rather pale iipright leaves,
abundantly produced from intravaginal shoots and open, erect or somewhat
flexuous, usually purplish panicle 5-10 cm. long. Culm striate, scabrous
immediately below the panicle and nodes ; internodes usually equaling or
exceeding the sheaths. Leaf -blades flat or more often convolute, scabrous
below, hispid-pubescent above, mucronate-pointed, those of the sterile shoots
1-2 dm. long, by about U mm. broad, those of the cauline leaves much shorter,
the upper one often reduced to a mucro ; ligule short, rounded or truncate,
reduced to a narrow band or fringe on the leaves of the sterile shoots ;
sheaths striate, scabrous, the lower ones membranous and persistent.
Panicle branches spreading in anthesis, ascending in fruit, scabrous, the
longer lower ones seldom more than 3 cm. long, flower-bearing above the
middle. Spikelet compressed, ovate to ovate-lanceolate, rather acute, 4-7
(rarely more) flowered, 7-8 mm. long, 8-3* mm. broad; empty glumes
unequal, smooth, the lower 1 -nerved, lanceolate, acute or subacuminate, the
upper 3-nerved, ovate to ovate-lanceolate, acute, about 4 mm long ; flower-
ing glumes keeled, oblong-elliptical to oblong-ovate, obtuse, pubescent below
on keel and marginal nerves, smooth between the nerves, intermediate
nerves obsolete, about 5 mm. long ; palet much shorter than the flowering
glumes, sharply bidentate, sparsely hispid-ciliate on the keels ; rachilla more
or less hispid -pubescent.
Colorado: Trinidad (Crandall 15); Upper La Plata (Tracy, Earle and

Baker 4257) ; Manitou (Heller 3502),
Neiv Mexico: Mangus Springs (Rusby 452) ; Silver City (G-reene 438) ; Santa
Magdalena Moimtains (Vasey) ; Santa Fe (Heller 3530, 3561). "New
Mexico" (Fendler 932, at least for the most part, type).
Arizona: Dos Cabezos (MacDougal 787; Emersley) ; Flagstaff (MacDougal
2) ; Ash Fork (Rusby) ; San Rita Mountains (Pringle) ; San Francisco
Peak (Jones).
California: " Sta. CriisSonora" (Parry).

Grows on dry hills and mesas at an altitude of from 1,350 to 2,750 m.

POA FENDLERIANA ARIZONICA var. nov.

Differs from the species in being strongly glaucous throughout, in having longer,
rougher leaves, more unequal empty glumes, and in the firmer texture of the
flowering glumes.

Type specimen collected at Yavapai Creek, June 1883, by Dr. H. H. Rusby. It
was also collected by Dr. Rusby at Prescott, Arizona. A less glaucous plant,
with smoother, flat leaves, collected by Dr. Rusby (No. 452) in the San
Francisco Mountains, Arizona, may be the staminate form of the variety.

6

POA EATONI S. Watson, Bot. 40th Parallel, p. 386, 1871.

An erect, caespitose perennial 4-G dm. high, from short rhizomes, with rough,
more or less glaucous leaves and rather open, oblong, or somewhat ovate
panicle, 7 or 8 cm. long. Culm smooth, more or less roughened immediately
below the panicle, composed of two or three internodes, all biit the lowest
exceeding the sheath, the upper one long and miich exserted. Leaf -blades
flat or becoming conduplicate, scabrous below, and rough-hispid above;
those of the basal leaves spreading, 1-2 dm. long, those of the cauline leaves
seldom over 2 cm. long, the upper one often reduced to a mucro. Ligule
short, with a rounded or truncate apex, that of the leaves of the sterile
shoots reduced to a narrow band or fringe. Sheaths striate, the basal ones
persistent. Panicle branches scabroiis, erect or ascending, seldom more
than 3 cm. long, flower-bearing along the upper half ; spikelets compressed,
oblong, rather obtuse, 5-9-flowered, 7-10 mm. long, about 3 mm. broad.
Empty glumes subequal, acute, or subacuminate, the lower 1-, the upper
imperfectly 3-nerved, hispid on the keels ^Hl the length of the adjacent
flowering glume. Flowering glumes strongly keeled, the keels and lateral
nerves thickly pubescent below, hispid above, smooth between the nerves,
4-5 mm. long, the intermediate nerves obsolete.

Very closely allied to Poa fencUerimia but possibly distinct through its oblong,
usually more numerously-flowered spikelets and shorter, broader, spreading
leaves which, though rough, are not pubescent on the upper surface. It is
possibly more strongly rhizomatous but this can only be determined by
studies in the field. No specimens have been seen that can be referred to
this species except those from the original collection by D. C. Eaton at Cot-
tonwood Canyon, Wahsatch Mountains, Utah, where the plants grew in dry
rocky soil at an altitude of 2,000 meters. Most of the material that has been
placed by botanists under this species is true Poa fendleriana. Prof. J.
Macomi has collected a plant in British Columbia (No. 58, Spencers Bridge,
1889; and 91, and 93, Valley of Eraser River, 1875), which has much the
habit of P. eatoni to which it has been referred btit it has the longer leaves
(more or less hispid-pubescent above) and ovate to ovate-lanceolate spikelets
of P. fendleriana and seems to be an intermediate form. It is not improbable
that further study will result in the suppression of Watson's species.

Note.— In Circular 9, Div. Agrost., p. 3, the name Poa plaid folia was applied
to a species of the Rocky Mountain region, but since this combination has already
been used by Dr. O. Kuntze (Revis. 3:, pt. 3, sec. 2, p. 366, 1898) for a South
American plant, a new name is necessary for our species. It may be called Poa
glaiicifolia Scribn. and Williams.

Thomas A. Williams,

Assistant Agrostologist.
Approved :

James Wilson,

Secretary of Agriculture.

Circular No. 11— (Agros. 43.)

<5>

United States Department^f Agriculture,

DIVISION OF AGROSTOL(

^

[Grass and Forage Plant Investigations.] 'V/ O'ys.

(;> '^

THE FLAT TEA. X<S

DESCRIPTION. ^^

The flat pea is a variety of the woodland pea {Lathyrus 8ijh<es-
tris), which is not very different in appearance from the common
sweet, or everlasting, pea. It is a perennial, enduring sometimes for

Fig. 1.— Plot of tint pea in the Grass Garden on the grounds of the Agricultural Department.

twenty-five years or more, has strong, deep, much-branched roots,
and produces many trailing or climbing stems, which interlace in
great tangled masses. The rather handsome rose-colored flowers
are borne in loose clusters and are followed by pods not unlike those
of the common pea.

This pea is a native of Europe, and is most abundant in the cen-
tral and southern parts, extending into northern Africa. It is found
in thickets and hedgerows, and on rocky hillsides, blossoming through-
out the summer. A variety has long been cultivated for ornament
in English gardens under the name of everlasting pea.

2

HISTORY OF ITS INTRODUCTION.

In 18()2, Herr Wagner, an agriculturist of Kircheim-Teck in Wur-
temberg, Germany, had occasion to visit the Little Carpathian
Mountains, where he observed that, while all other herbage had
been burned up by the intense summer drought, the masses of stems
and fohage of Lathyrus sylvestris remained green and tender. It
occurred to him that here was a good fodder plant for dry soils, the
more so as this luxuriant growth was upon pure chalk rubble. But
when Wagner put the Lathyrus to practical test he found it want-
ing in two particulars: first, it contained certain alkaloids which
rendered it disagreeable and injurious to cattle; and second, the seed
is protected by an exceedingly hard coat, so that it may lie in the
ground for years without germinating.

Fig. a.— Flat Pea (Lathyius syhxstris ivag)icri).

Thinking that possibly these defects might be eliminated by culti-
vation, Wagner undertook the systematic improvement of the Lathy-
rus. As a result the bitter alkaloids were gotten rid of, and the
hard, vitreous seed coat was rendered comparatively soft and pervious
to moisture, so that germination took place in two or three weeks,
instead of as many years, after sowing. This alteration was accom-
plished by yearly transplanting, each time to a better soil.

The improved variety of Lathyims sylvestris thus brought into
notice was named Lathy ims s^jlvestris wagneri, in honor of the
cultivator. English agriculturists have named it "Flat Pea." It
has received much attention in Germany, but has not yet been culti-
vated on an extensive scale. A great deal has been published about
it, chiefly in newspapers and agricultural journals. In England it
has excited some interest, and in the arid regions of Cape Colony its

3

cultivation has been attended with seeming success. In the United
States practical agricultnrists have not yet taken it u}) to any g-reat
extent. Several of the State experiment stations, notably those of
Louisiana, Michigan, California, and Massachusetts, have grown the
flat pea in a small way and have made chemical analyses of it. In
Louisiana the success of the experiment has so far not been marked,
although the experimenter admits that there has not been time for a
fair trial of it. At the Michigan Station quite favorable results have
been obtained, the yield be-
ing good and cattle relishing
the fodder.

USES.

Owing to its power of
taking nitrogen from the
air, and therefore of grow-
ing: without the aid of ni-

trogenous

manures when
once well established, the
flat pea is thought to l)e es-
])ecially adapted to compar-
atively poor, sand}^, shaly,
or "chalky" soils. Hence
its greatest value should be
as a soil renovator to re-
store nitrogen to ground
which has been exhausted
by cereals and like crops,
or is otherwise lacking in
sources of protein. This is
a function largely dele-
gated in the United States
to the clovers and cowpeas.
The Laf]u/ri(S is said to be
a good binder for drifting, sandy soils, the strong, deeply penetrat-
ing roots forming a mesh which holds the particles of soil together.
The roots are sometimes 20 to 30 feet long.

It is claimed that the hay of the Wagner flat pea makes an excel-
hmt fodder, much rehshed by cattle, sheep, and hogs, and contains
more nutriment than most standard fodders. Owing to the high
rate of nutritive matter in Lathy rns hay, it is recommended that it
l)e mixed with one-third or one-half of straw. At the Michigan
Experiment Station it was found that cattle relished the green
fodder, but no report is made of its effect upon them. A writer in

FlC. -A.

Young plants (if flat pi'ii. The out- to tlie left
shows the root tubercles.

Der Landbote, a German agricultural publication, states that a
plantation of flat pea affords excellent "standing mast" for pigs,
and the flowers of the Lathyrus are much frequented by bees, the
hone}^ from this source being exceptionally fine.

CULTIVATION.

German authorities recommend that the flat pea should be first
planted in the seed bed, from which the young vines are transplanted
to the locations where it is desired they finally should grow. It is
estimated about 30,000 plants are needed for an acre. In a country
so rich in forage plants as the United States, and especially where
the methods practiced are so difi^erent from what they are in Euro-
pean countries, this manner of procedure is not likely to be followed.
More economical and expeditious methods must be sought. One
method recommended is that of planting the seeds in drills 15-20
inches apart, burying the seed to the deepth of 3 inches, deeper if the
soil is very sandy and the region dr}^, or less deep in heavy soils.
The land should be prepared for receiving this crop the same as for
wheat or corn, although it is claimed that the flat pea will grow in
regions or in soils where these crops will not succeed. The young
plants usually come to the surface within three weeks from planting
and during the first season make a small growth, 10 or 12 inches
perhaps. The land during this season should be kept free from
weeds and the vines cultivated. The second year the growth is more
vigorous and during the third year the plants attain their full size.
In the Gulf States the vegetation is perennial, but in the latitude of
Washington the plants die down in the winter, although they are
quite resistant to frost and are slow in starting in spring. Their
heaviest growth appears to be during the months of July and Au-
gust, and at this season the vines make a dense mat, completely cov-
ering the ground, often to the depth of 3 or -4 feet. This tangled
mass of vines is cut with difficulty and must be mown with a scythe
or cut with a sickle. The vines, when cut, dry out readily and the
making of the hay is a simple matter,

CHEMICAL ANALYSIS.

The analysis of the air-dried hay at the Michigan Agricultural
Experiment Station is as follows :

Per cent.

Water .. 7.99

Crude ash 8.32

Ether extract, fat, etc 3. 08

Crude fiber, wooded matter 36. 70

Nitrogen, free extract, starch, sugar, etc 27. 74

Crude protem 27. 17

100. 00

The following is an anal^'sis made at the California Experiment
Station with comparative analyses of the flat pea made in England,
and of alfalfa. It is the general conclusion that although the
amount of crude protein in Hat-pea hay is very high, the plant does
not CDmmend itself for forage :

Flat Pea.

station (cut ..ft,.,- bloom)
before bloom). '''"'^^ oioom).

.\llulfa.

Moisture .

63.48
8.18
1.63

13.77
9.70
3.18

100.00

.58.03

7.44

2.05

10.. 58

12.21

3.09

07.40

f^rnde Di'ott^in . -

5.91

( 'rudp fat --

1.15

(!;irboliv(lrates

12. U2

Crude til )er

Ash _.

10.51
2.95

Total

100.00

100.00

NOTES FROM EXPERIMENT STATIONS.

The California Experiment Station reports that the flat pea main-
tains a heavy growth with very little moisture, keeping green all
summer without irrigation. Cattle and horses will eat the hay, but
avoid the green vines. Sheep and pigs eat it readily green. Flat
peas grow best on upland, sandy soils.

The North Carolina Station reports that the flat pea requires three
or more years to secure a stand and then it is of no practical value
for North Carolina. "The flat pea we place next to sachaline as the
most loudly trumpeted swindle perpetrated upon the long-suffering
public in recent years."

The Alabama Station reports that flat pea grows 8 to 10 inches
high the first year, withstands light frost, continuing to grow
throughout the winter. Cows and horses crop the vines, especially
before other plants have started. ' It is regarded as good a soil
renovator as cowpeas.

At the Hatch Experiment Station, Massachusetts, the flat pea was
cultivated for two years at considerable expense without securing
any fodder.

The Nebraska Station reports that this fodder plant appears to be
too tender to stand ordinary treatment and does not reccommend it.

The Utah Station reports that Latliyrus sijlvestris dried \\\)
during the summer of LS!t;>, although it remained green several
weeks after the alfalfa plants growing with it were dead.

The Kansas Station reports that young seedling plants of the flat
pea withstood the drought satisfactorily.

In the Michigan Experiment Station report for 1895 it is stated
that the flat pea had grown with fair success for several years, but
the feeding tests showed that sheep and cows had a decided distaste
for the forage. Sheep confined on flat-pea pasture lost weight, and
cows fed in stable lost weight and diminished in production of milk
and butter-fat when either green or ensiled flat- pea fodder was fed
as a part of the ration.

CONCLUSION.

In localities where it will pay to go to considerable trouble in order
to establish a growth of some forage plant, as in some of the sandy
and arid regions of the West, especially in limestone and chalky
regions of the Southwest, flat pea may be considered as having some
possible value; but where other crops can be grown, such as Indian
corn, the sorghums, wheat, rye, barley, clovers, or the ordinary
tame grasses, it will not pay to cultivate this fodder plant. The
deeply penetrating, tenacious roots and perennial habit exclude this
plant from ordinary use in rotation.

Finally, it must be borne in mind that this plant, naturally grow-
ing upon sterile ground, was improved by continual transplanting to
better soil. If sown upon poor soil, is there no danger that it will
revert to its original state, regaining the injurious qualities lost in
the course of improvement ? The method of improvement — by con-
tinual transplanting to better soil — makes the probability of rever-
sion great. Indeed, those who advertise the plant most extensively
admit that many who raise seed for sale have, on account of careless
cultivation, placed on the market seed of inferior or unimproved
forms. Unless this can be stopped it is to be feared that the result
will l)e the entire reversion of the imi)roved variety.

F. Lamson-Scribner,

Agrostologist.
Approved :

James Wilson,

Secretari/ of AgrkuUiire.

Washington, D. C, April 38, 1899.

45

Circular No. 12. -(Agros. 44.)

%^'^^i

(Issued May 5, 1809.)

.R KO. 1^. -(Agros. 44.) ■^ A 47/ ussucu .uay

United States Department (^Agmfuljure,

DIVISION OF AGROSTOLOGi
[Grass and Forage Plant Invesiigation.]

RAPE AS A FORAGE PLANT. ^/>

Tliroiigliout a large portion of the United States farmers and st«^-
raisers could advantageously grow more of the succulent forage
crops for feeding stock during the summer and autumn months, when
the suppl}' of grasses and clovers is often limited. Such crops may
usually be grown on land that has already produced an early matur-

FiG. 1.— Rape, {frown at Mellette, South Dakota.

ing crop of some sort, such as oats, rye, or winter wheat. One of

the best of these succulent crops is rape {Brassica napus Linn.), a

plant closely related to cabbage, turnips, and several other garden

and field crops.

DESCRIPTION.

Rape is much like the Swedish turnip or rutabaga in H})pearance,
but the root is more like that of cabbage. The leaves are large,
glaucous, smooth, spreading, and variously notched and divided ; the
Howers are bright yellow, nearly one-half inch in diameter; the
seeds are produced in pods usually 2 inches or more long.

Under ordinary field conditions tlie plant reaches a height of from
1^ to 4 feet, and the strong-growing roots penetrate the soil to a con-
siderable depth.

Rape is either annual or biennial. The annual varieties (summer
rape) are grown chiefly for the seed, and have not been much culti-
vated in this country. The biennial varieties (winter rape) are
used largely for forage.

NATIVITY, USES, AND EXTENT OP CULTIVATION.

Rape, like the turnip, is a native of northern Europe, ranging east-
ward into Siberia. Although it has long been cultivated in the Old
World, it has received but little attention in America until within
comparatively recent years, and is now much more widely grown in
Canada than in the United States. Practically, all the rape grown
in this country is the winter or biennial sort, but in Europe, especially
in England, summer rape is widely cultivated. The seed yields
about 33 per cent of expressed oil, which is of value for lubricat-
ing, and is also used for lighting. The compressed rape-seed cake
is used as a food for stock and as a fertilizer. It is regarded as par-
ticularly valuable as a fertilizer for flax and turnips. The seed is
much used as a bird food. In this country rape is grown almost
exclusively for forage, being used chiefly for soiling and summer
and autumn pasturage.

Rape is best adapted to rather cool, moist climates, such as prevail
in portions of Canada and the northern United States. It can, how-
ever, be successfully grown as a forage crop in many of the warmer
and dryer sections. Thus in favorable seasons or with a small
amount of irrigation excellent crops of rape are grown in Wyoming,
Montana, the Dakotas, and other States in the so-called semiarid
region, and many instances are on record where good crops have
been produced without irrigation, under conditions of drought so
severe as to cause the failure of corn and other farm crops. In
parts of the South rape may be grown for late fall or winter forage.

SOIL REQUIREMENTS.

For its best development rape requires a rich, moist, loamy soil,
and will usually do well on any but light sandy soils and stiff clays,
such soils being usually deficient in vegetable matter. In general a
soil that will produce good crops of turnip, cabbage, wheat, and corn
will be suitable for rape.

Rape is a gross feeder and draws quite heavily on the nitrogen as
well as the mineral constituents of the soil, and hence should be used
in rotation with crops that feed largely on other elements of plant
food. For example, rape and fodder corn take about the same pro-
portions of nitrogen, potash, and phosphoric acid from the soil, and

experience has shown that corn does not do well after rape, unless
the land is naturally rich in these substances. Results ol)tained at
the North Dakota Experiment Station indicate that the growing of
a crop of rape on land that has been sown to wheat for a number of
years produces a decided increase in the yield of wheat from the suc-
ceeding crop. This is a point of much value in regions where wheat
is extensively grown.

VARIETIES.

All the varieties of rape that have come into prominence in Ameri-
can agriculture are winter or biennial sorts. Dwarf Essex or Eng-
lish rape has been most widely cultivated. Recently a variety has
I been placed on the market under the name of Dwarf Victoria rape,
or simply Victoria rape, which has given excellent results in New
England and also in the Northwest, yielding, as a rule, rather better
than the Dwarf Essex. At the New Hampshire Experiment Station
this variety is reported as yielding nearly 50 tons of green fodder per
acre, and yields of 25 to 30 tons per acre are reported from South
Dakota and elsewhere in the Northwest. Under average conditions
a yield of from 10 to 20 tons or more may be expected from either of
these varieties.

CULTURE.

Owing to the great variety of ways for utilizing rape and the
many places it may occupy in the rotation of crops on the farm,
there are numerous methods of culture that may be followed in grow-
ing it. When it is grown as the primary crop of the season the land
should be prepared by deep and thorough plowing, preferably early
in the preceding autumn. In some soils a second plowing should be
given in the spring before the seed is sown, but in soils that are nat-
urally loose and mellow, such as are found in portions of the North-
west, a simple stirring of the surface with a cultivator or disk-har-
row will often be sufficient. The land should be well pulverized by
harrowing before the seed is sown. When the land needs fertilizing
barnyard manure may be applied before plowing in the autumn, or
if the land is plowed twice the manure may be spread on during the
winter or early spring before the last plowing. Commercial fertili-
zers may be applied by harrowing in at the time that the land is be-
ing pulverized previous to seeding. Whatever treatment the land is
given in preparation for this crop it should be such as to afford a
deep, mellow seed bed, as free as possible from noxious weeds.

SEEDING.

Throughout the Northern States generally, seeding may take place
from the first of June or possibly earlier, to the middle or last of
July, according to the season and locality. In the South the seed may

be sown in September or early in October. Under favorable conditions
2 to 3 pounds of seed per acre will be sufficient and it will never be
necessary to vise more than 5 pounds per acre. The seed should be
planted in drills far enough apart to allow cultivation. In practice
the distance varies, but it is seldom less than :10 inches nor more than
32 ; 2-1: to 28 being perhaps the most satisfactory, all things consid-
ered. For planting small fields an 3^ of the common garden drills
will be found quite satisfactory, but for large fields a grain drill with
some of the feed hoppers closed may be used. When the ground is
clean and in proper condition otherwise, good results maj' be obtained
by using the grain drill with all feed hoppers open, and giving no
after cultivation. As a rule, however, it will be best to plant in
wide drills and give sufficient shallow cultivation to keep the soil in
good physical condition and destroy weeds. With favorable soil and
climatic conditions, good crops of rape may be obtained from broad-
cast seeding, but whenever there is any danger of the surface soil
becoming very dry during the time the seed is germinating or when
land is at all foul, drilling will give much better results.

AS A CATCH CROP.

When rape is grown as a secondary or catch crop it will not often
be possible to pay so much attention to the preparation of the soil
and the time and method of seeding, and quantity of seed used may
be varied to suit the circumstances. Often fine rape may be grown
on land that has already produced a crop of some of the early ma-
turing cereals, such as rye, oats, or barley. As soon as the crop of
grain is removed, the land is plowed or "disked" and at once seeded
to rape. Field peas and other early maturing forage crops, or rye or
winter oats that have been pastured off in spring may also be fol-
lowed by rape with profitable results.

Another practice which is coming into favor in some sections of
the country is to sow rape in the spring with some grain crop, such
as wheat, allowing the former to take possession of the field when
the latter has been removed. This method is especially satisfactory
when succulent forage is desired for fall feeding. Rape may also be
sown in the cornfield just before the last plowing, as is often done
with rye and winter wheat.

AS A WEED DESTROYER.

Aside from its value as a forage, rape is an excellent crop to grow
on fields that are foul with weeds. The late date at which the seed
may be sown allows the weeds to get well started before the final
preparation of the soil begins, they are further kept in check by the
cultivation required for the crop during its early growth and later

the rape plants shade the ground so completely as to keep the weeds
down. An excellent treatment for a foul field is to plow thoroughl,y
in late summer or early autumn and seed to rye or some other forage
crop to be pastured off during the fall, winter, or early spring. When
the crop has been pastured sufficiently and before the weeds have
produced seed, plow again, plant rape in drills and give thorough
cultivation. There are few weeds that will survive such treatment
and the land will have given profitable returns in forage in tlu^ mean-
time,

HARVESTING AND UTILIZING THE CROP.

The rape is usually ready for use in about 8 or 10 weeks from the
date of seeding. The general practice is to use it as a soiling crop
or as pasturage. Sheep and swine may be turned into the field and
allowed to remain until the rape is pastured off. Cattle may also be
allowed to run in the field, but as they waste much of the forage by
pulling up the plants and trampling them down it is a better plan to
cut the rape with a scythe or mower and feed it to the animals.

With sheep and cattle care should be taken at first not to allow
the animals to eat too much, as there is danger of injury from bloat-
ing. Hungry animals should not be allowed to eat their fill, and it
is not best to turn them into the rape when the leaves are wet.
There is no danger of bloating with swine. It is an excellent plan
to have the fields so arranged that the sheep and cattle have access
to an open pasture as well as to the rape. Animals should have free
access to salt at all times when being pastured on this crop.

FEEDING VALUE.

Rape has a high feeding value. It makes an excellent feed for
fattening sheep and swine and for producing an abundant fiow of
milk in milch cows. On account of danger of tainting the milk
many people do not feed it to the cows until after milking. Rape
can be used to good advantage as a part of the ration for animals
that are being fed in pens for market or for the show ring. It is
also a valuable food for young lambs at weaning time. By begin-
ning as early as practicable in the spring and seeding at interva.ls of
two or three weeks, a continuous succession of rape can be produced
throughout the period when the permanent pastures are most likel}'
to be short. Rape will endure quite severe cold weather and thus
will last a long time after the ordinary pasture grasses succumb
to the frost. By the use of this crop stock, can be gotten into good
condition for the holiday markets or for winter and there need be no
check in growth, fat, and milk production through insufficient suc-
culent food during the late summer and autumn months, as is
too frequently the case.

Under favorable conditions two or three cuttings may be made in
a single season from a field of rape grown as a primary crop. Mr.
W. H. Heidman, of Kalispell, Mont. , reports three cuttings the first
season with a heavy yield of forage. He allowed the plants to stand
the second season and obtained a fine yield of first-class seed. Not
much attention has been paid to growing rape for seed in this coun-
try, possibly because of the fact that in most localities where this
crop has been extensively grown the winters are so severe as to
destroy the plants. It seems however that there are localities where
rape can be profitably grown for seed and farmers might well devote
more attention to this feature of rape growing since most of that
now used is imported.

Thomas A. Williams,

Assistant Agrostologist.
Approved :

James Wilson,

Secretary of Agriculture.

Washington, D. C, Mmj l, 1899.

ClRCrLAR No. 13— (Agros. 45.)

United States Department of Agriculture^

DIVISION OF AGROSTOLOGY. ^ ^ ' ^'**<

[Grass and Forage Plant Investigation.] r> ■■

FLORIDA BE(i(iAR WEED.

(Also Kiiovvu as Beggar Weed, Florida Clover, (Jiant Beggar Weed.)

{Desmodimn tortuosum. )

DESCRIPTION.

Florida beggar weed is an erect annual, a native of the West
Indies, and perhaps also of southern Florida. It is a leguminous
plant with rather woody stalks from 3 to 8 or 10 feet high, bearing
an abundant leafage above and
when in flower tipped with much-
branched erect panicles, the as-
cending lateral branches often 8
to 12 inches long. The seeds are
borne in many - jointed prickly
pods, which break apart at matur-
ity and are carried about by stick-
ing to the bodies of animals or the
clothing of persons. The plant is
hairy throughout, and has trifoli-
ate leaves, the obliquely rhomboid
leaflets being from 2 to 4 inches
long. Florida beggar weed is
closely related to the beggar weeds
or beggar lice of northern wood-
lands and prairies. Being a sub-
tropical species, it is adapted to
cultivation either as forage or for
soil renovation in subtropical re-
gions. It now ranks in the esti-
mation of the planter with velvet
1 ,, , 1 -j^ 1 r Fid. 1— BeggarweediDeswixJiuin tortwmini'^

bean, thougli perhaps its sphere or

usefulness is not so extended as the latter.

BEGGAR WEED AS A FERTILIZER.

In common with all other leguminous plants, the beggar weed has
the power of taking nitrogen from the air by means of tubercle-
forming bacteria in its roots. Beggar weed may be used as a niti*o-
gen gatherer by the farmer, who is thus enabled to procure at small
expense large quantities of this most valuable fertilizer or plant food.

It has been reported by farmers who have grown beggar weed in
their orchards and corn fields that the texture and color of the soil
have been changed within two years as a result of plowing under
the annual crops of beggar weed. It grows best on light, sandy
soils and makes its rankest growth when fertilized liberally with
superphosphate and muriate or sulphate of potash. Sandy soils do
not retain an excess of phosphoric acid and potash in the same
manner as clayey soils unless there is a good deal of humus or
decaying organic matter present, and it is of the highest importance,
if the fertility of such light soils is to be increased, that large quan-
tities of some form of green manure be given them. This can best
be done by growing leguminous crops, such as beggar weed or
velvet bean, which not only supplj^ a rank bulk of vegetation, but
contain larger amounts of nitrogen than the non-leguminous crops.
Humus added to the soil makes it more retentive of moisture and
improves its physical condition ; and the humus acts as a reservoir
of surplus inorganic plant foods, preventing or retarding their leach-
ing into the drainage waters. A crop of beggar weed turned under,
will, when decomposed, retain near the surface in ready reach of the
roots of succeeding crops more of whatever fertilizers are subse-
quently applied. Besides adding a large amount of nitrogen to the
soil, the beggar weed takes up large quantities of lime and potash,
about one-half of the total amount of ash consisting of these two
elements. Analyses show that one ton- of beggar- weed ash contains
508 pounds of lime, 230 pounds of phosphoric acid, and 482 pounds
of potash, valued, when purchased in the form of commercial ferti-
lizers, at about twenty dollars. One ton of beggar-weed hay con-
tains about 38 pounds of nitrogen. It requires from 20 to 25 tons of
hay to yield one ton of ash. At a four-ton j^ield per acre, which is
not an unusual one, the fertilizing value of the crop would be about
as follows: 150 pounds of nitrogen, at 15 cents per pound, $22.50;
phosphoric acid and potash worth $5.25, making a total of $27.75.
A four-ton crop of beggar weed would therefore, if turned under as
green manure, supply an equivalent of half a ton of the best com-
mercial fertilizer for the use of the succeeding crop in the rotation.

• HOW TO sow THE SEED.

For a crop of seed, beggar weed should be sown at the rate of 5
or 6 pounds of clean seed per acre. If grown for hay, from 8 to 10
pounds should be used. It should not be sown until the ground is
warm and moist, and the clean seed is preferable to the pods because
of the more uniform germination and better stand which may be
obtained. The seed is about the size, shape, and color of red clover
and weighs about as much to the bushel. It is now on the market
at a price low enough to place it within the reach of any farmer.

If sown at the beginning of the summer rains the seed need not l)e
covered. It must not be buried too deeply else the young plants will
not be able to reach the surface. By sowing at the beginning of
summer two crops may be secured. If cut for hay at the time the
first flowers appear the roots will send up a second crop, which may
be saved for seed, and enough seed will scatter to insure a crop the
next season. The seed may also be scattered in the corn rows at
the time of the last cultivation or at the beginning of the rains in
June. Then, after the corn has been stripped or cut for fodder, the
beggar weed may be mown for hay or harvested for seed. The crop
should be cut for hay when it is about 3 or 4 feet high, or at the
beginning of the blooming period. If cut after full bloom many of
the lower leaves will have
fallen and much of the best
part of the crop will be lost.

ITS VALUE AS A
HAY CROP.

Beggar weed makes a
very fine quality of hay,
which is relished by all
classes of farm stock.

According to analyses
made at the Florida Ex-
periment Station of the
upjier parts of plants not
yet in seed, 100 pounds of
hay contained 19.4 pounds
of crude protein. When
cut after the seed had rip-
ened the crude protein had
decreased to 15.75 pounds,
the fat and non-nitroge-
nous extracts from 45.4
pounds to 42.7 pounds, and
the crude fiber or indiges-
tible portion had increased
from 19.0 to 26.5 pounds.
An average of two analy-
ses of the entire plant, including the woody stems, the one made
from plants in green seed, the other from those not yet seeding,
indicated 11.85 pounds of crude protein in 100 pounds of hay. This
may be taken as representing more nearly the crude protein value of
average beggar- weed hay than the larger (juantity shown to be pres-
ent in the upper, more leafy, and more tender portion of the plant.

Fig. 2,— Florida beggar weed, grown at the Mississippi
Agricultural Experiment Station— plants 7 to 8 feet high.

If in cutting this crop only the upper leafy parts were saved then the
higher percentage of protein might be taken as representing the value.

COMPARED WITH OTHER FORAGE PLANTS.

Digestion experiments with beggar-weed forage have not been
made, but judging from the comparison of the analyses with those of
red clover, the nutritive ratio would be about the same. There are
12.3 pounds of crude protein in 100 pounds of red clover hay, 15.3
pounds in crimson clover, 15.4 in soy bean, 16.6 in cowpea, and 14.3
in alfalfa. In the beggar weed the percentage of crude protein is
less than in red clover, because there is a much larger amount of
crude fiber due to the larger and more woody stems. The percent-
age of loss in feeding beggar- weed hay is accordingly greater. On
the other hand, the yield per acre is higher than that of red clover,
ranging from 3 to 5 or even 6 tons per acre, especially when two
crops are cut.

BEGGAR WEED AS A FEED.

Beggar weed hay may be fed to best advantage by adding to the
ration some coarse forage which contains a smaller amount of crude
protein and more carbohydrates. In this way all of the digestible
portion of the crude protein in the beggar weed may be utilized. If
more is fed than can be utilized, that is to say, if the nutritive ratio
of the ration is too narrow, the surpkis of crude protein will be
wasted. The crude protein is the nitrogenous muscle-making ele-
ment in the food, while the digestible carbohydrates, including a
portion of the fiber, and the fat, starch, gums, and sugars are used up
in producing heat and energy. Crude protein enters into the forma-
tion of lean meat and it is necessarj^ for a forage to contain enough
nitrogenous food to enable the animal to make new blood, tendon,
and bone, and a substantial increase in weight. Nitrogen enters
into the fibrin of blood, the albumen of muscle, the gelatine of bone
and tendon, and the casein of milk, and to a certain extent into the
surplus fat. For the production of new and the repair of worn out
tissues the presence of crude protein in the forage is essential. The
animal can not make satisfactory growth if fed a forage deficient in
crude protein. Hence the great value of leguminous forage crops.

After the seed crop has been harvested, the beggar weed comes
up again, the rowen supplying fine pasturage until killed by frosts.
It never becomes a bad weed. The seeds do not sprout until the
ground is warm, and it may be used as a rotation crop, following
early spring vegetables or corn, the seeds remaining in the ground
and making their appearance after these crops are out of the way.
As a hay plant it is superior to velvet bean on account of the ease
with which it may be cut with an ordinary mower. It is also a

I

better crop to sow in orchards for green manure because it is not,
like the velvet bean, a climber, and does not have to be kept out of
the trees.

This plant is one of the hosts of the root knot {Heterodora radi-
cola). Professor Rolfs states that the " cowpea is frequently
attacked, the velvet bean occasionally, and the beggar weed rarely."
The nodules formed by this worm resemble the tubercles formed by
the nitrogen-gathering bacteria but may be distinguished by micro-
scopical examination.

ADAPTABILITY TO LIGHT SOILS.

Grazing and the production of hay for home consumption or sale
are each year becoming more important industries in the Southern
States as a natural result of the diversification of crops. Great
quantities of hay were formerl}^ shipped South, it being rather a
common opinion among Southern farmers and planters that good
hay could not be grown. This opinion is, however, no longer held
and there are many progressive farmers in the South who have aban-
doned the cultivation of low-priced cotton in favor of a more remun-
erative hay crop. There is no good reason why the South should
not grow every pound of hay that is needed for feeding work stock,
nor is there any reason why butter, cheese, and meats should not be
produced. Many of the farms in the South are not suited to the
growth of timothy and red clover, but there are plenty of good
leguminous hay crops and pasture grasses which equal or surpass
them in yield or feeding value. Some leguminous forage crop should
be grown on every farm. With beggar weed, alfalfa, velvet bean,
soy bean, and cowpea, there is a good list to choose from. Beggar
weed is, perhaps, the best of these for the lighter, sterile sand}^ soils,
including the hammock and pine lands of Florida and the sandy
pine lands along the Gulf Coast.

Jared G. Smith,

Assistant Agrostologist.
Approved :

James Wilson,

Secret (iri/ of Agriculture.

Washington, D. C, May 6\ 1899.

rV e W i OH H

CiRCULAU Ko. 14— (Agros. 48.)

United States Department of Agriculture,

DIVISION OF AGROSTOLOGY.

[Grass and Forage Plant Investigations.]

THE VELVET BEAN.

(Mi(cii)i(i ulilin.)

The velvet bean lias witliiu the past three or four years attracted a
good deal of attention in the Southern States. It is probable that very
few forage or other farm plants have spread so rapidly during so short
a time. The uniformly favorable reports regarding its value in the
South may be taken as an indication of the remarkable interest shown
by farmers throughout the laud in the more extended cultivation of
leguminous crops. Every one now knows that the legumes, including
such plants as the clovers, alfalfa, beggar weed, cowpeas, and velvet bean,
are capable of enriching the land on which they are grown because
they are fitted by the tubercles on their roots to absorb nitrogen from
the air, while other crops, such as the cereals, grasses, sorghum, and
sugar cane, must have nitrogen provided for them. Because of this
ability to manufacture fertilizers, leguminous crops are of great import-
ance in the renovatiou and building up of worn or sterile soils.

ORIGIN OF THE VELVET BEAN.

The velvet bean is apparently a native of India and has been in cultiva-
tion as an ornamental garden plant ibr a good many years. It is believed
to have been first introduced into this country by the Department of
Agriculture for this purpose about twenty-five or thirty years ago. In
favorable localities it often forms vines 30 to 50 feet in length. It is an
excellent i)lant for quickly covering unsightly objects or arbors. The
purple flowers are borne in clusters at intervals of 2 or 3 feet at the
joints of the stem. These are followed by clusters of short, cylindrical
pods, covered with a black, velvety down which has given the name to
the plant. Ivach pod contains 3 to (J large, rounded, brown and wliite
mottled seeds. The pods are constricted laterally between the seeds
and are often more or less curved.

The value of tlie velvet beau as a forage plaut was accideutally dis-
covered about six or eight years ago. Beiug a uative of the tropics it
only matures seed iu Florida aud the lower half of the States immediately
along tlie Gulf coast. It will probably uot ripen seed uorth of a Hue
drawu from Columbia, S. C, 150 or 200 miles back from the coast to
San Antonio, Tex. Wherever it ripens seed it is considered to be equal
or superior to cowpeas, but where seed must each year be purchased it
does not equal that croxi.

Fig. 1. — Velvet bean; a flowers, b pods, aiid c beans.
SEEDING.

In Florida the seed is sown in drills 4 feet apart, dropping from two to
four seeds iu hills 2 feet apart in the row. The seed may be dropped in
furrows when the ground is plowed, and covered 2 or 3 inches deei^.
The crop should be cultivated several times. In orange groves and
orchards the beans may be sown in drills 4 or 5 feet apart and not less
than 5 feet away from the trees in order to keep the vines out of them.
They make a better mulch crop iu the orchard thau the cowpea because
when the vines are cut down by frost they form a tangled mass which
retains the leaves and protects the soil from rain and sun. The leaves
stay on the vines longer than on cowpeas. Farther north the seeds
should be sown tliicker in drills 2 or 3 feet apart, or broadcast at the
rate of 1 or 2 bushels per acre. Its range of profitable cultivation does
not extend beyond that of cotton and will not until its period of culti-
vation has been extended long enough to result in the origination of
new varieties suited to a wider range of soils and climates.

FERTILIZERS.

Velvet beau makes its best growth on the lighter, sandy soils. While
capable of increasing the amount of nitrogen it requires a liberal dress-
ing of i)hosplioric acid and potash in the form of superpliosphates and
muriate or sulphate of potash. It pays to feed the crop well because
on rich ground the gain of nitrogen through the increased crop of vines
is more than proportionate to the added cost of the potash and phos-
phoric acid.

YIELD.

The yields of hay are about the same as for the best varieties of cow-
peas on similar soils. The plant grows looser and bulkier and looks as
though it would yield twice as much hay as the cowpea, but experi-

Fi(i. 2.— Velvet bean, showing leaves, flowers, aud young pods. Grown in grass garden, Department
of Agriculture, Washington, D. C.

ments conducted at the Alabama Experiment Station prove that the
superiority in this regard is only apparent. The reported yields range
from 2 to 4 tons of hay per acre, or m<n-e in Florida, where two or three
cuttings are made during one season. Farther uorth the crop has not the
same recuperative ability and can only be cut once. The yield of seed
amounts to from 20 to 25 or 28 bushels per acre, about the same as for
the most prolific varieties of cowpeas.

FEEDING VALUE.

The beans have a high feeiliug value, as shown by analyses made by
the Florida Experimeut Station. The air-dried shelled beans contain
6.29 per cent fat, 18.81 per cent crude protein, and 53.5 per cent non-nitro-
genous extract. Cowpeas contain 51.4 per cent fat, 55.7 per cent non-
nitrogenous extract, and 20.8 per cent crude protein ; the peanut 39.G per
cent fat, 15.0 per cent non-nitrogenous extract, and 27.9 per cent crude
protein; while soy beans contains 10.9 per cent fat, 28.8 per cent non-
nitrogenous extract, and 34 per cent crude protein. Digestion experi-
ments have not been made either with the hay or seeds. Judging from
the chemical analyses, they are about equal in feeding value to cowpeas,
but are of less value than either peanuts or soy beans. An analysis
of velvet beau hav made at the North Dakota Experiment Station

Fig. 3.— Velvet bean in an orange grove near Earleton, Fla. Seed sown iu drills 5 feet apart.
Photographed August, 1898.

showed 5.3 per cent crude fat, IG per cent crude protein, 20.7 per cent
crude fiber, and 41.8 per cent nonnitrogenous extract. The plants were
just commencing to flower at the time the analysis was made.

VALUE OF FERTILIZER.

Comparative analyses of the vines, iallen leaves, and roots of the
velvet bean, Spanish peanut, and " unknown" cowpea were made at
the North Louisiana Experiment Station to determine the amount of
fertilizers contained in each crop. For the velvet bean 4,113 pounds
of vines and leaves contained 93.4 j)ounds of nitrogen, worth $14; 3,382
pounds of fallen leaves contained 58.2 pounds, worth $«8.73, and 173
pounds of roots without any stubble contained 2.7 pounds, worth 40

5

cents, making a total of 154.2 ponnds of nitrogen in tbe velvet beans
produced on 1 acre of ground, worth $2.3.13. An acre of peanuts con-
tained 193 pounds of nitrogen, worth $28.95, while the 108.5 jiounds of
nitrogen in an acre of cowpeas was worth 810.20. Similar analyses
have been made at the Alabama Experiment Station. Here a yield of
8,240 pounds of cured vines and fallen leaves and 1,258 pounds of roots,
including about 3 inches of stubble, contained 201 pounds of nitrogen,
worth, at 15 cents per pound, $30.15. There was 2.29 per cent nitrogen
in the cured vines and 1 per cent in the air-dried roots.

Experiments were also made at the Alabama Station to determine
the value of the velvet bean as a fertilizer, judging from the yield of
succeeding crops of oats and sorghum. The increased yield of sorghum
fodder was 3,272 pounds per acre over the yield on the plat which had
not been cropped the previous year — something over li tons, valued at
$12. The yield of oats grown on land where velvet bean stubble had
been plowed under was 38.7 bushels, and where velvet bean vines were
used 28.0 bushels, while land on which a crop of crab grass and weeds
had been plowed under only yielded 7.1 bushels, an average gain of
20.5 bushels of grain as a result of growing velvet beans on the land
the previous year. The average gain was about the same when cowpeas
were grown.

It is harder to plow under a crop of velvet beans than one of cowpeas,
on account of the tangled mass of vines. It is necessary to use a roll-
ing cutter, unless the farmer has a disk plow. At the Alabama Experi-
ment Station it was found that as good results were obtained from
plowing under the stubble as from plowing under a full crop of vines.
As a general rule, it may be considered a wasteful practice to turn
under the entire crop, because the feeding value of any leguminous
crop is always greater than its fertilizing value. A greater profit can
be secured in the form of marketable meat i)roducts without materially
lessening the influence of the leguminous crop on the succeeding one
in rotation.

USE OF THE EEANS FOR FOOD.

^"elvet beans have been used to some extent as hunmn food. The
general opinion, as expressed by corres})ondents of the Southern agri-
cultural papers, is that they are richer and less palatable than cowpeas.

The seeds are large, dififlcult to thrash, and the pod does not break
up readily. Special machines have been constructed for the purpose of
cleaning them. For feeding purposes it is a good idea to grind them
up, pods and all, thus saving the expense of thrashing. The meal may
be used as a concentrated feeding stuff in the same manner as cotton-
seed meal. It is said that cattle, sheep, hogs, and poultry are all very
fond of them, but horses ai)parently do not relish them.

Jared G. Smith,

Ai)proved : Assistant Ayrostolofjist.

James Wilson,

/Secretary of Agriculture.

Washington, D. C, May 13, 1899.

i

/

Circular No. 15— (Agros. 49). Issued July 14, 1899.

United States Department of Agriculfure,

DIVISION OF AGROSTOLOGY. •

[Grass and Forage Plant Investigation. ]N,^

RECEM ADDITIONS TO SYSTEMATIC AC^ROSTOLOGY.

In 1890 there was published under the title of " The True Grasses "
a translation of Edward Hackel's contribution to the great work of
Engler and Prantl, "Pflanzenfamilien," by F. Lamson-Scribner and
Effie A. South worth. Since the date of this publication a number of
important works on grasses have appeared and several new genera
have been described. Last year, 1898, Hackel published, in a supple-
ment to the German work above cited, the important recent additions
to literature and descriptions of the new genera which have appeared
since the publication of his first )vork. A few of these additions
appeared in "The True Grasses," but the recently-described genera
and important changes in nomenclature, not published in that work,
warrant their jsresentation at this time for the use of American
students of grasses. Naturally the contents of this circular are
based largely upon the "supplement" here referred to.

NEW LITERATURE.

A. et Cas. de Candolle, Monographije Phauerogamarum, Vol. VI. Andro-
pogonege, auct. E. Hackel, Paris, 1889.— H. Baillon, Histoire des Plantes; Mono-
graphie des (3rraunnee8, Paris, 189B.— O. Kuntze, Revisio Generum Plantaruni,
pars II (1891).— Bruns, der Grasembryo (Flora 1892).— Celakovsky, Ueber den
Aehrchenbau der brasilianischen Grasgattuiig t^treptocha'ta, in Sitzungsbericht
der Boelimischer G^sellschaft der Wissenschaften, 1889.— Derselbe, Das Rcdvic-
tionsgesetz dor Blatter das. 1894. Derselbe, Nejnovejsi badani a nazory o em-
bryn trav. (Die neviesten Forscbungen iiiid Ansichlen iiber den Grasembryo
(boehmiscben), in Vestnika Ceske Akadeniie Fr. Josefa V, 1896).— A. Schlic-
kum, Morphologie und anatomisb Vergleicb der Kotyledonen und ersten Keini-
blatter der Keinipflanzen der Monokotylen, Bibl. bot. Heft. 35 (1896).

ADDITIONS AND CHANGES.

After Zea L. add as a synonym Thalysia L. 1735.

After THp/sacKiii L. add as a synonym Dactylodes Zanoni Nonti. 1742.

After Coiv L. add as a synonym Sphceriinn L. 173.").

After Polliii ia Trin. add :

Ischnochloa Hook, f.— Spikelets very small in solitary clusters in the leaf-
axils, on a continuous rachis, arranged in pairs, one sessile, the other long pedi-
cellate, both hermaphrodite, lanceolate, longawned. First empty glume sub-
coriaceous, compressed dorsally, 5-6-nerved; the second similar, 3 nerved, both
awnless. Third empty glume wanting ; fiowering glume awned from the cleft
in the apex ; awn slender, geniculate. Stamens 3.

Species one (/. Falconeri Hook, f.) in the northwestern Himalayas; a delicate
grass, growing among moss, similar in habit to Arthraxon microphyUns, but
withoiit close relationship to that species. At first this would seem to belong to
section 2 of PoUinia, in which, forms occur without the third empty glume; but
in PoUinia the racemes are never solitary and the rachis is articulate.

The genus Manistiris Sw. has been renamed Hackelochloa by O. Kuntze be-
cause Manisui'is is used for RottboeUia L. f. Hackel had already stated this fact
in DC. Monogr. Phanerog. Vol. VI, p. 314, without drawing the same conclu-
sion as O. Kuntze.
After Rhytachne Desv. add Lepturopsis Steud., as a synonym.
After Andropogon L. add Sorghum L. 1735, as a synonym.

Under the subgenus ScMzachy-
rium add Heterochloa (Desv. as a
genus) as a synonym.

The subgenus Hypogyninm is
divided into two sections in DC.
Monogr. Phan. VI. : Euhijpogy-
nium and Pseudanthistiria; the
latter has been raised to a genus
under this name by Hook. f. in
Flora Indica.

Under the subgenus Arthrolo-
j)/;/.s insert Arthrostachys Desv. as
a synonym.

Under the subgenus Amplii-
lophis insert Botliriochloa O.
Kuntze, with one species (B. ana-
mitica) from Anam. The differ-
ence may only lie in the different
interpretation of the inflores-
cence.

Under the subgenus Dichan-
thinni add Diplasantlmm Desv.
as a synonym.
After Tragus add :
Monelytrum Hack. Spikelets
4-5, in closely approximate fasci-
cles, simiilating a 4-r) flowered
spike, the fascicles crowded into
long, thick, awned, and woolly
spikes somewhat resembling Alo-
pecnrus. First empty glume
e, rough-nerved, but without bristles, terminating in
a long, spreading awn. Flowering glumes almost of equal length, thin and
bristle-pointed. Palea obtuse. Lodicules none. Only the lowest spikelet of
each fascicle hermaphrodite, the second staminate, the bristle-form, uppermost
one sterile.
Species one, M. luderitzianum Hack., in German Southwest Africa.
Nazia Adans, 1764, is a synonym of Tragus Hall, 17(58.
After Scliaffncra add:

Fourniera Scribn. (figs. 1, 3). Distinctly dioecious. Spikelets solitary and
sessile along a continuous rachis, falling off at maturity, the two sexes unlike ;
staminate spikelets two flowered, the lower flower sessile, the upper raised on a
short pedicel, which is not prolonged beyond the flower ; empty glumes three, in

Fig. \.~F<)urnkra mcricana, male plant: o, spikelet;
h, second floret ; c, d, e, the three outer glnmes (c, cor-
responds to c in flg. 2); fif, the floral glume of the same ;
h, stamens. The details are all drawn upon the same
scale.

wanting, the second lar

a whorl, the two anterior ones shorter and narrower : flowerinj? j^himes o-nerved,
those of the n])per fiowers :5 toothed, teeth uwnlike. Stamen.s three. Pi.stillate
R])ikelets one flowered, with a three a^^^led prolongation of the rachilla beyond
the flower ; emi)ty glumes 3, equal, cuneate, whorled, flowering glumes on short
pedicels. 3-nerved. 3-toothed, the middle tooth longest; styles 2 ; stigmas plumose.
A delicate, much Ijranched,
creeping gi-ass with erect
spikes. i. J

Species one {F. mexicana
Scribn. ) in Mexico.

This grass is quite distinct
from other members of the
tribe. The significance of the
three empty glumes standing
in a whorl is not clear; in
the pistillate plant they are
attenuated l)elow into a short,
bearded pedicel and appear
almost like an involucre of
three rudimentary spikelets
(compare Themeda) .

Under Pdsjxihwi, Sect. I, Eu-
jiaspalum, insert Paspalan-
thinm Desv. as a synonym.

Under Panlcum insert, after
Trichachne, GrtDiieriiim Desv.V
as a synonym.

After Oplisuienus Beauv. add
Hippugrofitis'Rmn\}\\., 1749, as
a synonym.

Under Setaria add :

"The generic name Setaria
has been much discussed in
recent times from the fact that
it was originally used by Beau-
vois for a, speciea of Pen)iisetuni
and moreover had been used be-
fore by Acharius for a genus of
lichens. O. Kimtze combines
Si'f((>'ia with Chamairaplds
R. Br. and uses for Setaria
the latter, older name; Scrib-
ner, on the contrary, rightly
does not approve of this com-

FiG. ■2.~F()unii(ia nic.iicaiia, fenuile plant: «, tei-mi-
iial portion of rachis with two spikelets: /<, a si)ikelet:
«, (I, i\ outer glumes (f may represent a Klume-likc contin-
uation of tlie secondary axis supporting the sjjikelet, or
it may represent a second spikelet of a cluster or two):
/, flowing glume raised upon a shoi't joint of the rachilla
(stipe); y, palea. At tlie left of a is a :i-awned prolonga-
tion of the rachilla. The pistil Is shown in tlie upper
left-hand corner.

bination, and proposes for

Setaria the new name " CJuetochloa" now adopted by American authors.

After ChfctocJdoa add:

Setariopsis Scribn. Rept. Field Columb. Mus. (bot. ser.) i: 288, 1890. Plate
XI. Panicles narrow, interrupted, and spike-like, the short branches bristle-
pointed ; second glume broadly ovate or orbicular, auriculate or cordate at the
base, 11-13-nerved; third glume narrow, 11 -nerved, lyre-shaped, the margins
below becoming somewhat coriaceous at matiarity; the fourth or flowering
glume mtich shorter tlian the second and third, apiculate or mticronate-pointed.

Species two, in Mexico and Central America.

4

Fig. 'i.—Lrophorus it n isetitx : a, a portion of the rachis of one of the racemes ; h, a spikelet show-
ing back of the first and third glumes: o, a spikelet showing the many-nerved second glume ;
d, fourth or flowering glume seen from the back with the projecting styles and stigmas: e,
spikelet in fruit, showing the tirst and third glumes and the liroad wing-like margins of the
palea of the third glume: ;. the same as e, seen from the other side: (/. dorsal view of the fourth
glume in fruit.

Reduced by Hackel to a section of Setaria.

After Scfariopsis add:

Ixophorus Schlecht. (fig. 3), which Bentham and Hackel formerly placed with
Setaria, is a well-defined genus. Characters: Spikelets 3 flowered, the upper
flower hermaphrodite, the lower staminate, arranged in two rows along the
branches of a simple panicle, their pedicels provided wath a smooth, sticky, awn-
like bristle {chcetodadium). First empty glume very short; second somewhat
shorter than the third ; the latter o-nerved, much longer than the flowering glume
of the hermaphrodite flower, which is awn-pointed. The palea of the staminate
flower (in the axis of the third glume) has at maturity two broad, parchment-
like wings, determining the form of the spikelet. Very different from Chteto-
chloa in inflorescence, approaching in this respect the section Ptychoiihyllum of
Panicnm. Tlie viscid secretions on the bristles are worthy of notice.

Species two, in Mexico.

After Ixophorus add: Dissochondrus (Hillebr. as subgenus) O. Kuntze. Like
Cha^tochloa, but both flowers hermaphrodite with coriaceous, persistent flower-
ing glumes and palea ; the leaf -blades attenuate below, the mouth or throat of
the sheath with two lon^, narrowly -lanceolate lobes.

Species one (D. biflorus O. Kuntze) in the Sandwich Islands.

The synonym Oxyanthe Steud., under Pennisetum, probably belongs to Arnndo.

Under Oryza add: Bhynchnryza Baill., founded on Oryza snbulata Nees (from
Brazil), differs from Oryza only in the flowering glumes being lengthened into
a hollow beak, articulated by separation-membranes (instead of a deciduous awn)
and in the united lodicules. It might better be regarded as a subgenus of Oryza.

After Piptocho^tium Presl, add as a synonym Caryochloa Spreng.

The genus Brachyelytrum, through recent discoveries, has undergone an exten-
sion and is now be divided into three subgenera, as follows :

Sect. 1. Aphanelytrum Hack. Empty glumes diminutive, scarcely 0.5 mm.
long, often wanting. Flowering glumes subhyaline with short, subulate awn.

Species one {B. procxnnhens Hack.) Ecuador.

Sect. 3. Eubrachyelytrum Hack. Empty glumes short, firm in texture, the
upper about one-eighth the length of the flowering, which is long awned.

Species one {B. aristatum Beauv.), North America.

Sect. 3. PsevdobromusK. Schum. as a genus. Empty glumes from one-half to
two-thirds the length of the flowering glume, which is long-awned, firm, herba-
ceous.

Species two, in Africa: B. Africanum Hack., in Transvaal and B. silvaticum
(K. Schum.) Hack., in Kilimandscharo.

After C'rypsis Ait. add as a synonym PaUasia Scop.

Chaboissaea Fourn., Mex. PI. Enum., Gram. 112. 1886. Spikelets in narrow
panicles, 2-flowered, the lower flower hermaphrodite, the upper imperfect;
empty glumes unequal ; flowering glumes mucronate-pointed and ciliate on the
margins ; stamens 3 ; styles long ; stigmas phimose.

Species one in Mexico. Allied to Middcubergia.

Before Mibora Adans. insert Brousmichea Bal. False spike cylindrical ; empty
glumes solitary and united below at the margins for one-third to two-thirds their
lengtli. Flowering glume awnless, 1-nerved; palea nerveless. Lodicules none.
With the aspect of Sesleria.

Species one {B. seslerioides Bal.) in Tonkin.

After Garnotia insert the three following genera :

Woodrovia Stapf. (Hook. Icon, PI. 5: P1.2U7. 1896). Spikelets in slender
racemes which are subdigitately panicled; empty glumes compressed, 1-nerved
with a rounded, thickened keel, the first much exceeding the 2-lobed flowering
glume which bears a slender, geniculate awn between the lobes; stamens 2.
Annual.

6

Species one ( W. diandra Stapf . ) from India.

Garnotiella Stapf. Spikelets solitary along the short branches of an elongated
panicle, articulate with the pedicels. Empty glumes delicate, nerveless, not
keeled, the second mucronate-pointed. Flowering glumes very small, with a
long, geniculate and twisted awn from the cleft of the apex. Palea very small.
Lodicules wanting. Stamens 2. Habit like Garnotia.
Species one [G. philippinensin Stapf.) in the Philippine Islands.
Cyathopus Stapf. Spikelets solitary along the branches of an open panicle,
articulate on short pedicels, which are thickened and cup-shaped above. Empty
glumes 3-nerved, mncronate ; flowering glume somewhat shorter, more delicate,
5-nerved (the nerves disappearing below the apex), awnless. Palea hyaline.
Lodicules 3 ; stamens 3.
Species one (C sikkimensis Stapf.) in Sikkim-Himalaya.
Before Zenker a Trin. insert:

Massia Bal. {MegaJachne Thw. non Steud.). Panicle open; empty glumes
lanceolate, subulate-pointed, many nerved; flowering glumes with a stout, termi-
nal awn, finally becoming indurated together with the long, two-awned palea.
Species one (M. triseta Bal. ) from Ceylon to Tonkin.
After Spartina Schreb. add Chauvinia Steud. as a synonym.
After Chloris Sw. insert Biatherium Desv. as a synonym.
After Monochoete Doell. insert Doellochloa O. K. as a synonym.
Before Craspedorachis Benth. insert :

Willkommia Hack. Spikes densely flowered ; empty glumes flat, not keeled,
one-nerved. Flowering glume with pointed, short-haired callus at the base,
hyaline, short-awned. Palea somewhat shorter than the flowering glume,
obtuse. Lodicules wanting.
Species three, in southwestern Africa.
Before Tripogon Roth, place :

Pentarraphis Kiinth. {Polyschisfis Presl.) Spikes or rather groups of spike-
lets fascicle-like, consisting of 1-2 spikelets, and 1-3 awn-like, often two-cleft,
rudiments of a second or third, loosely arranged on the main axis of the inflo-
rescence. Spikelets 3-flowered, the upper flower generally staminate, the lower
hermaphrodite, the first empty glume of the spikelet very narrow (in the dried
condition), awn-like; in case but one fertile spikelet is present in each group
the first empty glume of the same forms, with the 4-5 awn like rudimentary
glumes, an apparently lateral fascicle of awns (resembling a deeply 5-cleft
glume), when there are two fertile spikelets there is between them a fascicle
of 3-4 awn like glumes. The flowering glume is always 3-awned. Low, turf-
forming grasses.

Species two, in Mexico. (See Scribner in Bull. Torr. Bot. Club, 17: 131,
plates 107 and 108. )

Lepidopironia Rich, is reduced to a synonym of Tetrapogon.
Opizia Presl, Hackel amplifies and recharacterizes as follows :
Staminate spikelets in 2-5 spikes, like those of section 1 of Bouteloua ; pistillate
spikelets in two rows, on short spikes, which are half hidden in the sheaths of
the lower leaves, one-flowered ; the first empty ghinie short, the second as large as
the flowering glume, the latter 3-clef t, 3-awned ; palea shortly 3-toothed or 2-lobed.
Above the fertile flower 1-2 sterile glumes, either 3 or manyawned. Low,
creeping grasses. (See Scribner in U. S. Dept. Agr. Div. Agros. Bull. 4: 9, f. 4.)
Sect. 1. Euopizia. First enapty glume of the spikelet very small. Flowering
glume of the pistillate flower as well as sterile glumes with 3 long awns. O.
stolonif era Presl. (Fig. 4.) Mexico.

Sect. II. Privglcochloa Scribn.* (as a genns). First empty glnnie of the pis-
tillate spikelet linear, little shorter than the second; flowering glume of the
fertile flower shortly 3-cleft; the 2-3 sterile ones many awned above. O.
Pringlei (Scribn.) Hack. (Fig. 5.) Mexico.

The characters of OrcuttiaY-Asey, should be modified as follows : Cfpspitose,
spreading annuals with large, many-flowered spikelets in terminal spikes ; flow-
ering glumes broad, many -nerved, and toothed or lobed at the broad apex.

Species two, in Southern California.

After Diplachne insert :

Pogochloa Moore. Spikelets laterally compressed, 5-flowered, two-ranked on
long panicle branches (the rows approximate below), sessile. Empty glumes
unequal, the lower 8-5-nerved, the upper
with 9 strong nerves. Flowering glumes
boat-.shaped, keeled, 3 nerved, shortly two-
toothed or almost entire, with a long, erect
awn between the teeth.

Species one (P. hrasilieusis Moore). In
the state of Matto Grosso, Brazil.

Under Eragrostis there is to be added
under

Sect. I. C'ataclastos. Cladoraph is French,
(as a genus), is a one flowered depauperate
form of E. spinosa Nees.

Ipnum Phil, is reduced to Diplachne.

After Eragrostis add :

Halopyrum Stapf. Spikelets many-flow-
ered, in narrow panicles with a hairy axis
breaking up at maturity. Empty glumes
shorter than the flowering ghimes, the lower
3- the upper 5-nerved. Flowering glumes
firm in texture, keeled, 3-nerved, mucronate-
pointed. Caryopsis broadly furrowed in
front. In other respects like Eragrostis,
with the section Platystachya of which it is
closely related. Bentham placed the species
belonging here in Eragrostis, Sect. Seleros-
tachija, a conglomeration of species not
closely related, which ought to be completely broken up.

Species one {H. mucronatum Stapf., Brizopyram mucronatum Nees, accord-
ing to ^i'AY>f.^Uniola nnicronata L. V) distributed along the coast of the
Indian Ocean from Ceylon to the mouth of the Zambezi ; a grass with rigid,
convolute leaves.

AntJi.ocJiloa Nees (in PI. Meyen, 164). (Stapfia, and Neostapfia Davy, and
Davycila Hack. ) Panicles densely flowered, ovoid or cylindrical and spikelike;
spikelets 2-5-flowered; flowering glumes thin membranaceous, broad, flabelli-
form, finely toothed and minutely ciliate-fringed, awnless ; lower glumes narrow,
sometimes wanting.

Species three, two South American; one, A. colusana (Davy) Scribn., from
California.

After Lasiochloa Kunth, add as a synonym Triboliuvi Desv.

Fig. 4.—()}jizia stoluiiifcra: a, female
spikelet; h, empty glumes; c, second
empty Kl«me; d, first empty glume; c,
palea and adnate sterile nxdimentary
floret (/); g, pistil; /), caryopsis.

* (Bot. Gaz. 21 : 1.37, plate 13.) MoncBCiou8,staminate and pistillate spikelets unlike. Staminate
spikelets one-flowered, in racemose, unilateral spikes on erect toranclies; pistillate spikelets on
short, basal branches: flowering glumes 3-awned, with rudimentary, many-awned empty
glumes above. Low, stolonit'erous perennial.

8

Fig 5-OjM-2ia prinalei : a, staminate spikelet, showing the back of the second glume, etc.;
h empty glumes of the staminate spikelet, more highly magnified; r, lower part of the stammate
floret showing the short continuation of the rachilla behind the palea ; d, apex of the flower-
ing glume of the staminate spikelet ; e, first glume of the pistillate spikelet ; /, second glume
of the same; a, pistillate spikelet, dorsal view, empty glumes removed; h, palea of the female
floret ; i, fourth glume of the pistillate spikelet, the glumes above these have successively fewer
awns; k] rachis, to which were attached three female spikes.

9

After C'///io.s?/r»,s- insert :

To section Phaloiia belongs Pteriinn Desv. (as a genus). For LtDiiavchkt
Moench., O. Kiintze has brought forward the older synonym Avhiivodea Boeh-
nier. 1760.

After Fesfiica insert:

Littledalia Henisley, probably belongs to Fe.slura (as a section?); from the
description and illustration there appears to be no other difference than the
T-nerved (in Festuca 5-nerved) flowering glume. Of these 7 nerves 3 extend
almost to the obtuse, unawned apex. Lltfh'ddJiii lihcticn Hemsh-y is a note-
worthy grass 2.5 cm. high, with large si)ikelets.

After Bromus add :

Duthiea Hack. Spikelets ;5-r) flowered, in simple, compact, almost capitate
racemes, short-pediceled, the pedicels of the lowest often with sul)t('nding
bracts ; empty glumes 5-7nerved, nearly as long as the first flowering glumes
or longer. Flowering glumes rounded on the back, many-nerved, two-lobed,
with a somewliat geniculated awn between the lobes, which is slightly
twisted below. Lodicules wanting. Ovary long and densely h airly, with a
style which is cleft above into very long, threadlike, projecting stigmas that are
covered with short hairs.

Species two: D. hroiiio'nles Hack, in Cashmere and D. oligostacJtya (Munro)
Stapf , in Afghanistan.

Enyaphi>i Trin. (as a section) is a synonym of Boissiera Hochst.

Instead of Kralikia Coss. Dur., O. Kuntze has introduced the name ^rca?i-
gelina. On account of the homonymous older genus of Covqiusitif of Shultz,
Cosson and Durieu had already changed the name to Kraiikella on the jjrinted
labels of the Dauphinese Society. Since, however, in Hofmann's work on the
Compositce the genus of Schultz is suppressed, the name may remain in use for
tlie genus of Gniin'mete.

Under Jouved Fourn. may be included a grass from Lower California described
by Vasey and by him made the type of a new genus, Rhachidospermum. (See
Scribner, in U. S. Dept. Agr., Div. Agros. Ball. 4: U.)

Under Moiwriiia Beauv. add as a synonym Ojthiuvinclht Desv.

Under HaynaJdia add the following note:

Instead of Hayiialdia Schur., Durand has introduced the name Dasypymim,
used by Cosson and Durieu as the name of a section, on account of the homony-
mous fungus genus of Schulzer-Mueggen. However, the latter originated the
same year (180G) as that of Schur. and it is not evident which name was pub-
lished first ; moreover, Haynahlia Schulzer is ignored by modern mycologists,
as for example, Saccardo and Schroeter.

Under Aniiidinarid add as a section, Ghtziophyton (Franch. as a genus),
which is connected with the section Tliamnocidanms on account of the sub-
tending leaves of the spikelets and panicle-branches, but which bears the panicles
on leafless, nodeless culms, breaking up by means of separation nieml)ranes,
while the leafy culms (and not rarely also the leafless) remain sterile, similar to
.4. faleata and A. khasiana, species found in the Himalayas.

Species one (^4. mirabilis (Franch.) Hack.) in Brazil.

After Arundinaria add:

Microcalamus Franch. Spikelets only 2-flowered, the upper flower hermaph-
rodite, the lower staminate or reduced to the two glumes, without a sterile
flower above the hermaphrodite. Empty glumes 2, short; flowering glumes of
the lower flower similar to the empty glume, that of the hermaphrodite flower
longer, narrower, keeled, bow shaped. Palea naked on the keels. Styles 2. long
and slender, free, their stigmas short, broad plumose. Herbaceous, not more

10

than 0.5 m. high, with rootstock, slender culms, and lanceolate leaves, the short
stalks of which are articulate with the sheath. Panicle few-flowered, terminal.

Species one (M. barbinodis Franch. ). In the Congo region.

After Phyllostachys add :

Fargesia Franch. This genns differs from Phyllostachys, of which it will be
much better classed as a section, principally on accoimt of the sessile (in
PliyUostachys jiedicellate) fruit-nodes, and ?> styles separated to the base (in
PliyUostuchys being grown together for quite a distance).

Species one {F. spathacea Franch.), in China.

Under Bambusa Schreb. insert Arundarbor Rumph. as a synonym.

To Bambusa probably belongs a form first described by Gamble as a new
genus Microcalamus (non Franch., whose name is older), then referred to
Arundinaria (as A. prainii). Habit of Arundinaria, but with 6 anthers. Seems
to be closely related to Bambusa senanensis Franch.

Sect. III. GuadneUa Franch. (As a genus) according to Franchet's more
recent and complete statements must be restored to generic rank.

Before Atractocaipa Franchet insert:

Thyrostachys Gamble. Spikelets two-flowered, aiTanged in twos or threes
along the panicle-branches, subtended by large bracts. Empty glumes 2 ;
flowering glumes broad, many-nerved. Palea of the lower flower 3-keeled,
2-cleft almost to the middle, with subulate divisions, those of the upper flow-
ers romided on the back, slightly or not at all cleft. Lodicules 2 ; stamens 6,
free. Ovary on short pedicel, depressed turbinate. Style long, with 3 plimiose
stigmas.

Species two, in Burniah.

Oreobambos K. Schum. Spikelets two-flowered, arranged in many-flowered
whorls, subtended by 2 broad bracts; empty glumes 1-2; flowering glumes
broad, many-nerved. Palea generally two keeled, rarely rounded on the back.
Lodicules wanting. Stamens 6, free. Ovary on a short, naked gynophore.
3-cornered-ovate, smooth, produced or attenuated into a long, hairy, undivided
style with simple (V) stigma.

Species one (O. Biichicaldii K. Schum.) in Usambara.

Before Subtribe C. Dendroealameoi add: Bonia Balansa, a doubtful genus of
the Eubambusece, is characterized as follows : Spikelets 3-4, sessile on the panicle-
branches, 3-4-flowered ; the empty glumes separated from the nearest flowering
glumes by a naked internode 1 cm. long, which they surround like a sheath ;
lowest flower articulate with this internode. Flowering glumes 7-nerved.
Palea 2-keeled. Lodicules 3. Stamens 3. Styles 3, grown together to the mid-
dle. Stigmas plumose, elongated. Caryopsis cylindrical.

Species one {B. tonkinensis Bal.) in Tonkin. The genus is very closely related
to Bambusa; the significance of the two lowest bracts (called empty glumes
by Balansa) is doubtful.

Giucii.Aii No. lii.-lAgros. 50.) Issued .July 1, \m).

United States Department of Agriculture,

DIVISION OF AGROSTOLOGY.
[Grass and Forage Plant Investigation.]

NEW SPECIKS OF NORTH AMERICAN (BRASSES.

Ill Circular N(j. '.» of this Uivision there was published i:) new
species of North American grasses belonging chiefly to the genus
Poa. Circular No. 10 was also technical in its character, being
devoted to Poa fendleriana and its allies. In the present circular
nine new or little known species of grasses are descri])ed.

1. ANDROPOGON TENNESSEENSIS Scribner. u. comb. (.1. provin-

cidlis var. tennenseensis Scribn. Bui. Tt'iin. Agr. Expr. Sta. 7: 2;J. 1894.)
Pedicellate splkelet iisually hennaplirodite, H-K) mm. long, the 1st glume lanceo-
late, acuminate, rounded on the back, ll-i:5-nerved, apex acute, snbaristate
( If minutely bimucronate, scabrous all over the Imck and especially on the
nerves and margins. Second glume T-9-nerved, lanceolate, acute or snb-
aristate. scabrous all over the back and ciliate along the hyaline, infolded
margins above. Third glume about as long as the outer ones, lanceolate,
acute, scabrous on the back above and fimbriate-ciliate along the margins.
Fourth glume sh(jrter than the third, apex bifid, ciliate on the margins
above, awned. Awn slender, a little twisted, '2-4 mm. long. Palea i-'i as
long as its glume. Stamens and pistil as in the sessile spikelet. Sessile
spikelet about 8-10 mm. long, more than twice as long as the pedicels. First
glume rigid and very rough-scabrous all over the back, ciliate-scabrous along
the keels above. Second glume compressed and strongly keeled, long-acu-
minate pointed, scabrous on the sides and very rough on the keels, ciliate
on the narrowly inflexed margins above. Third glume a little scabi-ous on
the back above. Racemes as in A. j^rovincialis. Hairs on pedicels and
joints yellow. Pedicels and outer glumes very rough scabrous.
Open fields. Knoxville, Tenn. Rare.

2. PUCCINELLIA SIMPLEX Scribn., sp. nov. (Fig. 1.)

A slender, densely csespitose, erect or spreading, glabrous annual, 0.5-2 dm. high,
usually branching near the base, with soft, narrow leaves and simple, race-
mose panicles 2-8 cm. long. Sheaths loose or somewhat inflated, mostly
longer than the internodes; ligule 2.5-8.5 mm. long, acute, hyaline; leaf-
blade 2-3.5 cm. long, flaccid, 1-2 mm. wide, mimitely scabrous toward the
apex, otherwise glabrous. Spikelets solitary, or 2-3 on short, appressed
branches, 2-3 flowered, about 5 mm. long, with unecpial, acute, glabrous
empty glumes, the second one 3-nerved, and acuminate flowing glumes
which are 5-nerved, inibescent on the back and 3-3.5 mm. long.

Woodland, California, collected by J. W. Blankinship, May fi, 1893. This species
is entirely distinct from any other with which I am acquainted. Its slender,
somewhat wiry culms, rigid, strict panicles and pubescent, acute flowering
glumes are its most striking characters.

