Tải bản đầy đủ - 0 (trang)
IV. The Role of Improved Varieties

IV. The Role of Improved Varieties

Tải bản đầy đủ - 0trang

53



HALF CENTURY OF WHEAT IMPROVEMENT I N UNITED STATES



regions may be recognized: (1) the hard red spring wheat region, (2)

the durum wheat region, which geographically is in the hard red spring

wheat region, ( 3 ) hard red winter wheat region, (4) eastern United

States, and (5) western United States. Figure 15 shows the distribution

of wheat in the United States, with each of these districts indicated.

The changes in varieties will be discussed separately for each region,

and following this an attempt will be made to evaluate these changes

in terms of increased yields, better quality, and reduction of wheatgrowing hazards. An accurate evaluation is, of course, a difficult or even

an impossible task. Occasionally, however, someone longs for the “good



~~



-



~



~



~~



FIG.15. Distribution of wheat seeded in the United States in 1949 aiid principal wheat-growing areas. Each dot represents 5,000 acres.



old varieties” and discounts the claims made for the new ones, just as

there may be others who tend to overemphasize them, For this reason

even rough estimates of improvements may be worth-while. Since lack

of space precludes a full discussion of each region, the discussion for the

first region to be considered will be somewhat more complete than that

for the others.

1. Improved Varieties of Hard Red Spring Wheat



The wheat industry in the hard red spring wheat region developed

with the varieties RED FIFE and BLUESTEM. RED FIFE found its wa.y into



64



S. C. SALMON, 0. R. MATHEWS, AND R. W. LEU-



the North Central States soon after the middle of the nineteenth century

and BLUESTEM appeared somewhat later. PRESTON under various names

appeared near the close of the century and was generally grown in some

localities. According to Stoa (1945), RED FIFE and related strains of FIFE

dominated the picture before 1900. By 1914 (Yearbook, U.S. Department of Agriculture, 1920, p. 559) slightly more than half the crop of

hard red spring consisted of BLUESTEM. On the basis of fairly extensive

experimental trials in Minnesota, Montana, and the Dakotas, previous

to 1900, those considered here, HAYNES BLUESTEM, RED FIFE, POWER, and

PRESTON, are believed to be representative of, or at least as good as, any

that were grown by farmers at that time.

The first important change in varieties after 1900 resulted from the

introduction of the famous MARQUIS wheat from Canada in 1912 and

1913. MARQUIS was so immediately and generally popula,r that it was

grown on more than 70 per cent of the hard red spring wheat acreage by

1919 and on nearly 90 per cent by 1929. It headed and ripened a t least

two to three days earlier than HAYNES BLUESTEM, and one or two days

earlier than POWER and RED FIFE. By virtue of this fact it escaped some

of the damage from stem and leaf rust in Minnesota and the eastern

Dakotas, and from drought in the western part of the area. Having

slightly shorter straw it lodged less. It was widely acclaimed by millers

and processors and is still regarded by many as the ideal variety from

the point of view of quality.

CERES, produced by the North Dakota Station, was distributed to

fa.rmers in 1926. Its particular virtue was slightly earlier maturity

(about one day earlier than MARQUIS) and resistance to the then prevailing races of stem rust. It has also been widely regarded as being more

drought-resistant than other varieties. To what extent this belief is due

to its earlier maturity and hence to its tendency to esca.pe the effect of

drought is not precisely known. It is perhaps significant, however, that

CERES survived definitely better than the others in laboratory tests a t the

North Dakota Station (Helgeson and Blanchard, 1940) in which young

plants of CERES and other varieties were subjected to high temperature

for a few hours. Also Bayles et al. (1937), Hubbard (1938), and Bartel

(1947) found differences between CERES and other varieties with respect

to certain physiological or morphological characteristics consistent with

relative yields under drought conditions. It would appear, therefore,

that CERES owes its high yield under drought conditions to something

more than early maturity.

