Studies IK Er.toHjger.ous Fungi. T. Fetch. 109 



Poey," III (191S), p. So. On a scale insect, Saissdia hemi- 

 sphaerica, on Achras Sapota, Cuba. 



T. tenuis Fetch in Ann. Perad. xu {1922), p. ^2^. On 

 Aspidiotus, Aleyrodes, etc.. Ceylon. 



T. sublintea Petch, loc. cU. p. 324. On an Ale\Todid, Corral, 

 Chili. 



T. barda Petch, loc. cit. On an Ale\Todid, Corral, Chili. 



TorrubieUa is a Cordyceps (Torrubia) without a clava, the 

 perithecia being borne on a weft of h\-phae, instead of being 

 embedded in a fleshy stroma. The generic description does not 

 reveal much difference from Ophionectria, and it would be 

 expected that specimens which came into the hands of the 

 earher mycologists, especially those on scale insects where the 

 entomogenous nature of the fungus is not alwavs immediately 

 e\-ident. would have been placed in Xtxtria, or later in Ophio- 

 nectria. But I have failed to detect an\- species of TorrubieUa 

 among the Xectriae in the herbaria of Kew and the British 

 Museum. 



On the other hand, the resemblance of the perithecia and 

 spores to those of Cordyceps, coupled with the occurrence of 

 the fungus on insects, ma^- have led to the inclusion of species 

 of TorrubieUa in that genus. 



Moller (Phycomyceten und Ascom3'ceten, 1901) declined to 

 admit the vahdit\- of Boudiers genus TorrubieUa. He stated 

 (p. 144) that Boudier had separated from Cordyceps, as Torru- 

 bieUa, those species which did not form a stroma outside the 

 insect, but produced only perithecia or a loose weft of hvphae. 

 He objected to that on the ground that there were onlv few 

 such species in comparison with the total number of Cordyceps 

 recorded, and nothing was gained by the separation. Of the 

 species of Cordyceps described and figured bv ^loller, Cordvceps 

 Jiavo-viridis, C. cristata, C. rhynchoticola, and C. gonyhpiicida 

 appear to be TorrubieUa. According to Moller, the last-named 

 species has not the smallest trace of a stroma, the perithecia 

 occurring on the insect %^'ithout any film of mvcehum; hence 

 he concluded that his species was different from TorrubieUa 

 aranicida Bond. Perithecia which have onlv a ver\' shght basal 

 weft of h\-phae do, however, occur in species which have 

 normally a well-developed bvssoid stroma, and so the character 

 on which Moller rehed mav not be constant. 



Moller further objected that the genus TorrubieUa was not 

 well founded, because of the existence of transitional forms 

 between TorrubieUa and Cordyceps. In Cordyceps flavo-viridis, 

 the perithecia are situated on repent strands of mvcehum 

 spreading from the floccose stroma which covers the host insect ; 

 Moller pointed out that in this condition the fimgus was Torru- 