2

3. DISTICHLIS TEXANA Scribn., n. comb. {Poa texana Vasey, Contrib.

Nat'l. Herb. 1: 60, 1890; Sieglingia wrightiiYasey 1. c, 269, 1893).
A rather stout, rigid, glabrous, creeping perennial, the upright branches 3-6 dm.

high, with flat leaves and densely flowered, narrow panicles 10-30 cm. long.

Staminate spikelets compressed, 5-7-flowered ; pistillate spikelets, subterete,

Fig. l.—PuccineUia tiimidex Scribn. a, A portion of tlie axis with the empty glumes ; />, Spike-
let, empty glumes removed ; c. Floret, showing a portion of the rachilla ; d, Palea ; e, Grain.

fusiform, 12-25 mm. long. Western Texas, New Mexico (1038, C. Wright).
[Mexico, No. 507, E. Palmer, 1898.]
The dioecious habit, the character of the inflorescence, the rigid subcoriaceous
glumes, long-exserted styles protruding from the apex of the floret, and
grain inclosed in the coriaceous base of the palea, clearly connect this grass
with Distichlis.

I

4. POA CURTIFOLIA Scribn., sp. nov.

A sleuder, densely ciespitose, uprig-ht perennial 1-H dm. high, with short, but
strong, creeping root-stocks, flat leaves and narrow panicles 4-8 cm. long.
Sheaths rather loose, glabrous; ligule 8-5 mm. long, lanceolate, hyaline,
acute: leaf blades 1-4 cm. long, those of the culm usually 1-2 cm. long, 2-3
mm. wide, scabrous on both sides and along the margins, especially near
the involute and somewhat hooded apex, uppermost leaf often very short or
nearly wanting. Spikelets about H mm. long, 8-")-flowered. usually :$-flower-
ed; empty glumes broadly lanceolate, acute, 3-nerved nearly eciual, the first
.")-() mm. long: florets rather remote on a slender, naked rachilla; first flow-
ering glume, about 0 mm. long, broadly lanceolate or oblong, acute, 5-nerved,
glabrous. Palea a little shorter than its glume, ciliate-scabrous along the
keels excepting near the base. Mount Stuart, Kittitas County, "Washington,
August, 1S98, Nos. 1148 and 1150 A. D. E. Elmer.

This species is remarkable for its short, blunt leaves resembling those of the Poa
alpina, from which, however, it is at once distinguished by the characters
presented by its spikelets.

5. DACTYLOCTENIUM AUSTRALIENSE Scribn.. sp. nov. (Fig. 2.)

A ca?spito.se, difliusely spreading, more or less branching annual, 1.5-2 dm. high,
with short, loose sheaths, flat leaves, and short spikes at the apex of the culm
or its branches. Culm and sheaths glabrous. Ligule short, ciliate- fringed;
leaf -blade 5-10 cm. long, flat, acute, ciliate on the margins near the base,
with a few scattering hairs on both surfaces springing from white papillae.
Spikes 3-6, 0.5-1 cm. long, densely crowded so that the inflorescence appears
capitate. Axis of the spikelets projecting a little beyond them, strongly
ciliate-scabrous on the keel and margins, which are white— the intermediate
space on either side being green. Spikelets 2-3-flowered, about 4 mm. long.
Second empty glume oblong, broadly obtuse, with a scabrous awn about
1 mm. long. First flowering glume about 3 mm. long, 3-nerved, strongly
keeled scabrous toward the apex, which is shortly acuminate-pointed.

Cultivated from seed collected in Ooduodatta, South Australia, received from
A. Molineaux.

Chiefly distinguished from Dactyloctenium cpgyptiacum by its more caespitose
and more depressed habit of growth, and especially by its much shorter
spikes, which are crowded into nearly globular heads at the apex of the
culms. The whole plant is also more slender and the axes of the spikes are
distinctly winged-margined.

6. PANICUM OVINUM Scribn. and Smith, sp. nov.

A slender, erect or ascending, csespitose perennial, 3-4 dm. high. Nodes 4 or 5,
glabrous, sheaths closely enveloping and shorter than the internodes, ciliate
on the overlapping margins above, otherwise smooth. Ligule a line of white
hairs about 1 mm. long. Culm leaves rather rigid, linear-acuminate, 7-9 cm.
long, 2-4 mm. wide, ascending, glabroLis beneath, scabrous on the margins
toward the apex, closely striate, but without prominent lateral nerves or
midrib except at the base below, basal leaves lanceolate, acute, 4-5 cm.
long, 6-8 mm. wide. Panicles terminal, 5-8 cm. long, shortly ex.serted, few-
flowered ; branches flexuous, single or in pairs, ascending ; pedicels widely
varying in length. Spikelets about 2 mm. long, elliptical-obovate, obtuse;
first empty glume less than one fourth the length of the spikelet, broadly
ovate, obtuse, glabrous; second and third glumes obovate, obtuse, promi-
nently 7-nerved and minutely pubescent between the nerves. Flowering
glumes nearly 2 mm. long, elliptical, ovate, obtuse, with a conspicuous
depression near the base.

Fig. 2.—Dactyloctenium auMraliense Scribn.— a, iimpty glumes; h, Spikelet witla empty
glumes i-emoved ; c, Palea ; '?, Spikelet of D. cvgyptiacum ; e, Axis of spike ; /, Axis of spike of
D. (cgyptiaann.

I

I

Typp collet-ted by F. W. Thnrow, Waller County, Texas, May 2."), 189s. Closely
related to J\inicum angiisffoliinn Ell., from which it is distingiiished by its
much smaller and minutely pubescent spikelets and the smooth sheaths and
leaves.

]. PANICUM INFLATUM Scribn. & Smith, sp. nov.

An erect or ascending, much-branched perennial, 2.5-4 dm. high, often rooting
at the lower nodes, with glabrous culms, pubescent nodes, loose or inflated
sheaths, lanceolate, spreading, obtuse leaves 5-7 cm. long, and ovate-pyram-
idal, exserted, rather densely flowered panicles 6-7 cm. long. Sheaths
longer than the inteniodes, ciliate-piibescent along the margins and tuber-
culate-dotted between the strijie; leaf -blades 8-10 mm. wide, abruptly
rounded at the base, very smooth on the back, scabrous on the margins and
along the nerves above. Axis of the panicle slender and the flexuous branches
sparingly glandular viscid. Spikelets 1.8 mm. long, broadly obovate, or
nearly spherical. First empty glume ovate, obtuse, less than one-fourth the
length of the sinkelet ; second and third empty glumes orbicular, obovate,
obscurely 5-7-nerved, the second a little shorter than the bro.adly ovate,
obtuse, flowering glume, which is equaled by the third : the culms, sheaths,
and empty glumes are purple.

Type No. 4(522, S. M. Tracy, collected at Biloxi, Miss., October, 1898. Closely
related to Panicum Undheimeri Nash and resembling it in both the fonn of
the panicle and the spikelets, but readily distinguished by its tuberculate,
inflated .sheaths, glandular panicle- branches, and obtuse leaves.

8. PANICUM THUROWII Scribn. & Smith, .sp. nov.

A rather stout, erect or ascending, pubescent or villous, .simple or sparingly
branched perennial, 8-4 dm. high, with 5-6 densely white-bearded nodes,
sparingly villous sheaths, lanceolate-linear leaves 7-10 cm. long, and ovate-
pyramidal, shortly exserted panicles, 7-8 cm. long. Sheaths shorter than
the villous internodes, striate and tuberculate above between the stria';
ligule a ring of stiff, white hairs 3-4 mm. long; leaf -blades flat, 7-10 mm.
wide, softly pubescent on the lower surface, glabrous above, unu.sually
ciliate on the scabrous margins toward the ba.se. There is a very densely
villous line on the back where the blade joins the sheath. Axis of the pan-
icle sparingly pilose ; the branches subflexuous, villous at the base. Spike-
lets narrowly elliptical, obtuse, 1.5-2 nmi. long. First glume one-fifth the
length of the 7-nerved, puliescent and obtuse second and third .glumes which
are .slightly .shorter than the lanceolate-elliptical, acute floral glume.

This species is named for Mr. F. W. Thurow, by whom it was collected in Wal-
ler County, Texas, June 5, 1898, No. 9. No. 11 collected May 10 is the same
species.

Very closely related to Panicum inibescens Lam., differing in its more densely
flowered, narrower panicles, smoother spikelets and simple culms. The
leaves and spikelets are purplish.

Also collected by H. W. Ravenel, Houston, Texas, 1869.

9. PANICUM OCTONODUM Scribn. & Smith, sp. nov.

A slender, ascending, unbranched, glabrous perennial 4-6 dm. high, with 8 or 9
brownish, smooth nodes, rigid, linear, acuminate, ascending or spreading,
flat leaves, 5-10 cm. long, and exserted, rather densely flowered, narrowly
pyramidal panicles 5-12 cm. long. Sheaths shorter than the internodes,
sparingly ciliate on the overlapping margin above; ligule a dense fringe of
white hairs, about 1 mm. long; leaf blades 4-7 mm. wide, obscurely 9-nerved
below, scabrous on the margins, acute. Axis of the panicle smooth,

brandies subflexuoiis, solitary, often fasciculately branched just above the
base. Spikelets 1.5 mm. long, elliptical ovate, obtuse, first empty glume
muiute, the faintly 5-7-nerved, obtuse second and third glumes slightly
exceeded by the obtuse, perfect floret.

Waller Coimty, Texas. Collected by F. W. Thurow, May 5, 1898.

Closely related to Panicnm iiitichnn Lam., from which it may be separated by
the greater number of nodes of the culm, the smooth leaves and sheaths,
and the glabrous spikelets. No. 565 Lindheimer (184r>) approaches this in
habit and leaf characters, but has the ovate, obtuse panicles abruptly
rounded at the base, the branches horizontal and more flexuous, and the
minutely puberulent second and third empty glumes exceeded by the flow-
ering glume.

F. Lamson-Scribner,

Agrostologist.

Approved :

James Wilson,

Secretary of Agriculture.

Washington, D. C, June 16, 1899.

Circular No. 17.— (Agros. 53).

United States Department of Agricnlture,

DIVISION OF AGROSTOLOGY.

[Grass and Forage Plant Investigation

CRIMSOX

DESCRIPTION

nown as scar-

Crimson clover {Trifolium incarnatum L.),
let clover, German
clover, Italian
clover, and carna-
tion clover, is an
annual plant na-
tive to southern
Europe and has
long been culti-
vated as a forage
crop in the warmer
portions of that
country. It is an
erect, tufted plant,
1 to 2 feet high,
with soft - hairy
stems and leaves,
and usually bright
scarlet flowers in
elongated heads.
The root system is
^vell developed and
penetrates deeply
into the soil and
the plant is a vig-
orous grower. The
seed is larger than
lliat of red clover,
oval in shape,
bright reddish-yel-
low when fresh
(paler in the white-
flowered variety)
and has a highly
polislied surface. The plants stool freely, many stems arising from
a single root.

Fig. 1.— Crimson Clover.

2

Several varieties are recognized in Europe, distinguished chiefly
by the color of the flowers and the time required by the plants to
reach maturity. Very little attention has been paid to these different
varieties in this country. A white-flowered variety is sometimes
seen on the market, but as a general rule only one sort is recognized
by seedsmen and this is usually sold under the name of crimson
clover.

CONDITIONS OF GROWTH.

Crimson clover will not stand severe freezing although it is one of
the so-called "winter annuals" and under favorable conditions
makes much of its growth during the cool, moist weather of fall,
winter, and early spring. Its strong-growing roots enable it to
secure nourishment in many soils so poor and thin that red clover
would fail entirely. While requiring a warm climate, it will not
endure severe drought. It thrives best on rich, rather sandy loams ;
but, when the conditions of moisture and temperature are favorable,
it gives good results on light sandy soils as well as on clays if they
are not too stiff and cold.

Crimson clover has come into prominence in this country within
comparatively recent j^ears. It can hardly be regarded as a success-
ful crop outside of the region from New Jersey west to the Allegheny
Mountains and south to eastern Tennessee and Texas. True, good
yields are often obtained in other sections, but can not be depended
upon year after 3'ear. In the middle and south Atlantic States this
clover is one of the best crops that can be grown for forage and soil
renovation. It has given good results in many portions of the Gulf
States but many failures are also reported. In the colder sections of
the country this clover is sometimes successfully grown as a summer
crop but it usually winterkills badly when sown in the autumn. At
the experiment stations in Rhode Island, New York, Ohio, Michigan,
Illinois, South Dakota, Nebraska, and other States in the North and
West, the general results of tests show that it is too tender for the
climate and is less valuable than red clover. From results recently
obtained at the Alabama Experiment station it seems very likely that
in many cases, especially in the South, failures with this crop are to be
attributed to the absence from the soil of the tubercle-forming organ-
isms which are necessary for the proper appropriation of nitrogen
by the plant. These organisms being supplied to the soil, excellent
crops were obtained where without them the result was a failure.

CULTURE,

Preparation of soil. — There are many ways of preparing the
soil for this crop, depending largely upon the purpose for which it is
intended. Little if any fertilizer is needed, except what may be

3

used in connection with the clover to prepare the land for the next
crop in the rotation. On very poor, worn-out soils it is usually a
good plan to give a moderate application of phosphate and potash.
When crimson clover follows a cultivated crop like corn no prepara-
tion is necessary other than that of the cultivation of the corn or
other crop. When the clover follows a crop of small grain it is
usually necessary to plow the land and this should be done just
before the time for sowing the seed. Cloddy land should be well
pulverized before seeding. A fine seed bed is essential to uniform
germination.

Seeding. — It is extremely important that a farmer should sow
none but fresh seed. It is an easy matter to test the germinating
power of the seed; it requires but a very short time and is well
worth the while of every farmer intending to sow in any quantity*.
Good fresh seed should germinate 90 to 95 per cent and under
proper conditions the sprouting should take place within three days.
American seed has given much better results than the imported, on
account of its greater vitality. The amount of seed sown per acre
varies from 10 to 20 pounds, depending largely upon the character
of the soil and the use to which the crop is to be put. Rarely a
smaller quantity may be used. The common practice is to use about
12 or 15 pounds.

The seed may be sown broadcast or with a drill. The former is
the most common in practice, perhaps largely from the fact that the
clover is so often sown on land already occupied by corn, potatoes,
or some other crop. The seed is often sown by hand, sometimes
from horseback, but better results will usually be secured by the use
of some machine like the Gaboon broadcast seeder.

Throughout the middle Atlantic States and the South generally
crimson clover may be sown any time from July 15 to September
15, and if the moisture conditions are favorable it may be sown
still later in the Gulf States. As a rule the best results are obtained
from seed sown in July and August. The seed may also be sown
in the spring, but, except in the colder northern States or for some
special purpose, the results are much less satisfactory. In the
North, spring sowing is necessary if this crop is grown at all. It may
sometimes be grown to advantage in this section as a temporary
crop in place of red clover.

Some farmers do not cover the seed at all ; but many failures have
resulted from following this method, attributed by some to heavj^
rains immediately after the seed is sown but more likely due to the
exposure of the tender, unprotected plants to the hot summer sun.

* A full discussion of the selection and testing of crimson clover seed will be
found in Circular No. 18, Div. Bot., U. S. Dept. Agric, August, 1899.

When its use is practicable, the roller is an excellent implement for
covering the seed. It presses the seed into the soil and insures
uniform germination. A light harrow or a brush drag may also be
used.

Harvesting. — Crimson clover may be harvested in the same way
as common red clover. It is sometimes difficult to cure properly,
as it is ready to cut at a time when wet weather is likely to inter-
fere. It may be cut for hay as soon as it comes into bloom and
should never be cut for this purpose later than when it is in full
bloom . The hairs upon the calyx of the flowers become hard and
stiff as the plant matures and are likely to prove troublesome to
animals eating the hay, forming "hair balls" in the stomachs of the
animals, which sometimes cause death. When harvested for the
seed, cutting should take place as soon as ripeness has been reached
to avoid loss through shattering. It is also well to thrash the crop
or put it under cover as soon as it is dry, as heavy rains may injure
the seed by causing it to sprout in the head.

Under average conditions from 8 to 15 tons of green, or 1^ to 3
tons of cured forage may be obtained from an acre of crimson clover.
Very poor Soil or drought may result in a lighter yield, while as high
as 20 tons are reported under favorable conditions. The yield of
seed per acre usually ranges from 8 to 12 or 15 bushels.

USES.

For hay. — When cut in proper season and well cured, the hay is,
if anything, superior to that of red clover, having much the same
chemical composition and a somewhat higher percentage of digesti-
bility. It is relished by all kinds of stock and is very highly prized
for feeding to animals that are required to do heavy work. Its
nutritive ratio varies from 1 :3.5 to 1 :4.

For soiling. — Crimson clover is an excellent crop for this purpose.
It is ready for use some time before red clover and at a time when
there are few other forage crops at hand. It is particularly valuable
for this purpose on dairy farms. In experiments made at the New
Jersey station, nearly one and one-half tons per acre of digestible
food, valued at $25.00, were secured. A ton of crimson clover in
proper condition for soiling contains about 325 pounds of dry matter,
of which about 5^ pounds of crude fat, 50 of crude protein, and 150
of carbohydrates are digestible. Rape is sometimes sown with the
clover when a soiling crop is desired and the resulting forage is
excellent.

For pasturage. — This crop readily lends itself to use for pastur-
age, especially in the early spring. It may be used alone or in connec-
tion with winter rye or rape. At the New Jersey experiment station

it was estimated that an acre of crimson clover six inches high con-
tained "sufficient food to properly nourish twelve cows for one
week."

For silage. — This is one of the best of the clovers for use in the
silo. The yield of forage is large, is easily handled, and it makes a
better quality of ensilage than most other legumes commonly grown
for this purpose. The silage is especially valuable for feeding dairy
stock.

For green manure. — This is one of the most important uses to
which this crop can be put. Its season of growth is such that it can
be used without in any way interfering with the production of the
primary crop of grain or vegetables and it affords a large amount of
fertilizing material. The herbage is heavy, the roots are abundantly
produced and penetrate deeply into the soil, and together these form
a large amount of vegetable mould, exerting a beneficial effect on
the physical condition of the soil as well as adding much nitrogen
and other valuable elements of plant food to the surface soil where
it will be available to corn, wheat, and other crops. It is estimated
that the average crop of crimson clover is worth from $15 to $30 per
acre for the fertilizing value of the nitrogen alone. An important
advantage which crimson clover has over cowpeas as a fertilizer lies
in the fact that it decays more readily and is less likely to "burn"
the soil when a heavy crop is turned under. Another advantage
arises from the possibility of producing the crop of clover between
the time of harvesting the regular crop of one season and the plant-
ing of the next. It may often be used along with cowpeas to good
advantage, the cowpeas being used as a summer crop and followed
by the clover for the winter.

As a soil cover. — In many sections of the country, especially
where the soil "leaches out" or washes badly, it is imperative to
have the land covered with vegetation during the time it is not occu-
pied with the primary crop, particularly during periods of heavy
rains. Crimson clover is peculiarly well adapted to use as a cover
crop. It may be sown in corn, tobacco, potatoes, cowpeas, sorghum,
'and many other crops after the last cultivation and, when these are
removed, will come on and occupy the land during the fall, winter,
and early spring and may be pastured off or plowed under in time for
the planting of the next crop. It is an excellent cover crop for use
in orchards, where it is also one of the best of sources of nitrogen for
the trees. Crimson clover in the orchard reduces the amount of
cultivation necessary to keep the weeds in check, and if the crop is
not needed for fertilizing the soil, it can be cut for hay, soiling, or
silage. The roots and stubble left on the ground from a crop of
clover cut when in full bloom at the New Jersey station contained

6

nearly 40 pounds of nitrogen, over 10 of phosphoric acid and over

14 of potash per acre. When the crop is allowed to mature the pot-
ash in the roots and stubble is increased but the nitrogen and phos-
phoric acid is reduced. This crop can be used in connection with
small fruits as well as with peaches, pears, and apples.

SUMMARY.

1. Crimson clover is an annual, not adapted to use in permanent
meadows and pastures and too tender for successful general cultiva-
tion outside of the Middle and South Atlantic and Gulf States. It
thrives best in warm, moist loams of at least moderate fertility and
makes most of its growth in the fall and early spring.

2. The seed should be sown in late summer or early autumn (July

15 to September 15) at the rate of from 10 to 20 pounds per acre on
a well prepared seed bed. It ma}'- be sown alone or with other
crops, a frequent practice being to sow with corn, potatoes, and like
crops at or after the last cultivation. The seed may also be sown in
the spring, but the results are not as satisfactory as when sown as
above stated.

3. Crimson clover has a high feeding and fertilizing value and is
one of the best crops that can be grown in short rotations for forage
and soil renovation, lending itself readily to use for hay, pasturage,
soiling, silage, green manure, and as a soil cover to prevent leaching
and washing.

4. The crop should be cut for hay at or before full bloom, and for
seed as soon as ripe ; in the latter case it should be thrashed or put
under cover as soon as dry,

Thomas A. Williams,

Assistant Agrostologist.
Approved :

James Wilson,

Secretary of Agriculture.

Washington, D. C, August 19, 1899.

O

Circular No. 18. (Agros. .54.)

United States Department of Agriculture,

DIVISION OF AGROSTOLOGY.

[GRASS AND FORAGE PLANT INVESTIGATIONS.]
r. Lamson-Scribner, Agrostologist.

SMOOTH BROME-fiJRASS. S>

{Bromus inermis.)

Smooth brome has been known in Europe for over one hundred
and thirty years. The early agriculturists did not consider it of any
value because of its creeping rootstocks, thus resembling the dreaded
couch grass. In 1884,
however, Stebler and
Shroeter demonstrated by
experiments conducted for
thirty years at Magocs,
Hungary, that it had the
power to withstand long
periods of drought when all
other grasses succumbed.

These experiments re-
sulted in introducing the
grass into cultivation
Hungary where it still re-
tains the^ngjaae^ Huhga
riaii.-bfomeVgralis. ^^ ,
tr^LuctionAli^ \]bjlie Jliiite*
States\ relist "i''l)^^i^ ^akm
place aboH^^he same^i4eL
for we fino^s^tf^^Sfet
offered for distHbtif
Bulletin 22 of the Califor-
nia Experiment Station,
issued November 5, 1884,
and the statement made
that ' ' our experience indi-
cates that it will lo well
here either with o^ without
irrigation." S^"ace then
the experime? .c stations of
Colorado, Minnesota, Man-
itoba, Nor^Ji Dakota, South Dakota, and the different grass stations
of the Agricultural Department, Washington, D. C, have conducted
extensive experiments and all speak of it in the highest terms.

Fig. 1.— Smooth brome-grass showing rootstocks, seed-
head, splkelet, and parts of flower.

GENERAL CHARACTERS.

Smooth brome-grass is a native of Europe and Asia, ranging from
France eastward into Siberia, growing along roadsides, river banks,
borders of fields and woods, and upon sterile hillsides and pastures.
It is a vigorous, hardy perennial with strong, creeping rootstocks,
smooth, upright, leafy stems, one to four feet high, and loose,
open panicles, or "seed heads" four to eight inches long. In a few
years it forms a very tough sod, soon crowding out other grasses,
clovers, and weeds. Its remarkable drought-resisting qualities have
proved it to be the most valuable grass for dry regions where other
grasses could hardly exist.

As it is thoroughly permanent and grows with wonderful rapidity,
producing heavy crops and luxuriant pasture, its value to the far-
mers of dry regions can not be over estimated. All kinds of stock
eat it with relish and the chemical analyses made show that it is
rich in flesh-forming ingredients, much more so than timothy. It is
very hardy and not injured by severe spring and fall frosts when
once established. As it starts to grow very early in the spring be-
fore any of the grasses upon the native prairies show any signs of
life, and remains green and succulent far into November, it will
supply the long-felt want of early spring and late fall pastures.

DISTRIBUTION OF SEED.

Since 1896 six hundred and three experimenters have received seed
of smooth brome-grass from the Department of Agriculture, Wash-
ington, D. C. This distribution does not include the State experi-
ment stations to some of which large quantities were sent. The
packages sent out varied in amount from one quart to thirty-five
pounds. The larger part of this seed was purchased from Russia
by the Secretary of Agriculture and distributed directly by this
Division, or through the Section of Seed and Plant Introduction.
Almost every State in the Union is represented in this distribution
as shown by the following table :
Table I.— Number of experimenters and distribution of smooth brome grass seed.

Fiscal Year.

eg

a

03
<

c
1
<5

i

a
<

1

o

o
o
Q

ai

2
o

'Si

o

0)

C2

0
4
2

6

I— t

0
2
2

4

ai

a

0
6
0

6

eS

e

0
5
1

6

s.

1

H

1

a

0
6
0

,6

rf

O

V

■3

1

be

2

o

i

'S

a

'^

S
0

1
1

9
-/.

i

1896-97

0
2
0

2

0

7
0

7

0
4
0

4

0
10

8

18

13

27

1

42

0
0
0

0

'I (, <^ ' 0

0

1897-98

2 0
0 1

^1*

4

1898-99 -

2

Totals

1

r -- <?

6

.1

Table I. — Number of experimenters and distribution of smooth brome-grass seed-

Continued.

Fiscal Year.

03

a

B
c
o

%

1

0
0

1

8

0

11

2
13

■2

o

1
8
1

10

c
p

o

■4-:)

z
1

3

1

5

o

o

0

34

4

38

d

3
o

1

9
0

10

a

o

cS

a

o

1

6
2

9

(3
O
b£
<ii

Si

O

1

28
4

33

eg

'S
S3

>

1
1

0
5
0

5

c8

.3
£

§
02

0
0
2

2

o

02

1

22
1

24

05

s

0
9
3

12

03

o3
M

4
27
10

41

0
6
0

6

i

s

.'1

2

4
5

11

i

0
0

1
1

g

be

a

1

0

33

4

37

■Jo
C
o

1
'^

1

11
1

13

c

1

1896-97

1897-98

1898-99

Totals -.-

6

37

2

45

1

28

9

38

0

12

5

17

The States receiving the largest amounts of seed as shown by the
table, were Kansas, Montana, Texas, Colorada, Nebraska, North
Dakota, Washington, and Oregon, in the order named. The North
and South Dakota experiment stations have also systematically dis-
tributed large quantities of their home-grown seed throughout their
State. In the former State it has been so distributed that it is now
in the hands of farmers in all but two counties in the State.

METHOD OF SEEDING.

Smooth brome-grass will grow on almost any soil, but its produc-
tiveness depends upon the degree of fertility. It seems to germin-
ate well on rich, moist land, but is also able to produce a crop where
the soil is poor and the conditions unfavorable for the growth
of other grasses. The land should be fall plowed and put in good
condition by disking and spring-tooth or peg-tooth harrowing as the
land may require. The seed may be sown as early as wheat is seeded,
or it may be sown as late as the first of August or September if the
land is kept from drying out and from growing weeds, by an occa-
sional surface cultivation. In some parts of California, Washing-
ton, and Oregon it has been found better to sow in the fall during
the months of October and November.

Some farmers have reported that they had succeeded in sowing
the seed with a drill but the large majority abandoned it after a trial
and sowed broadcast. The seed is very light and chaffy, weighing
only fou' -^n pounds to the bushel, which makes it difficult to feed
satisfactorily through an ordinary seed-drill. The most common
plan is broadcast by hand at the rate of from eighteen to

twent}- j^oiuidB jor acre and harrow it in thoroughly with a peg-tooth
harrf w.

If pro^jer sec. .nils could be secured, there is a saving of seed, a

bet.-T I'liauce for germination, and a less likelihood for the grass to

be^-ome "hide-bov-ad," owing to its spreading rootstoocks, than when

s a broadcast The seed may be procured from any of the large

i firms at tlv rate of about $2.25 per bushel of 14 pounds. When

1 o is ci ' vdng a pasture or meadow and is limited in means,

the'following method is recommended by the South Dakota Experi-
ment Station :

" Purchase enough seed to sow a small parcel of land, say an acre.
Use good clean land and keep out all weeds. Save the seed sown on
this piece of land and sow it the next spring. In a few years a field
of the desired size can be obtained at a comparatively small expense.
By mixing red clover, alsike, alfalfa, millet, or other short-lived or
more tender forage plants with the smooth brome a larger piece of
land can be sown, and as these die out the brome-grass will spread
and finally fill the field."

VALUE FOR HAY.

The yield of hay from smooth brome-grass varies from one to four
and a half tons per acre according to climatic conditions, method of
seeding, and fertility of soil. The quality of the haj^ is excellent,
fully equaling that of timothy in palatability and nutritive qualities.
In order to obtain the best product, the hay should be cut at time of full
bloom. One important feature which distinguishes smooth brome-
grass from other hay grasses is that it does not deteriorate rapidly after
the flowering period and even if cut when the seeds are ripe the hay
will have lost but little of its nutritive qualities, owing to the fact that
after the seed-bearing stem has grown up a large number of leafy
shoots spring up from the base. After furnishing three or four crops
of hay the sod thickens up too much for a good growth of stems.
This thickening occurs sooner if the grass is allowed to ripen seed,
than it does when it is cut for hay, or if it has been seeded heavily
at first.

VALUE OF PASTURE.

Smooth brome-grass is essentially an early spring and late fall
pasture grass. After the hay crop has been taken off a heavy
growth of aftermath or second growth springs up. It seems to be
especially adapted for permanent pastures. After it has produced
several crops of hay it thickens up, forming a very compact sod and
a heavy growth of leaves. If one desires to use it for pasture at
once it will be better to seed it thickly at the rate of about twenty-
five pounds per acre.

As shown by the numerous testimonials received from all parts of
the United States it is relished by all kinds of stock. Cows are very
fond of it and are said to prefer it to timothy and even clover.

Mr. Gluyas, of Hofflund, Williams County, N. Dak., has tested
the palatability of this grass for horses. Some seed was accidentall}^
scattered over the native prairie sod. As a result, about two square
rods of smooth brome-grass appeared in bunches. Hort:ies which
had access to the pasture cropped the smooth brome-grass close and
left the prairie grass around it although the latter was sti^ll green
and in good condition.

J

Fig. 2.— Smootli hrome-grass.

KILLING THE SOD.

On lands where frequent rotation is desired smooth brome-grass
should not be sown. Its creeping rootstocks resemble to some extent
those of the common couch, or quitch-grass, and for this reason it is
not so easily killed by turning under as the more common grasses
used in rotations. Up to the present time the seed has been so
scarce and expensive that few farmers who have secured a good
field of the smooth brome have felt inclined to destroy the sod, so
that our knowledge on this subject is limited.

The results of investigations carried on along this line at the Min-
nesota Experiment Station and the Manitoba Experiment Station at
Brandon, prove that the sod could be thoroughly and successfully
killed. It was found by these stations that a crop of hay can be
harvested and taken from the land, and if the sod was plowed over
immediately afterwards and backset in September, that at the latter
date the grass would be all dead. When the grass was allowed to
ripen seed, however, it was found that the new shoots at the base
had gained such a foothold that when the sod was plowed under at
this stage it was not killed at the time of backsetting in the fall.

HARDINESS.

Smooth brome-grass will withstand extreme changes in the tem-
perature without injury. Its ability to produce good pasture during
long periods of drought far exceeds that of any other cultivated
variety. In Canada where it had been exposed to a temperature of
several degrees below zero and not covered by snow it was entirely
uninjured. Out of seven or eight hundred varieties tried at the
Kansas Experiment Station this proved to be the best. Without
doubt it is the grass for the semiarid regions of the West. From
the reports received it is evident that it is very little influenced by
the changes of cUmate. It does well in California, Kansas, Mon-
tana, North and South Dakota, Tennessee, Utah, Wyoming, and
all parts of Canada.

RESULTS OP TRIALS IN FORTY STATES.

Out of six hundred and three experimenters receiving seed of
smooth brome-grass two hundred and sixty have complied with a
request from the Department for reports as to its success or failure.

During the seasons of 1898 and 1809 report blanks have been sent
out to all those receiving seed of smooth brome-grass for trial, con-
taining the following questions :

Kind, conditions, and preparation of soil ?

Date and method of planting ?

Cultivation, if any ?

Date of harvesting and stage of maturity reached when harvested?
Date of full bloom ?
Date of ripening ?
Yield per acre (if practicable) ?
Quality of product ?
Notes on growth, probable value, etc.

The following table will indicate the number of reports received
from farmers in the different States during the years 1898 and 1899:

Table II. — Number of reports received from experimenters in the different States.

FiSl'Al. Ykar.

1

<

o

N
<

0

1
1

<

0
0

0

1
1

o

6
3

9

6

1
S

o
O

6
3

9

-a

s

0
0

—
0

d

O
<V

1
0

1

d

1

0
0

0

o

c

1
2

3

3

c

h-t
1

2

3

'^

a
a

'■5
t— t

c8

o

1— 1

3

OJ

c

1

'3

c

_c3

3

o

3

bi
1

o

CO

a

•1— 1

2
2

4

1

u

0

1
1

!

o

0
0

0

c3

3
o

1898

0
0

0

1
0

1

4

1

5

18
23

41

0
2

2

0
0

0

0

1

1

1
1

2

0
0

0

2
3

5

20

1809

10

Totals

30

Fiscal Year.

05

6

12

18

i

"-5

1
^;

0
0

0

8

M
m

12;

1

6

7

o
1

0
1

1

c

2

CS

O
o

;?;

2

1

ce
+j
o

a

9
13

o

S
o

1

0

1

a

o

o

2
2

4

o
be

0)

u
O

6
5

d

C
cS
>;

c

C

A*

Ph

0

1

c
p

3

o

'3

o
cc

0
0

1

i

o

7
15

>

03

g
H

5
2

03
M

Eh
1

8

3

1

1
1

4^

r-«

O

i

0
0

d
g

8
6

g

c

S

2
3

5

1

6
4

10

-3

-a

o

1808

1899 ----

137

Totals

--

22

1

1

1

0

2

2

7

9

4

Z

3

14

200

By comparing Tables I and II it will be seen that in most cases
the number of reports received correspond to the number of experi-
menters receiving seed, the most coming from Kansas with 41, Mon-
tana 30, North Dakota 22, South Dakota 22, and Nebraska 18. In
California, Washington, and Oregon smooth brome will succeed with
or without irrigation. In Colorado it retains its fresh green appear-
ance until December, affording excellent pasture. During the severe
droughts in Kansas, Montana, and Nebraska it dies down and appears
dead but as soon as rain falls it becomes green again. It is now well
established in the Dakotas and is grown extensively both for hay and
pasture. In Indiana and Ohio it is said to make about the same
growth as orchard grass but withstands dry weather much better.

Sufficient experiments have not been carried on in the South to
enable us to state its value for that part of the country, but it is
probable that it will be found of. considerable value for winter grazing.

8

The following are a few of the reports which will show the
great value that this grass has become to many of the States :

T. E. Pearce, Edger'ton, ' Johnson County, Kans. : The land used was high
prairie soil on the bluff of a creek, part of which is underlaid closely with rock.
The soil was deeply plowed, disked, and harrowed down fine. The seed was
sown broadcast on April 19, 1898, and then harrowed in lightly as the ground
was in a very moist condition. I thought it had died out in the fall and so
reported to you, because wild grasses had completely covered it up. In the
spring of 1899, however, it woke up like a sleeping giant and covered the gromid
with dense foliage 1 foot high. It did not produce much seed but proved to be
an excellent pasture grass. As the grass docs not get tough like most grasses
do, it is very tender to eat. The green growth starts very early in the spring
and I think from what I have seen of it so far, that it will be a fine pasture and
hay grass.

W. W. Heideman, Kalispell, Flathead County, Mont.: The land used was a
rather light sandy soil plowed this spring. The seed was sown April 13, 1898,
without a nurse crop, harrowed three times and then rolled. It bloomed August
5 and ripened September 15; while other grasses were drying up it retained a
healthy green color all through the summer.

W. S. Delano, Custer County, Nebr.: The seed was sown broadcast May 14,
1898, and covered by light harrowing. One-third of it was sown with barley as
a nurse crop. It made a very good stand but that sown with barley was almost
a total failure. On account of the drought the growth was light. In the spring
of 1899 it was pastured and then later a crop of hay was mowed July 10. After
this cutting its tops dried and died down. In October, however, it started again
from the crown and at this date, October 31, it is 3 inches high. It is an excel-
lent grass, starting earlier in spring than alfalfa and thickens into a solid turf.
As all stock like it, it promises to be an excellent grass for this section.

Messrs. Guill Bros., Chico, Butte County, Cal.: The seed was drilled in by
hand on March 4 and covered two inches deep. The soil is a sandy loam and had
been put into excellent condition by fall plowing and harrowing with a spring
tooth harrow. The crop was cultivated three times. The vitality of this seed
is remarkable. We had no rain to wet the ground for six montlis, from May to
November, yet there was none of the grass that died from drought. The grass
made a growth of about 8 inches during the season. During the autumn and
early winter it had continued its development and is now in excellent condition.
A plot of this grass was sown on October 21, broadcast and harrowed in, and is
looking very well at the present time. It has withstood some of our severest
winter weather without any ill effects.

W. R. Gluyas, Hofflund, Williams County, N. Dak.: The seed was sown broad-
cast at intervals from April 15 to November 10, 1898, on well prepared sandy
loam and harrowed sufficiently to cover seed. On August 27, 1899, it was har-
vested for seed. When it had attained its full growth it was five feet 6 inches
high and the yield per acre of hay would have estimated 4^ tons. It is an excel-
lent hay and pasture grass, withstanding all the severe climatic conditions and
is relished by all stock. When first sown it grows very slowly for a long time
and does not make any crop that can be harvested the first year. It stands
without any equal for both hay and pasture in the dry belt.

SUMMARY.

Smooth brome-grass is a vigorous hardy perennial with strong,
creeping rootstocks, valuable alike for hay and pasturage.

9

The land should be fall plowed, disked and harrowed thoroughlj',
and the seed sown in the spring except in California, Oregon, and
Washington, and probably in the Southern States, where it is prefer-
able to sow in October or November. Sow broadcast at the rate of
18 to 20 pounds per acre and harrow in thoroughly.

The yield of hay varies from 1 to 4^ tons per acre according to
climatic conditions, amount of seed sown, and fertility of the soil.
In quality it is equal to timothy, both in regard to palatability and
nutritive value.

It is essentially an early spring and late fall pasture grass, produc-
ing a very compact sod and a heavy growth of leaves. All kinds of
stock eat it with relish.

It has remarkable drought-resisting qualities and is the most suit-
able grass yet introduced for the dry regions of the West and North-
west. When once established it will withstand a temperature of
many degrees below zero without being injured.

Where frequent rotation is desired smooth brome-grass should not
be sown as it is not so easily killed as the more common grasses used
in rotations.

P. Beveridge Kennedy, Ph. D.,

Assistant in Division of Agrostology.
Approved :

James Wilson,

Secretary of Agriculture.

Washington, D. C, December 1, 1899.

t

*

Circular No. 19.-(Agros. 56.) Issued January 2, 1900.

United States Department of Agriculture,

DIVISION OF AGROSTOLOGY.

[Grass and Forage Plant Investigations.]

NEW OR LITTLE KNOWN MEXICAN GRASSES.

Nearly all ' the grasses enumerated in this circular were collected
by Mr. C. G. Pringle in 1899. While the collection is a small one,
it is remarkable for containing several species of Panicum of the
heleiophylhe, or so-called dichotoraous, group which have not before
been represented in Mr. Pringle's Plantce Mexicance.
ISCH^MUM LATIFOLIUM Kmith. Rev. Gram. 1:168. 1835.
Under tlie spray of the Cascade in the Barranca of Texola near Jalapa, altitude
1,100 m. (3,500 feet), April 30. C. G. Pringle, No. 8106. 1899.

PASPALUM CANDIDUM Kvmth. Mem. Mus. Par. 2:68. 1803. (See Kew

Index. )
Barranca of Texola near Jalapa, state of Vera Cruz, altitude 1,100 m. (3,500

feet), April 30. C. G. Pringle, No. 7884. 1899.

PANICUM PILOSUM Sw. var. MACRANTHUM var. nov.

Secondary axes or branches of the panicle (longer lower ones) 3.5 cm. long,
pilose, with papillate hairs about 2 mm. long. Spikelets 2.2 mm. long, the
outer glumes strongly scabrous on the keel near the apex, as is the fruiting
glume and palea.

Swamps near Jalapa, State of Vera Cruz, altitude 1,230 m.. May 21. C. G.
Pringle, No. 8195. 1899.

PANICUM LAXIFLORUM Lam. Encycl. 4 : 748. 1797. {P. jalapenseKth. '!)

Low, densely cespitose perennial, 1.5-3 dm. high, with crowded, lanceolate,
acute, pilose leaves and spreading, ovate panicles 4-5 cm. long. Culms much
branched near the base, glabrous. Nodes bearded with spreading hairs.
Sheaths pilose with soft, spreading or reflexed hairs. Ligule a dense fringe
of hairs about 1 mm. long. Leaves lanceolate, acute, 3-6 cm. long, 6-1 0 mm.
wide, pilose on both surfaces with soft hairs, ciliate on the margins with
long, spreading, papillate liairs. Axis of the panicle glabrous or pubescent.
Spikelets oblong, obtuse, 2 mm. long, first glume broadly obtuse, about one-
third the length of the spikelet, 3-nerved, the second and third glumes promi-
nently 7-nerved, pubescent with short spreading hairs between the nerves.

Gravelly banks near Jalapa, State of Vera Cruz, altitude 1,250 m. (4,000 feet),
March 29. C. G. Pringle, No. 8083. 1899.

Nearly identical with the grass from the Southern States which has been referred
to P. laxiflorum Lam.

PANICUM INFLATUM Scribn. & Smith. Circular 16, Div. Agros. 5, July,
1899.

Gravelly banks near Jalapa, State of Vera Cruz, altitude 1,250 m. (4,000 feet).
Date not given. C. G. Pringle, No. 7883. 1899.

2

PANICUM VISCIDELLUM Scribn., sp. n.

A slender, ascending or erect, finally branching perennial, G-10 dm. liigb, with
numerous, bearded nodes, pubescent internodes, pubescent sheaths, lanceo-
late, acute, pubescent leaves, and ovate, exserted panicles 5-7 cm. long.
Leaves 5-8 cm. long, 1-2 cm. broad, cordate-clasping at the base; ligule
pilose. Panicle branches somewhat viscid, the lower ones 2.5-3 cm. long.
Spikelets 1.8 mm. long, obovate, obtuse, or subacute, the 7-nerved second
and third glumes glabrous or with a few scattering hairs.

Gravelly banks near Jalapa, State of Vera Cruz, altitude 1,250 m. (4,000 feet).
C. G. Pringle, No. 8089. 1899. In thickets near Mirador, October. Lieb-
mann. No. 323. 1841.

Related to P. viscidum, but stems much more slender, leaves shorter and less
rigid, panicles smaller, as are also the spikelets, which are "nearly smooth.
Fournier in his enumeration of the grasses of Mexico refers this grass to P.
commelincefolmm Rudge, and cites P. midtiflorum Ell. and P. microcarpon
"Michx."as synonyms. I have not Rudge's work, "Plantse Guianse," in
which P. commelincefolmm is ilhistrated, and upon which illustration Foiir-
nier based his determination of Liebmann's plant, but oiir grass is certainly
not P. midtiflorum Ell. (P. polyanthes Schultes), nor does it agree with
available descriptions of P. commelincefolmm. P. microcarpon Ell., Sk. Bot.
S. C. & Ga., p. 127, not Muhl., is the grass now usually referred to P. har-
bvlatum. Mx.

PANICUM MULTIRAMEUM Scribn., sp. n.

A rather slender, tufted perennial, 20-30 cm. long, with the glabrous culms
densely fasciculate -branched above, bearded nodes and glabrous or thinly
pilose sheaths which are bearded at the throat ; lower ciilm leaves 5-8 cm. long,
those on the branches much shorter and narrower, sparingly ciliate near the
base, pubescent beneath, minutely scabrous along the margins which are very
narrowly cartilaginous. Panicles loosely flowered, 2-4 cm. long; spikelets
about 2 mm. long, obtuse, and 7-nerved, the second and third glumes thinly
pubescent, the fourth glume subacute ; the broadly obtuse ; first glume one-
third to nearly one-half as long as the spikelet.

Gravelly hills near Jalapa, State of Vera Cruz, altitude 1,250 m. (4,000 feet).
C. G. Pringle 7882, 1899. Orizaba, State of Vera Cruz, February 17.
Jared G. Smith, No. 593. 1892.

Allied to Panicum ciliosum Nash but smaller, nodes more distinctly bearded
and leaves less ciliate.

PANICUM ALBOMACULATUM Scribn., sp. n.

A rather slender, erect, sparingly branched perennial, 6-8 dm. high, with striate
sheaths, short, ciliate ligules and spreading panicles 12-16 cm. long. Nodes
glabrous, the overlapping margins of the sheaths very densely ciliate or
sub-villous. Leaf-blade 7-12 cm. long. 5-10 mm. wide (when dry), very
acute, rounded at the somewhat clasping base, scabrous on the nerves below,
glabrous above, sharply serrulate scabrous on the narrowly cartilaginous
margins which are ciliate near the base. Axis and branches of the panicle
glabrous, the lower longer branches 8-10 cm. long. Spikelets ovate, obtuse,
2.5 mm. long; first glume obtuse, 1 -nerved, sub-remote, clasping the pedicel,
the second and third glumes 7-nerved, thinly pubescent with short hairs,
about equaling the smooth and shining fourth glume ; the third glume has
a thin, short palea. The exposed portion of the culm and sheaths purplish,
the latter (in the type) white-spotted with small, oblong spots or blotches.

Dry rocky hills, Patzcuaro, State of Michoacan, October 10. C. G. Pringle, No.
5203. 1892.

Allied to Panicum scahrinsculuvi Ell. but readily distinguished by its larger
spikelets, less densely flowered panicles, and distinctly cartilaginous, ser-
rulate leaf margins. The plant throughout is more slender.

PANICUM POLYCAULON Nash. Torr. Bui. 24; 200, April, 1897. Low
places, borders of swamps, Minatitlan, State of "Vera Cruz, June 30.
Jared G. Smith, No. 555. 1893.

MUHLENBERGIA SETARIOIDES Fourn., Mex. PI. Enum. Gram. 84.

Under the spray of the Cascade in
Barranca of Texola near Ja-
lapa. State of Vera Cruz, alti-
tude 1,100 m. (3,500 feet),
April 30. C. G. Pringle, No.

8096. 1899.

MUHLENBERGIA ALAMOS-
ANA Vasey. Coult., Bot.
Gaz. 16: 146, 1891. Mossy
cliffs, Sierra de Tepixtlan near
Cuernavaca, altitude 2,800 m.
(7,500 feet), February 8 and
March 14. C. G. Pringle, No.
6994. 1899.

SPOROBOLUS PILIFERUS

(Trin.) Kimth. Enum. Plant.
1 : 211, 1833. Fields near Ja-
lapa. State of Vera Cruz, alti-
tude 1,250 m. (4,000 feet). C.
G. Pringle, No. 7881. 1899.

AVENA MICRANTHA sp. nov.
(Fig. 1.)

A very slender, densely cespitose,
upright perennial, 2.5-3.5 dm.
high, with soft leaves 6 — 15
cm. long, 1-2 mm. wide, lig-
ules 5-6 mm. long, and loosely
flowered simple panicles 5-7
cm. long. Sheaths glabrous;
leaves pubescent above,'
smooth beneath, becoming
involute when dry ; branches of

the panicle capillary, spread- Fig, l .-Avena micrantha.

ing or ascending, 1-3-flowered, the longer lower branches
empty glumes unequal, lanceolate, the first about 8 mm. long, 1-nerved, the
second about as long as the flowering glume, 3-nerved near the base, thin
membranous and abruptedly pointed ; flowering glume glabrous, 5-nerved,
rounded on the back, 2-toothed at the apex, the teeth awn-like ; callus rather
densely bearded, hairs stiff, the longer ones 3-4 mm. long ; awn arising below
the apex of the flowei-ing glume, slender, geniculate, twisted below the gen-
iculation, about 17 mm. long; palea as long as the flowering glume, rather
rigid, margins rounded, inflexed, apex subhyaline, the two nerves extending
into subulate awn-like teeth.

Scribn.
1-3 cm. long;

I

\

Cool, mossy cliffs, Sierra de Tepixtlan near Cuernavaca, State of Morelos, alti-
tude 3,300 m. (7,500 feet), February 5. C. F. Pringle, No. 8018. 1899.

AVENA STIPOIDES Scribn., sp. n.

A very slender, erect, somewhat wiry perennial 5-6 dm. high, with linear, erect
leaves, and loosely few-'flowered, simple panicles 5-10 cm. long. Sheaths
shorter than the internodes, very minutely strigose-pubescent ; ligule 5-8
mm. long, hyaline; leaves involute-setaceous, at least when dry, 1-3.5 mm.
wide, 1-3 dm. long, scabrous. Spikelets about 13 mm. long exclusive of the
awn; empty glumes unequal, thin, scarious, 1-nerved, acute, the first about
4 mm. long, the second 5.5 mm. long, flowering glume 11 mm. long, 5-nerved,
slightly roughened on the nerves above, 3-toothed at the apex, teeth awn-
like, awned on the back below the 3-toothed apex. Awn geniculate, twisted
below, attached about two-thirds above the base, 13-14 mm. long. Palea
equaling the glume, the two nerves extending into subulate, awnlike teeth.
Callus hairs 1-3 mm. long.

Sierra de San Felipe, State of Oaxaca, altitude 3,130 m. (10,300 feet), September
19. C. G. Pringle, No. 4905. 1894. Distributed as Mulilenhergia stipoides
Trin.

This grass is closely allied to Avena micrantha, No. 8018, but is at once distin-
guished by its shorter empty glumes.

The strictly one-flowered spikelets of this and the last species is a character
which would lead one to place these grasses in the tribe Agrostidece ; but
the densely hairy callus and rather rigid 5-nerved flowering glume which is
deeply 3-toothed at the apex, and the dorsal, geniculate, and twisted awn
formed by the union of three of the nerves suggest relationship with Arena
as does the character of the empty glumes, and I have tentatively placed
these species in that genus.

POA PRATENSIS L. Sp. PI. 67. 1753.

Mountains near Jalapa, State of Vera Cruz, altitude 1,750 m. (5,700 feet),
April-May. C. G. Pringle, No. 7880. 1899.

SITANION BREVIFOLIUM J. G. S. Bui. 18, Div. Agros. 17, June 24,
1899. Cerro Ventoso, above Pachuca, State of Hidalgo, altitude 3,600 m.
(8,500 feet), August 18. C. G. Pringle, No. 6944, 1899.

F. Lamson-Scribner,

Agr^ostologist.
Approved :

James Wilson,

Secretary of Agriculture.

Washington, D. C, December 19, 1899.

i

o

C G T A M » C A L^

United States Department of Agrici

Circular No. 20. (Agros. .57.)

DIVISION OF AGROSTOLOGY.

[Grass and Forage Plant Investigations,]
F. Lamson-Scribner, Agrostologist.

EXPERIMENTS WITH FORAGE PLANTS IN ONTARIO.

The following brief report by Dr. P. Beveridge Kennedy, an assis-
tant in the Division, on a portion of the proceedings of a meeting of
the Ontario Agricultural and Experimental Union, held at Guelph,
Ont., December 6-8, 1899, and which Dr. Kennedy attended offici-
ally, will beof value to many farmers in our northern States. — F.L.8.

The large majority of the experiments were carried on with the
cereals,' root crops, and horticultural plants, while grasses and forage
plants were only investigated and experimented with to a small
extent.

The following table indicates the gradual development of the Ontario Agricul-
tural and Experimental Union since its foundation. While in 1886 only one
experiment was undertaken by 13 experimenters there have been carried on in
1899 twenty-three different experiments by 3,485 experimenters:

The diflflculty met with in the proper receipt of reports from the free distribu-
tion of seeds is met with by the Ontario Experimental Union to a larger degree
than in our Division here.

Cooperative experiments in .agriculture for the Province of Ontario.

Years.

1886
1888
1891
1892
1893
1894
1895
1896
1897
1898
1899

No. of
ditterent

No. of

experi-
ments.

experi-
menters.

1

12

1

90

Vi

203

12

754

13

1,204

14

1,440

15

1,699

16

2,260

18

2,835

19

.3,028

28

3,485

Satisfac-
tory
reports.

8
40
126
295
416
504
513
501
610
667
739

It will be seen that out of 3,485 experimenters only 739 sent in satisfactory
reports during the season of 1899.

This in a large measure is due to many farmers feeling too keenly their lack of
educational ability to write \x\} a report or to fill out the blanks forwarded to
them. Sometimes, however, it is due to neglect, or owing to the experiments
having resulted in a failure they have not cared to inform anyone of it.

There are 3,505.422 acres of land in Ontario devoted to the growth of hay and
clovers.

Within the last thirteen years 72 different varieties of hay and clovers have
been experimented with by the Ontario Agricultural College, while 8 varieties
have been tested throughout Ontario.

2

The following grasses have been tested during the season of 1899:

Grasses.

Variety.

Tall Oat Grass . -

Timothy

Orchard Grass . .
Meadow Fescue-

Height,
first season.

QUANTITY OF HAY PER ACRE.

Second
season.

Inchex.

11.8

6.3

10.8

10.2

Tons.
3.0
2.7
2.0
2.1

Third sea-
son.

2.3
2.1
1.6
1.3

It is of interest to note that a much larger crop is obtained during the second
season than the third. Tall Oat Grass producing 3 tons to the acre during the
second season and only 2.3 tons during the third season. The same may be said
of Timothy, Orchard Grass, and Meadow Fescue.

An interesting experiment on the diffierent clovers was carried oh during the
present year. The following table indicates the average height of the clovers
for three years during the first season and the tons of gi-een and dry hay per acre
produced during the second season :

Clovers.

Variety.

Mammoth Red.
Common Red ..

Alsilce

Lucern

Height,

first season

(Average

for three

years).

Inches.
9.2
7.3

6.8

7.8

QUANTITY PER ACRE, SEC-
OND SEASON.

Green hay. Pry hay.

Tons.
7.3
6.0
5.6
5.3

Tons.
3.6
2.4
2.4
2.0

Mammoth clover it will be seen produces a much larger amount of fodder
both in the green and dry state than the common Red, AJsike, or Lucern clovers
in Ontario.

In regard to special crops for green fodder the following results have been
obtained for Hairy Vetch, Common Vetch, and Grass Peas :

Vetches.

Variety.

Hairy Vetch

Common Vetcli-
Grass Peas

Length of
plant (inclies).

Tons per acre

(5 tests).

41
29
25

9.0
6.9
5.1

It will be seen that Hairy Vetch in amount of forage and tons per acre far
exceeds the common Vetch or the Grass Pea. The latter, however, is said to be
weevil proof, which is a great advantage in the production of seed.

Three varieties of the early soy beans have been tested and may prove of con-
siderable value in some parts of Ontario.

Soy beans.

Variety.

Medium Green

American Coffee Berry
Extra Early Dwarf —

Estimated
value.

72
100

YIELD PER ACRE.

Straw (tons).

2.6
1.4
1.1

Grain
(husliels).

22.4

21.3
12.7

3

While in the District of Columbia, and in the South generally, it is possible
to g-row three or even four crops of soy beans in a season ; in Ontario only one
crop can be grown from May to September. From the table it may be gathered
that the Medium Green soy bean produces the largest yield of straw as well as
the largest amount of grain.

Three varieties of Millets were tested with the following results :

Millets.

Variety.

Japanese Panicle --.
Japanese Barnyard.
Hungarian Grass. . .

Estimated
value.

Two years.
100
73
63

Tons per acre of green hay.

Five tests.
4.4
3.7
4.0

Thirty tests.
6.4
5.9

4.«

Japanese panicle heads the list with 6.4 tons per acre of green hay.

CRIMSON CLOVER.

The immense value of the use of Crimson Clover as a cover crop was exempli-
fied by Mr. Ghent of New York State.

By delay in tilling the soil in early spring a loss of about 200 tons of water per
acre takes place in a single week, while it has been estimated that an acre of
meadow grass eliminates from the soil 106 tons of w^ater per acre in twenty-four
hours in the month of Jiine.

The soil may be greatly aided in the conservation of moisture by the aid of a
cover crop. There are many leguminous crops which may be used for this pur-
pose but Crimson Clover has been found especially valuable in the North. Till
very early in the season and continue doing so until the crops are mature, then
save the moisture by the use of a cover crop. Crimson Clover is an annual,
grows rapidly, and being a legume the growing of it enriches instead of impov-
erishes the soil. The following table indicates the value of tillage with the use
of Crimson Clover :

Tillage vnth and without crimson clover.

Water.

Nitrogen.

Humus -

Available phosphoric acid

With three
crops of clover.

Per cent.
15.00
.21
3.94
.015

Without
clover.

Per cent.
8.75

.13
1.91

.008

The above shows the great difference in composition of a soil which has had
three crops of clover grown upon it and one which has had the same amount of
tillage but where no clover has been used. When such results can be obtained
the necessity of artificial fertilizers may well be questioned.

P. Beyeridge Kennedy, Ph. D.,

Assistant in Division of Agrostology.
Approved :

James Wilson,

Secretary of Agriculture.

Washington, D. C, December 15, 1899.

A

i

Circular No. 21.— (Agros. 58.)

United States Department of Agriculture,

DIVISION OF AGROSTOLOGY.

[Grass and Forage Plant Investigations.]

COOPERATIVE RANGE GRASS AND FORAGE PLANT EXPERIMENTS AT

HIGHMORE, S. DAK.

INTRODUCTORY.

In the early part of the present year the Secretary of Agriculture
received a communication from Director Shepard of the South Da-
kota Experiment Station, asking the cooperation of the Department
with that station in a series of experiments with drought-resisting

*

grasses and forage crops with a view to finding varieties suitable for
use in the range regions. Through Professor Shepard the South
Dakota authorities made a proposition to furnish land for the experi-
ments and properly equip the Station, asking the assistance of the
Department of Agriculture in planning, instituting, and carrying on
the experiments.

The proposition meeting the approval of the Secretary of Agricul-
ture, Mr. Thomas A. Williams of the Division force was sent to
South Dakota with instructions to complete the arrangements for
cooperation and to assist the authorities of the South Dakota Station
in planning and instituting experiments for testing such drought-
resisting grass and forage crops as seem likely to be of value on the
northwestern ranges.

The results of the work for the first season are most satisfactory
both as to the plan of cooperation and the tests undertaken. The
authorities of the South Dakota Experiment Station and also of the
State Agricultural College are doing all in their power to make the
work a success. The same can be said of the people of that section
who are interested in the question of forage supply. The cordiality
with which all parties have entered into this work, the representa-
tive character of the site selected for the Station as to both soil and
climatic conditions, ^d the practical nature of the experiments, give
every promise that the results accomplished will be of great value to
the people of the northwestern range region.

Mr. Louis W. Carter, a graduate of the South Dakota Agricul-
tural College, was appointed special agent on the rolls of the Divi-
sion and placed in charge of these cooperative experiments and the
results of the work of the present season are appended.

The grounds of the Highmore Cooperative Grass and Forage
Plant Experiment Station consist of one hundred and seventeen
acres, being that part of section eleven, town one hundred and twelve
north, range seventy-two east, lying north of the Chicago and North-
western Railroad. This land was deeded to the South Dakota Ex-

2

periment Station by the people of the county in which Highmore is
situated for the purpose of testing dry -land grasses and forage plants
and experiments in range renovation. The soil of the land selected
and the climatic conditions are fairly representative of the great
range region of western Nebraska, South Dakota, North Dakota,
Wyoming, and Montana. It is on the high rolling prairie which
separates the valleys of the Missouri and James rivers and about
equidistant from each. The soil is a rich, sandy Ipam containing
little or no alkali and, with sufficient moisture, would produce good
yields of the ordinary farm crops. The subsoil is the yellow bowlder
clay characteristic of the drift soils in the Northwest. The Station
is equipped with a seed house 14 by 22 feet, 14- stories high, with a
lean-to 14 by 22 feet for a tool shed, a cistern and the necessary tools
for planting and cultivating the crops.

Between 40 and 50 acres lying close to the railroad were broken up
in 1884 and 1885 and cultivated for a number of years. This land
was then left idle and grew up to weeds and native grasses or "went
back," as the common saying is. It had not been plowed for six or
seven years. Beginning on the south side, series of plats were laid
off, each series being eight rods wide. These were again divided into
quarter-acre plats, each plat being 5 by 8 rods, and numbered con-
secutively, beginning at the east. There are 30 plats in each series.
The spries are lettered A, B, C, etc. Series A and B and about one-
half of series C were plowed 6 inches in depth with a common stirring
plow. About thirty permanent plats were sown. Most of these
made a good growth but those sown broadcast are mixed with the
native grasses, western wheat grass {Agropyron spicatum) being
most prevalent. Quite a number of annuals were planted and the
rest of the ground plowed was sown to millet or planted to corn to
keep the weeds down.

Rain and snowfall at Highmore, S. Dak., for the years 1897, 1898, 1899, as observ-
ed by Mr. J. C. Stoner.

1897.

1898.

1899.

Month.

February .
March —

April

Rain or
snow.

May

June

July

Augiist

September

October

November .

December.
Total -

Snow...
...do....
Rain ...
Snow...
Rain ...
Rain ...
...do....
...do....
...do....
...do...-
Snow...
Rain ...
Snow...
Rain ...
Snow...
J Snow..
) Rain ..

Inches.

Month.

8.
11.

1.6
13.

2.32
.51

2.79

3.75

1.45
.41

2.

1.05

4.
.08

3.5
41.5
13.88

Rain or
snow.

Inches.

Month.

February.. Snow... 2.

March Rain ... 1.40

April - do — 1.51

May ...do — ' 4.17

June ...do.... 1.67

July .•..do.... 2.33

August L..do — .62

Rest of year missing.

Total, Feb. to Sept. 1 :

Snow, 2. inches.
Rain, 11.70 inches.

Rain or
snow.

January ... Snow..

Rain
February..' Snow...

March ; Snow...

April |.-.do —

Rain ...
May Snow...

Rain ...

June do —

July ,...do —

August ;...do —

September . j . . . do —
October do —

Inches.

2.6

.11
.50
14.50
6.

1.4

.5

3.75

4.2

1.81

2.19

.48

.81

Total to October 22 :

Snow, 23.6 inches.
Rain, 14.-67 inches.

NOTES ON THE PLATS.

Plat A (1). — Thirty-four rows on the east side of this plat were sown to
smooth bunch grass {Poa laevigata). Seed collected by Shear and Bessey in
Colorado in 1898. Part of the seed was sown broadcast and part of it was drilled.
Very little of the seed germinated on the ground sown broadcast, and this was
pulverized Jime 30 to destroy weeds. The seed drilled did not make a very good
stand, but made a healthy growth. Seed sown May 10 ; sprouted and up May
33 ; October 23, the plants were two to three inches high and still green.

I 1

I

q:

66\

O

30

30
29

29

28

28

27

27

26

26

25

25

24

24

23

25

22

22

21

21

20

20

19

19

18

18

17

17

16

16

15

15

14

14

14

13

13

13

12

12

12

II

II

II

10

10

10

9

9

9

8

8

8

6

7

7

6

6

5

5

5

4

4

4

3

3

3

2

2

2

1

1

'

s>

-^

■^N

117.15

Native PraiTie

Cooperative Range

Grass and forage Plant

Experiment Station,

HiGHMORE S.D.

B C

S cale ; Ti 6 i tich = 1 ro d . ^-^^ ^'^^'^'^

Ten rows on the west side of this plat were drilled to bunch redtop {Poa biick-
leyana). Seed sown May 10; sprouted and up May 26. All the young plants
were destroyed by drought, and the groimd was cleaned of weeds June 30.

Plat A (2). — The east half of this plat was planted to bunch grass {Poa sp. ?),
the seed of which was collected in Wyoming by Mr. Griffiths in July, 1898.
Sown broadcast May 10; sprouted and up May 31. The stand is very thin.

The west half of this plat was sown in a like manner to a robust form of Cana-
dian blue grass {Poa comjiressa) , collected by Mr. Griffiths at Norfolk, Nebraska,
in 1898. Sown May 10; up May 31. Stand thin.

Plat A (3). — Nevada blue grass {Poa nevadensis). One-half of this plat
was drilled and the other half sown broadcast. Seed collected by Williams
and Griffiths in Wyoming in 1898. Sown May 10; sprouted and up May 31.
The stand on part of the drilled portion was so thin it was sown to millet to keep
weeds down. The part left made a good, healthy growth. The stand on tne
part of plat sown broadcast was also very thin. Mown July 39 to kill weeds.
October 23, the plants were two to three inches high and still green.

Plat A (4).- — Oregon brome {Bromus unioloides) . This plat was sown broad-
cast May 10. Seed from Oregon. Sprouted and up May 22. This seed germi-
nated well, but the plat was very weedy. August 5, it was twelve to eighteen
inches high ; August 21, seed was ripe and one-half bushel gathered by hand ;
October 23, six to eight inches high and still green.

Plat A (5). — The last row on the west side of the plat was sown to Buffalo

bunch grass (Festuca scahrella). Seed collected by Shear and Bessey in Colorado

in 1898. Seed sprouted, but the young plants died as a result of the drought.

'The second row from the west was sown to Langsdorff's reed-bent grass {Cala-

magrostis langsdorffli). Seed failed to germinate.

Six rows near the west side of this plat were drilled to Safflower (Carthamus
tinctorius). Seed drilled May 10 ; sprouted and xip May 21. Seed came from
Professor Hansen, of the South Dakota Agricultural College. Grew to be two
or three feet high. Seed ripened in September. This plant is grown in Russia
for the oil extracted from the seed.

Rows 9 and 10 from the west were sown to "Pajsa" No. 493. Seed failed to
germinate.

The east four-fifths of this plat were sown broadcast to short-awned brome
grass (Bromus breviaristatus). Seed collected by Mr. Griffiths in Wyoming in
1898. Sown May 10 ; sprouted and iTp May 24. The plat was mowed July 29 to
kill weeds. This grass made a good stand, and on October 23 it was two to three
inches high and still green.

Plat A (6). — This plat was drilled to King's fescue (Festuca kingii) . Seed col-
lected by Williams and Griffiths in Wyoming in 1898. Sown May 12 ; sprouted
and up May 24. This grass made a good stand. The dry weather did not seem
to affect it. October 23, the plants were three to four inches high and still
bright green.

Plat A (7). — This plat was selected as a site for the seed house and only part
of it sown to three kinds of imported millets. Two of these failed to germinate.
The millet that grew was a black broom-corn millet (South Dakota Experiment
Station No. H-9). This millet made a good growth, 24 to 30 inches in height,
and was not badly affected by the dry weather ; ripened, and was cut August
12. It yielded a quantity of seed but no reliable estimate of the yield per
acre could be made on such a small piece of ground. This is evidently the same
as the millet grown on plat C-12.

Plat A (8).— The first four rows on the east side of this plat were sown to the
woodland rye-grass (Elymus glaucus) . Seed collected by Williams and Griffiths
in 1898. Drilled May 11; sprouted and up May 28. August 5 it was eight to
ten inches high and of a bright green color. October 24 still bright green. The
fifth and sixth rows in this plat were drilled to bearded wheat grass (Agropyron
caninum) . Seed collected by Shear and Bessey in Colorado in 1898. Seed drilled
May 11, but failed to germinate. ,

The seventh, eighth, and ninth rows were drilled to bearded wheat grass.
Seed from U. S. Grass Garden in 1896. Failed to germinate. The next nine
and one-half yards east of drilled portion were so\vn broadcast May 13 to feather
bnnch grass {Stipa viridula). Seed collected by Griffiths in South Dakota 1898.
Sprouted and up May 30. Stand thin, mown July 29 to kill weeds.

The east half of this plat was sown broadcast to bearded wheat grass (Agro-
pyron caninnm). Seed collected by Williams and Griffiths in 1898 in Wyoming.
Seed sown May 11 ; sprouted and up May 25. Fair stand. Mown July 29 to kill
weeds. October 23, curing on the ground but still green at the bottom.

Plat A (9).— This plat was drilled May 11 to giant rye-grass {Elymus conden-
satus). Seed collected by Williams and Griffiths in Wyoming in 1898. Sprouted
and up May 24. July 26, ten inches high. August 5, twelve inches high and
very badly rusted. October 24, leaves all dead, killed by frosts.

Plat A (10). — This plat was sown broadcast to six lots of slender wheat
grass (Agropyron tenerum). Five of these were collected by Williams and
Griffiths in Wyoming and Montana and one lot by Shear and Bessey in Colorado.
They all made a fine stand. No difference could be detected this year. Seed
sown May 13 ; sprouted and up May 22. Mown July 29 to kill weeds. October
24, not very high but still green.

Plat A (11). — This plat was drilled to slender wheat grass {Agropyron tene-
rum) . Seed collected by Shear and Bessey in Colorado in 1898. Drilled May 12 ;
sproiited and iip May 28. There was a great difference in parts of this plat.
July 2G, six inches high ; some heading out. August 5, first five rows headed
out, twelve to fourteen inches high. In the next twenty rows the plants are
darker green, smaller, and later, just beginning to head out. Rest of plat same
as first five rov/s. October 24, rusted badly but still partly green.

Plat A (12). — This plat was sown broadcast to wild timothy [Mulilenhergia
racemosa). Sown May 11; sprouted and up June 6. August 5, three to four
inches high ; very thin stand and very weedy. Mown Jiily 31 to kill weeds.
October 24, all dead, killed by frosts.

Plat A (13). — Eight yards on the east side of this plat were sown to curly
niesquite {Hilaria cenchroides). Sown broadcast May 11; sprouted and up
June 1. Very thin stand. Mown July 31 to kill weeds.

The next eight yards of this plat were sown broadcast to blue grama {Botde-
loua oligostachya) . Seed collected by Williams and Griffiths at Billings, Mon-
tana, 1898. Seed sown May 11; sprouted and up May 27. Stand thin and plat
weedy.

The next five yards in this plat were sown broadcast to blue grama (Boute-
loua oligostachya). Seed collected by Williams and Griffiths, Montana, 1898.
Seed sown May 11 ; sprouted and up June 4. Mown July 31 to kill weeds. Thin
stand.

The rest of this plat, four paces, was sown to King's fescue {Festuca khigii).
Seed collected by Williams and Griffiths in Wyoming, 1898. Sown broadcast
May 13; sprouted and u.p May 26. Plat was mown July 31 to kill weeds.

Plat A (14). — Blue grama (Boiiteloua oligostachya) . Seed on the west half
of plat came from Walla Walla, Wash. Sown broadcast May 11. Thin stand.
Mown July 31 to kill weeds.

Plat A (15). — Mixed grama. Seed from Texas. Sown broadcast May 11 ;
sprouted and up June 8. Thin stand. Mown July 31 to kill weeds.

Plat A (16). — First twenty-two rows on east side drilled to false couch grass
{Agropyron j)^eudorepens) . Seed from Texas. Drilled May 12; seed failed to
germinate. The next eleven rows were sown to annual saltbtish {Atriplex holo-
carpa). Seed grown in U. S. Grass Garden, at Washington, in 1898. Drilled
May 12; seed failed to germinate.

The next twenty-four rows in this plat were sown to silvery saltbush (Atri-
plex argentea). Seed collected by Williams and Griffiths in Montana, 1898.
Drilled May 13; seed failed to germinate.

The last six yards on west side of plat were sown broadcast to wire bunch
grass (Agropyron divergens). Seed collected by Williams and Griffiths in
Wyoming in 1898. Sown May 13. Thin stand. Mown July 31 to kill weeds.
October 24, three to four inches high and still green.

Plat A (17 and 18). — Hairy or sand vetch (Vicia villosa). Seed imported
from Russia by the U. S. Department of Agriculture. Drilled May 12 ; sprouted
and up May 26. Sown too thick ; nearly every seed germinated. July 26, two
to four feet long. Almost all killed by hot winds July 15 to 20. Very weedy.
Around the edges next to paths it made a good growth and blossomed in August
and September, but failed to mature seeds. October 24, still green and would
make good pasture.

Plat A (19 and 20).— Sown broadcast May 23 to]Turkestan alfalfa {MecUcago
sativa var. turkesfanica) . Seed from Section of Seed and Plant Introduction, No.
991. Sprouted and up May 29. The alfalfa grew nicely until the first of July,
when it was six to eight inches tall. The dry weather in July and August
stopped the gi-owth and damaged it badly.

Plat A (22 and 23). — Western wheat grass (Agropyron spicatum). Seed col-
lected by Williams and Griffiths in Wyoming in 1898. Sown broadcast May 11 ;
sprouted and up June 4. Mown July 31 to kill weeds. Good stand. October 24,
three to four inches high and still green.

Plat A (21 and 24 to 29, inclusive).— Were sown broadcast May 10 to smooth
brome grass {Bromus inennis) . Seed grown at the South Dakota Experiment
Station, at Brookings, S. Dak. Sprouted and up May 21. Jtily 26, good stand
from four to six inches tall. Mown July 31 to kill weeds. October 24, still
green, five to six inches high and very thrifty. This grass has made the best
showing of any of the grasses sown broadcast.

Plat B (1). — Original prairie. Broken in June. Not planted.

Plat B (2). — North half of this plat was drilled to Hagi {Lespedeza bicolor),
introduced from Japan by the Section of Seed and Plant Introduction. Very
thin stand. Grew to be twenty-four to thirty inches high but did not come into
bloom. Had a single stalk and was very woody.

Plat B (3).— Earliest ripe fodder corn. Seed from J. A. Salzer Seed Co.
Planted in rows 22 inches apart. May 24. Sprouted and up June 1. Part of
this plat was drilled, part planted in hills 14, 28, and 42 inches apart in the rows.
The drilled portion suffered from drought in Jiily and only grew to be from two
to three feet high. The thinnest planting made the coarsest fodder and the
yield seemed to be heavier. The wind blew the corn together so that the differ-
ent sections of plat could not be weighed separately. Tasseled July 27. Cut
September 11. Weighed October 20.

Yield on one-quarter acre 400 pounds, or at the rate of 1,600 pounds per acre.
This plat is on some of the lowest ground under ciiltivation. The tallest corn
was from three and one-half to four feet high.

Four rows on the west side of this plat were sown to yellow uiilo maize. Seed
furnished by Mr. J. C. Stoner, of Highmore. It was old and failed to germinate.