CERES was grown on nearly 35 per cent of the hard red spring wheat

acreage by 1934. The acreage of MARQUIS had been declining in the

meantime, and HAYNES BLUESTEM, POWER, RED FIFE, and PRESTON had



HALF CENTURY OF WHEAT IMPROVEMENT I N UNITED STATES



55



practically disappeaxed. In about 1935 stem rust race 56, which attacks

CERES as well as all the other varieties grown a t that time, appeared in

epidemic proportions. Both MARQUIS and CERES were badly damaged in

the severe epidemic of that year, although yields of CERES were about

double those of MARQUIS.

The famous THATCHER variety, produced in cooperative experiments

with the Minnesota Station, was grown on 20,000 acres in 1935, having

been released to farmers the previous year. THATCHER produced 20

bushels per acre or more in excess of CERES and MARQUIS on farms that

year; and thereafter increased in acreage as rapidly as seed supplies

would permit. It occupied more than 40 per cent of the spring wheat

acreage by 1939, being a t that time the most widely grown variety of

hard red spring wheat. THATCHER was not only far superior to all others

with respect to resistance to the prevailing races of stem rust but was

also early (about two days earlier than CERES and three to five days

earlier than MARQUIS), lodged but little, and proved to be of excellent

quality. Because of its earliness it found a place not only in the eastern

half of the area. where stem rust is important but also in the western

Dakotas and Montana, where it is still extensively grown.

But THATCHER is unusually susceptible to leaf rust and in 1938 and

again in 1941 yielded substantially less in experimental trials than other

new varieties, principally RIVAL and PILOT, which had been included in

comparative trials a t some of the North Dakota Stations as early as 1933.

Both of these varieties were distributed to farmers in 1939 and soon

replaced a considerable part of the acreage of THATCHER in Minnesota

and the eastern half of the Dakotas.

MIDA was distributed by the North Dakota Station in 1944. Like

RIVAL and PILOT, it was resistant to the prevailing races of leaf and stem

rust, headed and ripened early in comparison with other varieties, did

not easily lodge, and had excellent quality. It also increased rapidly and

by 1949 was the leading variety of hard red spring wheat.

MIDA, RIVAL, and PILOT are bearded wheats. CADET, a beardless variety

having stiff straw and excellent quality, was distributed to farmers in

1946. It heads and ripens three to four days later than THATCHER or

about the same time as MARQUIS. It is grown only sparingly in western

North Dakota and in some counties in the northern part of North Dakota.

Other varieties which should be mentioned include REGENT, RESCUE,

and REDMAN, introduced from Canada ; NEWTHATCH, produced in cooperative experiments a t the Minnesota Station ; and HENRY, a t the Wisconsin Station. REGENT was grown on less than one-half million acres in

1949, mostly in North Dakota, REDMAN on about 135,000 acres, and NEWTHATCH on about 250,000 acres scattered throughout North Dakota and



56



S. C. SALMON, 0. R. MATHEWS, AND R. W. LEUEEL



western Minnesota. HENRY is grown mostly in Wisconsin. RESCUE is

the only commercial variety grown in the United States that is resistant

to wheat-stem sawfly. About one million acres are grown in the sawflyinfested area of northwest North Dakota and in northeast and north

central Montana. Two new varieties, RUSHMORE and LEE, were released

to farmers in 1949 and 1951, respectively.

Resistance to leaf rust has been an important objective i n the breeding of all varieties since the distribution of THATCHER. As a consequence,

all of the newer varieties are resistant to the races of leaf rust that were

generally prevalent when they were released. Since then new races or



100

90



10



0

191s



1814



I929



1934

VEPlR



1939



1944



1949



FIG 16. Changes in acreages of important varieties of hard red spring wheat

in the northern Great Plains (Minnesota, North Dakota, South Dakota, and Montana) from 1919 to 1949.



previously unimportant ra.ces have multiplied, with the result that none

of them, with the possible exception of LEE, can be said to be generally

resistant in the field.

