FIELD CEOPS. 339 



first hybrid generation have already been noted (E. S. R., 25, p. 436). From 

 data herein presented on the characters of the second hybrid generation, the 

 author draws conclusions which follow : 



Out of 316 plants, 43 were distinctly earlier in flowering than either the 

 Lyon or velvet beans grown in the same field, while 44 were much later. This 

 character evidently segregates. The pods and beans of some plants were much 

 larger than those of the Lyon bean, while those of other plants were smaller 

 than in case of the velvet bean. This character segregates, possibly with two 

 factors. 



The hairs on the shoots segregated into dark, velvety, and other mixed or 

 lighter classes which were not readily separable in the field. The ratio of 

 lighter to dark velvety hairs was 256 : 51. Dark velvety shoots were more or 

 less closely coupled with smooth convex leaves, dark velvety pods, partly with 

 late flowering, and in some cases with abnormal flowers in small bunches, 

 dropping of flowers, few pods to the bunch, and few seeds to the pod. The 

 purple color of the wings and, to a less degree, of the standard segregated in 

 the ratio of 3:1. The hairs on the pods segregated into long stiff, short stiff, 

 and long dark weak hairs with some intermediates. 



The lengths of the 5-seeded ix)ds gave a curve divided in the ratio of 3:1. 

 The opening of pods appeared to be a dominant character. The ratio of plants 

 with all closed pods to those with one or more open pods was 1:3. The 

 average weight of one seed reckoned as a percentage of the average weight of 

 a 5-seeded hull gave a number for use in comparison of the proportion of hulls 

 to seeds. This figure is low in the Yokohama bean, higher in Lyon, and highest 

 in the velvet beau. Among 115 segregates it gives a curve with two maxima, 

 one beyond the velvet bean. The average number of beans in the velvet and 

 Lyon pods were about the same, but the plants of the second generation hybrid 

 varied greatly in both directions in this respect. 



The crops of good seeds produced by the segregates showed much more varia- 

 tion than did those produced in the same field by parent strains of Lyon and 

 velvet beans. Some plants pi-oduced seed which appeared more subject to mold 

 than was the case with either Lyon or velvet, and the same was perhaps true 

 of wrinkled seed. Seeds from different plants, presumably of the same pure 

 line, showed gi'eater variation in average dimensions than did different pickings 

 from the same plant. Beau length was correlated with pod length and segre- 

 gated in the ratio of 3 : 1. Breadth and thickness did not visibly segi-egate inde- 

 pendently of the lengths, but the shorter beans were broader and thicker rela- 

 tively to their lengths than were the long seeds. The average breadth was 

 about the average of the Lyon and velvet strains. Mottling perhaps resulted 

 from 3 independent factors, and the ratio of mottled to unmottled beans was 

 apparently 63:1. 



A statement of the technique used in the crossing of stizolobiums is followed 

 by brief notes on pure lines of velvet beans, some corn crosses, and analyses of 

 the juice of West Indian sugar canes. B. 208 stood highest in Brix reading and 

 in percentage of sucrose as indicated by the polariscope. 



Corn production, G. Roberts and B. J. Kinney (Kentucky Sta. Bui. 163, pp. 

 173-200, figs. 4)- — Directions are given for corn production in Kentucky, includ- 

 ing soils and fertilizers, seed improvement, cultural methods, and harvesting. 

 Census figures showing the yield of corn in Kentucky in 1909, by counties, are 

 appended. 



The results of a test of rates of seeding and methods of planting are regarded 

 as indicating " that in a favorable season 3 stalks per hill will give a higher 

 yield than 2 stalks, and that drilled corn yields better than corn planted in 

 56022°— No. 4—12 4 