Plat B (4) . — Salzer's Superior fodder corn. Seed furnished by J. A. Salzer
Seed Co. Planted May 24 in rows 22 inches apart, hills 14, 28, and 42 inches
apart in the row. As in the preceding plat the corn planted farthest apart made
the largest growth. Sprouted and up June 1. July 27, three to four feet high.
Very rank growth. Most of it tasseled otit before September 1. Cut Septem-
ber 11. Four to seven feet high. Weighed October 20; weight for one-quarter
acre 810 pounds, or at the rate of 3,240 pounds per acre.

Plat B (5, 6, 7, and 8).— Sown broadcast to Turkestan alfalfa {Medicago sa-
tiva var. turkestanica) at the rate of 35 pounds per acre. Seed from Section of
Seed and Plant Introduction, No. 991. Sown May 21. Sprouted and up May 27.
Made a fine growth up to July 4. Badly damaged by hot winds July 15 to 20.
Six to ten inches high July 1. Did not grow any taller. Still alive and fresh
October 24.

Plat B (9).— Earliest ripe fodder corn. Seed from J. A. Salzer Seed Co.
Planted May 24, in hills 42 inches apart. Sprouted and up June 2. July 27,
three to four feet high. Thin stand. Cut September 11. Had quite a number
of small ears. Weighed October 20 ; weight, dry fodder, 310 pounds, or at the
rate of 1,240 pounds per acre. This fodder was badly blown about by the winds.

Plat B (10). — Jerusalem corn. Seed from J. A. Salzer Seed Co. Half of this
plat was drilled and half sown broadcast. Stand was thin, seed did not germi-
nate well. Three rows on east side of plat were left to ripen seed and yielded
one peck. Ripened in September. The corn on part of plat sown broadcast
grew to be two or three feet high ; that drilled, three to four feet. Cut Septem-
ber 11. Weight estimated 300 pounds or at the rate of 1,200 pounds per acre.

Plat B (11).— Salzer's Superior fodder corn. Seed from J. A. Salzer Seed Co.
Planted in hills 42 inches apart May 24. Sprouted and uj) May 31. Made a
good gi'owth. July 27 was four feet high. Cut September 12; five to six feet
high at time of cutting. Weight October 20, 500 pounds, or at the rate of 2,000
pounds per acre. The land being higher than plat B (4), this corn did not make
as thrifty a growth as on that plat.

Plat B (12). — East half of plat planted to Wisconsin amber cane. Seed from
J. A. Salzer Seed Co. Drilled May 25 ; sprouted and up June 14. About 75 per
cent of a stand. July 27, two and one-half to three feet tall ; badly damaged by
hot winds. Cut September 13. Three rows left to ripen seed. About one-half
was badly affected by smut.

Plat B (12). — West half of plat planted to hairy vetch {Vicia villosa) and
Jeriisalem corn. Sown broadcast May 25. Vetch sprouted and up June 2 ; corn
sprouted and up June 8. Badly damaged by drought in July ; very weedy.
September 1, corn and vetch dead except a little near the edges of the plat.

Plat B (13). — East half of plat. Dwarf Victoria rape. Drilled in rows one
foot apart May 25 ; sprouted and up May 30. Seed so^vn too thick. July 27, one
to one and one-half feet high. Did not grow any after July. Yield, four to
five tons of gi-een fodder per acre.

Plat B (13). — West half of plat. Dwarf Victoria rape and Wisconsin amber
cane. Sown broadcast May 25 ; rape sprouted and xrp June 2 ; cane sprouted and
up June 13. Plat was sown too thick; cane all died. Rape six inches high.
Rape did not grow any after July.

Plat B (14).— Salzer's Superior sand vetch {Vicia villosa). East half of plat
sown in drills two feet apart ; west half of plat sown broadcast. Sown May 25 ;
sprouted and up May 31. July 27, drilled portion, one to one and one-half
feet long; did not grow any more. That sown broadcast died from hot winds
in July.

Plat B (15). — Salzer's Superior fodder corn. Seed from J. A. Salzer Seed Co.
Planted in hills 42 inches apart May 25; sprouted and up Jime 4. July 27,
75 per cent of a stand; four feet high. Cut September 13; five to six feet
high. Weight, October 10, 500 pounds, or at the rate of 2,000 pounds per acre.

Plat B (16). — Earliest ripe fodder corn. Seed from J. A. Salzer Seed Co.
Drilled May 25, in rows 42 inches ax)art; spfouted and up June 3. Tasseled
out July 27; three to three and one-half feet high. Cut September 11. A
few ears of com on stalks. Weight, October 20, 800 pounds, or at the rate of
1,200 pounds per acre.

8

Plat B (17).— Hairy vetch {Vicia villosa). East half of plat drilled ; west half
sown broadcast. Sown May 26 ; sprouted and up May 30. July 27, in blossom ;
one to two feet long on di-illed portion. Part of plat sown broadcast, dried
out and destroyed by dry weather. A few plants next to paths made a good
growth and blossomed, but did not mature seeds.

Plat B (18).— Kafir corn. Seed from J. A. Salzer Seed Co. Half of plat
drilled and half sown broadcast. Sown May 26; sprouted and up June 12. Ger-
mination poor. Portion sown broadcast killed by weeds and dry weather.
Drilled corn two to two and one-half feet high ; part of it left for seed, part cut
September 13. None of that left for seed filled. Weighed October 20 ; yield at
the rate of 1,500 pounds per acre.

Plat B (19).— Combination plat— sand vetch, alfalfa, and Kafir corn. Sown
broadcast and drilled May 26 ; sprouted and up June 1. Portion sown broadcast
dried out and died. July 27, drilled portion— corn, two feet high : vetch, one to
two feet long; alfalfa, eight to ten inches. Did not grow any after July.
October 20, alfalfa and sand vetch still green but not growing.

The rest of Series B was planted to fodder corn and cultivated to keep the
ground clean.

Plat C (1 and 2).— New breaking, not planted.

Plat C (3 to 6). — Common millet {ChcEtochloa italica). Sown for feed at the
rate of twelve quarts per acre. Sown Jime 7. Sprouted and up June 16. These
plats were on low ground and the millet made a good growth one to two feet
high. Cut for hay August 8. Yield 2,961 pounds.

Plat C (7 to 10, inclusive).— One acre. Common millet {Cha'toddoa italica).
Sown at the rate of sixteen quarts per acre. Sown June 7. Sprouted and
up June 16. Cut for hay August 8. This millet was on high ground and suffered
badly from drought in July. Yield 1,740 pounds.

Plat C (12).— Red Orenburg broom-corn millet [Panictim. miliaceum). Seed
from Section of Seed and Plant Introduction, No. 2960. Sixteen rows on east
side of plat drilled June 16. Sprouted and up June 23. Very poor germination.
July 27, one and one-half feet high; low and spreading; seed red; heads com-
pact and heavily seeded. About two quarts of seed were saved of this variety.

Plat C (12).— Black Russian broom-corn millet {Panicum miliaceum). Seed
from Section of Seed and Plant Introduction, No. 2795. Eighteen rows drilled
June 16 Jiily 27, one and one-half feet high; very rank grower. September 1,
two to two and one-half feet high. Seed ripe. This is a black-seeded millet and
the best of the Russian millets tried. One and one-half quarts of seed were
raised.

Plat C (12). — Red Veronezh broom-corn millet {Panicum miliaceiim). Seed
from the Section of Seed and Plant Introduction, No. 2796. Twenty-five rows
drilled June 16. Poor germination. July 27 one and one-half feet high. Plants
low and spreading ; heads long ; seed red. Three quarts of seed were raised.

Plat C (13).— Tambov broom-corn millet {Panicum milaceum). Seed from
Section of Seed and Plant Introduction, No. 2794. Fifteen rows drilled June 16.
Sprouted and up June 23. Very poor germination of seeds. July 27, headed
out, one foot high. Plants low and spreading. Two quarts of seed raised from
this niimber.

Plat C (13).— Red Russian broom-corn millet {Panicum miliaceum). Seed
from the Section of Seed and Plant Introduction, No. 2797. Fourteen rows
drilled June 16. Sprouted and up June 23. Poor germination. July 27, headed
out, one and one-half feet high.* Heads long and spreading: seed red. Two
(|uarts of seed of this millet were saved.

9

Plat C (14). — Ktu'sk millet {Chcetochloa italica). Seed from the Section of
Seed and Plant Introduction, No. 2798. Thirty-one rows drilled Jnne 16. Sprout-
ed and up June 24. July 27, very thick; one foot high. August 1, headed out
sixteen to eighteen inches high. Damaged somewhat by hot winds. Heads
one to three inches long. Eight quarts of seed were saved.

SUMMARY.

The preceding notes record the results of the season's work at
Highmore, S. Dak. There are a few conditions which must be
kept in mind in studying the results. The land slopes from west
toward the east, the west side of the Station being a dry ridge. The
land had "gone back" and had not been plowed for six or seven
years and had become full of foul seeds. Russian thistles, pigeon
grass, and western wheat grass had complete possession of the
ground. The plats sown broadcast could not be weeded and the
young plants had to contend against the weeds without aid. The
annuals (corn, rape, vetches, etc.) did not yield as well as they un-
doubtedly would have, had the ground been cultivated for the past
four or five years. The land was packed from the trampling of
stock and when plowed was lumpy and could not be made into a
satisfactory seed bed in so short a time. The season also was pecul-
iar. A study of the moisture records will show that while there was
the usual amount -of rainfall it came early and late in the season and
July was both dry and windy.

The pressing need of this section of the country is winter feed,
either hay, fodder, or pasture. Some of the bunch grasses from the
higher altitudes in Wyoming and Montana, such as bunch redtop
{Poa buckleyana) and smooth bunch grass {Poa Icevigata) give
much promise for winter pasture, while Nevada blue grass {Poa
nevadensis) and King's fescue {Festuca kingii) give promise of
both hay and pasture. The frost does not affect them until very
late in the season. The favorable growth and behavior of smooth
brome grass {Bromus inerinis) this year as well as previous experi-
ence here and elsewhere in the Northwest with this grass shows it
to be a good hay and pasture grass. On the highest, driest ground
of the Station it made a good stand and kept green and thrifty dur-
ing the driest weather. Oregon brome has made a fine showing and
deserves a thorough trial on account of its excellent yield of seed and
forage, and drought-resistant qualities. The native wheat grasses
furnish the larger percentage of the hay in this section and under
cultivation and favorable conditions will undoubtedly increase in
productiveness.

One of the results of the investigation of the forage problem of
the northwestern ranges is a demonstration of the need of united and
continued effort along the line of experimentation designed to test
the adaptability of the various native and introduced grass and

10

forage crops to the conditions which prevail on the ranges, to learn
more as to the best methods of growing these crops and of improv-
ing and maintaining the general productiveness of the ranges them-
selves. Much can be accomplished by the individual efforts of the
ranchmen and others concerned ; but all will agree that the best re-
sults can be secured only through the united, systematically planned
investigations of the representatives of the General Government and
the State experiment stations in close cooperation with the farmers
and ranchmen. The grass and forage plant investigations at High-
more, South Dakota, are being conducted along these lines, and the
results are highly gratifying.

Some of the millets introduced from the plains of Russia, and some
of the varieties of sorghum and fodder corn, indicate the possibility
of obtaining profitable returns in forage for winter use. With vari-
eties sufficiently hardy to withstand the drought, or early enough to
produce a crop of forage before the drought begins, the question of
winter feed will be settled and the production of milk and butter
through the winter months assured. By the cultivation of a rela-
tively small area of land to these crops, the ranchman can produce
enough forage to bring his stock safely through the winter, and the
native pastures and meadows will be left in much better condition
because of the lighter drain on their resources.

One point can not be too much emphasized, and that is the careful
handling of the range so as to keep it up to the highest limit of pro-
ductiveness. Experiments in range renovation have begun by prop-
erly fencing the Station grounds. The difference already seen where
there has been a summer's rest, should be an object lesson to every
stockman. Experiments along different lines will be carried on next
season, such as scarifying and sowing grass seeds on -the native sod

and scarifying alone.

F. Lamson-Scribner,

Agrostologist.

Approved :

James Wilson,

Seci^etary of Agriculture.

Washington, D. C, December 20, 1899.

O

d

Circular No. 22.— (Agros. 59.)

United States Department of Agriculture,

DIVISION OF AGROSTOLOGY. UBRARY

NRW YOPK

[Grass and Forage Plant Investigations,] _.

BOTANfCAL

— ^ GARDEN

GRASS AND FORAGE PLANT INVESTIGATIONS ON THE PACIFIC COAST.*

INTRODUCTION.

The Department of Agriculture has from time to time carried on
investigations of the grasses on the Pacific coast, but the first really
systematic work of this kind dates from a personal visit made to that
section during the season of 1898. On this trip a number of the
more important localities were visited, especial attention being de-
voted to the investigation of the present conditions and needs of the
range region of central and eastern Washington and Oregon and the
areas subjected to the shifting sands along the coast region and at
various points along the Columbia River.

Throughout the drier sections east of the Cascades, the carrying
capacity of the ranges has been very much reduced through drought
and overstocking, and one of the questions of greatest importance in
this region has to do with the improvement of the range and bring-
ing it back, if possible, to its original productiveness. The present
worn-out condition has undoubtedly resulted from a combination of
circumstances. During the years* when the ranges were at their
period of greatest productiveness, they were grazed to their utmost
capacity, and upon the advent of the series of dry seasons, which
naturally resulted in light yields of forage, it was necessary to over-
graze in order to keep the stock alive. Then, again, the open win-
ters made it possible to keep the stock on the range a much greater
portion of the year than previously, and close grazing followed at a
season when the grasses could least endure it — namely, in early
spring. The grass was given no opportunity to recuperate, and the
present worn-out condition of the ranges resulted.

The large number of horses which have ranged over some sections
of the country have had much to do with the present depleted con-
dition, while in other sections much injury has been caused by sheep.
To any one visiting these grazing regions it is at once apparent that

* This report is based upon personal observations and upon communications
received from time to time from Mr. A. B. Leckenby, a special agent of this
Division in charge of investigations on the Pacific coast. A report in detail upon
the varieties grow^n by Mr. Leckenby is in prej)aration. — F. L. S.

2

drought-resisting grass and forage crops were of the utmost impor-
tance as well as improved methods of range management. In many
sections the native grasses have been driven out and their places
taken by weeds to such an extent that reseeding is about the only
way that the lands can be restored to anything like their original
productiveness. To do this, grasses and forage crops capable of en-
during severe drought and other hardships are necessary. It seemed
desirable to select some point or points where experiments could be
conducted, having for their object the testing of drought-resisting
grass and forage crops and of the range conditions. With this end
in view, a number of localities were visited. At Yakima and Walla
Walla, both in the State of Washington, it was found that some
work had already been undertaken ; that at the former place under
the auspices of the Northern Pacific Railway Company, and that at
Walla Walla by the Oregon Railroad and Navigation Company.
The officials of each of these companies expressed a readiness to coop-
erate with the Department in continuing in this work of testing
grasses and forage crops and range improvements, and went so far
as to offer to turn over for the use of the Department the lands occu-
pied by their respective experiment stations and any others that
might be deemed suitable or necessary for making such tests. For
the station at North Yakima this offer included tools, team, and
other appliances used at the station. These offers seeming to afford
a desirable opportunity to carry on these much needed investigations,
their acceptance was recommended. This recommendation meeting
with the approval of the Secretary, the Division took charge of the
experimental work at these places, and special efforts were made to
collect in quantity the seed of native grasses and forage crops that
seemed so desirable to be tested in connection with the work of range
improvement.

The investigations on the Pacific slope were continued during the
season of 1899, both in the field and at North Yakima, Walla Walla,
Rowena, and other points. The Agrostologist and two other mem-
bers of the office force of the Division staff spent a considerable por-
tion of the summer in the field studying the native grasses of the
region and other questions connected with the forage problem.

At various points along the immediate coast and on the Columbia
River, the question of fixing the' drifting sands is an important one,
and much time has been spent in studying the conditions which
exist there and in endeavoring to secure practical means of holding
the shifting sands in place. At some points along the river the sand
occurs in such abundance and drifts so badly that it is a decided
menace to orchards and farm crops in the immediate' vicinity, and
seriously impedes traffic by forming drifts over railroads and other
avenues of commerce. In the course of our investigations a number

of native plants have been discovered which promise to be of great
value as sand-binders, and some of them also as sources of forage.
Among the more important of these native sand-binders are sea-
side, or Astoria, blue grass {Poa macrantha) ; sand blue grass {Poa
leckenhyi) ; sea lyme grass {Elijmus arencunus) ; yellow lyme grass
{Elijmus flavescens); small sand lyme {Elymiis m^enicolus), simi-
lar to the preceding; and a sand-binding sedge {Car ex macro-
cephala). Other grasses that are being tested as to their value as
sand-binders in this region are big sand grass {Calamovilfa longi-
folia), Marram grass {Ammophila arenaria), Bermuda grass (C^w-
odon dactylon), and Johnson grass {Sorghum halapense).

EXPERIMENTS AT NORTH YAKIMA AND WALLA WALLA.

In the experiments undertaken at North Yakima and at Walla
Walla, a large number of grasses and forage plants have been tested
as to their ability to withstand drought, and, in many cases, also
have been grown under irrigation. As far as possible these stations
have been used in the production of seed to be distributed to the dif-
ferent points in the region where their value under existing conditions
can be tested. At North Yakima the land used in these experiments,
although rich, is quite rocky, and hence difficult to cultivate, and
the general conditions are much less favorable than those which pre-
vail at Walla Walla. At the latter place, the soil is a rich, volcanic
ash, and in fine condition for the various operations connected with
cultivation, and is so graded that irrigation may be resorted to when
necessary. During the present year a considerable amount of seed
has been pi'oduced, particularly at the Walla Walla station, and this
will be used in our investigations for the coming year.

VARIETIES.

In the experiments made at Walla Walla during the past year
about 150 varieties of grasses and forage crops have been tested.
These include most of the commonly cultivated sorts, varieties
recently introduced by the Department of Agriculture through its
special agents in foreign countries, and many native species, partic-
ularly those of the Rocky Mountains and the region east of the Cas-
cades. Most of the seeding was done during the first half of April.

Of the varieties tested some 25 or more have shown themselves to
be adapted to the conditions which prevail in the Northwest and
have in most cases given excellent results without artificial watering
of any sort. These include varieties adapted to all the various uses
in feeding stock ; varieties suitable for the ordinary meadows and
pastures and also those adapted for use on the drier uplands. They also
include annual varieties suitable for use in short rotations and varie-
ties adapted to soils strongly impregnated with alkali. The common

clovers, particularly alsike and mammoth red clover, have given
very satisfactory results and the same may be said regarding tim-
othy, meadow fescue, and tall meadow oat-grass. Smooth brome
grass has produeed the same excellent results here that it has given
elsewhere in tb (Northwest and gives promise of being to the drier
sections of this re; ;ion what the blue grass is to Kentucky and tim-
othy is to the northern States. Of the more recent introductions the
Japanese wheat grass promises to be of great value, particularly
for winter pasturage, while the varieties of alfalfa from Turkestan
and northern Africa seem to possess great powers of adaptability to
the conditions which prevail in the semiarid regions.

Some of the best results have been secured from our native grasses.
A species of brome, closely related to the rescue grass, has given
good yields of seed and forage and seems likely to prove as valuable
for the Pacific coast as rescue grass is for the South. Several of the
native wheat grasses show wonderful adaptability to cultivation and
are destined to assume an important place on our list of forage-pro-
ducing plants. Four of these deserve especial mention, namely,
western wheat grass {Agropyr^on spicatum), meadow wheat grass
{A. pseudorepens), slender wheat grass {A. tenerum), and bunch
wheat grass {A. divergens). The first three species are suitable for
meadows and may be grown either with or without irrigation. The
bunch wheat grass is a native of the dry uplands and is likely to
prove one of the best grasses for reclaiming the worn-out ranges.
The plants thrive under conditions of extreme drought and afford
excellent pasturage for all kinds of stock. This is the bunch grass
of the great grazing regions west of the Rocky Mountains, and for-
merly occupied extensive areas affording much pasturage. Another
native grass which does well under cultivation and which will un-
doubtedly prove valuable in reseeding the ranges is blue grama,
known in Montana as buffalo grass. It is perfectly hardy, responds
quickly to cultivation, resists trampling of stock, and affords
first-class grazing. Under favorable conditions of soil and moisture
it may also be cut for hay. This grass, however, has one drawback
in that the seeds are difficult to handle in the various harvesting and
seeding operations.

Of the annual grasses Japanese barnyard millet and black Russian
broom-corn millet are most deserving of special mention. Both of
these made excellent yields of forage and seed and are of undoubted
value to this section.

In addition to the above-mentioned grasses and forage plants which
may be regarded as of undoubted value to the Pacific coast region,
about twenty sorts were successfully grown at Walla Walla, the
present season, but further experimentation is necessary to decide as
to their real value. Most of these, while they made a good growth

of foliage, either failed to produce seeds or shbwed other characters
which may possibl.y exclude them from the list of varieties of prac-
tical utility for general cultivation. Johnson grass made a good
yield of forage, but is objectionable to some becau! e of its habit of
growth, which, in the South, renders it at times a serious pest in
cultivated fields. Both big blue-stem and bushg^ blue-stem thrive
under cultivation, but it is quite difficult to secure seeds that will
germinate. The Metcalfe bean made an excellent growth, covering
the ground thickly with its leaves and stems, and produced flowers,
but did not mature seed. It showed great sensitiveness to frost and
stock did not seem to relish the forage obtained from it. Gram, or
chick pea, and lentil produced good crops of seed and may prove
valuable for use in connection with grain and other feed stuffs in
fattening stock.

Quite a number of other grasses, some twenty in all, gave results
of sufficient promise to deserve further study and experimentation.
These include a number of the indigenous species of Poa, Elymus,
Paspaluin, and Mulilenhergia, which in their native condition are
valuable for hay and pasturage. The more important of these are
mutton grass {Poa fendoleriana), smooth bunch grass {Poa Iceviga-
ta), Wyoming blue grass {Poa wheelef^ii), Canadian lyme grass
{Elymus canadensis), smooth paspalum {Paspalum Iceve), and
wild timothy {Muhlenberg ia racemosa).

Of the sand-binding grasses which were tested during the season,
Marram grass has made a fine growth and will, no doubt, prove as
valuable here as it is on the Atlantic coast. Bitter panic grass
{Panicum ainai^um) made a good growth, but failed to mature seed,
and for this reason may prove to be of less value here than it is along
the coast of the Southern States. Reed canary grass {Phalaris
arundinacea) has made an excellent showing, particularly in situ-
ations that are liable to be submerged for some time ; for example,
along the Columbia River. Even where submerged six weeks dur-
ing the summer, excellent growth was made early in the spring and
late in fall, affording a large amount of forage. In addition to serv-
ing as a sand and soil binder, this grass has also made a good show-
ing on the drier land of the experimental plots at Walla Walla, and
also at Pullman on the grounds of the State Experiment Station.
Small sand lyme grass {Elymus arenicolus) has made an excellent
growth at Walla Walla from roots transplanted from Grants, Ore-
gon, indicating that it will grow on ordinary soil as well as dry,
sandy ground, where it occurs naturall3^ It will undoubtedly be of
great value as a sand-binder. Big sand grass {Calamovilfa longi-
folia) has not given very good results and may prove vmsuitable to
the conditions which prevail on the Pacific coast. Seaside blue grass
{Poa macrantha) has given excellent results wherever tried and,

6

especially at tlie Walla Walla station, seemed to endure the hot, dry
weather perfectly. In addition to being one of the best native sand-
binders of the coast region, this grass gives promise of being of great
value for forage. Seed sown on the sand dunes of the Great Lake
region the past season made a good showing, indicating the possibil-
ity of introducing this grass into the sandy regions of the interior.
Indian millet {Eriocoma cuspidata), although not coining very
well from the seed, shows indications of being valuable as a sand-
binder, especially on relatively high and dry situations. Although
the forage is rather harsh, it is often eaten by stock which are espe-
cially fond of the rather abundantly produced seeds.

Mr. A. B. Leckenby writes to the Agrostologist, under date of
December 11, 1899:

I am delighted to tell yon that the Elymus arenarius is a grand siTccess, where
the water does not remain on it too long, as is also Elymus arenicolus and Ely-
mus flavescens. You would be delighted to see the grasses growing through seven
feet of drifting sand. The Elymus arenarius is particularly pleasing because of
its luxuriant growth. I planted about one acre with Elyiiius flavescens, Thurs-
day and Friday, with roots ijrocured from the Dalles. I am sending you by
same mail some roots of this same grass.

Of the various saltbushes tested, Australian saltbush {Atriplex
semibaccata) has given the best results of any of the introduced
sorts, and white or sweet sage {Eurotia lanata) is most promising
of the native series. The latter grows naturally on the dry, sterile
soils of the ranges and adapts itself readily to cultivated conditions
producing an abundance of seed. It will undoubtedly prove of great
value for use in reclaiming the worn-out ranges.

In November some very interesting observations were made at the
Walla Walla station regarding the behavior of different grasses and
forage crops toward the early frosts. A variety of brome grass
secured at Portland, Oregon, was not at all injured by the early
frosts and remained fresh and green, while nearly related varieties
from the Rocky Mountains suffered quite severely. Seaside blue
grass and other of the Poas, or blue grasses, showed little, if any,
effects of the frosts, while the grama grasses and the blue-stems
early became dry and brown. Bearded wheat grass {Agropyy^on
caninum) remained fresher and greener than either slender wheat
grass or western wheat grass. Giant rj'-e-grass {Elymus conden-
satus) was injured while other rye-grasses, such as Canadian rye-
grass and Terrell grass, were very slightly affected. Turkestan
alfalfa was not injured at all, while Oasis alfalfa from Africa suf-
fered considerably and the commonly-grown variety of alfalfa was
slightly injured. Such annuals as bur clover, lentil, and gram were
not injured, indicating their possible value for fall and early winter
grazing. Smooth brome and Oregon brome were not at all injured;
and were growing finely.

i

GRASS AND CLOVERS AVAILABLE FOR USE IN THE UPPER
PACIFIC COAST REGION.

Judging from the results of the experiments conducted at North
Yakima and Walla Walla and the information acquired through
personal observation of our own field agents and correspondents who
are cooperating with the Division, the following grasses may be
suggested as available for use in the upper Pacific coast region :

Grasses and clovers for permanent meadoivs and pastures where the average
conditions of soil and climate prevail— Alsike, mammotli clover, meadow fescue
{Festuca 2)ratensis), red fescue {Festuca rubra), reed fescue {Festuca arundi-
nacea), tall meadow oat-grass {Arrhenathervm elatius), western wheat grass
{Agropyron spicatum), meadow wheat grass {Agrojryron pseudoi^epens) , slender
wheat grass {Agropyron tenerum), smooth brome {Bromus inermis), Kentucky
blue grass (Poa pratensis), perennial rye-grass {Lolixim perenne), Turkestan
alfalfa (Medicago sativa var. turkestanica) .

As deserving of trial: Wild timothy [Muldenhergia racemosa), bearded wheat
grass (Agropyron caninnrii), smooth bunch grass {Poa Icevigata), mutton grass
{Poa fendleriana), Terrell grass {Elymus virginicus), Wyoming blue grass {Poa
ivheelerii), ])ale bunch grass {Poa lucida), and Oasis alfalfa.

Hardy annuals.— Rescue grass {Bromus unioloides), Japanese barnyard millet
{Panicum crusgalli), Japanese wheat grass {Brachypodium jajjonicum), black
Russian broom-corn millet {Panicum miliaceum), bur clover {Medicago denticu-
lata), spring vetch {Vicia sativa), hairy vetch {Vicia villosa).

The following are suggested for further trial :

Crowfoot or goose grass {Eleusine indica) African millet {Eleusine coracana),
gram [Cicer arietinum), \ent.il {Ervum lens).

Grasses for the dry uplands.— Bine grama {Bouteloua oligostachya), side oats
grama {Bouteloua curtipendula), bunch wheat grass {Agropyron divergens),
Nevada blue grass {Poa nevadensis), sheep fescue {Festuca ovina), King's fescue
{Festuca kingii), white or sweet sage {Eurotia lanata), sainfoin {Onohrychis
sativa), smooth brome.

Sand Binders.— Sea, lyme grass {Elymus arenarius), slender sand lyme grass
{Elymus arenicolus), yellow lyme grass {Elymus flavescens), Astoria blue grass
{Poa macrantha) , Leckenby's blue grass {Poa leckenbyi).

F. Lamson-Scribner,

Agrostologist.
Approved :

James Wilson,

Secretary of Agriculture.
Washington, D. C, December 22^ 1899.

I

• y'lKcri.Ai! No. 2:; — (Autos. 6:i).

United States Department of Agriculture,

DIVISION OF AGROSTOLOGY.

[Grass aiid Finite Plant Investigations.]

V. LAMS0N-SCRIJ5XEK, Agrostolouisl.

PROGRESS OF EXPERIMENTS IN FORAGE CROPS AND RANGE
IMPROVEMENT AT ABILENE. TEX.

[Ill a great many sectious of the Southwest the natural rauges and jjaslnreis have
lieen nearly or entirely destroyed by overstocking. The nativ<' grasses have died
out through excessively close grazing, and their places have been taken by useless
weeds and shrubs. Owners of what have heretofore been fine grazing properties have
tlitis snffered great loss in being deprived of their pastures, and are now engaged in
an effort to somewhat restore the former conditions, if possible.

The Division of Agrostology in 1898 commenced a three-years' experiment with a
view to helping in this matter. A body of land (i40 acres in extent, near Abilene,
'I'cx., Avas kindly loaned by its owner, and tlie citizens of that town cheerfully
assisted in the movement by having the property placed under fence, building water
tanks, etc. Mr. II. L. Bentley, special field agent of the Division of Agrostology,
was placed in charge, and has been carrying ont the plan of ex]>erimentation devised
by the Agrostologist of the De]iartment.

'J'liis is Mr. Bentley's second report of jirogress of experiments in range improve-
ment, the first one having been submitted and iirinted in December, 1898, as Circu-
lar No. 8, and entitled " Experiments in IJange Imi»rovements.'" — F. L. S.]

Tliere have been iiuuiy serious difficulties to contend with during the
past season. In the first place the months of January and February,
especially the latter, were very unfavorable for experimental work. In
order to explain the conditions that prevailed here, a statement show-
ing the protracted and extreme cold during the month of February is
l)resented.

Tabic showituj the dales diiruKj February, 1S99, when tlie temperature
WHS below the freezing point.

Date.

Fel). 1
2
3
4
5
6
7
S
9
HI

Degrees. !

18
25
14
14
12
10
14
13
13

Date.

Degrei's.

,I>.11

]

12

(3

13

1)

14

26

15

22

J8

30

22

28

23

17

24

■'(I

26

•'(1

9

There were only eight days during the mouth when the temperature
was above 32 degrees.

A large number of seeds were sown during the month of October,
1898, and others during the January following, but as a result of the
extreme cold weather during 1^'ebruary, everything that had come up
was killed by March 1.

The month of January was very dry. Tlie following table shows the
precipitation during that month:

Date.

Inch.

0.06
.10

.24

Jan. 11 .

23 .

27 .

Date.

Inch.

Jan. 4 ...

-

0. 02

1 .08

q

! .01 1

'Jotal for the month, 0..")1 inch.

I>uring February there was only 0. 01 inch of rain, which fell on the
4th, and during March there was only 0. 04 inch, which fell on the lltli.
The ground in the pastures was not in condition to be harrowed, and
the grass garden could not be plowed and made ready to receive
seeds until April. During that month the rainfall was about normal
for the season. The following table shows the precipitation during this
month:

Date.

Inch.

Date.

Inch.

Apr.

.*!

6

0. 30
.56

15

.78

Ki

.24

Apr. 20 0. 38

21 1 .30

oo I .40

'total lor the month. 2.96 inches.

The rain on April 5 and 0 put the pasture ground in fair condition
for harrowing, and on April 10 that work was begun and continued,
with some interruptions, until completed.

During the autumn of 1898 the 10 acres in the grass garden had been
plowed deep with a turning plow. March 15 following, notwithstand-
ing the dry weather, the ground was in fair condition; as a result of
the freezing during February it was easily pulverized, but was very
cold and contained but little moisture. However, in anticipation of
rain, the work of seed planting was begun March 15 and continued
tlirough April until May 4, Many of the seeds planted in March did
not germinate at all, which was most probably due to the fact tliat the
ground was very cold and the vitality of the seed thereby destroyed.
The seeds that did germinate made a very vigorous growth. Tlu^ same
conditions, however, that induced such growth also induced a corre-
sponding growth of weeds. The grass garden was laid oft' in plots 20

feet wide, separated into subdivisions according to the (quantities of
seeds in baud. Tbe plots and subdivisions were marked by rows of
sorgliuni, as a rule, and the grass seeds were sown broadcast. It was
impossible to use plows in fighting weeds, aiul the work of getting rid
of them had to be done by hand. It was necessary for the weeders to
be in the plots, and as the giouiul was very wet a portion of the grasses
in the plots was also destroyed in the effort to get rid of the weeds,
either by trampling or by being pulled up with the weeds. Tbe rain-
fall during May aggregated 4.02 inches, much above normal. During
June it aggregated 5.45 inches. During July there was a sudden fall-
ing off in precipitation, as shown by the following table.

Date, Iiicli. ' Date. lucli

Ill.ll.

Date.

0 52

.Tills- 17

10

24

.July 1 0.52 .Tills- 17 ». '-'O

.50

'J'dlal IVii- I III- iMiiiilh, 1 . :;(S iiiclit's.

By the 1st of August everything in the station garden had begun to
show the inevitable effect of the drought. Hot winds were blowing
during all of July, and the succulent garden i)lants presented the
appearance of having been scalded with hot water, and in some
instances were killed. During August the rainfall was only normal,
being 0.10 of an inch on the 16th. In September there was a fraction
more, viz, 0.44 of an inch on the 7th. During the first twenty-five
•lays of October there was a precipitation of only 0.01 inch, falling on
the IGth.

In view of the hot winds and the protracted dry weather it is remark-
able that anything in the garden survived. October 2C the first rain-
fall of any consequence since June 30 (when tlie precipitation was 2.01>
inches) occurred; the precipitation was 2.S0 inches. During November
there were several rains, viz:

1

1

1

Date.

Inrh.

1
0. 18 !

Date !

Ineli.

Xciv.

, 1 ..
18 ..
211 ..

' N..V,

. 21 ..

29 ..

1

0. 12
. 90 1

... . .06

I. 10

1

'I'lital lor the iiidiitli, 2.36 inches.

Of tin; plants in the garden, the several varieties of alfalfa, sul!a,
sainfoin, and hairy vetch greened out wonderfully, and have since
then made vigorous growths, in spite of the frosts during the months
of November and December. Following the slight rainfall of Novem-
ber 1, there was a killing frost (the first of the season) on the night of
the 2d, and there was much freezing weather during the months of

November and December, as will appear by the following table, sbowing
the dates when the temperature was as low as 32 degrees :

Date.

Degrees.

Date.

Degrees.

Not. 3..

4..

Dec. 4..

5..
14..

28
30
32
29
26

Dec. 15

26

27
28
32
32

19

20

......

21

oo

In view of the extremes of heat and cold and dry and wet weather,
it will be seen that there were serious difficulties to contend with
during the season of 1899.

RANGE IMPROVEMENT.

METHODS EMPLOYED.

The section of 640 acres was divided into six pastures of 80 acres
each, two of 40 acres each, one of 70 acres, and the remaining 10 acres
were set apart for testing such grasses or forage plants as fairly
promised to be of practical value in the semiarid regions of Texas.

The work as planned by the agrostologist in 1898 was as follows:

Pasture No. 1. — No treatment except to keep stock oft' until June 1,
pasturing the balance of the season.

Pasture No. 2. — Cut with a disk harrow and kept stock oft' until

June 1, pasturing the balance of the season.

Pastures Nos. 3 and 4 (of 40 acres each). — Grazed alternately, the

stock being changed from one pasture to another every two weeksj

thus allowing the grasses a short period for recovery after each grazing]

Pasture No. 5. — No treatment except pasturing until June 1 an(
keeping stock off the balance of the season.

Pasture No. 6. — Left as a check, without any treatment whatevei
except to keep stock oft" during the first season.

Pasture No. 7. — Dragged with an ordinary straight-toothed harrow
and stock kept oft" during the first season.

Pasture No. 8. — Disked and stock kept oft' during the first season]

The 70-acre pasture, No. 9, was not grazed. Seeds of a number o|
wild and cultivated varieties were sown directly upon the sod.

Three of the most experienced stockmen of Texas personally inspecte(
every acre of the section March 23, 1898, and unanimously reported thai
its utmost capacity for supporting cattle at that time was 1 head oj
mixed stock to every 10 acres, or 40 head to the section. The rainfall
during 1898 was exceptionally light, and practically nothing wag
accomplished that year on the station except the cultivation of the|
native sod — that is to say, a disk harrow was used in two of the 80-acre
j)astures to cut up the sod as thoroughly as i>racticable. An ordinar;^
iron-tooth harrow was used on one 80-acre i)asture, and the surface ol

the land was scarified as thoioughly as was possible uiidei' the circum-
stances. There is a rather heavy growth of mesquite trees growing in
each of the pastures, and on account of them it was not possible to
scarify the surfa(;e of every acre.

The theory was, that by cutting the surface as deep as was practica-
ble with the harrows (disk and tooth) the roots in the ground would be
given better chance for development, the runners from the grasses
would find soft ground in which to take root readily, the rain would
sink into it instead of running oft', and the seeds that fell would find
a suitable jdace in whicli to germinate. In suite of the drought of

»

Yui, i,_ri:m 111' (UO-iicrc tract used (or tlio range experiments at Abilene, Tex. (Scale 3 inches to

tlie mile.)

1898, during wliich there was less rainfall even than during the mem-
orable dry year of 1887, theie was a marked improvement in the range,
treated as above mentioned. The harrowing was done during March and
April, 1898, and, fortunately, the rainfall was reasonably good just after
the work Avas completed and the autumn rains were nearly up to the
nornuil, although the sum of precipitation for the season was very much
below normal. A year from the first inspection, the same three stock-
men made a second visit to the station and unanimously reported a very
marked improvement over the conditions as tliey had found them in
March, 1898. They reported in March, 1899, that the capacity of the

6

section to support stock was tlien 1 bead of mixed cattle to 10 acres,
or ()4 bead to the sectiou. Later, iu April, 1899, two of the men (Messrs.
Middleton and Bryan) made another inspection of the pastures after
the long spring rains had commenced falling, and reported that 1 head
of mixed cattle to every 8 acres (80 head to the section) was the then
capacity of the section. It is on this basis that cattle are now being
held and will be held during the next year in the station pastures.
This gain of 100 per cent in capacity to support stock is phenomenal,
considering the drought of 1898, and had the pastures been stocked
to their full capacity, as recommended by the inspectors, I would
have taken the result as finally demonstrating the great value
of cultivating the sod as explained above. But on account of the
scarcity of water for stock in the pastures it was not possible to keep
in them the proportion of animals contemplated, hence the pastures
were rested for several months, and precisely at the season when the
grass seeds were maturing and dropping. In determining, therefore,
what factors entered into the experiment resulting in so marked an
improvement in the range conditions, the fact of the resting of the
pasture must not be overlooked. Indeed, I am satisfied that the
improvement in the said range conditions has been due largely to this
period of rest. My conclusions are: (1) That it Avill pay farmers and
stockmen of Texas, especially in the semiarid districts of the State, to
cultivate their pastures by use of disk and iron tooth harrows; (2) that
it will pay them to rest their pastures periodically during the seasons
when the grass seeds are maturing and falling to the ground.

COST.

A farmer or stockman having his own teams and hands can do the
work incident to these experiments when practically nothing else can
be done on the farm or ranch, and at a nominal cost. The best time
for such work is believed to be in the early spring, just before or Just
after rains. But it is quite probable that it will pay also to have sim-
ilar work done in the autumn. The difiiculty that is likely to be met
with in making these experiments in the fall is that the rains at that
season are not very regular in this sectiou, and are generally followed
by frosts; hence, it is quite ]>ossible that the damage caused by expo-
sure of the grass roots, especially where the disk harrow is used, may
more than ofiset the advantage that may be secured. The matter is
worth testing and will be tried as soon as practicable.

CATCHING WIND-BLOWN SEEDS.

Another experiment has been instituted on one of the station pas-
tures that promises to result favorably, namely, catching the grass
seeds that are blown about by the winds.

The prevailing winds during the summer months in central Texas

I

I

are from the south. Many of the grass seeds, as they fall to the jiround,
except where the surface has been scarified or plowed, are blown by
the winds onto other lauds. In this way such seeds are often lost to
the grounds on which they are growu and matured. In order to save
these seeds to the pastures producing them, a part of one of the station
pastures was selected, and about every 12 feet furrows have been
plowed from east to west, the idea being, (I) that the seeds falling. to
the ground, if blown at all by the winds, will be caught in these
furrows and in that way saved to the pastures, (2) that the storm
waters falling on such pastures, instead of being allowed to waste by
running off into creeks and bottoms, will also be caught in the furrows.
This work was done in May, 1899. IJy June 30 the furrows had
caught a great many seeds, and, as the result of the surface irrigation
incident to the rain having been collected in them, the grass immedi-
ately proximate to such furrows was much greener and more vigorous
than that farther away. Indeed, in approaching the pasture wo treated
the furrows could easily be traced by the eye a half mile or more away
by their fresh, green appearance.

TRANSPLANTING CrKASS ROOTS.

Another experiment in the improvement of range conditions has
been the transidantiug of grass roots. There were considerable areas
ill the pastures of the station that were bare of all vegetation a year
ago, the result of overstocking, drought, insects, and prairie dogs.
Several native turf grasses were selected the past spring and the sods
were taken up and planted in a few of these naked spots, just before
rains. In every instance they have grown well and promise to cover
the naked spots in a short time. The following grasses were used in
this experiment : Curly mesquite {H liana cenchroides), needle grass
{Aristidafasciculata), cotton-top [Panicum lachnanthum), wild timothy
(Mtihlenheryia racemosa), Galleta or black grama {HiJaria mutica), h]ue
grama (liouteloua oligostachya), side-oats grama (Bouteloiia curtipen-
dula), Canada rye grass {Elynms canudensls)^ and everlasting grass
{Eriochloa ijunctata). All of these are natives of central Texas and
occur in nearly every county. There are many other native grasses
and forage plants here that may be used to good advantage in the

same way.

GARDEN WORK.

In the 10 acres set apart for experiments with seed of grasses and
forage plants the tests made this year have not been altogether satisfac-
tory, chiefly owing to the dryness of the season, but results of practical
value have been secured.

The soil is thin and dry, as a rule, and part of the ground is too
rocky for cultivation. Tlic bed of a small stream extends through the
east side of the garden and drains the moisture from the lower lands;
next to it is the dry soil, characteristic of the upland. The stream has

water in it only in.mediiitely after rains; it is a "dry branch" as
understood in Texas, About 100 different varieties of seeds were
received from tlie Division of Agrostology and tbey bave been carofnlly
tested, in many cases with satisfactory results.

SALTBUSHES.

Seeds of several varieties of saltbusbes were planted, but only a few
of them germinated. The plantings were made March 16 and April 18.
The seed of Nelson's saltbusb {Atriplex pahularis) came up, but only
a few plants were saved. Of iS'uttairs saltbush [Atriplex nvttaUii)
only a few vigorous plants were secured. About a dozen weak plants
of the shad scale [Atriplex canescens) came up, but they soon died. Of
all the varieties of saltbusbes that were tested the annual saltbush
from Australia [Atriplex holocarpa) gave the best result. The seeds
were planted March 16 and April 18. A number of vigorous plants
were secured, but most of them ^ere destroyed in one night by a
prairie dog that made its way into the grass garden through a woven '
wire fence. The two plots of saltbush planted were on thin land.
The growth was not very vigorous, being only from 12 to 18 inches, but
the seed development was surprisingly great. As a drought-resisting ^
plant too much can not be said in its favor. Seed formed and matured
in the long, hot, dry summer. Up to December 1 the bushes continued
green, but by December 15 the plants were nearly dead. On each of
the varieties except the shad scale there were many thousand seeds,
and although twice in July and August they were carefully stripped of
seed, the ground under each bush is now covered with hundreds of the
seeds that have ripened and fallen off". It is said that the saltbush
will mature several crops of seeds in a season, and this is not hard to
believe considering the results of the tests in the station garden this
year. It withstood the utmost extremes of drought. November 1, after |
four months of dry weather, with the hot winds which at that season *
prevail in this section and in all the western counties of Texas, when
nearly everything else in the grass garden Avas showing the effects of
the heat and the dry weather, the saltbush appeared nearly as vigorous
as it did during the wet weather. The saltbusbes are recommended
especially for the sections of Texas in which the soils are impregnated
with alkali. The annual saltbush is valuable; it is an excellent forage
plant for soibng sheep aud cattle, and horses will eat both the bush
and the seed.

LEGUMES.
ALFALFA OR LtTCERN.

Seeds of alfalfa [Medicar/o sativa) were sown in March, 1808, on the
freshly broken sod. A rather thiu stand was secured and when it had
grown to about 6 or 10 inches in height, the long dry spell began and
checked all further growth; indeed, it gradually dried up on its roots
until scarcely a stem of it was to be seen. The following fall and win- ^

J

9

ter the prairie dogs and rabbits destroyed pretty nearly all the roots
left by the drought. In the early spring of this year (1899) a few of the
steins were noticed after the ground had been plowed, and they were
carefully protected. Without other water than the natural rainfall the
plants developed remarkably. By June 1 they had grown to be from
18 to 21 inches tall, at which time the crop was cut. On July 2 it was
cut again, many of the stems measuriug from 24 to 28 inches. July 22
a third crop was ready fur cutting, some of the stems measuring from
18 to 20 inches in length. It is true the rainfall here from April 1 to
July 1 had been above normal, but I am strongly inclined to believe
that alfalfa will thrive in this section without irrigation. April 25,
1899, one pint of seed was procured and sown in a rather low but not
at all wet plot of ground, attbrding an excellent stand. The growth
continued to be vigorous up to about July 20, by which time the severe
drought had set in. The foliage was dried and parched by August 1,
and even the roots appeared to be dying. October 26 the drought was
broken by a good rain, and by the 30th green leaves began to start
from the roots. By the last of November the new foliage was from 8
to 10 inches in height. There were several white frosts and one or two
freezes in November, but December 1 this plot was as green as in
spring time, and the new growth was from 12 to 18 inches tall, and is
even yet (December 30) green and vigorous looking.

TURKESTAN ALFALFA.
{Medicago satiia iiirlestaiiua. )

The seed of the Turkestan alfalfa was sown in April on rather thin
dry land and only a partial stand was secured. It grew well until the
drought set in, but soon began to show indications of injury. By August
1 the plants began to drop their leaves, and by September 1 the roots
also appeared to be dead or dying. After the rains (October 26) the
roots were slow to revive, but by November 15 they had put out a new
growth. By December 1 this growth was from 11 to 12 inches tall, but
less strong than the common alfalfa. At this date the plants are green
and vigorous.

OASIS ALFALFA.

A small package of oasis alfalfa from Tunis was received through the
Division of Agrostology. Three plantings were made, one in February,
1899, which, on account of the severe frosts that followed a day or two
later, failed to germinate. In April one half of the balance of the seed
was sown in a low but not damp plot, and what was left was planted
on higher land. Excellent stands were secured from each of these
last two plantings. The growth was vigorous from the start, but by
July 30 the plants were looking wilted and by August 1 the foliage
had withered badly and the roots were apparently dead. I reported
about that time that I feared the roots were all dead. The rain that
fell October 26 quickly revived them; by November 5 fresh leaves

10

were putting out, aud by Isioveuiber 15 tlie new growth was from 8 to
12 inches tall. On December 1 the growth was from 12 to 20 inches
high and at this date (December 30) the plants are still green and
vigorous. In approaching the station garden the vigorous growth of
these plots can be distinctly noticed at a distance of half a mile. The
results of the ex])eriments with this particular variety of alfalfa are
YTg^y gratifying and it seems likely to do well in this section without
other moisture than a normal rainfall. To date, the oasis alfalfa is more
satisfactory than the other varieties.

RED CLOVER.

Seeds of this variety were sown in two plats and good stands secured
in each, one in low the other in higher land, both dry. The plantings
were both made in April. In each case the growth was vigorous at
the start, but on account of the early spring rains a very heavy growth
of weeds resulted, that choked out many of the clover roots. In spite
of this difficulty, by May 5 there was a vigorous growth from 8 to 16
inches high, that continued green through June and as late as July
20. By August 5 the hot, dry winds had very much injured the plants;
by September 1 the foliage was withered, and soon even the roots were
mostly dead.

WHITE CLOVER.

This variety failed to yield a good stand, and at no time during the
year did it do as well as the red clover. The weeds got the start
of it in the early spring, and when the drought set in the foliage soon
withered, and by September 1 most of the roots were dead.

CRIMSON CLOVER.

The seed of this variety was sown in a low flat plot near a dry
"branch." The early spring rains filled the branch and kept it full of
water several weeks. A fair stand was secured, but the weeds soon
injured it materially. Soon after the drought set in the plants began to
languish, and by September 1 the roots were dead. The foliage turned
yellow during the early spring rains, evidently the result of too much
moisture. j

MAMMOTH CLOVER. 1

This variety was also tried without obtaining a first-class stand and
the weeds soon thinned that which was secured. The growth was
never very vigorous and gave way entirely soon after the drought
began. 1

BUR CLOVER.
(Medieaijo denticitlata.)

Only a few seeds of this variety were sown, therefore only a few
bunches of clover were secured. These grew up nicely and continued
green until about August 1, when they were checked by hot, dry winds

i

11

A inontli later most of tlie roots were apparently dead. Quite recently
(December) some of the roots liave taken on new life and are coming-
out nicely. They are at this date not only alive, but have developed
new foliage in spite of the freezing weather since the October rains.
I desire to mal<e another test of this variety of clover during 1900, and
am inclined to think that it will do well here under normal conditions.

VETCHKS.

Few of the farmers of Texas are familiar with the vetches. Two
varieties were successfully tested during 1899, viz, spring vetch (Vicia
sativa) and hairy vetch ( Vicia villosa). Seeds of Stolley vetch ( Vieia
leaven n-ortim) were sown in February, but the freeze that followed a
few days later and continued through the month destroyed them. In
the same month seeds of the other two varieties named were sown, but
few of them germinated and none of the plants (not vigorous at any
time) survived. The later plantings (March 17 and April 10) did well.
The spring vetch ( Vicia sativa) is an annual trailing herb and grew
from 18 to 20 inches tall. The hairy vet(;h has been successfully grown
without irrigation in the moist coastal regions of Washington, on the
dry ])rairies of South Dakota, and the rich loamy soils of the Gulf
country. Both of these varieties, the spring vetch and the hairy vetch,
if sown in Texas in April or May, will be ready for harvest by the
middle of August or first of September. Texas farmers are strongly
recommended to try both of them. This year seed sown broadcast, in
drills, and in hills all gave satisfactory results.

«rL,LA.

(Hedj/saruiii coroiiarinm).

This is a perennial legume, native to southern Italy, that should be
known to and cultivated by our Texas farmers. The main plot was
planted May 4 and is on the side of a dry, gravelly hill. From the
start the growth was vigorous and by the time the hot, dry winds
began in July the foliage was from C to 12 inches long. The roots
penetrated the ground to a considerable depth and up to about August
1 it seemed that this crop Avas going to stand the drought well. By
tlie 15th of that month, however, it began to show the effects of drought
aud by September all the foliage was Avithered and dead. From that
time until late in October there was no evidence of vitality in the
plants. Following the rain of October 20 the roots revived very (juickly
and by November 10 the new growth was from 3 to G inches tall; since
then the roots have been growing down into the earth and the foliage
has been pushing up, and tlie plot is still green. Roots were sent to
the Department of Agriculture recently that measured 10 to 14 inches
in length with foliage from 5 to S inches long. The recent freezes, none
of them very severe, however, do m)t appear to have made any impres-
sion either on the roots or foliage. The good showing made by this

12

forage plant on this upland soil, with the severest drought ever known
here, certainly indicates that it may prove of much value in this section.
There can be no doubt of its great value in portions of Texas less sub-
ject to long, hot, dry spells than this immediate section.

SAINFOIN.
(Onobrychis saiira).

This plant, known also as esparcette, is a deep-rooting, perennial
legume that thrives on dry calcareous soils which are not favorable for
either clover or alfalfa. Seeds were sown May 4, and the plants which
were from 12 to 18 inches high stood the dry and hot weather better
than the vetches, clover, or alfalfa, and (juite as well as the sulla.
There is no better forage plant for barren hillsides.

COWPEAS.

About sixteen varieties of cowpeas were tested this year, all of them
giving satisfactory results. The seed was planted in drills 3 feet apart,
and cultivated as long as the season permitted by stirring the surface
of the ground between the drills and about the roots. All the plant-
ings were made on the same date, April 11, as there was but little seed
of each variety. The plants grew nicely and all matured seed, none of
them doing as well as they would have done had the season been more
favorable. Seed of each variety was harvested, and will be tried again
in 1900. The vines failing on account of the hot, dry winds, were cut
and stacked in July, and several bales of excellent hay much relished
by cattle and horses were secured.

CANADA FIELD PEA.

Seeds of the Canada lield pea were planted April 10, resulting in
only a very indifferent stand. The vines grew well, luxuriantly in
fact, bloomed well and matured seed abundantly. Both for the vines,
which make an excellent forage, and for the seeds, which are quite as
rich as the well-known garden pea, which it much resembles in appear-
ance, this variety will doubtless prove of much value to farmers and
stockmen in central Texas.

MUMMY PEA.

A few seeds of this variety were planted April 11 in a heavy soil.
Only a few germinated. A vigorous growth of vines was secured, but
only a few seeds matured.

BLUE PEAS.

A small planting of this variety was made April 11 on the bank of
the "dry creek" elsewhere referred to. During the next ten days the
rainfall was excessive, the creek overflowed, and most of the vines were
destroyed. The few that were left grew luxuriantly but bloomed
sparingly, and only a few i)ods of peas matured.

13

VELA'ET IJEAN.

Tlie plot of velvet bean in the uarden witlistood the hot, dry spell of
weather better than anything else except the soy bean, up to about
August 15, but by September 5 the leaves were drying u]> and drop-
j)ing, and by October 1 the plants were mostly dead. The seed was
planted in hills about 4 feet apart each way. The vines covered the
ground completely and vsome of them grew to be 10 or li* feet long. Xo
seed was matured.

SOY BEAN.

The soy bean promises to be one of the best foraae plants for this sec-
tion. Three varieties, early, medium, and late, were tested in the station
garden. The early variety grew to be from 8 to 12 inches high and the
plants were well tilled with pods and seed. The medium variety grew to
be from 8 to 12 inches tall and produced a very heavy yield of seed. The
late variety proved a vigorous grower. The plants were from 24 to 30
inches high, but they matured only a few beans. This variety did not
stand the dry, hot weather quite as well as the velvet bean.

METCALFE BEAN.

(Phaseohis retusus.)

A few seeds of the Metcalfe bean were received from the Division of
Agrostology and a few from Mr, J. K. Metcalfe, of Silver City, IS". Mex.
Three plantings were made, March 21, April 11, and April 18,
respectively. It was given the best possible chance, in the hope that
it would do well in this section. Good stands were secured, and the
vines grew nicely, some of them reaching 8 or 10 feet in length. Up
to July 1 there was every indication of good results, but very soon
after the drought began the vines showed evidence of injury, and by
August 1 most of them were either dead or very much dried up.

NATIVE HORSE BEAN,

(Phaseolua angustissimus. )

Seed was also received from Mr. Metcalfe of a beau which stockmen
here recognized as that known in New Mexico and Arizona as the horse
bean. The seeds were planted April 11 and produced only a few
plants that grew vigorously and promised good results, but they were
injured by drought and produced no blooms, although the vines con-
tinued green until early in August. By October 15 they were all dead
and the roots did not revive after the rain appeared in the fall.

LENTIL.

From seeds planted April 11 a good stand was secured. The growth
was not vigorous, only from 6 to 8 inches tall. Seed formed, but the
plot was on the bank of the "dry creek" mentioned, and in an over-
flow all the plants were destroyed.

14

SORGHUM.

Several varieties of sorghum were tested iu the garden, ibe non-
saccharine as well as the sweeter varieties. All the saccharine varie-
ties did well. The Dhonra corns, Milo maize, Jerusalem corn, "(lyp"
corn, Kafir corn, and other varieties of the nou saccharine sorghums all
did as well as any reasonable farmer could desire. Satisfactory crops
were secured from each of the following varieties: Big Amber, Ked
Amber, Early Black Amber, and Orange cane; Milo maize, ''Gyp"
corn, Jerusalem corn, Eed Kafir corn, White Kafir corn, White Brandl-
ing Dhoura corn, and Brown Branching Dhoura corn, all planted
April 25. All of the varieties of sorghum tested in central Texas
have proved fairly well adapted to the conditions of climate ordinarily
prevailing here. They do better during dry years than anything else
in the line of grains or forage crops that are cultivated. They, as a
rule, yield large crops, and there is no difficulty iu curing and pre-
serving them. The idea of baling the hay, noted elsewhere, was to
determine if they could be handled successfully in that way. If babng
can be practiced they can be grown here to great advantage, and ought
to possess considerable commercial value. ISTo richer stock food can be
grown in central Texas, and if the hay can be shipped to advantage it
will command good prices farther west.

TEOSINTE.

Teosinte was successfully cultivated in the station garden this sea
son. A long season of hot weather and a rich, rather moist, soil are
necessary for its full development. One plot of it was in soil that was
neither very moist nor very rich, and it made a vigorous growth in
spite of the dry weather. It somewhat resembles sorghum in appear-
ance, but has a more branching habit, usually making from 20 to 00
shoots from a single seed and growing from 5 to 10 feet tall. It will
prove a valuable forage plant in all parts of this State, to be fed green,
as silage or as fodder.

One planting was made April 10, in hills 4 feet apart each way. By
July 1 it had met in the rows ; but by July 15 the drought had set in, and
by August 1 the teosinte had begun to suffer seriously. After that date
no seed heads were formed, and those that had previously formed did
not develop further. By August 15 the fodder had become yellow and
dry, though the roots were still alive. Had a good rain fallen on them
even as late as September 1, I am confident they would have revived
and put forth a further growth of foliage. However, by October 20,
when the drought was broken, even the roots were dead. Early in
July some of the stalks were cut, cured, stacked, and later baled.
The bales have kept sweet, and those still on hand are all that could
reasonably be desired for forage purposes. The stalks are small, the
leaves abundant, and the hay is soft. Cattle and horses relish it very
much. A yield of 10 to 15 tons per acre can probably be secured
here under normal conditions. As the hay is softer and more easily

15

baled and handled than the sorghums, it will pay faruiers and .stot-k-
meii in central Texas, and in all parts of Texas east and south of this
section, to try to grow teosinte for forage.

MILLETS.

Several varieties of millet have been tested in the station garden
this year with satisfactory results in each case.

JAPANESE BARNYARD MILLET.

( I'anicum cnis-'/alii. ,

On March 1.") seed was sown in rather thin land on a hillside, wliich
had been broken deeply in October, 1898, and not i)lowed since. A good
stand was secured, and by April 18 the millet covered the ground with
vigorous growth. On May 11 weeds were abundant and threatened to
choke out the ndllet. .lune 3 the millet was cut, affording a splendid
yield. The new growth started up immediately, and July 1 a stcond
cutting was made — yield rather light. The millet began a third growth,
but the drought of July was too severe, and it did not make another
croi). The lirst crop grew to be 3 feet tall and seeded abundantly.
Seed sown in another plot April 10 gave quite as satisfactory results.
Tnder normal conditions this variety of millet will probably yield three
('roi)S a year.

SHAMA MILLET.

( I'anicuiii frumciiliiii'iim).

Seed was planted October 4, 1898, and germinated well, but the
winter freezes destroyed all the roots. Other plantings were made
March 17 and April 10, 1899. Splendid stands were secured, the
growth was vigorous, 18 to 24 inches tall, and the seed development
was very good. Several bales of hay were made from the plat, which
is equal to the best hay grown or imported into this section. Stock are
veiy fond of it and the yield is good. One cutting only was secured
before the drought set in. The second growth gradually dried up dur-
ing July and August, and by September 1 was to all appearances
dead.

PEARL MILLE'I'.

A good stand was secured from seed sown April 18. Some of the
stalks grew 7 feet tall with seed heads 8 to 10 inches long. There was
an abundance of fodder, which was cut, cured, and baled, making rich
soft hay that was relished by stock. It was a success here notwith-
standing the drought.

liROOM-CORN MILLET.

Planted in April. The yield was large and the seeds matured before
J Illy-

16

GERMAN MILLET.

This variety has been fully tested here during past j'ears and is justly]
regarded as of great value.

DWARF ESSEX RAPE.

Seed of this plant was sown April 18. A splendid stand was secured
The growth was vigorous and promised well, but before the conclu-
sion of the experiment the plants were destroyed by prairie dogs.

EXPERIMENTS WITH GRASSES.

' Seeds of a large number of varieties of grass were received from the |
Division during the year. Some were planted October 18, 1898, some
in January, 1899, and others early in February, 1899, but on account
of the very severe freezing weather that continued through the latter
month many of them either failed to germinate or were killed.

SMOOTH BROME.
(Bromiis inervtis.)

This grass has been tested in several of the Southern States and has
given good results. Three plantings, all of Eussian seed, were made |
October '4, 1898, November 15, 1899, and April 10, 1899, respectively. 1
The seeds first planted germinated, but the grass was killed down the "
following February. The results from the other plantings were only
partially satisfactory. Fair stands were secured. The growth was
rather strong and stood the dry hot weather in July and August
reasonably well, but by September 15 it showed the effects of the
drought badly. The roots, though still alive, did not revive after the
autumn rains began.

SIDE-OATS GRAMA.
(Bouteloua curiipendtila.)

This is a native grass and the few seeds saved in 1898 were planted
April 18, 1899. A good stand was secured, the grass did well, stood
the drought well, and the roots revived after the autumn rains began, |
although the grass is now not green, having cured on the ground as is
its habit when growing on the open range. This is one of the most
valuable of all the native grasses of central Texas.

BUFFALO GRASS.

(Bulbilis dacty hides. )

This grass is frequently mistaken by stockmen for grama grass. It
is a native of all this section and furnishes more rich stock feed than
any other variety of so-called grama known here. It can be grown
successfully from thu seed. Land well seeded down to it, along with
curly mesquite and needle grass, makes an ideal pasture. Plantings

17

were made October 4, 189S, January, 1899, and April 18, 1899. Those
of October 4 and January gave no results. Some seeds germinated,
but tbe young roots were killed by tbe following February freeze.
The planting of April 18 gave very satisfactory results. A fair stand
was secured and the grass stood the drought well. By October 15 the
grass had dried up and promised to make good winter pasturage.

OREGON RESCUE ORASS.

(Bromiis Hiiioloides.)

The few seeds of this variety received were planted May 4 and an
excellent stand was secured, although the growth was at no time vig-
orous. The seed heads began to form when the stalks were 0 to 8
inches tall. By September 1 the grass showed in a marked degree
the effect of the hot, dry winds of July and August, and by October 1
the foliage was all dead. The roots did not revive after the autumn
rains began, although many of them are still alive.

BERMUDA GRASS.
(Cynodon dactylon.)

Sowings were made October 4, 1898, January 8, 1899, March 16,
1899, and April 18, 1899. There were no results from the first two.
A few seeds of the March sowing germinated, but the plants soon dis
appeared entirely. A rather thin stand was secured from the April
sowing and the grass grew nicely, continuing green until August,
when it showed evidence of suffering on account of the hot, dry
winds. By October 1 it had disappeared. The roots did not revive
after the autumn rains began, although a few are still alive. This
plot is on a rather high point; soil thin and gravelly.

GIANT RYE-GRASS.

{Ehjmus condemaius.)

Seeds of this variety were sown March 15 and April 11. Oidy a few
germinated, and the grass was choked out by weeds.

CANADA RYE-GRASS.

{Ehjmus canadensis.)

This grass was not grown at the station, but an experiment made
with it the past season is of interest and demonstrates its value in this
section. A small plot of ground outside of the grass station contained
this grass, which in tlie early part of January, 1899, was green and
from 4 to 5 inches tall, while nearly all the other varieties of grass in the
same neighborhood were dried up and gave no indication of growth.
It was arranged with the farmer, who had a lease on the held, to turn
17014— No. 23 li

18

over that small spot for experimentation; it was cleared of brush
and the past year's growth of weeds, the surface broken with an iron-
tooth harrow, and the soil about the roots loosened. By March 1 there
was a splendid growth of grass that kept at least two weeks ahead of
any other variety. During June a heavy crop, well headed out and ripe,
was cut. A month later a second growth of this wild rye was ready to
be cut, and bundles that measured from 20 to 30 inches in length were
secured. Later a third crop was cut from the same plot, measuring 15
inches in length, that, when cut, was as green as in springtime, not-
withstanding the fact that since October 26 tliere have been several
freezes. During the long, hot, dry spell from early in July to late in
October the wild rye grass gave little evidence of injury from the
drought and was about the only grass in this section that continued
green. There is no better native hay grass in this section. Stockmen
and farmers in all parts of Texas are recommended to secure seeds of
wild rye and cultivate the grass for both hay and pasturage. A
l^astnre of 100 acres of it as it is now growing near the station would
furnish abundant, rich pasturage for from 30 to 40 head of cows, after
having produced two crops of hay during the year.

MEADOW FESCUE.

{Fesluca pratensis. )

Only one planting was made (November 15), affording a fair stand,
and the grass lived through the summer, but by September 1 was to
all appearances dead. The roots are still alive, but have made little or
no autumn growth.

GRAPEVINE MESQUITE.

(Panicum ohinsum.)

The sowings of this grass were made March 17 and April 10. Every
seed must have germinated, as splendid stands were secured. The
seeds were saved by me in 1898. The grass is native to this sec-
tion of Texas and grows in most of the counties of central Texas. It
seems to prefer low, moist ground, but frequently grows luxuriantly
on high ground. Both of the plots in the station garden this year
were on high land. It has long, creeping, jointed stems that root
wherever they touch the ground. The leafage is rather scanty and
harsh. The seed is produced abundantly and clings to the stems far
into the autumn. The native growth in this section is from 18 to 24
inches in height. Two bales of the cured hay were secured. Stock
seem to relish it. None of the grasses grown this year in the station
garden resisted the drought more successfully. One plot planted to it
was not cut this year and is now covered with the grass, dried up of
course, but the roots are alive. It is well worth cultivation for pasture
as well as hay.

19

BALING LEGUMES AND SORGHUMS.

One experiment in baling legumes and sorghums, made this year,
is well worth particular mention. The seasons in this section are
very favorable as a rule to the devolo])meut of a vigorous growth of
pea and bean vines, although they do not always mature seeds. These
vines well cured make rich forage, but as the leaves fall off badly,
they have little commercial value. To determine if these crops would
bear baling several varieties were secured and planted in the grass
garden. They made vigorous growths, and bore heavy crops of peas
and beans. When the fruits were nearly full grown, and before they
began to turn yellow, the vines were cut and carefully cured as hay.
Later a few bales were made of each. Eecent examination showed that
the hay was as sweet as when first baled. Since very large yields of
these legumes can be grown in all parts of Texas nearly every year,
the crops can be given a decided commercial value if carefully cured
and baled.

A similar experiment was tried with the sorghums and other coarse
forage crops, such as teosinte and pearl millet. When the stalks were
from 3 to 5 feet tall, aud while the seeds were " in the dough," several
bundles of each variety were cut, cured, and stacked, aud later baled,
to determine if they could be handled in that shape to advantage.
The bales have kept as sweet as those of the well recognized hay grasses
of this section, and, as in the case of the legumes, the experiment was
a decided success.

FAIR EXHIBITS.

In September last an exhibit was made of native and other grasses
and forage plants grown during the year. Several boxes of baled hay
were prepared for the exhibit at the Paris Exposition in 1900, and 53
bales of excellent hay, each representing a difterent variety, were made
up and turned over to the managers of the district fair held here in
October. The exhibit attracted a great deal of favorable attention
from farmers and stockmen of this section, who became much inter-
ested in the station work. Heretofore many of them have either known
nothing of such work or have paid but little attention to it. In the
future they will watch the experiments more closely. They are man-
ifesting especial interest in the work intended to improve the range
conditions by cultivation of the native pasture grasses.

SUMMARY.

In spite of the adverse conditions under which they have been con-
ducted, the experiments here have yielded results of great value. They
have demonstrated the availability of alfalfa (especially oasis alfalfa),
sulla, sainfoin, smooth brome, Canada rye grass, Terrell grass, and
others for use in permanent pastures and meadows; of the vetches,
cowi»eas, velvet bean, soy benn, teosinte, and a larger number of varie-
ties of the sorghums and millets for annual or temporary pastures, and

20

as sources of coarse forage, either fresh or cured; of saltbush for alkali

soils; of the gri^^ii'^s, Canada rye grass, grapevine mesquite, curly

mesquite, galleta, and needle grass for reseeding the worn out ranges.

They have shown the feasibility of range improvement by resting and

scarifying the land and by sowing hardy native and introduced grasses.

The farmers and ranchmen are beginning to understand that they can,

at comparatively small expense, greatly improve their ranges, and by

cultivation of the many excellent grasses and forage plants tested this

year in the station grounds can add very much to the productive

capacity of their ranches and farm pastures. They are beginning to

recognize the fact that there are very many native grasses and forage

plants that are well worth careful attention and that others of almost,

if not quite, equal value are being imported from the semiarid regions

of the Old World and can be successfully cultivated here. Many of

them propose, during 1900, to test on their owu places the dittereut

varieties of alfalfa, vetches, sulla, sainfoin, teosiute, saltbush, the

saccharine and nousaccharine sorghums, and the best native and

foreign grasses.

Respectfully,

11. L. Bentley,

Special Agent, in Charge of Experiments at Abilene, Tex.

Approved :

James Wilson,

Secretary of Agriculture.

AVashington, D. C, December 30, 1899.

O

CiKCVLAR No. 'Jl.— (Agros. 64.)

UNITED STATES DEPARTMENT OE AGRICULTURE,

DIVISION OF AGROSTOLOGY. ' '

[Grass and forage plant investigations.]
F. Lamson-Scribner, Agrostologwt. GAiRDFN

[Corn is everywhere the king of fodder crops, and cowpeas are to the South what
the clovers are to the North. One supplements the other, and combined as they
may be in the silo, they constitute a ration of high feeding value. The practical pos-
sibility of growing and harvesting the two crops together is set forth in the present
circular prepared at the request of the Secretary of Agriculture by Mr. W. Gettys, a
prominent and successful farmer and dairyman of Tennessee. His methods of
handling cowpeas and corn for silage and fodder will be of interest to every farmer
in the South who is seeking improved systems of farming and stock feeding. —
F. L.-S.]

COWPEAS AND CORN FOR SILAGE AND FODDER.

OBSERVATIONS ON CORN AND PEA STOVER.

In the South, at least in Tennessee and some of the other border
States, the silo has become a necessity to the dairyman and live-stock
breeder. It is the compensating hand of nature reaching- out to us
and makino- o-ood some of the natural deficiencies found in our South-
ern agriculture, and enabling us to compete successfully with the
West in making beef and butter. Nothing can till up the gap made
by a short summer crop, bridge over a fall drought, or draw reluc-
tant spring into the lap of winter so w^ell as good silage. When
the spring floods have drowned out the regular crops on the lowlands
or so delayed their growth that they will not mature in due season,
then the silo comes to the planter's relief and enables him to utilize his
crops, as he can in no other way, before they are caught and ruined
by the early frost. Or should the drought cut short his hay crop, he
still has the chance at some of the numerous catch crops, such as sor-
o-hum, millet, cowpeas, and corn. Any of these forage plants, if the
season is at all favorable, will advance far enough to make a fair crop
of silage. Low, wet, bottom lands that remain useless for anything
else till midsummer can profitably be used in this way. Even large

I

mill ponds, that are in the winter and earl}' spring- seasons applied to
grist and saw-mill power, may be dried off in the late spring and their
beds cultivated in time for a crop of corn silage. The soil of such
lands is especially rich in all the elements requisite for large yields of
such crops. To convert these catch crops that I have named into hay
is, in our uncertain climate, a very unsatisfactor}', and often a very
unprofitable, part of the farmer's work. It requires a succession of
hot, dry days to cure for safe-keeping such coarse provender, and even
when hay caps are resorted to, the crop is badly damaged if the
weather takes an unfavorable turn before it is put under shelter. But
all these ills may be cured, regardless of the weather, by the use of the
silo.

Many silage growers secure fairly successful yields by turning early
wheat stubble land and sowing cowpeas, and succeed in getting the crop
into the silo before frost. This is an uncertain dependence, for its
success hangs on having a good growing season all through. Yet it
rarely fails to pay the expenses of the effort, because if the season is
such as will not make mature silage or hay, the planter will have valu-
able late pastures for his stock and the land will be benefited by the
peas and be read}" for another crop of small grain without further
plowing. No other kind of stock food, green or dry, grain or stover,
will make milch cows respond at the pail and churn so readily as a late
growth of peas pastured off' in the fall at a time when everything else
seems to fail to arrest the natural tendency of the cow to decrease her
flow of milk. The dairy product, however, will be somewhat unfavor-
ably affected in its flavor by too much of this feed.

During the fifteen years that the writer has been using the silo he
has endeavored to ascertain the best kind of crop for silage, and in
doing so has grown nearly all the forage plants that were thought to be
adapted to this climate and soil, taking into consideration always the
quality as well as the quantity of the product, and the purposes for
which that product was to be used- viz. making winter food for a
breeding herd of dairy cows.

SORGHUM UNDESIRABLE FOR SILAGE.

The several kinds of sorghums have been tried under the different
methods of planting and cultivation, and while the production was
large, the quality of the silage was not good.

COWPEAS ALONE NOT SATISFACTORY.

Most of the varieties of cowpeas have also been tried and found,
when grown alone on rich soil, expensive to handle and injurious to
the flavor of a first-class dairy product, although among the best for
a large production of milk and butter and for feeding beef and stock

cattle. Probably no other kind of silage will i-arry dry cattle and
young stock through the winter in such fine appearance as will that
made from cowpeas.

CORN ALONE TOO EXHAUSTING TO THE I>AND.