The more important of the above changes are shown graphically in

Fig. 16. The geographical distribution of each of the widely grown

varieties in 1949 is shown graphically by Clark and Ba.yles (1951). It

will be noted that MARQUIS and CERES now occupy only about 5 per cent

each of the hasd red spring wheat acreage. This is mostly in sections

of Montana where rust is not a factor. THATCHER declined rapidly fol-



HALF CENTURY OF WEIEAT IMPROVEMENT IN UNITED STATES



57



lowing severe leaf rust damage in 1938 and 1941, being largely replaced

by RIVAL and PILOT and these in tu rn by MIDA soon after its release to

farmers in 1944. A final evaluation of these varieties, as indeed of all

others, in view of the prospective damage from race 15B of stem rust

must await additional information.

a. Are the Improved Varieties Superior to the Old? What do these

changes mean to growers and consumers of hard red spring wheat? To

what extent have they resulted in larger yields and better quality, if

any? Specifically, what would be the production of hard red spring

wheat today if farmers had to depend on the varieties known to them

before 1900 1 Any one familiar with the agricultural history of the area

can scarcely doubt that they have been tremendously important. The

evidence for this statement consists mostly of relative yields, quality

characteristics, and resistance to disease, insects, and weather hazards of

varieties grown in comparable experimental trials. These trials have

been conducted for many years in each of the states that produce most

of the spring wheat.

When MARQUIS was first included in these trials in about 1913, HAYNES

BLUESTEM, POWER (POWERS FIFE), RED FIFE, and PRESTON were the principal varieties.

The results of experiments in which these varieties and others were

compared have been reported by Atkinson and Donaldson (1916) and

by Morgan and Bell (1926) for trials a.t Moccasin and Havre, Montana,

respectively ; by Towle (1930) for Sheridan, Wyoming ; and by Hardies

and Hume (1927) for Brookings and Highmore, South Dakota. Reports

of early varietal trials in North Dakota have been published by Lanxon

(1919), Walster (1920b), Thompson (1921), Ruzicka, (1922), Kuenning

(1925), and Moomaw (1925). Most of the North Dakota experiments

have been summarized by Stoa (1921) and by Stoa et al. (1927). Results at Minnesota Stations have been reported by Wilson and Arny

(1930). Most of the early experiments up to about 1919 have been summarized by Ball and Clark (1916) and by Clark et al. (1922).

CERES was recognized by research workers as a promising variety before it was distributed to farmers in 1926 and was generally included in

experimental trials after 1925. As a result of the larger co-ordinated

program initiated in 1928, THATCHER, RIVAL, PILOT, MIDA, and others were

included in comparative trials throughout the region. Reports of many

of these trials have been published by Klages (1931), Hayes et al. (1936),

Stoa (1951), Hume (1940), Swenson (1942), Waldron et al. (1942),

Harris et al. (1947), and Walster and Nystuen (1948). Waldron (1945)

has presented a n interesting tabulation of relative yields by five-year

periods. Clark (1931-1950) has tabulated and presented in mimeo-



58



S. C. SALMON, 0. R. MATHEWS, AND R. W. LEUKEL



graphed form the principal field plot and nursery data for ea.ch station

each year.

CERES was first included in comparative yield trials in 1923, THATCHER

in 1929, RIVAL and PILOT in the mid-thirties, and MIDA in 1940 or 1941.

Many varieties were discontinued as soon as their inferiority had been

demonstrated. HAYNES BLUESTEM and RED FIFE have been included in

recent yield trials only a t Dickinson, North Dakota,, POWER only a t Fargo,

and PRESTON not at all. MARQUIS was discontinued in many of the experi~ ~ CERES has been discontinued from some of

ments in the early 1 9 4 0 ’ and

them. Although this procedure usually is justified under the circumstances, i t is sometimes unfortunate when, as in the present case, there

is need for long-time comparable records, either to evaluate varieties or

to elucidate the principles on which varietal adaptation depends. Thus

direct comparisons between MIDA and the old varieties HAYNES BLUESTEM,

POWER, RED FIFE, and PRESTON can be made at very few places and for a

few yeaxs only.

