MUTATIONS. 23 



It is evident from the diversity of the propjeny that the parents 

 were hybrids in all the cases listed. The number of plants grown in 

 each case is too small to secure definite proportions, but it is clear 

 that the color of the pubescence and the color of the seed behave in 

 Mendelian fashion. The same is probably true of the flower color, 

 which was counted in only one case. 



There is thus furnished a clear explanation of the origin of many 

 of the new varieties at the Arlington Experimental Farm that were 

 at first mistaken for accidental admixtures. It also accounts for the 

 diversity of the population exhibited in manj'- introduced varieties 

 notwithstanding the apparent uniformity of the seed. 



It must not be supposed from the foregoing account that hybrids 

 are common in soy beans. At Arlington the test rows are grown 

 contiguously, so that there is great opportunity for cross-pollination. 

 Nevertheless, the percentage of hybrids that occur is very small, per- 

 haps not one individual in two hundred. 



Thus far the hybrid plants have been detected mostly by the color 

 of the seed. In a number of cases none of the progeny has seed 

 similar to the parent; or, in other words, the color of heterozygote 

 seeds is often unstable. Among the most striking of such heterozy- 

 gote seeds (PI. VIII) are yellow with a single narrow transverse 

 band of brown; yellow or green, with an irregularly star-shaped 

 brown or black figure centering at the hilum; and green or yellow 

 more or less suffused with a smoky color. Some of the last breed 

 true, but most of them do not. 



Heterozygote plants, especially where the seeds are largely or 

 wholly yellow, are often distinguishable by the unusual form of the 

 pods near the tips of the branches. These are more tumid than the 

 other pods and the seeds more crowded. Such pods may also be 

 thinner in texture and much less hairy. Illustrations of this phe- 

 nomenon are shown in Plate VII. 



MUTATIONS. 



The origin of new varieties of soy beans without hybridization has 

 apparently occurred in certain cases that have come under our obser- 

 vation. From a theoretical standpoint there can be no doubt that 

 the fundamental diversity in a plant, especially when normally self- 

 pollinated, is brought about by other causes than hybridization. It 

 is self-evident that there must be two different varieties to cross 

 before crossing can become effective in producing new varieties. 

 Most soy-bean varieties when pure remain very constant to type, so 

 that any chance variation is quickly detected. There are two cases 

 in which the evidence is fairly satisfactory that a brown-seeded 

 variety arose as a mutation from a yellow-seeded sort. 



197 