Corn alone, as we grow it for the silage in the South, is, perhaps,
as nearly a perfect, all round silage crop as is possible to be had, but
it is exhausting to the soil. To remedy this objection the effort was
made to find some forage plant that would grow well with the corn
and, while adding quality and quantity to the yield, would add nitro-
gen to the soil. For this purpose tests were made with various
legumes and a suitable crop found in the cowpea which adds quantity
to the crop, and being a renovater of the soil and a conserver of its
fertility, counteracts or counterbalances the ^exhausting effects of the
corn on the land. Especially is the (piestion of soil renovation important
to the South now, since of late years it has become next to impossible
to secure a stand or a good crop of red clover, even on land that once
grew enormous yields of that most valuable forage plant. In this
agricultural emergency the cowpea has become a boon to a large sec-
tion of the South for both hay and silage, and the area in cultivation
has widened in Tennessee wonderfully within the last few years.

Some writers have doubted the advisability of growing cowpeas for
ensilage on account of its alleged inferior quality as food. But the
writer's experience leads him to believe that if the crop is grown
properly, harvested at the right stage of maturity, evenly mixed in
the silo with an equal portion of corn, and fed in connection with good
hay or other provender, the quality of the combined crop will produce
milk and butter of first-class quality, which is well known to be the
most severe test that can be applied to any cattle food. The question
of the quality of the silage is of much importance, since the hist five
years have witnessed a great increase in the number of silos built and
used in the South, especially in east Tennessee, and perhaps the
greater part of their contents is fed to dairy cattle, a class of stock
that demands the best of health-giving food, and whose dairy product
must be of unexceptional quality.

I

WHIP-POOR-WILL PEA THE BEST.

Of all the varieties of cowpeas the whip-poor-will has been found
the best for silage purposes, because, when planted with corn, it grows
rank enough; does not entangle the corn so nuich as the others, and
hence damages it less; is the most easily harvested, and in this climate
and soil yields more grain and ripens its vine more uniformly than
the ranker growers, such as the "Clav," ^'Unknown," "Wonderful,"'
and other runners. The ''Whip" pea, strictly speaking, is a bunch

pea and not a climl)er or runner, but with us, in favorable soils and
seasons, it does climb in many instances as high as 10 feet, and bears
fully matured pods in profusion to the top.

I

; .

-<5f-

FiG. 1.— Three stalks ot corn covered with a grcnvlh of whip-poor-will cowpeas,
AN IDKAL TENNESSEE CROP FOR SILAGE DESCRIBED.

My own silage crop of "Whip" peas and corn grown together the
past season was, to all appearances, an ideal one. A 20-acre field
of chocolate land, lying in, and on the slopes of the valley, was well

broken and prepared, and the corn planted the middle of May with-
out fertilizer. The corn, a laro-e Southern variety, was put in with
a one-horse drill, in rows -ii feet apart and from 9 to 16 inches apart
in the row. This width of row was given to admit plenty of sunshine
and air. After the first cultivation and during the first days of eTune,
when the corn was about 6 inches high, the peas were planted between
the corn and in the same row with the corn, with a hand planter, com-
monly used for replanting purposes. The delay in planting the peas
is deemed necessary to give the corn a good start ahead of the peas,
the latter being of more rapid growth. If the corn rows are straight
the peas may be put in more expeditiously with the one-horse drill, if
care is taken to keep the same close up to the corn row; the object
being to have both crops grow so as to be harvested together with the
corn harvester, and bound into bundles, without waste, ready for the
wagons. The season was favorable for both corn and peas, the stand
was good, and after three cultivations the crop was all that could l^e
desired. 1 have seen larger crops of both, when grown separately,
but never such a yield of grain and provender of both plants combined
and in such perfect condition for harvesting quickly and economically.
The corn stood up well, and the peas, confining themselves to the row,
grew over much of the field to the base of the corn tassel, fully grown
and matured pods being found on the vines 10 feet from the ground.
The accompanying cut (Fig. 1), showing a man of ordinary height
standing by the stalks, gives a fair idea of the growth of the crop.
The harvesting began September 8, when the grain of the corn was
beginning to glaze and when three-fourths of the pea pods were ripe,
using a corn harvester (with four mules attached) that cut and bound
the crop in bundles ready for the silage cutter. Few noxious weeds
or grasses could stand under such a growth, so that after a satisfac-
tory harvest and a few days of gleaning by the herd of cows
the field was ready, without further preparation, for the disk harrow
and wheat drill, and is now covered by what promises to be a very
good and very cheaply grown crop of wdieat. This land is ordinarily
good for a yield of from 40 to 50 bushels of corn and from 18 to 25
bushels of wheat ppr acre. From this year's crop a considerable por-
tion of the riper grain was pulled from the corn before cutting, for
fattening hogs, hence no estimates were made by which the weight of
the crop could be indicated, a fact I now regret. Doubtless compara-
tive failures of this silage crop will now and then occur, as with all
crops, but I am sure, from my own experience of several years, that
all good corn-growing seasons will bring paying yields of this combi-
nation, grown as above described, and it will meet more fully than
anything else of the kind the wants of the Southern silage grower.

6

SILOING AND FEEDING THE CROP.

The crop was put into the silos in the usual way. The field being
half a mile away, four two-horse wagons, with one hand to each wagon
and two, sometimes three hands, loading, were required to keep the
cutter going. The knives were set to cut half inch in length, the cut-
ter being preferred to the shredder, because it did its work more
cheaplj^ and quickly. The silos being side b}'^ side, the work of filling-
was all done at one setting of the cutter by simply shifting, when
required, the table chute at the top of the carrier over the pits, and
were both filled as nearh^ at the same time, as possible, on alternate
days, so as to allow the contents to heat and settle. In this way a
greater quantity can be put in the same space. Two men were kept
in the pits all the time, tramping the sides well. When full the con-
tents were covered with some of the chafl:y, less valuable part of the
same material, without putting on anj^ weights. They were well
tramped once a da}" for a week, after which thej were let alone till
time for using.

The feeding should begin at the top, using the windows in the sides
for removing the silage as the pits are emptied. This crop of silage,
as that of former years, has kept well, comes out of the pits with fine
color and flavor, and is much relished b}'^ the cows, about 40 pounds a
da}', in two feeds, being allotted to each cow. As silage of any kind,
of itself, is not considered a perfectly balanced ration, the cows receive
in addition, during the winter, 4 quarts of wheat bran, 2 quarts of
corn meal, and 1 pound of cotton-seed meal, night and morning,
together with one feed of corn-and-pea stover. Good clover hay would
be preferred to the roughage named, if it could be had. In feeding a
breeding or dairy herd during the winter, silage, however good or
plentiful, should always be supplemented by some kind of mill feed
and well-cured hay or other stover. When the increase of the herd
is the greatest consideration, the cows forward in calf and the breeding
bulls should be given silage sparingly.

THE QUALITY OF THE SILAGE ON TRIAL.

The quality of this kind of silage and its influence on the character
of the dairy product are now receiving a fair trial. The cows are
apparently in perfect health, and the cream from the herd averages
33^ per cent of butter fat. A portion of this goes to a near-b}'' city,
where it is scientifically manufactured into various kinds of food for
the table. The balance is made into butter and sent to customers in
a Southern city, some of whom have been using my product for
eighteen years. Thus far no complaint from either source has been
heard and none is anticipated.

THK GROWING CROP DIVERTED TO OTHER USES.

Should there })e, as there was in the writer's case, more of the
standing crop of corn and peas grown together than is needed for
silage, as already described, it can be very i)rofitably used as a dry
provender. Allow the crop to stand till the proper stage of maturity
is reached, which is shown by the l)rown husk of the corn and the
yellow leaves and ripe pods of the pea vines. Then cut with a corn
harvester and shock as with ordinary corn, putting about sixteen hills
to the shock. To guard against molding at the center, go through the
field two or three days ahead of the harvester and make small shocks,
which will give a dry nucleus or center to the main shock and prevent
all danger of damage. See that all shocks stand up well, and that
they are well tied together at the top with binder twine. Select dry,
cool weather in November or December, the sooner the better, pro-
vided the cornstalks are dried out enough to keep in the mow, then
haul to the barn and run the crop through the "corn shredder and
busker."" In one operation the peas will be knocked from the pods
and carried, with the small proportion of corn that is shelled in the
process, into sacks to be taken to the mill and ground for stock food;
the ears of corn husked and dumped into the wagons, to be carried to
the crib; and the corn fodder and pea vines torn to pieces and con-
veyed up through the blowpipe into the hayloft, to be used as winter
stover. Such stover is now being fed, in connection with silage, with
as good results, and with as little waste as would be with the ordinary
hay, such as is made and sold in this part of the country. The work
mules and dry cattle get no other roughage and do well.

ALWAYS PLANT PEAS WITH THE CORN.

Considering this necessity of growing corn m some manner for live-
stock food and as a rotation crop, the planter should aim to secure as
large yields of provender to the acre as possible with a minimum
amount of exhaustion to the soil. Therefore none of our uplands in
the South, and particularly within the bounds of what is known as the
"Piedmont region," where the cowpea seems to have its home, should
be put into corn without peas being planted with it. This advice will
hold good, whether the crop be intended for silage or dry provender.
Peas add to the quality, as well as to the quantity, of the yield and
restore to the soil its fertility. To harvest and put away this com-
bined growth quicklv and economically requires improved agricultural
implements that are especially adapted to the purpose, such as the
corn harvester, silage cutter, and corn shredder and harvester.
These every well-to-do stock grower and farmer should own or be able
to hire.

8

COWPEA HAY FROM STUBBLE LAND,

It ma}^ be of interest and some benefit to Southern stockmen and
planters to recount here the experience of the writer in growing a catch
crop of pea hay from stubble land. The wheat was taken off as early
as possible after harvest, and in June the land was turned and well
harrowed and rolled, the season being favorable for such work.
Whip-poor-will peas, a bushel to the acre, were sown with an ordinary
wheat drill. The seed being defective, the stand was not good, but
the growth was rapid and promised a rich crop of grain. As harvest
time approached the thought occurred to the writer that, while there
might be a medium-sized crop of good ha}^ secured from the land, cut
and put awa}^ in the usual manner, there should be more of the grain
saved and utilized than is possible when the crop is handled as ordi-
nary hay, knowing from past observation that nuich, if not most, of
the grain is scattered and lost by the time the ha}^ reaches the feed
manger. If peas are to be our dependence in the future, in the
absence of clover, we must discover the best method of reaping the
greatest benefit from their production. The crop was permitted to
stand before cutting a few da3"s longer than it would have been for
making hay, so as to give a greater proportion of the pods time to
ripen. It was then cut with a mower and let lie in the hot sun thirty-
six hours, when it was thrown into light windrows with a haja-ake, the
raking being done after sunset, sometimes by moonlight, when the
plant was tough, to avoid shattering the leaves and grain. After
remaining in the windrows twent3^-four hours, it was thrown into small
cocks and left exposed to the sun and air for thirty -six hours longer,
when it was loaded on wagons and taken to the barn. No thrasher
suitable for the work being available, the crop was run through the
silage cutter. The cutter had a shredder attachment, but this, requir-
ing too much speed, was removed and the cutter knives used as with
silage. To save the thrashed peas separately from the stover was
the next problem encountered. For this purpose a section 8 feet long
was cut out of the bottom of the carrier and a suitable wire screen
tacked in place of the solid bottom, which allowed the peas to drop
through into a wagon bed on the ground underneath, while the pea
stover was carried on up 36 feet into the barn loft, where it was well
scattered and mixed with other drj^^ feed.

The peas were damaged but little in the process, and after being
run through a hand pea cleaner were spread out on a granary floor to
dry a few days before sacking. The hay was not so good as it would
have been if cut earlier, but this was more than counterbalanced by the
money value of the peas secured, which was greater than the value of
the wheat crop just removed. The land was much benefited by the
growth of the peas, and required only the doubling of the disk harrow

9

to prepare an excellent .seed bed for the wheat that was to immediately
follow. Inasmuch as wheat was to follow wheat, the greater part of
the labor and expense of this catch crop of peas was but the prepara-
tion of the land for the succeeding- crop of wheat, leaving a very
small expense account chargeable to the crop of peas. Of course, the
harvesting and housing of the crop in this manner required favorable
weather, but not more so than would have been required to put away
successfully the crop in the ordinary way. In such work as this the
planter is more or less dependent on the weather for his success, this
being one of the incidents of his calling.

If it were possible again to grow such tine crops of clover on the
lands in Tennessee, and elsewhere in the South, as we secured years ago,

Fig. 2.—X tiekl of corn and whip-poor-will cowpeas.

the questions discussed and experiences given in this paper would be
of less importance than they are now. But as this can not be done,
and as the cowpea has come to be regarded as the substitute for
clover, for the purpose of fertilizing and renovating the soil, and for
furnishing food for live stock, the planters and stock growers in these
States are compelled, by self-interest, to study the question in all of
its bearinp-s, in order to ascertain the best manner of cultivating and
saving this crop. And the question is made more important by the
fact that this substitute is so varied in its useful purposes and capable
of serving the planter in so many different kinds of combinations with
his other farm crops. The writer's advocacy of the cowpea for the
needs of the South and his reasons for growing and using it in so
17716 2

10

many ways are not based on a study of any scientific analysis giving
its comparative feeding value for stock. He has been content in con-
sulting only the good effect the plant has produced for him, as shown
in the marked improvement of the soil, the continued good health of
his cattle, and their increase, and the quality and quantity of their
product at the pail and churn. All of which observations and expe-
riences lead to the belief that no single grain of the plant was e^^er
cast intelligently into Southern soil without profit to the sower.

Respectfully.

^\. Gett'^

Approved:

James Wilson,

Secretary of Agriculture.

Washington, D. C, Febrimrij 26, 1900.

O

)

Circular No. 35.— (Agros. 08.)

United States Department of Agriculture,

DIVISION OF AGROSTOLOGY.

[Grass and Forage Plant Investigations.]

U. S. Department of Agriculture,

Division of Agrostology,
Washington, D. C, April 9, 1900.

Sir : I have the honor to transmit herewith and recommend for publication as

a circular of this Division the information now in hand relative to the Turkestan

alfalfa, imported and distributed by you in 1898 and 1899.

Respectfully,

F. Lamson-Scribner,

Hon. James Wilson, Agrostologist.

Secretary of Agriculture.

TURKESTAX ALFALFA.

{Medicago sativa var. Turkestanica. )

This variety of alfalfa v^as secured from Russian Turkestan by-
Prof. N. E. Hansen while engaged as special agent of this Depart-
ment in 1898. The localities from which the seed was collected rep-
resent a wide range of climatic conditions, as follows : Inventory No.
469, from the Turkestan Agricultural Society, Turkestan ; 079, from
Bokhara; 991 and 1101, from Tashkend; 999, from Uralsk Agricul-
tural School; 1150 and 1151, from Djarkent; 1109, from Merke; 1295,
from Samarkand — all in Turkestan, and 1159, from Kopal, Siberia.

THE CLIMATE OF TURKESTAN.^

Russian Turkestan is about two-fifths as large as the United States,
with a population of 3,000,000, and yet for the most part it is a vast
tract of country consisting of either steppes or marshlands. A con-
siderable portion of the country is not adapted to settled agriculture ;
so that the inhabitants are mainly employed in cattle raising. The
climate of Turkestan, far removed as it is from the ocean and closed
in on nearly every side by wide stretches of dry land, is not unlike
that of many of our interior States. The summers are very hot,
cloudless, dry and long, lasting from five to five and one-half months ;
the autumns and springs are mild and rainy, and the winters usually

' The Industries of Russia, 3 ; 444, 1893.

2

open, with but little snow. Severe frosts occur sometimes, but these
very rarely last long. In many places the evaporation exceeds by
many times the quantitj' of moisture that has fallen in the course of
the year. From May to September very little rain falls, often only
1 . 1 inches, and in many parts of the steppe regions, which are re-
moved from the mountainous districts, there is absolutel}' no rain
during the entire summer season. Drj^ winds from the north and
northeast prevail during the summer, with a temperature of about
loO F. (40 degrees Celsius), which generally dry up the whole Tur-
kestan basin until vegetation can not exist without irrigation. The
extent of watered lands in Turkestan is, in comparison with its whole
area, by no means great, not more than two and one-half per cent,
which is altogether insufficient for the subsistence of its population.
This deficiency is partially redeemed by the growing of wheat, barley,
and millet, which depend on the winter moisture in the soil and a suf-
ficient fall of rain during the spring.

MERITS OF TURKESTAN ALFALFA OVER THE EUROPEAN VARIETIES.

The following extracts from an article by Prince V. I. Massalski,
of the Russian Department of Agriculture, show its great value to
the arid regions of Russia over the European varieties : ^

Lucern clover (Medicago sativa var. Turkestanica) is the chief forage in use
throughout Central Asia, and to the settled population of Turkestan is of the
highest importance, since diiring the summer it forms the chief, and in winter,
prepared in the shape of hay, the only fodder for cattle. It is of all the greater
importance because, within the regions populated by settled inhabitants, there
are no meadows. Soft herbs and other grasses that grow up in the early spring
in certain parts of the steppes are quickly dried up by the hot rays of the sun
and give place to coarse, prickly stubble, or, in any case, to less nutritive grasses
that are in general tmfitted for sheep, camels, or stepjie cattle, and still less fitted
for horses or the cattle of those who are settled in the oases, and are thus closely
confined to the foreland or rivers, in most cases far removed from the steppes.

Massalski describes the native methods of cultivation and irriga-
tion, and continues —

The native lucern would seem to be a cattle fodder that can not be replaced
in countries as dry and hot as Turkestan and the Transcaspian Province. Par-
allel experiments that have been made in the Merv oases, in the Transcaspian
Province, in growing native and French lucern, under widely different conditions
of water supply have shown that the native lucern, particularly where there is
a lack of water, is vastly superior to the French in the crop it yields, and that it
is able to grow satisfactorily with a minimum supply of water, a supply so small
that the European lucern would perish with drought. It possesses a very large
root system, and its leaves are covered with thick down. This, in conjunction
with a deeply channeled leaf, enables the plant on the one hand to imbibe the
moisture from the deeper layers of the soil ; and on the other hand, to exhale it
in very small quantity.

The Industries of Russia, 3; 459, 1893.

DISTRIBUTION OF THE SEED.
(See Table I. )

In 1898 and 1899, 1,111 packages of the seed of Turkestan alfalfa,
of from 2i- to 20 pounds each, were distributed by the Secretary of
Agriculture through the Section of Seed and Plant Introduction of
the Division of Botany and through this Division. This distribution
included experimenters in forty-seven States and Territories. Un-
fortunately, owing to the primitive conditions prevailing in the
region in which the seed was collected, it was not clean when pur-
chased, and had to be carefully inspected and freed from all weed
seeds before being distributed. This occasioned such a delay in
sending the seed out that for most localities it was too late for spring
sowing. Many experimenters planted the seed immediately upon
receiving it, and thus subjected the young and tender plants to the
dry, hot summer weather, which in many cases killed the alfalfa, as
the roots had not a sufficiently strong hold on the ground to enable
them to withstand the drought. Some have been holding the seed
for a favorable season in which to sow, and have not yet planted it,
but expect to do so the present season.

As this variety of alfalfa was introduced especially for the semiarid
regions, the largest consignments of seed were sent to the following
States: California, 48 packages; Colorado, 77; Idaho, 21; Kansas,
97; Montana, 38; Nebraska, 49; New Mexico, 30; North Dakota,
26; Oklahoma, 37; Oregon, 27; South Dakota, 20; Texas, 318;
Washington, 28; and Wyoming, 21, The number of experimenters
to whom seed was sent in each of the States and Territories and the
number and character of the reports received are given in Table I.

REPORTS RECEIVED FROM VOLUNTEER EXPERIMENTERS.

A total of 466 reports have been received, and of this number 237
are satisfactory and 229 unsatisfactory. (See Table I.) A report
in this experiment has been termed "unsatisfactory" unless the
Turkestan alfalfa has proved in some quality to be superior to the
common alfalfa. Included in the 229 unsatisfactory reports are 76
which state that owing to the seed being received too late in the
season, or because of the unfavorable weather or conditions, it has
not yet been planted.

Table I. — Distribution of seed of Turkestan alfalfa and number and character

of reports received from experimenters.

States.

Alabama 1

Arizona

Arkansas

California.-

Colorado

Connecticut

District of Columbia

Florida

Georgia

Idaho

Illinois

Indiana

Indian Territory —

Iowa -

Kansas

Kentucky

Louisiana

Maine -

Maryland

Massachusetts

Michigan

Minnesota

Mississippi

Missom'i.--

Montana

Nebraska

Nevada

New Jersey

New Mexico

New York.

North Carolina

North Dakota

Oklahoma —

Ohio

Oregon .-.

Pennsylvania

Rhode Island

8outh Carolina

South Dakota

Tennessee

Texas

Utah

Virginia

AVashington

West Virginia

Wisconsin

Wyoming

Total

Number
of experi-
menters.

Number 1
of reports

satis-
factory.

Number
of reports
unsatis-
factory.

Total.

11
13

1
1

1

2

3

11

2

4

6

48

8

11

19

77

11

18

29

1

10
5
6

21

3
2

1

4

3

2

1

.3

7

9

1

3

4

12

5

2

7

4
17

3

3

6

97

39

22

61

9

1

2

3

7

2

4

6

1

1

1

2

1

2

3

■ 5
3

10

4

2

6

11

.»

3

5

16

3

5

8

30

22

i

29

49

14

12

26

10

4

1

5

1

30

3

5

8

10

2

1

3

4

1

4

5

26

8

4

12

37

8

8

16

15

1

3

4

27

6

5

11

8

4

1

5

1

4

20

1

1

1
3

1

9

12

7

3

2

5

318

43

60

103

18

4

4 ,

8

16

1

1

28

5

6

11

6

1

1

2

9

1

2

3

21

8

6

14

1,101

237

229

466

i

The following testimonials from the experimenters will in some
degree indicate the value of Turkestan alfalfa in the different States.

ARKANSAS.

Mr. C. T. Burns, Lynn, Lawrence County :

Alfalfa is practically unknown here. The seed received was sown under favor-
able conditions and came np well, but about the time it began to show green
over the field, we had a long drought which killed out many of the plants. I
think if I can get a stand it will be of great value, and on rich land the produc-
tiveness will exceed any of the grasses. I should have sown earlier.

Mr. H.N. Jameson, Calhoun, Columbia County :

Alfalfa has never been sown before in my immediate neighborhood. I selected
a thoroughly drained, rich, loamy, well-prepared soil, which received no fertili-
zers. The seed came up well, but the young plants were washed over by a water-
spout on the 12th of May and ruined. I believe alfalfa will grow here.

CALIFORNIA.

Mr. J. J. Dean, MoDeta, Los Angeles Connty :

I prefer the Turkestan alfalfa to our common alfalfa, as it was up eighteen
days earlier; is not so woody; has more and larger leaves and grows more rap-
idly. It is hardy, keeps green, and is not as easily injured by frosts as the other
varieties. For dairies it is the best. My cows prefer it. The Turkestan alfalfa
that I sowed June 5, 1898, I have cut nine times this year (1899), the growth
being about 18 inches each time. The crop was irrigated with deep-well, sul-
phur water.

Mr. M. Hartley, Dehesa, San Diego County :

I received the seed too late to sow in 1898, which was a dry year with us, and
so also was 1899. The creek did not run through my place, so the bottom land
was wet only on the surface. However, I sowed the seed in March after a good
rain. It came up beautifully and grew very well for about two months, when
the vegetation dried up on the highlands and caused the qiiail and rabbits to
flock in from all sides, and they destroyed almost every growing thing, includ-
ing the Turkestan alfalfa.

Mr. W. H. Old, Chualar, Monterey County :

There is no particular variety of alfalfa grown in this part of the country, and
very little attention is given to any kind. I secured a beautifiil stand of the
Turkestan alfalfa, 90 per cent or more of the seed germinating. It seems to have
the power of withstanding drought well, as last year was unusually dry. It is
hardy and extra early.

Mr. J. R. Robinson, Lancaster, Los Angeles County :

The Turkestan alfalfa is doing well for the first year. It was cut three times.
I shall save the seed for further use, as it is difficult to determine its value from
a single year's experiment. It was not affected by a temperature of from 100 to
110 degrees.

COLORADO.

Mr. F. E. Ewing, Hugo, Lincoln County :

I have experimented with Turkestan alfalfa for two seasons. The past season
was unusually dry, yet this variety of alfalfa withstood the drought well. The
growth was only about 12 inches high, but the plants were strong and vigorous.
The crop was not irrigated at any time, and we had but one good rain during
the season.

Mr. Adolph Froelich, Aroya, Cheyenne County :

A sandy loam, broken last season, was plowed about 6 inches deep this year.
The seed was sown broadcast. May 9, 1899, when the ground was very dry, and
harrowed in. We had less than the usual amount of rain ; in fact, the driest
season in eighteen years. The Turkestan alfalfa began to bloom on August 1,
and I cut it with a scythe and left it on the ground. It soon grew up again 18
inches high. Part of it was irrigated when sown, and again on the 8tli of
August. That irrigated has a fine stand, but that not irrigated was a total fail-
ure, only twenty-five plants coming in a plot 150 by 100 feet. This variety held
its own better after it came up than our common alfalfa would have done ; that
is, where a good stand was first obtained by irrigation.

6

Mr. D. Y. Hamill, Seguro, Huerfano County:

The first year, 1898, the Turkestan alfalfa proved to be a quick, strong grower.
I cut it when it was 2 feet 2 inches high. It afterwards made a growth of 2 feet
before the winter set in. The second season, 1899, was so cold and dry that the
Turkestan alfalfa made only one-half a crop. After cutting, it leafed out and
remained gi-een till winter. Native grasses— oats, barley, and wheat— did not
make one-half stand. Most of the common alfalfa sovm in 1898 died in July or
earlier, but the Turkestan alfalfa, which was sown on the level ground, grew
well. One-sixth of that grown on the slopes died.

Mr. William Raymann, Deertrail, Arapahoe County :
I am in a very poor position this year to give you information, as we had such
a very bad season— no rains until too late to be of any use. I plowed the Tur-
kestan alfalfa, never thinking it would grow ; but must confess I was surprised,
as some of it came up. The crop suffered from an unusually long drought.

Mr. M. J. Weyand, Sedgwick, Sedgwick County:
The Turkestan alfalfa is of a darker-green color, has larger leaves, and grows
taller. It will be a famoas plant in two more years. It is early, grows rapidly,
and would bring a good crop of hay even in a short season. The plant is perfect,
but our water is scarce at times.

GEORGIA.

Mr. E. J. Hartman, Orr, Gilmer County :

The soil was well prepared and the seed sown broadcast about the end of May
and the field lightly harrowed. It gave two good cuttings of excellent quality.
This alfalfa has done wonderfully well on our soil. I have a fine stand now and
hope to get excellent crops next year.

INDIANA.

Miss Kate A. Drake, Elkhart, Elkhart County :

A poor, sandy soil was plowed and harrowed. The seed was sown in the fall
of 1898, after a rain. It came up nicely and grew well. We had a very severe
winter, and some of it froze out, but a large part of it came on again in the
spring. This grew well, some of it reaching 18 inches in height. The severe
drought caused it to dry up, but did not kill it, as a part of it has come up this
fall (1899) and looks very promising. I hope it will be a success. I would like
to send you a report next summer after I see if it dries up again. I believe it
will be adapted to our soil and prove a valuable acquisition to the drought-
stricken iiarts of the country.

Mr. M. F. Eastman, North Vernon, Jennings County :
A clay loam was well prepared and the seed sown broadcast in March, 1899.
The plat, 10 by 200 feet, has been cut three times this season, which has been
very dry. The yields were very heavy at each cutting, estimated at 2 tons per
acre. I believe it to be a paying crop if a catch can be secured, but, owing to its
slow growth, it is difficult to get a stand.

KANSAS.

Mr. A. Y. Bentley, Wallace, Wallace County :

A good sandy loam that had been in cultivation for several years was used.
The seed was sown May 12, 1898, and we had rain on the 13th and 14th. I mowed
it when about 7 inches high and left it on the ground. The grasshoppers kept
it eaten down to the crown, but it came up again and is a good stand now. I
sowed several patches of common alfalfa in the spring and the grasshoppers took
them all just as they did the Turkestan, which is the only patch that survived.

Mr. Ben. Browu, Natoma, Osborne County :

A rich, light, prairie soil was plowed and harrowed twice. The seed was sown
about May 27, 1898. A good stand was up by June :i. It grew well until checked
by drought in July and August, but this did not seriously injure it. Since the
heavy rains of September 8 to 14, it has grown from 8 to 10 inches high, and is
now green. The field looks promising for future crops. I have 80 acres of com-
mon alfalfa, which I cut four times each season, getting from 4 to 6 tons to the
acre. It does not give a full crop until the third or fourth year. If the Turkes-
tan beats this it will do well.

Mrs. Sarah J. Gilmore, Norcatur, Decatur County :
The Turkestan alfalfa was sown on high upland and received no irrigation.
In the summer of 1898 we were sixty-five days without rain, and the alfalfa
lived through it, although the plants were small. I think it much better than
the common alfalfa to withstand the drought.

Mr. J. G. Gray, Ottawa, Franklin County:

The Turkestan alfalfa is earlier, hardier, more productive, and grows more
rapidly than the common alfalfa.

Mr. H. C. Hollowell, Barnes, Washington County :
A dry, black loam was plowed early and harrowed. The seed was sown broad-
cast about May 20, 1898, and harrowed in. The growth was short, but it stood
the dry weather well. I think it better than our common alfalfa.

Mr. C. H. Jackson, Kidderville, Hodgeman County :
The seed was sown May 12, 1898. The alfalfa grew about 6 inches high by the
first frost. The dry weather tested it severely, but it would green up now if we
had a little rain. This is the only variety that seems to offer a chance for a crop
in this part of the State.

Mr. Don. F. Lyman, St. Francis, Cheyenne County :
The Turkestan alfalfa differs from the common variety in having smaller
stems and larger roots. It is very hardy, early, quite productive, and grows
until very late in the season. The growth of this alfalfa last summer was sur-
prising, as the season was very dry and hot. I believe it is going to be a success
here.

Mr. Mads. Olson, Mullinville, Kiowa County:

The soil was second sod in good condition. The seed was drilled in the last day
of May, 1898. It came iip well and was nicely in bloom when the grasshoppers
came and destroyed it.

Messrs. Wm. B. Sutton and Son, Russell, Russell County :
A i)iece of new black loam was broken and finely cut with a disk. The seed
was sown broadcast about May 1. The season was very dry, but I obtained a
fine stand, 8 inches high. It withstood the drought well, and is now growing
quite nicely. This has been a poor season to test it. A four-year old field of
common alfalfa made only one small cutting in June. There was no rain in
June, July, or August.

Mr. Theo. Swartz, Salina, Salina County :

The Turkestan alfalfa seems to have finer stems and smaller leaves, which
are consideral)ly darker green in color than the common alfalfa. The seed came
up evenly, made a good start, and looks fine at present, comparing very favor-
ably with the common alfalfa sown at the same time. As I only sowed in May,
1899, it is too sfK)n for me to judge of its special qualities.

MASSACHUSETTS.

Messrs. Jas. H. Gregory & Son, Marblehead, Essex County:

We planted a small sample of Tm-kestan alfalfa from your Department, and
noted that it made quite 50 per cent more plant and stood the winter better than
the common alfalfa planted by its side.

MINNESOTA.

Mr. C. F. Miller, Faribault, Rice County :

I have experimented two seasons with the Turkestan alfalfa and find it very
productive. I think from its behavior that it will succeed in the Northwest.
The common alfalfa is ruicertain here.

MONTANA.

Messrs. W. W. Gamble & Son, Burton, Choteau County :

A gravelly land on bench land was well plowed and harrowed. The seed was
so\^Ti broadcast, May 19, 1898, with a hand-seeder. No cultivation except irri-
gation was given the crop. It was in fall bloom in July and ripened in August.
The quality of the product was good. This seems to be a better plant in its
growth than the ordinary alfalfa. The winter of 1898-99 seemed quite trying to
alfalfa, with a tendency to winter-kill. In the spring of 1899 our ordinary al-
falfa came up bunchy in appearance, and although it recovered itself later in
the season, the growth was irregular. The Turkestan alfalfa, however, grew
even and uniform throughout the season under precisely the same conditions.
It is probably a better variety than the ordinary sort sold here.

Mr. W. H. Heideman, Kalispell, Flathead County :

The soil was very light, sandy, and open, varying to a black, sandy loam. The
land was plowed thoroughly, harrowed, then rolled and harrowed again. The
seed was sown broadcast and the land then smoothed with a harrow and rolled.
It came up May 18, and was in full bloom by Jiily 28. The crop was cut with a
mower along with the weeds. So far as the experiment has gone it is very prom-
ising. I think the alfalfa has done remarkably well for the first year. The sea-
son was very hot and dry, with no rain after June, but the drought did not seem
to affect it. It was about fourteen inches high on July 38, 1898. The winter of
1898-99 was very severe, but none of it was winter-killed. The fore part of the
season of 1899 was very dry, but a great deal of rain has fallen since August 1st.
The plat of Turkestan still continues to do well. It was pastured by hogs till
May 28 and then let go for seed. It made a great gi'owth, but did not fill well,
the cold weather probably preventing the production of seed. After cutting,
the alfalfa grew up again rapidly. I think it is going to be more valuable than
the common variety. It seems to do well on alkali spots. I would advise all to
sow it as soon as they can secure seed.

Mr. Len. Lewis, Lewis, Meagher County:

Dry, limestone, bench land was iised. A good, strong growth was secured. It
seems to be hardier than the ordinary alfalfa. I think it is highly satisfactory.

Mr. Emory Vine, Miles City, Custer County :

The Turkestan alfalfa wintered well and is coming on this spring as if it
meant business. At the present time it has a decided advantage over our com-
mon French alfalfa. It shows green several rods away, while the common
variety sown last season has not started to green yet.

NEBRASKA.

Mr. J. A. Anderson, Harrison, Sioux County:

A sandy soil in good condition was thoroughly prepared and irrigated. The
seed was sown with oats, May 15, 1898. The quality of the product was good,
and I think it will make a valuable forage plant in this section. It did better
than the ordinary alfalfa.

Mr. W. Benjamin, Banksville, Red Willow County:
From the short experience I have had with the Turkestan alfalfa, I believe it
will succeed in our dry climate when it once gets acclimated. My experiment
was handicapped by jack-rabbits, cotton-tails, and grasshoppers destroying the
plants.

Mr. T. R. Butler, Beaver City, Furnas County :

The outlook is encouraging. I can hardly tell yet whether it has any special
qualities over the common alfalfa, but I think it has. It came up well, pro-
ducing a good stand when the common alfalfa was killed out by drought.

Mr. J. W. Williams, Weeping Water, Cass County:
A thin, black loam was plowed and harrowed. The seed was sown broadcast.
May 2.5, 1898. The alfalfa made a very heavy growth ; withstood our short
drought very well and is still in excellent condition. It did much better than
our common alfalfa for the first year.

NEVADA.

Mr. H. C. Campbell, Skelton, Elko County :

The Turkestan alfalfa does not seem to require as much water as the common
alfalfa.

NEW MEXICO.

Mr. J. A. Gishwiller, Roswell, Chaves County:

A good, clay loam was plowed three times and harrowed fine. The seed was
sown broadcast, July 13, 1898, and covered with a harrow. I secured a good
stand. The gi'owth was strong and vigorous, somewhat stronger than that made
by our common alfalfa sown on adjoining land. Another season will be neces-
sary to compare it well with the common alfalfa.

NORTH DAKOTA.

Mr. Joe Horsky, Anamoose, Mc Henry County :

No other varieties of alfalfa are grown in this vicinity. The Turkestan alfalfa
seems to be hardy both as to drought and cold so far. .

Mr. W. Loughland, Nesson, Williams County:

I am plea.sed to report very favorably on the Turkestan alfalfa received from
your Department. I had seven acres sown to the common alfalfa, which was all
killed off in the winter of 1898-99, while the Turkestan variety, which was sown
May, 1898, was not affected in the least. This winter, 1899-1900, we are having
no snow, and I am looking forward to its withstanding the frosts. If it does
that, it is just the plant that is wanted here.

Mr. CM. Lovett, Pingree, Stutsman County :

The seed of the Turkestan alfalfa gave a poor stand at first, but thickened up
remarkably well after the fall rains. It seems to be easier to get a catch with
his variety than with the common alfalfa. I think it is promising.

10

Mr. C. F. Miller, Harmon, Morton County:

Alfalfa has never been tried in this neighborhood to my knowledge. The Tur-
kestan alfalfa was sown May 2, 1899. The plants came up within ten days and
grew vigorously till injured by frost on June 1. During two weeks of warm
weather they started again from the roots and grew until December, when
growth was checked by frost.

OKLAHOMA.

Mr. C. L. Boyd, Redmoon, Roger Mills County :

A black, sandy, bottom soil was broken early in February, rebroken at the
time of seeding, and harrowed down fine. The seed was sown broadcast. May
5, 1898, and the land was then harrowed and rolled. Tlie (luality of the Tur-
kestan alfalfa was the finest I have ever seen. I sowed along side of it a plot of
the common alfalfa, and there was a marked difference between them, the Tur-
kestan variety growing much ranker. I think the Turkestan alfalfa the more
valuable variety.

Mr. W. A. Row^an, Gallienas, Beaver County :

A sandy soil was plowed, well pulverized, and put in fine condition. The seed
was sown broadcast, May 9, 1898, and harrowed in. The field was irrigated about
May 35, and again on June 20. I secured a good stand, but the alfalfa does not
mature sufficiently the first year to make a good crop. It was sown by the side
of common alfalfa and given the same care. The Turkestan variety seems to be
stronger and hardier. The value of a good stand would be about $30 per acre.

OREGON.

Mr. F. T. Byrd, Pilot Rock, Umatilla County :

I found the Turkestan alfalfa a superior drought-resisting plant. My experi-
ment was seriously interrupted by the continued freezing and thawing last
winter. The plants grew splendidly through the longest and hottest summer
ever known in Oregon.

Mr. A. Wintermier, Silvies, Harney County :

A decayed lava and clay soil was used. The Turkestan alfalfa came up very
nicely with an even stand. When about three inches high a severe frost cut it
down. It grew up again, however, but was a second time cut down by frost,
and still it grew to be one foot high. Apparently it will not stand frost as well
as the French alfalfa, but seems to be better adapted to dry soils. Another year
will determine what it will do in this part of the country.

PENNSYLVANIA.

Mr. J. O. Brown, Pittsburg, Allegheny County:

Alfalfa of any kind is a new plant here. I have tried Wisconsin, Nebraska,
and Turkestan seed, and I think the last-mentioned germinates best and is the
most hardy. The few plants that were not destroyed by grasshoppers in the
summer of 1898 started to grow earlier in the spring of 1899 than the American
alfalfa, and, I think, stood the extremely cold winter better. The hot, dry
summer seemed to have about the same effect on both kinds, but I think the
Turkestan variety recovered quicker in the fall. I have gathered sufficient seed
from the sowing of 1899 to double the amount of ground.

11

SOUTH DAKOTA.

Mr. H. C. Clifford, Casey, Ziebach County:

In May, 1898, you placed in my charge Turkestan alfalfa seed. I planted it
May 16 on four different jilats.

Plat No. i.— A black, sandy loam was used. I gave it plenty of water, and
the alfalfa came up in four days and grew well. On July 12, when H feet high,
it was cut, a plat 40 l)y 1 00 feet, yielding 1 ,200 i)onnds. I cut it again September
20, when about the same height as before, and it yielded 1,500 pounds. I think
it away ahead of the American alfalfa, as a two-year old field of alfalfa, such
as we commonly grow in South Dakota, did not yield any better, if as well.

Plat No. .?. — A gravelly and sandy loam was used. The alfalfa came up
nicely, although I did not water it ; growing to a height of from 4 to 6 inches,
where it remained and kept green, although the season was very dry and hot.

Plat No 3.— A rich, black loam was used. The plants grew 2 or 3 inches tall
and then died out.

Plat No. If. — A white, gumbo land was used. The plants came up but did not
amount to anything.

Plats Nos. 3 and 4 were, sown with oats. The oats grew well, and my opinion
is that alfalfa will do better alone, as the oats seem to smother it. Nos. 2, 3,
and 4 got only the moisture from the natural rainfall, which was merely
sufficient to start it. As to whether it will endure the winter or not, I can not
say as yet.

Mr. W. H. H. Phillips, Brookings, Brookings County:
A black loam, with a clay subsoil, was well harrowed and made fine. The
seed was sown broadcast about May 20, 1898. In March, 1899, it was covered
with barnyard manure. On May 17, 1899, it froze somewhat, which gave the
weeds a start. I mowed it early in Jiily. Since then it has made a good growth.
The first year it made but little growth, but now (October, 1899) it is very
promising.

Hon. H. C. Warner, Forestburg, Sanborn County:
A black, sandy loam was plowed deep in the spring and well prepared. The
seed was sown broadcast April 24, 1898, and harrowed in. The stand was per
feet, and notwithstanding the drought of summer, and the cold, open winter,
without snow, the plants came through in perfect condition. The Turkestan
alfalfa was not cut this season, judging that it is better to let* it become well
established first. Every root of common alfalfa growing by the side of it was
killed.

TEXAS.

Mr. Severin Ball, Lytle, Atascosa County :

A black, sandy land was used. I planted the seed the latter part of October,
and would have sown earlier, biit the ground was too dry. The alfalfa seemed
to grow very slowly during the winter ; apparently the weather was too cold for
it to grow rapidly. As it is only a few months since it was sown, I can not say
much about it yet. It seems, however, to be vigorous and long rooted at the
present time.

Mr. J. W. Cartwright, Amarillo, Potter County :
The seed was sown the latter part of September, 1899. The alfalfa came up

nicely, and now looks well, with four inches of snow on the ground. I am quite

proud of this, and will report later on.

12

Mr. T. A. Coleman, San Antonio, Bexar County :
The Turkestan alfalfa grows luxuriantly here and makes fine hay.

Mr. Wm. Fancher, Taopi, Fisher County :

A dark, sandy loam, with clay subsoil and containing some alkali, was used.
The seed was sown October 15; 1899, in very dry weather, but it came up well
and produced a good stand by November 1 . Thus far the Turkestan alfalfa has
done well in this section. I think it will stand the drought well.

Mr. B. H. A. Groth, Selma, Bexar County :

A loam with plenty of lime in the soil was used. The soil in this section does
not need drainage, from the fact that for the most part there is nothing to drain,
and when there is, the subsoil, a sandy loam, will take it up. The seed was
sown April 8, 1899. Last season we suffered from a severe drought. From the
1st of July to the 1st of November there was no rain, and from the middle of
July to the last of August the weather was extremely hot. I can as yet answer
only to the hardinesss of the Turkestan alfalfa. It is green now (January 2,
1900), and we have had some ice on water in vessels. I believe the Turkestan
alfalfa will withstand our dry climate and furnish us a good forage plant.

Mr. J. Q. Hanna, Olga, Nolan County :

The 1898 sowing on what we call black land was a complete failure. The
1899 sowing is showing well on the sandy land. That it is now standing leads
me to be very hopeful. I do not know whether the seed will mature here with
out irrigation ; the common alfalfa will not. I have faith that the Turkestan
alfalfa will prove a success here. If so, its value can not be overestimated. It
will be worth millions to the stockmen of Western Texas.

Mr. George Hoeffert, Schulenburg, Fayette County :
A heavy, black soil was used. The seed was planted about the middle of
March, at the same time as some California alfalfa. The Turkestan variety
came up sooner and produced a better stand on the same soil. It seems to be
earlier, hardier, more drought-resisting, and more valuable than the California
alfalfa.

Mr. S. R. Jeffery, Graham, Young County :

A sandy loam was used. I planted the Turkestan alfalfa in drills in the fall
of 1898. It came up perfectly, but a very severe winter followed and killed out
the greater part of it. In the spring it came rather late and very thin. I did
not cut the plants nor allow them to be grazed. They grew to be two and one-half
feet high and seeded. We had the driest and hottest summer for years, but the
stems of the Turkestan alfalfa kept green through it all, and when our fall rains
came numerous shoots came up all around the old plant. It is now, December
28, from eight to twelve inches high, looking very green and fresh. I think it
will be a great plant for this section of the State.

Mr. Geo. Loeloff, Hellemans, Bexar County :

A dry, black soil mixed with clay was used. The seed was sovra, October 1,
1899, so that the plants could be compared with Colorado seed. The leaves are
similar but somewhat broader in the Turkestan variety. On examination of the
seedlings of both varieties thirty days after sowing, when three inches high, the
Turkestan alfalfa appeared more robust, more leafy, stronger and deeper rooted
than the Colorado alfalfa. Further comparison at a later date, December 33,
showed these differences to be even more pronounced,

J

13

TEXAS.

Mr. W. S. Marshall, Channing, Hartley County :

I believe that the alfalfa from Turkestan is a more valuable plant for this
section than the common variety. It withstands the drought much better, seems
to be hardier, starts very early in the spring, and is very productive.

Mr. P. P. McDermatt, St. Joe, Montague County :

A black, waxy, somewhat sandy land was used. The seed was planted in the
spring of 1899. I have not grown this variety long enough to give a full report
on it. I secured a good stand and it withstood the extreme drought of last
summer. I expect to get seed from it the coming season.

Mr. H. B. McCurry, Fate, Rockwall County :
The Turkestan alfalfa seems to be a splendid drought-resister.

Mr. R. T. Shelton, Anna, Collin County:

A well-drained, black, waxy soil was used. The seed was sown April 15, 1899.
The drought began this summer in July and lasted until the middle of October.
The native variety was dried up, while the Turkestan remained green.

Mrs. H. P. Thomas, Ector, Fannin County :

The seed was sown by the side of the native variety. The Turkestan alfalfa
made the stronger growth, being almost twice as large. It stood the summer
drought very well.

Mr. J. W. Tillerson, Roseland, Collin County :

A black, waxy soil was used. The Turkestan alfalfa is the only variety we
have tried. It seems to be early, hardy, and productive, and will be valuable
for hay and pasture.

Mr. Arnott West, Brownwood, Brown Count}^ :

A rich, deep, black, slightly ashy soil was used. The quality of the product
was good, and altogether throughout the season would produce from eight to
ten tons per acre. This variety stands the dry weather splendidly in this part
of the country if it has a deep, well-drained soil. Alfalfa should be grown here
exclusively. I am feeding two hundred steers on it and they are getting fat.

UTAH.

Mr. James Lof thouse, Paradise, Cache County :

The seed was sown broadcast, May 28, 1898, on a clay loam. The alfalfa
bloomed July 25, but was destroyed by stock before ripening seed. It stands
drought better than the kind of alfalfa commonly grown here. It grew eighteen
inches high, and outgrew the common alfalfa under similar conditions.

Mr. Samuel Roskelly, Logan, Cache County :

A dry, sage-bush upland that had been cul^^ivated in wheat for three years,
was well plowed and han-owed. The seed was sown broadcast May 9, 1898, and
harrowed in lightly. I consider this one of the best and most valuable fodders
for dry farms, which are not irrigated, in this western country. I have raised
common lucern for twenty years, but for drought resistance I think this is
better. My experience is that seed should not be raised from first year's growth,
as cutting before maturity causes it to send the roots down to find moisture.

14

WASHINGTON.

Mr. N. S. Dickinson, Clover, Okanogan County:

The soil was black loam with a clay subsoil. The seed was sown May 15, 1898.
and irrigated the first season. The summer of 1899 I let it go to seed, but lost it
all on accoiint of heavy rains. I think it is a good variety for this climate. It
seems to be earlier than the kind commonly sown here and makes a finer quality
of hay.

Mrs. M. A. Kuelin, Valley, Stevens County :

A sandy soil was fertilized and irrigated, and the seed planted in drills about
June 10, so as to avoid frosts while the plants were young and tender. The field
was irrigated about every eight days. I have a very good stand of alfalfa, and
am satisfied this is a good section for growing it. There is no alfalfa grown in
this vicinity that I know of, and few people seem to know of its forage value.
My Turkestan alfalfa was not affected by drought or cold. I have great faith
in it as a food for all kinds of stock.

WEST VIEGINIA.

Mr. Ira C. Snider, Avon, Doddridge County :

The land is a loose, well-drained, sandy loam. The seed was sown April 3,
1899. I believe the Turkestan alfalfa will prove quite successful in this country
for its hardiness, and its deep-rooting qualities will enable it to withstand the
drought well.

WYOMING.

Mr. Edwin M. Howard, Sarver, Bighorn County :

I am growing the common variety and the Turkestan alfalfa, and am satisfied
the latter is the more siiccessful. I can see a difference in the bloom, and the
hay is of a finer quality. It starts two weeks sooner than the common alfalfa,
grows taller, and will produce more to the acre.

Mr. J. M. Rowsier, Farrall, Crook County:

The Turkestan alfalfa looks very much like the common variety, except that
the stems are more slender and taller and grow finer. The experiment has
proven that it is especially able to resist drought.

THE RESULTS OF EXPERIMENTS WITH TURKESTAN ALFALFA BY THE

STATE EXPERIMENT STATIONS.

Seed of Turkestan alfalfa was sent to Experiment Stations in the
following States :

Alabama, Arizona, California, Colorado, Connecticut (Storrs Sta-
tion), Florida, Illinois, Indiana, Iowa, Kansas, Kentuckj', Louisiana,
Maryland, Massachusetts, Michigan, Minnesota, Missouri, Montana,
Nebraska, Nevada, New Jersey, New Mexico, New York (Cornell
and Geneva Stations), Ohio, Oklahoma, Oregon, Pennsylvania,
Rhode Island, South Dakota, Texas, Utah, Washington, and
Wyoming.

Reports have been received from the Experiment Stations as
follows :

I

15

ALABAMA.

Cdiichrake E.vpvriinent Station, Uniontoirn.

The Turkestan alfalfa was planted in March, 1899, and cut for hay three times
during the season. It stood the dry fall and winter freezes fairly well, but not
so well as our common Medicago sativa. The plat is now (March 2H, 1900) in a
flourishing condition with plants about 8 inches high, which ^\^ll be ready to
cut for hay in about fifteen days.— H. Benton, Director.

CALIFORNIA.

State Experiment Station, Berkeley.

Seed of Turkestan alfalfa sent to the California Experiment Station by the
Section of Seed and Plant Introduction, Department of Agriculture, was for-
warded by me to the substations and also planted at the Central Station. This,
was in March, 1898. The season was dry. to very dry, throughout California,
and even in favorable seasons alfalfa should be planted in the autumn, A\'ith the
first rains, or not later than January.

At the Central Station the seed was sown on good soil and had more rainfall
than at other places. The Turkestan variety showed no especial peculiarity iu
the season's growth, excepting that it was more compact than other alfalfa.

The seed sent to the Amador County substation (elevation 2,000 feet) in the
Sierra foothills gi-ew excellently without irrigation, but was badly eaten down
by rabbits.

At Pomona in the Chino Valley, Los Angeles County, the Turkestan alfalfa
was tested on alkali soil and on arid, sandy soil. In both of these localities it
proved superior in drought resistance to the common variety.

The best test of this alfalfa was at Paso Robles substation, San Luis Obispo
County. The total rainfall of season was about 5 inches, distance to water
nearly 200 feet, and the soil shallow and very poor, on deep hardpan. The only
two species of forage plants which endured these conditions and remained green
during the summer without irrigation were Turkestan alfalfa and Australian
saltbush {Atriplex semibaccata) . The latter made growth and furnished more
fodder, but the former promises well and deserves much more extended trial
under similar conditions.— C. H. Shinn, Inspector of Stations.

COLORADO.

state Experiment Station, Fort Collins.

Of the Turkestan alfalfa furnished by the Department of Agriculture last sea-
son most was distributed by Professor Cooke of the Agricultural section. Some
was tried at the substation at Rocky Ford and at Cheyenne Wells. A good part
of the seed was sown on the elevated ridge between the Arkansas and Platte
rivers known as the Divide. The results have been rather negative, that is,
the plant has not been especially better than our alfalfa, which is the common
plant grown in Colorado under irrigation. It was a poor stand at Cheyenne
Wells which is not irrigated and did not do so well on the Divide, which is also
without irrigation.— L. G. Carpenter, Director and Irrigation Engineer.

CONNECTICUT.

Storrs Experiment Station, Storrs.

The Turkestan alfalfa received from the Section of Seed and Plant Introduc-
tion was sown April 21, 1898. The soil is a medium heavy loam, with good
natural drainage. The land was plowed, harrowed, and raked before seeding.
For several years previously the plat of ground had been fertilized, mainly with
mineral fertilizers. Method of Seeding: Oats were first sown at the rate of
about 2i bushels per acre (designed as a protective crop), the alfalfa seed was
sown on a fine seed beed, immediately after sowing the oats, at the rate of about
50 pounds per acre. The oats and alfalfa both germinated well and made a
good growth for two or three weeks. We had a severe drought, lasting from
the early part of May till after the middle of June, and during this period the
alfalfa made a very slow growth. The oats were cut ofl: rather high June 15,
and again July 18. The weeds by this time had become quite thick, and owing
to the weak condition of the alfalfa ])lants. most of them died out during the
latter part of the summer.— C. S. Phelps, Vice- Director.

16

MASSACHUSETTS.

Hatch Experiment Station, Amherst.

The Turkestan alfalfa was sown side by side with alfalfa seed from a number
of different sources, as follows : Arizona, Colorado, California, Utah, and Kansas.
The seed from all sources was sown in rows in well-drained, medium loam.
The plants were hand cultivated and hand weeded to give them a good start.
The Turkestan variety proved among the least hardy. The plants were seriously
affected by a parasitic growth upon the leaves, which appeared to be a rust
which seriously weakened them, and at the end of the first winter nearly all
were dead. The seed which came from Kansas and Utah gave the best results. —
Wm. F. Brooks, Agriculturist.

FLORIDA.

State Experiment Station, Lake City.

A portion of the Turkestan alfalfa seed was sown upon new ground, prepared
fur the purpose, but, either through unadaptation to our climate or the exceed-
ingly dry weather, it failed to germinate, so we did not succeed in securing any-
thing like a satisfactory stand. My impression from so slight a trial is that it
is not adapted to the dry soils of Florida, except where irrigation is possible. —
H. E. Stockbridge, Agriculturist.

ILLINOIS.

State Experiment Station, Urbana.

The Turkestan alfalfa did not jjrove upon our grounds more hardy than other
varieties, and seemed unable to withstand the conditions. — A. D. Shamel,
Assistaiit in Agronomy.

INDIANA.

State Experiment Station, Lafayette.

A small area, perhaps one-fortieth of an acre, was seeded to Turkestan alfalfa,
with seed obtained from the Department of Agriculture, Washington, D. C, in
1898. I recall that my former assistant thought this made a better growth than
the home-grown alfalfa seed. By some oversight, however, he left no written
record concerning it. My present assistant tells me that it did not stand the
winter of 1898-99 as well as the home-growai seed, and that it did not conse-
quently make as large a growth in 1899. It also seemed to be more affected
with some leaf disease. This, however, is common to all alfalfa that we have
grown here, the second season. I think it due rather to the dry conditions and
the hardpan layer from one and one-half to three feet below the surface of the
soil. — W. C. Latta, Agriculturist.

IOWA.

State Experiment Station, Ames. \

In the spring of 1898 one-quarter of an acre was seeded with Turkestan alfalfa,
received from the Department of Agriculture. A similar area was seeded with
the common alfalfa, 20 pounds of seed per acre being used in each case. Both
crops flourished during the first season, and by fall the ground was completely
occupied with plants. In January and February of 1899 the weather was un-
usually severe, so much so that clover throughout the State was killed. Not a
solitary plant of the common alfalfa lived through the winter, and only about
one dozen survived on the Turkestan plot, a stand which we did not consider
sufficient to justify its in allowing it to remain.

This experience would lead us to believe that the Turkestan is slightly more
hardy than the common variety, but not sufficiently so to withstand the severe
conditions which sometime prevail in this State.— James Atkinson, Assistant
in Agriculture.

KENTUCKY.

state Experiment Station, Lexington.

The Samarkand alfalfa seed was planted (three-eighths of a pound to one-eight-
ieth of an acre) May 21, 1898. It produced only a fair stand, which, August 21,
was six inches high— an even growth, but rather spindling. All through this
season it looked weak and unpromising. In 1899 it appeared much stronger.
Under date of May 23, 1899, I find the following among my notes: "This plot,
started from seed furnished by the Department of Agriculture, is now in excel-

I

17

lent condition, averaging about 10 inches high and densely covenng the ground
with a fine green growth, among which some purple heads can be seen. I can
see no difference between it and other alfalfa." August 23, 1899, the plat was
again noted as in excellent condition and the growth about 18 inches high, with
some bloom. It was cut, and material was secured for chemical analysis. We
have had in the experiment farm for many years a plat of alfalfa started from
seed obtained from an easteni seedsman. It looks and behaves much like Samar-
kand alfalfa. Both withstand drought well ; both hold their own against weeds
very well. The older plat has furnished two cuttings during some specially
favorable seasons, but at other times would j)roduce but a scant cutting. It has
held its own, however, better than the red clover jjlanted near it. It must be
added that the soil is rather sterile for this region, and otherwise is not well
adapted to the plant. We are starting new i)lats in better land this spring and
may have a more favorable report to make later. — H. G arm an. Botanist.

LOUISIANA.

State Experiment Station, Audubon Park.

The Turkestan alfalfa received by us and distributed by the Section of Seed
and Plant Introduction was diily received and planted. It was rather late in the
season to plant alfalfa, which is usually done here in early October, and from
this cause and, perhaps, from defect in the seed, we did not obtain as good a stand
as we usually do. However, we have still growing over an acre of this crop from
the seed sent us. There are growing near it several acres of alfalfa from our
home-grown seed, and the difference in growth is very noticeable. Our home-
grown gives us from six to eight cuttings per year ; the Turkestan has given only
three to four. During the winter just passed we have not cut the Turkestan at
all ; in fact, it remained almost stationary from last October up to the i^resent
time. It has now the appearance of ])utting on a new growth. In the mean-
while, we have cut the home-grown alfalfa twice during the winter. The plants
seem to be identical, but I fear the Turkestan variety is not yet acclimated to
oiir environments. We intend this year to let it go to seed in order that we may
experiment afresh with the seed grown from that sent us. — W. C. Stubbs, Director.

MARYLAND.

State Experiment Station, College Park.

W« made every effort to get a stand of the Turkestan alfalfa, but failed to do
so, partially on account of the failure of the seed to germinate and partially on
accoiint of the very dry weather which prevailed throughout the growing season
of 1898, so that, on the whole, our test with the seed was a failure. — H. J. Pat-
terson, Director.

MISSOURI.

State Experiment Station, Columbia.

The alfalfa seed from Turkestan, furnished by the Department of Agriculture,
was sown April 26, 1898, on carefully prepared seed bed and lightly covered with
a hand rake and rolled. Only a small proportion of the seed germinated, and
many of the plants died apparently from the excessive heat of July and August.
The stand was further reduced by the alternate freezing and thawing of the fol-
lowing spring, leaving so small a number of the plants that the plats were aban-
doned. This alfalfa does not seem to be quite so hardy as the common alfalfa.

H. J. Waters, Director.

NEBRASKA.

state Experiment Station, Lincoln.

On May 24, 1898, a one-tenth acre plat of ground was seeded to Turkestan
alfalfa by drilling with a press drill in rows 6 inches apart. The seed began to
come up on May ;^0, giving a good stand. It was cut three or four times during
the summer to keep do^vn the weeds, but no croj) of alfalfa. In 1899 it was cut
on June 15, yielding at the rate of one and seven-hundredths (l.UT) tons of hay per
acre. This was considerably less than the yields obtained from ordinary alfalfa
in nearby fields cut at the same time. The other fields were older, but one sowni
in 1897 gave in 1898 a yield of two and one-half tons of hay to the acre and in 1899
two and one-quarter tons. Our experience with the Turkestan alfalfa would
indicate that it does not produce as heavy yields as does the ordinary variety.

18

It seems, however, to have a somewhat smaller leaf and thinner stem, which
would indicate a somewhat higher feeding value. During the winter of 1898-99,
which was a very severe one in this region, the Turkestan alfalfa suffered abso-
lutely no loss from winter-killing, while the ordinary alfalfa killed out to some
extent. — T. L. Lyons, Acting Director.

NEW JERSEY.

state Experiment Station, Neic Brunswick.

The Turkestan alfalfa sent to us by the Department of Agriculture was given
a trial on a plat of ground containing one-fifth of an acre. The soil was a me-
dium clay loam, well drained. The seed was sown May 14, 1898, at the rate of
thirty pounds per acre, and a good stand was secured. The yield the first season

(1898) was at the rate of 7.5 tons per acre from two cuttings. The second season

(1899) four cuttings yielded at the rate of 9.42 tons per acre. Four-fifths of an
acre of the variety which is commonly grown was sown at the same time as the
Turkestan and yielded the first season (1898) at the rate of 8 tons per acre from
two cuttings, and the second season at the rate of 20.2 tons per acre from four
cuttings, and was much more vigorous than the Turkestan variety. The results
indicate that the common variety is superior to the Turkestan for this locality. —
C. B. Lane, Assistant in Dairy Husbandry.

NEW MEXICO.

State Experiment Station, Mesilla Park.

On the 13th of April, 1898, six pounds of Turkestan alfalfa seed were sown
broadcast on one-fourth of an acre. The seed was covered with an ordinary
tooth-harrow, and on the same day it was irrigated to produce germination.
By the 4th of May the alfalfa was growing quite well, but the ground had baked
and cracked considerably ; and in order to help the weaker plants break through
the crust, another irrigation was given on this date. Nothwithstanding the fact
that the seed had been sown late in the season, the germination was good and
uniform. The first crop was cut June 13, when the alfalfa was about a foot
high but was not yet in bloom. On the 17th, four days after cutting, it was
irrigated to start the new growth. On July 21 the second crop was cut, but
owing to the lack of water to irrigate with and the drought, the alfalfa made a
short crop. During the rest of the season the alfalfa made a very short growth,
as the drought continued and there was no water for irrigation. The Tiirkestan
alfalfa, under similar conditions, will grow as well and yield as much as our
common alfalfa {Medicago sativa) in this region. There seems to be no material
difference between the two varieties in their drought-resisting qualities. — Fabian
Garcia, Assistant AgricuUurist.

NEW YORK.

Cornell Experiment Station, Ithaca.

The Turkestan alfalfa seed, which was received by the Cornell University Ex-
periment Station, was planted on a gravelly loam soil in May, 1898. The soil
was one especially subject to effect of droughts, and no fertilizer has been ap-
plied for the past five years. The alfalfa seed germinated quickly, and the plants
made good growth. Three cuttings were made in 1898. During the winter of
1898-99 a portion of the alfalfa was killed by the ice which formed over it. Seed
was sown upon the patches killed out, and a good growth was secured during
the season of 1899. We cut the alfalfa four or five times during the season.
From what I have seen of it, I should say that for our soil it is not as valuable
as our common alfalfa or lucern. — L. A. Clinton, Assistant Agriculturist..

OHIO.

State Expe7'iment Station, Wooster.

"We sowed the Turkestan alfalfa seed April 29 on a piece of land previously
well prepared. The soil would be designated as clay loam, with a subsoil
definitely clay. This alfalfa made a fair stand, but by the first of July had
nearly all died out or disappeared. This has been the common experience with
alfalfa here. It does not seem to be adapted to the soil, I think, for the reason
that it is rather too tenacious in its make up. It has been our greatest difficulty
to secure a stand of alfalfa, but when we get it started it does well and holds
well in the ground. — J. Fremont Hickman, Agricidturist.

I

}

t

I

19

OKLAHOMA.

State Experiment Station, Stillwatcv.

The Turkestan alfalfa seed received from the Department of Agriculture was
sown alongside of some Kansas-grown seed. Last season was a poor one for
the alfalfa yields, though the stand was not seriou.sly affected. No comparative
yields were obtained, and at this time no appreciable difference exists between
the stand and vigor of the plants on the two plats, both starting off vigorously
and giving jiromise of good yields this season. — John Fields, Director.

RHODE ISLAND.

state Expevivient Station, Kingston.

The Turkestan alfalfa, of which the seed was sent here in 1898, was tested on
the station grounds. It germinated well and came through the winter of 1898
and 1899 in excellent condition, showing itself as hardy as the other five lots of
alfalfa, the seed of which came from different States and was also being tested.
In yield it was a fair average with the others. A part was sown on ridged and
a part on level land, to see under which method of cultivation it would winter
best. So little difference was noticeable as to make no practical difference in
favor of either method. — J. A. Tillinghast, Assistant, Field Experiments.

SOUTH CAROLINA.

state Experiment Station, Clemson.

I received the Tvirkestan alfalfa too late for spring planting in 1898. Planted
plants on alluvial river bottom in July, August, and September following. The
July and September plantings vegetated well, the August planting poorly. All
plants disappeared during the late fall and winter, probably on account of the
very wet season, as alfalfa sown near by failed in the same way. No experi-
ment was made on upland. — J. S. Newman, Agricnlturisf.

SOUTH DAKOTA.

State Experiment Station, Brookings.

I have just made a careful examination of the Turkestan alfalfa sown in the
spring of 1898, and find that it has stood the winter well. All the plants seem
to be alive and are begining to make a strong, vigorous growth.

The winter of 1898-99 was a very severe one on all kinds of meadows, and nearly
all clover in this locality was completely killed out. The Turkestan alfalfa,
however, was not injured. The past winter was very mild and open and not so
trying upon meadows, but there was no snow to protect the alfalfa ; it has come
through this, its second winter, in fine shape and is now well established, and is
certainly the most promising variety of alfalfa yet tried in this locality. —
E. C. Chilcott, Vice Director and Agricidturist.

WYOMING.

State Experiment Station, Laramie.

Following is a brief report of our results with Turkestan alfalfa from seed
furnished by the Department :

Sown in spring of 189.'). — Not cut that season.

In 1896 the plat was cut only once (September 21), and the yield was small.

In 1897 it was allowed to go to seed, but was planted so thickly that the seed
failed to ripen here, as the seasons are very short.

In 1898, cut August 3; yield, 6,446 pounds per acre. Cut a second time,
September 15; yield, 943 pounds per acre. Total yield, 7,389 pounds per acre,
cured hay.

In 1899, cut July 20; yield, 4,717 potmdsper acre. Cut a second time Septem-
ber 18 ; yield, 8, 14o pounds per acre. Total yield, 7, 862 poimds per acre, cured hay.

Average yield last 2 years, Turkestan alfalfa: First cutting, 5,581 pounds;
second cutting, 2,044 pounds; total for season, 7,625 pounds cured hay.

Average yield last 2 years, common variety: First cutting, 4,620 pounds;
second cutting, 1,710 pounds; total for season, 6,330 i)Ounds cured hay.

The winter of 1898-99 was unusually severe. The plat of Txu-kestan alfalfa
was not affected, while the common variety suffered vdnter killing to a consid-
erable extent. — B. C. Buffum, Vice Director and Agricidturist.

20

SUMMARY.

The results thus far reported are at first glance apparently quite
contradictory, particularly with respect to the tests made at the
State Experiment Stations. If these reports are analyzed closely,
however, it will be seen that the results are what might be expected
from our knowledge of the Turkestan alfalfa and the natural con-
ditions under which the variety has been developed. The reports
from the region west of the Mississippi River and north of Kansas
and CaUfornia indicate that this variety is hardier and more pro-
ductive than that commonly grown in this region. It seems to
endure drought better, ife not so easily affected by freezing, and gives
better results on strongly alkaline soils. In the East, however,
where there is a heavy rainfall and where heavy soils predominate,
this variety seems to be little, if any, superior to the French or
Chilean varieties ; in fact, it seems certain that, in some localities at
least, it is less valuable. In the South so few tests have been made
that no definite conclusions can be drawn, the reports from some
sections being favorable to the Turkestan alfalfa, while those from
others indicate that the commonly grown varieties are the most val-
uable. In the extreme Southwest the results are as yet quite contra- j.
dictory, and further experimentation is needed. ■I

The seed of Turkestan alfalfa will germinate much quicker and
the plants start into growth earlier under the same conditions than
common alfalfa. The plants are more leafy, grow more rapidly, and
have a stronger, more vigorous root system. Another advantage
which the Turkestan variety has is that the stems are more slender
and less woody, the plants making a more nutritious hay of finer
quahty. That it will withstand drought under the same conditions
better than ordinary alfalfa seems certain from the reports of the
experimenters. In the West and Northwest, at least, it seems to be
more productive, both with and without irrigation.

There is every indication that the introduction of this variety will
result in a very marked extension of the area devoted to the cultiva-
tion of alfalfa, and this, too, in sections where the forage problem is
one of paramount importance, and where heretofore climatic condi-
tions have largely prohibited the growing of leguminous forage crops.

P. Beveridge Kennedy, Ph. D.,

Expert and Agent in Charge of Experimental Work.

Circular No. 26.— (Agros. 71.)

United States Department of Agriculture,

DIVISION OF AGROSTOLOGY.

[Grass and Forage Plant Investigations,]
F. LAMS0N-SCRIBNP:R, Agrostologist.

RESCUE (iKASS.

Bronnis iniioloides (Willd. ) HBK.

DESCRIPTION.

An annual, or sometimes perennial grass, growing three to four feet
high under favorable conditions and sending up several stems from
the same base. The pani-
cle is usually large and
spreading, bearing much
flattened spikelets about
one inch long and one-
quarter of an inch wide,
composed of seven to ten
florets overlapping each
other. The flowering
glumes are rather coarse
in texture, strongly
nerved, and usually bear-
ing a short i>oint or awn
rarely exceeding an eighth
of an inch in length. The
"seed," in the commercial
sense of the term, repre-
sented in fig. 1, b, consists
of the flowering glume
with the inclosed palet and
grain.

Ther(? are several forms
in cultivation , varying
somewhat in the size of the
panicle and the length of
its branches, also in the
length of the awn on the
flowering glume. T he
form illustrated in the fig-
ure is the one to which the

name Bromus .schrader-i has been especially applied,
from the typical form only in having the branches somewhat longer
and more drooping.

Fi(i. 1.— 7Jro)iiu.s M/M'i/niWcs (Wilkl.) HBK. Rksciik-
GRASs: o, outer or oniply ^lunle^<; h, side view of one
of tlie florets; c, palea.

It differs

HISTORY.

Rescue grass was known to botanists in Europe in the latter part
of the eighteenth century. In 1806 it was described and figured by
Willdenow from specimens grown in the botanical garden at Berlin
from seed sent from Carolina, where it had apparently been intro-
duced. The grass is a native of South America, where it is quite
widely distributed. It also appears to be native in Central America
and Mexico and possibly southern Texas, but apparently became
distributed in our Southern States through the agency of the Euro-
pean settlers. The first attempt to introduce it into cultivation to any
extent, so far as we can learn from published records, was made by
B. V. Iverson, of Columbia, S. C, in 1853 and 1854. He wrote some
glowing accounts of it and was quite successful in introducing it in
various parts of the South at the rate of $5 per peck. He also gave
it the name, "rescue grass." It was also called in his honor "Iver-
son's grass."

About this time it was also introduced into Australia, where it was
very popular for some time under the name of California prairie
grass, this name having arisen apparently from the erroneous idea
that it was a native of California. It is still used quite extensively
in Australia under various names, as Australian oats, Australian
brome, and prairie grass, and is highly regarded by writers on Aus-
tralian grasses.

It was also introduced into France with very extravagant state-
ments, made mostly by people hoping to derive profit from its sale.
It was called Schrader's brome-grass, a translation of one of its bot-
anical names, Bromus schraderi. This name is also more or less
prevalent in this country. It has also been called arctic grass.

CULTIVATION AND USE.

This grass is adapted to cultivation in the Gulf States and has
been tried with success as far north as North Carolina. Its value is
in its use for winter pasture and hay. It has been tried at several of
the experiment stations in the South and spoken of very favorably. In
eastern Texas it is rather common as a volunteer crop and is spoken
of very highly by some of the farmers. It grows best on a rich,
loamy soil, and will do well in somewhat shady locations. On light,
poor soil it produces bat a scanty growth, and for pasture in such
soil it is inferior to rj^e. To secure the most satisfactory results, the
land should be well plowed and harrowed, and the seed drilled or
sown broadcast and harrowed in, using 30 to 40 pounds of seed per
acre.

The seed is rather expensive at present. It is sold by most of the
larger dealers and is quoted at from 25 to 30 cents per pound, or 10

to 25 dollars per hundred-weight. The seed weighs about 16 pounds
per bushel. After having seeded a small area, the grower may find
it profitable to grow his own seed rather than pay 25 or 30 cents per
pound for it.

The seed should be sown in late August or early September, so as
to be ready to germinate as soon as the first fall rains come. In case
of a dry autumn the crop will be late, but under favorable conditions
a heavy stand will be produced, furnishing excellent pasture from
December to April or May ; or if it is desired for hay, one or some-
times two crops may be secured.

The grass is naturally an annual, producing its seed and then
dying, but if prevented from seeding by continuous cutting or pas-
turing, it will survive several years and produce well; but as the
grass dries up during the summer, the use of the land during that
period is practically lost. Results giving the most general satisfac-
tion in growing this grass may be secured by pasturing it until
spring and then letting it reseed itself. After it has matured its
seed, the land may be plowed and sown, preferably to cowpeas or
Japan clover, which should be harvested in time to allow the rescue
grass to start again with the first autumnal rains. Excellent volun-
teer crops may be secured in this way for several years.

FEEDING VALUE.

Oats, rye, barley, and hairy vetch are the other principal annual
plants grown with more or less success for winter forage in the region
to which rescue grass is adapted ; so that it must be compared with
these in determining its relative value. Professor Tracy, writing
of this grass in Bulletin No. 20, of the Mississippi Station, says:
"Sown with equal care it will give a better winter pasture than will
either oats or rye, and in the spring can be plowed in with equal
advantage as a fertilizer." For hay, Professor Phares says it is
equal to a good stand of oats.

Professor Brunk, of the Texas Agricultural College, in writing of
it, says : "It makes more forage in February and March than any
other grass tried. When cut for hay in April it produced about two
tons per acre." He places it second in a list of grasses for winter
pasture in Texas, placing reed canary grass, which is a perennial,

first.

Professor McCarthy, in reporting on this grass in Bulletin No. 73,
of the North CaroHna Experiment Station, says: "It requires rich,
moist, light soil. On poorer soils it is much inferior to common rye
for winter grazing."

