Tải bản đầy đủ - 0trang
II. Acreage, Yield per Acre, and Production in the United States
HALF CENTURY OF WHEAT IMPROVEMENT IN UNITED STATES
also increased considerably in North Dakota and in the Pacific Northwest, especially in Idaho and Washington, generally into drier and less
productive areas. The California acreage and production are only about
half what they were before 1900.
The curve of Fig. 1 showing yields per acre is of special interest
because yields per acre are often used to measure or indicate technological improvements. They are reasonably good indices in countries in
which acreage remains fairly constant or where the productivity of the
new acreage does not differ materially from the old. They may be mis-
FIG.1. Production of wheat in the census years 1839, 1849, 1859, and 1869 and
average acreage, production, and yield per acre by ten-year periods in the United
States from 1870 to 1949.
leading, however, in a country such as the United States, where the
acreage has greatly increased into areas where the conditions for growth
are quite different. If an improvement reduces cost per acre, thereby
permitting a larger expansion on less productive land, average over-all
acre yields may actually be reduced.
The primary objective of the United States farmer has been to grow
more bushels a t a minimum of cost and inconvenience. He takes considerable pride in growing a good crop, but he is vitally concerned with
large yields per acre only to the extent that it contributes to his net income. The profit from wheat relative to that of other crops a1so has a
marked influence. If a new variety or a better cultural method results
5. C. SALMON, 0. R. MATHEWS, AND R. W. LEU-
in larger yields per acre, it may mean that wheat can be grown a t a profit
on less productive land. Given reasonably high prices or prospects of
such prices, the natural tendency is to increase production. A decrease in
crop acreage such as has taken place in the eastern United States may
mean a larger proportion of wheat on better land and hence larger overall yields per acre, even though there may have been no improvement
The fact that yields per acre are based on harvested and not seeded
acres should also be considered. I n some years much wheat wm seeded
tha.t was not harvested, as shown in Fig. 2. The abandonment was espe-
FIQ.2. Acreage of winter and of spring wheat seeded in the United States but
cially heavy during the drought years 1933-1937, when the average exceeded 20,000,000 acres o r 28 per cent of the seeded acreage. Abandonment of winter wheat was heavy in 1912 and 1928, largely because of
winterkilling in Ohio, Indiana, and Illinois ; in 1917, because of winterkilling in Nebraska and Kansas; an d in 1925, because of winterkilling
in Washington, Oregon, Montana, and Texas.
Finally, it should be noted that the environmental conditions under
which wheat is grown today are different from those of fifty years ago,
even in the same areas. I n some cases soil fertility has declined and in
many cases there has been an increase in insect, weed, and disease pests,
in accordance with the well-known principle that concentration and con-
HALF CENTURY OF WHEAT IMPROVEMENT IN UNITED STATES
tinuation of a particular crop favors the pests peculiar to that crop.
Research unquestionably has aided in keeping these under control ; without research, acreages and yields per acre would probably be less than
they were in 1898.
The decline in yields per acre from 1900-1909 to 1930-1939 shown
in Fig. 1 is due largely to expansion into drier arem of the western
Plains and to the unprecedented drought during the mid-thirties. The
marked per acre increase for the 1940-1949 period, on the other hand,
was due in part to more favorable weather, to better vaxieties, to more
extensive use of fertilizers and pesticides in some sections, and to more
timely operations made possible by mechanization. Favorable prices
during this period, which permitted and encouraged the use of technological improvements, and better informed farmers as compared with earlier
periods are some of the other factors that should be recognized. It is
not possible on the basis of available information to evaluate these various factors separately, but some evidence of the importance of particular
ones will be presented later.
Figures 3,4, 5, and 6 give similar information for the principal wheat
states in the southern Great Plains, in the northern Great Plains, in the
Eastern States, and in the Pacific Northwest, respectively. Ea,ch of
these areas is relatively homogeneous as compared with the United States
as a whole.
a. The Xouthern Great Plains. Yields per acre in the southern Plains
have been relatively constant in spite of an enormous expansion in acreage into western Kansas, Nebraska, Oklahoma, and the Panhandle of
Texas. These are areas which up to about World Wa r I were generally
considered too dry to produce whea.t economically. A period of unusually high precipitation from about 1905 to 1915, aided and abetted by
real estate promoters and other enthusiasts, stimulated a n extensive influx of settlers and homesteaders into these drier areas. Many of them
failed because the information on which a sound agriculture could be
based was completely lacking, as we now know.
