ENTOMOPHTHOREAE OF THE UNITED STATES. ]53 



show that these characters, as regards the formation of i-esting spores, are not sufR- . 

 ciently distinctive. Yet I doubt if this proves to be the case; and, should it prove 

 otherwise, 1 believe that the subgenus should be separated as a genus from other Em- 

 pusae. 



Under Entom,ophthora, which I have used in brackets as a subgenus, I have included 

 all forms characterized by the production of typically digitate conidiophores ; differing 

 from recent German writers in not considering this character of geneiic value, as well 

 as in omitting even as a subgenus the name Lamia given by Nowakowski to a single 

 form {JE. Guilds). 



In his extended paper on Entomophthoreae,^ ISTowakowski summarizes the generic 

 distinctions of Empnsa, JEntomopMTiora and of his genus Lamia, as follows: 



1. EsTOMOPHTHOitA. iuoludiug ovispora, curvispora, conioa, and Aphidis. 



Fungus growth, one-celled or with filamentous branches. 



ParapJiyses, rhizoids and columella, present. 



Conidiopliores, branched. 



Resting spores, zygospores (three species), az^'gospores (two species). 



2. Lamia including the single form Culicis. 



Fungus growth, filamentous. 



Paraphyses, present. 



lihizoids and columella, absent {nie ma). 



Conidiophores, unbranched. 



Resting spores, azygospores (borne terminally). 



3. Empusa including Grylli, Fresenii and Muscae. 



Ftingtts growth, filamentous. 

 Paraphyses and rhizoids, absent. 

 Columella, present or absent. 

 Conidiophores, simple. 

 Resting spores, azygospores. 



"Whether ]N"owakowski in his text gives more satisfactory characters for the genus 

 Lamia than are shown in the above table I am unable to say; but, from the data here 

 given, the presence of paraphyses (which I have apparently overlooked in examining the 

 species) is the single point which separates it from Lmpusa as defined below it. The 

 opinion of Brefeld that the form should be separated as a connecting link between Em- 

 pusa and EntomopMTiora, based upon a tendency to a digitate type observable in the co- 

 nidiophores and upon the presence of rhizoids is more readily understood, yet singularly 

 enough, and erroneously I think, I^owakowski affirms the absence of both these points 

 of structure. In my own opinion, LJ. Culicis cannot be separated from LJ. Muscae by other 

 than specific distinctions; the points of similarity of the two being decidedly greater 

 than their points of difference. The same may be said of the two species subsequently 

 described as E. papillata and E. apiculata which bear somewhat the same relation to 

 E. Grylli that E. Culicis does to E. Muscae; each having rhizoids and showing a slight 

 tendency, in the case of LJ. apiculata at least, to a digitate type of conidiophores. 



Taking the genus EntomopMhora, as defined by JSTowakowski in the same table, the 

 digitation of the conidiophores is apparently the only exclusive difference of importance 

 which separates it from Empusa. Even here LJ. Culicis and E. apiculata tend to break 



' I. c. B, p. 176. 