Its nutritive value is high. Comparing the chemical analyses of
the grass with those of rye and oat fodder, it is found that it contains

a larger percentage of protein and fat than either. Its nutritive ratio
is 1.6, showing it to be a well-balanced ration for stock.

Experience has shown that though rescue grass can not be offered
as a panacea for all the troubles of the Southern farmer as was first
claimed, it can be safely recommended as a valuable addition to the
winter forage plants of the South, either for hay or pasture.

Cornelius L. Shear,
Approved : Assistant Agrostologist.

James Wilson,

Secretary of Agriculture.

Washington, D. C, Maij 21, 1900.

O

/

Circular No. 27.-(Agros. 74.) (Issued Dec. 4, 1900.)

United States Department of Agriculture,

DIVISION OF AGROSTOLOGY.

[Grass and Forage-Plant Investigations.]
F. Lamson-Scribner, Agrostologist. ,. ^"V

AGROSTOLOGICAL NOTES. ^^^ ttV)^^

INTRODUCTION.

In the present circular, to which the general title "Agrostological Notes " is
given, are considered :

1. Grasses in the Herbarium of Dr. H. Muhlenberg.

2. Two New Species of Eatonia.

3. A New Variety of Panicum nashianum.

4. Nomenclature Notes.

5. Notes on Melica and Stipa.
The circular is scientific in its character and is of interest to the student of

systematic Agrostology. The examination of Muhlenberg's grasses by Mr.
Merrill and myself has helped to clear up some doubtful points in synonymy.
The several papers here presented are too short to be issued in bulletin form, but
it has been thought that the matter was of sufficient importance to warrant its
publication as a circular, thus avoiding needless delay.— F. L.-S.

1. THE (JKASSES IN THE HERBARIUM OF DR. H, MUHLENBERG.

By F. Lamson-Scribner and Elmer D. Merrill.

In 1817 Dr. H. Muhlenberg published his " Descriptio Uberior
Graminum et Plantarum Calamarium Americse Septentrionalis. "
The collections upon which this work was based are now the prop-
erty of the American Philosophical Society, and at present are
deposited in the Academy of Natural Sciences of Philadelphia.
Through the courtesy of the Conservator of the Botanical Section
of the Academy opportunity was given to make a careful exami-
nation of this collection. The specimens are preserved in folios
lOi by 15 inches, which are inclosed in wooden byok-like cases,
opening by a sHde upon one side. They are well preserved, and
although often fragmentary, at least in the case of North American
material, they are sufficiently ample for identification. The Grami-
I nece, Cyperacece, and Juncacece occupy three boxes and are con-
secutively numbered from 1 to 414. The classification is according
I to the Linn^an system. No data accompany the specimens in most
cases excepting the numbers, which are the same as those of the list
of the species contained in the folio, written upon the outside sheet.
These names are not always the same as those under which the

2

that of his work and there is seldom any doubt as to the plant
intended. It not infrequently happens, however, that more than
one species is included in a fold. When this occurs, it is usually
possible to determine the plant referred to, by careful examination of
the description. Unfortunately many of Muhlenberg's species are
not represented in his herbarium. There has long been a doubt as
to the identity of a number of Muhlenberg's species through a lack
of knowledge of his types, and it has not infrequently happened that
they have been wrongly referred or the names cited as synonyms
under species with which they are in no wise related. On this
account the matter here presented will be of interest from the fact
that it serves to clear up many points of synonymy.

"151 Panicum pungens M. 97 Elliott, 358." This was published as Panicum
setaceiun Mnhl. Descr. 99. Panicum neurantJmm G-riseb., is the same. A
specimen in this sheet bears the label " P. uniflorum n. s. ? "

" 153 Panicum glaucum M. 98 a" = Chcetochloa glauca (L.) Scribn.

"153 Panicum glauco aff. aristis purpur. Mon. 99 a. b." This was published
by Muhlenberg, Descr. 101, without name, Schrader giving it the name
Setaria affi)iis in Roemer & Schultes Mantissa 2 : 276. 1834. One specimen
on this slieet, labelled Panicum polystachion, is ChivtocJdoa corrugata parvi-
flora (Poir. ) Scribn. & Merrill. Another specimen is Chcetochloa imberbis
(Poir.) Scribn.

"154 Panicum lavigatum Elliott 353 M. 99 c " = Chcetochloa imberbis (Poir.)
Scribn. A very young form with undeveloped spikelets, exactly matched
by a specimen in Herb. U. S. Department of Agriculture, collected in the.
Dismal Swamp, Virginia, by G. McCarthy, 1883.

"155 Panicum viride Mon. 100 " = CJmtochloa viridis (L.) Scribn.

"156 Panicum verticillatum Mon. 98, h " = Chcetochloa verticillata (L.) Scribn.

"157 Panicum italicum M. 101 a," = Chcetochloa italica (L.) Scribn.

"158 Panicum germanicum M. 101, 6, Elliott m''= Chcetochloa italica german-
ica (Mill.) Scribn.

"159 Panicum hirtellum Elliott 85, Mon. 100 h" = OpUsmenus hirtellusR. & S.

"160 Panicum crus-galli M. 103 a." There are two forms in this cover— one
the common long-awned, long-branched form of Panicum crus-galli L. ; the
other, Panicum crus-galli muticum Vasey.

"161 Panicum crus-galli var. humifusa M. 103, 6." There is nothing in this
cover. »

"163 Panicum hispidum M. 103 b, Elliott 25 " = Panicum. walteri Pursh.

"163 Panicum colonum v. serotinum M. 103 a"^=Panicum colonum L.

"164 Panicum dimidiatum Walter, Ell. 478." This was published by Muhlen-
berg, Descr. 108, as Panicum ivalteri, and is P. cligitaroides Carpenter.

" 165 Panicum molle ? Mon. 77 c " = Eriochloa mollis (Michx.) Kunth.

" 166 Panicum latifolium Mon. 104 a." This cover contains both forms of Pan-
icum latifolium L. (P. porterianum Nash and P. pubifolium Nash) and also
a form of Panicum commutatum Schultes.

" 167 Panicum scoparium M. 104 b, macrocarpon Ell. 481." In this cover is one
specimen of true Panicum scoparium Lam. (P, viscidum Ell.) and one speci-
men of Panicum pauciflorum Ell,

"168 Panicum clandestinum M. 105 b " = Panicitm clandestinum L., as under-
stood by modern botanists.

" 169 Panicum nitidum (microcarpon) M. 105 'b"=Paniciim polyanthes Schiiltes.

"170 Panicum depauperatum M. 100. " This cover contains the form now rec-
ognized as Panicam depauperatum, and also specimens of Panicum lineari-
folium Scribn.

"171 Panicum verrucosum M. 107, 6 " = The form so considered to-day.

" 172 Panicum dichotomum M. 107 a, b." In this cover are several specimens
of Panicum dicltotomniii viride Vasey, the form now considered as true
Panicum dichotomum Linn, by common consent. The tall form of Pani-
cum lucidum Ashe is also represented, and also several unrecognizable
scraps.

"173 Panicum laxiflorum Lamarck, M. \0^ a," = Panicum barbidatuni Michx.
This was publislied by Muhlenberg, Descr. 114. 1817, as Panictim discolor
Sprengel. In this cover are also a few scraps of the form generally referred
to Panicum pubescens Lam. by recent authors.

"174 Panicum flexuosum V M. 108 b." This form was published by Muhlen-
berg in his Descriptio 115, as Panicum discolor co-species vel varietas major,
and is exactly identical with Panicum mattamusketense Ashe, Journ. E.
Mitch. Sci. Soc. 15:45. 1898. Ashe's name must, however, be retained as
major is preoccupied by Panicum nitidum var. majus Pursh, Fl. Am. Sept.
1 : 67. 1814.

"175 Panicum pubescens M. 109." This cover contains true Panicum scopa-
rium Lam. (P. viscidum Ell.), and also several plants erroneously referred
by botanists to Panicum pidjescens Lam.

"176 Panicum melicarium M. 112." This species has previously been referred
to Panicum Mans Elliott, but is Sporobolas junceus (Michx.) Kunth.

" 177 Panicum agrostoides M. 111." This cover contains the form generally
referred to Panicum agrostoides Muhl., and also specimens of Panicum
elongatum Pursh, not Salisb.

" 178 Panicum divergens Elliot 358, M. 112, 6" = Panicum cognatum Schultes.

" 179 Panicum virgatum M. 112." This cover contains several specimens of the
form considered to-day as true Panicum virgatum L. and also a specimen of
Panicum amarum Ell.

" 180 Panicum rostratum M. 113, anceps 'M.x.'= Panicum anceps Michx. There
is also a specimen of Panicum agrostoides Muhl. in this cover.

"181 Panicum geniculatum (dichotomiflorum) M. 114 " = A small branched form
of P. 2>roliferuni L.

"182 Panicum capillare L. M. 115a" = The common large panicled form of
this species.

" 183 Panicum capillare minor " = Panicum flexile (Gatt.) Scribn. Mr. Nash'
referred the wrong plant to Panicum minus (Muhl.) Nash. The synonymy
of this form should be as follows :

Panicum philadelphicum Bernh. ; Trin. Gram. Pan. 216. 1826, as synonym;
Nees, Agrost. Bras. 198. 1829. {Panicum diffusu.m Pursh, Fl. Am. Sept.

11:68. 1814, not Swartz, 1788; P. cajnllare minus Mnhl. Descr. 124. 1817, not
P. pubescens minor Poh\ in Lam. Encycl. 4:272. 1816, nor P. minus Nash,
Bui. Torr. Bot. Club, 22 : 421 . 1895. P. capillare flexile Gatt. Tenn. Fl. 94. 1887 ;
P. Jlexile (Gatt.) Scribn. Bui. Torr. Bot. Club, 20:416. 1893.) We believe

1 Bui. Torr. Bot. Club, 1 : 421. 1895.

L

Panicum philadelpliicum to be the earliest tenable name for this species, as
the description applies to the form here referred to that species, and,
moreover, the statement is made by Nees that he had seen specimens in
Schreber's Herbai'ium (which were doubtless from Muhlenberg), and at the
same time he cites Panicum capillare minus Muhl. , as a synonym. For the
form referred by Mr. Nash to Panicum minus we propose the following :
Panicum minimum (Engel.) n. comb. {Panicum, capillare minimum Engel. ;
Scribn. Tenn. Agr. Exp. Sta. Bnl. 7:44, Jig. 40, 1894; P. minus Nash, Bui.
Torr. Bot. Club, 22:421. 1895, not P. capillare minor Muhl. Descr.- 124. 1817.)

" 184 Panicum M. 116 a, N. Angl." This is described on page 37 of the

" Descriptio " without name. Roemer & Schultes gave it the name of Pani-
cum spretum. Mantissa, 2 : 248. 1824. This is exactly Panicum eatoni Nash,
Bui. Torr. Bot. Club, 25 : 84. 1898, and should be referred to Panicum niti-
dum Lam. (See Bui. 24, Division of Agrostology.)

"185 Panicum aquaticum M. 116 b, Elliott S8" ^Panicum gibbum Ell.

"186 Panicum ciliatum M. 117, Elliott i80"^Panicum ciliatum Ell.

" 187 Panicum strigosum M. 117, Elliott 497." This species is in the same rela-
tive position in Muhlenberg's Herbarium as is the description of Panicum
cartilagineum in his ' ' Descriptio. " There is in this cover of Panicum strigo-
sum one specimen of Panicum nashianum Scribn., one specimen of Panicum
scabriusculum Ell., and several specimens labeled '' Panicuyn aureum," which
are now referred to Panicum laxiflorum Lam. Panicum nashianum
Scribn. = ? P. cartilagineum Muhl.

"188 Panicum, guinea-grass polyg. " = PamciM>i maximum Jacq.

"189 " = 188.

"190 Panicum near enslini. " This cover contains both Panicum baldwinii Nutt.
and P. tvrightianum Scribn.

"191 " This cover contains several forms of Panicum commutatum

Schultes.

"192 Panicum parvulum M. 110, 6." In this cover are several unrecognizable
scraps, and also three plants with a label attached, bearing the name Pani-
cum deustum. We consider this to be the type of Panicum tenue Muhl. Descr.
118, and it is exactly identical with Panicum trifolium Nash. The synonymy
of this species is as follows :

Panicum tenue Muhl. Descr. 118. 1817. {Panicum deustum Brickell et Enslin,
1. c. 119, as synonym; Panicum Won R. & S. Mant. 2: 250. 1824; Panicum
macrum Kuuth, Rev. Gram. 1 : 40. 1835; Panicum trifolium Nash, Bui.
Torr. Bot. Club, 26: 580. 1899.) Panicum tenue Muhl., was changed by
Roemer & Schultes and by Kuntli because of an earlier use of the name
by Roxburgh, Catalog. 1813, but we believe this earlier publication is a
nomen nudum, the description not being published until the issue of his
Flora Indica 1 : 306. 1820, in which case Muhlenberg's name Panicum tenue
should be retained. It is possible that this species is true Panicum ensifolium
Baldw. ex Ell. Sk. Bot. S. C. and Ga. 1 : 126. 1817. A careful search in the
herbarium of the Academy of Natural Sciences of Philadelphia, revealed no
Panicum ensifolium from Baldwin.

"193 Panicum simplici affini Baldw. 866 Milium" = 1

"194 Panicum dubia Baldw. " This cover contains unrecognizable scraps of
several species.

The following species of Panicum, considered by Muhlenberg, are not repre-
sented in his herbarium:

11 Panicum Muhl. Descr. 106. 1817 (sine nomine). Evidently a form of

Panicum cms-galli L.

12 Panicum Muhl. Descr. 106. 1817 (sine nomine). Evidently a form of

Panicum crus-galli L.
25 Panicum nervosum Muhl. Descr. 116. 1817, not Lam. {P. commiitatum

Schultes. )
27 Panicum Muhl. Descr. 118. 1817 (sine nomine) [Panicum Muhlenhergi-

aunm Roemer & Schultes Mant. 2 : 230. 1824.)
33 Panicum densum Muhl. Descr. 122. 1817.
36 Panicum acuminatum Swartz, in Muhl. Descr. 125. 1817. (Panicum Muhleti-

bergii Sprengel, Syst. Veg. 1: 143. 1825; P. sprengelii Kunth, Rev. Gram.

1: 39. 1835.)
39 Panicum Muhl. Descr. 127. 1817 (sine nomine). {Panicum tremulum

Spreng. ex Roemer & Schultes, Mantissa 2: 237. 1824.)
41 Panicum cartilagineum Muhl. Descr. 128. 1817. ?
"201 Eleusine filiformis" published as Eleusine sparsa Muhl. Descr. 135.

\>i\1 .=■ Leptochloa mucronata (Michx.) Kunth, a depauperate form.
" 206 Poa viridis " Muhl. Descr. 138. 1817.= Poa pratensis Linn.
"207 Poa stolonifera " Haller in Muhl. Descr. 139. 1817.= Poa trivialis Linn.
"214 Poa capillaris var." Poa capillaris co-species, Muhl. Descr. 145. 1817.=

Eragrosth frankii Steud.
"215 Poa hirsuta" Muhl. De.scr. 145. 1817 .= Eragrostis pectifiacea (Michx.)

Steud. A specimen of Eragrostis trichodes (Nutt.) Nash, is also in the

sheet.
"216 Poa reflexa Ell. 361." Published as Poa refracta Muhl. Descr. 146.

\8\1 .^=Eragrostis refracta (Muhl.) Scribn.
" 223 Poa stricta uniflora." This was published as Poa ? uniflora Muhl. Descr.

151. 1817, and is the same as the species now called Sporobolus serotinus

(Torr.) A. Gray. The synonomy should be as follows:
Sporoblus uniflorus Muhl. new comb. (Poa ? uniflora Muhl. Descr. 151.

1817; Agrostis serotina Torr. Fl. U. S. 1:88. 1824; Sporohohis serotinus

A. Gray. Man. Bot. 577. 1848).
"235 Festuca prostrata." This was published as Festuca procmnhens Muhl.

Descr. 163. 1817, and is the same as Leptochloa fascicularis (Lam.) A. Gray.
"238 Festuca clandestina" Muhl. Descr. 163. 1817.= A depauperate form of

Leptochloa fascicularis (Lam.) A. Gray.
"Festuca brevifolia" Muhl. Descr. 167. 1817 .= Triplasis purpurea (Walt.)

Chapm. In this sheet are also specimens of Panicularia acutiflora (Torr.)

Kuntze.
" 248 Bromus ciliatus" Muhl. Descr. 169.=i?romMs kalmii A. Gray. There are

fragments of several other species in this sheet.
" 252 Aristida oligantha." This is the form published by Muhlenberg as Aris-

tida raceinosa Descr. 172. 1817, and is Aristida purpurascens Poir. Aristida

oligantha Muhl. Descr. 173, is not represented in the herbarium.
"256 Elymus villosus " Muhl. Descr. 175. 1817— Muhl. in Willd. Enum. 131.

1809.= Elymus striatus Willd., at least in part. Specimens of Elymus Mr-

sutiglumis Scribn. are also in the sheet, but Muhlenberg's description does

not apply to this species.

6

"258 Elymus glaucifolius " Muhl. Descr. 177. 1817— Muhl. in Willd. Enum.
131. 1809. = Elymus canadensis glaucifolms A. Gray.

" 261 Elymus ciliatus" Muhl. Descr. 179. 1817. In this sheet there is only the
spike of the plant, which is evidently a form of Elymus striatus Willd. ,
with unusually narrow outer glumes.

"267 Avena glumosa" Muhl. Descr. 184. 1817 .^ Danthonia sericea Nutt., not
Danthonia glumosa Beativ.

"278 Erianthus brevibarbis "' Muhl. Descr. 193. 1817 . = Eriarithus stricttis
Baldw. ex Elliott, Sk. Bot. S. C. and Ga. 1:39. 1816. There are a few
short hairs at the base of the pedicellate spikelets.

"387 Holcus odoratus" Muhl. Descr. 273. 1817 .= Savastana odorata (L.)
Scribn.

"389 Holcus saccharatus" Muhl. Descr. 275. 1817, is a form of cultivated Sor-
ghum, called broom corn.

"390 Holcus cernuus" Muhl. Descr. 276. 1817, is also a form of cultivated
Sorghum.

"391 Chloris monostachya " Muhl. Descr. 286. 1817 .= Campidosus aromaticus
(Walt.) Trin.

"394 Atheropogon apludoides" Muhl. Descr. 287. 1817 .= Botiteloua curtipen-
dida (Michx. ) Torr.

"402 Andropogon, Elliott, glaucus M. 264 b." A. glavcns Muhl. Descr. 278.
1817.= A slender form of Andropogon glomerahis (Walt.) B. S. P.

"407 Andropogon purpurea." Published as Andropogon purpurascens Muhl.
Descr. 282. 1817 .= Andropogon scoparius Michx.

2. TWO NEW SPECIES OF EATONIA.

EATONIA PUBESCENS Scribn. & Merrill sp. nov.

An erect or ascending, tufted, pubescent perennial 4 to 8 dm. high, with linear
leaves and exserted contracted panicles. Culms densely pubescent below,
with short, spreading or reflexed matted hairs, sometimes nearly smooth
above ; nodes smooth ; sheathes shorter than the internodes, strongly striate,
densely pubescent, with hairs similar to those of the culm; ligule promi-
nent, generally hyaline, obtuse, somewhat lacerate, 3 mm. long, rather
densely pubescent on the back ; leaf -blades linear, 10 to 18 cm. long, 4 to 6 mm .
wide, slightly auriculate at the base, very strongly scabrous on both sur-
faces and margins, often somewhat pubescent beneath, with short spread-
ing hairs. Panicle lanceolate in outline, 10 to 18 cm. long, rather densely
flowered ; common axis sparingly pubescent ; branches erect or ascending,
the lower ones 5 to 6 cm. long. Spikelets 3 to 3. 5 mm. long, 2-flowered ; empty
glumes very uneciual, the first linear, scabrous on the keel, nearly 2 mm.
long, the second 2 to 2.2 mm. long, broadly obovate, obtuse or truncate,
scabrous on the keel and on the two faint lateral nerves ; flowering glumes
lanceolate, acute, sparingly scabrous at the apex, slightly punctate, sessile
one about 2.5 mm. long, the pedicellate one slightly shorter.

Type specimen collected by S. M. Tracy at Starkeville, Miss., April 30, 1891,
distributed as Eatonia pennsylvanica (DC.) A. Gray. Other specimens
referable to this species, Hinson Springs, Tex., L. C. Johnson, 1886; Pales-
tine, Tex., 45 E. N. Plank, April, 1895.

p

I

This species is related to Edtonia obtusata (Michx.) A. Gray, but is readily dis-
tinguished by its densely pubescent culms, sheaths, and ligule.

EATONIA ARISTATA Scribn. & Merrill sp. nov.

An erect nearly glabrous, perennial, with very long, narrow leaves, elongated,
contracted panicles, and rather small spikelets, the second flowering glume
bearing a short awn. Culms and nodes smooth ; sheaths shorter than the
internodes, striate, often sparingly pubescent between the striae; ligule
hyaline, obtuse, 1.5 mm. long; leaf blades linear, glabrous, 10 to 18cm. long,
about 2 mm. wide. Panicles 15 to 25 cm. long, interrupted, the branches erect
or ascending, the lower ones 5 to 7 cm. long, somewhat flexuous. Spikelets
3 to 3.5 mm. long, 2 to 3 flowered; first glume very narrow, linear, slightly
scabrous on the keel, about 2 mm. long; second glume 2.5 mm. long, broadly
ovate, acute, slightly scabrous on the keel; sessile floret about 2.5 mm. lung,
acute, very slightly scabrous and punctate, its palea nearly equaling the
glume ; pedicellate spikelet shorter, scabrous, bearing a scabrous awn just
below the apex 1 to 2 mm. in length, which is sharply geniculate, forming
nearly a right angle with the glume.

Type specimen collected in South Carolina by A. H. Curtiss in 1875.

This species closely resembles Eatonia filiformis in its very long, narrow leaves
and panicles, but is distinguished from that species by its larger spikelets
and awned second floret. It is most closely related to Eatonia pallenn, Init
is at once distinguished by its elongated and very narrow leaves, smaller
spikelets, and geniculate awn of the second flowering glume.

This plant was considered by Vasey ^ as being identical with the species dis-
ciissed as a hybrid between Eatonia pennHylvank-a and Trisetinn palvstre,
but it is very distinct from that species. This form, like Eatonia pallens,
shows the close relationship of Eatonia to Trisetum.

EATONIA PALLENS (Spreng. ) Scribn. & Merrill, new comb. {Air a pallens
Spreng. Mant. Fl. Hal. 33. 1807; A. pallens Muhl. Descr. 84. 1817, at least
in part; Trisetum p)alustre Torr. X Eatonia jjennsylvanica (DC.) A. Gray;
Vasey, Bot. Gaz. 9:165. 1884; and Scribn. 1. c. 167. J?y. 1.)

An erect perennial 6 to 10 dm. high, with linear leaves, long contracted panicles
and pale spikelets, the second flowering glume bearing a slender awn at the
apex. Culms and nodes glabrous ; sheaths shorter than the internodes, stri-
ate, glabrous; ligule about 2 mm. long, hyaline, obtuse, lacerate; leaf
blades thin, 7 to 16 cin. long, 3 to 5 mm. wide, scarcely narrowed at the auricu-
late base, striate, slightly scabrous on the margins and nerves above, other-
wise glabrous. Panicle pale, exserted, lanceolate in outline, 12 to 18 cm. long,
the branches f asciciilate, erect or ascending, slender and somewhat flexuous,
the lowers ones 5 to 6 cm. long, naked below, rather densely flowered above.
Spikelets 4 mm. long, 2-flawered; empty glumes very unequal, the first 2.5
mm. long strongly-compressed, scabrous on the keel, hyaline on the margins,
lanceolate, acute and about 0.6 mm. broad when spread out, the second
glume 3 mm. long, much broader than the first, scabrous on the keel, broadly
obovate, about 1.7 mm. broad when spread out; first floret sessile, its flower-
ing glume 3.3 mm. long, lanceolate, acute, scabrous on the keel, peculiarly
and finely punctate. Palea hyaline 2.5 mm. long, cleft at the apex. Sec-
ond floret pedicellate, its rachilla somewhat hispid, similar to the sessile
floret except in being slightly shorter and bearing a slender .scabrous some-
what flexuous awn about 3 mm. long, immediately below the apex.

iBot. Gaz. 9:166. 1884.

8

Type locality: " Pensylvania. " General distribution: in wet meadows, Penn-
sylvania and Virginia, June-Jiily.

The original description of this species and also Muhlenbnrg's description are

given below.

"Aira pallens aristata, panicula contracta, valvulis calycinis insequalibus,
flosculo altero aristato, altero mutico. Folia linearia, glabra, rigida,
subinvoluta, vaginis pubescentibus.

E. Pensylvania 3Iiihlcvb." Sprengel, Mantissa Flora Halensis 33. 1807.

"Aira pallens. Culmo tripedali erecto nodoso. Foliis lineari-lanceolatis
glabris, nodis glabris. Ligula retnsa ciliata. Vagina substriata. Pan-
icula contracta nutante. Ramis paniculse 5, ternis geminisque scabris,
ramulis alternis. Cal. gluma 2-valvis corolla minor, nervosa, scabra,
valvula una latiori ; altera lineari utraque acuminata niutica, bi- et tri-
flora, flosculo nno pedicellato, omnibus hermaphroditis. Cor. gluma
3-valvis puncticulata, flosculi sessilis mutica, pedicellati aristata, arista
paulo infra apicem sive dorsali contorta, pedicello flosculi Isevinec pubes-
cente. Stam. 3. Pist. 2. Semen glabrum. Radix perennis. Habitat in
pratis humidis et varietas omnino mutica in sylvis, floret Jtinio et iterum
Sept. Aira Pennsylvanica, Sprengel. Habitus Avence sesquitertioi, cf .
Athena j)alustris, Michanx." Muhlenberg Descriptio 84. 1817.

In Muhlenberg's herbarium there are two sheets, one labeled ' ' 128 Aira pallens "
and the other " 129 Aira pallens var. aristata." In the first sheet are sev-
eral fragments of Deschampsia flexuosa Trin., Poa debilis, Torr. and an
Eatonia, probably Eatonia yiitida (Spreng.) Nash., while in the second sheet
there are specimens of Trisetum palustre Torr., and the grass described by
Vasey as a hybrid between Eatonia pennsylvanica and Trisetum, palustre,
the form here taken up as Eatonia pallens. It is at once evident from an
examination of Muhlenberg's description above, that this last form is the
plant described by Muhlenberg as Aira pallens. Muhlenberg erred in citing
Aira pennsylvanica Spreng., as a synonym, as this species is awnless and is
Eatonia nitida (Spreng.) Nash.

Sprengel's description of Aira pallens does not exactly apply to our plant, as he
speaks of its subinvolute leaves and i^ubescent sheaths. It is very probable
that he had other species under the same name, possibly Eatonia nitida or
some species of Trisetum, as is the case in Muhlenberg's Herbarium. This
is more probable, because Sprengel received the material on which the spe-
cies was based, from Muhlenberg.

This form was discussed by Vasey and also by Scribner ^ as a hybrid between
Trisetum palustre and Eatonia pennsylvanica, and in this discussion the close
relationships of Eatonia and Trisetum are pointed out with the conclusion
that the genus Eatonia should properly be placed in the Avenea' next to Tri-
setum and not in the Festucea', where it now is. In considering this form as
a species in the genus Eatonia, it is necessary to modify the character of the
genus so as to include this awned form. See Scribner, U. S. Dept. Agr.,
Div. Agros. , Bui. 20 : 135, fig. I04, 1900. This species is at once distinguished
from Eatonia peymsylvanica (DC.) A. Gray, to which it is most closely
allied, by its awned second -flowering glume.

The only specimens we have seen referable to this species is the one in Muhlen-
berg's Herbarium and a specimen collected at Hunting Creek, Alexandria,
Va., by Dr. Vasey, June 6, 1884.

J

1 Bot. Gaz. 9. 165-169. 1894.

I

S. A NEW VARIETY OF PAMCUM NASHIANUM.

Panicum nashianum patulum Scribn. & Merrill var. nov.

Culms much brancliecl, slender, often purplish, puberulent or short pubescent
throughout ; sheaths much shorter than the internodes, rather densely pubes-
cent, with short, spreading hairs, ciliate on the margin ; leaf blades densely
short-pubescent on both sides, papillate ciliate, with few long hairs on the
rather strongly cartilaginous margins. Rachis and branches densely puberu-
lent. Spikelets 2 to 3. 5 mm. long ; first glume thin, obtuse, about one-third as
long as the spikelet ; second and third glumes densely pubescent, with short,
spreading hairs.

Type specimen: 1296 Robert Combs, Braidentown, Manatee County, Fla., Sep-
tember 3, 1898.

General distribution : In fertile hammock and pine woods, Florida and Missis-
sippi, March to September. Other specimens referred to this species : Flor-
ida: Old Town, 858, 859 Combs, 1898; Grasmere, 1169 Combs, 1898; Lake
City, 132 (in part) Combs and Rolfs, 1898; Lake Alfred, T. Holm, 1893;
Jacksonville, 140 T. H. Kearney, 1895. il/missi^jji ; Biloxi, 4586, 4587 S. M.
Tracy, 1898.

This variety is closely related to the species, intermediate forms occurring, and
approaches Panicum demwmm. Trin. nearer than any other North American
plant we have seen. From the former it is at once distinguished by its
pubescent culms, sheaths, and leaves, and larger densely pubescent spikelets.
From the latter it is distinguished by its much smaller one-nerved first
glume and densely pubescent first and second glumes. The spikelets of
Panicmii demissuvi are glabrous.

4. NOMENCLATURE NOTES.

PANICUM RAMISETUM Scribn. nom. nov. Panicum suhspicatum Vasey,
U. S. Dept. Agr. Div. Bot. Bui. 8:25, 1889, not Desvaux, Opuscules 84, 1831.

AGROPYRON OCCIDENTALE Scribn. new comb. {Agropyron glaucum occi-
dentale Scribn. Trans. Kans. Acad. Sci. 9 : 119. 1885 ; A. spicatmn Scribn. &
Smith, U. S. Dept. Agr. Div., Agros. Bui. 17:298, ^c/. 594, 1899, not Festuca
spicata Pursh, Fl. Am. Sept. 1 : 83. 1814; Agropyron smithii Ryd. Mem. N. Y.
Bot. Gard. 1 : 60. 1900. ) Rydberg states that he examined the type of Festuca
spicata Pursh, in the herbarium of the Academy of Natural Sciences of
Philadelphia, and found that this form had been wrongly interpreted in
Bulletin No. 4 of this Division. He, therefore, applied the name Agropyron
smithii to this species, under the impression that Agropyron glaucum occi-
dentale Scribn. was a nomen nudum. A description of this variety was pub-
lished, however, in the Trans. Kan. Acad. Sci. 9: 119. 1885, and according
to the Rochester rules, the name occidentale will have to be retained for this
species.

AGROPYRON OCCIDENTALE PALMERI Scribn. new comb. Agropyron
spicatmn 2Kthneri Scribn. & Smith, U. S. Dept. Agr., Div. Agros., Bui. 4:33.
1897; A smithii imlmeri Heller, Cat. N. Am. PI. ed. 2, 3. 1900.

AGROPYRON OCCIDENTALE MOLLE Scribn. new comb. Agropyron spi-
catum molle Scribn. & Smith, 1. c. ; Agropyron molle Ryd., Mem. N. Y. Bot.
Gard. 1 : 64. 1900.

ELYMUS BOREALIS Scribn. nom. nov. Elymus ciliatus Scribn., U. S. Dept.
Agr., Div. Agros., Bui. 11:57, jjZ. JG, 1898, not Muhl. Descr. 179. 1817.
Elymus ciliatus Muhl. is not listed in Index Kewensis.

10

5. N0TE8 ON MELICA AND STIPA.

Note.— The following notes were made by Prof. C. V. Piper while studying the
types of certain Western grasses in the Gray Herbarium, Cambridge, Massachu-
setts, and later at the herbarium of the Division of Agrostology.— F. L.-S.

MELICA BELLA Piper, nom. nov. (M. bulbosa Geyer in Hook. Journ. Bot.
and Kew Gard. Misc. 8 : 19. 1856, nomen nudum ; M. bulbosa Geyer in Gray,
Proc. Am. Acad. 8:409. 1872, nomen nudum; M. bulbosa Geyer in U. S.
Dept. Agr., Div. Bot., Bui. 13 : 63, jjZ. 63, 1893, not Melica bidbosa Geyer in
Thiirber in S. Wats. Bot. Cal. 2 : 804. 1880.)

The fact seems to have been overlooked hitherto that both the first and second
publications of the name Melica bulbosa Geyer are nomina nuda. The first
publication of the name, with a description appended, is that of Thurber in
the botany of California, and while the species there described is not the
original plant of Geyer at all , nevertheless the name must stand for the
plant there described.

The first description of Melica bella, the original species of Geyer, is in U. S.
Dept. Agr., Div. Bot., Bui. 13, Vasey's "Grasses of the Pacific Slope." It
is somewhat variable, but its caespitose habit distinguishes it from its imme-
diate allies. Geyer's plant, which may be designated the type of Melica bella,
and which is in the Gray Herbarium, was collected in a " rocky ravine. Upper
Platte." It is matched by Cusick's No. 900a, from Union County, Oregon.

STIPA THURBERIANA Piper, nom. nov. (S. occidentalis Thurb. in Wilkes
U. S. Explor. Exped. 1 7 : 483. 1874, not U. S. Geol. Explor. 40th Par. 5 : 380.
1871.)

Much confusion has arisen in the names of the above two Stipas. Thurber first
described as Stipa occidentalis a plant collected by Pickering and Brecken-
ridge in Washington on the "N. branch of the Columbia," which was not
published, however, until 1874. In the mean time he had identified and
named Californian plants of Bolander's collection as Siqxt occidentalis, one
of which, No. 5038, from ' ' Yosemite Trail, " was taken by Watson as the
type of StijM occidentalis published in the Botany of the King Expedition
in 1871.

Dr. Vasey in 1882 detected the fact that the form described as Stipa occidentalis
in the U. S. Geol. Explor. 40th Par., 1871 was different from that described
in Wilkes U. S. Explor. Exped. 1874, but in segregating them, imfortu-
nately renamed the one which had first been published, namely, the plant
described in the U. S. Geol. Explor. 40th Par. 1871.

It is clear, therefore, that the name '' Stqxt occidentalis'' must pertain to the
plant published in the U. S. Geol. Explor. 40th Par. 1871, and consequently
the plant of the Wilkes' U. S. Explor. Exped. 1874, is here renamed.

Stipa occidentalis Thurber, in Watson U. S. Geol. Explor 40th Par. 5 : 380. 1871.
{S. stricta Yasey, Bui. Torr. Bot. Club, 10:42. 1883, not Lamarck, Tabl.
Encycl. 1:158. 1791. S. stricta sparsifloraY asey, Contr. U. S. Nat. Herb.
3:51. 1892. S. oregonensis Scribn., U. S. Dept. Agr. Div. Agros. Bui. 17:
\.ZQ,fig. 42G, 1899.)

Circular No. 38— (Agros. 77.)

United States Department of Airricnlture,

DIVISION OF AGROSTOLOGY,

[Grass and Forage-Plant Investigations.]
F. Lamson-Scribner, Agrof

GRASSES AND FODDER PLANTS ON THE POTOMAC ELM'S.

During the past summer the Division of Agrostology has grown a large vari-
ety of native and imported gi-asses and fodder plants upon the island in the Poto-
mac River, which is just south of the city of Washington and locally known as
the "Potomac Flats." The soil consists of dredgings from the river bed and is
very rich. It is an ideal place for growing the coarser annual fodder plants and
has afforded an interesting field for studying the new perennial grasses and clo-
vers, both of native varieties and those imported from foreign countries through
the Section of Seed and Plant Introduction of this Department and grown here
for the first time. The behavior of a number of grasses from oiu- Southwestern
States and Territories is especially worthy of note.

Curly mesquite, from the plains of Texas and Arizona, made a wonderful
gi'owth and produced a large crop of seed; turnip grass, from New Mex-
ico, made a surprisingly rich growth and seeded heavily ; water grass exhibited
a growth which maintains for it all the claims which have been made respecting
its forage value ; and sprangle, a native of Texas and regions west of that State,
promises to be one of the most productive and finest hay grasses grown on our
trial grounds. A number of species from Australia have manifested a ready
adaptability to our climatic conditions and on the soil of the Flats grew finely.
Button grass from the interior plains of South Australia grew with remarkable
vigor and seeded most abundantly. It is not unlikely that this grass may have
high value for portions of the warmer and drier regions of the Southwest.
Mitchell grass, another Australian species, regarded by stockmen in the interior
of New South Wales as one of the best of all native grasses both for its drought-
endtiring qualities and its fattening properties, made a vigorous growth and
appeared to be perfectly at home in its new surroundings. Bermuda grass, the
king of pasture grasses for the South, grew readily from seed sowni early in the
season and made a most luxuriant growth, the mass of leaves and stems cover-
ing the ground knee-deep with a rich bed of pasturage. A number of the Aus-
tralian saltbushes grew finely and showed great possibilities for forage produc-
tion of their class. The growth of the clovers and alfalfa was inferior compared
with that of the grasses, but the vetches and different varieties of peas and beans
succeeded well, while such plants as Kafir com, several varieties of pearl millet
and teosinte grew with great vigor and yielded an astonishing amount of forage,
especially the last named. A detailed report by Mr. C. R. Ball, assi.stant in the
Division, who had charge of the work of the Division on the Flats, is here pre-
sented and its publication recommended, not only for the interest attached to
the several species cultivated, but to place the work on record.— F. Lamson-
Scribner.

2

INTRODUCTION.

The experiments noted herein are exclusive of those made in the
grass garden on the grounds of the Department, and relate wholly
to the work done on about 4 acres of land on the Potomac Flats.
These 4 acres constitute a part of an area of some 22 acres, which
has been used during the past season by several divisions in the
Department for conducting experiments with plants and vegetables,
and which was primarily set apart for the use of the Division of
Botany. The land in question is on an island just south of Wash-
ington, D. C. , near Long Bridge, and was originally a tidal marsh. Its
present condition is the result of dredging out the channel between it
and the main land, now forming the city water front, and pumping
the mud over upon the marsh. The surface was thus raised from
two to ten feet above high tide. This work was done by the War
Department, and the use of the land by the Agricultural Depart-
ment was through permission of the Secretary of War. The land
is exceedingly rich, being the city drainage and harbor deposit of
many years' accumulation.

On the higher, well-drained areas the soil is loose and porous,
washing badly with heavy rains, but on the lower portions it forms
a hard surface crust as it dries after having been packed by floods.
As might be expected from the origin of the island, the subsoil is
very porous, so much so that the surface water frequently breaks
suddenly through the firmer upper crust into subterranean cavities,
new openings of this character being apparent after every heavy
rain. It will be readily seen that the soil is peculiar to the island,
and can in no way be considered typical of this or any other region.
This fact does not especially injure its value for comparative
experiments.

The ground was cleared and plowed in the spring of 1899 and
fallowed during that season. In the late fall it was plowed three
times, at intervals of about three weeks, and left rough over winter.
In the spring of the present year an Acme harrow was used fre-
quently, being run over the field in both directions to keep down, the
weeds. Ever since the formation of the land some ten years ago,
weeds have been growing and seeding unrestrained, and they cov-
ered the surface with a dense and luxuriant growth. On the higher
part the most abundant were sweet elder {Sambucus Canadensis
L.), velvet leaf {Abutilon Avicennae Gaertn.), lamb's quarter
{Chenopodium album L.), purslane {Portulaca oleracea L.), and
morning glory {Convohmlus sepium L.) On the low ground,
wormseed {Chenopodium anthelminticum L.), and a species of
heartsease or smartweed (Polygonum) were so abundant and so vig-
orous as to call forth the most strenuous efforts to prevent their

choking the young forage plants entirely. A constant warfare was
waged on the pests during the entire season, and it is to be hoped
that they will prove less troublesome another year.

For several reasons the conditions under which the experiments
of this year were carried out were far from satisfactory, a fact
which gives greater value to the good results obtained and no doubt
explains some of the failures, partial or entire. Work was not
begun until the end of April, and was then greatly retarded by a
lack of assistance and equipment, due to the insufficiency of the
appropriation which could be used for that purpose. The months of
May and June were very wet, as will be seen from the following
rainfall tables :

Daily precipitation at Washington, D. C, from April to November, 1900.
(Data furnished by the U. S. Weather Bureau.)

Date.

April.

May.

June.

July.

Aug.

Sept.

Oct.

Nov.

1

.08
3.48

.30
Ti

.01
.01
T

.06

•z

T

3

.21

.75

4

.04

.04
T
T

.03

'""".'35"

T

5

e.

T

T

.44
T

8

2.84

.34

9

.41

.02

.25

10

11

.03
..53
T

T
T

.03

12

.17
.78
.56
.03
1.00
1.68

.01

.08
.03

13 _

"'"'."19"
1.27
1.68

.19
.23

14

IH

16

.07
.04

17

.02'
.68
.12

"""'.'6i"

3.15
T

18

T

iq

T

.14

.40

"'""19"
T

""'.'os'

.32

20 ■

.13
..38
T
.11
.87
.07

T

T

21

.22
.01
.09

.01

22

T

23

.07

.06
.20

24

.14

.04

T

.08
.24
T

.31

25

.12

2fi

.31

27

T

1.06

28

T
.06

.15

29

30

T

.04

.29

31

T

4.02

1.74

10.94

2.28

Total

1.25

4.61

1.44

2.15

» " T " as used throughout this table denotes a trace of rainfall.

The extremely heavy rains of May 19, June 3, and June 8 did
much damage. The first one, coming after the planting of a large
part of the higher ground, fell with such rapidity and violence as to
wash away both seed and soil from some plats and mix the seed on
several others so thoroughly as to entirely vitiate their integrity and
necessitate replanting. Each of these rains covered the lower field
with standing water several inches deep, which drained away but
slowly from the level land or sloping in from the river. This
seriously delayed the planting on that area.

During July and August the weather was very hot and dry,
affording excellent opportunity to observe the drought-resisting
properties of the various crops. Some very interesting results were
obtained in this connection. Many of the plats, however, had been
sown so late that the young plants had not become firmly established
before they were overcome by the arid conditions which followed.
A large number managed to just about hold their own during those
months and afterward made a good growth.

The land allotted to this Division was divided by a roadway into
two nearly equal fields, numbered 9 and 10. Each field was then
laid out into plats a rod square, with paths 2 feet wide running in
both directions. There were about 480 of these plats. Each row of
plats running lengthwise of the field was given a series letter. A, B,
C, etc., and the plats in each series were numbered consecutively.
The first of the series in field 9 would thus be 9 — A — 1, the second
9 — A — 2, and so on for row A, The plats in the next row would be
9 — B — 1, 9 — B — 2, etc. The same scheme was used for field 10.

PERENNIAL MEADOW AND PASTURE GRASSES.

About 150 plats were devoted to experiments with grasses other
than the millets and sorghums. Quite a large number of those sown
earliest were rendered valueless for test purposes by a heavy rain
falling just after seeding, washing arway a large amount of seed and
mixing some of the remainder. This was most unfortunately the
case in a series containing 25 brome grasses, from which much in
the way of comparative results had been hoped. No report is made
here on several of our common cultivated grasses, such as red tops,
orchard grass, etc., which were included in the experiment. These
are well known perennials, and may better be reported on in another
season.

Bermuda grass (Cynodon Dactylon (L. ) Pers.) — This plat was intended to
demonstrate that Bermuda could be profitably started from seed, and it was a
complete success. The seed was from the Division supply, purchased in 1899,
and was planted June 6 at the rate of 20 pounds per acre. By the middle of
July there was an abundance of the creeping stems 3 to 6 inches long. Two
weeks later the plat presented a luxuriant mass of foliage about 4 inches high.
On the 20th of August the creeping branches covered the ground completely for
a distance of 2 or 3 feet on all sides of the plat, which was a dense mass of grass
12 to 18 inches high, and was ready for cutting. The sides of the plat were now
cut back to their original position. Two months later, on October 20, the
foliage was still soft, luxuriant, and perfectly healthy, while the creeping stems
had again reached a length of 4 feet. Had the plat been cut for bay at the
time the stems were cut back it would have been ready for a second cutting
before this date. Pieces of the stem trimmed off in August had covered the
entire surface of an adjacent vacant plat with wiry stems which have as yet
produced few leaves. Several severe frosts during the first two weeks of Novem-
ber killed most of the foliage.

As stated, this plat was seeded at the rate of 20 pounds per acre, which is 2 or
21 times as thick as is generally recommended. The resulting growth was very
thick, much thicker than was necessary to secure a good stand and an abundant
hay crop. At the rate of 8 pounds per acre, with seed costing 75 cents per
pound, the expense per acre is only $6. This ought, with well prepared soil and
careful seeding, to result in a good hay crop, free from weeds, the same season.
The ordinary method of planting the roots, while primarily less expensive,
usually results in a weedy field, with little or no crop the first year. Where
seed is used, the difference in amount of hay or pasturage obtained the first year
should more than pay for the seed.

Turnip grass {Panicum bulbosum H. B. K.).— Two square rods were planted
on June 6 with seed raised in the grass garden in 1899. In five weeks the plants
reached a height of about 6 inches, of vigorous growth and a beautiful glaucous
color. By the end of July they were nearly U feet in height, and this rapid,
sturdy growth was maintained throughout August. Early in September the
grass flowered at a height of about 5 feet, the seeds in the large open panicles
ripening by October 1, and 9 pounds were gathered on the 5th. By the 20th of
October the plants, still green and vigorous, began putting out small lateral
panicles to replace the large ones cut for seed.

An adjacent plat, planted with seed from what was called Panicum avenaceum,
proved identical with the above in every respect.

This grass makes a very rapid, vigorous growth, and, if planted thickly and
cut before maturing, it should yield an abundance of excellent hay. It promises
to be of great value in the Southwest and possibly also for other sections. Its
luxuriant growth on the "Flats" was hardly more striking than that in the
poorer soil in the grass garden on the Department grounds.

Sprangle {Lex>tochloa duhia Nees.).— One square rod was planted June 7 from
seed raised in the grass garden in 1899. It sprouted quickly, and the plants made
a fine growth, reaching a height of 8 inches by the middle of July and of 2 feet
by the 1 st of August. A month later the grass stood 4 feet high, strong and
healthy, and it was in full flower early in September. Owing to its isolated
position, it was badly lodged by heavy winds. Seed was gathered October 5 and
the plat mowed. Two weeks later a good aftermath was observed.

For softness and abundance of foliage and fineness of stem, this grass is unri-
valed by any other which grows to the same size— 5 feet or over. This I regard
one of the most important grasses from the Southwest yet experimented with
and have no hesitation in recommending its extended cultivation.

Large water grass (Pasjmhim dilatatum Poir. ) . — This vakiable grass was sown
May 28 on five fractional plats equivalent to 2 square rods. The seed, which was
obtained from an eastern seed firm, germinated slowly, and not until the end of
June were the plants above ground. During the remainder of the season growth
was vigorous, and by the middle of October the plants were 3 feet high and
strongly tufted. The basal and culm leaves made a dense growth of foliage 18
inches to 2 feet high, above which rose the naked flowering culms, on which the
seed was nearly ripe. On November 17, after several heavy frosts, the foliage
was injured but little and later seed was still ripening.

Another square rod planted on June 6 from seed gathered by the writer on the
Red River in Louisiana in 1898 came up at the same time and grew a little more
vigorously, ripening its seed at the same time.

This is one of the largest of the water grasses and has long been recognized as
valuable for wet meadows and pastures in the South. It can be highly recom-
mended and shovild be more generally used.

Jaragua {Andropogon mfus Kth. ).— One square rod was sown June 6 with
seed received from Brazil by the Section of Seed and Plant Introduction (No.

6

3891). It germinated fairly well and grew slowly during tlie hot, dry months
of July and August. On the 30th of October the grass had reached a height of
4 to 5 feet, a luxuriant mass of long, soft, and flexible basal leaves, but with no
sign of flowering culms. The foliage was killed by the early November frosts,
but it is hoped that the roots will survive the winter and establish a permanent
plat.

Dr. J. F. de Assis-Brasil in his work, "Cultura dos Campos,"

speaks of it as follows :

It reproduces itself readily both by seeds and creeping rhizomes. The stem
grows to the height of 12 feet at the time of flowering. As the stems are at that
time rather hard, it is best to pasture the meadows occasionally in order to keep
down the flowering stalks. An analysis made by Dr. Trovassos indicates a
higher percentage of protein and carbohydrates than in the best leguminous
plants, but for the purposes of this note it is sufiBcient to say that this indigenous
grass produces an extremely rich forage.

This grass may prove of great value in the South and Southwest.

Louisiana grass, carpet grass {Paspahmi compressum Nees.).— One plat was
sown May 23 with seed purchased by the Division in 1899 and presumably of the
crop of that year. At the end of seven weeks, or by the middle of July, it had
just begun to germinate. A second plat was sown on the 2d of June with seed
secured in 1898. By the middle of July this presented a close, even growth
about 2 inches in height, and this lead over the first plat was maintained
throughout the season. In the middle of October the grass on both plats was
about a foot high and both were in flower. A firm turf had been formed by
the creeping stems, and the foliage on the erect culms was healthy and abundant.

Both plats were situated on rather low ground where the water stood longest
after rains. In the South where this grass is native it grows abundantly on open
hillsides in the pine barrens as well as in the lower and more shaded lands along
the creeks and branches. It is valuable for lawns, and especially so for pastures.

G-rapevine mesquite {Panicum obtusum H. B. K.).— Two plats were planted
June 7. The seed did not germinate until late in July, when a thin stand ap-
peared. During the rest of the season the growth was vigorous, the erect shoots
reaching a height of 18 inches and producing considerable foliage. The stolons,
or runners, were very abundant and extended from 5 to 6 feet in every direc-
tion, binding the soil very closely. The foliage was killed by frosts early in
November.

It is native of the Southwest, extending into Mexico, and is worthy of experi-
ment in those regions. A plat of this grass in the garden on the Department
grounds has persisted for five years resisting the cold winters and making a good
growth each summer.

Mitchell grass {Astrebla pectinata F. v. Muell.).— One plat was planted May
26 with seed received from Prof. J. H. Maiden, Queensland, Australia. It ger-
minated readily and a fair stand was obtained. By the middle of July the strong,
leafy plants were 12 to 15 inches high and the first heads were appearing. Two
weeks later foimd the height increased to 2 to 2i feet, and some of the seed was
ripe. Production of seed continued throughout the season, the last being
gathered October 20, at which time the plants were about 3 feet high and the
foliage very abundant. Frosts in early November killed the foliage.

This grass is very highly regarded as a hay and forage plant on the interior
plains of Australia. It is very productive, and the strong root system enables it
to withstand long periods of dry weather ; so that it should be of value in our
Southwestern States. It should be noticed that most of the growth was made
during July and August, which were exceedingly dry and hot months.

Button grass {Dactyloctenium anstraliense Scribn.). — Two plats were planted
with seed received from Australia nnder the name of "Munyeroo," or " Peta-
chartes " grass. The grass on these two plats was identical in habit and appear-
ance with the plants grown in the grass garden in 1896 from Australian seed and
reproduced each year by volunteer seeding. Some of the grass garden seed of
1899 was planted on two near-by plats, but it was either washed away or most
of it failed to germinate, as only a few weak plants appeared. The button grass
was sown May 26, and by July 14 the circular prostrate tufts were a foot in
diameter and loaded with flower heads. On July 30 the first gathering of seed
was made. From that time until the plants were killed by frosts in November,
seed was produced in great abundance and gathered at intervals of about two
weeks. The diameter of the individual plants increased to about 18 inches.

Two adjacent plats, planted also on May 26, with a mixture of Australian
grass seeds were found to contain a small Panicum, considerable Eragrostis
brownii, and about half button grass. The rate of development during the
season was very much the same as that of the plats just described. The single
plants were, however, much larger, the branches being about 2 feet long and
rooting strongly near the base, the whole plant being some 4 feet in diameter.
The color was light green, and relatively much more foliage and less seed was
produced than in the other plats. This should make it of more value as a forage
plant and soil holder.

Button grass is a native of the hot and dry interior regions of Southern Aus-
tralia, and it may be of great use in similar regions in this country.

Curly mesquite {Hilaria cenchroides H. B. K.).— Three plats were planted
June 7 with seed gathered in 1897 ; germination was good and the growth was un-
checked by dry weather. By the middle of July the grass was 2 inches high and
the runners were abundant. Two weeks later the runners had reached a length
of 2 feet or more, striking root every four or five inches and sending up sturdy
new plants which filled all the vacant space between the original seedling plants
and made a strong, close growth, excellent for pasturage purposes. The erect
culms were about 6 inches high and in full flower. One plat was taken up in
September and transplanted to the Division's exhibit on the Pan-American
Exposition grounds at Buffalo, N. Y. The plants on the other two ripened seed
about October 1, at which time the fruiting culms were nearly dead, but the
basal portions and the young plants were still vigorous. This is one of the most
valuable grasses of the cattle ranges of the Southwest, and its successful propa-
gation by seed is of great economic significance.

Black heads {Pappophorum nigricans R. Br.). — This well-known Australian
forage grass proved to be the chief ingredient of a package of seed of mixed sand
grasses received from that continent through Prof. J. H. Maiden. It was planted
May 23 on one of the driest plats on the grounds. Germination was prompt, and
the growth continued without check during the dry months of July and August,
when the plants reached their maximum size of a foot to 15 inches and were
fruiting abundantly. The production of fresh growths of leafy branches and
seeds was continued until stopped by the November frosts.

This grass is very highly esteemed for pasturage in the drier parts of Australia.
It probably would never be valuable as a hay grass on account of its low, tufted
habit, but it should prove a desirable addition to the pasture grasses of our drier
regions.

A grass very closely resembling "Black Heads" both in appearance and habit
of growth occurs on the cattle ranges of Arizona, where it is highly valued by
cattlemen.

Grama grasses (Bouteloua). — One plat of Side-oats grama (B. curtipenclula
(Mx.) Torr.), one plat of Blue grama (B. oligostachya (Nutt.) Torr.), and two

8

plats of Bristly grama {B. hirsuta Lag.) were planted on May 21 on the highest
ground of the field.

All three of these valuable gramas germinated rather poorly and the Bristly
grama quite slowly also, but all were uninjured by the drought, and all made a
vigorous growth and produced an abimdance of both foliage and seed. Their
value as native pasture grasses in the Southwest is well known.

Blue grasses (Poa).— Twenty plats were sown to different species of Poas,
mostly from western seed. Some failed entirely, owing to poor seed, dry weather,
and unsuitable soil. Others made a very satisfactory progress, considering the
conditions.

The soil of the Flats can scarcely be considered suitable for testing the moun-
tain grasses. Several plats planted with forms of Kentucky blue grass from
Oregon and Washington gave very promising results. Six plats of Canada blue
grass, of two different lots, made a good growth and are well established for
another season.

Fescues (Festuca). — Nineteen different lots of fescues, comprising fourteen
species, were planted June 7 and 8 on the lowest part of the field. Two lots of
King's fescue {F. Kingii Scribn. ) failed to germinate. Most of the smaller
ones, such as forms of red fescue, sheep's fescue, and slender fescue, did not do
very well. From a few scattering tufts to a meager half stand was secured,
the growth ranging from 1 to 3 inches. Two plats of tall meadow fescue from
Russia (S. P. I. Nos. 1180 and 1337) both came up well and grew slowly during
the entire season, reaching a height of 1 foot. A plat of reed fescue {F. arundi-
nacea), from grass-garden seed of 1899, made the best stand and greatest growth,
about 2 feet. Two plats of rough fescue (F. scabrella) from seed grown at our
grass station at Walla Walla, Wash., presented a fine stand and a vigorous
growth of long, soft leaves, 12 to 18 inches in height. With a variety of hard
fescue from the same source a fine stand was secured with a sturdy growth of
1 foot. *

Palm-leafed panic {Panicum plicatum Lam.). — This ornamental grass was
planted on June 2, from seed raised the previous year in the grass garden. The
two plats were on very low ground and were wet for some time after planting.
On one plat and part of the second the seed came up quickly, while on the
remainder of the second it did not germinate for several weeks. Growth on
the first was fairly constant during the season, and the plats pi'esented a very
pleasing appearance. By the middle of October the grass was 3 to 4 feet high
and the long, slender panicles appeared. The folded or plicate leaves were 12
to 18 inches long, about 2 inches wide, and very handsome. The foliage was
entirely killed by the first hard frosts.

Some roots have been potted with the intention of trying its value as a house
plant. By proper trimming it may probably be made to resemble quite closely
the common Pandanus. It is valuable as an ornamental grass.

SUMMER AND SOILING CROPS.
MILLETS.

More than 50 plats were devoted to the different kinds of millets,
exclusive of pearl millet, which will be found in another place
(p. 10). The millets are divided into three groups, the foxtail
millets, the broom-corn millets, and the barnyard millets, each rep-
resenting a different species with its varieties.

Foxtail millets (ChaetocMoa italica (L.) Scribn.).— Thirty-one plats were
seeded on May 17 and 18 to the different forms. Common millet, German millet.

9

Golden Wonder millet, and Hungarian millet. Except for a few which washed
out and were later reseeded, very satisfactory results were obtained. The
plants varied in height from 3 to 3* feet for some of the common millets to 6
feet for some of the German and Golden Wonder millets. Fully one-third of
them matured seed in from To to 78 days. These were chiefly Common millets.
Most of the remainder matured in from 80 to 90 days. Three German millets
took 103 days and one required 110 for ripening its seeds.

One plat of Common millet was badly affected by a leaf-spot disease, while a
second one contained a number of smutted heads.

Broom-corn millets {Pmiicmn miliaceum L.).— Eighteen plats of these millets
were sown May 17 and 18. Most of the seed was secured through the Section of
Seed and Plant Introduction from China, Russia, and Asia Minor. There were
many decided differences observable in varieties grown in the different plats.
The plants varied greatly in regard to the thickness, height, and amount of
branching of the culms, hairiness, the size and habit of the panicle, and the
time of maturing. Two of the largest, 3i to 4^ feet tall, with large, slender,
drooping panicles 10 to 14 inches long, were very slow in growth, the seed
ripening August 7, eighty-one days from planting. One was from Asia Minor
(S. P. I. No. 3665), the other from China (S. P. I. No. 3867). Another, similar
in habit but smaller, from Russia (S. P. I. No. 1387), was equally slow. All of
the remainder, including three from American-grown seed, matured between
July 31 and July 35, or in from sixty five to sixty-nine days from the time of
seeding. On one plat only was any smut observed.

Barnyard millets (Pajiicitm crMS-firoZ7i L.).— The barnyard millets are culti-
vated forms of our common barnyard grass. Four plats were seeded on May
17 and 18. One (9-E-ll) with 1899 seed from the Washington Experiment
Station ripened on August 14, and one (9-E-3) with Russian seed (S. P. I. No.
2798) on August 10. Both grew to a height of three feet, but the Russian plant
was much more slender in habit. Another (9-F-8b), sown with seed from the
Division supply, was 4i to 5 feet high by the Ist of August, but had produced
no heads. It made a very luxuriant, leafy growth, but died about the end of
September without having flowered.

SORGHUMS {Sorghum vulgare L.).

Two plats of brown dhoura (9-G-3 and 4), the second of which was mixed
with yellow milo maize, gave good results. The plants grew 6 to 7 feet in height,
producing an abundance of stocky, much-branched stems full of leaves. The
short, stout panicles are both lateral and terminal. A plat (9-G-7) of white
African millet sown with Texas-grown seed made a splendid growth of about 7
feet, with abundant foliage, the lower leaves 3 to 4 inches wide and 3 feet long.
Several plats of sorghum grew from 8 to 9 feet tall and produced a most succu-
lent forage.

TEOSINTE (Euchlcena luxurians).

Teosinte stands unrivaled for quantity of forage produced. Two plats
(9-H-9-10) were sown May 31 with seed from France (S. P. I. No. 3024). By
the middle of July it was 3 feet high, and on August 1 it had reached 6 feet in
height and so luxuriant in growth that the rows, planted 3 feet apart, could no
longer be distinguished. By October 15 it was 10 feet high and an almost solid
mass of culms and foliage. The leaves were about 3 inches wide and 3 feet long
and crowded on the stems, which were 1 inch in diameter. About November 1
the male flowers began to appear, but soon afterward frost killed the plants.
Two adjoining plats, of which the label was unfortunately lost, made an equally
satisfactory growth.

10

PEARL MILLET (Pennisetum spicatum).

One plat was planted May 18 with seed obtained from a Richmond, Va., firm.
On July 14 a fine stand of plants 3 to 4 feet high was noted. By August 1 it
stood 6 to 7 feet tall, very full of leaves, but with no heads. Two weeks later
heads began to appear. On October 18 it was 10 to 13 feet high. The culms
were much branched but rather slender in habit, and the slender, spreading
branches gave the plants a bushy appearance.

Another plat (9-G-13) sown with seed received from College Station, Tex.,
presented a striking contrast. The plants were 8 to 10 feet high and much tuore
strict in habit than the preceding. The culms were miich branched at the top,
but the branches were appressed and bore no spikes. The spikes on the central
stems were erect and 10 to 16 inches long.

. MISCELLANEOUS FORAGE PLANTS.

SALTBUSHES.

Seed of several native species were planted May 12, but after the
heavy torrential rain of the 19th, none of the seed could be found;
certainly none of it ever grew. Some of the Australian species and
several Kochias, closely related plants from Australia, were also
lost in the same way.

Australian saltbush {A. semibaccata R. Br.). — One plat (9-A-l) planted May
12 with seed from California (S. P. I. 3922) germinated quickly and well, and
made excellent growth. On July 3 it was 8 inches high, and July 31, after a
month of very hot weather, the branches had grown to be 12 to 16 inches in
length. Growth continued throughout the season, the vines on October 17 being
3 to 4 feet long and making a dense mat about 6 inches in depth on the ground.
This was the slowest to produce fruit of any species grown this year. On the
date last mentioned hardly enough had been formed to pay for picking.

Three moi-e plats were planted June 9— two with seed from the Division sup-
ply and one with the same seed as the plat just described. All were up by the
1st of July, and during that hot month each made a growth of about 1 foot, a
rate maintained until the middle of October, when they had caught up with the
plat sown a month earlier — at least in size and vigor, but had just begun to
fruit.

Gray saltbush (-4. halhnoides hmdl.) .—One -plat (9-A-ll) was planted on May
12 with seed from California (S. P. I. 3925). On June 9 the plants were 1 inch
high ; on July 3 they were 6 inches high, and on July 31 they were fine and vig-
orous, 12 inches in height. Two plats sown July 9 with seed from the same lot
produced plants which were aboiit 5 inches shorter than the preceding through
the summer, but had reached an equal size by October 17, when all were robust,
spreading bushes fully 2 feet high and loaded with the spongy fruits which had
been ripening and falling to the grotxnd in such quantities that a bushel was
taken up from each plat at this time. The plants continued to fruit until nearly
the end of November, when they were pulled.

Annual saltbush {A. holocarpa F. v, Muell.).— One-half a plat (9-C-lOa) was
sown May 16 with seed from the United States grass station at Abilene, Texas,
1899. Germination was prompt and growth was rapid. On Jiily 3 the stout
bushes were 6 to 8 inches high, and fruit on some was nearly full grown.
August 1 they were 12 to 15 inches high, widely branching and full of fruit,
some of which was ripe and falling off. The plants grew to be 16 to 18 inches

11

high and fruited abundantly until they were pulled, November 28. On October
17, U bushels of fruit were taken up from the ground under the plants, and on
November 28 about 1 bushel more was secured, the total yield being at the rate
of 800 bushels per acre.

Slender saltbush (-4. leptocarpa F. v. Muell. ).— One plat (9-A-lO) was planted
May 12 with California seed (S. P. I. 3926). On June 9 the plants were 1 inch
liigli, and on July :} they were 6 to 8 inches high and beginning to spread out on
the ground like the Australian saltbush, which this plant closely resembles.
Two more plats (9-A-8 and 9) were sown June 9 with seed of the same number
and made a similar growth, equaling those from the first seeding by the middle
of October, when all presented a dense mat of prostrate vines 3 feet or more in
length and fruiting abundantly. They remained unharmed by frosts until the
end of November, when they were taken up for seed.

BURNET (Poterium sanguisorba L.).

A plat (9-C-ll) was seeded on May 18, and on June 6 the first plants appeared.
On July 3 they were 3 to 5 inches high, but their growth was checked by the
heat and increased only 2 or 3 inches during the month. With the coming of
the September rains they began a vigorous growth, and soon the plants had a
diameter of about 2 feet and continued green and fresh until December 1.

RAPE {Brassica napus L.).

One plat (9-C-7) was sown May 16 with French seed (S. P. I. No. 1449), which
germinated very quickly and by June 6 gave a good stand of healthy plants 3
inches high. On July 3 they were 1 to 2 feet high, strong and vigorous, the
largest in flower. A second plat (9-A-14) started June 9, came up well, and
grew rapidly at first, but was badly burned by the hot weather of July, standing
only 6 to 10 inches high at the end of that month and making almost no growth
later.

LEGUMES.

More than 200 plats were devoted to the experiments with legumi-
nous plants, for the most part with seed obtained from abroad through
the Section of Seed and Plant Introduction. The remainder were
from the Division stock.

ALFALFA {Medicago sativa L.).

Fifteen different lots of alfalfa were sown. The seeds were grown in different
parts of the United States, in the Argentine Republic, in Italy, France, and
Bavaria, in Turkestan, Siberia, and China, and in Egypt. Some of them were
planted on May 14, some on May 24, and some as late as June 22. The ground
selected was on the highest and best-drained portion of field 9. All the seed
germinated well, and the young plants made a vigorous start, but in the last
days of June and during July they were badly burned by the intense heat.
Growth was completely checked for over a month, and the uppermost leaves
became yellowish and finally died. The comparative vitality of the plants in
the different plats could best be judged by the way in which they recovered from
their injury.

Two plats (9-B-3 and 6) sown May 24 with Arizona seed seemed less vigorous
after the burning than did most of the others and were quite badly affected with
spot disease. Plat 9-C-14 was sown June 22 with Argentine seed (S. P. I. No.
3507), and plats 9-C-17 and 18 were seeded the same day with No. 3508. On

12

July 3, when most of the early planted plats were burned badly, these were just
germinating in excellent condition. By August 1 they were about 6 inches high
and slightly burned. On October 18 they stood 12 to 15 inches tall, healthy and
vigorous, some in flower, and with little or no spot disease to be found.

Four plats were sown May 23 with seed from Southern Europe. Plat 9-B-13
was planted with Italian seed, plat 9-B-14 with Bavarian seed, plat 9-B-15
with seed from the south of France, and plat 9-C-15 with seed from Provence,
France. All were badly tip-burned early in July, and did not grow any more
until the end of that month. On October 17 the Bavarian stood nearly 18 inches
high, while the others were only 1 foot high, but all except the Italian were
weak and dying. None were affected with spot disease except the Provencal,
and that but slightly.

Plat 9-B-17 was sown May 14 with seed from Bokhara, Turkestan (S. P. I.
679). This suffered severely from burning and made only 1 foot of growth dur-
ing the season. In October much spot disease had developed and the lower
leaves were dying. Plat 9-B-5 was sown on May 24 with Samarkand seed
(S. P. I. No. 1295). This made the same growth during the season as the last,
but remained quite green until the last of November. Very little spot disease
was observed. Plats 9-B-l and 2, planted from the same seed on June 21, were
just coming up on July 3 when most of the plats were badly burned. By the
middle of October they were 14 to 18 inches high and quite healthy in appear
ance. Plat 9-C-4 was sown with seed of Turkestan alfalfa [31. sativa Turkes-
tanica) (S. P. I. No. 1169). It was injured by heat as badly as any and recov-
ered as slowly. The growth amounted to 12 inches by October 17, when the
lower leaves were dying, but none was affected by spot disease.

Plats 9-C-56 and 6 were planted May 16 with seed from China (S. P. I. 1152).
The results were the same as with Turkestan alfalfa. Plat 9-B-4, sown May 24
with Egyptian seed, was badly burned, but made good recovery, and reached a
height of 12 to 18 inches by October 18, when it appeared vigorous and healthy.
No spot disease was found upon it.

CLOVERS.

Only five species were grown this year. They were all planted on
the lower part of the land, where they made a very satisfactory
growth.

Alsike clover {T. hybridum L.). — Plats lO-F-5 to 7 were seeded May 29 with
American seed. On June 11 the plants were just up and the stand was excel-
lent. By the 1st of July it was about 6 inches high. During July the growth
was considerably checked but the foliage remained healthy and bright. Increase
in size commenced with the September rains. By October 16 it stood 12 to 15
inches high, flowering and seeding abundantly. November 17 it was still fresh
and green, notwithstanding several frosts.

Crimson clover (T". Incamatum Ij.) . — Five plats, 10-1-14 to 18, were sown on
June 1 with seed from the Division supply. It was well up on June 11 and,
except for a slight burning in July, maintained a steady healthy growth through-
out the season, reaching a foot in height in October. On November 17 the plants
showed no effects of the severe frosts.

Hungarian clover {T. pannonicum Jacq.). — Plat lO-F-2 was seeded to this
May 29. Only a few plants appeared and made biit a slow growth. By October
16 each plant was a sturdy tuft of crowded stems about 6 inches high. It seems
perfectly hardy here for it was not injured by November frosts. It is a native
to Southern Europe, where it is said to be much earlier than red clover but less
relished by stock.

13

Mammoth clover {T. medium L.). — Two plats, lO-F-8 and 9, were planted
^lay 29. This germinated quickly and grew well, showing but little effect of
the drought. By October 16 the plants were 2 feet high and seeding well, but
were beginning to die. November 17 the most of the foliage was dead but no
injury had been done by frost.

Strawberry clover [T. fragifermn L.).— Plat 10-F-l was sown May 29 with
Russian seed (S. P. I. No. 1018). On June 19 the first plants appeared. These
grew slowly during the summer, showing no injury by heat. On October 16 the
plants were in vigorous healthy condition, 5 to 6 inches high, making a thick,
close growth. No flowers were produced. November 17 a few of the younger,
tenderer leaves were killed by frost but the mass of the foliage was not injured.

Bush clover (Le.y>edeza sp.).— Plats lO-F-14 and 15 were planted May 29 with
Japanese seed of L. seriaea Benth. (S. P. I. No. 3121). The plants came up
thinly and grew steadily through the hot spell, reaching a height of 2 feet in
October. The plant is perennial, erect, with slender ascending branches, short,
sessile leaves and white flowers borne sessile on the main stem and branches,
nearly hidden by the numerous small leaves. The stems are tough and woody,
but aside from this the plant should make good forage if palatable.

Another species {L. bicolor Turcz.) called "Hagi" in Japan (S. P. I. No. 2923)
was planted May 29 on plats lO-F-16 to 18. It grew very much as the preceding
but attained to the larger size of 3 feet. This species is very much branched
with longer silky leaves, and flowers in an open terminal panicle. It has much
more foliage in proportion to the stems and branches and would make more and
better forage. It is also a perennial. The leaves were mostly killed by frost in
November, by which time most of the seed was ripe.

Bird's-foot clover (Lof^fs comiculatus L.).— A half plat, 10-G-3a, was sown
with Russian seed (S. P. I. No. 1065) on May 31. Only a few plants came up,
but by their prostrate habit they soon covered the surface of the groiind. By
August 3 the plants were about 18 inches across and buds were appearing.
September 20 the plants measured 2| feet in diameter and were covered with
bunches of bright yellow flowers. A month later these had disappeared and
the seed was neai-ly ripe. The younger leaves were killed by November frosts,
but the plat still looked green and fresh.

This little plant is highly e-^teemed in Europe for use on light, sterile soils and
in dry places. It has become naturalized in the Southern States. The foliage
is relished by both cattle and sheep.

Swamp lotus (Lotus uliginoms Schkuhr.).— A two thirds plat, 10-G-2b, of
this was started with French seed (S. P. I. No. 1468) on May 31. A fair stand
resulted and the plants grew slowly until the heat of July checked them for a
time, burning the tender foliage somewhat. On September 20 they were 6
inches high, healthy, and beginning to flower. Growth continued but no seed
set. Frost killed only the more delicate yoiing leaves.

The swamp lotus is a native of northern Europe, where it is valuable for wet
or swampy land. It has been used successfully in similar places in the northern
United States.

Sulla {Hedysarum coronarium L.). — Plat lO-G-4 was planted May 31 with
Division seed and plats lO-G-5 to 7 on the same date with French seed (S. P. I.
No. 1518). A good stand of each was secured by June 11. July 6 the plants
were about 3 inches high and noticeably affected by the dry weather, from
which they recovered only slowly. In Augiist rapid, vigorous growth com-
menced, and by October 16 they were over 1 foot high and quite healthy in
appearance. November 18 found them 18 inches in height and but little injured
by frosts.

14

PEAS AND BEANS.

Field pea {Pisum arvense L.). — Four lots of field peas were sown May 39. All
were imported seed (S. P. I. No. 1173 and Nos. 1485, 1486, and 1487) from France
and Russia. Only a thin stand was obtained, and while they did not appear
affected by the drought, their growth was always slow. The vines finally
attained a length of 18 inches to 2 feet and most of them flowered, but none
ripened any seed. All died in the latter part of September. Many of them
were affected by downy mildew.

Bitter flatpea {Lathyrus sativuslj.) .—Threei)la,ts (10-D-10tol2) were planted
on May 28 and two plats (lO-E-13 and 14) on May 29 with Russian seed (S. P.
I. 1175). The plants came up promptly and grew vigorously during the sum-
mer. The plants were 16 to 20 inches high by the end of July, with an abun-
dance of flowers and some full-grown pods. By October 1 they were nearly all
dead and the seed ripe.

This species has not been cultivated in this country to any extent, because the
seeds contain an alkaloid which is said to be poisonous. It would probably be
of some value in the Southwest.

Chicharaca {Lathyrus tingitanus L.). — Plat 9-D-13 was seeded on May 28
with seed from Algeria (S. P. I. No. 3289). Only a thin stand was secured, but
the growth was very rapid. The plants were unaffected by the heat and drought
and reached a length of 3 to 4 feet by the end of the season. They produced
numerous flowers during July, but no seeds set and the pods did not develop.
On November 15 the vines were still green and healthy in spite of several severe
frosts. This species closely resembles the bitter flat pea, but grew much larger.
It has not heretofore been cultivated in this country. It was introduced from
Algeria, where it is used as a winter forage, and is said to be free from the poi-
sonous proi)erties found in so many of the flat peas. It is cultivated in the
Canary Islands and Morocco also, and is said to be hardy in Southern France
where exposed to a temperature of 26° F.