6. The Northern Great Plains. Excepting the ten-year period ending
in 1949, yields per acre have declined almost constantly in the northern
Great Plains, owing (1) to expansion into drier areas west of the Missouri River and (2) to increasing damage from stem and leaf rust, from
scab, and from weeds in the eastern half of the area. The severe drought
of the mid-thirties plus a severe stem rust epidemic in 1935 and extensive
damage from leaf and stem rust in 1937, 1938, and 1941 are primarily
responsible for low yields during those years. Both the northern and
southern Plains have been favored by above-average rainfall during the
past ten years. However, Heisig et al. (1945) have shown that yield
S. C. SALMON, 0. R. MATHEWS, AND R. W. LEUKIIL
FIG.3. Average acreage, production, and yield per acre of wheat in Kansas,
Nebraska, Colorado, Oklahoma, and Texas for 1866 to 1869 and by ten-year periods
from 1870 to 1949.
FIG.4. Average acreage, production, and yield per acre of wheat in Minnesota,
North Dakota, South Dakota, and Montana for 1866 to 1869 and by ten-year periods
from 1870 to 1949.
HALF CENTURY OF WHEBT IMPROVEMENT IN UNITED STATES
per acre trends in North Dakota and in Kansas for the period 1920-1945
are definitely upward even after they are adjusted for precipitation and
temperature effects. The indicated increase for Kansas during this period is slightly more than 2 bushels per acre and that for North Dakota,
about 4 bushels per acre. An important factor in the northern Plains
is the relative freedom from damage by stem rust since 1940.
o. The Eastern states. I n the eastern United States the acreage of
wheat has constantly declined to but little more than half that of the
peak period, 1880-1889. Yields per acre have constantly increased.
FIQ.5. Average acreage, production, and yield per acre of wheat in the Eaatern States for 1866 to 1869 and by ten-year periods from 1870 to 1949.
partly as a result of growing wheat on the more productive land while
reverting the poorer land to pasture and timber. Baker (1937) showed
a marked decline in the acreage of all harvested crops i n the eastern
United States, especially from 1919 to 1928. Johnson (1929) also mentions elimination of much of the poorer wheat land as one of the reasons
for the increase in yields of wheat in Pennsylvania. Lamb (1932, p. 15)
has shown that in Ohio wheat now occupies a smaller per cent of the
improved land than formerly. Wheat in the Eastern States is grown
almost exclusively in rotations with other crops. As Lamb (1932)
pointed out, the adoption of a rotation system of farming, and then of
longer rotations involving more spring grains, resulted in a decrease in
the wheat acreage. The constant and very material increases in yields
C. SALMON, 0. R. MATHEWS, AND R. W. LEUKEL
per acre in the eastern United States has been due principally to more
fertilizers, better varieties, and more timely operations made possible by
d . The PacificNorthwest. Figure 6 shows the acreage, production,
and yields per acre in Washington, Oregon, and Idaho. Acreage increased constantly until 1920-1929 and then leveled off a t slightly less
than 5,000,000 acres. Yields per acre are relatively high, partly because
most wheat in this area is grown on summer fallow. Yields per acre
gradually increased up to 1930-1939, and much more rapidly thereafter,
owing in part to more favorable weather and also to a considerable de125
FIG.6. Average aareage, production, and yield per acre of wheat in Washington,
Oregon, and Idaho for 1866 to 1869 and by ten-year periods from 1870 to 1949.
gree to better varieties produced during this and earlier periods, and
to better cultural methods.
e. Development of the Durwm Wheat Ilzdwtry. The durum wheat
industry of the United States has been developed almost entirely since
1900. According to Carleton (1900, p. 19) a few thousand acres were
grown before that time in Texas and still are, although the amount is
insignificant in relation to the total. Also a small acreage was grown in
North Dakota from seed brought in by Russian emigrants at least as
early as 1893 (Shepperd and Ten Eyck, 1902). At the present time
more than 85 per cent of the durum wheat is grown in North Dakota,
principally in the eastern part of the state, west of the Red River Valley,
HALF CENTURY OF WHEAT IMPROVEMENT I N UNITED STATES
but extending into western Minnesota and northeastern South DakotaThe principal advantage of the durum wheats is resistance to the races
of leaf and stem rust that prevailed previous to 1950 and a n ability to
outyield common wheats in the durum area.
According to Ball and Clark (1918) durum wheat was first introduced into the United States in 1855 but never took hold, largely because
it lacked a market and partly perhaps because it was tried only in the
Eastern States, where it is not adapted. The early development of the
industry was due largely to the initiative and vision of M. A. Carleton,
cerealist of the United States Department of Agriculture, who made a
FIO.7. Average acreage, production, and yield per acre of durum wheat in the
United States by five-year periods from 1920 to 1949.
trip to Russia in 1898-1899 and again in 1900 and made a thorough study
of the durum wheat production in that country. He was greatly impressed with the similarity of the climate and soil of this area to that of
the Great Plains of the United States and recommended the growing of
durums in the latter area. He was particularly impressed with what he
believed to be their drought resistance, and accordingly recommended
them for the drier portions of the Great Plains. He states (1901, p.
