1920] THAXTER— FUNGUS-PARASITES 7 



As far as can be determined from the series of specimens ex- 

 amined, there seems to be no continuous increase of the fungus in 

 diameter after the original proliferation of the primary stage, which 

 gives rise to the sporogenous region. This is indicated by the fact 

 that this region, as soon as spore formation has begun, is surrounded 

 completely by sterile indurated structures, incapable of radial 

 extension, and also by the fact that the bristles of the host, which 

 are surrounded by the filaments of the preliminary stage, are not 

 bent down as by an advancing margin, but retain their normal 

 position, and may even be seen projecting beyond the hy menial 

 surface of mature individuals, as in fig. 10. 



Although each individual must produce an enormous number of 

 spores, this very curious type does not appear to have been very 

 successful in propagating itself effectively; for although its hosts 

 are densely gregarious and live under conditions which should be 

 very favorable for the communication and development of such 

 parasites, hardly more than i per cent of the individuals in an 

 infected nest appear to bear the fungus. Snyder, who was the 

 first to observe the type species of this parasite and to whom I am 

 greatly indebted for the original material examined, informs me 

 that he has found this ratio of infection more or less constant in 

 material from a number of different sources, and Smulgan, 

 who has also kindly communicated material from the Boston 

 region, makes a similar estimate. In the case of the second species, 

 described from the Island of Grenada, I have also found almost 

 exactly the same percentage of diseased individuals among the 

 several thousand hosts examined. 



It does not appear seriously to inconvenience the insect on 

 which it grows, and the only indication of injury is a slight brown- 

 ing of the tissue immediately opposite the primary attachment, 

 as shown in fig. 14, although all the cells of the tissue lying immedi- 

 ately below the integument are hyper trophied, wherever the fungus 

 is in contact with the host, often assuming a rather regular palisade- 

 like structure, similar to that shown in fig. 10. It is most con- 

 spicuous when growing on the abdomen (fig. 9) , where it is likely 

 to assume a more regular and rounded form, being suborbicular, or 

 more often transversely elongated, with an even or sometimes 



