or other local conditions would tend to be equalized by the alternate-row method of comparing the two systems of culture. But more numerous and wider skips often appear in wide-spaced rows on account of injury to seedlings, due to greater exposure in the early stages of their development. Fewer plants are injured in single-stalk rows because of the mutual protection afforded by the thick-standing seedlings, and such injuries as occur are rendered unimportant because enough plants remain to thin as desired. This feature has been previously pointed out as one of the many advantages of later thinning.1

There were 16 rows in the experiment, the two systems being compared in alternate rows. Two pickings were made, but the yields from only the first picking were reported. These are shown in Table IV.

TABLE IV.-Row yields of the first picking obtained in a single-stalk culture experiment with cotton conducted in 1915 on the Killarney Plantation at Mounds, La.

Table IV shows that in this experiment single-stalk rows yielded more cotton in every instance than adjoining old-system rows, the total increase for the first picking being 39 per cent. Judging by the appearance of the rows in August, when counts of immature bolls were made by Mr. Watson and the writer, showing 75 to 100 per cent more bolls on the single-stalk rows than on the others, it is believed that the difference in yield from the second picking, had it been recorded, would have been even greater than 39 per cent. But even if it were no greater, or considerably less, the difference in the first picking is of marked significance, since the experiment was located in a section of maximum boll-weevil infestation, where earliness in the cotton crop is of the greatest importance.

The Boney experiment.-This experiment was on sandy soil located on the farm of Mr. R. K. Boney, at Duckport. Half-and-Half cotton was planted under fairly desirable conditions (date and rate not reported) and a good stand was secured. Definite details concerning the time of thinning the old-method rows and their condition when thinned were not reported. The plants in these rows were spaced,

1 See Cook, O. F. Leaf-cut, or tomosis, a disorder of cotton seedlings. In U. S. Dept. Agr., Bur. Plant Indus. Cir. 120, p. 29-34, 1 fig. 1913.

75103°-18-Bull. 526-2

however, 18 to 24 inches in the row. When inspected on June 3, the old-method plants were branching normally, indicating fairly early thinning. At this time, also, the plants in the single-stalk rows were rather too far advanced for the best results, numerous fruiting branches being visible. Although thinning was completed within the two days following, on June 4 and 5, it is reasonable to assume that more significant results would have been obtained had thinning been done several days earlier. When thinned the plants were 8 to 12 inches high with 6 to 8 leaves. With a long-staple variety, this would have been more nearly the proper time to thin, but with Halfand-Half, a short-staple cotton, it unquestionably was too late. The plants were spaced about 8 inches apart.

There were 12 rows in the experiment, and the two systems of culture were compared in alternate rows. Two pickings were made, the yields of each row for each picking being as reported in Table V.

TABLE V.—Row yields obtained in a single-stalk culture experiment with cotton conducted in 1915 on the farm of R. K. Boney, Duckport, La.

Table V shows that the yield per row at each picking favored single-stalk culture in all but 4 instances, in one of which the yields were identical. In total yield, single-stalk culture led in four of the six rows. The total yield at each picking favored single-stalk culture by 10 and 6 per cent, respectively, the total increase for both pickings being 9 per cent. This difference is considerably less than was forecast by boll counts made on August 14, when it was found that single-stalk rows had from 40 to 50 per cent more bolls than oldmethod rows. The chief reason for this discrepancy between the boll counts and the yields reported may lie in the probability that more bolls opened earlier on the single-stalk rows than on the oldmethod rows and more of the cotton on the former was lost before picking time. This explanation appears the more credible if we take into account the fact that varieties of the type used in this experiment are lacking in stormproof qualities.

The Webb experiment.-Mr. R. C. Webb, of Delta Point, planted Money Maker cotton on sandy soil on April 20 at the rate of three pecks per acre. The stand secured was poor generally, there being also a number of skips in each row. Owing to nonuniformity in germination, there were, moreover, plants of various sizes in the single-stalk rows at thinning time. The smaller plants were still too young to have had their vegetative branches completely suppressed, while the larger plants had reached the stage where their fruiting branches were being suppressed. Since it was necessary to leave plants of all sizes in the row in order to have the spacing as uniform as possible, some allowances should be made in considering the results of the experiment.