11): “They stand foremost among all wheat crops in their excellent
adaptation to heat and drought.” In a map of the United States published in 1900, he pictured the durum wheat area as comprising north
central Texas, western Oklahoma, and extreme southwestern Kansas. I n
S. C. SALMON, 0. R. MATHEWS, AND R. W. LEUKEL
a later ma,p (1901,p. 20) the recommended area extends in a wide belt
comprising most of the Great Plains from Mexico to the Canadian border.
As we now know, durums are not so resistant to heat and drought as
many adapted common wheats. They produced slightly higher average
yields than the common spring wheats in the drier areas, but the difference was not enough to offset the difference in market price. Carleton
recognized their leaf-rust resistance but considered this relatively unimportant. He did not realize that they were also resistant to stem rust
until after the severe and widespread epidemic of stem rust in 1904.
FIG. 8. Average prices of hard red spring and durum wheats at Minneapolis,
Minnesota, for 1903 to 1905 and by five-year periods from 1906 to 1950. (Prices
for hard red spring wheat are for No. 1 Northern Spring, and for durum, No. 2
Amber Durum from 1903 to 1933 and No. 2 Hard Amber Durum from 1934 to 1950.)
Official estimates of the acreage and production of durum wheat as
distinct from hard red spring wheat were first made for North Dakota,
South Dakota, and Minnesota in 1909 and for the United States as a
whole in 1919. Carleton (1901) estimated that proba.bly 75,000 to
100,000 bushels were produced in 1901,and Ball and Clark (1918)state
that about 50,000,000 bushels were produced in 1906. The acreage, production, and yield per acre from 1919 to 1950 are shown in Fig. 7.
Acreage and production, it will be noted, reached a peak in the period 1925-1929 and then declined sharply to about half of the maximum,
principally because of a marked price differential in favor of the common spring wheats. Another important fact, as pointed out by Waldron
HALF CENTURY OF WHEAT IMPROVEMENT IN UNITED STATES
(1947), was the development of early-maturing, rust-resistant, highyielding common spring wheats beginning with CERES, eliminating most
of the advantages enjoyed by durum wheats. The prices were much less
for durum than for common spring wheat until about 1911, then nearly
equal up to and during World War I, followed with a decline soon after
the war, as shown in Fig. 8. Since 1930 the price of durum has been
about equal to, or above, that of hard red spring wheat.
The relatively low yields per acre during the 1930-1934 and 19351939 periods were due largely to unfavorable weather. The higher yields
for the 1940-1944 and 1945-1949 periods were due to more than usually
favorable weather and to the absence of severe damage from stem rust.
2. The Preresearch Era
Mr. William White has suggested that if one wants to study capitalism one of the best ways is to go where there isn’t any. Much can be
learned about research in the same way. Since wheat production before
the present century in what are now the principal producing areas was
largely without benefit of research, a study of this period should be informative. Malin (1944) has provided useful background information
for such a study in his agricultural history of four counties (Riley,
Qeary, Dickinson, and Saline) in east central Kansas, and much additional information is available in the reports of state boards of agriculture and agricultural experiment stations of the various states, in the
reports of the United States Department of Agriculture, and in the farm
press. Kansas is an especially fertile field for study because it is now
the leading state in wheat production, because it was one of the first
of the states of the Great Plains to be settled, and because the problems
of the early Kansas wheat grower were at least as numerous, aa important, and as difficult of solution as those of any other state.
a. Early Developments in Kansas. According to Malin, wheat was
grown in eastern Kansas as early as 1839, when 100 acres near Topeka
were harvested. It was seeded near Junction City, about 140 miles west
of Kansas City, in the fall of 1855, and increased rapidly thereafter in
the surrounding territory. Settlers coming to Kansas brought seed supplies, implements, and the methods and philosophies that prevailed in
the country from which they emigrated. Most of these were poorly
adapted to the new environment. Some adjustments or changes were
made quickly, but others came about very slowly.
Slow but powerful oxen broke much of the prairie for the earliest
settlers, but horses and mules were soon substituted for them. The tractor was not used for land preparation until well along in the present
century. Broadcasting the seed was a common practice; as we now
6. C. SALMON, 0. R. MATHEWS, AND R. W. IIETSKEL
know, this must have resulted in many heart-breaking disappointments
because of poor stands and winterkilling. It was not until about 1869
or after some fifteen or twenty years of experience that farmers were
sufflciently convinced of the advantage of drilling to ship in a substantial number of drills. Considerable wheat seems to have been broadcast
even as late as 1880. Soft winter wheat and spring wheat, the latter a t
first predominating, were the only kinds grown. Both are much more
susceptible to injury by rust, and the soft winter wheat is more easily
winterkilled than are the hard red winter varieties generally grown now
or even than were those that were common a t the close of the century.
