Jan. 15. 1920 Further Studies of Sorosporella uvella 403 



and bearing similar conidia. In the following description of Sorosporella 

 uvella it will be seen that the method of development, including the in- 

 ternal formation, coherence, and size of the resting spores, the yeastlike 

 reproduction in culture, the character of the conidiophores, the shape 

 and size of the conidia, etc., so readily conform with analogous characters 

 of Acremonium cleoni, w^hich, furthermore, parasitizes the same insect, 

 that there can be no doubt as to the identity of the two species. 



A third species, Fusariwm acremoniopsis Vincens {22), should be con- 

 sidered in this connection; and while the measurements of the bodies 

 which Vincens terms its chlamydospores (3.5 by 4 microns) as well as the 

 measurements of its conidia (4 by 7 by 2 by 3 microns) do not coincide 

 with similar structures of Sorosporella uvella, there are nevertheless many 

 points of resemblance between the two fungi which cannot be disre- 

 garded and which show, in the writer's opinion, that they, though per- 

 haps not identical, are at least very closely related. 



Vincens described his fungus as a parasite of the "vergris" (cut- 

 worm ?) from Brazil. The insect when dead was dark brown in color, but 

 no external fructification was evident. Internally, however, Vincens 

 found mycelia bearing short, undifferentiated conidiophores at the tips 

 of which cylindrical, one-celled conidia were produced singly. His 

 illustration of such hyphae is very similar to certain structures found by 

 the writer when Sorosporella uvella was grown upon artificial nutrients 

 (PI. 52, O). It must be noted, however, that no such structures were 

 ever observed in cutworms infected by the same fungus. 



Vincens placed fragments of the caterpillar in a moist chamber; and 

 from these fragments conidiophores developed, the branches of which 

 had a tendency to group themselves verticillately, a tendency, it will be 

 shown, that is noticeable also in the conidiophores of Sorosporella uvella. 

 The cylindrical spores borne on these conidiophores were largely uni- 

 cellular, but a certain number were curved and provided with one or 

 more septae. 



At a later date Vincens discovered, in dried-out fragments of the cater- 

 pillar, large numbers of brown, globular spores, grouped in spherical or 

 oval masses, the illustrations of which resemble very closely the resting- 

 spore aggregations of Sorosporella uvella, which, as the writer has in- 

 dicated, are distinctive bodies and serve at once to differentiate Soro- 

 sporella from all other entomogenous fungi. 



There is, therefore, considerable resemblance between Fusarium 

 acremoniopsis and Sorosporella uvella; and in spite of the fact that the 

 measurements of their respective reproductive bodies are dissimilar, as 

 well as the fact that the conidia of F. acremoniopsis are said to be some- 

 times septate, the writer believes that Vincens' fungus, though perhaps 

 not 5". uvella, is at least some closely related species. 



A few other papers, chiefly in the form of taxonomic notes, such as 

 those of Thaxter {21), Danysz and Wize (5), and Lakon {12), have 



