528 RESEARCHES ON FUNGI 



what advantage does a Rust fungus gain by having its basidia 

 curved ? In the Hymenomycetes, the basidia are closely packed 

 together to form a hymenium, and they form a very compact layer 

 with a comparatively smooth outer surface. Owing to the extreme 

 crowding of the hy menial elements, it is necessary that the sterig- 

 mata should have a terminal position on each basidium. More- 

 over, since the sterigmata are always parallel to the basidium-axis 

 and perpendicular to the surface of the hymenium, the spores which 

 they produce are from the first directed away from the hymenium. 

 The spores, when discharged, must therefore be shot perpendicularly 

 outwards from the hymenium, i.e. in the best direction to give them 

 a chance of escaping from the fruit-body. I consider that the form 

 of the basidium in typical Hymenomycetes is perfect for secur- 

 ing efficiency in the production and liberation of spores from the 

 hymenium. Its special features are: (1) its compact body, (2) its 

 non-septate condition so that there is no expenditure of material 

 in the production of interior cell-walls, (3) the short sterigmata in 

 the construction of which but little material is used up, and (4) the 

 delicate adjustment in the spatial relations of the sterigmata and 

 spores. The sterigmata and spores are set as closely together as 

 possible, having regard to the necessity of freedom for each spore 

 during its development and discharge. If these four points are 

 considered, it will be seen that the basidium in the Uredineae is not 

 nearly so highly organised, for in the basidium of a Rust fungus we 

 find : (1) that the basidium is often considerably elongated, (2) that 

 three or four septa are always present which divide the end of 

 the basidium into four cells, (3) that the sterigmata are often rela- 

 tively long, and (4) that the sterigmata and spores are not closely 

 and compactly arranged in relation to one another but, relatively, 

 are sprawling. However, I am of the opinion that, although 

 hymenomycetous basidia are admirably fitted for hymenomycetous 

 fruit-bodies with their highly organised hymenia, yet uredinous 

 basidia are just as admirably suited for their function in connection 

 with uredinous fungi. In the Uredineae there is no hy menial layer 

 such as we find in the Hymenomycetes. Instead, the basidia, on 

 developing from the teleutospores, remain separated from each 

 other. Now, in order that the sporidia shall be shot away from 



