F. KiDD AND C^ West 12:> 



Halsted(2ti & 27) sliovvs that with Soy lieans the size of the seed is 

 controlled by the number of seeds in the pod and by their position in the 

 pod. The basal seeds and the seeds in one-seeded pods are the heaviest, 

 but these produce the poorest plants. Accordinfi to Halsted the seed,« 

 which occur in the middle jiosition in the pod give the most vigorous 

 plants. 



Working with Phaseohi>t vulgaris J. A. Harris (36, 37, & 39) found 

 that a correlation existed between the number of seeds in a pod and 

 the seed-weight, and that the seed-weight is influenced by the position 

 of the seed in the pod. In an earlier study with the same species 

 Harris (29) found that seeds of medium weight are more capable of de- 

 veloping into fertile plants than those ])roduced from the larger or 

 smaller seeds of which the weight exceeds or falls below the average. 

 With regard to germination, however, this observer (30) states that the 

 larger the seed the longer it takes to germinate. 



The same author (32), working with Staphylea, obtained indecisive 

 results. 



In the sugar beet two to five or more fruits are associated to form a 

 so-called "clump."" In this "clump"' the upper flower produces the 

 largest seed, the lateral seeds are markedly smaller. Briem(5) has 

 conducted a series of experiments to determine what effect the difference 

 in seed-size has upon the subsequent plant and its yield. He sowed the 

 individual seeds of a "clump" separately and compared the resulting 

 plants with one another throughout the whole course of their develop- 

 ment. The largest seed produced the largest seedling, the finest beet, 

 and lastly, the most fertile seed-parent. The figures for the relative 

 weights of five seedlings of a "clump" in one case were 100, 74, 67, 51 

 and 46 respectively. The full-grown beets derived from a single " clump " 

 weighed 1156, 859, 574, 344, and 310 grams and furnished respectively 

 241, 167, 202, 239, arid 104 grams of seed at the end of their second 

 year. Townsend and Rittue(69) found that sugar-beet plants produced 

 from "single-germ" seeds, which appear to be strictly localised on the 

 parent plant, not only gave a higher (about 30 %) percentage of germi- 

 nation, but were more vigorous and possessed greater vitality than those 

 produced from the so-called "multiple-germ" seeds. 



In a classical experiment many years ago Nobbe(57) showed in the 

 case of wheat and barley that variation in the size of the grains on the 

 ear is correlated with their position on the ear. A random distribution, 

 such as would occur if genetic factors were concerned, does not occur. 

 The largest grains occur in the middle-length of the ear. In the same 



