Studies in Enionwgenoits Fungi. T. Peich. iii 



a 5/t/&«m-like conidial stage, which belongs to the genus 

 Hirsutella. 



There are kno^Ti, therefore, the following species, which have 

 either been recorded as Torrubiell^, or should probablv be re- 

 ferred to that genus. 



On spiders — Torrubidla aranicida Boud., T. flava Fetch, 

 Cordyceps gonylepticida MoUer, Hdminthascus arachnophthorus 

 Transzch., and Barya montana Rac. 



On Lepidoptera — Torrubiella rostrata P. Henn., T. ochracea 

 Pat., T. sericicola v. Hohnel, and Cordyceps crisiata MoUer. 



On Rhynchota — Cordyceps rhynchoticola Moller. 



On undetermined insects — Cordyceps flavo-virid is Moller. 



On "Coccids" — Torrubiella tomentosa Pat., T. rubra Pat. 

 and Lagh., T. luteorostrata Zimm., T. brunnea v. Keissl., 

 T. Lecanii Johnst., T. tenuis Petch, T. sublintea Petch, and 

 T. barda Petch. 



The present account deals chiefly with the last group. It is 

 based on a series of specimens collected in Cevlon and the Eastern 

 Tropics, and others from South America kindl\- furnished bv 

 Professor Thaxter. 



\ery many of the species of the genus Cordyceps have been 

 described from single specimens, and, except in the case of a 

 few weU-known species, the herbarium material is scant v, or is 

 represented in one herbarium only. The host insect, in many 

 cases, is not gregarious, and it may hapjjen that only one 

 example of the fungus can be found. The same is true of the 

 species of Torrubiella on Lepidoptera and Arachnida. The latter 

 are usually found, in the tropics, on the under side of leaves, 

 attached to the leaf by a weft of mvcehimi, and under such 

 circumstances, seeing that the insects can travel about at ran- 

 dom, it is scarcely to be expected that a nimiber of examples 

 of the fungus will be found together. 



In one locahty in Ceylon, however, there is a marked exception 

 to the theorv^ just enunciated. The species concerned is Cordy- 

 ceps dipterigena B. and Br. It is found in the jtmgle at Hakgala, 

 where it occurs on flies which are, as a rule, attached to the 

 lower side of small branches or the imder surface of leaves, 

 and if one searches carefully, if one be found, there are usuallv 

 at least half a dozen. That has been mv experience on several 

 occasions during the last twelve years, and circumstances 

 suggest that the insects are infected during the period when 

 they are in close association, i.e. in the lar\-al stage. 



Species of Torrubiella on scale insects (includinsr Aleyrodidae) 

 can usually be collected in fair quantity, when they do occur, 

 because, in general, large numbers of the insects occur together 

 on the same host plant 



