92 BRAINERD: HYBRIDS OF THE PALMATA GROUP 
Dr. Greene’s conception of a species, which without hybrid- 
izing appears ‘under forms enough to make three,” is quite 
contrary to the latest inductions of biology. The law that ‘“‘like 
produces like,” even through seed (when pure or not affected by 
foreign pollen), is confirmed by numerous experimental tests. 
If we except the extremely rare cases of observed mutation, the 
offspring of a species or of a natural variety is always uniform.* 
Whenever a miscellaneous offspring appears, we may be confident 
that at no very remote past, species or varieties, as unlike or 
more unlike than the offspring, were sexually united in the 
parentage.t 
Several other examples of V. papilionacea X triloba might be 
cited, but I name only two from well known collections: No. 34, 
North American Violaceae, Greene and Pollard; ‘‘V. palmata 
dilatata Ell.” New Springville, Richmond Borough, N. Y., 
Wm. T. Davis, July 17, 1903; also no. 5108, colony 3, Violets of 
Philadelphia and vicinity; ‘ V. te i dilatata Ell.” Sherwood, 
Pa., June 17, 1903. 
Viola sororia X triloba hyb. nov. 
The confluence of these two species is generally recognized, 
and it is so common that many students of Viola have regarded 
them as forms of the same species; just as, because of a similar 
confluence, V. sagittata has been held to include V. fimbriatula. 
The shallow and obscure lobes of the hybrid leaf are the same as 
in V. papilionacea X triloba, but the foliage is never glabrous. 
But more satisfactory than this negative test is the discovery of 
the intermediate forms in a region from which V. papilionacea 
is absent. Along a shady limestone ledge in Orwell, Vt., where 
V. triloba and V. sororia were abundant, but no V. papilionacea, 
was found in 1904 a large colony of intermediates that will pass 
muster as V. sororia X triloba. In this case the most satisfactory 
evidence would come from the artificial production of the hybrid; 
this I trust will soon be attempted. 
* Dimorphism, such as th . 
f both staminate and pistillate 
plants, or of long-styled and short-styled forms, belongs to another class of 
phenomena. 
t See DeVries, H. Species and varieties. 
Lectures 6 & 7, on stability and 
vicinism. 
