American Breeders Association 7: 118-127 (1912)

COMPETITION IN CEREALS
E. G. MONTGOMERY
Lincoln, Nebraska

Competition as a factor in modifying the character of plant populations, by means of destroying or hindering the weak or least fit to survive under the particular environment, has been recognized as one of importance since Darwin pointed out its effective workings.

While thus recognized as a natural means of maintaining the vigor and strength of native vegetation, it has not been recognized as a means of maintaining the yield and vigor of our cultivated crops. Our common cereals have been cultivated for many thousand years with practically no attention to selection or grading until quite recent times. If any change has taken place, there has been a slow improvement, so far as we are able to judge by the limited information at hand. It is possible that the custom of placing in the soil seeds for two or three times as many plants as are really necessary to occupy the land has resulted in a continuous natural selection of the strongest and most productive.

In the fall of 1907 a series of small winter wheat plats were put out at the Nebraska Experiment Station for the purpose of securing some data on the amount of natural elimination of plants taking place under various rates of planting, and also to determine to what degree plants coming from undeveloped or small seeds might be eliminated when planted in competition with plants from large well developed seeds.

The plan adopted was to plant seeds of wheat at various distances, namely, 1/4 inch, 1/2 inch, 1 inch, and 1 1/2 inches apart. This was done with two varieties—Turkey Red and Big Frame. Then to test the effect of competition on plants growing from large, well developed and poorly-developed seeds, respectively, seeds of the two varieties were alternated. For example, large seed of Turkey Red would be alternated with undeveloped seed of Big Frame at the various rates from 1/4 inch to 1 1/2 inches. A reciprocal series was also planted, using poorly developed Turkey Red, in competition with well developed Big Frame. The series were duplicated and alternated with check plats, there being 96 plats all told. The experiment was repeated with winter wheat in 1909 and 1910, and with two varieties of oats in 1910, using large and small seeds, but all well-developed. The wheat plats were completely winterkilled in 1910.

Decrease in Plants from Planting to Harvest.—The number of seeds planted per plat at each rate were 672 when 1/4 inch apart, 336, 168 and 112, respectively, at the other rates. The number of plats in a series, including checks, was 48, and repeated, making 96 in all. The plats each consisted of 4 rows 36 inches long and 8 inches apart, or the area of the plat was approximately 32 by 36 inches.

On October 22 the living plants in each case were counted, and again at harvest time all were counted. For ease of comparison all data were reduced to the basis of 100 grains planted in the following table:

TABLE 1.—Combined data for Turkey Red and Big Frame wheat, showing survival of plants at various rates of planting, 1908.

Distance of planting in inches.	Seeds planted.	Number of plants October 22, 1907.	Number of plants harvested 1908.	Decrease from October to harvest.
1 1/2	100	89	77	12
1	100	87	70	17
1/2	100	83	66	17
1/4	100	83	60	23
Average	100	85.5	68.2	17.2

For every 100 seeds planted there was an average of 85.5 live plants October 22, and 68.2 plants came to maturity. Between October 22 and harvest time there was an average decrease of 17.2 plats from various causes. That some of this decrease was due to competition is shown by the fact that about twice as many were lost under thick planting as thin.

In 1909 the experiment was continued but under unfavorable conditions. The fall and winter proved to be very dry, so the fall growth was rather small, and winterkilling was heavy. The data show that for every 100 seeds planted there were 74 live plants December 2, 41 at harvest time, and an average decrease of 33 plants from fall to harvest. Again the greatest relative decrease is in the thicker plantings.

TABLE 2.—Combined data for Turkey Red and Big Frame wheat, showing survival of plants at various rates of planting, 1909.

Distance of planting in inches.	Number of seeds planted.	Number of plants December 2, 1908.	Number of plants harvested.	Decrease from December 1908, to harvest.
1 1/2	100	76	51	25
1	100	78	44	34
1/2	100	74	37	37
1/4	100	69	31	38
Average	100	74	41	33

In 1910 the winter wheat plats were winterkilled and no data were secured. A series of oat plats, planted in the same manner, gave data similar to that secured with winter wheat, as shown in Table 3.

