4IO 



BOTANICAL GAZETTE 



[JUNE 



The pollen grains in their enlarged state, owing to germination, measure 

 70X50^11.4 The walls are reticulately thickened and the thinner areolae 

 have in many cases given way and become perforations. The protruding 

 endospore can be seen in two cases to contain cells (Jig. 2, /), but their 

 nature is difficult to determine. They are probably of the same nature as 

 the cells that can be seen attached to one of the free antherozoids (fig, 2, Z^- 



In the absence of any evidence as to their nature, I think it best to 

 suggest that they are possibly of fungal origin, but, on the other hand, they 



Fic 



Fig. 2 



Fig. I. 



genostoiftd 



otdes 



cular bundles of the integument. 



Fig. 2, — The pollen chamber and its contents from the same ovule, X232; w/, 

 tongue of nucellar tissue which supports the pollen chamber; nw, nucellar wall of 

 the pollen chamber; a, antherozoid; as same cut across; e, endospore protrusion; 

 f S P^y P^y P^y four polkn grains; t, t^, tissue which is possibly fungal. 



Camera drawings from a slide in the R. H. College Collection, C N. 63. Locality, 

 Dulesgate. 



bear a remarkable resemblance to the tissue observed by Renault in the 

 pollen grains of Cordaites. If we had a case in which each of these cells 

 gave rise to an antherozoid, it would then be obvious that they were game- 

 tophytic. The central body in each cell has the appearance of a nucleus, 

 but is possibly merely the carbonized remains of the whole protoplast. 



4 By an oversight the dimensions were given ten times too small in Benson, loc, cit. 