Winter flat pea {Lathyrus deer L.). — Two plats (lO-D-14 and 15) were planted
on May 28 with seed from France (S. P. I. No. 1459). On June 5 a good stand
appeared. By the middle of July the plants were 5 to 6 inches high and some-
what bui'ned. Toward the end of the month they revived a little and made
some new growth, but produced no flowers, and later died. It is used in Ger-
many and France as an early spring forage.

Everlasting flat pea {Lathyrus sylvestris L.). — Two plats (lO-D-3 and 4)were
planted on May 28 with Russian seed (S. P. I. No. 2802), and a half plat (10-D-6a)
with French seed (S. P. I. No. 1460). Both lots came up thinly and grew slowly
but steadily, reaching a height of about 16 inches by October 15. On November
1 6 they were still fresh and green after several heavy frosts.

Improved everlasting flat pea (L. sylvestris wagneri). — Seed from France (S.
P. I. No. 1461) was l^sed in planting a half plat (10-D-6&). In growth and
appearence it did not differ from the above. This plant has been widely praised
as a valuable forage during recent years ; but, it must be said, that experiments
in various parts of this country do not establish its claim.

G-ray milk-pea {Galactia canescens Benth.). — Ahalf plat (10-G-2a) was planted
May 31. The seed did not germinate until June 19, and then the growth was
quite slow. On July 15 the seedlings were 3 inches high and bush-like in habit.
By August 3 the twining branches were 6 to 8 inches long ; September 20 their
length had increased to 1 or 1+ feet, and a few flowers were out. On the 16th of
October the fruit was well set and the plants still fresh. By the middle of
November the plants had succumbed to the frosts without ripening more than
a small part of the seed. It is native to the dry region from Texas westward,
and for that part of the country should be valuable.

15

Chick pea, garbanzo {Cicer arietinum L.)-— Four plats were sown May 28 with
French seed lO-C-12 and 13 with S. P. I. No. 2137, and lO-C-14 and 15 with No.
2376. Plats lO-C-16 and 17 were planted with Morocco seed (S. P. I. No. 2977),
which failed to germinate. Both lots of French seed germinated well and made
rapid, healthy growth. By the end of July the vines were 1^ feet in height,
nearly done flowering, and bearing an abundance of nearly full-grown pods.
September 20 the vines were almost all dead, and the pods were then picked and
thrashed, only to find that but few of them contained good seed. Some were
empty, and many contained moldy, immature seed.

Green gram {Phaseolus mungo L.). — Plats lO-C-18 to 20 were planted May 28
with Russian seed (S. P. I. 1385). It germinated quickly and gave a fair stand
by June 5. On July 12 the plants were 12 to 15 inches high. The leaves were
considerably eaten by insects. July 27 the vines were very vigorous in appear-
ance, 2 feet or more in length, and making a tangled mass of very leafy forage.
Flower buds were appearing. On the 22d of September nearly half the seeds
were ripe and were then gathered. The older leaves were beginning to wither.
On October 15 the seed was all ripe, but the vines were still green. Twenty-five
pounds of seed were produced.

Two-thirds of a plat (10-H-3a) was planted on June 1 with seed of what was
called Chinese soy bean (S. P. I. No. 2873). Germination was very irregular,
but a fair stand was finally secured. Its habit and appearance soon showed it
to be the green gram. It made about the same growth as the one just described,
but the vines died a little earlier.

Slender bean {Phaseolus angiistissimus Gray) . — Plat 10-D-l was planted May
28 with seed grown in the grass garden in 1899. Only a few of the seeds ger-
minated, but the plants grew rapidly, reaching a length of 3 feet by Septem-
ber 20 and twining on each other in the absence of other support. It was then
in full flower. On October 15 the seeds were nearly ripe, and the vines were
dying. This bean is a native of the Southwest and may be worthy of cultivation
in that region. The slender, twining vines grow vigorously, are much branched,
and produce an abundance of foliage and fruit, which should make good forage.

Lablab {Dolichos lablab L.). — Several varieties of this bean from Italy and
Algeria were planted June 1 on low ground. They came up well and made
a most luxuriant growth of vines during the season. None of them seemed
injured or even checked by the drought. The long, running shoots reached a
length of 8 to 10 feet and produced an abundance of flowers in September. Fruit
was matured only on one or two of the dark-leaved, purjjle-flowered varieties.
All were killed by the November frosts.

Lentils (Ei-vum lens L. ). — Plats lO-F-19 and 20 were planted May 29 with seed
from Smyrna (S. P. I. No. 3658). A fair stand appeared. On June 11 the plants
were 2 to 3 inches high, and on July 16 the height was 6 to 8 inches. By July 27
they were badly twisted about by winds and were beginning to die. On Sep-
tember 20 they were entirely dead.

Seed (S. P. I. No. 1467) planted May 81 on plats lO-G-18 to 20 gave a good
stand, but the plants grew very slowly, took on a dried and sickly appearance,
and finally died after reaching a height of 4 inches. No flowers were produced.
Plats lO-G-16 and 17, planted on the same date with No. 1466, gave scarcely
better results. The plants grew a little taller and lived a little longer, but also
produced no flowers.

On May 31 plats lO-G-13 to 15 were sown with seed (S. P. I. No. 1183) from
India. On June 11 the plants were 2 inches high and the stand excellent. By
July 16 the plants were flowering at a height of 6 inches. August 3 they were
still flowering, though badly beaten down and covered by dirt. On August 27
they were entirely dead and the seeds ripe.

16

Velvet bean {Mucuna utilis Wall.).— Plats 10-F-lO and 11 were planted May
29 with seed from Florida (S. P. I. No. 4333) and plats lO-F-12 and 18 with seed
from the Division supply. Germination was fairly prompt, the young plants
appearing by June 11, but the stand was poor. By the middle of July the vines
were 3 feet long and growing vigorously. On September 20 the plats presented
a tangled mass of vines about 3 feet deep and running out for 8 to 10 feet on all
sides — a most luxuriant growth. October 16 they were still growing with no
sign of flowers. October 25 showed the more tender and exposed foliage killed
by the light frosts and early in November all were dead.

Soy bean {Glycine hispida Maxim.). — Several lots of soy beans were planted
June 2. The series of plats on which they were sown was intersected by a wet
depression in which the plants were noticeably smaller and less healthy in color.
As no lot was situated entirely in the depression, the comparative value of the
tests was not impaired. There were some striking differences in the time of
maturing and in size and habit also.

Seed of a variety from China (S. P. I. No. 2869) wiih light green leaves and
blue flowers made a growth of nearly 3 feet and matured about September 12.
The variety " Kaiyuski daiszu " from grass garden seed had also light green
leaves, very much wrinkled, and white flowers. It matured at the same time.

A variety from Paris (S. P. I. No. 5039) produced rather slender plants with
dark foliage. Fruit ripened September 20. This was called ' ' Early Black. "

Two Japanese forms (S. P. I. Nos. 4912 and 4913) were sturdy bush-like plants
about 2^ feet high and ripened their seeds about October 10.

Another Japanese variety, called "Best Green" (S. P. I. No. 4914) produced
stout, heavy plants which grew to be 3 feet high and matured very slowly. On
October 16 the vines were yet fresh and green and the seed was not ripe. Most
of it matured from November 7 to 13.

Two Chinese varieties, " Chm Pi Do" (S. P. I. No. 2871) and "Heh Pi Do"
(S. P. I. No. 2872) were planted May 31. The first grew to a height of 4 to 6
feet. The plants were at first erect but finally bent over from their own
weight. On October 16 the fruit was well formed but not ripening, and the
plants were still vigorous. On November 17 the frost had destroyed most of
the foliage. Only a few of the seeds were fully ripe. The second variety reached
a height of over 4 feet and matured a little earlier than the preceding.

LUPINES AND VETCHES.

Eight different lots of lupines were planted on May 29. The seeds of one were
collected in Wyoming in 1898. The others were all secured in France and
Russia by the Section of Seed and Plant Introduction. All were failures from
the start, as not more than a dozen germinated on any plat, and most of these
succumbed to the heat and insect attacks. Not more than a dozen plants on
all the plats lived to produce seeds, of which the greater part were eaten by
insects.

Thirty-six different lots of vetches, comprising fifteen different species, five
native and ten foreign, were planted May 22, 26, and 28. The different lots of
Giant Vetch {Vida gigantea Hook.) and Stolley Vetch {V. leavenworthii T. & G.),
both native species, failed to germinate, as did also one lot of Vicia dnmetonmi
from France. Five different lots of horse bean (F. faba L.) resulted in a very
poor stand of weak plants, which were badly injured by heat. The few plants
which survived made a growth of li feet and some produced flowers. Three
lots of Narbonne Vetch ( F. narbonensis L.) gave similar results. Two lots of
the one-flowered vetch ( V. monantha) gave fair stands, but the plants were dried
up at the height of 5 inches. They recovered somewhat later, but made little

growth, spending tlie remainder of the season in dying slowly. Two lots of the
Big-seeded vetch ( V. macrocarpa Bert.) from France resulted about the same, but
grew 10 inches high and persisted longer. It closely resembles Spring Vetch.
Two lots of Narrow-leafed vetch ( V. angustifolia) gave only a few small plants,
which soon died. A half plat of the American vetch (F. americcma Muhl.),
planted with Wyoming seed of lS9b, gave but a few plants, which made little
growth, were badly eaten by insects, and finally died early in October.

Vicia Bivonea Rafin. — Plat lO-A-2 was planted May 22 with seed from
Australia, 1897 (No. 2:3740 of Professor Maiden). A thin stand resulted, but the
young plants grew finely, reaching 6 to 8 inches in height by July 12. They
showed no sign of injury by the heat. By September 1 they were a foot high,
forming a den.se, tangled mass, but died soon after without producing flowers.

Black bitter vetch (T^ ervilia). — One plat (lO-A-20) planted May 29 gave a
thin stand, as did plats lO-D-19 and 20 planted with seed from France (S. P. I.
No. 1452). The plants grew slowly but did not seem affected by the heat. By
October 15 the plants were a foot high and still green but had produced no
flowers. This species is adapted to hot dry climates. It is cultivated in the
Barbary States, where it produces large crops.

Scaiiet vetch {V.fulgens Batt.).— Plats 10-B-ll and 12 were planted on May 28
with seed from France (S. P. I. No. 1514) and plats lO-B-13 and 14 with another
lot (S. P. I. No. 4:i36) from the same source. In each case a thin stand was
obtained. On July 12 each was 6 to 8 inches high, the first entirely healthy, the
second somewhat burned at the tender tips. By the end of the month no differ-
ence could be observed between the two. On September 17 they were 1^ to 2 feet
high, a healthy vigorous growth of slender, tangled vines. October 1 5 they stood
about :3 feet high and showed no signs of flowering. On November 17, the
vines, while matted down somewhat, showed no injury from the frosts. This
annual comes from Algeria, where it is known as a rapid grower and is consid-
ered one of the most valuable forage plants.

Spring vetch ( V. saliva L. ). — Nine different lots of this vetch were sown, and
the results in all cases were nearly the same. The stand was not more than one-
half in any case. The young jjlants were healthy, but were soon checked by the
hot weather and quite badly burned. They recovered slowly from this and made
new growth, but were always feeble, reaching a final height of about 10 inches
and none of them producing flowers. By the end of September most of the
plants were dead.

MogoUon vetch ( V. e.vigua Nutt. ).— Plat 10-C-l was sown May 28 with seed
of the crop of 1897 and plat lO-C-2 with seed of 1898. About half the seed gi-ew,
and the young plants, like those of the wild vetch, grew slowly throughout the
hot months and showed no actual injury. The plants lived imtil the middle of
October, at which time they had reached a length of from 18 inches to 3 feet,
but had produced no flowers. This plant is native to the Southwest from New
Mexico to California.

Vicia Jiirsiita Koch.— Plat 10-A-:5 was planted May 22 with seed of this vetch
raised at Ophir, Wash., in 1899. A thin stand appeared, and on June 19 the
plants were 1 to 2 inches high. July 12 foimd them 4 to 5 inches high, green
and healthy, and with no sign of injury from the heat. On July 27 the plants,
though no larger, were still uninjured, but subsequent growth was slow. Sej)-
tember 17 sliowed a height of 7 inches, but no sign of flowers. On October 15
the plants were beginning to take on autumn colors. November 27, after several
severe frosts, they were still alive.

This vetch is a native of Asia and has been sparingly introduced into this coun-
ti-y from Europe.

18

Milk vetches.— The seeds of several native species of milk vetches from the
Uocky Mountains and of two imported species were planted June 1. Of the
natives only one germinated at all, and that produced but a half dozen plants.

Morton's milk vetch {Astragalus Mortoni Nutt?).— Plat lO-I-l was sown with
seed from the Division supply. Only a few came up, and these made but a
slow growth, although they did not appear to be affected by the heat— at least
the foliage was not injured. The plants are prostrate, forming round mats
about three feet across. On November 17 they were still growing, after severe
frosts, but had not flowered. This species is native to the Rocky Moiintains and
westward to the coast. It produces an abundance of foliage and may prove of
some value in that region.

Astragalus falcatus Lam. — Two lots of French seed were sown June 1. Plat
10-1-4 was planted with S. P. I. No. 1448, and 10-1-6 and 7 with No. 5084. A
thin stand was secured and a slow growth resulted. October 16 the plants were
one foot high, fresh and healthy in appearance, but without flowers. Novem-
ber 17 they were unharmed by the heavy frosts. This species is native to the
Caucasus, and may prove of value in the Rocky Mountain region.

Gtenge, rengeso {Astragalus sinicus'L.).— Flat 10-1-5 was planted with seed
from Japan (S. P. I. No. 3725). On June 11 a good stand had appeared. About
July 1, when the plants were only two inches high, they were checked and
burned by the heat, and recovered only slowly. October 16 the plat presented
a very healthy appearance. The plants were then six inches high and a few
clusters of blue flowers were out. November 17 flowers were still appearing,
but the upper leaves had been killed by frosts. The rest of the plant was
uninjured.

C. R. Ball,

Assistant in charge of Grass Oardens and

Experiments within the District of Columbia.
Approved :

James Wilson, Secretary.
Washington, D. C, December 13^ 1900.

Circular No. 29. ^^ (Agros. T8.)

United States Department of Agriculture,

DIVISION OF AGROSTOLOGY.

[Grass and Forage-Plant Investigations.]
F. Lamson-Scribner, Agrostologist.

THE GRASSES IN ELLIOTT'S "SKETCH OF THE BOTANY OF SOUTH

CAROLINA AND GEORGIA."

One of America's most distinguished botanists was Stephen Elliot, born at
Beaufort, S. C, in 1771. He was a man of marked ability, and began his career
as a legislator, in which capacity he served his State for many years, taking a
leading part in many of the important measures of his day. Although con-
stantly engaged in public and engrossing financial business, he did not neglect
literary and scientific pursuits. He found time during his busy life to prepare
what he modestly termed a "Sketch of the Botany of South Carolina and Geor-
gia," a work of 1,344 pages, written in Latin and English, and containing a dozen
finely executed plates, illustrating nearly 50 species of grasses and sedges. The
work is an enduring monument to Elliott's scientific skill and acumen. Not
until one has prepared a book where almost every line contains a statement of
fact learned from original observation, can he fully appreciate the amount of
patience and labor involved in the preparation of such a work as the ' ' Sketch of
the Botany of South Carolina and Georgia." For many years Elliot's "Sketch "
was the only botanical authority for the plants of the region it embraced, and
to-day it remains indispensable to the working systematic botanists of our
country.

Elliott began the publication of his work in parts. The first part, including
pp. 1-96, was published in October, 1816; the second, which carried the work to
pp. 222, was issued in February, 1817. This second part contained descriptions
of the grasses, and is, therefore, the one most immediately interesting to agros-
tologists. ' The entire work was finally issued in two volumes, the first volume,
containing parts 1-5, was published in 1821, as indicated by the title page, the
second in 1824.

The following notes presented here for publication are based upon a careful
examination of the collection several years ago by myself, the notes then taken
having been recently verified by Mr. Elmer D. Merrill, who was by my direction
sent to Charleston for that purpose. I wish to thank the authorities of the Col-
lege of Charleston for kindly extending to the Department the privileges accorded
to it, enabling us to accomplish our object.

NOTES ON THE SPECIES.

Many species noted below were described by Elliott as new, but
some had been published by foreign botanists who had received speci-
mens from American collectors, and a few of the earlier published
species taken up by Elliott were misunderstood by him. There is
much existing doubt respecting the identity of some of Elliott's

' Anthoxanthum and the species of Erianthus were published in the first part.

2

species of grasses, and as the only means of clearing up this con-
fusion and preventing further error it was necessary to examine
the specimens upon which they were based. Fortunately Elliott's
herbarium has been preserved and is now the propert}' of the College
of Charleston. The herbarium consists of 28 volumes of folios, 12
by 23 inches in size, and the specimens are for the most part attached
to the sheets and labeled both on the outside of the fold and on the
inside. Often several species are in one sheet, but in such cases
each specimen is separately labeled. A part of the herbarium, espe-
cially that containing the grasses, is in very good state of preserva-
tion, while in several of the volumes of other plants nearly all the
specimens have been destroyed by insects. The Gramineae occupy
exclusively one volume and portions of three others.

Anthoxanthum odoratum Linn. ; Ell. Sk. 1 : 37.

Erianthus alopecurioides Ell. Sk. 1 ■.S8.=Erianthus saccharoides Michx.

Erianthus brevibarbis Michx.; Ell. Sk. 1:39. On this cover is also the name
" Erianthvs contortns," which is scratched out. The specimen in the Her-
barium of Elliott is matched by No. 1426 A. P. Anderson, South Carolina,
October 1, 1897, in the U. S. National Herbarium.

Erianthus strictus Baldwin ; Ell. Sk. 1 : 39. This is the same as a specimen in
the Herb. Acad. Nat. Sci., Phila., so labeled by Baldwin.

Erianthus contortus Ell. Sk. 1 : 40. This name appears in the Herbarium of
Elliott only on the cover with Erianthus brevibarbis noted ^bove.

Cenchrus echinatus L. : Ell. Sk. 1 : 93.

Cenchrus tribuloides L. ; Ell. Sk. 1 : 93. In this cover is true Cenchrus tribuloides
L., and a specimen of Cenchrus macrocephalus (Doell.) Scribn.

Spartina juncea Ell. Sk. 1 -.M.^Sjiartina patens (Ait.) Muhl.
Spartina polystachya Michx. ; Ell. Sk. 1 : 95.

Spartina glabra Miihl. : Ell. Sk. \:96.=Spai'tina strictu maritima (Walt.)

Scribn.
Arundinaria macrosperma Michx. ; Ell. Sk. 1 : 96.
Muhlenbergia diffusa Ell. Sk. 1 :98.
Muhlenbergia erecta Muhl.; Ell. Sk. \ ■.98.:=Brachyelytntm erectum [Qchreb.)

BeaiTV.

Trichodium laxiflorum Michx. ; Ell. Sk. 1 : 99.=AgrostishyemaHs (Walt.) B. S. P.

Trichodium perennans Walt. ; Ell. Sk. 1 : 99=A form of the grass now referred
to Agrostis altissima (Walt.) Tuckerm.

Leersia virginica Willd. : Ell. Sk. ] :WO=Homalocenchrus virginicus (Willd.)
Britton.

Leersia lenticularis Michx. ; Ell. Sk. 1 : 100. =Homalocenchrus lenticularis
(Michx. ) Kimtze.

Leersia oryzoides Linn.; Ell. Sk. \ ■AO\.:=Homalocenchrus oryzoides (Linn.)

Poll.
Phalaris americana Ell. Sk. 1 : 102. ^=Phalaris caroliniana Walt.
Aulaxanthus ciliatus Ell. Sk. 1 : 102.=AntJiaenantia villosa (Michx.) Benth.

Aulaxanthus rufus Ell. Sk. \ .103.= Anthaeiiantia nifa (Ell.) Schultes. Speci-
men not seen.

Milium paspaloides Ell. Sk. 1 : \04.=Paspalti7n paspaloides (Michx.) Scrlbn.
This species is in Elliott's Herbarium under the name " Milium coi-nigatum."
With it is a note by Ravenel sajnng "This is M. paspaloides Ell. Sk."
Paspalum setaceum Michx. ; Ell. Sk. 1 : f04. Not seen.
Paspalum deblie Michx. ; Ell. Sk. 1 : 105. Not seen.

Paspalum ciliatifolium Michx. ; Ell. Sk. 1 : 105. A form with long narrow leaves,
which are regularly ciliate along the margins, otherwise smooth. Sheaths
hairy at the throat. Racemes solitary. Spikelets very smooth, 2 mm. long.
Paspalum dasyphyllum Ell. Sk. 1 : 105. Leaves 15 cm. long, 5 to 7 mm. wide,
pubescent on both sides with stiff, rather long, spreading, papillate hairs.
Rachis of the racemes about 1 mm. broad, undulate. Spikelets slightly
pubescent, about 1.5 mm. long, crowded.
Paspalum praecox Walt. ; Ell. Sk. 1 : 106. This resembles Paspalum purpura-
cens. Spikelets about 2 mm. long, in four rows. Rachis flat, as broad as
the rows of spikelets. Racemes 3, ascending.
Paspalum laeve Michx. ; Ell. Sk. 1 :106.

Paspalum floridanum Michx. ; Ell. Sk. 1 : 107. In Elliott's Herbarium the plant
under this name is identical with the specimen under Paspalum laeve except
in having the sheaths somewhat pubescent.
Paspalum plicatulum Michx. ; Ell. Sk. 1 : 107. Specimen not seen.
Paspalum purpurascens Ell. Sk. 1 : 108. = Paspalum boscianum Fliigge.
Paspalum dlstichum Linn. ; Ell. Sk. 1 : 108. Specimen not seen.
Paspalum vaginatum Ell. Sk. 1 : 109. = Paspalum membranaceum Walt.
Ceresia fluitans Ell. Sk. 1 : 109 = PaspaZrwH mucronatum Muhl.
Phleum pratense L. ; Ell. Sk. 1 : 110. The only specimen under this name in
Elliott's herbarium is from Sullivan's Island, and is Polypogon monspeliensis
(L.) Desf.
Alopecurus geniculatus L. ; Ell. Sk. 1 : 111.

Panicum cenchroides Ell. Sk. 1 : 111 = Cenchrus viyosiiroides H. B. K.
Panicum laevigatum Muhl.; Ell. Sk. \: 112 = Chaetochloa imberbis (Poir.)
Scribn. This species is in Elliotts herbarivim under the name ' ' Paniciim
glaberimnm," and is exactly matched by a specimen in the herbarium of
the U. S. Dept. Agriculture, collected in North Carolina by G. McCarthy in
1889, distributed as Setaria glauca var. laevigata Chapm.
Panicum glaucum L. ; Ell. Sk. 1 : 112 = Chaetochloa glauca (L.) Scribn.
Panicum glaucum var. glaucum? Ell. Sk. 1 : im = Chaetoehloa cornigata parvi-

flora (Poir.) Scribn. & Merrill.
Panicum glaucum var. flavescens Ell. Sk. 1:113 = A yellow-awned form of

Chaetochloa glauca (L.) Scribn.
Panicum glaucum var. purpurascens Ell. Sk. 1 : 113 = A short-awned form of

Chaetochloa imberbis (Poir.) Scribn.
Panicum corrugatum Ell. Sk. \ : m = Chaetochloa corrugata (Ell.) Scribn., the
form so considered in Bui. 21:22, fig. 11, of this Division, and matched by
No. 108 T. H. Kearney, Apalachicola, Fla , July 15, 1895, in the U. S. Nat.
Herb.
Panicum italicum Walt. : Ell. Sk. 1 : 115 = Chaetochloa magna (Griseb.) Scribn.
Panicum crus-galli L. ; Ell. Sk. 1 : 114 = The form so considered to-day. Var.
muticum is the awnless form. Var. aristatum is not represented in the
herbarium. Var. hispidum is Panicum tralteri Pursh.

4

Panicum walteri Ell. Sk. \ : 115 = Panicuvi digitarioides Carpenter. In this
sheet is a label bearing the name " P. dimididiatum." Muhlenberg consid-
ered this species under the same name {Panicum ivalteri). See Scribner &
Merrill, Circular 27:2, of this Division.

Panicum hirtellum Michx. ; Ell. Sk. 1 : 115 = OpliHinenus hirtellus R. & S.

Panicum gibbum Ell. Sk. 1 : 116.

Panicum molle Michx. ; Ell. Sk. 1 : 116 = Eriochloa mollis Kunth.

Panicum gymnocarpum Ell. Sk. 1 : 117.

Panicum geniculatum Muhl. ; Ell. Sk. 1 : IIH ^= Panicum prolifenim Lam.

Panicum anceps Michx. : Ell. Sk. 1 : 118. Not seen.

Panicum hians Ell. Sk. \ : IIS ^ Pajiicum melicarium Michx.

Panicum latifolium Walt. ; Ell. Sk. 1 : 119=Panicum latifolmtn Linn, (excluding
reference to Sloane). A specimen of Panicuvi currani Ashe, is also in the
cover.

Panicum scoparium Lamarck; Ell. Sk. 1:119. Recent investigations have
shown that Panicum scoparium Lam., has been entirely misunderstood by
American botanists and that it is the form described by Elliott as Panicum
viscidum. See Scribner & Merrill Bui. 24 : 34 of this Division. A descrip-
tion of Elliott's plant is given below.

Panicum ravenelii Scribn. & Merrill, U. S. Dept. Agr. Div. Agros. Bui. 24:34.
January 9, 1901. Panicum scopainum of. Ell. Sk. Bot. S. C. andGa. 1 : 119. 1817,
not Lam. Encycl. 4 : 144. 1797. — An erect or ascending, caespitose perennial
3 to 6 dm. high, with broad, ascending leaves, large spikelets and generally
few flowered panicles. Culms at first simple, later becoming branched,
rather densely papillate-pubescent with ascending or nearly appressed hairs ;
nodes bearded with a dense ring of short spreading hairs, and with a smooth
ring immediately below ; sheaths shorter than the internodes, or the upper
ones crowded and overlapping, striate, distinctly papillate-pubescent, densely
bearded on the back at the apex ; ligule a fringe of rather stiff hairs 3 to 5 mm.
long; leaf-blades lanceolate, acute, slightly narrowed at the rounded and
somewhat clasping base 8 to 13 cm. long, 1. 5 to 2 cm. wide, ciliate on the carti-
laginous margins, rather densely pubescent beneath with soft hairs, glab-
rous above or rarely with few scattered papillate hairs, nerves faint above,
very prominent on the lower surface. Panicles about 10 cm. long, the
branches spreading or the lower ones often erect; rachis and branches
densely pubescent with short hairs. Spikelets 4 mm. long, ovate, obtuse,
slightly pubescent with short spreading hairs; first glume acuminate,
about one-third as long as the spikelet; second and third glumes 9 to 11
nerved ; flowering glume very smooth.

Type specimen in the Herbarium'of Elliott, College of Charleston, Charleston,
» S, C. , type locality. South Carolina.

General distribution: in shaded fertile soil. District of Columbia south to
Florida and Texas, May to September, not common.

Specimens examined : District of Columbia : near Eastern Branch, L. F. "Ward,
June 16, 1881 ; near Chain Bridge, E. S. Steele, 1899. Tennessee : no locality.
S. M. Bain, 1893. South Carolina: no locality, Stephen Elliott, type in
Herbarium of Elliott; Aiken, W. H. Ravenel— F. L. -Scribner, 1894; Florida:
• Apalachicola, A. W. Chapman, 2994a Biltmore Herbarium; Monticello, 305
R. Combs, August 4, 1898. Arkansas: "Northwest Arkansas" 31 F. L.
Harvey; Benton County, E. N. Plank. Texas: Harvester, No. 7 F. W.
Thurow% June, 1898; Waller, 7a F. W. Thurow, 1898.

This species is readily distinguished from Panicum pauciflorum Ell., and P.
scrUmerianum Nash, to which it is most closely related, by its much broader
pubescent leaves and larger spikelets. Of the specimens cited above, S. M.
Bain, 1898; Steele, 1899; 2994a Biltmore Herbarium, and the specimens from
Ravenel have been carefully compared with Elliott's type.

Panicum pauciflorum Ell. Sk. 1 : 120. A distinct species, related to Panicum.
scrilmerianiini Nash. The same as the specimen in the Herbarium of
Columbia University, so labeled by Elliott = Pa?w'ciMu oligosanthes Schialtes.

Panicum virgatum Linn. ; Ell. Sk. 1 ; 120.

Panicum amarum Ell. Sk 1 : 121. The specimen in the Herbarium of Elliott

under this name is a robust

form of Panicum virgatum

Linn. From Elliott's descrip-
tion, however, it is very evident

what plant he had in mind. The

true Panicum amarum Ell., is

represented by the following

specimens in the U. S. National

Herbarium. Virginia: Ocean

View, Geo. Vasey, August 5,

1890; Virginia Beach 2063, 2064

T. H. Kearney, jr., October 6,

1898; 3090, T. A. Williams, Sep-
tember 24, 1900. Florida: Key

Largo, A. H. Curtiss, November

7, 1894 ; Lake Worth Inlet, 5527,

A. H. Curtiss, September 2,

1895; Indian River, 3578 Cur-
tiss; Cape Malabar, Curtiss,

September, 1879. The small

form which was called Panicum

amarum minor Yasey & Scribn.

U. S. Dept. Agr. Div. Bot. Bui.

8 : 38. 1889, is here proposed as

the following species.
PANICUM AMAROIDES Scribn.

& Merrill sp. nov. {Panicum

amarttvi minor Vasey & Scribn.

U. S. Dept. Agr. Div. Bot. Bui.

8:38. 1889, not P. capillare

minor Muhl. 1817.) (Fig. 1.) A

glabrous, glaucous, non-caespi-

tose perennial, 3 to 8 dm. high

from horizontal root-stocks,

with thick, often involute

leaves and narrow, contracted

panicles. Culms erect, or ascending, often somewhat branched below, stout,
the lower internodes very short; nodes smooth; sheaths loose, glabrous,

crowded, and overlapping ; ligule a dense fringe of soft white hairs 2 to 3

mm. long; leaf -blades very thick, smooth, lanceolate, acuminate, not nar.

rowed at the rounded base, 1 to 3 dm. long, 5 to 10 mm. wide, strongly invo-
lute at least above. Panicles 1 to 2.5 dm. long, few flowered, narrow, the
branches appressed 1 to 7 cm. long. Spikelets glabrous ovate acuminate 5
to 6 mm. long ; first glume clasping the base of the spikelet, strongly nerved

Fig. \.— Panicum amaroides Scribn. & Merrill, a,
shows habit of the plant; c, a spikelet somewhat
enlarged; rt, the perfect floret showing the flower-
ing-glume, palea, stamens, and stigmas; e, ante-
rior view of the same with the palea closed; h, a
portion of panicle of P. amarum Ell.

6

PANICUM AMAROIDES Scribn. & Merrill sp. nov.— Continued.

three-fourths or sometimes as long as the spikelet, usually scabrous on the
keel near the apex; second glume slightly exceeding the third, 7-nerved
similar to the first ; third glume thin, faintly 7-nerved, inclosing the flower-
ing glume and palea, subtending a thin palea of equal length which bears a
staminate flower; flowering glume elliptical-ovate, smooth, 3 to 3.5 mm.
long. Palea as long as the glume and of similar texture, plane or nearly so.
Type specimen collected at Fortress Monroe, Va. , by Dr. George Vasey, in 1879.

General distribution: In sands
along the coast, Connecticut to
Florida and Mississippi. August
to October,

Specimens examined: Connecticut:
New Haven, J. A, Allen, Sep-
tember 30, 1876; A. L, Winton,
jr,, 1887; D, C, Eaton, no date.
Netv York: Staten Island, A. A.
Tyler, September 17, 1895 ; North-
ville, L. I,, H. W, Young, Nov.
5, 1873, Delaicare: Sand dune,
no locality, W, M. Canby, Sep-
tember. Maryland: Bay Ridge,
F, L,-Scribner, September 3, 1897,
Virginia : Fortress Monroe, Geo.
Vasey, 1879 (type) ; G, McCarthy,
1883; Ocean View, Geo, Vasey,
Aug. 5, 1890; Virginia Beach,
3089 T, A, Williams, September
24, 1900. North Carolina: Ocra-
coke Island, 3317 T, H. Kearney,
jr,, October 17, 1898; Brunswick
County, 211 G. McCarthy, August
11, 1885, 3Tississip2n : Horn Is-
land, 3854 S. M, Tracy, August
30, 1894-September 6, 1897,

This species is readily distinguished
from Panicnni amarum Ell. by
its much smaller size, smaller
panicles, larger spikelets, and
longer first glume. In its habit
of growth it is strikingly different
from Panicum amarum. The fol-
lowing notes are from the field
observations of Mr, T, A, Wil-
liams: Panicum amamm grows
in densely caespitose bunches.

Fig, ^.—Panicum ovale Ell. a, h, c, spikelets; d,
anterior view of the third glume, showing small
palea; e, dorsal view of the fourth or fruiting
glume; /, anterior view of same showing back of
palea.

from 9 to 18 dm. high, from ver-

tical root-stocks. Panicum amaroides is not at all caespitose, sending up
solitary culms from long horizontal root-stocks, and rarely attains a height
of 8 dm. At Virginia Beach, where both species were observed, Panicum
amaroides was in full bloom and at the same time Panicum amarum was in
fruit or late bloom. The former species was closely grazed wherever it was
accessible, while the latter was avoided by stock.

PANICUM AMAROIDES Scribn. & Merrill sp. nov.— Contimied.

Paniciiin amaroides is an excellent sand binder, and although not caespitose, it
covers considerable areas, holding the sands by means of its strong hori-
zontal rootstocks.

Panicum scabriusculum Ell. Sk. 1 : 131.

Panicum nervosum Muhl. ; Ell. Sk. 1 : \22=Pat)icum covivmtatina Schultes, and
is exactly matched by the following specimens in the National Herba
rium: South Carolina, Aiken, H. W. Ravenel, June 1, 18(57; Florida, A. W.
Chapman, no locality or date; 366 (in part) Chapman, no locality or date;
Jacksonville, A. H. Curtiss, September.

Panicum multiflorum Ell. Sk. 1 : l22=Panicitwi polyanthes Schultes (P. micro
ca?-j>on Muhl. Descr. 111. 1817.)

Panicum ovale Ell. Sk. 1 : 123.
{Panicum erytJtrocarpon Ashe,
Journ. E. Mitch. Sci. Soc. 16:
90, 1900 is the same. ) (Fig. 3. )
The specimen in the herbarium
of Elliott to which this label
is attached is matched by the
following specimens: North
Carolina : near Wilmington,
W. W. Ashe, June, 1899 (co-
type Panicum erythrocarpon
Ashe). Sotith Carolina: Ai-
ken, H. W. Ravenel, May.
Florida: A. H. Curtiss, 3583
N. Am. Plants (in part).

Panicum lanuginosum Ell. Sk.
Bot. S. C. andGa. 1:123. 1817.
{Panicum tennesseense Ashe,
Journ. E. Mitch. Sci. Soc. 15:
53.1898). (Fig. 3.) An erect
or ascending, finally much-
branched pubescent perennial
1.5 to 8 dm. high, with weak
culms, thin leaves and slightly
exserted, usually many-flow-
ered panicles. Culms genicu-
late below, papillate-pilose
with few weak, spreading
hairs; nodes yellow, sparingly
bearded with few reflexed
hairs, and generally with a
smooth ring immediately be-
low ; sheaths shorter than the
internodes, rather loose, striate, papillate-pilose with scattered spreading
hairs ; leaf-blades spreading or ascending, thin, soft, lanceolate, 3 to 8 cm.
long, 4 to 10 mm. wide, acute, slightly narrowed to the roimded base, scabrous
on the margins, appressed-pubescent beneath with scattered, short, papil-
late hairs, and often also on the upper surface, giving the leaves a peculiar
lustre or sheen. Panicles 4 to 8 cm. long, broadly ovate or subpyramidal,
pale green; rachis pilose; branches capillary, spreading, the lower ones often
densely flowered and interlaced, giving the panicle a characteristic matted

Fig. ■i. — Pftnicum lamidhiomm Ell. a, h, spikelets,
in a the .short first glume is seen; c, outline of ante-
rior view of the fruiting glume showing back of
palea.

8

Panicum lanuginosum Ell. Sk. Bot. S. C. and Ga. 1 : 123— Continued.

appearance; secondary panicles included, few flowered. Spikelets ovate,
obtuse, 1.5 mm. long, first glume hyaline, acute, one-fourth as long as the
spikelet ; second and third glumes, rather strongly striate, pilose with rather
long, weak, spreading hairs.

Type specimen in the herbarium of Elliott ; type locality, Georgia.

General distribution : Low thickets, woods and swamps, Maine to Illinois,
south to Kansas and Mississippi, May to August.

Specimens examined: Maine: Orono, M. L. Fernald, 1893; Hiram, 48, 49 H. W.
Merrill, 1897; East Auburn, 9 E. D. Merrill, 1898. Vermont: Brandon,
F. H. Knowlton, 1882. Massachusetts: Framington, 741, 743 E. C. Smith,
1899; Salem, J. H. Sears, 1883. Rhode Island: Providence, T. J. Battey,
1886. Connecticut: New London, C. B. Graves, 1898; Portland, 134 F.
Wilson, 1897. New York: Dresden Station, 4a C. H. Peck, 1898; Ganse-
voort, 7a Peck, 1897; Albany County, 7 Peck, 1897; Niagara County, 2, 3
E. C. Townsend. Pennsylvania: Easton, T. C. Porter, 1898. District of
Columbia: F. L.-Scribner, June 7, 1894; 33 T. H. Kearney, 1897. Virginia:
no locality, 9 T. Holm, 1888. Michigan : Belle Isle, 597d, O. A. Farwell,
1892. Tennessee: Knoxville, F. L.-Scribner 1891; La. Vergne County, 7087
Biltmore Herbariiim (co-type of Panicum tennesseense Ashe) ; Knox Coimty,
T. H. Kearney, 1892; Franklin County, 30 H. Eggert, 1897; Georgia: no
locality, Dr. Baldwin (type). Mississippi: Fairport, 3208 S. M. Tracy,
1897 Macon, 3223 Tracy, 1897. Illinois: Peoria, F. Brendel. loim: Leb-
anon, 35 Ball & Sample, 1898; Ames, 45 C. R. Ball, 1896. Missouri: Shan-
non Coimty, 746, 760 B. F. Bush, 1894. Kansas: Cherokee County, 682,
A. S. Hitchcock, 1896.

This species is characterized by its lax habit of growth, thin leaves which gen-
erally have a peculiar lustre or sheen, pale green generally somewhat impli-
cate panicles, at least in herbarium specimens, and long-pubescent spikelets.
It is most closely allied to Panicum unciphyllum. Trin. (P. puhescens of
authors not Lam. ) and is distinguished by the characters noted above. The
type specimen is exactly matched by specimens collected at Knoxville, Tenn. ,
and Brookland, D. C, by F. Lamson-Scribner, cited above. The specimen
from Elliott in the herbarium of Columbia University, noted in Bui. Torr.
Bot. Club 27 : 595, 1900, as being identical with Panicum orangensis Ashe, is
very different from the specimen in Elliott's herbarium which must be con-
sidered as the type of the species. Ashe's name {Pa?ncum orangensis)
should be retained for that species. Panicum tennesseense Ashe is only the
branched state of P. lanuginosum Ell.

Panicum viscidum Ell. Sk. \:\2Q = Panicum scoparinm Lam., see Scribn. &
Merrill, U. S. Dept. Agr. Div. Agros. Bui. 24:34. 1901.

Panicum dichotomum L. ; Ell. Sk. 1 : 124. In this cover are several unrecogniza-
ble scraps and a specimen of Panicum harhidatum Michx.

Panicum villosum Ell. Sk. 1 : 124. This is exactly identical with the form so
considered by Nash, Bui. Torr. Bot. Club, 23: 147. 1896. The synonymy of
this species is as follows:

PANICUM CONSANGUINEUM Kunth, Enum. 1:106. 1833. {Panicum vil-
losum Ell. Sk. Bot. S. C. and Ga. 1 : 124. 1817, not Lamarck, 1791; Panicum
georgianum Ashe, Journ. E. Mitch. Sci. Soc. 15:36. 1898; Panicum cahoo-
nianum Ashe, ibid., 113. 1898.)

Panicum sphaerocarpon Ell. Sk. 1 : 125.

Panicum pubescens Lam. ; Ell. Sk. 1 : 125 = Panicum pseudopubescens Nash.

9

Panicum strigosum Mulil. ; Ell. Sk. 1 : 126. The type is well matched by No.

3597 A. H. Curtiss. A specimen from Elliott in the Herbarium of Columbia

University is the same. {Panicum Jongipedunculatuni Scribn.)
Panicum ciliatum Ell. Sk. 1 : 126.
Panicum ensifolium Baldw. ; Ell. Sk. 1 : 126. (Fig. 4.) This is exactly the form

described by Ashe as Panicum cutlibertii, Joum. E. Mitch. Sci. Soc. 15:48.

1898. The grass described by Ashe,

Journ. E. Mitch. Sci. Soc. 1. c, 46.

as Panicum ensifolium Baldw., is

Panicum tenue Muhl.
Panicum barbulatum Michx. ; Ell. Sk.

1 : 127. This is very distinct from

the form considered as Panicum

harhuJatum Michx. to-day, and is

proposed below as a new species.
PANICUM SUBBARBULATUM

Scribn. & Merrill sp. nov. {Panicum

barbulatum of Ell. Sk. Bot. S. C. and

Ga. 1 : 127. 1817, not Michx. Fl. Bor.

Am. 1 : 49. 1803.

An erect or ascending, glabrous, caespi-
tose perennial, 4-9 dm. high, with
small panicles, bearded nodes, and
pubescent spikelets. Culms wiry,
simple, or becoming branched above ;
nodes bearded with reflexed hairs;
sheaths much shorter than the inter-
nodes, ciliate on the margins ; ligule
a short ciliate ring ; leaf -blades lan-
ceolate, primary ones 5-6 cm. long,
5-8 mm. wide, spreading, the second-
ary ones 3-4 cm. long, 3-5 mm. wide,
naiTowed at the base, acute, scabrous
on the margins, the nerves rather
prominent. Panicles exserted, ovate,
4-7 cm. long, the branches ascending.
Spikelets ovate, obtuse, 2 mm. long,
pubescent; first glume about one-
fourth as long as the spikelet.

The type of the species in the herbarium
of Elliott is well matched by the fol-
lowing specimens: 3600a A. H. Cur-
tiss, Sumpter County, Fla., margins
of pine-barren ponds; 3388 S. M. Tracy, June 13, 1897, Morrisonville, Miss.

This species is readily distinguished from Panicum barbulatum Michx., by its
smaller size, smaller leaves, and panicles and larger pubescent spikelets.

Panicum microcarpon Muhl. ; Ell. Sk. 1 ; 127= Panicum barbulatum Michx.

Panicum nitidum V Lam. ; Ell. Sk. 1 ; \2^.-= Panicum erectifolium Nash.

Panicum melicarium Michx. ; Ell. Sk. 1 : 128. Not in the herbarium. There is,
however, a specimen of Sjwrobolus junceus in the collection, evidently from
Muhlenberg, so labeled, but corrected to Agrostis j%i,nceiis by Elliott.

Panicum debile Ell. Sk. 1 ; \2^. = Panicum x^errucomm Muhl.

Fig. 4.- Panicum enttifoUiDn Baldw. a, b, c,
spikelets; '?, anterior view of the third
ghime, showing small palea; e, /, fruiting
glume.

10

Panicum angustifolium Ell. Sk. 1 : 129. This is well matched by 4615 S, M.
Tracy, Biloxi, Mississippi, June 14, 1898. Culm leaves 15-30 cm. long, 4-6
mm. wide, very long acuminate and hairy at the base. Panicle few-flowered.
Spikelets 3 mm. long, acute.

Panicum divergens Miahl. ; Ell. Sk. 1 : ISO = Panicum cognatimi Schultes.

Digitaria sanguinalis Ell. Sk. 1 : lZ\=^Panicuvi sanguinale Linn.

Digitaria villosa Walt. '? Ell. Sk. 1 -.1^2 = Panicum serotinum Trin.

Digitaria filiformis Ell. Sk. 1 : 132 = Panicum flliforme Linn.

Digitaria dactylon Ell. Sk. 1 : \3'3 = Cynodon dactylon Pers.

Agrostis arachnoides Ell. Sk. 1 -.134: = Agrostis elliottiana Schultes.

Agrostis tenuiflora Ell. Sk. 1 -.134:^ Muhlenhergia diffusa Schreb.

Agrostis sericea Ell. Sk. 1 •.\t5 = Agrostis hyemalis (Walt.) B. S. P. Specimen
from Schweiuitz.

Agrostis trichopodes Ell. Sk. 1 :Vdo.^=Muhlenbergia trichopodes (Ell.) Chapm.

Agrostis decumbens Muhl. ; Ell. Sk. 1 AZQ -^ Agrostis verticillata Vill.

Agrostis dispar Michx. ; Ell. Sk. 1 : 136. Not in the herbarium. Agrostis dispar
Michx. in the herbarium of the Paris Museum of Natural History, is a form of
Agrostis alba Linn. , identical with Agrostis stolonifera of European botanists.

Agrostis alba L. ; Ell. Sk. 1 : 137.

Agrostis juncea'? Michx. ; Ell. Sk. 1 : \Z1 =^ Sporoholus junceus Kunth.

Agrostis clandestina Muhl. ; Ell. Sk. 1 : 138. This species is not represented in

the herbarium.
Agrostis indica Ell. Sk. 1 : V3%-=^Spordbolus indicus R. Br.
Agrostis virginica Ell. Sk. 1 : 1 39 = S/9oro5oZMS virginicus Kunth.
Stipa avenacea Ell. Sk. 1 :139.

Stipa stricta Lam. ; Ell. Sk. 1 : 140. This species is not represented in the her-
barium.

Stipa capillaris Lam. ; Ell. Sk. 1 : 140. This species is represented in Elliott's
herbarium under the name Stipa sericea and is Mulilenhergia. capillaris Trin.

Aristida spiciformis Ell. Sk. 1 : 141.

Aristida dichotoma Michx. ; Ell. Sk. 1 : 141.

Aristida gracilis Ell. Sk. 1 : 142. 'A specimen in the herbarium with the follow-
ing label "Aristida genicidata Rafin. n. sp. '? Rafinisque, Long Island," is
the same. Aristida geniculata was published by Rafinesque in American
Monthly Magazine, 2:119. Dec, 1817.

Aristida stricta Michx. ; Ell. Sk. 1 : 142.

Aristida lanosa Mulil. ; Ell. Sk. 1 : 143.

Andropogon ciliatus Ell. Sk. 1 : \i4=^ Andropogon nutans avenaceus Hack.

Andrpogon nutans Ell. Sk. 1 : 144:^ Andropogon nutans linneanus Hack.

Andropogon ambiguus Michx. ; Ell. Sk. 1 : 14:5 ■=Gyninopogon ambiguus (Michx.)

B. S. P.
Andropogon melanocarpus Ell. Sk. 1 : 146.

Andropogon scoparius Michx. ; Ell. Sk. 1 : 146. This is in Elliott's herbarium
under the name '' Andropogon purpiirascens."

Andropogon ternarius Michx. ; Ell. Sk. 1 : \4:1 ^= Andropogon argyrceus Schultes.
Andropogon argenteus Ell. Sk. 1 : 1^-^ Andrpogon argyrceus Schultes.
Andropogan vaginatus Ell. Sk. 1 : \4^ = Andropogon virginicus L., and is under
this name in the herbarium.

11

Andropogon dissitiflorus Michx. ; Ell. Sk. 1 : 149 = A form of Andropogon glom-

eratus (Walt.) B. S. P.
Andropogon macrourus Michx. ; Ell. Sk. 1 : \'iSi = Andropogon gJomeratus (Walt.)

B. H. P.
Andropogon tetrastachys Ell. Sk. \ : \bO := Andropogon virginicus tetrastachys

(Ell.) Hack.

Andropogon furcatus Miihl. ; Ell. Sk. 1 : 150.

Aira pallens var. aristata Mn,hl. : Ell. Sk. \ ■.\ol=zTrisetum pennsylvanicum
(Linn. ) Beauv.

Aira flexuosa Ell. Sk. 1 : lol = Deschamjma Jlexiiosa (L. ) Trin.

Aira purpurea Walt. ; Ell. Sk. 1 : 153=: Triplasis purpurea (Walt.) Chapui.

Aira capillacea Lam. ; Ell. Sk. 1 : 152. This species is not represented in the
herbarium.

Aira triflora Ell. Sk. 1 : 153 = Poa, sp. indet.

Aira obtusata Michx. ; Ell. Sk. 1 : loS ^= Eatonia obtnsata A. Gray.

Aira mollis Mulil. ; Ell. Sk. 1 : l^i = Eatonia nitida (Spreng. ) Nash.

Melica glabra Michx. ; Ell. Sk. 1 : \M = Melica mutica Walt.

Dactylis glomerata L. ; Ell. Sk. 1 : 155.

Poa capillaris L. ; Ell. Sk. 1 : ir)Q = Eragrostis capillaris Nees.

Poa tenuis Ell. Sk. 1 : 156 = A very small-flowered form of Eragrostis capillaris

Nees.
Poa hirsuta Michx. ; Ell. Sk. 1 : 157. Not in the herbarium.
Poa parviflora Pursh. ; Ell. Sk. \:\f)l-=Panicularia nervata (Willd. ) Kuntze.

Poa conferta Ell. Sk. \ : lol ■= Eragrostis glomerata (Walt.) L. H. Dewey. It
is in the herbarium under the name " Poa glomerata."

Poa annua Linn. ; Ell. Sk. 1 : 158.

Poa autumnalis Muhl. ; Ell. Sk. 1 : 159.

Poa viridis Muhl; Ell. Sk. 1 : IhQ = Poa pratensis L.

Poa angustifolia Ell. Sk. 1 : 160=: Poa pratensis Linn.

Poa tenella Linn. ; Ell. Sk. 1 : \Q0 = Eragrostis jiUosa Beauv.

Poa eragrostis Ell. Sk. 1 : IQl = Ei'agrostis 2)nrsliii Schrad.

Poa nitida Ell. Sk. 1 : 10,2 = Eragrostis nitida (Ell.) Chapm.

Poa refracta Muhl. ; Ell. Sk. 1 ■.W2 = Eragrostis refracta (Muhl.) Scribn.

Poa fluitans Ell. Sk. 1 : IQS = Panicnlaria fluitans (Linn.) Kuntze.

Poa reptans Michx. ; Ell. Sk. 1 : IQS = Ei-agi-ostis hypnoides (Lam.) B. S. P.

Poa rigida Ell. Sk. 1 : \Qi^= Scleropoa rigida (Kunth) Griseb.

Poa quinquifida Pursh; Ell. Sk. ] : IM = Triodia seslerioides (Michx.) Benth.
On the sheet are also the following names: " Poa flava Linn. ; ' "Poa sesle-
rioides Mx."

Poa ambigua Ell. Sk. 1 : 165= Triodia ambigua (Ell.) Benth.

Briza eragrostis Ell. Sk. 1 ; Hi') ^ Eragrostis major Host.
Uniola paniculata Ell. Sk. 1 : 166.

Uniola spicata L. ; Ell. Sk. 1 : \(j()^= Distichlis spicata (L.) Greene.
Uniola latifolia Michx. ; Ell. Sk. 1 : 167.
Uniola nitida Baldwin; Ell. Sk. 1:167.

Uniola gracilis Michx. ; Ell. Sk. 1 :168= Uniola la.ca (L.) B. S. P.
Festuca tenella Pursh; Ell. Sk. 1:168. This species is not represented in the
herbarium.

13

Festuca polystachya Michx. ; Ell. Sk. 1 : 169. This species is not represented in
the herbarium :=Lep^oc/i/oa/ascic«Zaris (Lam.) A. Gray.

Festuca myurus L. ; Ell. Sk. 1 : 169.

Festuca parviflora Ell. Sk. 1:170. This is apparently a young, undeveloped
form of Festuca octoflora Walt., although it may prove to be distinct.
Empty glumes nearly equal ; flowering glumes nearly smooth, about 3 mm.
long ; awns about equaling the glumes in length.
Festuca duriuscula L. ; Ell. Sk. 1 : \1\ = Festuca pratensis Linn.
Festuca grandiflora Lam. ; Ell. Sk. 1 : 171. Not in the herbarium.
Festuca unioloides Willd. ; Ell. Sk. 1 : 171. Not in the herbarium.
Festuca nutans Ell. Sk. 1 : 172. In the same cover is a specimen of Festuca

obtusa Spreng.
Bromus secalinus L. ; Ell. Sk. 1 : 172.
Bromus ciliatus L. ; Ell. Sk. 1 :173.

Bromus purgans L. ; Ell. Sk. 1 : 173. Not in the herbarium.
Avena spicata Ell. Sk. 1 ■.V!4: = Danthonia spicata (L.) Beauv.
Avena pennsylvanica Ell. Sk. 1 : 174: = Arrhenatherum elatius (Linn.) Beauv.
Eleusine mucronataV Michx. ; Ell. Sk. 1 : llo = Leptochloa mucronata Kunth.
Eleusine indica Ell. Sk. 1 : 175.
Eleusine cruciata Ell. Sk. \ ■.176 = Dactyloctenium aegyptium Willd. It is in

the herbarium under the name Chloris mucronata.
Monocera aromatica Ell. Sk. \ : 177 = Camjmlosus aromaticns (Walt.) Trin.
Chloris petr^a Michx. ; Ell. Sk. 1 : 178.
Rottboellia dimidiata? Ell. Sk. \: 179 = St enotaphrum secundatum (Walt.)

Kuntze.
Elymus virginicus L. ; Ell. Sk. 1 : 180.

Elymus striatus Ell. Sk. 1 : 180 := Elymus canadensis Linn.
Elymus europaeus Ell. Sk. 1 : 181. Not in the herbarium.
Elymus hystrix Ell. Sk. 1 : 481 = Asperella hystrix (L.) Hiimb.
Panicum verticillatum L. ; Ell. Sk. 1 : Adv. 1. Not in the herbarium.
Panicum walteri Pursh. ; Ell. Sk. 1 : Adv. 1. This is the same as Panicum crus- ;

gain hispidum Ell., page 114.
Panicum fusco-rubens Lam. ; Ell. Sk. 1 : Adv. 4. Not in the herbarium.
Panicum striatum Lam. ; Ell. Sk. 1 : Adv. 4. Not in the herbarium.
Panicum diffusum Swartz. ; Ell. Sk. 1 : Adv. 4. Not in the herbarium.
Panicum nodiflorum Lam. ; Ell. Sk. 1 : Adv. 5. Not in the herbarium.
Agrostis pauciflora Pursh. ; Ell. Sk. 1 : Adv. 5. Not in the herbarium.
Agrostis cinna Ell. Sk. 1 : Adv. 5=Cmna latifolia Griseb.

Agrostis lateriflora Michx. ; Ell. Sk. 1 : Adv. 6. The specimen in this sheet is
Calamagrostis canadensis (Michx.) Beauv.

F. Lamson-Scribner,

Agrostologist.
Approved :

James Wilson,

Secretary of Agriculture.

Circular No. 30. - (Agros. 79.) Issued March 8, 1901.

United States Department of Agriciiltupe>

DIVISION OF AGROSTOLOG^<^*'^V»* * ^ ^'^
[Grass and Forage Plant Investigations^*^ ». "* ^ VO r^lT

NEW OR LITTLE KNOWN (aiASSES.

The following descriptions are of new or little known species of
grasses which have appeared in recent collections from several
sources and their publication as a circular of this Division is recom-
mended. The notes on Danthonia intermedia are by the late
Prof. Thomas A. Williams, and were made just before his death.

Setariopsis latiglumis (Vasey) Scribn. Field Columb. Mus. Bot. Ser. 1:289.
PI. XI. December, 1896 [Setaria latiglumis Vasey, Bui. Torr. Bot. Club,
13:229. 1896).

Culms 1 m. high; nodes bearded with erect, appressed hairs; sheaths, at least
the lower ones, papillate-pilose, long-hairy at the throat ; leaves thinly hairy
or pilose on the upper surface, the lower ones pilose on the back, at least
towards the base which is long dentate; the broader leaves 1.5 to nearly 2
cm. wide, long-acuminate-pointed and gradually tapering to a very narrow
base. Panicle about 30 cm. long, rather densely flowered ; branches erect,
appressed, the lower ones 4-5 cm. long.

This description is drawn from No. 8412. C. G. Pringle, 1900, collected in Iguala,
Guerrero, Mexico. Pringle's specimens are much more robust and coarser
than the type, No. 117a, E. Palmer, 1885. In Palmer's plants the leaves are
5-10 mm. wide, and the comparatively few flowered panicles 6-10 cm. long.
The spikelets, however, in Palmer's plant are somewhat larger than those
in the specimens above described. There is such a close resemblance, how-
ever, in the character of the essential parts that I must regard them as rep-
resenting a single species.

ICHNANTHUS APICULATUS Scribn. sp. nov.

A slender, wiry, ascending perennial, more or less extensively creeping and root-
ing at the lower joints, with flat, lanceolate, acute leaves 3-8 cm. long, and
loosely flowered, spreading panicle 14-20 cm. long, the base partially included
or barely exserted from the uppermost leaf sheath. Culms striate, glabrous
or with a pubescent line along one side ; nodes finely and shortly pubescent ;
sheaths striate, ciliate along the margin, papillate-pilose towards the apex
and pubescent at the summit on the back, where it is somewhat contracted ;
ligule very short, the margin shortly and finely ciliate; leaf blade 4-8 mm.
wide, rounded at the base and gradually tapering from near the middle to
the very shai-ply acute apex, sparingly papillate-pilose on both surfaces or
nearly glabrous, margins minutely and sharply scabrous and with a few long
hairs near the base; lower panicle branches ascending, 8-12 cm. long, the
upper branches more spreading, gradually becoming shorter. Spikelets on
rather slender pedicels, often in pairs— one short pedicellate, the other sup-
ported on a longer pedicel — oblong-ovate in outline, 3-3.5 mm. long, about
0.5 mm. broad, obtuse. Outer glumes prominently nerved, glabrous — the
first, 3-nerved, obtuse, two-thirds to three-fourths as long as the spikelet;

2

second and third glumes 5-nerved, broadly ovate, oblong, obtuse, minutely
pubescent at the apex, slightly exceeding the fourth or flowering glume,
which is about 2.5 mm. long, contracted at the base and depressed on the
back, the obtuse and somewhat hooded apex shortly apiculate. Palea glab-
rous, about as long as the glume which it resembles in color and texture.
Type specimen from near Jalapa, State of Vera Cruz. C. G. Pringle, No. 9208,
May 13, 1900. Allied to Ichnanthus lanceolatus Scribn., from which it is at
once distingiiished by its narrower, sessile leaf blades, more obtuse and
somewhat shorter spikelets, and distinctly apiculate fourth glume.

Agrostis nana (Presl) Ktmth, Enum. 1:266. 1833. {Tricodium nanmn J. S.
Presl in C. B. Presl Reliq. Haenk. 1: 243. 1880.) On cliffs near Fort
Bragg, Mendocino County, Cal., No. 6159, Jos. Burtt Davy and Walter C.
Blasdale.

The type locality of this species is doubtfully given by Presl as Peru. The
original specimens might, however, have come from the California coast,
where Haenke also made collections. Agrostis nana Kunth, as represented
by specimens in the herbarium of the Missouri Botanical Garden and illus-
trated by figure 2, plate 84, in the Tenth Annual Report of Missouri Botani-
cal Garden, is a much smaller plant than that collected by Davy and
Blasdale ; but the habit of growth, and size and character of the spikelets are
identical, and the very full description given by Presl applies so closely to
our plant that there can be little doubt of its identity. The characters of
the Mendocino plant are as follows :

A densely caespitose, wiry, glabrous perennial 1-2 dm. high, with involute
filiform leaves and strict, few-flowered panicles 2-4 cm. long. Ligule 1.5 mm.
long, cleft in three divisions, the lateral divisions auriculate. Leaf -blades of
the culm about 2 mm. wide. The axis of the panicle and the closely
appressed branches are 1 cm. long or less, strongly scabrous. Spikelets on
short, rigid, appressed, scabrous pedicels ; outer glumes nearly equal or the
first a little longer than the second, about 3 mm. long, scabrous on the back,
especially along the keel, broadly lanceolate and acute ; flowering glumes
about 1.8 mm. long, oblong truncate, and erose dentate at the apex, awnless
or very short-awned just above the middle. Palea minute or wanting,
callus short, obtuse.

The only manifest difference between our plant and that of Presl is in the veg-
etative characters which are most likely to vary.

Agrostis pringlei Scribn., U. S. Dept. Agr. Div. Agros. Bui. 7:156, fig. 13S.
1897. Along cliffs at Point Arena, Mendocino Coiinty, Cal. , No. 6030, Davy
and Blasdale. 1899.

These specimens represent a multi-branched form of the species, with panicles
more densely flowered than in the type, but the essential characters of the
spikelets, root habits, and foliage are those of Agrostis pringlei. This species
is remarkable for the long hairs at the base of the floret. These hairs are
1-2 mm. long.

AGROSTIS VIRESCENS MICROPHYLLA (Steud.) Scribn. n. comb.

{Agrostis viicrophylla Steud. Syn. PL Gram. 164. 1854). Dry places in

meadows near Sherwood Valley, Mendocino County, Cal., No. 5142, Davy

and Blasdale, 1899.
Culms very slender, 1-1.5 dm. high, apparently annual. Leaves 1-2 cm. long;

ligule about 2 mm. long. Panicle densely flowered, 1-3 cm. long, purplish.

Empty glumes lanceolate, scabrous, long acuminate, subulate pointed ; the

first 3.5-4 mm. long, little exceeding the second; flowering glumes 1.5 mm.

long, scarcely one-half the length of the outer glumes, awned on the back near

the middle. Apex bifid, each division minutely two-toothed by prolonga-
tion of the lateral nerves. Awn ascending from near the middle of the
glumes on the back, 4-4.5 mm. long, rather stout below, geniculate and
projecting beyond the glumes.

While these plants are very much smaller than Agrostis virescens H. B. K. , the
differences are in the vegetative rather than in the essential characters.
A. virescens in the typical form is 20-54 cm. high, with a more open panicle,
longer and broader leaves, but the spikelet characters of the variety here
noted, are so closely identical with those of the species that the plant can
not be regarded as representing anything more than a variety. Specimens
here described have the appearance of being annuals or seedlings, and this
may account for their diminutive size.

AGROSTIS DAVYI Scribn. sp. nov.

A rather slender, erect, or ascending perennial, 6-9 dm. high, with narrow,
scabrous leaves and loosely flowered, narrow panicles 1-2 dm. long. Culms
and sheaths scabrous, ligule about 10 mm. long, hyaline, scabrous on the
back, at least below, decurrent. Leaf -blades 8-18 cm. long, 3-6 mm. wide,
scabrous on both sides. Spikelets 3.5-4 mm. long, outer glumes lanceolate,
acuminate, very acute, scabrous on the back, especially along the keel,
nearly equal; flowering glumes about 3 mm. long, oblong, trimcate,
5-nerved, lateral nerves projecting as very short teeth or setae, the mid-
nerve usually excurrent at or above the middle as a short, slender, incon-
spicuous bristle. Callus hairy on the sides, the hairs 1-1.5 mm. long.

In brush on hillsides near Point Arena, Mendocino Coimty, Cal., No. 6062 Davy
and Blasdale. May-August, 1899.

The spikelets of this species closely resemble those of Agrostis pringlei Scribn. ,
and this resemblance is carried to the remarkably long calhis hairs, liut in
other respects the species are very distinct. The culms in Agrostis davyi
are simple, and much taller, and the panicles longer and more loosely
flowered.

STIPA LEMMONI (Vasey) Scribn. n. comb. {Stipa pringlei var. lemmoni
Vasey, Contr. U. S. Nat. Herb. 3:55. 1892). Near Harris, Humboldt
County, Cal., No. 5348, Davy and Blasdale; dry rocky slopes. Long Valley,
No. 5287, Davy and Blasdale.

A rather slender, rigid perennial 4-6 dm. high, with narrow, involute leaves and
strict, few-flowered panicles 10-12 cm. long. Culms glabrous ; sheaths stri-
ate, glabrous, shorter than the internodes. Ligule 1.5 mm. long, roiinded-
obtuse, entire, decurrent. Leaves 2-3 mm. wide, 5-10 cm. long (those of
the innovations longer and narrower), glabrous beneath, pubescent and
strongly striate above, closely involute when dry, divergent at the apex.
Panicle branches erect, the longer lower ones in twos and threes, 2-3 cm.
long, few-flowered. Spikelets rather large and pale green or straw-colored ;
empty glumes nearly ecpial, about 12 mm. long, broadly lanceolate, long
acuminate-pointed, 3-5-nerved, scarious excepting the nerves; flowering
glume oblong, about 7 mm. long, obtuse, 2-toothed at the oblong apex, thinly
pilose all over with appressed hairs. Callus broadly obtuse, shortly bearded.
Palea broad, nearly as long as the glume, and similarly hairy. Anthers naked.

This grass has been referred to Stipa viridula, but it is very distinct from that
species in its larger spikelets, longer palea, fewer-flowered panicles, fewer,
shorter, and narrower leaves. In addition to the above-cited specimens, this
species is represented in the National Herbarium by specimens from West
Klickitat County, Columbia River, Washington, No. 146, W. N. Suksdorf,
May 16, 1885, and the type No. 5456, J. G. Lemmon, Mohawk Valley, Plumas
County, Cal., May, 1889.

I

STIPA LEMMONI JONESII var. nov. A small form of the species with rather
more slender culms and panicles, and spikelets, with the outer glumes 7-8
mm. long and flowering glumes about 6 mm. long. This variety is repre-
sented in the National Herbariimi by Nos. 3563, L. Schoenfeldt, collected at
Laguna, Cal., June 4, 1894; 49, J. W. Blankinship, collected at Mount San
Hedrim, Mendocino County, Cal. ; 3298, Marcus E. Jones, collected at Emi-
grant Gap, Cal., June 28, 1882 (type). There is also a specimen from Mari-
posa County, Cal., collected by J. W. Congdon, June 4, 1897.

BOUTELOUA PRINGLEI sp. nov.

A rather slender, apparently csespitose and erect perennial 10-14 dm. high, with
flat, pilose leaves, and many, usually spreading or reflexed spikes scattered
along the common axis forming a somewhat one-sided raceme 10-30 cm. long.
Culms and nodes glabrous, sheaths papillate-villous, at least above, with long,
lanate hairs ; lower portion of the sheaths usually glabrous ; ligule very short.
Upper surface of the leaves strongly papillate-pilose along the nerves, less
strongly so on the under surface. Spikes 30 to 50 or more, approximate along
the common axis, the lower ones somewhat remote, axis of spikes 0.5-3 cm.
long, bearing 6-20 spikelets, apex more or less deeply cleft, the divisions
subulate ; these axes sometimes branched again, the secondary axes bearing
2-3 spikelets. Spikelets 4-5 mm. long ; outer glumes very unequal, the first
nearly subulate, 3-4 mm. long, the second lanceolate-oblong, 4-5 mm. long,
short-awned at the apex, silky hairy on the back, strongly 1 -nerved, hairs 1
mm. long or more ; flowering glumes nearly 4 mm. long, 3-nerved, 3-toothed
at the apex, the teeth about equal in length, the lateral ones awn-like, rather
densely pilose hairy on the back, especially along the margins. Palea as long
as its glume, 2-toothed, finely pubescent on the back. Rudiment very small,
usually reduced to a mere pedicel. Mountains of Iguala, Mexico, No. 8374,
C. G. Pringle, 1900.

This species is very closely allied to Bouteloua curtipendula, from which it is at
once distinguished by its densely villous-hairy sheaths, and pilose outer and
floral glumes.

BOUTELOUA HIRTICULMIS Scribn. sp. nov.

An erect or ascending, caespitose perennial, 4-6 dm. high, with long leaves, hir-
sute culms, and 2 to 4 spreading, often reflexed, densely flowered purple
spikes 3-5 cm. long. Culms simple, striate, very densely hirsute below with
spreading hairs, about 2 mm. long, the upper portion of the culm glabrous
or nearly so ; nodes bearded with few short appressed hairs, or nearly glab-
rous; sheaths usually shorter than the internodes, striate glabrous, or the
lower ones, which are crowded, with few scattered hairs ; ligule a very short
ciliate ring; leaf -blades linear, firm, 1-2.5 dm. long, 3-4 mm. wide, glabrous
beneath, usually with few scattered papillate hairs on the upper surface,
scabrous on the margins. Panicle exserted, axis sparingly pilose; spikes
with the short strongly-bearded pedicels, the rachis continued beyond the
spikelets, the continuation 1.5-2 cm. long, awn-like. Spikelets numerous,
crowded ; empty glumes lanceolate, acuminate, the first thin, smooth, about
3 mm. long, the second about 5 mm. long, with a row of dark or black
glands on each side of the mid-nerve, each emitting a long hair ; flowering
glume pilose with appressed scattered hairs, about 5 mm. long, 3-lobed, the
lobes awn-pointed, the middle one 3 mm. long, the lateral ones about 2 mm.
in length ; sterile rudiment on a short x^edicel, naked, consisting of three
imperfect glumes and three scabrous awns.

Type specimen collected in the Sierra de San Francisquito Mountains, Lower
California, No. 11 T. S. Brandegee, September 29, 1899; same locality, No.
30 T. S. Brandegee, October 18, 1890.

The following specimens are also referable to this species, although they differ
in several minor respects from the type: Chipias, Mexico, No. 3131 (in part)
E. W. Nelson, September, 14, 1895; Oaxaca, No. 1259 E. W. Nelson, Sep-
tember 38, 1894; Mexico, No. 201, E. Palmer, 1896.

This species is closely related to Bouteloua hirsuta (H. B. K. ) Lag., but is at
once distinguished by its usually greater size, more numerous spikes, longer
leaves, and especially by its densely hirsute culms.

DANTHONIA AMERICANA Scribn. nom. nov. {Danthonia grandiflora
Philippi, Anal. Univ. Chil. 568 (1873). Not Hochst. ex. A. Rich. Tent. Fl.
Abyss. 2:418. 1851.

A slender, densely caespitose perennial 2-4 dm. high, with short, slightly inflated
sheaths, narrow mostly involiite leaves and simple panicles of 1-4 large
spikelets. Culms and nodes very smooth ; sheaths much shorter than the
internodes, pilose-pubescent, at least the lower ones, long bearded at the
throat ; culm leaves 3-7 cm. long, 1-2 mm. wide, mostly involute, the lower
ones sparingly pilose ; leaves of the innovations involute-filiform, 8-10 cm.
long, pilose with long spreading hairs. Axis of the panicle and pedicels
puberulent. Spikelets 12-18 mm. long; empty glumes as long as or slightly
exceeding the florets, lanceolate-acuminate, the narrowed apex obtuse,
7-9 nerved ; flowering ghame exclusive of the awn and teeth 6-8 mm. long,
rounded and glabrous on the back, densely silky villous along the margins
from just above the base to a little above the middle, the hairs crowded in
little tufts, broadest above the middle, abruptly 2-toothed, teeth very slen-
der, bristle-form 4-6 mm. long; callus narrow, about 1 mm. long, barbate,
the hairs from 0.5 to nearly 3.5 mm. long; awn very slender, twisted below,
7-9 mm. long. Palea rather broad, about 6 mm. long, shortly ciliate on the
keels.

This si^ecies is distinguished from D. iinispicata with which it has been con-
founded, by its shorter and narrower leaves, the presence of 2-3 or more
spikelets in each culm, comparatively shorter and less acuminate outer
glumes, broader and more abruptly acuminate flowering glumes, the teeth
of which are much longer and more slender, and in the more abundant hairi-
ness of the flowering glumes.

Specimens examined : California : Walkers Valley, Mendocino County, May 25,
1899, "very abundant," No. 5038 J. Burtt Davy; Kawah meadows, Tulare
County, altitude 9,300 feet. No. 5347 C. A. Purpus. In this specimen the
spikelets are only 10 mm. long; Silver Lake, Amador County, Hansen,
June 30, 1899. Oregon: Cougar Peak, August 8, 1896, No. 189 Coville and
Leiberg. Washington : Falcon Valley, in low grounds, July 3, 1885, No.
150 W. N. Suksdorf. In these specimens the outer glumes are IS mm.
long, a little longer than in the type and the callus hairs are a little longer
but the plants are otherwise the same. Near Montesano, Chehalis County,
June 8, 1898, No. 3908 A. A. and E. Gertrude Heller. There is also a speci-
men in the National Herbarium from Washington, collected by Suksdorf
in 1880. Canada : Crevices of rocks, Nanaimo, Vancouver Island. No. 39
J. Macoun, June 13, 1887. Chile: Province of Nuble, No. 256 Herb. Philippi.

DANTHONIA THERMALE Scribn. sp. nov.

A rather slender, erect, densely caespitose perennial with narrow, more or less
pilose leaves and densely few-flowered panicles 3-5 cm. long. Sheaths, at
least the lower ones, papillate pilose; ligule a dense fringe of short hairs,
those near the margin 3-3 mm. long. The lower leaves, especially those of
the innovations conspicuously pilose, the upper surface rather densely hairy.
Branches of the panicle 1-3-flowered, erect, pedicels scabro-pubescent. Outer
glumes nearly equal, 12-13 mm. long, 3 mm. wide, 5-7-nerved, apex acute;

6

flowering glumes, including the very sharp-pointed or subulate teeth, 5 mm.
long, thinly pilose, hairy on the back, more densely so on the margin. Cal-
lus with a few short hairs. Awn about 7 mm. long. Teeth at the apex of
the flowering glume subulate, 1.5-2 mm. long.