It appears that the best that can be done in the present case is first

to compare MARQUIS with the old varieties HAYNES BLUESTEM, POWER, RED

FIFE, and PRESTON; then CERES with MARQUIS; THATCHER with CERES;

RIVAL, PILOT, and MIDA with THATCHER; and finally to. make such direct

comparisons as are possible between the newer varieties and the old.

Both methods suffer from certain defects, but the former has the advantage of including most of the yield data.

(1)Relative yields, Table VII shows the comparative yields obtained

by these various comparisons from nearly all unirrigated varietal field

plot experiments conducted for five years or more in central and western

Minnesota, North and South Dakota, northern Wyoming, and central

and eastern Montana. They do not include nursery plot trials except

for a few stations where in recent years replicated multiple-row nursery

plots have been substituted for field plots. The newer varieties include

only those which were grown on 250,000 acres or more in 1949. One

variety, RESCUE, which was grown on more than 250,000 acres is not

included for reasons stated later. The varieties are arranged from left

to right roughly in the order in which they were released to farmers.

MIDA is compared with THATCHER rather than with RIVAL and PILOT, since

the number of stations and years in the two cases are almost identical.

RIVAL and PILOT are considered together, since they are simi1a.r in plant

type, dates of heading and ripening, resistance to rust, and yield.

The data are presented separately for the eastern half of the area,

where leaf and stem rusts axe dominant yield factors, and for the western half of the area, where drought and high temperatures are more

important. The 100th meridian is the approximate dividing line, al-



HALF CENTURY O F WHEAT IMPROVEMENT I N UNITED STATES



59



though data from three stations, Highmore, Eureka, and Mandan, which

lie slightly west of this meridian are included in the eastern section because rust-resistant varieties at these stations have generally outyielded

susceptible varieties. It should not be assumed, however, that this is a n

accurate classification; rust has a t times caused serious loss in the western Dakotas and even in Montana, and drought is not unknown even as

far east as St. Paul, Minnesota.

It will be seen that in the eastern section MARQUIS produced higher

average yields than any of the four varieties that preceded it and that in

turn CERES has yielded more than MARQUIS; THATCHER more than CERES;

and RIVAL, PILOT, and MIDA more than THATCHER. The same is true in

the western section except that RIVAL, PILOT, and MIDA are not higheryielding than THATCHER. Of these latter varieties only PILOT and MIDA

are grown extensively west of the 100th meridian, and they, only in western North Dakota. The differences in yield in the original tables, which

are summarized in Table VII, were remarkably consistent in most cases.

In the eastern section MARQUIS failed to a.verage more than HAYNES BLUESTEM, POWER, RED FIFE, and PRESTON in only one trial, viz, St. Paul, Minnesota, for 1913-1915 and for 1918-1927. I n this case PRESTON outyielded MARQUIS by 0.4 bushel. On the average for all years a t each

station CERES outyielded MARQUIS, THATCHER ou tyielded CERES, MIDA

outyielded THATCHER,

and RIVAL or PILOT failed to outyield THATCHER

a t only two stations and then by less than a bushel in each case. Likewise a t all stations in the western area MARQUIS on the average outyielded

HAYNES BLUESTEM, POWER,RED FIFE, and PRESTON;CERES outyielded

MARQUIS;and THATCHER outyielded CERES. I n the remainder of the

trials in the western region, i.e., those in which RIVAL, PILOT, and MIDA

are compared with THATCHER, the average yields were often but not

always in favor of THATCHER. This is i n line with the comparative

acreages of these varieties and THATCHER west of the 100th meridian.

The question naturally arises as to whether a useful estimate of the

over-all gains in yield may be had by adding the successive gains.