TABLE 3.—Combined data for two varieties of oats showing survival of plants at various rates of planting, 1910.

Distance of planting in inches.	Number of seeds planted.	Germination.	Number of plants harvested.	Decrease from planting to harvest.
2	100	—	91	9
1	100	—	75	25
1/2	100	—	64	36
1/4	100	—	64	36
Average	100		73.5	26.5

The per cent of germination was not recorded. There was an average of 73.5 per cent as many plants harvested as seeds planted, and only 64 per cent with the thickest plantings.

It appears from data cited that there is a decrease of about 15 to 30 per cent in the number of plants from spring to harvest. There are many causes of this decrease, such as insects, diseases and accidents of various kinds, but after these are accounted for it is evident that we still have a loss of plants through competition in the thicker plantings. It seems natural that the plants eliminated should be those that from any cause are weak or slow in development. Plants from undeveloped seeds are usually smaller in early growth, although later the plants may become vigorous enough. As mentioned early in the paper, in certain plats well-developed seeds were alternated with undeveloped seeds at the various rates, for the purpose of determining

whether the more vigorous plants from the large seeds would tend to eliminate the less vigorous plants from the poor seed. Table 4 is a summary of data of 48 plats of winter wheat, all rates of planting.

TABLE 4.—Combined data for Turkey Red and Big Frame winter wheat, to show effect of competition of two grades of seed, 1908.

Description of seed.	Manner of planting.	Number of seeds planted.	Number of plants October 22.	Number of plants harvested.	Decrease from October to harvest.
Well-developed, plump	Alone	100	87	72	15
Undeveloped	Alone	100	80	65	15
Well-developed, plump	In competition	100	89	79	10
Undeveloped	In competition	100	85	60	25

a The advantage was not as pronounced in other tests. See table 8.

Both grades of seed showed the same decrease from October to harvest, when planted alone, i.e., each grade in a plat to itself. However, in competition, when the two grades were alternated, plants from the well-developed seed showed a marked advantage.^a Evidently the initial advantage gained by the more vigorous plants from large plump seed enabled them to crowd out the less vigorous plants when planted in competition. Observations made in the field show that early in the season when rapid growth first begins, whether with fall-sown or spring-sown crops, if for any cause a plant is slow in starting, and it is located between quick-starting plants, the latter will soon shade it and have advantage in various ways, that the slower plant is permanently retarded, whereas a similar slow starting plant, among others of its kind, may fully recover in time and at harvest be as vigorous as any.

TABLE 5.—Combined data for Turkey Red and Big Frame winter wheat, to show effect of competition of two grades of seed, 1909.

Description of seed.	Manner of planting.	Number of seeds planted.	Number  plants December 2, 1908.	Number of plants harvested, 1909.	Decrease from December to harvest.
Well-developed, plump	Alone	100	74	43	31
Undeveloped	Alone	100	67	34	33
Well-developed, plump	In competition	100	81	50	31
Undeveloped	In competition	100	75	36	39
Average			74.2	41	33.2

The data for 1909 are irregular, owing to the unfavorable conditions under which grown and severe winterkilling. All grades were so severely winterkilled that little competition could take place in the spring growth, but again the highest per cent surviving plants was with the best seed in competition and the greatest elimination among plants from poorly developed seeds in competition.

In 1910, Swedish Select (white) and Garton No. 70 (black) were sown in competition, alternating large and small seeds at various rates from to 2 inches, and sowing large seed and small seed alone at the various rates—48 plats in all.

TABLE 6.—Large and small seeds of two varieties of oats, sown al various rates and in competition, 1910.

