June 14, 1907] 



SCIENCE 



941 



of Viola, and also from the seeds of about 

 twenty pure species. The behavior of these 

 two classes of seedlings was surprisingly un- 

 like; the offspring of the pure species re- 

 sembled each other closely, but the offspring 

 of the hybrids were often much unlike each 

 other and unlike their immediate parents, re- 

 verting in some qualities to one parent of the 

 hybrid and in other qualities to the other 

 parent, and this in a great variety of ways. 



The species involved in these experimental 

 cultures all belong to the group commonly 

 known as 'blue stemless violets,' of which 

 V. palmatdj V. cucullata and F. sagittata are 

 familiar examples. Of this group there are 

 about twenty species in the northeastern 

 United States. I may not repeat here the 

 proof, published elsewhere,' that these closely 

 allied species, when growing together, freely 

 interbreed. I will merely say that the facts 

 to be presented in the present paper furnish 

 most positive confirmation of the opinion that 

 these anomalous plants are hybrids. 



One of the corollaries of Mendel's law is 

 that each pair of contrasting characters in a 

 hybrid works out its effects, for the most part, 

 independently of all other pairs. As in New- 

 ton's 'Law of the Coexistence of Motion,' the 

 final result is but the summing up of the vari- 

 ous component movements taken separately. 

 It will be simpler for us, therefore, in de- 

 scribing the behavior of violet hybrids, to con- 

 sider each pair of characters by itseK, taking 

 up, in order, the divergence that occurs in 

 respect to leaf-outline, in respect to pubes- 

 cence, in respect to size, and lastly in respect 

 to color of capsules and of seeds. 



1. A striking illustration of diversity in leaf- 

 form was seen in the offspring of Viola cucul- 

 lata X septemloha. This hybrid was published 

 by Mr. Bicknell in September, 1904, as a spe- 

 cies, Y. notabilis. It has been found in five 

 different stations, always growing with the re- 

 puted parents. In June, 1904, I received 

 from New Jersey one of these plants that I 

 have grown now for three seasons. From its 

 cleistogamous capsules, which of necessity give 

 pure cultures, I collected seeds in 1905 that 



^See Rkodora, VI., 213-223; VIII., 6-10, 49-60. 



furnished the following summer ten vigorous 

 plants. These bore in August and September 

 an abundance of cleistogamous flowers that 

 matured seeds; several plants bore also in the 

 autumn petaliferous flowers. 



The leaves of the parent species are very 

 dissimilar, that of V. cueullata being broadly 

 heart-shaped and pointed, that of V. septem- 

 loha (V. Brittoniana) primarily 3-parted, with 

 the segments 2-4-lobed. The hybrids of these 

 two species in all the five known stations 

 exhibit a fair compromise in leaf-outline be- 

 tween the two quite unlike leaves of the pa- 

 rents, and closely resemble each other. They 

 show about the same number of lobes as in 

 V. septemloha, but the lobes are shorter and 

 broader, the sinuses only half as deep. But in 

 the offspring of this hybrid we have in addi- 

 tion to plants with this compromise leaf -form, 

 plants with leaves but slightly lobed and show- 

 ing the cordate base and acute apex of V. 

 cucullata, and still other plants in which the 

 leaf-outline is almost a complete reversion to 

 V. cucullata. In the living plants that dis- 

 play, each, twelve or more leaves of these sev- 

 eral patterns, the impression of dissimilarity 

 is most striking. 



Another marked ease of diversity of leaf- 

 outline in the progeny of the same hybrid was 

 seen in V. fimbriatula X septemloha. This is 

 Mr. Pollard's V. Mulfordce, found first in 

 1902 at Hempstead Plains, Long Island. It 

 has turned up during the past season in two 

 other stations, but always in close proximity 

 to the two parents. As before, the leaf of the 

 hybrid is markedly intermediate between the 

 very unlike leaves of the parent species; but 

 not so with the leaves of several plants that I 

 raised from this hybrid the past summer. The 

 seedlings all came from the self -fertilized cap- 

 sules of plants sent by Miss Mulf ord from the 

 type station, but appeared most dissimilar in 

 foliage, though growing side by side under the 

 same cultural conditions. Some plants bore 

 leaves like those of the parent hybrid; others 

 bore leaves resembling those of V. septemloha 

 in width, but those of V. fimbriatula in having 

 only basal lobes ; and still other plants were in 

 leaf-pattern complete reversions to V. fimhria- 

 tula. 