Clearly such a procedure is not justified if there is a change in yield

levels. Leaf and stem rust have occurred much more frequently since

1913 than before; hence, part of the differences in yield merely offsets

the tendency for yield levels to decline. There is also evidence to indicate, as pointed out by Stoa (1951), that differences in yield between

rust-resistant and rust-susceptible varieties tend to become less as new

races of the rust organism appear. On the other hand, the earlier heading and ripening of the newer vasieties has enabled them to escape some

of the damage that would otherwise have occurred because of rust,

drought, and high temperature. To the extent that increase in rust



TABLE VII

Comparative Yields of Important Varieties of Hard Red &ring Wheat in the Northern Great Plains

Item



Number

of



and

section



station

years



Eastern section



29

59

69

199

157

135

147

99



Average yields, bn./acre

HllPNEs



REI)



FIFE



POWEB



BLUESTEX



PBJCSTON



15.0

17.8

19.1



MAEQUTS



19.7

20.3

20.0

16.6



CEBEIS



19.9

18.8



THATCFKEE



21.2

23.7

23.5

24.9



RIVAL



MIDA



PILOT



26.5

25.3

28.3

~



Difference compared with



-2.5



MARQUIS



-4.7



-0.9



$3.3

+2.4



CFaES

THATCHER



Western section



+2.8



67

60



45



13.5

19.1

15.1



22

119

86

35

68

36



14.3



16.3

20.6

16.5

16.0

18.7



20.7

20.5



21.2

21.9

22.6

27.2



+1.8



+3.4



20.7

22.2

26.8

-



Diff ercnces compared with

MARQUIS

CFaES



THAWHE%



-1.4



-1.5



-2.8



-1.7



+2.0

+0.7



-1.2



-0.4



-0.4



* The original data from which this table wau prepared have been publisbed, as indicated in the text, or they were obtained in cooperation witb

the Agricultural Experiment Stations of Minnesota, Sooth Dakota. North Dakota. Montana. snd Wvominlr.



H A L F CENTURY OF W m A T IMPROVEMENT IN UNITED STATES



61



damage and new races of rusts have been important, adding successive

gains in yield would not be a valid procedure. Fortunately, this can be

determined to some degree by comparing the newer varieties THATCHER,

RIVAL, PILOT, and MIDA with certain of the old varieties a t two stations,

Fargo and Diclrinson, North Dakota.

It is we11 known that yields in experimental trials are often higher

than those in farmers’ fields and that differences between varieties often

increase as yields increase. To correct for this tendency, yield differences between varieties in the original tables for the varietal pairs shown



FIG.17. Relation between differences in yield in varietal pairs and the yield

level in experimental field trials.

( r = 0.9818.

N = 6675)

y = 0.97 .104X.



+



in Table V I I were tabulated and averaged for different yield levels. The

same was done for each of the other three wheat regions, and since the

results were substantially alike, they were combined and are shown here

as a graph in Fig. 17. Observed differences for hard red spring wheat

were then adjusted on the basis of this graph to a yield level of 14.5

bushels per seeded acre, which is approximately the axerage yield of all

hard red spring wheat in Minnesota, North Dakota, South Dakota, and

Montana for the period 1941-1950. It may be noted from Fig. 17 that

the relation between yield levels and yield differences is linear and that

yield differences decline almost precisely 1 bushel for each 10 bushels

difference in yield level.



62



S. C. SALMON, 0. R. MATHEWS, AND R. W. LEUKEL



Table VIII shows the average gains for each successive new variety

as compared with the one or more that preceded it before and after adjustment to yield levels and the estimated accumulated gains. The latter

are derived by adding the successive gains. The estimated gain shown

for MARQUIS in this table is the weighted average of those shown in

Table VII.