Description of seed.	Manner of planting.	Number of seeds sown.	Number of plants harvested.	Decrease.
Small seed	Alone	100	60	40
Large seed	Alone	100	65	35
Small seed	In competition	100	68	32
Large seed	In competition	100	74	26

In this case the large seed in competition did not show a greater advantage than when the large and small were sown alone. For some reason, in every case with both wheat and oats, two varieties in competition have given a greater number of plants at harvest or greater yield than when either variety was sown alone.

Table 7 is a summary of data of the number of plants surviving at the various rates of seeding for three years.

TABLE 7.—Number of plants harvested for 3 years, on basis of 100 grains planted, at various rates.

Distance of planting in inches.	Number of seeds planted.	Winter wheat 1908	Winter wheat 1909	Oats 1910.	Average.
1 1/2	100	77	51	91	73
1	100	70	44	75	63
1/2	100	66	37	64	56
1/4	100	60	31	64	52
Average	100	68.2	41	73.5	61

The results indicate that about two-thirds as many plants are harvested as seeds planted, as an average of all rates of planting. The number surviving, where the planting is thin, is about 40 per cent greater than where the planting is thick. This greater decrease in the thick planting is apparently due to the effect of competition among the plants.

FIG. 1. THREE SETS OF KHERSON OAT PLANTS.
Grown at three rates of seeding, namely. 4 peaks per acre, 8 pecks. and 16 pecks. At this time (plants 8 inches in height) the plants from the thin seeding were producing tillers freely, while very little tendency to tiller was shown by the plants from thick planting.
FIG. 2. SIXTEEN TYPICAL KHERSON OAT PLANTS.
From plant sown at rate of 16 pecks per acre. Plants about 8 Inches high. At the right are about 4 plants that probably will not survive until harvest, or, if they do. will not produce more than a few seeds. In Fig. 3 are shown the relative size of the smaller plants at harvest time.

Table 8 is a summary of data where large, well-developed seed was compared with small or undeveloped seed.

As an average for 3 years when the small or undeveloped seed are planted alone, 53 plants were harvested for each 100 seeds and 60 plants from the large well-developed seed. This shows that there was considerable difference in the original quality of the seed. When planted in competition there is a higher percentage of survivors in both cases—a fact noted every year. The survivors from large seed in competition, however, were 68, or 13 more than from the small seed, while planted alone the advantage was only 60 to 53—an advantage of 7 plants. Since the total reduction in number of plants was 41 per cent on an average, the advantage of large seed over small is not large. It seems that there are almost as many weaklings susceptible to the effects of competition among the plants from large seed as small.

Effect of Rate of Planting on Size of Plant.—Table 9 affords an excellent example of the relation of rate of planting to yield and size of plant.

	FIG. 3. RELATIVE SIZE OF LARGE AND SMALL PLANTS AT HARVEST. [KHERSON OATS.]
	From plat sown at rate of 16 pecks per acre. When the plants are about 8 inches high the stronger begin to grow rapidly, while the weaker plants either die or make slender growth producing only a few seeds.

TABLE 8.—Number of plants harvested for 3 years on basis of 100 grains planted at various rates.

Description of seed.	Manner of planting.	Number of seeds planted.	Number of plants harvested.
			Winter wheat, 1908.	Winter wheat, 1909.	Oats 1910.	Average.
Small or undeveloped	Alone	100	66	34	60	53
Large, plump	Alone	100	73	43	65	60
Small or undeveloped	In competition	100	00	34	68	66
Large, plump	In competition	100	79	50	74	68
Average			69	41	67	59

TABLE 9.—Oats 1910. Effect of rate of planting on size of plat.

	Spacing of plants.
	2 inches.	1 inch.	1/2 inch.	1/4 inch.
Number of seeds planted	100	100	100	100
Number of plants harvested	91	76	64	64
Average yield per plat (72-144-288-576 seeds per plat)	94	96	87	98
Yield per plant	1.72	1.09	0.62	0.35

The yield per plat was not materially influenced by the rates of seeding, as the yield per, plant seemed to make full adjustment to conditions, thus producing maximum yield under the thin planting.

