162 ROLAND THAXTER ON THE 



occurrence that T have noticed was among caterpillars in Cambridge late in October. 

 The tendenc^^ of the affected hosts to crawl upwards before death, instead of concealing 

 themselves, as one Avould naturally expect, is noticeable in this, as in other Empusae, 

 and results in a far more effective dissemination of the spores. The GeuthojjMlus above 

 mentioned, which usually lives under logs and in similar situations, had crawled up a bush 

 about six feet high and hung suspended from the topmost twigs; grasshoppers also almost 

 invariably crawl nearly to the top of the culms of grass before death, and ai'e thus very 

 conspicuous and easily collected. Mr. Miyabe informs me that he has seen grasshoppers 

 in this position in Japan; but whether the Empusa by which they were attacked is E. 

 Orylli or not, I am unable to say. 



For some mention of European epidemics caused by "E. aidicae'' wiiich have been no- 

 ticed as f;\r back as 1835, reference should be made to the papers of Cohn and Bail al- 

 ready cited. 



Empusa Tenthredinis (Fresenius). 

 PI. 15, figs. 49-55. 

 EntornophtJioi'd Tenthredinis, Fresenius Z. c. B, p. 205, figs. 51-58. Sorokin I.e. C, p. 212, 

 fig. 652. 

 Conidia broadly ovoid, tapering slightly towards the apex and with a prominent, rather 

 narrow papillate base; 25 X 35,a-35 X 55/^ (62.5/^ = maximum length, sec. Fresenius). 

 Conidiopliores simple, coalescing over host. Secondary conidia like the primary. Rest- 

 ing spores unknown. Host attached to substratum by its legs. Khizoids not observed. 

 Hosts. Hymenoptei-a: larvae of Tenthredinidae. 

 JJalntat. Kittery, Maine; Europe. 



I have referred to this species a form found early in September upon a small Tentliredo 

 larva feeding upon Scutellaria in a swampy situation among woods. The larva when 

 found was hanging flaccid by its prolegs, and the conidiophores were just beginning to 

 emerge in small tufts over its body. The spores are somewhat smaller than those de- 

 scribed by Fresenius, nor is their difference from E. Grylli as decided as represented in 

 Fresenius' plate. A slight tendency to taper more ecpuilly from the middle of the spore 

 towards the base and apex, as well as its usually more delicate papilla, may separate it 

 from Grylli to which it is closely allied. It is ap])arently rarely met with even in Europe, 

 the only mention that I have seen beyond that of Fresenius being a i-eference by Cornu^ 

 to an Empusa on Tenthredo larvae. The i-eference and figure in Sorokin^ are taken from 

 Fresenius. 



Empusa conglomerata (Sorokin) ?. 

 PI. 15, figs. 56-62. 



Entomox>1dhora conglomerata, Sorokin I. c. A; I. c. B, p. 388, Taf. xiii, figs. 1-11; I. c. 



C, p. 228, figs. 571, 591, 592. 

 ? Empusa Grylli, Nowakowski (in part) I. c. B, p. 168. 



W. c. A, p. 189. ^l.c. 