TABLE VIII

Estimated Gain in Yield for Improved Varieties East and West of the 100th

Meridian (Approximately) before and after Adjustment for Yield Levels

East of 100th meridian West of 100th meridian



Item

Average gain f o r :

MARQUIS over HAYNES BLUZS.l'EM,

POWER, R.ED FIFE, and PRESTON

c m s over MARQUIS

THATDEER over cmms

RIVAL over THATCH=

PILOT Over THATOHER

MIDA over THATCHER



Total accumulated gain f o r :

CEREB over H A ~ BLUESTEM,

S

POWER, FZD FIFE, and PRESTON

m v a ~over RAYNES BLUESTEM,

POWER, RED FIFE, and PRESTON

PILOT over HAYNES BLUESTEM,

POWER, RED FIFE, and PRESTON

MIDA over HAYNES BLUESTEM,

POWER, RED FIFE, and PRESTON

THATCHER over HAYNEB BLUESTEM,

POWIGR, RED FIFE, and PRESTON



No. of

station

years



Before After

Adjustment,

bu./acre



128*



2.2



199

153

135



3.3

2.4

2.8



147

99



1.8

3.4



1.5

2.8

1.8

1.7

0.8

2.1



4.3



No. of Before After

station Adjustment,

years

bu./acre



98*

119

86

35

68

36



2.0

2.0

0.7

1.2

-0.4

-0.4



1.9

1.5

0.1

0.0

0.0



0.0



3.4



7.8

6.9



3.5



8.2



3.5



6.1



3.5



* The number of station years for the comparison of M A R Q U Iwith

~

the older varieties ia not

the sum of those given i n table VII since in many cases more than one was compared with

M A E Q U I ~ in the same experiments.



The gains estimated by direct comparison and by adding successive

gains a t Fargo and Dickinson are compared in Table IX. In some cases

the direct comparison indicates a lesser gain and in others a greater gain

than is obtained by adding successive gains. On the whole, the agreement seems to be as good as could be expected considering the relatively

short testing periods in some cases and the sharp fluctuations in yield

differenws from year to year due in the main to whether rust was or

was not a factor.



HALF CENTURY OF WHEAT IMPROVEMENT IN UNITED STATES



63



In this connection it should be noted that differences between rustresistant and rust-susceptible varieties have been in general substantially

less during the last ten years than in previous years, owing presumably

to less damage by rust. Since the estimates of increases secured by accumulating gains as in Table VII are based on a much larger number of

trials representing more years and more stations, it would appear reasonable to consider them more reliable for estimat.ing the increases for the

area as a whole than are those for Fargo and Dickinson alone (Table

I X ) . They accordingly are used to estimate the increases in production

due to the improved varieties, as shown in Table X.

TABLE I X

Estimates of Increased Yields a t Fargo and Dickinson, North Dakota, by Accumulating Gains and by Direct Comparison before and after Adjustment to Average

Yield Levels



Item and location



Number

of

station

years



Fargo, N . Dak.

Average gains

MARQUIS over POW=, 1913-1914, 1916-1950

C E ~ E Sover MARQUIS, 1923-1950

THATCHER Over CERES, 1930-1950

MIDA over THATCHER, 1940-1950

MIDA over POWER

By accumulating gains

By direct comparison, 1940-1950

THATCHER over POWER

By accumulating gains

By direct comparison, 1930-1950

Dickinson, N . Dak.

Average gains

MARQUIS over HAPNES BMESTEM and RED FIFE, 1913-1950

CERES over MARQUIS, 1923-1950

THATCHEIL over OERXS, 1930-1950

MIDA Over THATCIIER, 1940-1950

MIDA over HAYNES BLUESTEM and RED FIFE

By accumulating gains

By direct comparison, 1940-1950

THATCHER over HAYNE#SBLUESTEM and RED FIFE

By accumdating gains

By direct comparison, 1930-1950



37

28

21

11



Estimated

gaius

Before After

Adjustment

bu./acre



+2.6

+3.5

+1.4

+1.8



1.4

2.6

0.3

0.5



11



9.3

6.4



4.8

5.1



20



7.5

6.2



4.3

5.7



$1.9

+2.2



1.9

2.1

1.3

0.9



38



28

21

11



+1.3



+1.5



11



6.9

6.2



6.2

5.6



21



5.4

4.2



5.7

4.2



Tài liệu bạn tìm kiếm đã sẵn sàng tải về

IV. The Role of Improved Varieties

Tải bản đầy đủ ngay(0 tr)

×