Competition of Varieties.—As a result of the experiment, some data were secured showing the effect of competition one variety may have on another. In the case of the winter wheat, the following table shows that in 1908, when Big Frame was sown alone under the conditions of this experiment, it yielded most, but when sown in competition with Turkey Red the latter was most productive.

In the case of oats in 1910, whenever the black oats (Garton No. 70) was sown alone, it outyielded the white (Swedish Select), but when sown in competition the white returned the greatest yield.

TABLE 10.—Competition of varieties.

Variety.	Manner of planting.	Number of seeds planted.	Number of plants harvested.	Total yield.
Turkey Red	Alone	2676	1429	Grams. 885
Big Frame	Alone	2576	1784	985
Turkey Red	In competition	2676	1836	1190
Big Frame	In competition	2576	1712	826

TABLE 11.—Comparative yields of two varieties of oats when sown alone and in competition. Yield per plat in grams.

	Spacing of plants.
	2 inches.	1 inch.	1/2 inch.	1/4 inch.	Average.
Garton No. 70 (black)	136	143	121	146	137
Swedish Select (white)	116	116	96	119	112
Garton No. 70 (black)	102	111	110	92	104
Swedish Select (white)	142	142	155	164	151

There is a possible explanation in the early habits of growth of the two oat varieties. The black oats tend to spread out close to the ground, and make little growth in height for several weeks, while the white oats grows rapidly in height at first. When mature, both varieties are about the same height. It appears that the tall growing habit of the white oats in the early stages gives it an initial advantage not fully lost later.

Corn has always been grown rather thin, so the plants are free from competition to a large extent. Natural selection through competition has not been effective. In order to secure data on the effects of competition in corn a series of plats were started in 1905, using Hogue's Yellow Dent corn and planting at 1, 3, and 5 grains per hill. Each lot has been planted continuously at the same rate since, the results for three years being given in Nebraska Bulletin 112. In 1910, after five years continuous selection, the following results were secured:

Rate at which continuously planted in past.	Rate planted in 1910.	Yield per acre.
Grains per hill.	Grains per hill.	Bushels.
1	3	59.7
3	3	61.3
5	3	65.7

This shows a gain of 6 bushels per acre, or 10 per cent, as a result of growing the seed under competitive conditions.

SUMMARY

While the experiments are not extensive enough to fully justify conclusions, yet the data are so suggestive that it seems worth publishing at this time. These data verify the following conclusions:

1. That as the rate of planting is increased, the percentage of plants surviving until harvest is gradually decreased. The total decrease in cases sighted, amounting to 39 per cent, and the relative survival of the thinnest stand (2 inches) and thickest stand (1/4 inch) averaged 73 and 52 per cent, respectively, or a decrease of 29 per cent.

2. When large plump seeds are planted in competition with small or poorly developed seeds, the decrease in the large plump seed is 32, the small or shrunken 45 per cent, while the same grades planted alone decreased 40 per cent and 47 per cent, respectively. All grades show a marked decrease due to various causes in addition to competition, but when the two grades are planted in competition the stronger plants have an advantage in number of survivors.

3. When two varieties are planted in competition, one variety is very apt to have an advantage, which, if continued, would in time practically replace the other. It appears also that the one yielding best alone will not always be the one surviving under competition.

4. Nature seems to have a way of eliminating the weaklings (under our present system of sowing cereals) whether they come from large or small seeds. Since this elimination has been going on for ages, it does not seem that an artificial method of seed separation (as a fanning mill or screens) could increase the efficiency of seed, especially since our method of thick seeding allows nature to eliminate one-half the plants each year without affecting the yield.

5. The desirability of separating pure strains is emphasized by the suggestion that when left in competition the best yielder (when placed alone) may not dominate, but, on the other hand, a poor yielder [may] be best able to survive competition.