56 



Mycologia 



ful search was made in the field for clues that might lead to the 

 discovery of alternate forms for the true aecia. In two cases 

 most promising indications were detected. Aecidium passi- 

 floriicola was found a number of times so intimately and exclu- 

 sively associated with Puccinia Scleriae that little doubt seems to 

 remain regarding their genetic connection. Living host plants 

 for both rusts were sent to Lafayette, Indiana, and apparently 

 good telial material for cultures, but the teliospores could not be 

 made to germinate, and no cultures were secured. A. tubulosum 

 was also taken intimately associated with Puccinia substriata, but 

 this connection also needs confirmation by cultures. It is a 

 curious situation that with only five or six true unattached aecia 

 that may be considered heteroecious, there are some twice as 

 many species of grass and sedge rusts on the island, which pro- 

 duce telia in sufficient abundance to indicate that they may occa- 

 sionally complete their life cycle with aecia. Most of the grass 

 and sedge rusts of Porto Rico, however, evidently reproduce in 

 that region only by uredinia. 



It is interesting to note increased results obtained by systematic 

 efforts in collecting a certain class of plants. If we compare the 

 collections obtained by Clinton (1904), Holway (191 1), Stevens 

 (1913 and 1915), and Whetzel-Olive (1916), all by well-informed 

 uredinologists and excellent observers, it will be noted first that 

 the two later sets contain about five times as many species as the 

 two earlier sets. This is accounted for in part by the length of 

 time employed. But the two earlier collectors evidently did not 

 see many rusts in the field to collect, while Stevens required 

 nearly two years to secure as many species as Whetzel-Olive ob- 

 tained in about two months. All these collectors came to the 

 island without direct knowledge of tropical vegetation, and did 

 not know the plants by name. Thus the published list of the 

 Stevens collection, which Whetzel and Olive had in their pos- 

 session, did not help much at the time, because for the most part 

 they were not able to recognize the plants at sight to which the 

 names applied. It is highly probable that the difference in results 

 can largely be accounted for by a gradual, and quite likely, un- 

 conscious recognition of a marked difference between the macro- 

 scopic appearance of the majority of tropical rusts and those of 



