336 EXPERIMENT STATION EECOED. [Vol. 41 



tilizer experiments with cotton on Coastal Plain soils, conducted in coopera- 

 tion with farmers for the period indicated in continuation of similar wor'c 

 previously noted (B. S. R., 30, p. 636). The results secured may be sum- 

 marized as follows : 



The average increase, following the use of cottonseed meal, was greater than 

 that from acid phosphate in 68 per cent of the experiments, equal to it In 16 

 per cent, and less in 16 per cent. Cottonseed meal resulted in an average in- 

 crease greater than that secured from kainit in 73 per cent of the tests, equal 

 to it in 7 per cent, and less in 20 per cent. Acid phosphate surpassed kainit 

 in 52 per cent of the tests, equaled it in 14 per cent, and was less in 34 per cent. 

 With cottonseed meal as a basis, the relative increases in crop yields, due to 

 the use of 200 lbs. cottonseed meal, 240 lbs. acid phosphate, or 200 lbs. kainit, 

 were as 100 : 45 : 34. As an average of a much greater number of experiments, 

 however, it is stated that potash is of much more relative importance, and acid 

 phosphate slightly more so, than indicated above. 



A 100-lb. application of kainit was found to be nearly as effective and, on 

 the average, more profitable than a 200-lb. application. Kitrate of soda, ap- 

 plied at the rate of 100 lbs. per acre where the plants were from 6 to 8 in. 

 in height to make a complete fertilizer, was found to be slightly more effective 

 than a 200-lb. application of cottonseed meal applied in a complete fertilizer 

 before planting. The maximum average increase in yield and the highest 

 average profit were obtained from an application of 240 lbs. acid phosphate 

 and 100 lbs. of kainit per acre before planting, and 100 lbs. of nitrate of soda 

 when the plants were from 6 to 8 in. high. 



Transmission of characters in potatoes, C. von Seelhobst {Jour. Landw., 

 66 {1918), No. 3, pp. lJtl-162). — In continuation of work previously noted (E. S. 

 II., 16, p. 3G0), the author presents further evidence regarding the transmis- 

 sion of yielding qualities in potatoes by the selection of seed from high-yield- 

 ing strains. It is stated that almost always the use of small seed tubers from 

 large mother plants resulted in higher yields than the use of large tubers from 

 small plants. Furthermore, the yield expressed as a multiple of the seed em- 

 ployed increased markedly with a decrease in the size of the seed tubers. 



Missing hills in potato fi.elds: Their effect upon the yield, F. C. Stewabt 

 {New York State Sta. Bui. ^59 {1919), pp. 45-69, figs. S). — Experimental work 

 conducted during 1918 is described in which an effort was made to determine 

 liow much of the loss in yield of potatoes, due to missing hills, is compensated 

 by increased yields of the adjoining plants. Each of 3G0 tubers was divided 

 lengthwise into two equal parts, and the resulting seed pieces planted in ten 

 rows, 36 in. apart and 15 in, apart in the row in groups of four, with blank 

 spaces of 30 in. between the gi'oups. Each group contained the pairs of seed 

 pieces from two tubers, so that one meml)er of each pair adjoined a missing 

 hill. The difference in yield between the exterior plant and its mate, the 

 interior plant, was taken as a measure of the effect of the missing hiU. Data 

 were secured from 351 pairs of plants. As a check on the experiment, an at- 

 tempt was also made to ascertain the extent of the variation in yield occurring 

 between the plants from two seed pieces from the same tuber when grown 

 under parallel conditions. For this purpose the pairs of seed pieces were 

 planted 15 in. apart in two continuous rows without blank spaces, and data 

 secured from 85 pairs of plants. The weight and number of tubers per plant 

 are recorded in tabular form and the results summarized as follows : 



Based on the weight of the total yield, the exterior plants outyielded the 

 interior ones on the average 23.2 per cent, indicating that in the case of a 

 skip containing a single missing hill, the two adjoining plants together make up 

 46.4 per cent of the loss ju total yield, and a little more in the yield of market- 