Type specimen No. 6140, Aven Nelson and Elias Nelson, collected on the dry
slopes about the geyser formations in Norris basin, Yellowstone Park, July
25, 1899. Same locality. No. 169 E. D. Merrill, August, 29, 1900. Other
specimens examined, apparently belonging to this species, are No. 307,
"Williams and Griffiths, collected near the edge of Lolo Hot Springs, Mon-
tana, September 17, 1898. This is a small form, 12-24 cm. high. No. 449,
David Grifiiths, collected in abundance in ravines near Sundance, Bear
Lodge Mountains, Wyoming, August 9, 1897 ; No. 896, David Grifiiths, Sun-
dance, Bear Lodge Mountains, Wyoming, 1898; No. 982, David Griffiths,
collected in clearings in parks. Bear Lodge Lountains, Sundance, Wyoming,
July 21, 1898

This species is most closel> related to Danthonia spicata of the East, from which
it may be distinguished by its more rigid and erect habit, larger spikelets,
broader and more conspicviously nerved empty glximes, and larger flowering
glumes, the teeth of which are much prolonged and subulate iiointed. The
flowering glumes are also much more conspicuously hairy above on the back
and margins.

In the latter part of August, 1900, I found this grass growing on geyser forma-
tions in the Norris basin, doubtless in the same locality as that from which
the type was collected. It grew in scattered tvifts in the white shelly depos-
its close to the geyser, where one would scarcely expect to find any vegeta-
tion at all, and, in fact, where there was little else than this grass to be
found. The specimens from Montana, above cited, were collected within
two feet of Lolo Hot Springs. The differences in the spikelets between this
plant and Danthonia spicata of the East, together with the geographical
range and peculiar habitat, are our reasons for distinguishing it as a species.

Danthonia intermedia Vasey.

This species was described by Dr. Vasey ^ from specimens collected at Mount
Albert, Lower Canada, July 26, 1881, by O. D. Allen, and the original
description reads in part as follows :

"Radical leaves, 6 to 10 inches long, narrow; cauliue leaves, 2 to 6 inches,
pubescent, especially on the sheaths."

The material in the National Herbarium shows wide variation in the character
of the pubescence. In the type specimen the long white hairs are but
sparsely scattered on the sheaths and under surface and margins of the
leaves. In other specimens (Griffiths, 486, and Henderson, 3064b) the pubes-
cence is dense and shaggy, while on the other hand specimens not otherwise
distinguishable (Henderson, 1315, 3064a; Williams, 2770) are quite destitute
of hairs, except at the throat of the sheaths and on the leaves of the sterile
shoots, and in the latter instance they are very scattering and easily over-
looked. The species also varies considerably in the width of the leaves and
size of the spikelets.

Specimens examined: Canada: Mt. Albert, O. D. Allen, July 26, 1881; (type)
No. 16 and 31, J. Macoun, August 26 and 27, 1882; mountains north of Grif-
fin Lake, British Columbia, 46a., J. Macoun, August 5, 1899; Milk River
Ridge, Alberta, No. 13070, J. Macoun, July 19, 1895; head of Qu'Appelle,
Great Plains, No. 120, J. Macoun, July 11, 1879.

Washington: Stevens Pass, Cascade Mountains, No. 760, Sandberg & Lei berg,
August 16, 1893 ; Loomiston, No. 553, A. D. E. Elmer, August, 1897.

' Bui. Torr. Bot. Club. 10 : 52. 1883.

Oregon: Mt. Hood, No. 1315, L. F. Henderson, August 24, 1884.

California: No. 6104, H. N. Bolander; No. 1866, J. G. Lemmon, Tnolomne River,

Yosemite National Park, August, 1897.
Idaho: Beaver Canyon, No. 302, C. L. Shear, June 27, 1895, and No. 2057, P. A.
Rydberg, June 27, 1895 ; De Smet Mission, Coeur d'Alene Mountains, No.
1008, J. B. Leiberg, June 21, 1895; Salmon River near Petit Lake, No. 3568,
L. F. Henderson, July 30, 1895; without locality, Nos. 3064a and b, L. F.
Henderson, 1895, representing both the smooth (No. 3064a) and the hairy
(30646) forms of the species.
Montana : Lower basin, Gallatin River, J. W. Blankinship, July 8, 1898.
Wtjoming: Crazy Woman Creek, Big Horn Mountains, No. 2770, T. A. Williams.
August, 1897 ; Bear Lodge Mountains, near Sundance, No. 486, D. Gr.fftths,
August 10, 1897.
Colorado: No. 404, J. Wolf, 1873; Silverton, No. 1241, C. L. Shear, August 6,
1897 ; Red Dirt Divide, Steamboat Spring Road, Routt County, No. 1354,
Shear & Bessey, July 31, 1898.
The gi-ass is usually found in rather dry meadows and on mountain sides, reach-
ing an elevation of over 3,000 m. towards the southern limits of its range
and descending to 700 m. toward the northern limits.
DANTHONIA INTERMEDIA CUSICKII T. A. Williams, var. nov.
Distinguished from the type of the species by its larger size, longer, flatter,
softer, less scabrous leaves quite destitute of pubescence, except occasionally
on the younger leaves of the sterile shoots, loose, glabrous, shining sheaths,
rather larger spikelets and usually more open panicle.
Type specimen No. 2427, Wm. C. Cusick, Oregon, 1899, growing in partial shade
at an altitude of about 1200 m. Other specimens referable to this variety
are No. 2047, Cusick, head of Dirt Creek, Southern Blue Mountains, Oregon,
July 19, 1898; No. 2244, P. A. Rydberg, Mystic Lake, Montana, July 25, 1895;
No. 586, C. L. Shear, and No. 2338, P. A Rydberg, Beaver Canyon, Idaho,
August 7, 1895.
This form apparently grows in moister situations than the species.
DANTHONIA SPICATA LONGIPILA Scribn. & Merrill, var. nov.
A slender form, 3 to 4 dm. high, with small, few-flowered panicles and smaller
spikelets than in the species. Leaf blades very narrow, involute, pilose
throughout with scattered ascending hairs about 2 mm. in length. Basal
leaves 1 to 2 dm. long, those of the culm shorter. Empty glumes slightly
unequal, the fii-st 8 to 9 mm. long, the second about 1 mm. shorter. Flow-
ering glumes 2 to 2.5 mm. long, pilose.
Type specimen collected in Benton County, Arkansas. No. 38, E. N. Plank.
Nos. 40, 62, and 97, Plank, from the same locality, are the same. This vari-
ety is distinguished from the species by its smaller spikelets and pilose
leaves.
DANTHONIA EPILIS Scribn. nom. nov. Danthonia glabra Nash, Bui. Torr.
Bot. Club, 24 : 43. 1897. Not Dantlwnia glabra Philippi Anal. Univ. Chile,
Pi. Neuv. Chilenas, 94:30. 1896.
DESCHAMPSIA CURTIFOLIA Scribn. nom. nov. Deschampsiabrachyphylla
Nash, Mem. N. Y. Bot. Gard. 1 : 37. 1900. Not Deschampsia brachyphylla
Philippi Anal. Univ. Chile, PI. Neuv. Chilenas, 94:23. 1896; Deschampsia
brevifolia R. Br. App. Parry's Voy. 291. 1821. Not Aira brevifolia Bieb.
Fl. Taur. Cauc. 3:63. 1819.
CALAMAGROSTIS PERPLEXA Scribn. nom. nov. Calamagrostis nemo-
ralis Kearney U. S. Dept. Agr. Div. Agros. Bui. 11:26. 1898. Not Cala-
magrostis nemoralis Philippi Anal. Univ. Chile, PI. Neuv. Chilenas 94 : 18.
1896; C. Porteri of Vasey, Dudley Cayuga Fl. 125. 1886. Not A. Gray.

8

I

CALAMAGROSTIS LUCIDA Scribn. nom. nov. Calamagrostis laxiflora
Kearney U. S. Dept. Agr. Div. Agros. Bill. 11: 34. 1898. Not Deyeuxia
laxiflora Philippi Anal. Mus. Nac. Chile Bot. 84. 1891 ; Calamagrostis lax-
iflora Pliilippi Anal. Univ. Chile PI. Netiv. Chilenas 94 : 18. 1896.

Agrostis pseudointermedia Farwell, Ann. Rept. Com. Parks and Boulevards,
Detroit, Mich., 11:46. 1900. Agrostis intermedia Scribn. Tenn. Agr. Exp.
Sta. Bui. 7^:76. 1894. Not Agrostis intermedia Balb. Elenco, 85. 1802.

TRISETUM SHEARII Scribn. nom. nov. Trisetum argenteum Scribn. IT. S.
Dept. Agr. Div. Agros. Bui. 11:49. fig. S, July 20, 1898. Not Trisetum
argenteum R. & S., 1817, nor Schur 1860. Named for C. L. Shear, who col-
lected the grass in Las Animas Canyon near Silverton, Colorado in 1897
(No. 1214).

MELICA IMPERFECTA PUBENS Scribn. var. nov. A slender form 9-12
dm. high with the sheaths and lower surface of the leaves below strongly
pubescent with spreading or sometimes matted hairs.

Type specimen collected by T. S. Brandegee, No. 64, Santa Cruz Island, Cali-
fornia, 1888.

MELICA CEPACEA (Phil.) new comb. {Festuca cepacea, Phil. Linnsea 33:
297. 1864-65; Melica acuminata Bolander Proc. Calif. Acad. 4:104 1870.
Calif, and Chili.

PANICULARIA NERVATA PARVIGLUMIS Scribn. & Merrill, var. nov.
A lax, glaiacous grass 8-9 dm. high, with long thin leaves, open pale green
panicles and very small spikelets. Leaf -blades 1-3 dm. long, about 5 mm.
wide, thin, glaucous, scabrous above and on margins. Panicle 1-2 dm. long,
the lower branches often 1 dm. long spreading or ascending. Spikelets 2-2.5
mm. long, generally about 1 mm. wide, 3-5 flowered; empty glumes thin,
triangular, the first 0.5 mm. long, the second nearly 1 mm. long; flowering
glumes about 1.3 mm. long, 0.5 mm. wide, obovate, obtuse, or truncate,
strongly nerved.

Type specimen collected in rich low woods, Racine, Wisconsin, No. 36 S. C. i
Wadmond, June 26, 1900. The same form has also been collected on the '
county line, Racine County, by Mr. Wadmond, July 14, 1900. This variety
is distinguished from the species by its smaller and more lax flowered and
narrower spikelets, shorter and narrower flowering glumes.

F. Lamson-Scribner,

A(/r()sf<>I()(/>sf.
Approved :

James Wilson,

Secretctrij.

Circular No. 31.

Agros. 82.

United States Department of Agricnltnre,

DIVISION OF AGROSTOLOGY,

[Grass and Forage Plant Investigations.]
F. LAMSON-SCRIBNER, Agrostologist.

BERMUDA GRASS.

Bermuda grass {Cynodon dactylon) is well known throughout
the Southern States. It is a native of tropical regions of the East,
and was introduced into
the United States at an
early period. It has since
spread over the region from
Maryland to Missouri and
Texas, and is locally abun-
dant from New Mexico to
southern California. Al-
though its name would in-
dicate that it came from
the Bermudas, it is well
known in Europe, and is
thought to have originally
come from Southern Asia.
Bermuda grass is said to
have been first noticed in
this country about 1825 by
General Bethune, of Geor-
gia, who planted it in many
places throughout the
South.

There are many local
names for Bermuda,
among which are reed
grass, scutch grass, Ba-
hama grass, and, in the
region of Washington,
wire grass. In Australia

it is called couch grass. fig. l.--Bennudagra.ss(C'|/Hod<./i tktc(y<o»o.

It is a standard grass in the South, but can not be grown success-
fully north of Virginia and Oklahoma.

DESCRIPTION.

Bermuda grass is a low perennial grass, spreading extensively by-
creeping stems. These stems may be on the surface of the soil, or
commonly more or less buried, sometimes to the depth of several
inches. Under favorable circumstances they may extend 5 or 6 feet
with lateral branches of a foot or more. At intervals of an inch or
two, roots are produced, and usually a leafy stem is thrown up to
the height of a few inches. The flowering stems are upright, naked ,
above, and have the flowers in slender, one-sided spikes at the sum- 1
mit. (See fig. 1.) These spikes are from 1 to 2 inches in length,
and are in clusters of four or five, although there may be more or
fewer, according to conditions under which the grass grows.

Where it obtains a foothold, Bermuda grass spreads with rapidity,
and in exposed situations tends to drive out other vegetation. It
does not thrive in the shade, but will endure great extremes of heat
and drought. It adapts itself to a great variety of soil conditions,
growing on sand, clay, black loam, or even on strongly alkaHne soils,
and will endure a large amount of moisture or even inundation. It
does not usually produce fertile seed in the United States, except in
the extreme South. The seed upon the market is mostly imported
from the West Indies or other tropical regions. Professor Toumey
reports that it seeds abundantly in Arizona, and occasional plants
with apparently mature seeds have been found as far north as Phil-
adelphia.

ST. LUCIE GRASS.

This is a variety of Bermuda grass which is much used in Florida
and somewhat elsewhere as a lawn grass. It differs from Bermuda
in having the propagating stems more upon the surface of the soil
and in the lighter green color of its foliage. It is said to be more
resistant to frost and to keep green in winter longer than Bermuda.
It is reported to have withstood a temperature of 10" below zero in
Tennessee and to remain green through heavy frosts. This variety
has been grown successfully upon the grounds of the Department of
Agriculture, surviving the severe winter of 1898-1899.

BERMUDA AS A PASTURE GRASS.

Bermuda is the most valuable of all the grasses for pasture in the
South. It will stand trampling of stock, is very nutritious, and
thrives on soils too poor for the successful cultivation of other crops.
It is preeminently a summer grass, the length of its season depend-
ing upon the latitude. In Mississippi it furnishes grazing from the
middle of May to the middle of November. In the Gulf States,
where grazing is desired through the entire season, it is recommended

to combine Bermuda with bur clover {Medicago maculata). In
this case the Bermuda sod is scarified about September 1 with a cut-
away or disc harrow, and 15 to 20 pounds of bur clover sown per
acre. The clover ^rows during the winter and disappears in the
spring when the Bermuda appears.

In California Bermuda grass has obtained a foothold in the south-
ern part of the State and thrives on all kinds of soil. Mr. Leckenby
reports that it furnishes feed during nine months of the year, and
recommends it for land not suited for other purposes. When other
grasses are mixed with Bermuda, these are likely to be replaced by
the latter when subjected to continued grazing, especially on poor
soils. Professor Tracy, of Mississippi, states that Bermuda and
Japan clover should be the foundation of pastures, especially upland,
through the Gulf States. On the black soils of Mississippi and Ala-
bama he recommends that sweet clover be added. In the course of
a few years a Bermuda pasture becomes somewhat sod-bound. To
renovate such a pasture and keep it in good condition, it should be
plowed and harrowed in the spring every three to five years.

BERMUDA GRASS FOR HAY.

On fertile soil the growth becomes very luxuriant, and may reach
a height of 2 feet or more. It can be cut two or three times during
a season, and yields a nutritious hay of high feeding value. The
yield under favorable conditions may be as much as two to four tons
per acre, and even as high as ten tons during the season is reported.

The following treatment is recommended by Professor Tracy:

"After the last cutting in the fall, plow the land and sow with oats or vetch,
or a mixture of the two. The soil should be thoroughly harrowed both before
and after the sowing, and if possible smoothed off with a heavy roller, in order
to give a level surface for mowing. The oats and vetches give a crop of hay in
May, and by October the Bermuda may be cut."

Red clover is often sown when Bermuda is first planted, in order
to increase the yield of hay. Like other grasses, it responds readily
to the application of stable manure or other fertilizers.

FORMATION OF PASTURE OR MEADOW.

The grass may be started from seed or cuttings of the creeping
stems. To start a pasture from seed the ground should be carefully
prepared and sown in early spring with good seed at the rate of 6 to
8 pounds per acre, and pressed in with a roller. If sown just before
a rain the rolling is unnecessary. The seeds are small (about 118,000
to the ounce), and should not be covered too deeply. Trials on the
Potomac Flats at Washington, D. C, using 20 pounds of seed per
acre, gave excellent results. Less quantity of seed may be used, but

the stand is not so likely to be complete. On account of the high
price of seed, and the necessity of a thorough preparation of the
soil, pastures and meadows are more often started from cuttings.
To prepare cuttings the sod is gathered and cut into small pieces
with a feed cutter or other similar machine, or a wooden block and
hatchet can be used if only a small quantity is needed. Since most
of the propagating stems are near the surface, it is necessary to
shave off a layer of sod only an inch or two thick. If cuttings are
wanted in large quantities, the sod can be plowed and the roots har-
rowed into windrows or piles. In all cases care should be taken not
to allow the roots to get dry. The cuttings may be planted at any
time of the year in the South except the coldest winter months, but
the work is usually done in March. If a meadow is desired, more
care should be taken in the planting of the cuttings to insure a level
surface for the mowing machine. The cuttings are planted by
dropping them at intervals of a foot or two in shallow furrows and
covering with the next round of the plow. This can be done when
the field is plowed, the cuttings being dropped every other round or
every third round. Or the field can be prepared first and the cuttings
dropped upon the surface and pressed in with the foot as they are
planted. For meadows it is best to go over the land with a roller
after planting. For pastures, when a smooth surface is not neces-
sary, it is sufficient to plow shallow furrows every 2 to 4 feet and
drop the cuttings therein, covering them with the foot or by turning
the soil back over them with the plow.

Professor Tracy remarks: "So easily may Bermuda grass be
propagated that good stands can be secured b}^ scattering a dozen or
more sods to the acre and cultivating the land in corn or cotton two 1
or three years, when the grass becomes distributed in the field."

i

AS A SOIL BINDER.

On account of its creeping habit of growth, Bermuda grass is an
excellent plant to prevent the washing of soils along ditches, ravines,
embankments, or other similar places, and also to prevent the drift-
ing of sand upon sand dunes. It has been used for the latter pur-
pose with excellent results upon sand dunes of the coast of Southern
California. It is not stout and vigorous enough to hold large shift-
ing dunes, but it will cover sandy soil and prevent its blowing.

FOR LAWNS.

Bermuda is the best known lawn grass of the South. It has all
the desirable qualities of a lawn grass except that of holding its
color during the winter. It turns brown upon the approach of cold
weather, and is rather late in becoming green in the spring. It is,

however, a good turf former, has a good color and a fine texture
(fig. 2), and under the usual lawn treatment forms an ideal lawn,
except during the winter months.

St. Lucie grass, as mentioned above, has, for the purposes of lawn
making, certain advantages over the common variety, one of which
is its remaining green later in winter. On account of its habit of
of growing more upon the surface it can be more easily eradicated,
and is not so likely to become a pest.

The quickest method for the production of a lawn, if the extent of
surface is not too great, is to transplant sod if that can be obtained.
A lawn can, however, be formed by the methods given under mead-
ows. To insure a perfect stand and uniform appearance, it may be
necessary to sprinkle or to hand- weed. It should be kept closely
mowed and rolled.

METHODS OF ERADICATION.

The very qualities which render Bermuda so valuable as a pasture
grass serve to
make it an ag-
gressive and pes-
tiferous weed. On
account of its tend-
ency to spread and
insinuate itself into
land where it is not
wanted, and to per-
sist in fields which
are to be used for
other purposes, it
has, in many cases,
not been utilized to
the extent that its
good qualities
would indicate.
However, it can be
eradicated from a
field with compar-
ative ease by proper cultivation. Since it will not thrive in the
shade, it is only necessary to smother it out by some quick-growing
crop. A method recommended by Southern agriculturists, and
which may be modified to suit conditi<jns, is to plow the land after
the last crop of hay is cut, if the field is a meadow, or about this
season if it is a pasture. Sow the field to oats, wheat, or other
thick-growing crops. When this crop is harvested, plow the land
immediately and plant to cowpeas. It is probably best to plant

KiG.'J. -Acloselycul twrfof Hcrniuila (ii-ass(a8 seen fruinabove,
designed t« show texture).

these in drills and cultivate them until the vines meet, after which
they will shade the ground and prevent the growth of Bermuda.
Usually this treatment is sufficient to completely destroy the Ber-
muda ; but if not, the process can be repeated.

FEEDING VALUE.

Bermuda grass is much relished by all kinds of stock, both when
fresh and in the form of hay. Experience has shown that its milk-
producing quahties, ton for ton, are fully equal to timothy. The
feeding tables show that it contains more protein than most of the
grasses commonly cultivated for forage.

SUMMARY.

(1) Bermuda grass is a native of the tropics, but widely introduced
throughout the Southern States. It can not be successfully grown
much north of Virginia.

(2) It is a creeping perennial, which will grow upon a great vari-
ety of soils and will endure extreme conditions of temperature and
moisture.

(3) It is the best pasture grass for most parts of the South, especi-
ally upon poor land.

(4) On rich soil it produces an abundance of nutritious hay, two
to four cuttings of which can be made each season.

(5) It can be started by thoroughly preparing the land and sowing
in the spring about 6 to 8 pounds of seed per acre, pressing it in
with a roller. It is usually more satisfactory to plant root-cuttings,
or bits of sod 2 inches square, about 2 feet apart each way.

(6) It is a valuable soil binder, and is used to hold steep slopes and
to prevent washing of soil or blowing of sand.

(7) It is the best-known lawn grass in the South, and is to be
recommended for this purpose, especially the variety known as "St.
Lucie grass."

(8) It can be eradicated from fields by proper cultivation of shad-
ing crops, especially cowpeas.

(9) Ton for ton the feeding value of its hay is equixl to that of
timothy. It is rich in protein, and is one of the most nutritious of

the cultivated grasses.

A. S. Hitchcock,

Approved : Assistant Agrostologist.

James Wilson,

Secretary.

Washington, D. C, March 21, 1901.

CIRCULAR No. :».— (Agros. 83.) Issued April 22, 1901.

United States Department of Agriculture,

DIVISION OF AGROSTOLOGY.

[Grass and Forage Plant Investigations.] ,

SOME ARIZONA GRASSES.

The material ou which the following list of Arizona grasses is based
was collected bj' Prof. David Griffiths and Director R. H. Forbes while
pursuing their investigations of the forage conditions in Arizona, in
cooperation between the Division of Agrostology and the Arizona
Agricultural Experiment Station. Unless otherwise stated, the speci-
mens cited below were collected by Professor Griffiths during the
months of September and October. 1900. — F. Lamson-Scribner.

TRACHYPOGON SECTJNDTJS (Presl) Scribn. n. comb. {Heterojiogon secundus
Pre!<l, Rel. Haenk. 3o5. 1830; Trachypogon montufari Nees Agrost. Bras. 342.
1829, not H. B. K. 1815; Trachypogon jjohjmorphus secundus Hack, in DC. Monog.
Phan. 6: 326. 1889.)

Dragoon Mountains, No. 1839; near Pearce, No. 1934.

Elionurus barbiculmis Hack, in DC. Monog. Phan. 6: 339. 1889.

Near Pearce, Nn. 19.56. »

ELIONURUS BARBICULMIS PARVIFLORUS Scribn. var. nov. Sessile spike-
lets about 5 mm. long, broader in proportion to its length and less acuminate-
pointed than in the species, in which the sessile spikelet is 8-9 mm. long.

Dragoon Mountains, No. 1849.

Andropogon contortus L. Sp. PI. 1045. 1753. {Heteroj^ogoii contortus R. & S.
Syst. 2: 836. 1817; H. Mrtus Pers.)

Dragoon Mountains, No. 1844; Vales, No. 1678; near Pearce, No. 1947.

The first glume of the pedicellate spikelet glabrous or sparingly tuberculate-pilose
near the apex.

Andropogon hirtiflorus feensis (Fourn.) Hack, in DC. Monog. Phan. 6:372.
1889. {A. feensis Fourn. Mex. PI. 2: 62. 1881.)

Dragoon Mountains, No. 1851.

Andropogon torreyanus Steud. Nom. ed. 2, 1:98. 1840. {A. glnucm Torr., not
Muhl.; A. KK ccJk in I i( Irs torreyanus Hack, hi DC. Monog. Phan. 6:495. 1889.)

Papayo Reservation, without number.

Sessile spikelet .3-4 mm. long.

Andropogon leucopogon Nees, Linna^a, 19: 694. 1845. (-1. saccharokles leuco-
p<,g<m Hack, in DC. Monog. Phan. 6: 496. 1889.)

Tombs-tone, No. 1982; Dragoon Mountains, No. 1854; Sulphur Spring Valley, No.
1639; Rincon Mountains, No. 1784.

Andropog-on halepensis (L. ) Brot. Fl. Lusit. 1: 89. 1804.

University grountl.«, Tucson, Xo. 1533.

Nazia aliena (Spreng.) Scribn. U. S. Dept. Agr. Div. Agros. Bui. 17: 28, fig.

324. 1899. (Lappago aliena Spreng. Neue Entd. 3:15. 1822.) Western

prickle-grass.
Near Pearce, Xo. 1952; Benson, Xo. 1832.
Nazia racemosa (L.) Kuntze, Rev. Gen. PI. 3': 357. 1898. {Cenchrus racemosus

L. Sp. PI. 1049. 1753; Lappago rncemosn Honck. Syn. 1: 440.) Prickle-grass.
University grounds, Tucson.
Hilaria jamesii (Torr. ) Benth. Journ. Linn. Soc. 19:62. 1881. {Pleuraphis

Janu'si) Torr. Ann. Lye. X. Y. 1: 148, t. 10. 1824.) Black bunch-grass.
Sulphur Spring Valley, Xo. 1647, R. H. Forbes.
Hilaria mutica (Buckl. ) Benth. Journ. Linn. Soc. 19:62. 1881. {Pleuraphis

mutica Buckl. Proc. Acad. Xat. Sci. Phil. 1862: 95. 1862.) Black grama.
Benson, No. 1998; Vales, Xo. 1679. The margins of the outer glumes densely fringed

with short silky hairs, prominent nerves of the broad outer glumes three,

diverging above as in the type, which, however, has 5-7 nerves in the outer

glumes.
Sulphur Spring Valley, Xo. 1647. In this specimen the outer glumes of the stami-

uate spikelets are about 6 mm. long, oblong, obtuse or subacute; nerves 3-5,

running nearly parallel. The shape of the glumes ap])roaches those of H.

jamesii, but the size of the spikelets and character of the pubescence at the base

of the group are those of H. mutica.
Hilaria cencliroides H. B. K. Xov. Gen. et Sp. PI. 1:117, t. 37. 1815. Curly

mesquite.
Tucson, Xo. 1534; Rincou Mountains, X^o. 1583; Cochise, Xo. 1883.

Paspalum distichum Linn. Amcen. Acad. 5:391. 1759. Knot-grass.

Tucson, Xo. 610.

Eriochloa aristata Vasey, Bui. Torr. Bot. Club, 13: 229. 1886.

University grounds, Tucson, X^os. 1612, 1546. Xot previously reported from the

United States. Introduced from Mexico.
Eriochloa annulata (Fliigge) Kunth, Rev. Gram. 1:30. 1835. {Pasjyalus annu-

lafus Fliigge, Monog. 135. 1810; Helopus annulatus Xees, Agrost. Bras. 17.

1829— Doell in Mart. Fl. Bras. 2^: 124, t. 19. 1871-1877.)
University grounds, Tucson, September 5, 1900, Xo. 1516.
This distinct and widely distributed species has not previously l)een reported from

the United States, probably introduced. It is represented in the U. S. National

Herbarium by specimens from Cuba, Australia, and India.
Eriochloa punctata (L.) W. Hamilt. Prodr. PI. Ind. Occ. 5. 1825.
University grounds, Tucson, No. 1537; Benson, No. 2002.
Eriochloa punctata minor Vasey, Contr. U. S. Nat. Herb. 3: 21. 1892.
Sulphur Spring Valley, R. H. Forbes, October, 1900, Xo. 1641.

PANICTJM ARIZONICUM Scril)n. & Merrill sp. nov. [Panicum (sine nomine)
Scribn. Bui. Torr. Bot. Club, 9: 76. 1882; P. fasciculatum dissitiflonun Vasey, in
herb. 'Sot P. di.mfiflorum Steud. 1841.)

A tufted, nearly glabrous perennial (?) 3 to 6 dm. high, with narrowly lanceolate
leaves, open panicles, and pubescent spikelets. Culms prostrate below and root-
ing at the lower nodes, glabrous except just below the panicle, as are the numer-
ous branches; nodes glabrous. Sheaths shorter than the internodes, smooth,
rather loose; ligule about 1 mm. long, ciliate; leaf-blades 8 to 15 cm. long, 5 to
8 mm. wide, cordate and clasping at the base, gradually tapering to the acumi-
nate apex, thin, but tirm, minutely serrulate-scabrous on the promment, carti-

kfiinoiis margins an<l with few scattered, stiff, papillate hairs on the margins
and often on the midnerve beneath. Paniele at length exserted, 18 to 20 cm.
long, the axis an<l the erect or ascending simple branches pilose with rather
stiff, spreading hairs, the lower ones 5 to 8 cm. long, bearing the approximate but
not crowded racemose spikelets in pairs (the upper ones solitary), one nearly-
sessile and often imperfectly developed, the other on a pedicel nearly its own
length; pedicels pilose. Spikelets oblong or obovate, somewhat pointed, 3-3.5
mm. long, the three outer glumes pubescent with short, spreading hairs; first
glume deltoid, obtuse, enclosing the base of, and about one-third as long as the
spikelet, .5-nerved; second and third glumes prominently ners-ed, subequal, as
long as the flowering glume, the second 7-nerved, the third 5-nerved, slightly
reticulate-veined above; flowering glume about 3 mm. long, pointed an<l finely
transversely rugose.

Type specimen collected on mesas near Camp Lowell, Santa Cruz Valley, Arizona,
465 C. G. Pringle, 1881.

Other specimens examined: Arizona: Tucson, 1596 D. Griffiths, September, 1900;
Bowie, 29 J. W. Toumey, October 1, 1896; no locality, 353 J. G. Lemmon, 1882;
sandy plains near the Mexican boundary, C. G. Pringle, August 29, 1884; along
streams, Santa Catalina Mountain, 3062 J. G. Lemmon and wife, August, 1883.
Xeic Mexico: Mangos, 0. Metcalfe, September 14, 1897.. Texas: near :Madeira
Creek, Presidio Co., G. C. Nealley, September, 1892. Mexico: Guaymas, 159,
208 E. Palmer, 1887; no locality, 250 E. Palmer, 1897. Loiver California: San
Jose del Cabo, 18 T. S. Brandegee, September 4, 1890.

This species was described by Scribner,^ without name, as '' Paaicum {Virgariu)
allied to P. fuscum Sw.," but is certainly distinct from that, and from related
species. In some respects it resembles Panicum velutinosum of Trin. Icon. 2:
t. 180, but is distinct from the plant there described and also very .listinct from
P. velutinomm Xees, Agrost. Bras. 121. 1829, which is certainly different from the
plant so considered by Trinius.

Panicum arizonicum is at once distinguished from related species by its panicle char-
acters and especially by its pointed pubescent spikelets.

PANICUM ARIZONICUM TENUE Scribn. & Merrill var. nov.

A small, ca-spitose form 1.5 dm. high or less, with smaller leaves and panicles than
in the species, the panicle branches 1 to 3 cm. long. Spikelets as in the species.

Type specimen collected at Fort Huachuca, Arizona, by T. E. ^Yilcox in 1894. Other
specimens examined: Mexican boundary line, south of Bisbee, Arizona, 1072
E. A. Mearns, October 9, 1892. Tcn/.s.- El Paso, M. E. Jones, September 10,
1884.

PANICUM ARIZONICUM L^VIGLUME Scriljn. & Merrill var. n<jv.

A low, ca-spitose, much branched form, which differs from Panicum arizonicum tenue
only in having glabrous si)ikelets, but which is clearly related to Panicum ari-
zonicum rather than to P. fuscum.

Type specimen collected at Mescal, Arizona, 1810 David Griffiths, October, 1900.
Other specimens examined: Cochise, 1810 Griffiths, October, 1900; Fort Hua-
chuca, T. E. Wilcox, 1894; Johnson's Ranch, 11 miles east of San Pedro River,
:Mexican boundary, 1694 E. C. Merton, August 7, 1893.

PANICUM ARIZONICUM MAJOR (Vasey) Scrilm. & Merrill n. comb. [Pani-
rnin jusrum major Vasey, U. S. Dept. Agr. Div. Bot. Bui. 8: 26. 1889). A stout,
erect form 6-9 dm. high, with larger leaves, which are scabrous-punctulate or
nearly smooth, scal)rous-punctulate sheaths, and larger, more l)ranched panicles.

Southwestern Chihualma, 3Iexico, Dr. E. Palmer, August to November, 1885.

1 Bui. Torr. Bot. Club, 9: 70. 1882.

The species in this group are much confused and no two botanists have interpreted
them alike. The following notes on the several species may be of interest.
Panicum fuscum Swartz is apparently the most common and widely distributed
species in the group and is represented in the U. S. National Herbarium by
specimens from Florida, Texas, New Mexico, Mexico, St. Croix, Porto Rico,
Martinique, Honduras, and Costa Rica. It is distinguished jjy its ascending
panicle branches, dark-brown, somewhat obtuse, glabrous spikelets, the outer
glumes of which are prominently reticulate-veined. See Trin. Icon. 2 : t. 206.
1829. The plant figured by Trinius is doubtless P. fasciculatum Sw. which evi-
dently is only a large form of P. fnKnnn.

PANICUM FUSCUM RETICULATUM (Torr. ) Scribn. & Merrill n. comb. (P.
recticulatum Torr. in Marcy's Explor. Red R. La. 299. 1853; P. fasciculatum
reticulatum Beal, Grasses 'N. A. 2: 117. 1896.)

Tucson, Nos. 15^5, 1616; Papayo Reservation, No. 1754.

This variety is distinguished from the species only by its contracted panicles an<l
should perhaps be considered only as a form of Pan icum fuscum Sw. It is rep-
resented in the U. S. National Herbarium by specimens from Texas, Arizona,
and Mexico.

Panicum grossarium Linn. Amccn. Acad. 5: 392. 1759. {Panicum adspersum
Trin. Gram. Panic. 116.1826; Icon. 2: 1. 169, 1S29.) This species is very distinct
from Panicum fuscum Swartz, and is at once recognized by its larger, lanceolate
leaves, 15-20 cm. long, 1.5-2 cm. wide, ciliate sheath margins, scabrous, not papil-
late rachis, and larger spikelets, which are 4 nun. long, prominently pointed,
the outer glumes sparingly tuberculate-hispid, not reticulate-veined. It is well
represented by the following specimens from Florida: St. Augustine, 176 T. H.
Kearney, jr., July 24, 1895; 6705 A. H. Curtiss, August 3, 1900; Sanibel Island,
292 J. H. Simpson, March, 1891; Key West, 5431 Curtiss, June 22, 1895; Sand
Key, Cape Sable, 3606 Curtiss, July 1880; no locality, A. AV. Chapman. A small
form of this species was also collected on l>allast at Philadelphia, Pennsylvania,
byF. Lamson-Scribner, Seiitember, 1881.

This is apparently a distinct species and is so considered by Grisebach, Fl. Brit. "\V.
Ind. 546, although some of the Florida specimens referred to here are much
larger than the plant he described.

Panicum carthaginense Swartz, Prodr. Yeg. Ind. Occ. 22. 1788, is apparently rep-
resented by a specimen collected on ballast at Mobile, Alabama, by Dr. C. Mohr,
September 16, 1891. This species is of doubtful value and should perhaps be
considered as a synonym of Panicum fuscum Swartz or as a variety of that species.

Panicum capillare Linn. Sp. PI. 86. 1753.

University grounds, Tucson, No. 1520; near Pearce, No. 1938; Cochise, No. 1918, a
small form related to Panicum barbijyulriiKttum Nash.

Panicum cognatum Schultes, Mant. 2: 235. 1824. (P. dirergens Muhl. 1817, not

H. B. K. 1815; P. autumnale Bosc. 1825.)
Benson, No. 1833; Dragoon Mountains, No. 1862.
Panicum colonum Linn. Syst. Nat. ed. 10, 870. 1759.
Papayo Reservation, No. 1655; University grounds, Tucson, Nos. 1518 and 1534,

forms approaching Panicum crus-gaUi Linn.

Panicum crus-gaUi Linn. Sp. PI. 56. 1753.

Benson, No. 1993; Papayo Reservation, No. 1671; Fairbank, No. 1970; Tucson, No.
1616. All these numbers represent the awnless form, which has been called var.
muticum by various authors, although all are somewhat diffei'ent from the Euro-
pean specimens and those of the eastern United States so referred.

Panicum hallii Yasey & Scribn. Bui. Torr. Bot. Clul), 11: 61. 1884.
Mescal, No. 1813.

Panicum obtusum H. B. K. Nov. Gen. et Sp. PI. 1: !>8. 1815.

Tucson, N..S. 1514, 1546; Benson, No. 2006; near Pearce, No. 1935; Sulphur Spring
Valley, No. 1645 R. H. Forl^es, October, 1900.

Panicum sang-uinale Linn. Sp. PI. 57. 1753. {Si/afherisma ><aaijidnalis Nash.)

University grounds, Tucson, No. 1517.

Panicum saccharatum Buckl. Prel. Rept. Geol. and Agrl. Snrv. Tex. App. 2.
1866. {P. larhniiiithniiiToYV. 1857, not Hochst. 1855.)

Near Pearce, No. 1942; Benson, No. 1997; Fort Lowell, No. 1572; Dragoon Moun-
tains, No. 1871; Rincon Mountains, No. 1782.

Chgetochloa composita (H. B. K.) Scribn. U. S. Dept. Agr. Div. Agros. Bui.
4:39. 1897— Scribn. & Merrill, ibid. BuL 21:27. 1900. {Setaria composUa

H. B. K.)

Benson, No. 2003; Tucson, No. 1511, small form.

Cheetochloa grisebachii (Fourn.) Scribn. 1. c— Scri])n. & Merrill,!, c. 35. {Setaria

gvi.sebiiddi Fourn. )
Common in sandy soil, Sulphur Spring Valley, Wilcox to Cochise, No. 1901.
ChEetochloa macrostachya (H. B. K. ) Scrilm. & Merrill, U. S. Dept. Agr. Div.

Agros. Bui. 21:29. 1900. {Sdarla macrostachya H. B. K.)
Near Pearce, No. 1944.
Chgetochloa viridis (Linn.) Scribn. U. S. Dept. Agr. Div. Agros. Bui. 4:39.

1897. {Panicum viride Linn.; Setaria viridis Beauv.)
University grounds, Tucson, No. 1526.

Cenchrus montanus Nees in Royle, 111. Bot. Himal. 416. 1839.
University grounds, Tucson, No. 1530. Introduced from Asia.

Cenchrus tribuloides Linn. Sp. PL 1050. 1753.

Fort Lowell, No. 1560.

Aristida sequiramea Scheele, Linntea, 22 : 343. 1849.

Cochise, No. 1880.

Aristida wquiramea Scheele has been variously referred to A. purpurea Nutt., .1.
fendkriana Steud., A. loiigiseta Steud., and A.fasciculata Torr., 1)ut is a very dis-
tinct species. It is most closely related to .1 ristida purpurea Nutt. , but is readily
distinguished by its usually larger size, larger spikelets, shorter empty glumes,
and especially by its very strongly tul)erculate-hispid flowering glumes. No.
1880 Grithths, cited above, differs from typical A. wquiramea in having nearly
awnless empty glumes.

Aristida americana Linn. Amcen. Acad. 5 : 393. 1760. {Aristida fasciculatu Torr.
Ann. Lye. N. Y. 1 : 154. 1824; Aristida dispersa Trin. & Rupr. Agrost. 3 : 129.
1842.)

LTniversity grounds, Tucson, No. 1528.

From a careful study of the synonymy of this species, it is evident that Aristida
americana Linn, is the correct name for this species.

ARISTIDA AMERICANA BROMOIDES (H. B. K.) Scribn. & Merrill, n. comb.
{Aristida hrumuidcs II. B. K. N<jv. Gen. et Sp. PI. 1 : 122. 1815.)

Benson, No. 1835; Rincon Mountains, No. 1808; Pearce, No. 1926; Sulphur Spring Val-
ley, 1632 R. H. Forbes, October, 1900.

This is the common, low, csespitose form of the species, and scarcely worthy of spe-
cific rank.

Aristida divaricata Humb. (k Bonpl. in Willd. Enum. 99. 1809.

:yiescal, No. 1816.

Aristida divergens Vasey, Contr. U.S. Nat. Herl). 3:48. 1892. {A. schiediana

minor Vasey.)
Dragoon Mountains, No. 1872; near Pearce, No. 1838.

6

Aristida havardii Vasey, Bui. Torr. Bot. Clnb, 13 27. 1886.

Cochise, Xo. 1885.

Aristida schiediana Trin. & Rupr. Agrost. 3 : 120. 1842.

Drag! lull ^[ountaiiii^. No. 1866.

Muhlenbergia gracillima Torr. Pac. R. R. Rept. 4^ : 155. 1857.

Rincon Mountains, No. 1805; Mescal, No. 1811; Dragoon Mountains, No. 1857.

Mulilenbergia porteri Scribn. in Beal, Grasses N. A. 2 : 259. 1896. (J/, iexana

Thurb. in Coult. :Man. Bot. Rocky Mt. Reg. 410. 1885. Not Buckley. 1862.)
Vales, No. 1872; Tucson, No. 1602. A form with unusually large panicles.
Muhlenbergia vaseyana Scribn. Rept. Mo. Bot. Gard. 10 : 52. 1899. {M. dis-

tichophylla of American authors. Not Kuntli. )
Dragoon Mountains, No. 1876; near Pearce, No. 1857 (small form).
Lycurus phleoides H. B. K. Nov. Gen. et Sp. PL 1 : 142. 1815.
Pearce, No. 1959; Dragoon Mountains, No. 1838.
Sporobolus airoides Torr. Pac. R. R. Rept. T : 21. 1856.
Sulphur Spring Valley, No. 1638 R. H. Forbes, October, 1900; Benson, No. 2005;

Tucson, No. 1502; Wilcox, No. 1904, abundant about an alkali lake bottom, a

valuable grass, called "fine saccaton."
Sporobolus argutus (Nees) Kunth, Enum. 1:215. 1833. {Yilfa arguta Nees

Agrost. Bras. 395. 1829.)
Benson, No. 1490; AVilcox, sandy ground, No. 1896.
Sporobolus auriculatus Vasey, Contr. U. S. Nat. Herb. 3 : 64. 1892. {S. asperi-

folius brerlfoUus Vasey, ibid. 1 : 56. 1890. Not S. hrevlfoUus Nees. )
Dragoon Mountains, No. 1864. A peculiar form with nearly all the spikelets

2-flowered.
Sporobolus confusus (Fourn.) Vasey, Bui. Torr. Bot. Club, 15 : 293. 1888. ( Vilfa

coiifusa Fourn. Mex. PI. 2 : 101. 1881. )
Dragoon Mountains, No. 1846.
Sporobolus cryptandrus (Torr.) A.Gray, Man. Bot. 576. 1848. {Agrostis cryp-

tandra Torr. Ann. Lye. N. Y. 1 : 151. 1824. )
Fairbank, No. 1981; Fort Lowell, No. 1581; Sulphur Spring Valley, No. 1636 R. H.

Forbes, September, 1900.
SPOROBOLUS STRICTTJS (Scribn.) n. comb. {Sporobolus cryptandrus stricta

Scribn. Bui. Torr. Bot. Club, 9 : 103. 1882. )
Benson, No. 2004; University grounds, Tucson, No. 1532; Rincon Mountains, No.

1796; Wilcox, No. 1893.
This form is readily distinguished from Sporobolus cryptandrus A. Gray, by its strict,

erect, densely flowered, wand-like panicles, the lower portion of which is

enclosed in the inflated leaf-sheath.
Sporobolus wrightii Munro in Scribn. Bui. Torr. Bot. Club, 9 : 103. 1882.
Sulphur Spring Valley, No. 1637 R. H. Forbes, October, 1900.
Sporobolus utilis Torr. Pac. R. R. Rept. 5^ : 365. 1857. ( Vilfa saccatilla Fourn.

Mex. PI. 2 : 101. 1881; Sporobolus saccatilla Griseb. 1. c.)
Sulphur Spring Valley, R. H. Forbes, No. 1635.
Epicampes rigens Benth. Journ. Linn. Soc. 19 : 88. 1881. {Cudm i.Mcroura Thurb.

1880. Not Kunth. 18.35. )
Dragoon Mountains, No. 1848. Forms large tussocks and is invariably closely

cropped by stock.
Polypogonmonspeliensis (Linn.) Desf. Fl. Atl. 1:67. 1798. {Alopecurusmonspe-

//V/;.-<;.s^ Linn. Sp. PI. 89. 1753.)
Tucson, No. 1619.

Cynodon dactylon (L. ) Pers. Syn. 1 : 80. 1805. {Panicum dactylonJAnn. ^i). V\.

85. 1753; Capriola dactylon 0. Kuntze, Rev. Gen. PI. 2 : 764. 1891.)
University grounds, Tucson, No. 1520.

Chloris elegans H. B. K. Nov. Gen. et Sp. PI. 1 : 166, f. 49. 1815. {Cldorlx a/ha

Presl, Rel. Ilaenk. 1:289. 1830.)
Cochise, No. 1923; near Pearce, No. 1929; University grounds, Tucson, No. 1521;
Mescal, No. 1826; Dragoon Mountains, No. 1852; Sulphur Spring Valley, No.
1643 R. H. Forbes, October, 1900.

Chloropsis mendozina (Phil.) Kuntze, Rev. Gen. PI. 3- : 348. 1898. {Cidorls?
mendozhui Phil. Anal. Univ. Chile, 36 : 208. 1870; Cldorojjsis hlanchard'iaini J.
Gay in herb. Paris; TricMoris bhoidtardiaua Scribn. Bui. Torr. Bot. Clul), 9 : 146.
1882; Triddoris fasciculata Fourn. Mex. PI. 2: 142. 1881; TricMoris verticillata
Vasej', U. S. Dept. Agr. Div. Bot. Bul. 12:^ : 2.r,, jil. 25. 1891; Chloropsis fasci-
cnlata O. Kuntze, Rev. Gen. PI. 2:771. 1891; Chloropsis hlanchardiana O.
Kuntze, 1. c. ; Tricldoris mendocina Kurtz, ]Meni. Facult. Ci. Univ. Cordoba,
1896: 37. 1897; Leptochloris greggHM.nnroiwh.arh.)

Benson, No. 1969.

Note — Kuntze' has the following note regarding this genus: " Trichlora und -is sind
nicht nebeneinander ziilassig, sodass das synonym von Hackel giltig wird.
Hackel schrieb: bei den " Gartnern" als Chloropsis oder Chloridop.ns (ein unpub-
licirten name) bekannt. Durand ind. schreibt, Chloridopsis "Gaertn." sodass
man den botaniker namens Gaertner damit verstehen miisste."

Chloropsis Hack. 1887. "Hort." in Engler & Prantl, Nat. Pfi. 2^ : 59. 1887, as
synonym; Chloridopsis Hack. 1. c, as synonym; TricMoris Fourn. Mex. PL
2:142.' 1881. Not Tn'cWora Baker. 1877.

Kuntze is, however, evidently wrong in his conception of species in this genus, as he
recognizes six, holding Chloropsis fascirulata, C. hlanchardiana, and C. mendozina
as distinct species, although they are identical. There are apparently two well-
marked species in this genus, Chloropsis mendozina (Phil.) O. Kuntze and C.
jilxrifora (Fourn.) O. Kuntze (TricMoris pluriftora Yourn.), both of which grow
in North and South America. For an admiral^le discussion of the species in this
genus see F. Kurtz, Col. Fl. Arg. 46-50. 1900 (reprint from Bol. Acad. Nac.
Cordoba 16. 1900).

Bouteloua aristidoides (H. B. K.) Griseb. Fl. Brit. W. Ind. 537. 1864. [Dinehra

aristidoides H. B. K. Nov. Gen. et Sp. PI. 1 : 171. 1815.)
Mescal, No. 1818; University grounds, Tucson, No. 1527; near Pearce, No. 1928;

Cochise, No. 1922.

Bouteloua bromoides (H. B. K.) Lag. Gen. et Sj). Nov. 5. 1816. {Dinehra bro-

moides H. B. K. Nov. Gen, et Sp. PL 1 : 172, t. 51. 1815.)
Rincon Mountains, No. 1670; near Pearce, No. 1949.

Bouteloua curtipendula (Michx.) Torr. in Emory, Notes Mil. Recon. 153. 1848.

[Cldoris cu.rtipcndala'SHch'S.. 1803; Boideloua racemosa Lag. 1805.)
Benson, No. 1834; Dragoon Mountains, No. 1837; Cochise, No. 1907.

Bouteloua eriopoda Torr. Pac. R. R. Rept. 4^ : 155. 1857.
IMescal, No. 1825; Rincon Mountains, No. 1669.

Bouteloua havardi Vasey, Proc. Am. Acad. 18 : 179. 1883.
Mescal, No. 1S29.

Bouteloua hirsuta Lag. Varied. Cienc. Lit. Art. 2 : 141. 1805.

Dragixtu Mountain:^, No. 1865; near Pearce, No. 1948; Rincon ^lountains, No. 1809.

iRev.Geii. PI. 2: 771. 1891.

8

BOXJTELOUA MICRANTHA Scribn. Oc Merrill sp. nov.

A very s^leiider, erect, cwspitose, glabrous perennial 7-9 dm. high, with sparingly-
branched culms, which are geniculate below, 6-10 very slender, erect or
ascending, densely flowered spikes 2-2.5 em. long, and very small spikelets.
Sheaths shorter than the internodes, glabrous; ligule a ciliate fringe about 2
mm. long; leaf-blades 5-8 cm. long, about 3 mm. wide, with few long papil-
late hairs on the margins and often on the upper surface. Panicle 1.5-2 dm.
long, the spikes about equaling or the upper ones exceeding the internodes, com-
mon rachis smooth. Pedicel of the spikes about 2 mm. long, minutely short-
pubescent; the rachis flattened, scabrous on the margins. Spikelets crowded in
two rows, secund, purplish, 2-2.5 mm. long; empty glumes unequal, the first
hyaline, lacerate at the apex, a1x)ut 1 mm. long, the second firm, purplish, ovate,
pubescent, Avith few short, stiff hairs, cleft at the apex, the midnerve prominent,
excurrent as a stout awn 0.5 mm. in length; flowering glume about 2 mm. long,
pilose with rather long, ascending hairs below and on the margins, smooth above,
three-cleft, three-awned, the awns from the cleft less than 1 mm. long, scarcely
exceeding the lobes of the glume, the two middle lobes broad, obtuse, the lateral
ones very narrow, acute. Palea equaling the glume, 4-toothed, the teeth ciliate.
Awns 2, very short. Sterile rudiment on a pedicel about 0.5 mm. long, with a
tuft of few short hairs at the apex, consisting of three glumes and three awns
which are somewhat longer than those of the flowering glume.

Type specimen collected at Fort Lowell, Arizona. David Griffiths, September 1900,
No. 1556. No. 244 of E. Palmer's ^Mexican collection for 1897 is the same.

This species has been confused with Bouteloaa polystadnja and B. inicrostachya, but
is strikingly different from those species in habit of growth. It is perhaps most
closely related to Bouteloua rothrocHi, but is distinguished by its very slender
habit, smaller, narrower spikes, much smaller spikelets, and shorter awns.

Bouteloua oligostachya (Nutt.) Torr. in A. Gray, Man. Bot. ed, 2, 553. 1856.

{Atlteropogon oligostadtyus 'Suit. Gen. 1 : 78. 1818.)
AVilcox, No. 1905; near Pearce, No. 1980; University grounds, Tucson, No. 1552;

Dragoon Mountains, No. 1863; Sulphur Spring Valley, R. H. Forbes, No. 1640.

Bouteloua polystachya (Benth.) Torr. Pac. R. R. Rept. 45:366, t. 10. 1847.

{Cliondrosium jiohi-itadtyumBenth. Bot. Voy. Sulph. 56. 1841.)
Wilcox, No. 1897, abundant on sandy ground; Benson, No. 2009; University grounds,

Tucson, No. 1524; Sulphur Spring Valley, R. H. Forbes, No. 1646; Rincon

Mountains, No. 1785.

Bouteloua vestita (S. Wats.) Scribn. Contr. U. S. Nat. Herb. 2 : 531. 1894. {B.

polystiickya vestita S. Wats. Proc. Am. Acad. 18 : 177. 1883.)
Mescal, No. 1819; Pearce, No. 1927.

Dactyloctenium aegyptium (Linn.) Willd. Enum. 1029. 1809.
University grounds, Tucson, No. 1515.
liBptocliloa dubia (H. B. K.) Nees, Syllog. Ratisb. 1:4. 1824. {Chloris ditbia

H. B. K. Nov. Gen. et Sp. PI. 1 : 169. 1815; Diplachne dubia Scribn. Bui. Torr.

Bot. Club, 10 : 30. 1883.)
Cochise, No. 181; near Pearce, No. 1943; University grounds, Tucson, No. 1539, a

young plant.
Leptochloa fascicularis ( Lam. ) A . Gray, Man. Bot. 588. 1848. ( Festuca fascicularis

Lam. Tabl. Encycl. 1 : 189. 1791; Dipladine fascicularis Beauv. Agrost. 160.

1812.)
Fairbank, No. 1971.

Leptochloa mucronata (Michx.) Kunth, Rev. Gram. 1 : 91. 1835. {Eleusinemucro-
wda Michx. Fl. Bor. Am. 1 : 65. 1803. )
University grounds, Tucson, No. 1513.

9

lieptochloa mucronata pulchella Scribn. Bui. Torr. Bot. Clulj, 9 : 147. 1882.

Benson, No. 1989. This variety is the small tufted form and extends from Missis-
sippi to Arizona and IMexico. From the specimens in the U. S. N'ational Her-
barium it is evident that its small size is due to habitat.

L.EPTOCHLOA PIL.OSA Scri1)n. sp. nov.

A c-espitose, rigid, s<jme\vhat purplish perennial about 3.5 dm. high, with papillate-
bispid sheaths and terminal panicles of many slender, spreading spikes 3-6 cm.
long. Culms glabrous, somewhat branched below; sheaths usually ekceeding
the internodes, rather loose, pilose with long, weak, spreading hairs, each from a
dark-colored glandular papilla; ligule a lacerate fringe al)out 2 mm. long; leaf-
blades 5-12 cm. long, 4-6 mm. wide, rigid, ascending, acute, subauriculate at
the hi^se, often somewhat involute, and with few scattered papillate hairs on the
upper surface, glabrous beneath. Spikelets about 2 mm. long, 2- to 3-flowered;
empty glumes subequal, acute; flowering glumes obscurely emarginate, not
mucronate, about 1.5 mm. long.

Type specimen collected in sandy soil, Dappan, Travis Co., Texas, 294 J. E. Bodin,
Septend)er, 1891.

No. 1603 Griffiths, Tucson, Arizona, September, 1900, is apparently the same, differ-
ing only in its shorter leaves and panicle branches.

This species is closely related to LeptocMoa mucronata, but is at once distinguished by
its rigid leaves and papillate-pilose sheaths.

Iieptochloa pringlei Beal, Grasses N. A. 2 : 426. 1896. {Diplachne pringlei Vasey,
1. c.) Dragoon Mountains, No. 1850; Rincon Mountains, No. 1801.

Leptochloa viscida (Scribn.) Beal, Grasses N. A. 2:434. 1896. {Dqdachne
viscida Scribn. Bui. Torr. Bot. Clul), 10 : 30. 1883.)

Benson, No. 1988; No. 1968, from the same locality, is a lax form of this species.

Pappophorum wrightii S. Wats. Proc. Am. Acad. 18 : 128. 1883. (P. horeale
Torr. Pac. R. R. Rept. 4^ : 155. 1857. Not Griseb. 1853.)

Mescal, No. 1827; Rincon Mountains, No. 1673; Vales, No. 1668; University grounds,
Tucson, No. 1525; Cochise; No. 1908; Pearce, No. 1925.

Pappophorum vaginatum Buckl. Prel. Rept. Geol. and Agrl. Surv. Texas, App. 1.
1.S6(). [P. aperttua :\Iunro in Scribn. Bui. Torr. Bot. Club, 9 : 148. 1882.)

Fairbank, No. 1969; University grounds, Tucson, No. 1538.

As Buckley's name is the earliest one published for this species it is here taken up, as
both Buckley's and Munro's species were based on the same plant, No. 803 C.
Wright, Texas, 1849.i At the time the description of /''. apertinn was publi-shed it
was supposed that P. vaginatum was only an herbarium name, although it had
been published many years before. On account of its very obscure publication
it was not listed in Index Kewensis.

Cottea pappophoroides Kunth, Rev. Gram. 1 : 281, /. 3;.'. 1835.

Near Pearce, No. 1939.

Arundo donax Linn. Sp. PI. 81. 1753.

No label; probably University grounds, Tucson.

Triodia mutica (Torr.) Scribn. Bui. Torr. Bot. Club, 10:30. 1883. {Tricuspid

mutka Torr. Pac. R. R. Rept. 4 : 156. 1857.)
Tombstone, No. 1976; Mescal, No. 1812, frequent on gravelly Ijanks.

TBIODIA PILOSA (Buckl.) n. comb. {Uralepls {Trieuspis) jjilosn Buckl. Proc.
Acad. Nat. Sii. Phil. 1862 : 94. 1862; Trieuspis acuminata Munro in A. Gray,
1. c, 335; Sierjlingia acuminata 0. Kuntze, Rev. Gen. PI. 2 : 789. 1891; Sieglingia
pilom Nash in Britt. & Br. Illus. Fl. 3 : 504. 1898.)

Tombstone, No. 1975.

1 Scribn. Bui. Torr. Bot. flub. 9 : 14S. 1882.

204:82—01 2

I

10

Triodia pulchella H. B. K. Nov. Gen. et Sp. PI. 1 : 155, t. 47. 1815. {Sieglingia

indcheUa Kuntze, Rev. Gen. PL 2 : 789. 1891.)
Tombstone, Xo. 1978; Rincon 3Iountain.«, No. 1665; Tucson, Nos. 1503, 1807.

Scleropogon brevifolius Philippi, Anal. Univ. Chile, 34 : 205. 1870 — Sert. Mendoc.

2 : 48. 1871. ( Trimspis vionstrosa Munro in Hemsl. Diagn. PI. Nov. Mex. 56.

1880; Lesourdia mtdtifora Fourn. Bui. Soc. Bot. Fr. 27 : 102. 1880; L. Kanvin-

skyana Fourn. 1. c. ; Scleropogon karicuisl-yanus Benth. in S. Wats. Proc. Am.

Acad. 18: 181. 1883.)
Dragoon Mountains, Nos. 1855, 1856; Cochise, No. 1912; Sulphur Spring Valley, R.

H. Forbes, No. 1630.

Eragrostis abyssinica Schrad. Linnpea, 12 : 450. 1838.
University grounds, Tucson, Nos. 1422, 1542. Introduced.

Eragrostis lugens Nees, Agrost. Bras. 505. 1829.

Dragoon Mountains, Nos. 1843, 1853.

Eragrostis major Host. Gram. 4 : 14, pi. 14. 1809.

Mescal, No. 1814; Papayo Reservation, No. 1653; University grounds, Tucson, No.

1523.
Eragrostis purshii Schrad. Linuaea, 12 : 451. 1838.
Benson, No. 2010.

Distichlis spicata (L.) Greene, Bui. Calif. Acad. Sci. 2 : 415. 1887. {Uniola spicata
L. Sp. PL 71. 1753; D. maridma Raf. Journ. Phys. 89 : 104. 1819. )

Tucson, Nos. 1504, 1505; Wilcox, No. 1899. Abundant and was the means of carry-
ing many cattle through the dry spell, although not eaten ordinarily.

Poa pratensis Linn Sp. PI. 67. 1753.

University grounds, Tucson, No. 1512, a depauperate form.

Festuca octoflora Walt. FL Carol. 81. 1788. {Fe.stxca ienella Willd. Enum. 1 : 113.

1809. )
On high peaks near Pearce, No. 1958.
Bromus unioloides (Willd.) H. B. K. Nov. Gen. et Sp. PL 1 : 151. 1815. {Festuca

unioloides Willd. Hort. Berol. 1 : 3, pi. 3. 1806.)
University grounds, Tucson, No. 1531.

Elymus canadensis Linn. Sp. PL 83. 1753.
Papayo Reservation, No. 1660.

Elmer D. Merrill,
Assistant hi Charge of Collections.

Approved:

James Wilson,

Secretary of Agriculture.
Washington, D. C, Aj?ril 8, 1901.

O

Circular No. 33. — (Agros. 84.)

United States Department of Agriculture,

DIVISION OF AGROSTOLOGY.

[Grass and Forage Plant Investigations.]

RANGE GRASS AND FORAGE PLANT EXPERIMENTS AT HIGHMORE,
S. DAK.— REPORT OF PROGRESS.*

The cooperative range grass and forage plant experiments at High-
more, S. Dak., begun in the spring of 1899 and conducted in cooper-
ation with the South Dakota Experiment Station, were continued
during the season of 1900, and the present circular is based upon the
report of Mr. Louis W. Carter, who has had immediate charge of the
work. In 1899 there were cultivated T-t plats, including -12 varieties of
grasses and forage plants, all but 12 of which were perennials. In
1900 57 varieties were grown, of which 27 were not grown in 1899.
These were as follows : Meadow fescue, red fescue, Broiiiua en^cfus^
B. hdmil^ B. teetorum, orchard grass, Washington blue-grass, Japan-
ese barnyard millet, white Russian broom-corn millet, Samarkand
alfalfa, French alfalfa, Australian saltbush, bitter vetch {Lathyrus
sativus)^ white milo maize, Egyptian corn, March rape, Canadian 13'me
grass, blue Canadian lyme, goose wheat, oats, barley, speltz, emmer,
Russian buckwheat, Viciafaha^ white soy beans, sweet clover.

The season of 1900 was peculiar in that, although the rainfall
was larger than usual, the season as a whole was very unfavoraljle.
The preceding winter was dr}^ and without snow until the latter
part of ]\[arch, when there was a fall of 8 inches. This, together
with a fall of 3.2 inches of rain at the end of April, furnished
favorable conditions for the germination of seeds, but was followed
T)y a protracted drought of about seven weeks. Inuring that period
hot winds blew a part of the time, the prairie turned j-ellow, ponds
dried up, and wells began to fail. The drought was general over
South Dakota and all grasses and grains sufiered, but cultivated crops,
like corn and potatoes, when well tended, resisted the drought and

*The report for 1899 is embodied in Circular No. 21 of this Division, entitled
Cooperative Range Grass and Forage Plant Experiments at Ilighmore, S. Dak.

suffered little. Rains in the first half of Julj" were followed b}^ a
period of hot, drj^ weather, accompanied b}' hot winds, which produced
serious injur}^ and was fatal to corn which was drilled or planted
thickly. Corn in hills and well cultivated resisted the drought,
which fact shows that methods of culture tending to conserve the soil
moisture are of great importance. There was a rainfall of 7.2 inches
in August and 4.39 in September.

The report includes a record by plats, following the S3^stem pre-
sented in the first report.

Plat A (1). — Smooth bunch grass {Poa hrvigata). After the June rains the grass
resumed its growth, interrupted by the spring drought, and formed quite a sod, but
did not get over 6 inches high and did not form heads.

A part of the plat was plowed up and sown to the seed of this grass, but the young
plants were all killed by the drought.

Plat A (2). — East half. Bunch grass (a form of Poa Ixvigata). Good growth in
May, but did not head out. Formed a good sod latter part of summer.

The west half, in Canadian olue grass {Poa compressa), made a good growth early
in the season and headed out June 1, but formed no seeds.

Plat A (3). — Nevada blue grass {Poa nevadensis) . Condition similar to the
preceding.

Plat A (4). — Oregon blue grass {Bromus unioloides). Eesown. Seems to be a
winter annual. Self-sown seeds germinated in the fall, survived the winter, and
headed out in June.

Plat A (5). — Short-awned brome grass {Bromus breviaristaius) . Poor stand.

Plat A (6). — King's fescue {Festuca kingii). Fair growth, but did not head out.

Plat A (8). — East half. Bearded wheat grasa {Agropyr on caninum). Did no;
head out, but grew fairly well.

Most of the west half in feather bunch grass {Stipa viridida), but growth poor.

Plat A (9) . — Giant rye grass {Elymus condensatus) . Growth poor.

Plat A ( 10) . — Six lots of slender wheat grass {Agropyron tenerum ) . Made a growth
of about 6 inches.

Plat A (11). — Slender wheat grass {Agropyron tenervm). This plat, which was
drilled, stood the drought better than the preceding, which was sown broadcast.

Plat A (12). — Wild timothy {Muhlenbergia racemosa). Growth poor, but headed
out in Sejitember. Killed by frost September 20.

Plat A (13).— A mixture of curly mesquite {Hilaria cenchroides) , blue grama
{Bouteloua oUgostachya) , and King's fescue {Festuca kingii). Growth poor.

Plat A (14) . — Blue grama {Bouteloua oUgostachya) . Growth poor.

Plat A (15) . — Mixed grama grasses. Growth poor.

Plat A (16, 17, 18). — Sown to Bromus inermis, B. erectus, B. kalmii, B.tectorum,
Agropyron pseudorepens, Dactylis glomerata, Muhlenbergia racemosa, and Poa nevadensis,
but all failed to germinate.

Plat A (19,20). — Turkestan alfalfa {Medicago sativa turkestanica) . Grew well
except during the midsummer drought.

Plat A (22,23,24). — Western wheat grass {Agropyron spicatum). Growth fair,
but did not form heads.

Plat A (21,25 to 29). — Smooth brome grass {Bromus inermis). Growth good
except during summer drought.

Plat B (1). — Planted to potatoes in preparation for next year.

Plat B (2). — Planted in sugar corn.

Plat B (3). — Planted to millets; west side to Japanese barnyard millet. Growth
good, nnally reaching a height of 4 to 5 feet. One of the best forage plants tested

this year. Another portion of plat planted to white Russian broom-corn millet
(S. P. I., No. 1387). Growth poor. The remainder of the plat planted to red vero-
nezh broom-corn millet (S. P. I., No. 2796). Growth very poor.

Plat B (4 ) .—Samarkand and French alfalfa. Drilled in rows 1 foot apart. Growth

good in both cases.

Plat B (5, 6, 7, 8) .—Turkestan alfalfa {Medicago sativa turkestanica) . Wintered

well. Growth fair.

Plat B (9).— Turkestan alfalfa (S. P. I., No. 991). Sown broadcast at the rate of
28 pounds per acre. Suffered from the midsummer drought. Killed by frost Sep-
tember 20.

Plat B (10).— Australian ssLlthnsh {{Atriplex semibaccata) . Drilled May 10 on the
following plan: 11 rows on the west side, seed pressed in; 11 rows in the center, seed
covered one-half inch; 11 rows on the east side, seed 1 inch deep. Seed did not
sprout till June 20. Only a few plants came in at west side. In the center the stand
was about 50 per cent, while on the east side the stand was about 25 per cent. Made
a growth during the latter part of the season, but was all dead by the 1st of November.

Plat B (11).— Bitter vetch {Lathyrus sativus). Made a fair growth and stood the
drought well.

Plat B (12).— Milo maize {Andropogon sorghum). Yellow variety cut September
10. Weighed, November 3, at the rate of 6,336 pounds per acre. White variety cut
September 15. Weighed, November 3, at the rate of 12,693 pounds per acre.

The yellow milo maize formed heads, but the white did not.

Plat B (13).— Egyptian corn {Andropogon sorghum) . Stand poor and growth fair.
Formed heads August 18. Cut September 11, when about 5 feet high. Weighed,
November 3, at the rate of 3,120 pounds per acre.

Plat B (14). — Wisconsin amber cane {Andropogon sorghum) . Stand, about 50 per
cent. Yielded at the rate of 2,960 pounds per acre.

Plat B (16) .—Hairy vetch ( Vida villosa). Failed to germinate.

Plat B (17).— Rape {Brassica napus). West half, March rape, drilled May 10;;
not successful. East half, dwarf Victoria rape, drilled May 10, made a rank growth
and yielded, September 1, green fodder at the rate of 26,880 pounds per acre. A
single plant weighed 23 pounds.

Plat B (18 to 24).— Sown to the following grasses, but all were killed by drought:
Hard fescue {Feducascahrellu), red fescue {Festuca rubra), wild timothy {Mahlmbergia
racemosa), Washington blue grass, rye grass {Elymus canadensis glaucifoUus) , smooth
brome grass {Bromus inernus), and Western wheat grass {Agrojnjron spicatiim).

Plat B (25 to 30).— Smooth brome grass {Bromus inermis). A very poor stand.

Plat C (1) .—Planted to potatoes in preparation for next year.

Plat C (2, 3) .—Goose wheat. Severely injured by drought and Russian thistles.

Plat C (4, 5).— Oats. Killed by drought.

Plat C (6 to 11). — Russian speltz. Growth poor. Yielded at rate of 4 to 5 bush-
els per acre.

Plat C (12).— Emmer {Triticum dicoccum). S. P. I., No. 2789, on part of plat,
remainder sown to Russian buckwheat {Fagopyrumfagopyrum), S. P. I., No. 2801, but
in both cases the young plants were killed by the drought.

Plat C (13). — Planted to various varieties of imported wheats, oats, barley, and
speltz, bitter vetch (S. P. I., No. 1175), horse beans ( Vidafaba), and white soy beans,
but all were killed by drought.

Plat C (14).— Red Orenburg broom-corn millet, S. P. I., No. 2960 {Panicum milia-
ceum). Growth poor, but stood the drought better than any of the broom-corn mil-
lets except the black (Plat C, 19).

Plat C (15) .—Tambov broom-corn millet, S. P. I., No. 2794. A red variety. Badly
damaged by the hot winds.

Plat C (16).— Red Russian broom-corn millet, S. P. I., No. 2797. Similar to
preceding.

Plat C (17). — A white variety of foxtail millet {ChsetocMoa italica). Made a fair
growth. Cut September 6.

Plat C (18).— A red variety of the preceding. Made a good growth.

PlatC (19).— a black variety of Russian broom-corn millet, S. P. I., No. 2795.
Made a fair growth and withstood the drought well.

Plat C (20 to 30). — Sown to common millet to ke^p down the weeds.

An additional series (D) was broken up and planted this season, with the inten-
tion of carrying on experiments in green manuring, but owing to the dry weather
they were not successful. Series E was broken up and sown broadcast to common
millet to keep the ground clean.

Series F.— A portion of this had not been plowed for eight years and was very
weedy in places. This portion Avas pulverized last May and sown broadcast with
Bromus hiermis and then dragged. The seed germinated about September 1, and the
results will be noted the following season. The remainder of this plat is broken in
preparation for the next year.

Series H.— Numbers 4 and 5 were treated with 40 loads of manure, which was
harrowed in. These plats showed an increase of grass of about 30 per cent.

Plat H (6). — Scarified and sown to Bromus iuerrn'm.

Plat H (7). — Scarified and sown to Nevada blue grass {Poa nevadensis) .

Plat H (8). — Scarified and sown to King's fescue (Festuea king'd).

Plats H (9, 10, 11, 12). — Scarified and left to be sown the following spring.

From the preceding it will be seen that the grasses mentioned in the
report of 1899 as giving promise have in most cases confirmed the
estimates placed upon their value. Among these may be mentioned
smooth bunch grass {Poa IcBvIgata), Nevada blue grass {Poa neva-
densis), King's fescue {Festuea Mngii), and Oregon brome grass
{Bromus unioloides). On account of the extremel}^ dry weather many
of these did not form seed, but they nevertheless made a fair growth
and thickened into more or less of a sod. A few other perennial
grasses grew well in spite of the unfavorable conditions, such as
bearded wheat grass {Agropyron caninum)^ slender wheat grass {A. tene-
rum), and western wheat grass {A. sj>(catum). Canadian blue grass
{Poa compressa) made considerable growth early in the season and
headed out by June 1, though it formed no seed. This grass, how-
ever, is not so valuable from an agricultural standpoint as some of the
others. Smooth brome grass {Bromus inermis) made a favorable
growth, though, like most of the other perennial grasses, it suffered
from the summer drought, but it renewed its growth upon the return
of more favorable conditions. This is undoubtedly one of the best
grasses for the dry regions of the Northwest for both hay and pasture.

Turkestan alfalfa has grown fairly well, but so far it has not justi-
fied the claim of great superiority over the ordinary sort.

The Australian saltbush {Atrlplex semihaccata) germinated poorly,
but withstood the drought quite well. As it does not survive the
winter it must be treated as an annual and sown each spring. Except
on alkali soil, it probably is not to be recommended in this latitude, as
there are other more promising annual plants for ordinary soil.

Among the annual grasses that furnish coarse forage there are sev-

eral that proved able to produce a crop in spite of the extreme condi-
tions. Japanese barnyard millet is especiall}^ to be recommended.
Common or foxtail millet also did well. The broom-corn millets made
a fair growth, ))ut the varieties tested thus far are inferior to the
common millet.

The sorghums, including amber cane, Egyptian corn, and milo
maize, in several cases produced a large amount of fodder, a white
variety of milo maize yielding at the rate of 6 tons per acre (Plat B, 12).

Dwarf Victoria rape made a rank growth and yielded green fodder
at the rate of 13 tons per acre.

Many varieties of grasses which Avere tried failed to get sufficient
start before the summer drought and succumbed to the dry weather.
These will be given further trial and doubtless some will prove of

value.

It is evident that certain of the perennial grasses tried can be used
for permanent pasture. But in case of prolonged periods of drought
it is necessary to supplement the pastures by soiling crops. For this
purpose Japanese barnyard millet, common millet, the sorghums, and
rape can be used to advantage. Smooth brome grass and some of the
other perennial grasses can usually be depended upon for hay. In
addition to these hay can be obtained from the millets and cane.

F. Lamson-Scribner,

Agrostologist.
Approved.

James Wilson,

Secretary of Agriculture.

Washington, D. C, Ajyril^, 1901.

O

CiKCULAR No. 34— (Agros. 85.) Issued May 24, 1901.

United States Department of Agriculture,

DIVISION OF AGROSTOLOGY.

[GRASS AND FORAGE PLANT INVESTIGATIONS.]

F. LAMSON-SCRIBNER, Agrostologist.

ARISTIDA PURPUREA NUTT., AND ITS ALLIES.

