170 MR. F. €. HORRELL ON THE STERIGMATA 
Of these gentian violet and Congo red gave decidedly the 
best results; the red staining the walls very sharply and making 
the sterigmata very distinct, whilst the gentian stained the 
nuclei only. This combination, however, was difficult to use, as 
the Congo red quickly washed out the gentian. Staining for 
24 hours in a strong aqueous solution of gentian violet, followed 
by about 10-20 seconds in a spirit solution of Congo red, gave 
the finest results. 
Some of the pieces were cut so as to show the basidia in 
longitudinal section, and others so as to show the tops of the 
basidia with their sterigmata in end-on or surface view. 
In all the preparations there could be no doubt that on nearly 
all the basidia there were only two sterigmata developed, those 
bearing three sterigmata not forming more than 1 or 2 per cent. 
of the whole, and those with four not 1 per cent. Notwith- 
standing this small relative proportion, however, the entire 
number of basidia seen bearing undoubtedly 8 and 4 sterigmata 
was considerable, but in no case was there any evidence of the 
spores being borne in two successive generations as W. G. Smith 
has described. When 8 or 4 sterigmata were present they all 
seemed to be of the same age, and basidia bearing more than two 
sterigmata were not more frequently present in old plants than 
in young ones. The basidia vary somewhat in shape, and con- 
siderably in size, even in the same lamella. They are club-shaped 
bodies about 25-34 u in length and from 61-13 p in width at the 
club-shaped end, and about 3 u at the attenuated base. Each 
sterigma is a conical outgrowth, tapering from a round base to à 
fine point on which the spore is borne, and varies in length from 
23-5 u. Between the basidia are numerous nearly cylindrical 
paraphyses about 20-27 p in length and 5 in width. The dark 
brown elliptical spores are, when ripe, about 83 p by 62. Below 
the hymenium are several layers of pseudo-parenchymatous 
cells, succeeded by the cells, running longitudinally, of the 
trama. , 
My investigations have clearly shown that, at any rate in cult- 
vated specimens, some degree of variation does occur. ‘There 
does not seem to be any reason why the same should not 
happen in wild plants, and possibly to a still greater extent, for 
their environment is subject to greater variation (e. f. P the 
