PAPERS ON BOTANY 1*1 



to be smaller. Exact measurements proved that the extremes 

 in size are the same in both cases, the smallest being - 3 /* by 4/* 

 and the largest 4.5 n by 6/a, but that the average size of the 

 type spores (3.53 /><• by 5.16 /*) is slightly smaller than the 

 Havana material, (6.56 /* by 5.4 /*). 



Apparently the specimens of Triplasis purpurea attacked by 

 the smut are different from those which are not. They are 

 only one-third as tall as the herbarium specimens, but are 

 more freely branched ; the internodes are so shortened that 

 the sheathes overlap; the panicles are all partially or wholly 



enclosed in the sheathes instead of the largest ones being long 

 exserted; the spikelets are somewhat larger; the ovaries have 

 developed into rather large ergot-like bodies about 4 mm. 

 long and 1.5 mm. in diameter, and the glumes, lemmae, and 

 paleae are membranous and paler in color, but the number, 

 shape and arrangement of these flower parts are the same. 



However, these differences are not all due to the action 

 of the fungus. It will be remembered that the collection was 

 made late in the fall. Enclosed and cleistogamous spikelets 

 are characteristic of late specimens in the genus Triplasis, in- 

 cluding Triplasis purpurea 4 . This would account for the lack 

 of exserted panicles and probably for the difference in tex- 

 ture of the flower parts. But the abnormal development of 

 the ovaries is entirely due to the fungus. 



In order to understand the occurrence of Ustilago sieglin- 

 giae in Havana we must consider its geography. It is situated 

 in a sand deposit in the midst of a broad glacial flood plain. 

 Its geologic conditions are therefore local, so that the smut 

 must have reached it by transportation through short distances 

 from sand patch to sand patch. It is surprising then, that it 

 has not been collected from some sand area intervening be- 

 tween Punta Rassa and Havana, two places in which the 

 meteorological conditions are very different. If, as is possible, 

 it attacks only the autumnal forms of its host, it has probably 

 not been obtained from these other localities because collectors 

 have searched for it earlier than it appears. We believe that 

 it can be found in such regions when the conditions are the 