Losses from rust and winterkilling were severe (Malin, 1944). Experimental evidence secured in recent years indicates that the soft winter
varieties then grown yielded no more than two-thirds as much, and the
spring wheat no more than one-third or one-half as much, as the TURKEY
wheat grown somewhat later. But nearly twenty-five years’ experience
was required to induce farmers to abandon spring wheat and nearly
thirty years elapsed after the introduction of TURKEY wheat before
farmers generally were convinced that it wa8 better than the soft winter wheats. Contrast this with recent Kansas experience in which three
varieties, PAWNEE, COMANCHE, and WICHITA, almost unknown to Kansas
farmers in 1944, occupied 71 per cent of the acreage in wheat in 1952;
or with the spread of THATCHER i n the northern Great Plains from a
nominal 20,000 acres in 1935 to nearly 12,000,000 acres in the United
States and 5,500,000 acres in Canada by 1941.
Little was known about the best preparation of the land or time of
seeding in the Great Plains or about the control of rust, smut, grasshoppers, chinch bugs, and Hessian fly. Time of seeding was a debatable
question as late as 1920. It still is, though to a very minor extent as
compared with fifty to seventy-five years ago. Much wheat was sown
too early and winterkilled. This caused reaction to the other extreme,
which also resulted in winterkilling or a late harvest and in more than
average injury from rust, drought, or high temperature. Much of the
wheat dried out before winter because it had been seeded on land that
had not been prepared until near seeding time, and the remainder often
produced low yields because the growth of weeds between harvest and
seeding had used up the available moisture and nitrates in the soil. Substantial improvements in tillage were made from time to time, but there
was no concerted effort to prepare the ground immediately after harvest
until about the second decade of the present century. Experiments by
the Kansas, Nebraska, and Oklahoma Experiment Stations showed very
material advantages from early soil preparation, and the advent of the
tractor made this possible.
HALF CENTURY OF WHEAT IMPROVEMENT IN UNITED STATES
Losses from winterkilling were frequent and severe before the present
century. Malin states that abandonment of winter wheat acreage ranged
up “to 75 or 80 percent and probably oftener than not was 25 and 50
percent.” This, no doubt, was due to failure to get stands, to drought,
grasshoppers, and Hessian fly as well as to freezing during the winter
or early spring. He mentions winterkilling specifically as having been
recorded in nineteen of the forty-five years between 1855 and 1900. It
certainly has not occurred in anything like the same proportion of yeass
since 1900 to a sufficient degree to attract notice in the farm press or
official reports. McColloch (1923) has noted eight outbreaks of Hessian
fly in Kansas u p to 1916, six of which occurred previous to 1900 and
each of which was generally more widespread and severe than the one
that preceded it.
Malin mentions damage from rust in five of the forty-five years, stem
rust in two years, and kinds not specified in others. Remembering that
late-maturing, rust-susceptible varieties of spring wheat comprised a n
important part of the acreage u p to about 1880 and that rust-susceptible
varieties of soft winter wheat made u p most of the remainder until about
1900, it seems probable that rust caused more frequent damage than is
here suggested. Carleton (1896, p. 498) mentions the “blasting effects”
of rust, principally stem rust, in the “Southern latitudes,” and states
that wheat growing in vast areas of Texas had been abandoned on account of rust. Although Carleton mentions Texas specifically, he certainly had Kansas in mind, considering that one of the principal
objectives of a breeding program inaugurated there in 1899 was to produce rust-resistant varieties.
The shift to hard winter wheat also resulted in a n improvement in
quality, including plumper grain higher in test weight and yielding
more flour per bushel. Malin cites da.ta compiled by McFarland of the
United States Department of Agriculture for the years 1876-1883 in
which test weights ranged from 53.1 to 57.3 and averaged 54.6 pounds
per bushel. The areas represented by these data are not known, but it
appears that test weights of wheat have been much improved. The
average test weight of hard red winter wheat inspected a t Eansas City
for the years 1923-1932 was 59.1, and the least for any year (1923)
5. EaAy Wheat Growing ilz Nebraska. In Nebraska spring wheat
predominated until after 1900, and winterkilling of the soft winter wheat
was even more severe than in Kansas.
*“Grain Inspectors Letter” June 17, 1933.
U.S. Department of Agriculture.
Grain Division, Bur. Agr. Econ.,