1906] Brainerd, — Hybridism in the Genus Viola 57 



674,) has included under V. villosa cordifolia seems to be this. Both 

 supposed parents were abundant in 1903 at the Sherwood station in 

 a dry woodland; but the trees have since been cut. 



It remains to speak of several cases in which unlike forms are pro- 

 duced by the crossing of two doubtfully distinct species. 



28. V. FIMBRIATULA X SAGITTATA. — The most conspicuous dis- 

 tinctions between these two species are found in the foliage. They 

 may be thus stated: pure V. sagittata has narrower leaves, dilated 

 and incised basal lobes, longer petioles, and is glabrous; (cf. fig. c of 

 pi. 66 with fig. c of pi. 67). The progeny of the cross instead of pre- 

 senting a halfway state between these divergent characters, is often 

 quite like one parent in some of these features and quite like the other 

 parent in other features. Thus in this particular cross we have four 

 groups of forms: 



A. The breadth of leaf of V. fimhriatula and the glabrous surface of V. 

 sagittata. (See pi. 68, fig. c.) 



B. The leaf-outline of V. sagittata and the pubescence of V. fimbriatula. 

 (Fig. d.) 



C. A compromise in leaf-outline, forming a deltoid leaf, either with 

 incised lobes at the base (fig. a.), or without (fig. e.), and with 

 either much or little pubescence. 



D. The lower half of the leaf may be broad and the upper half abruptly- 

 narrowed (fig. b.), and this either with incised lobes at the base or 

 without, and with a varying amount of pubescence. 



In this plate only a few of the variant forms are given. One needs 

 to visit a station where hundreds of specimens can be seen to appre- 

 ciate this tendency to sport. A fine station for this was discovered 

 by Miss Ryan, on the north side of a railway embarkment across a 

 salt marsh, at Giant's Neck, East Lyme, Conn. The plants, or their 

 seeds, were evidently brought there with the soil, and no other violets 

 were found there. Other stations are: "Gristmill Park," Plainville, 

 Conn., Luman Andrews, 1904; Blue Hill Reservation, Milton, Mass., 

 H. A. Purdie; Lindenau, near New Brunswick, N. J., H. D. House. 

 — Various forms have been seen from other stations, but I have 

 visited only the four above named. In all these forms there is little 

 or no impaired fertility. 



29. V. EMARGiNATA X FIMBRIATULA. — Here wc have a similar 



