﻿520 Dodge: Effect of host on Gym 



but there also occur numerous two-celled spores mixed in with 

 the others. Certain specimens confined to the leaves were dis- 

 tributed by Seymour and Earle as G. hiseptatum var. foUicolum 

 (Econ. Fungi 244). Ellis also found this variety at Newfield, 

 New Jersey (N. Am. F. J^/p). Farlow's comments clearly set 

 forth the essential facts in regard to the two forms, and I quote the 

 following paragraph from his early work on the Gymno sporangia of 

 the United States (3) : "The spores of the present species (G. hisepta- 

 tum) are characterized by the great variability in the number of 

 cells of which they are composed. The most usual number is 

 three or four, two are rather common and occasionally there 

 are as many as six. . . . The spores of the present species when 

 full grown are not easily mistaken for those of any other species, 

 but the young tufts on the leaves often bear spores which are all, 

 or nearly all, two-celled. I have received specimens from Mr. 

 Ellis, with the fungus confined to the leaves, and it was difficult 

 to say to what species to refer it. Large sets of specimens collected 

 at Newton, however, show that while the young spots on the leaves 

 may have principally two-celled spores, those on the smaller 

 branches have about an equal proportion of two- and three-celled 

 spores, and the still older branches have a large proportion of 

 three-celled spores. In short, the variability is so great that with- 

 out a large set of specimens, one would have difficulty in convincing 

 himself that the extreme forms belonged to the same species." 

 It should be noted that Farlow has apparently seen spores with 

 more than two cells in the leaf son. Thaxter (20) first showed 

 the connection between G. hiseptatum and Roestelia Botryapites 

 on Amelanchier, but I do not find that he produced the infection 

 by using spores from son on the leaves. 



Kern (9) describes this leaf form on Chamaecyparis as a new 

 species under the name G. fraternum, on the basis of "Its habit 

 and morphological characters distinct from G. hiseptatum." 



Jackson (8) and Meineke (12) have found witch 's-brooms on 

 Libocedrus, supposed to be caused by G. Blasdaleanum, which has 

 hitherto been thought to be strictly foliicolous (9, 10). Jackson 

 used spores from such specimens and found that the aecidia 

 produced were the same as those derived from infections with 

 spores from the leaf form. This species furnishes an example of 