The species of Aristida belonging to the section with very une-
qual empty glumes, occurring in the United States, never have been
clearly understood, and an attempt is here made to show the rela-
tionships of the many species, or supposed species, which have been
described by various authors. While in some instances the conclu-
sions are not wholly satisfactory, and some species are interpreted to
include many different forms, it has been thought that the only
proper way to treat this group is on broad lines. Under the present
tendency to multiply species, some authors would doubtless add
many new species in this group ; but we are of the opinion that such
a multiplication of species would add to rather than lessen the diffi-
culty in determining the limits of the several species. There appears
to have been a tendenc}^ to overlook old published species which, by
reason of their very short or imperfect descriptions, are doubtfully
known, and in the present paper an attempt is made to correlate
the old species and varieties, with the belief that, even if these are
wrongly interpreted, such work is often of more value than promis-
cuous publications of many new species with no attempt to determine
the limitations of those already published. Among the old species
and varieties in this group are Aristida americana L,., A. fascicu-
lata Torr., A. purpurea Nutt., A. dispersa T. & R., A. purpurea
berlayidieri T. & R., A. purpurea hookeri T. & R., A. fendle-
riana Steud., A. longiseta Steud., A. roiiieriana Scheele, A. cequi-
ratnea Scheele, A. filipendula Buckle}', and A. curtiseta Buckley.
— F. Lamson-Scribner.

There are in the U. S. National Herbarium the types of the several
species and varieties described by Dr. Vasey, and also cotypes of the
following species: Aristida longiseta Steud. (978 Fendler), A.
fendleriana Steud. (1)73 Fendler), A. purpurea berlandieri T. &R.
(1777 Berlandier), A. purpurea hookeri T. & R. (293 Drummpnd),
and, apparently, A. cequiramea Scheele (562 Lindheimer). Through
the courtesy of Mr. Stewardson Brown, curator of the botanical sec-
tion of the Philadelphia Academy of Natural Sciences, we were able
to examine the types of A. purpurea Nutt., A. jiJipendida Buckley,
and A. curtiseta Buckley. Mr. George V. Nash kindly matched
specimens with the type of Aristida fasciculata Torr., in the
herbarium of Columbia University.

Of the above-mentioned species, Aristida americana L., was first

2

taken up for the plant which lung passed under the name of Aris-
tida disj^ersri T. & R., by Scribner.^ A. fasciculata Torr., is a
synonym of this species. A. fasciculata and A. purpurea ha^ve
previously been confused, although they are not even closely related.
The confusion of these two species doubtless occurred from the fact
tlvAt fasciculata was interpreted to apply to habit of growth rather
than to the inflorescence. Nearly all of the other species here con-
sidered have at one time or another been referred to A. purpurea
or A. fasciculata, either as synonjmis or as varieties. A. cequi-
ramea Scheele, treated here as a variety of Aristida purpurea^ is
perhaps worthy of specific rank.

In studying this group, one of the chief difficulties has been the
large amount of young or imperfectly developed material in the
herbarium, which, in many cases, we found impossible to determine
satisfactorily ; and, in addition to this, the great variability in habit,
doubtless due to environment, added to the difficulty in determining
the limits of the several species. It is hoped that the present
paper will at least lead to some uniformity in determining the species
here considered, as heretofore the names Aristida piupurea, A.
fasciculata,, A. fendleriana, and A. longiseta, etc., have been
variously applied, first to one form, then to another.

A. Culms branched, awns not exceeding 2 cm. in length.

Aristida americana Linn. Amoen. Acad. 5:393. 1759. (Pugill. Jam. PI.)
(Aristida fasciculata Torr. Ann. Lye. N. Y. 1 : 154. 1834; Cha'ta)-ia fascicu-
lata Schult. Mant. 3:578. 1826; Aristida dispersa Trin. & Rupr. Agrost.
3:129. 1842.)

The original description of this species in Amoenitates Academicae is as follows :
" Aristida americana. (j7'a«!cn Festncam referens. Cnlmns ramosus. Differt
ab Aristida ascensionis quod radii panicuJa' sint indivisi. Spiculis alternis
e flosculis sessilibiis, quornm valvula calycina purpurascens est."

Munro'^ says regarding the specimen in the Linnean Herbarium: ''Aristida
americana L. Am. Acad. 5 : 393. From Jamaica, Browne. This is called
.4. disjjersa Trin. \ but Linnteus's name ought to take precedence. Kunth
has misplaced the Linnean synonym in Enfriana juncifolia." Linna?us
compares Aristida americana with his A. adscensionis, to which the form
here considered, is closely related, not only in habit but also in spikelet char-
acters. The description of Aristida americana in later editions of Linnaeus's
works is very different from the original one given above, and this may
explain why Kunth referred this species to Eutriana (Bouteloua).

The type of Aristida fascicidata Torr., in the Herbarium of Columbia University,
bears the label, "Aristida fascicidata Torr., Forests of the Canadian,
Long's Expedition, Dr. James," and has been variously interpreted by dif-
ferent authorities. This specimen is well matched by No. 66 E. Palmer,
Guaymas, Mexico, 1887, distributed as Aristida dispersa Trin. & Rupr.

' Bui. Torr. Bot. Club 9:87. 1888.

= On the irlentiflcatlon of the Grasses of Linnaeus's Herharium, Journ. Linn. Soc. Bot. 6:149.

1862.

Aristida amerieana Linn., is an extrenielj- variable species and extends from
Kansas to Arizona, Lower California, Mexico, and the West Indies. It is
a ctespitose, much branched perennial 'i to 6 dm high, with rather short
involute leaves and many-flo\vered i)anicles 5 to 15 cm. long. Spikelets 8 to
10 mm. long, fasciculate; empty glumes usually purplish, uneiiual, the first
one-half to two-thirds as long as the second, which about equals the flower-
ing glume. Awns subequal, 1 to 1.5 em. long.
Specimens Examined. — A'rtH.sa.s; sandy soil, Comanche Coimty, 888 A. S. Hitch-
cock, 1896. re.ra.s-.- " Western Texas," 741 C.Wright, October, 1849, also
30 V. Havard, September, 1883, Limpia Canyon, 163 G. C. Nealley, Sep-
tember, 1892; Abilene, H. L. Bentley, 1899. Neir Mexico: Mangos Valley,
J. G. Smith, September 22, 1896; Organ Mountains, G. R. Vasey, 1881.
Arizona: Fort Huachuca, Dr. T. E. Wilcox. 1894; Santa Catalina Moun-
tains, 13 J. W. Tourney, April 3, 1894. Mexico : Guaymas, 66 E. Palmer,
1887; Mexican Boundary, near White Water, 2264 E. A. Mearns, September
11, 1893; Saltillo, State of Coahuila, 388 E. Palmer, September, 1898; also
1352, E. Palmer, 1880; Durango, State of Durango, 535 E. Palmer, August,
1896; Chihuahua, 390 C. G. Pringle, August, 1885. Loicer California: 504
E. Palmer, 1887. Dutch West Indies: St. Croix, 64 A. E. Ricksecker,
November 11, 1895. Forto Rico: 'o^uanica, 3438, 3766 P. Sintenis, 1886.
Aristida amerieana bromoides (H. B. K.) Scribn & Merrill, U. S. Dept. Agr.,
Div. Agros., Cir. 32:5. 1901. [Aristida bromoides B.. B. K. Nov. Gen. et
Sp. PI. 1:122. 1815; Aristida dispersa bromoides Trin. &. Rupr. Agrost.
3:130. 1842; C'/icefaHa 5romoides R. & S" Syst. 2 : 396. 1817).
A low tufted form 1 to 3 dm. high, with slender, simple or branched culms,
short involute leaves and densely-flowered purplish or pale panicles 2 to 5
cm. long. Spikelets as in Arnstida amerieana.
In dry soil, Texas to California (Mexico).

Specimens EXAMINED : Texas: Guadaloupe Mountains, 67 V. Havard, 1881. New
Mexico : between Santa Fe and Canoncito, 3754 A. A. & E. G. Heller, June
33, 1897; Santa Fe, 1325 A. L. Mulford, September 16, 1895, Socorro, G. R.
Vasey, 1881. Arizona: Santa Catalina Mountains, 14 J. W. Toumey, April
20, 1894; near Bisbee, 1027 E. A. Mearns, October 5, 1892, also 861 Mearns,
September 14, 1892; Quitovaquito, 2759 E. A. Mearns, January 30, 1894; Con-
gress, C. R. Orcutt, April, 1896; Ehrenberg, W. M. Canby, September, 1874;
Valley of the Colorado, 542 E. Palmer, 1876. Nevada .- Rioville, 5034 M. E.
Jones, April 12, 1894. California: near Saratoga Springs, 259 Coville &
Funston, February 2, 1891, The Needles, 3788 M. E. Jones. May 3, 1884;
Santa Catalina Islands, T. S. Brandegee, May 16, 1890. Lower Calif urnia :
near Indian Wells, 20:53 C. R. Orcutt, February 5, 1890: San Julio, 12 T. S.
Brandegee, 1889; Guadalupe Island, 669, 675 E. Palmer, 1889; Cedros Island,
665 Palmer, 1889; Santa Rosalia, 270 E. Palmer, 1890; no locality, 503 E.
Palmer, 1887. Mexico : Sonora, Guadalupe Canyon, 2033 E. C. Merton, Aug-
ust 27, 1893; Guaymas, 270 E. Palmer, 1890; Valley of Mexico, Federp.l
District, 6227 C. G. Pringle, November 4, 1895.
This variety is distinguished from Aristida amerieana by its smaller size, smaller
panicles, usually less branched culms and more ctespitose habit of growth.
We have assumed this form to be Aristida bromoides H. B. K., but may be
wrong in so interpreting it. The form here considered is doubtless Aristida
wa?ia Steud. Norn. ed. 2. 1:131. 1840.

B. Culms simple, awns 2 to 8 cm. in length.

Aristida longiseta Steud. Syn. PI. Gram. 420. 1855. [Aristida fascicidata
nuttallii Thurb. in Beal, Grasses N. A. 2:208. 1896.)

A densely tufted glaucous, glabrous perennial 1 to 4 dm. high, with numerous
involute, basal leaves, erect simple culms and long, usually purplish setae.
Sheaths shorter than the internodes, striate ; ligule a ciliate fringe; leaf-
blades strongly involute, wiry, 2 to 11 cm.' long, the basal ones numerous,
often recurved, culm leaves one or two, similar to the basal ones. Panicles
few-flowered, exserted, the branches solitary or 2 or 3 at the lower nodes,
ascending, at least in herbarium specimens, each bearing 1 to 3 or 4 flowers.
Spikelets rather large, purplish; empty glumes unequal, the first shorter
than the flowering glume, slightly scabrous on the keel, acute or with a
very short mucronate tip, 11 mm. long; second glume much exceeding the
flowering glume, about 20 mm. long, otherwise like the first; flowering
glume about 14 mm. long, including the densely pubescent callus, which is
about 1 mm. long, smooth below, scabrous above. Setae nearly equal,
scabrous, about 7 cm. long.

Type locality. New Mexico, No. 978 A. Fendler, 1847.

General distribution. — In dry soil, South Dakota to Montana, south to Texas,
New Mexico, and Arizona (Mexico). May to September.

Specimens examined.— South Dakota: Stearns, 33 E. J. Wallace, August 2,
1896. Montana: Upper Big Horn River, 184 J. W. Blankinship, July,
1890. Wyoming: Devil's Tower, 544 David Griffiths, August 13, 1897.
Colorado: Grand Junction, 5476 M. E. Jones, June 21, 1894; Denver, 11
J. M. Holzinger, 1896, also 31 Alice Eastwood, July, 1891 ; Walsenburg 790
C. L. Shear, July 10, 1896; Fort Collins, E. D. Ball, June 26, 1898, also 518
C. S. Crandall, July 1, 1892; Ar boles, 14 C. F. Baker, June 29, 1899; Jules-
burg, 24 E. N. Plank, June, 1896; McElmo Canyon, \lh Alice Eastwood,
June, 1892. Utah: Springdale, 5249 M. E. Jones, May 16, 1894; Silver Reef,
5163 M. E. Jones, May 4, 1894; Antelope Island, 1297 S. Watson, June,
1869. Nebraska : North Platte, 2025 P. A. Rydberg, June 22, 1895, also 279
C. L. Shear, same date ; near Thedf ord, 1 300 P. A. Rydberg, September 8,
1893; Lavaca, 817 J. M. Bates, July 14, 1898; Longpine, 1121 J. M. Bates,
July 23, 1898. Kansas: Rockport, E. Bartholomew, June 12, 1889; Osborne
City, 158 C. L. Shear, July 9, 1894; Ellsworth County, 589a A. S. Hitchcock,
July 12, 1895; Riley Cr-unty, 263 Hitchcock, July 15, 1895; Ulysses, 63 C. H.
Thompson, June 27, 1893. Texas: San Diego, J. G. Smith, June 11, 1897;
Kerrville, J. G. Smith, June 22, 1897; Palestine, 49 E. N. Plank, April, 1895.
Arizona: Tucson, J. W. Toumey, 1892; "North Arizona," J. G. Lemmon,
1884; Port Apache, 575 E. Palmer, June, 1890; Ash Fork, 16, 17 J. W.
Toumey, Jime, 1892. New Mexico: No locality, 978 A. Fendler, 1847
(cotype); Rosa, 152 C. F. Baker, June, 1899; Mangos, J. K. Metcalfe, May,
1897; White Sands, 404 E. O. Wooton, August 28, 1897; Santa Fe, 3535
A. A. & E. G. Heller. May 15, 1897; Socorro, 74 E. N. Plank, August, 1895.
Mexico : Chihuahua, 473 C. G. Pringle, October 8, 1885.

This widely distributed species has been variously referred to Aristida purpurea,
A. fendleriana, A. fascicidata, etc., but is readily distinguished by its
simple, erect, slender culms, few, large spikelets, acute or but slightly
mucronate empty glumes and long, purplish setae.

Steudel describes Aristida longiseta as being but 3 or 4 inches high, but in all
other pai ticulars his rather full description applies to the form here consid-
ered. It is evident that Steudel based his description on a very small form.
Fendler's No. 978 in the U. S. National Herbarium is 3 dm. high, while the
same number in the herbarium of the Philadelphia Academy of Natural
Sciences is but 1.5 dm. high.

ARISTIDA LONGISETA ROBUSTA var. nov.

A stout, densely tufted form, 8 to 4 dm. liigh, with robust culms, more rigid
inflorescence and usually longer leaves than in the species. Empty glumes
as in the species except that both are prominently cleft at the apex, bearing
in the cleft a scabrous awn about 2 mm. long: flowering glume nearly
smooth or only slightly scabrous.

General distribution.— In dry soil, South Dakota to Kansas, west to Wash-
ington and British Columbia. June to Aiigust.

Specimens examined.— So^j^A Dakota: Aurora Coianty, 35 E. N. Wilcox,
August 18, 1890; Redfield, 68 David Griffiths, August 29, 1896; Belle
Fourche. 404 Griffiths, August 4, 1897; Canning, 43 Griffiths, August 27,
1896; Lebanon, 252 Griffiths, July 23, 1896; Custer, 1118 P. A. Rydberg,
August 16, 1892; Rosebud, 34 E. J. Wallace, July 26, 1896. Iowa: Rock
Rapids, 410 C. R. Ball, July 6, 1897. Nebraska: St. Helena, 2G29, F.
Clements, June 27, 1893; Longpine, 1123 J. M. Bates, August 18, 1898;
Osborne City, 156 C. L. Shear, July 9, 1894. Wyoming: Clear Creek, 110
Williams & Griffiths, August 5. 1898; Devil's Tower, 517 Griffiths, August
12, 1897 ; Whalen Canyon, 54 0 A. Nelson, July 19, 1894. Montana : Billings,
222 Williams & Griffiths, August 30, 1898; Glendive, L. F. Ward, July 21.
1883; Missoula, 250 Williams & Griffiths, September. 1898; Indian Creek, 336
F. L. Scribner, July 5, 1883 (type) ; Prickly Pear Canyon, 559 R. S. Williams,
July 23, 1897. Idaho: T. E.Wilcox, 1881. CWorado ; Colorado Springs,
496 S. M. Tracy, 1887. Washington: Alma, 536 A. D. E. Elmer, July, 1897;
Rock Island. 439 Sandberg & Leiberg, July, 1893; Spokane, 906 Sandberg,
Heller & McDougal, August 16, 1892, also 2597 C. V. Piper, June 25, 1897.
British Columlna : Spencer's Bridge, J. Macoun, May 28, 1889.

This variety can be readily distinguished from the species by its more robust
culms, and especially by its rather prominently awned empty glumes.

ARISTIDA LONGISETA HOOKERI (Trin. & Rupr.). {Aristida j^urpurea
hookeri Trin & Rupr. Agrost. 3: 107. 1842).

A robust form 4.5 to 7 dm. high, with longer leaves and stouter culms than in
the species. Leaf -blades rigid, involute, 10 to 20 cm. long. Spikelets some-
what larger than in the species, the empty glumes entire or only slightly
toothed at the apex ; flowering glume glabrous, or nearly so.

General distribution.— In dry soil, Texas, New Mexico, and Arizona. May
to August.

Specimens examined. — Texas: No locality, 293 Drummond (cotype); Llano, J. G.
Smith, June, 1897, also 6 E. N. Plank, August, 1892; San Antonio, 45 Plank,
May, 1893, also V. Havard, 1882; no locality, 11 J. Reverchon, 1881. New
Mexico : Mesilla Park, 42 E. O. Wooton, Jime 19, 1897. Arizona : no local-
ity, J. G. Lemmon, 1884.

This variety is evidently related to Aristida longiseta rather than to A. x>nr-
purea, and is distingiiished by its larger size, longer leaves, and glabrous
flowering glumes.

ARISTIDA LONGISETA rENDLERIANA (Steud.). {Aristida fendleri-
ana Steud. Syn. PI. (4ram. 420. 1855; ,1. 2)nvparea fendleriana Va.sey
Contr. U. S. Nat. Herb. 3: 46. 1892; .1. fasciculata fendleriana Vasey in
Beal, Grasses N. A. 2: 207. 1896.)

A densely tufted form 2 to 4 dm. high, with simple, erect cnlms, numerous invo-
lute basal leaves and much shorter settf than in the species. Panicles strict,
few-flowered. Empty glumes unequal, acute, generally cleft at the apex and
bearing a short awn ; second glume ecjualing or slightly exceeding the some-
what scabrous flowering glume. Setaj subequal, 2 to 3 cm. long.

6

General distribution. — In dry soil, Montana and Wyoming, south to Texas,
Arizona, and California.

Specimens examined. — Montana: Billings, 223 Williams & Griffiths, August 30,
1898. Wyoming : Pine Bluff, 3617 A. Nelson, July 6, 1897. Colorado : Grand
Junction, 5469 M. E. Jones, June 21, 1894; Buena Vista, 605 C. L. Sheldon,
July 6, 1892; Durango, 12 Alice Eastwood July, 1891 ; La Porte, L. H. Pam-
mel, June 25, 1896 ; Salida, 953 C. L. Shear, August 2, 1896 ; Colorado Springs,
2113 T. A. Williams, July 20, 1896; Trinidad, 15 S. M. Tracy, 1887. Utah :
Glenwood, 70 L. F. W^ard, May 22, 1875; "South Utah," M'. E. Jones, 1880.
Texas: Dallas, 769 E. Hall, 1872; El Paso, M. E. Jones, April 21, 1884; no
locality, G. C. Nealley, 1887; J. Reverchon, 1880. New Mexico: 2015 C.
Wright, 1851-52; 973 A. Fendler, 1847 (cotype). Arizona: Cosnino, 4046 M.
E. Jones, August 9, 1884; Tucson. J. W. Tourney, 1892; Mormon Lake, 91
D. T. MacDovigal, Jime 11, 1898; Mt. Elden, 355 MacDougal, July 11, 1891;
Moki Reservation, 9 W. Hough, 1896; no locality, E. Palmer, 1869. Cali-
fornia: The Needles, 133 M. E. Jones, May 6, 1884.

This variety is fairly constant in its characters and is distinguished by its short
seta? and short second glume, which about equals the flowering glume in
length. In Aristida longiseta the second glume much exceeds the flowering
glume.

Aristida purpurea Nutt. Trans. Am. Phil. Soc. 5: 145. 1837. {A. purpurea
berlandieri Trin. & Rupr. Agrost. 3: 107. 1842; A. romeriana Scheele,
Linna-a22: 344. 1849.)

A tufted, glabrous perennial 3 to 6 dm. high, with slender culms, involute leaves
and lax pale or purplish panicles. Culms simple, leafy, glabrous, sheaths
shorter than the internodes; ligule a short ciliate fringe: leaf -blades 5 to 10
cm. long, involute, rigid, slightly scabrous. Panicles 5 to 15 cm. long, the
branches at first erect or ascending, often becoming somewhat spreading
and flexuous, short, solitary or two or three at each node, few-flowered.
Spikelets pale or purplish; empty glumes unequal, the first lanceolate, 6
mm. long, cleft at the apex ; the teeth acute, about 1 mm long, bearing
between them a scabrous awn about 1.5 mm. long. Second glume 12 mm.
long, exceeding the flowering glume, cleft at the apex; the teeth 1.5 mm.
long, the avvni 2 mm. long. Flowering glume 10 mm. long, smooth below,
tuberculate roughened above; callus bearded, about 0.5 mm. long. Awns
equal, purplish, 2 to 4 cm. long (description of spikelet characters drawn
entirely from the type specimen).

Through the courtesy of Mr. Stewardson Brown, curator of the botanical f^ection
of the Philadelphia Academy of Natural Sciences, we have been able to
examine the specimen on which Nuttall based his Aristida purpurea.

Specimens examined.— Indian Territory : 387 E. Palmer, 1868. Texas: Presidio,
66 V. Havard, 1881; San Angelo, J. G. Smith, July 12, 1897; San Angelo
to Stirling City, 382^ W. L. Bray, May 20, 1899; Austin, 767, 768 E. Hall,
May 12, 1872; Bracket, 267 W. M. Canby, March 21, 1900; Turtle Creek,
Kerr Coimty, 166 W. L. Bray, May 1. 1899; El Paso, 17a M. E. Jones, April
18, 1884; Llano, J. G. Smith, June 29, 1897; Laredo, 104 G. C. Nealley, Au-
gust, 1892; Bastrop, 43 E. N. Plank, July, 1892; Gregory, 1579 A. A. Heller,
April 14, 1894 ; Beeville, J. G. Smith, May 22, 1897 ; no locality, 3 S. B. Biick-
ley, 1883; 949 Berlandier; 1777 Berlandier (cotype of Aristida purpurea ber-
landieri Trin. & Rupr.). New Mexico: Eddy, J. G. Smith, August 4, 1897.
Arizona: Ash Fork, 47 H. H. Rusby, May 20, 1883 (exactly matches the
type) ; no locality, Wheeler's expedition, 1872.

Aristida purpurea Nutt., has been variously interpreted by different authorities

and many very different forms have been referred to it. It apparently
presents constant characters and, so far as onr herbarium material shows,
is confined to Indian Territory, Texas, New Mexico, and Arizona. It is
readily distinguished from Arisfida IcDigiseta Steud., by its lax halnt of
growth, slender culms, smaller spikelets. shorter awns, tuberculate-rough-
ened flowering glume and awned empty glumes.

Aristida purpurea micrantha Vasey, Contr. XJ. S. Nat. Herb. 3: 47. 1893. {Ai'is-
tidafasciciilata micrmithn Beal, Grasses N. A. 2: 207. 189(i).

A very slender form with pale, lax panicles, the Hexuous branches capillary,
spreading or reflexed ; spikelets smaller than in the species, the flowering
glume. 6 to 8 mm. long, very slightly scabrous or nearly smooth.

General distribution.— Texas to New Mexico (Mexico).

Specimens examined. — Te.ca.s.- Corpus Christi. 85 W. H. Ravenel, April JiO, 1869;
Western Texas, U3 C. Wright, October, 1849; San Diego, J. G. Smith, June
12, 1897; Del Rio, 82 E. N. Plank, December, 1891 ; no locality, S. B. Buck-
ley, 1883; G. C. Nealley, 1888, 1889 (type); 563 F. Lindheimer, 1846; 2U04
C. Wright, 1851-52.

This variety is distinguished from the species by its smaller spikelets, more lax
panicles, the branches often somewhat spreading and fiexuous, and usually
nearly glabrous flowering glumes.

ARISTIDA PURPUREA ^EQUIRAMEA (Scheele). {Aristida ccqniramea
Scheele, Limu^a 22 : o43. 1849. {Aristida filipendida Buckley, Proc. Acad.
Nat. Sci. Phila. 1862:93. 1862; Aristida purpurea californica Vasey,
Contr. U. S. Nat. Herb. 3:47. 1892; Aristida fascicidata californicaYsbHey
in Beal, Grasses N. A. 2 : 207. 1896. )

A rather stout, tufted, glabrous form 4 to 8 dm, high with simple culms, invo-
lute rigid leaves, erect or ascending panicle branches and very tuberculate-
scabrous flowering glumes. Ligiile a short ciliate ring; leaf-blades 8 to 15
cm. long, 1 to 2 mm. wide, involute. Panicles purplish, the branches capil-
lary, generally erect or ascending, sometimes somewhat flexuose, usually
many flowered, 3-5 at each node. Spikelets pale or purplish; second empty
glume twice as long as the first, equaling the flowering glnme, both cleft at
the apex, the midnerve excurrent as a scabrous awn 1 to 2 mm. long ; flower-
ing glume about 10 mm. long, strongly tuberculate-scabrous. Awns sub-
equal, 5 to 7 cm. long.

General distribution. — In dry soil, Texas to California.

Sppx'IMENS examined.— 2V.to.s-.- no locality, 1409 bis, J. Reverchon; 562 F. Lind-
heimer, 1846; G. C. Nealley, 1887; Abilene, H. L. Bentley, 1899. Arizotia:
J. G. Lemmon, 1882. California: San Jacinto, 1549 S. B. & W. F. Parish,
June, 1882; near San Bernardino, 2123 S. B. Parish, May 15, 1891, also 3668
S. B. Parish, May 20, 1895; Capon Valley, 5474 J. G. Lemmon, May, 1891.

This variety was based on a specimen collected in Texas by Lindheimer, no
number cited, but doubtless the form distributed under 562 as Aristida
cequiramea Scheele, as this specimen agrees perfectly with Scheele's desc-rip-
tion. The plant in the Herbarium of the Philadelphia Academy of Natural
Sciences under this number is identical with the same number in the U. S.
National Herbarium and is labeled by Buckley "Aristida filipendula," and
is the form to which Buckley's description applies. Another specimen in
the herbarium so labeled by Buckley is a form of Aristida americana Linn.
Aristida purpurea californica Vasey, was based on number 1549 S. B. and
W. F. Parish, cited above.

This variety is distinguished from the species by its larger size, usually more
densely flowered panicles, longer awns, more prominently awned empty
glumes and more strongly tuberculate-scabrous flowering glume.

8

ARISTIDA PURPUREA LAXIFLORA var. nov.

A lax, tufted form 3 to 6 dm. high, with open, few-flowered panicles. Panicle
branches mostly solitary, flexuous, capillary, 1 to 3 flowered. Spikelets as
in the species.

Type specimen collected in Texas, no locality. No. 12 J. Reverchon, May, 1881 ;
No. 13 Reverchon, same date, is identical with No. 12. A specimen from
Texas, no locality, collected by G. C. Nealley, in 1887, is referred here and
also a specimen collected at Tucson, Arizona, by J. W. Tourney, in 1893.

This variety is very closely related to var. crquiramea, and is only distinguished
by its very lax, open, few flowered panicles.

ARISTIDA PURPUREA CAPILLARIFOLIA var. nov.

A pale glaucous form about 4 dm. high, with lax, few-flowered panicles and
numerous involute basal and culm leaves 2 to 3 dm. long. First glume 6
mm. long, the second about 14 mm. long, mucronate and slightly exceeding
the scabrous flowering glume. Awns 3 to 4 cm. long.

Type specimen collected in Texas by G. C. Nealley. No locality given.

Aristida reverchoni Vasey, Bui. Torr. Bot. Club, 13:52. 1886.

A densely tufted glabrous perennial, 3 to 5 dm. high, with numerous wiry,
radical leaves 7 to 15 cm. long, and narrow spike-like panicles 10 to 15 cm.
long. Sheaths short; ligule a short ciliate ring; leaf -blades involute,
curved. Panicles purplish, the branches in pairs, uueqiial, the lower ones
2 cm. long, all appressed. First glume about 8 mm. long, the second 12
mm. long, about equaling the glabrous flowering glume. Awns about 2
cm. long, nearly e(iuai.

General distribution.— Texas to Arizona.

Specimens examined— Te.ras.- Rocky hills, Crockett County, 1237 J. Rever-
chon, May; Comanche Peak, 154 Reverchon, 1882; no locality, G. C.
Nealley, 1887, 18 Reverchon, 1885; Limpia Canyon, 146 G. C. Nealley,
September, 1892; El Paso, G. R. Vasey, 1881; Kerrville, J. G, Smith, Jime
21, 1897. New Mp.xico : Mangos, O. Metcalfe, September 2, 1897. Arizona:
Fort Hnachuca, T. E. Wilcox, September, 1894.

This species, although closely related to other species in this group, and
especially to Aristida longiseta fendleriana, is readily distinguished by its
narrow spike like panicle and relatively longer first glume.

Aristida reverchoni angusta Vasey, Contr. U. S. Nat. Herb. 3:46. 1892.

A form distinguished from the species by its much narrower panicles, a charac-
ter doubtless due entirely to habitat.

Specimens examined.— Tea-os; Comanche Peak, 10 J. Reverchon, Septem-
ber, 1881 (type) ; San Angelo, J. G. Smith, July 12, 1897; no locality, G. C.
Nealley, 1889.

Aristida curtiseta Buckl. Proc. Acad. Nat. Sci. Phila. 1862:92. 1862.

It is impossible to identify this species, as the specimen on which Buckley
based his description is a form in which the spikelets are aborted by a
species of Ustilago. It is probably Ari.stida pmyurea Nutt., or some
closely related form. Buckley's type is in the herbarium of the Philadel-
phia Academy of Natural Sciences.

Elmer D. Merrill,
Assistant in Charge of Collections.
Approved :

James Wilson,

Secretary of Agriculture.

Washington, D.C., 3Iay 3, 1901.

Circular No. 3o.— (Agros. S(j.) (Issued .June 18, 1901.)

United States Department of Agricultnre,

DIVISION OF AGROSTOLOGY.

(GRASS AND FORAGE PLANT INVESTIGATIONS.) j ffJf^/^liJY

F. Lamson-ScriBiNKU, Ar/rostolof/isL NEW YOWK

BOTANICAL

AfiROSTOLOfilCAL NOTES. GARDEN

INTRODUCTION.

Agrostological Notes was the general title of Circular No. 27 which contained
several short papers of purely technical nature and the same title is here adopted
for a series of papers of similar character as follows:

1. Some Species of Grasses published by by S. B. Buckley.

2. Notes on Calamovilfa.

3. Three New Species of Panicuvi.

4. A new Species of Poa.

5. Some Changes in Nomenclature.

1. SOME SPECIES OK «R.\SSES IM B1,ISHKD l!V S. B. BllKLEY.

By Elmer D. Merrill.

In ISGG Mr, S. B. Buckley published a small octavo volume of 81
pages, with an appendix of 4 pages, that was issued from the office
of the "State Gazette" at Austin, Texas, under the title "A Pre-
liminary Report of the Geological and Agricultural Survey of Texas."
The appendix, which is paged separately, is the part of the report of
interest to agrostologists. In it one genus and nine species of Texan
grasses, supposed b}^ the author to be new, are described. These
species, which for some reason were not included in Index Kewensis,
are as follows :

Schleropelta Buckley, Prel. Rept. Geol. and Agrl. Surv. Texas, App. 1. 1866.
=Hilaria H. B. K. Nov. Gen. et Sp. PL 1: 116. LSiT).

Schleropelta stolonifera Buckley, 1. c.=Hilaria ceucliroides H. B. K. 1. c. 117.

Pappophorum (Polyrhaphis) vaginatum Buckley 1. c. This is the earliest pub-
lication of this species and Buckley's name should be retained. Pappo-
phorum apertuiu Munro in Scribn. Bui. Torr. Bot. Club, 9: 148. 1882, is a
synonym. See Merrill, U. S. Dept. Agr. Div. Agros. Cir. 32: 9. 1901.

Cenchrus setoides Buckley, 1. c, 2=C'ej?(7//M.s inyosiiroidcs H. B. K. Nov. Gen.
et. .Sp. Pi. 1: 115. t. .15. 1815.

Calamagrostis (Deyeuxia) longirostris Buckley, 1. c. "Culm erect, a little
scabrous, half a foot high, sheaths a little scabrous, and about ecpial to the
internodes; ligula elongate, membranaceous and subentire; leaves plain or
convolute, scabrous, shortly pubescent above, 4 to 6 inches long, and 2 to 3
lines l)r()ad; panicle terminal, somewhat open and interrupted, .3 to 5 inches
long and 8 to 12 lines broad; rachis terete, pubescent; glumes subequal, lan-
ceolate, acute, longer than the florets, scrabrous on the back, margins and
apex hyaline ; tipper 1 to ;{ nerved, lower palea o to 5 nerved, shortly pilose
at the base, hyaline upon the margin, apex bifid, below the middle long
aristate; awns twisted, jointed and double the length of the florets; lower
internodes of the rachis about 8 linos in length : upper intervals 2 to 3 lines
long; long estrays naked near their bases and densely spiked near their
summits; spikelets about 3 lines long."

The identity of this species is in doubt.

Panicum (Tridachne) saccharatum Buckley, 1. c, 2. This name should be
retained. Panicum Icichnanthum Torr. 1857, not Hochst. 1855, is a
synonym.

Panicum glomeratum Buckley,!. c.,?>.=Panicum phityphylluDi Munro. Buck-
ley's description antedates that of Munro, but the name Panicvm glomera-
tum is preoccupied by Panicum glomeratum Moench, 1794, et al.

Panicum texanum Buckley, 1. c, 3. This is a distinct species and, so far as
known, has no synonyms.

Panicum repente Buckley, 1. c, 3.=Pa?u"cui»i obtusum H. B. K. Nov. Gen. et
Sp. PI. 1:98. 1815.

Panicum ciliatissimum Buckley, 1. c, 4. A very distinct and well-marked
species.

•2. NOTES ON CALAMOVILFA.

By F. Lamson-Scribner and Elmer D. Merrill.

CALAMOVILFA GIGANTEA (Nutt.) n. comb. Calamagrostis gigantea
Nutt. Trans. Am. Phil. Soc. II, 5 : 143. 1837.

A stout, erect glabrous perennial, 10 to 20 dm. high, from a stout creeping root-
stock, with rigid involute leaves and large, open panicles. Sheaths exceed-
ing the internodes, glabrous ; ligule a dense ciliate fringe of hairs about 1 mm.
long; leaf -blades about 3 dm. long, glabrous. Panicles 3 to 7 dm. long, the
branches spreading or ascending, the lower ones often 3 dm. in length, naked
below, flower-bearing toward their extremities. Spikelets 8 to 10 mm. long;
empty glumes unequal, glabrous, acute, the first shorter than the second.
Flowering glume about 8 mm. long, glabrous above and on the margins,
pilose below on both sides of the keel with rather long, appressed white
hairs, basal hairs copious, about one-half as long as the glume. Palea
about equaling the flowering glume, pilose along the margins, otherwise
glabrous.

The type specimen in the herbarium of the Philadelphia Academy of Natural
Sciences is ticketed by Nuttall: " Toxeumia gigantea, Calamagrostis Nutt.,
Arkansas, Salt River." This specimen, although covering the entire sheet,
represents but little more than the panicle of the plant. The entire plant
must have been very large. The panicle is over 2 feet long, and the wide-
spreading branches more than 1 foot in length, which are flower-bearing
toward their extremities, naked below. In many of the spikelets on this
specimen the empty glumes are about equal; in others, the second is often
slightly longer than the first — 10 mm. long, or somewhat less. The flower-
ing glume and palea are equal, the former being pilose -pubescent on the back
for two-thirds its length, the latter pilose-pubescent on the margins.

General distribution in sandy soil, Kansas to Arizona, July to October.

Specimens examined.— 7va?jsa.s; Moonlight, 183 C. H. Thompson, August 15,
1893; Riley County, 904 J. B. Norton, 1896; Stevens Coimty, 344 M. A. Carle-
ton. July, 1891. Oklahoma Territory: Kay County, Mark White. July 23,
1898. New Mexico: No locality or collector given. 1882. Arizona: Hol-
brook, H. H. Rusby, August, 1883; San Francisco Mountains, Loew, 1873;
Chalcedony Park, Myrtle Tuck, October 15, 1897; Moki Reservation, 41 Wal-
ter Hough, 1896.

This species was recognized by Torrey in Marcy's Explor. Red Riv., La. 300.
1853, and also in Pac R. R. Rept. 4^: 154. 1857, but was later reduced to
Calamagrostis lougifolia by Gray. It is closely related to Calamovilfa longi-
folia (Hook.) Hack., and all the specimens cited above have been so referred
in herbaria. It is distinguished by its larger size, large, open panicles,
elongated branches, larger spikelets, and in the pilose flowering glume and
palea.

Calamovilfa longifolia (Hook. ) Hack. True Grasses, 113. 1890: Calamagrostis
loncji folia Hook. Fl. Bor. Am. 241. 1840; Vilfa rigida Buckl. Proc. Acad.
Nat. Sci. Phila. 1862 : 89. 1802 ; Ammophila longifolia Bentli. in Vasey, U. S.
Dept. Agr. Spec. Kept. 63: 29. 188:1

This species e-xteiuls from We.stern Ontario and Indiana westward to Wyoming,
Montana, Idaho, and southward to Iowa, Nebraska, and Colorado. It is dis-
tinguished by its usually contracted panicles and glabrous flowering glume
and palca.

CALAMOVILFA LONGIFOLIA MAGNA var. nov. This variety is dis-
tinguished from the species mainly by its large size, 12 to 18 dm. high; in
this respect approaching Calainovilfa gigantea, but distinguished from that
species by its glabrous spikelets. It is represented in the National Herba-
rium by specimens collected on the lake shore at the mouth of the Kalama-
zoo River, Mich., by W. A. Taylor in 1894. It is very abundant and valua-
ble as a sand binder in the dunes about South Haven and Grand Haven,

Mich.

:t. THKKK SKW SPKCIKS OF PAMCUM.

By F. I.AMSON-ScRiBNER aiid Elmer I). Mkrrill.

PANICUM (VIRGARIA) BARTOWENSE sp. nov.

Panicum proliferum pilomm Griseb. PI. Cub. 232. 1860. •>. Not Panicinn pilos-
uiu Swartz, 1788, et al.

A stout, erect annual 12 to 10 dm. high, with very long leaves, tuberculate-hispid
sheaths and diffuse panicles 3 to 0 dm. long. Culms simple, glabrous, ."> to
7 mm. in diameter below. Sheaths about equaling or somewhat exceeding
the iuternodes; ligule a ciliate fringe about 2 mm. long; leaf-blades 4 to 6
dm. long, about 10 mm. wide, tapering to the acute apex, scarcely narrowed
at the rounded and somewhat clasping base, scabrous on the margins and
on the nerves above, glabrous beneath, or often with few scattered papillate
hairs on both siirfaces. Panicles pale or purplish, the branches at first erect,
finally spreading, the lower ones 3 to 4 dm. long; the primary axis and
branches scabrous. Spikelets nearly 2.5 mm. long, narrowly ovate, acute,
glabrous : first glume truncate, enclosing the base of the spikelet, less than
0.5 mm. long; second and third glumes subequal, acute, slightly exceeding
the flowering glume, both 7-nerved, the nerves rather prominent, the third
glume subtending a hyaline palea about 1 mm. long ; flowering glume less
than 2 mm. long, ovate, acute, smooth and shining.

Type specimen collected in wet, reclaimed swamps at Bartow, Polk County, Fla.
No. 1220, Robert Combs, September 29, 1898; No. 971, Combs, collected in
a water ditch at Homosassa, Citrus County, Fla., September 13, 1898, is the
same; No. 483, A. S. Hitchcock, collected in inundated fields, Myers, Lee
County, Fla. , July-August, 1 900, represents a later stage of the species with
a very diffuse, open panicle: No. 5380, A. H. Curtiss, low grounds, Palm-
beach, Fla., May 13, 1895, belongs here.

This species is closely related to Panicum prolifevum Lam., the most striking

difference being its tuberculate-hispid sheaths and very small first glume.
PANICUM SHASTENSE sp. nov.

A tufted, pubescent perennial about 3 dm. high, with erect or ascending leaves
and open, ovate, rather few flowered panicles. Culms geniculate below,
pilose with scattered long white hairs ; nodes bearded with spreading or
reflexed hairs and with a smooth ring immediately below. Sheaths close,
striate, shorter than the internodes, pilose with long spreading hairs ; ligule
a ciliate fringe about 2 mm. long; leaf-blades narrowly lanceolate, acute,
slightly clasping at the base, 5 to 7 cm. long, 0 to 8 mm. wide, pilose beneath
with weak spreading hairs, nearly glabrous above, scabrous on the margins.

Panicles about 8 cm. long, rachis and branches sparingly iiilose, branches
fasciculate at the nodes, ascending, few-flowered. Spikelcts ovate, obtuse,
2.5 mm. long, pale, or purplish at the base , first glume triangular, pubescent,
less than 1 mm. long ; second and third glumes about equal, densely pubes-
cent with short spreading hairs, the third glume enclosing a hyaline palea
a little over one-half as long as tbe glume; flowering glume ovate, 2 mm.
long.

Type specimens collected in a moist meadow at the edge of pine forests at Castle
Crag, near Mt. Shasta, California, by Louis A. Greata, June, 1899.

Related to the common and widely distributed Panicum uticijjliyUinn Trin.
(P. pubescens of authors, not Lamarck), but is at once distinguished by its
larger spikelets. In some respects it resembles Famcuvi seribnerianum
Nash, but it is not closely related to that species.

PANICUM FUNSTONI sp. nov.

A slender tufted, nearly glabrous perennial 4 to 7 dm. high, with numerous,
ascending leaves, slender erect culms, exserted ovate panicles and small spike-
lets. Culms at first simple, becoming much branched, sparingly appressed-
pubescent or glabrous. Sheaths shorter than the internodes, striate, glabrous
except on the ciliate margins ; ligule a prominent fringe of hairs ii mm. long ,
leaf-blades narrowly lanceolate, those of the primary culm 5 to 7 cm. long,
6 to 8 mm. wide, acute, slightly clasping at the base, glabrous except on the
scabrous margin which is often papillate ciliate at the base, those of the
branches much smaller, crowded, the upper ones often minutely puberulent
beneath. Panicles 4 to 5 cm. long, ovate, the rachis and branches glabrous;
secondary panicles small, few-flowered, mostly enclosed in the upper sheaths.
Spikelets ovate, 1.5 mm. long or slightly less; first glume very small, less
than 0.5 mm. long; second and third glumes rather densely pubescent with
short spreading hairs, the second slightly shorter than the flowering glume.

Type specimen collected on the bank of Kaweah River at Three Rivers, Tulare
Coimty, Cal., No. 1286, Coville & Funston, July 26. 1891. No. 142 Dr. Ezra
Michener, Sacramento, Cal., August 31, 1894, is also referred here, although
it differs from the type in its more lax habit of growth, thinner leaves and
glabrous culms.

This species is very distinct from any other from the P?,cific slope, and is appar-
ently not closely related to any of those of the Eastern States. It is distin-
guished by its tall, slender culms, numerous crowded leaves on the branches,
and small pubescent spikelets, the second glume being shorter than the
flowering glume. "We take pleasure in naming it for General Funston who
first collected it.

It is the Panicmn dicJiotomitm of the Botany of the Death Valley Expedition,
Contr. U. S. Nat. Herb. 4: 216. 1893.

4. A NKW SPECIES OF POA.

POA APERTA Scribn. & Merrill sp. nov.

A lax, loosely tufted, glabrous, glaucous, perennial, 4 to 5 dm. high, from creep-
ing rootstocks, with wiry spreading culms, linear leaves and open, few-
flowered panicles 4 to 8 cm. long. Culms terete, branching below, genicu-
late and rooting at the lower nodes. Sheaths striate, the lower ones mar-
cescent, crowded; ligule membranceous, acute, 2 to 3 mm. long, decurrent
as a rather prominent white sheath margin ; leaf-blades rather firm, striate,
minutely scabrous, glaucous, acute, those of the innovations 10 to 16 cm.
long, about 2 mm. wide, those of the culm shorter. Panicle exserted, ovate
in outline, rigid, branches spreading or reflexed, scabrous, tlie lower ones
2 to 2.5 cm. long, bearing about 3 spikelets, the upper ones solitary, shorter.

5

Spikelets pnrplisli, lanceoljite-ovate, acute, 5 t<. G inin. long, 3 to 4-flowere(l;
empty glumes, lanceolate, acute, glabrous except for the slightly scabrou.s
keel, slightly uneciual, the first 3.5 mm. long, the second somewhat broader.
4 mm. long; flowering glumes lanceolate, minutely pubescent throughout
and prominently webljy-liairy on the keel and lateral nerves below.

Type specimen collected at Telluride, Colo., No. !i«, C. L. and Wm. Shear. Sep-
tember 1, 1900, occasional in small loose tufts, on moist open mountain sides.
Altitude 2940 m.

This species bears a striking resemblance to Poa coinpressa L., in habit of
growtli, but otherwise is not closely related to tliat species. It is readily
distinguished by its terete culms, rigid, spreading or reflexed panicle
brani'hes, and spikelet characters.

... SOMK (II.\N<;ES IX XOMKSi LATl KK.

By El.MKR D. Mkrrii.l.

Panicum striatum (Linn.) Lam. Tabl. Encycl. 1:172. 1791 ; Encycl. 4: 738.
(err. typ., 748. 1797.); Holcits striatus Linn. Sp. PI. 1048. 1753: Gronov.
Fl. Virg. 135. 1739; Panicum gibbum Ell. Sk. Bot. S. C. and Ga. 1: IIG.
1817: Panicum aquaticnm Mnh\. Gram. 120. 1817; Panicum Muhl. 1. c. 127.
No. 39 (sine nomine) ; Panicum tremuhimUchnltes, Neue Ent. 2: 103. 1821 ;
Mant. 2:237. 1824; Syst. 1:319. 1825; Panicum elliottianum Schultes
Mant. 2: 25G. 1824; Panicum hydropliilum Schultes 1. c, 237.

While Lamarck does not inention Hvlcus striatus Linn., in connection with his
Panicnm striatum, his description is so ample as to leave no doubt as to the
identity of the plant he had in mind. Munro' says, ' ' Holcus striatus Gronov.
Virg. 135, is Panicum gibbum Ell." The descriptions of both Gronovius and
Linna?us apply to this species. Panicnm elliottianum was applied to this
species by Schultes, as he believed Elliott's name to be preoccupied by Pani-
cum gibbosum R. Br. Panicum Jiy(lro])liilum Schultes, was based on P.
aquaticum Muhl., which is the same as Elliott's Panicum gibbum.- Panicum
tremulum Schultes, was based on Panicum No. 39 Muhl. Gram. 127 (sine
nomine), and although this species is not represented in Muhlenberg's herba-
rium we do not hesitate in referring it to Panicum striatum. Muhlenberg
queries "An varietas P. aquatici ? sed folia multum angustioria. "

Panicum hemitomum Schultes, Mant. 2:227. 1824. {Panicum dimidiatum
"Walt. Fl. Carol. 72. 1788, not Liim. 1753; Panicum walteri Muhl. Gram.
108. 1817; Ell. Sk. Bot. S. C. and Ga. 1: 115. 1817, not Pursh, 1814; Pani-
cum carolinianiim Spreng. Syst. Veg. 1:310. 1825; Panicum carinatum
Torr. Boston Journ. Nat. Hist. 1: 137. 1835. Panicum digitarioides Carpen-
ter, Steud. Syn. PI. Gram. 75. 1854; Panicum curtissii Chainn. Fl. So. U. S.

573. 1800.
From examination of the herbaria of Muhlenberg and Elliott, it was found that
Panicnm walteri of both these authors is the grass now known as Pauimm
digitarioides.-^ Panicum hemitomum Schultes, was based on P. u-alteri
Muhl., and is the earliest available name. Panicum Carolinian nm Spreng.,
published one year later was ba.sed on Panicum walteri Ell. which is the
same as Muhlenberg's Panicum icalteri. Both of these names antedate J*ani-
cum digitarioides Carpenter, which in any case is untenable as it is ante-
dated by P. digitarioides Raspail, Kunth. Enum. 1:53. 1883.

* On the identification of the Grasses of Linn*us's Herbarium, Journ. Linn.
Soc. Bot. 6:53. 1862.
■^Scribn. & Merrill, U. S Dept. Agr. Div. Agros. Cir. 27: 4. 1900.
3U. S. Dept. Agr. Div. Agros. Cir. 27; 2. 1900. Ibid. Cir. 29:4. 1901.

6

SPOROBOLUS COMPOSITUS (Poir.) n. comb. Agrostis compositis Poir.
in Lam. Encycl. Sui)pl. 1: 254. 1810; Agrostis involutaMnhl. Gram. 72. 1817;
Agrostis longifolius Torr. Fl. U. S. 1 : 90. 1824; Sporoholus longifoluisWood,
Class-book, 77"). 18G1 ; Sjiorobohts asper Vasey, Contr. U. S. Nat. Herb. 3:
59. 1892, not Kmith. 1833.

Agrostis composita Poir., based on a specimen collected by Bosc in "Caroline,"
is evidently the earliest available name for this species, and is here taken up.
Agrostis ini-olnta Muhl. has been previously referred to Sporoholus asper,
but according to the specimens in Muhlenberg's herbarium and to his de-
scription, it is identical with the form described by Torrey as Agrostis longi-
folia, which it antedates.

Bromus pubescens Muhl. in Willd. Enum. 120. 1809; Gram. 169. 1817.

Tliis species, according to the specimen so labeled in Muhlenberg's herbarium,
now deposited in the Philadelphia Academy of Natural Sciences, is identical
with B. purgans Linn., and should be cited as a synonym of that species.

Bromus scabratus Scribn., U. S. Dept. Agr., Div. Agros., Bui. 13 : 46. 1898, is
a synonym of Bromus porteri (Coult. ) Nash. It is based on No. 3800 A.
Neison, Vermillion Creek, Wyo., July 20, 1897.

The above names were not cited in Bulletin No. 23, in which the North American
species of Bromus are monographed.

Chaetochloa hispida Scribn. & Merrill, U. S. Dept. Agr., Div. Agros., Bui. 21:
25, fig. 1^. 1900.

This species was based on a single specimen collected in Cuba by C. Wright,
January, 1865, but has recently been collected in Florida by A. S. Hitchcock,
No. 915, shell mounds, Marco, Lee County, July- August, 1900. This speci-
men is somewhat larger than the type, about 9 dm. high, with slightly larger
spikes, but otherwise the same.

Phalaris paradoxa Linn. Amoen. Acad, 4: 264. 1759; Sp. PI. ed. 2. 1665. 1763.

This species has been introduced into California and is represented by specimens
collected near Princeton, Colusa County, by J. Burtt Davy, May, 1898, and
at San Diego by T. S. Brandegee, January, 1901. It is considered by Jepson,
in his " Flora of Middle Western California," and is apparently quite widely
distributed in that State. It is a native of Mediterranean region of Europe,
Africa, and Asia.

Approved :

James Wilson,

Secretary of Agriculture.

Washington, D. C, June 4, 1901.

O

KEW YORK

Circular No. 36.-(Agros. 87.) ^^\' ^'^^'•■^^•»

United States Department of AfffMifftf

DIVISION OF AGROSTOLOGY.

(GRASS AND FORAGE-PLANT INVESTIGATIONS.)

O

LIST OF THE PUBLICATIONS OF THE DIVISION OF ACJROSTOLOGY.

ANNUAL REPORTS.

(The annual reports of the Chief of the Division of Agrostology
are bound with those of the Secretary of Agriculture, and since
1896 have also been published separately.)

Report of the Agrostologist for 1895. By F. Lamson-Scribner.
From the Report of the Secretary of Agriculture for 1895. Pp.
165-168.

Report of the Agrostologist for 1896. By F. Lamson-Scribner.
From Report of the Secretary of Agriculture for 1896. Pp.
103-108.

Report of the Agrostologist for 1897. By F. Lamson-Scribner.
From Annual Reports, Department of Agriculture for 1897. Pp.
161-166.

Report of the Agrostologist for 1898. By F. Lamson-Scribner.
From Annual Reports, Department of Agriculture for 1898. Pp.
145-152.

Report of the Agrostologist for 1899. By F. Lamson-Scribner.

From Report of the Secretary of Agriculture for 1899. Pp.

113-117.
Report of the Agrostologist for 1900. By F. Lamson-Scribner.

From the Annual Reports of the Department of Agriculture for

1900. Pp. 83-90.

YEARBOOK ARTICLES.

Grasses as Sand and Soil Binders. By F. Lamson-Scribner. Re-
print from the Yearbook of the U. S. Department of Agriculture
for 1894. Pp 421-436, figs. 100-110.

Grasses of the Salt Marshes. By F. Lamson-Scribner. Reprint
from the Yearbook of the U. S. Department of Agriculture for
1895. Pp. 325-332, figs. 75-79.

Canadian Field Peas. By Thos. Shaw. Reprint from the Year-
book of the U. S. Department of Agriculture for 1895. Pp.
223-232, figs. 46-48.

Grass Gardens. By F. Lamson-Scribner. Reprint from the Year-
book of the U. S. Department of Agriculture for 1895. Pp.
301-308, figs. 68-69.

Forage Conditions of the Prairie Region. By Jared G. Smith.
Reprint from the Yearbook of the U. S. Department of Agricul-
ture for 1895. Pp. 309-324, figs. 70-74.

Timothj^ in the Prairie Region. By Thomas A. Williams. Reprint
from the Yearbook of the U. S. Department of Agriculture for
1896. Pp. 147-154, figs. 29-30.

Division of Agrostology. By F. Lamson-Scribner. Reprint from
the Yearbook of the U. S. Department of Agriculture for 1897.
Pp. 160-175, pi. 1.

Lawns and Lawn Making. By F. Lamson-Scribner. Reprint from
the Yearbook of the U. S. Department of Agriculture for 1897.
Pp. 355-372, pis. 8-14.

Leguminous Forage Crops. By Jared G. Smith. Reprint from the
Yearbook of the U. S. Department of Agriculture for 1897. Pp.
487-508, pi. 31, figs. 17-24.

Sand-Binding Grasses. By F. Lamson-Scribner. Reprint from the
■Yearbook of the U. S. Department of Agriculture for 1898. Pp.
405-420, pis. 28-30, figs. 110-120.

Forage Plants for Cultivation on Alkali Soils. By Jared G. Smith.
Reprint from the Yearbook of the U. S. Department of Agricul-
ture for 1898. Pp. 535-550, figs. 125-128.

Succulent Forage for the Farm and Dairy. By Thos. A. Williams.
Reprint from the Yearbook of the U. S. Department of Agricul-
ture for 1899. Pp. 613-626, pis. 59-60.

Progress of Economic and Scientific Agrostology. By F. Lamson-
Scribner. Reprint from the Yearbook of the U. S. Department of
Agriculture for 1899. Pp. 347-366, figs. 5-9.

Our Native Pasture Plants. By F. Lamson-Scribner. Reprint
from the Yearbook of the U. S. Department of Agriculture for

1900. Pp. 581-598, pis. 76-79, figs. 75-85.

BULLETINS.

No. 1. — Notes on the Grasses and Forage Plants of the Southeastern

States. By T. H. Kearney, jr., Assistant Agrostologist. Pp. 28,

figs. 7. 1895.

This bulletin is divided into two parts, the first of which gives an alpha-
betical list of the grasses and other plants of importance as forage found ia the
sections visited, with brief economic notes on each. The second part contains
a classified list of the species found with remarks upon their distribution and
variations in forms.

No. 2. — Fodder and Forage Plants Exclusive of the Grasses. By

Jared G. Smith, Assistant Agrostologist. Pp. 58, figs. 56. 1896.

A revise of this bulletin was published in 1900, containing 86 pages, 46
figures, and 2 plates. The work contains brief descriptions of about 200 species
of native and introduced forage plants. The descriptions are brief and remarks
under each species afford such information as farmers and those interested would
be most likely to wish to know.

No. 3. — Useful and Economic Grasses. By F. Lamson-Scribner.
Pp. 119, figs. 89. 1896.

This bulletin contains an enumeration of the more important grasses of this
and other countries with brief descriptions and economic notes.

No. 4. — Studies on American Grasses. Pp. 43, pis. 5, figs. 15. Feb-
ruary 6, 1897.

This is a technical bulletin and contains the following papers : I. The Genus
Ixophorus, by F. Lamson-Scribner; H. A List of the Grasses collected by Dl'.

E Palmer in the vicinity of Acapulco, Mexico, 1894-95, by F. Lamson-Scnbner ;
III Some Mexican Grasses collected by E W. Nelson in Mexico, 1894-9."), by F.
Lamson-Scribner and Jared G. Smith; IV. Some American Panicums m the
Herbarium Berolinense and in the Herbarium of Willdenow, by Theo. Holm ; V.
Native and Introduced Species of the Genera Hordeum and Agropyron, liy F.
Lamson-Scribner and Jared G. Smith ; VI. Miscellaneous Notes and Descrip-
tions of New Species. The new species of Fourriiera is here described and illus-
trated as well as a number of new species from various sources.

No. 5.— A Report upon the Grasses and Forage Plants of the Rocky
Mountain region. By P. A. Rydberg and C. L. Shear. Pp. 48,

figs. 29. 1897.
This bulletin is based upon the field notes and general observations made by
the authors in Nebraska, Idaho, Montana, Utah, and Colorado during the sum-
mer of 189.1. The first part contains descriptions of the more important economic
species observed and a classified list of ail the species collected with notes upon
their distribution, etc., closes the bulletin.

No. G.— Grasses and Forage Plants of the Dakotas. By Thos. A.
Wilhams. Pp. 47, figs 10. 1897.
This bulletin contains general notes of the forage conditions in the Dakotas,
a general review of the species which are or may be important for forage and a
classified list of all the grasses and forage plants collected or observed in the
Dakotas in 1896 by the author, who was assisted by Messrs. M. A. Brannon,
E. N. Wilcox, and David Griffiths.

No. 7.— American Grasses, I. Pp. 319, illustrations 302. 1897.

A second revised edition of this bulletin was published in 1898, and a third
revise in which the descriptions were entirely rewritten and enlarged, was pub-
lished in 1900.

No. 8.— Studies on American Grasses. Pp. 20, pis. 9, fig. 1. May
1, 1897.
This is a technical bulletin and contains the following papers: "New or
Little-Known Grasses," by F. Lamson-Scribner; and "The Leaf Structure of
Jouvea and of Eragrostis obtusiflora," by Miss E. L. Ogden.

No. 9.— Notes on the Grasses and Forage Plants of Iowa, Nebraska,
and Colorado. By L. H. Pammel. Pp. 47, figs. 12. 1897.
This report contains, first, general observations upon the physical conditions
and important questions relative to forage production in the States named. The
second part is devoted to an enumeration of the more important grasses and
forage plants of these States with economic notes. In the third part the author
presents a classified list of the grasses of Iowa, Nebraska, and Colorado, col-
lected by him during the seasons of 1895 and 1896.

jq-Q. 10. — A Report upon the Grasses and Forage Plants of Central
Texas. By H. L. Bentley. Pp. 38, figs. 14. 1898.
This report contains brief accounts of the physical condition of central
Texas, the early condition of the cattle ranges and popular descriptions and
general observations upon the economic importance of a large number of the
grasses and forage plants native to the region.

No. 11.— Studies on American Grasses. Pp. G2, pis. 17, figs. 12.
July 20, 1898.
This bulletin contains ; " A Revision of the Native American Species of
Calamagrostis," by Thos. H. Kearney, jr. ; and "Descriptions of New or Little-
Known Grasses," by F. Lamson-Scribner.

No. 12.— A Report upon the Grasses and Forage Plants and Forage
Conditions of the Eastern Rocky Mountain Region. By Thos. A.
Williams. Pp. 78, figs. 30. 1898.
This bulletin is based upon the field observations of Mr. Williams and agents
of the Division in 1896 and 1897. The general topographical features of the re-
gion in (question are discussed; also the present aspect of the forage problem, and

the forage conditions on the ranges of South Dakota and northwestern Wyo-
ming. The cultivated grasses and forage plants are treated somewhat in detail,
as are also the native plants of the same class.

No. 13. — The Red Desert of Wyoming and Its Forage Resources.
By Aven Nelson. Pp. 72. pis. 5, figs. 24. 1898.

This bulletin treats of the topography of the soils and climate of the Red
Desert of Wyoming, the characteristic desert plants and forage plants of the
summer range, and closes with a list of the plants of the Red Desert based upon
specimens secured by the author during the summer of 1897,

No. 14. — Economic Grasses. By F. Lamson-Scribner. Pp. 85, pis.
3, figs. 91. 1898.

This is practically a revision of Bulletin No, 3, with the addition of some
new figures and plates. The work closes with a list of grasses for special soils
or uses.

No. 15. — A Report upon the Forage Plants and Forage Resources of
the Gulf States. By S. M. Tracy. Pp. 55, figs. 20. 1898.

This work treats of the soils and natural pastures of the Gulf States, forma-
tion of pastures, the leading forage plants and the more important hay and pas-
ture grasses and miscellaneous forage crops.

No. 16. — Grazing Problems in the Southwest and How to Meet Them.
By Jared G. Smith. Pp. 47, figs. 9. 1899.

This bulletin treats of the former condition of the ranges, their present con-
dition, and the causes which have led to their deterioration, the renewing of the
cattle ranges, the relation of land laws to range improvement, etc.

No. 17. — American Grasses, II. By F. Lamson-Scribner. Pp. 349,
figs. 325. 1899.

A revised edition of this work was published in 1901. This work is a con-
tinuation of Bulletin No. 7, and is uniform with the third edition of that bul-
letin in type and illustrations.

No. IS. — Studies on American Grasses: A synopsis of the Genus
Sitanion. By Jared G. Smith. Pp. 21, pis. 4. June 24, 1899.

Twenty-two species of the genus Sitanion are here described.

No. 19. — Structure of the Caryopsis of Grasses with Relation to
Their Morphological Classification. By P. B. Kennedy. Pp. 44,
pis. S. 1899.

No. 20. — American Grasses, III. By F. Lamson-Scribner. Pp. 197,
figs. 136. 1900.

This bulletin includes descriptions of the tribes and genera and a bibliog-
raphy of the works cited in American Grasses, I, II, and III.

No. 21. — Studies on American Grasses. The native American spe-
cies of Chaetochloa. By F. Lamson-Scribner and Elmer D. Mer-
rill. Pp. 44, figs. 24. March 8, 1900.

In this paper twenty-eight native American species are described, twenty-
three of which are natives of this continent. Six of these species are here pub-
lished for the first time.

No. 22. — Cooperative Experiments with Grasses and Forage Plants.
By P. B. Kennedy. Pp. 86, pis. 13, fig. 1. 1900.

In this bulletin the reports received from farmers and others to whom seeds
were sent for experimental i^urposes from the Division are presented.

No. 23. — Studies on American Grasses. A revision of the North
American species of Bromus occurring: north of Mexico. By
CorneHus L. Shear. Pp. 66, figs. 40. July 3, 1000.

This bulletin contains descriptions of thirty-six species and twenty-eight
varieties of the genus Browns. Three species and fifteen varieties are described
as new.

No. 24. — Studies on American Grasses. Pp. 55, figs. 23. January
9, 1901.

This bulletin contains the following papers : ( 1 ) " Some Recent Collections
of Mexican Grasses," by F. Lanison-Scribner and Elmer D. Merrill; (3) "Notes
on Panicum nitidiim Lam., Panicum scoparium Lam., and Panicnim pnbescens
Lam.," byF. Lamson Scribner and Elmer D. Merrill; (:{) "Miscellaneous Notes
and Descriittions of New Si^ecies, " by F. Lamson-Scribner and Carleton R. Ball.

CIRCULARS.

No. 1. — A Note on Experimental Grass Gardens. By J. G. Smith.
Pp. 4. 1895.

No. 2. — Hairy Vetch, Sand Vetch, or Russian Vetch. By F. Lam-
son-Scribner. Pp. 4, fig. 1. 1895.

No. 3.— Saltbushes. By Jared G. Smith. Pp. 4, figs. 3. 1896.
No. 4. — Renewing of Worn-out Native Prairie Pastures. By Thos.
A. Williams. Pp. 4, figs. 4. 1896.

No. 5. — Cowpeas. By Jared G. Smith. Reprint from the Yearbook
of the U. S. Department of Agriculture for 1896. Pp. 11. 1898.

No. 6. — Cultivated Vetches. By Jared G. Smith. Winter vetch,

hairy vetch, guinea vetch, and Dakota vetch, are described and

illustrated. Pp. 8, figs. 6. 1898. Revised.
No. 7. — Gram. Chick Pea, or Idaho Pea. By J. G. Smith. Pp. 4,

fig. 1. 1898.
No. 8. — Experiments in Range Improvement. By J. G. Smith.

Pp. 5, fig. 1. 1898.

Relates to the experiments conducted at Abilene, Texas, by Mr. H. L.
Bentley, Special Agent of this Division.

No. 9. — New Species of North American Grasses. By F. Lamson-
Scribner. Pp. 7, figs. 24. February 24, 1899.

Fourteen new species are described.

No. 10. — Poa fendleriana and its Allies. By Thos. A. Williams.
Pp. 6, fig. 1. April 15, 1899.

No. 11.— The Flat Pea. By F. Lamson-Scribner. Pp. 6, figs. 3.
1899.

Describes the Flat Pea {Lathyriis sylvestris), its method of cultivation and
uses.

No. 12. — Rape as a Forage Plant. By Thos. A. Williams. Pp. 6,
fig. 1. 1899.

No. 13. — Florida Beggarweed. By Jared G. Smith. Pp. 5, figs. 2.
1899.

No. 14.— Velvet Bean. By Jared G. Smith. Pp. 5, figs. 3. 1899.

No. 15. — Recent Additions to Systematic Agrostology. By F.
Lamson-Scribner. Pp. 10, figs. 5. July 14, 1899.
Gives descrii)tion of genera and cites synonyms not published in Hackel's
'• True Grasses." The contents of the circular are based largely upon a supple-
ment to Engler & Prantl's "Pflanzenfamilien." Fourniera mexicana, Ixophorus
unisetus, Opizia stolonifem, and Pringleochloa mexicana, are illustrated.

No. ID. — New Species of North American Grasses. By F. Lamson-
Scribner. Pp. 6, figs. 2. July 1, 1899.
Nine new species are described. Puccinellia simplex Scribn. and Dactyloc-
teniuvi australiense Scribn. are illustrated.

No. 17.— Crimson Clover. By Thos. A. Williams. Pp. G, figs. 3.
1899.

No. 18. — Smooth Brome-grass. By P. Beveridge Kennedy. Pp. 9,
figs. 2. 1899.

No. 19. — New and Little-Known Mexican Grasses. By. F, Lamson-
Scribner. Pp. 4, fig. 1. January 2, 1900.
Five new species are described.

No. 20. — Experiments with Forage Plants in Ontario. By P. Beve-
ridge Kennedy. Pp. 3. 1899.

No. 21. — Cooperative Range Grass and Forage Plant Expei-iments

at Highraore, South Dakota. By F. Lamson-Scribner. Pp. 10,

fig. 1. 1900.

Describes the experiments being made at Highmore, South Dakota, in
cooperation with the South Dakota Experiment Station.

No. 22. — Grass and Forage Plant Investigations on the Pacific Coast.

By F. Lamson-Scribner. Pp. 7. 1900.

Describes experiments made in 1899 with grasses and forage plants at Walla
Walla and at North Yakima, Washington, in cooperation with the Northern
Pacific Railroad and the Oregon Railroad and Navigation Companies.

No. 23. — -Progrevss of Experiments in Forage Crops and Range
Improvement at Abilene, Texas. By H. L. Bentley. Pp. 20.
1900.

Descriptions of the experiments with grasses- and forage plants made at
Abilene, Texas, under the immediate charge of H. L. Bentley, Special Agent.

No. 24.— Cowpeas and Corn for Silage and Fodder. By W. Gettys,
Athens, Tennessee. Pp. 10, figs. 2. 1900.

No. 25.— Turkestan Alfalfa. By P. Beveridge Kennedy. Pp. 20.

1900.

This circular includes reports on trials with Turkestan alfalfa from those to
whom seed was sent by the Department of Agriculture.

No. 26.— Rescue Grass. By C. L. Shear. Pp. 4, fig. 1. 1900.

Brief description of rescue grass (Bromiis unioloides), and an account of its
cultivation and value.

No. 27.— Agrostological Notes. Pp. 10. December 4, 1900.

This circular contains the following short papers : (1) " Grasses in the Herba-
rium of Dr. H.. Muhlenberg;" (2) "New Species of Eatonia;" (3) "A New
Variety of Panicum nashianum;" (4) " Nomenclatiu-e Notes;" (5) "Notes on
Melica and Stipa."

No. 28. — Grasses and Fodder Plants on the Potomac Flats. By
Carleton R, Ball. Pp. 18. 1900.

This circular gives a brief account of the grasses and fodder plants grown
by this Division on the Potomac Flats during the season of 1900.

;N'o. 29.— Grasses in Elliott's Sketch of the Botany of South Carolina
and Georgia. By F. Lamson-Scribner. Pp. 12, figs. 4. 1901.
The circular is based upon an examination of the grasses in Elliott's herba-
rium by F. Lamson-Scribner and Elmer D. Merrill. Two new species are
described.

No. 30. — New or Little-Known Grasses. By F. Lamson-Scribner.
Pp. 8. March 8, 1901.
Seven new species are described.

No. 31.— Bermuda Grass. By A. S. Hitchcock. Pp. 6, fig. 1. 1901.
No. 32.— Some Arizona Grasses. By Elmer D. Merrill. Pp. 10.
April 22, 1901.
This circular is based upon specimens collected during the summer of 1900
by Profs. R. H. Forbes and David Griffiths, of the Arizona Experiment Station.
Three new species and five new varieties are described.

No. 33.— Range Grass and Forage-Plant Experiments at Highmore,
South Dakota. Report of progress by F. Lamson-Scribner. Pp.
5. 1901.
This circular is a report of the progress of the range experiments carried on

by this Division in cooperation with the South Dakota Experiment Station at

Highmore, South Dakota.

No. 34.— Aristida purpurea Nutt. and Its Allies. By Elmer D. Mer-
rill. Pp. 8. May 24, 1901.
Aristida fasciculata Torr. and Aristida dispersa Trin. are referred to Aris-
tida aviericana Linn. Aristida longiseta Steud. is recognized with three varie-
ties, and Aristida imrpm-ea Nutt. is also recognized with three varieties, two
of which are new.

No. 35. — Agrostological Notes. By F. Lamson-Scribner and Elmer

D. Merrill. Pp. 6. June 18, 1901.
This circular contains the following short articles: (1) " Some Species of
Grasses, "published by S. B. Buckley; (2) "Notes on Calamovilfa;" (3) "Three
New Species of Panicum ; " (4) "A New Species of Poa : " (5) " Some Changes m
Nomenclature."

No. 36. — List of the Publications of the Division of Agrostology.
By F. Lamson-Scribner. Pp. 8. June 29, 1901.

farmers' bulletins.

No. 31.— Alfalfa, or Lucern. By Jared G. Smith. Pp 24, figs. 3.

1895.

Several editions of this bulletin have been published. It contains a full
account of the history, habits of growth, methods of cultivation, harvesting,
etc., of alfalfa.

No. 50.— Sorghum as a Forage Crop. By Thos. Williams. Pp. 20,

fig. 1. 1897.

Treats of the extent of cultivation in the United States, varieties, conditions
of growth, methods of harvesting, etc.

No. 58.— Soy Bean as a Forage Crop. By Thos. A. Williams. Pp.

24. 1898. Revised 1899.

This bulletin treats of the varieties, methods of culture, harvesting, value
as a soiling crop, as a hay crop, as a soil renovator, etc. There is an appendix
on the "Soy Bean as Food for Man," by C. F. Langworthy, of the Office of Ex-
periment Stations.

8

No. 66. — Meadows and Pastures. By Jared G. Smith. Formation

in the Middle Eastern States. Pp. 28, figs. 9. 1898. Revised

1899.

This bulletin treats of grasses as soil builders, fertilizers for grass lands,

methods of preparing the soil, manner of sowing the seed, varieties of hay

grasses, clovers for meadows and pastures, and some giass mixtures.

No. 72. — Cattle Ranges of the Southwest. A history of the exhaus-
tion of the pasturage and suggestions for its restoration. By H.
L. Bentley. Pp. 32, figs. 9. 1898.
This bulletin treats of the eai'ly and present condition of the Texas pastures,
especially those of central Texas, the obstacles to the renewal or improvement
of the ranges, how the ranges may be renewed. There is also a discussion of the
remaining grasses and forage plants native to the Southwest.

No. 89.— Cowpeas. By Jared G. Smith. Pp. 16. 1899.

No. 101.— Millets. By Thos. A. Williams. Pp. 28, figs. 6. 1899.
This bulletin treats of the place of millets among the farm crops, of the cul-
tivated varieties of foxtail millet, barnyard millet, broom corn millet, the cul-
ture of millet, and their uses and value.

No. 102. — Southern Forage Plants. By F. Lamson-Scribner. Pp.

48, figs. 14. 1899.

This bulletin treats of the formation and care of pastures, soiling and fodder
crops, the more important grasses for hay and pasture lands, leguminous forage
crops, and miscellaneous forage plants.

No. 108. — Saltbushes. By P. Beveridge Kennedy. Pp. 20, figs. 9.

1900.

This bulletin treats of the general characteristics of the introduced and
American saltbushes, chemical composition of saltbushes and their feeding
value, and miscellaneous plants for alkali soils.

F. Lamson-Scribner,

Agrostologist.
Approved :

James Wilson,

Secj^etary of Agriculture.

Washington, D. C, June 12^ 1901.

-:>

New York Botanical Garden Librar

illllllilllllllllll

5185 00259 7746

^%