There were 30 rows in the experiment and the two systems of culture were compared in alternate rows. The old-method rows were thinned on May 20, when the plants were 4 to 5 inches high and had 3 to 4 leaves. The single-stalk rows were thinned on June 6, when the plants were 6 to 12 inches high and had 6 to 10 leaves. The plants in the old-method rows were spaced 18 to 24 inches apart, while the single-stalk plants were spaced at about 8 inches wherever the stand permitted such close spacing.

Two pickings were made, the yields from each row for each picking being as shown in Table VI.

TABLE VI.-Row yields obtained in a single-stalk culture experiment with cotton conducted in 1915 on the farm of R. C. Webb, Delta Point, La.

Table VI shows a difference of 19 per cent in favor of the old system of culture at the first picking. The lower yield of the single-stalk rows at this picking doubtless resulted, as already explained, from the fact that the thinning was done too late in the case of some

plants and too early in the case of others in the same row. However, as often happens in such cases, the difference was practically offset by the yields of the second picking, which favored single-stalk culture by 15 per cent. There was practically no real difference in total yield, though the old system had a very slight advantage. This was contrary to what was to be expected, in spite of the poor stand, as counts of immature bolls made on August 14 indicated an increase for singlestalk culture of 15 to 20 per cent.

BOSSIER PARISH.

Two experiments were conducted in Bossier Parish, both in the immediate vicinity of Benton. One of these was on the farm of Mr. E. S. Burt, parish agent, and was conducted by Essie Jackson, a tenant; the other was on the farm of Mr. W. B. Wilbourn.

The Jackson experiment.-Cotton of the Brown variety was planted April 7 on low beds 4 feet apart. The soil, a sandy loam, was in good condition and a full stand was obtained. Wet weather prevailed during April, followed by dry weather during May, this being unfavorable to good growth.

The old-method rows were thinned on May 17, the plants, which were 3 to 4 inches high with 4 to 5 leaves, being spaced 18 to 24 inches apart. The single-stalk rows were not thinned until June 9, when the plants, being 10 to 14 inches high with 6 to 10 leaves, were spaced about 8 inches apart.

There were 26 rows in the experiment, and the two systems of culture were compared in alternate rows. Two pickings were made. The yields of each row for each picking and the total yields are reported in Table VII.

TABLE VII.-Row yields obtained in a single-stalk culture experiment with cotton conducted in 1915 by Essie Jackson on the farm of E. S. Burt, Benton, La.

Table VII shows a slight increase in the total yield from the first picking in favor of the old method, though in only 4 of the 13 instances did old-method rows yield more than single-stalk rows. In 7 instances the yields were identical. It is believed that the reason for this difference in yield at the first picking may be found in the fact that the single-stalk rows were thinned much later than was advisable, some suppression of the lower fruiting branches already having taken place. By the time the second picking was made this handicap had been overcome and the single-stalk rows yielded 22 per cent more cotton than the others, making a total increase in yield of 7 per cent.

An interesting point in regard to the value of single-stalk culture under conditions conducive to unusually rank growth was observed in connection with this experiment. A hogpen once stood near one end of several of these rows and the soil here was much richer than elsewhere in this field. The cotton plants grew to heights of 5 to 7 feet and put on a dense foliage. The plants in the early-thinned wide-spaced rows threw out long vegetative branches, and when nearing maturity these settled to the ground, in many cases breaking away from the main stalk and in all cases obstructing the space between the rows. It was difficult also to pick cotton from these plants, and owing to its being allowed to touch the ground much of the cotton was spoiled. The single-stalk plants, on the other hand, stood erect, their fruit well off the ground and easily accessible. Moreover, the space between these rows was left open enough to allow cultivation.

The Wilbourn experiment.-Cotton of the Brown variety was planted on April 7 in low beds in rows about 4 feet apart, and a fairly good stand was secured. Wet weather prevailed during the remainder of April, followed by drought during May, which was not favorable to good growth.

The thinning of the old-method rows was done on May 17, when the plants were 3 to 4 inches high, with 2 to 4 leaves, while the single-stalk rows were not thinned until June 9, when the plants were about 8 inches high and had 6 to 8 leaves. The plants in the oldmethod rows were spaced 18 inches apart and those in the singlestalk rows 6 to 8 inches.

There were 40 rows in the experiment and the two systems of culture were compared in alternate rows. Only one picking was made. The yields obtained from each row are reported in Table VIII.

Table VIII shows that in no instance did single-stalk culture yield more than the old method with which it was compared, the total increase for the old method being 17 per cent. This is absolutely the reverse of what was indicated on August 16 by boll counts