4o8 Journal of Agricultural Research voi. xvni, No. s 



stage of Cordyceps sphecocephala, and the occurrence of an analogous 

 structure in Sorosporella suggests that it also represents the imperfect 

 stage of some species of Cordyceps or an allied type. 



A second series of infections was made with army worms, Cirphis 

 unipuncta, and as in the instance described above, freshly dead insects 

 were employed. The artifically infected larvae were, however, nearly all 

 in the pupal stage at the time of death, the pupa case forming a thick, 

 resistant, chitinous envelope about their dead bodies. Such pupae were 

 placed in moist chambers at periods varying from one week to two 

 months after death; and although they were completely filled with 

 resting spores which were quite normal in appearance, as was shown by 

 subsequent examination, they produced no external conidiophores, nor 

 did any of the resting spores germinate. When the pupa cases were 

 torn open, however, permitting a free circulation of air about the fungus, 

 germination was readily induced. 



The results of the foregoing experiments show, therefore, that the 

 resting spores or chlamydospores do not necessarily require a long period 

 of rest before germination, a period that may be necessarily more or less 

 protracted in certain other fungi, but that if suitable conditions obtain, 

 germination may take place at once. The tests also show that when 

 insects die in the pupal stage, germination does not take place at once 

 unless the thick chitinous wall is broken. 



In order to determine the longevity of the resting spores as well as 

 to ascertain whether or not fresh insects could be infected with them, 

 other tests were performed as outlined below. 



Infected cutworms which died in June, 1916, were kept in the labora- 

 tory in ordinary pasteboard pill boxes until August 2, 191 7, when a 

 number of resting-spore masses were removed from them and placed 

 in hanging drops in Van Tiegham cells. On August 6, they began to 

 germinate; and subsequently conidiophores and an abundance of conidia 

 were formed. This result demonstrates that the resting spores may be 

 viable after they have been subjected to desiccatory conditions for 14 

 months. 



In October, 191 7, a number of infected army worm pupae were buried 

 out of doors about 2 inches beneath the surface of the soil. On March 8, 

 1 91 8, they were exhumed and examined. So far as could be determined 

 microscopically, no change had taken place; they appeared just the same 

 as when buried the autumn before. One pupa was broken open and spore 

 masses placed in Van Tiegham cells, while the pupa itself was placed in 

 a moist chamber in the labora,tory. On March 1 1 the resting spores in 

 the Van Tiegham cells began to germinate, and conidiophores could be 

 seen with the hand lens on the pupa in the moist chamber. The winter 

 of 1 91 7-1 8 was exceptionally severe, yet the chlamydospores were 

 apparently unharmed. Their germination shows clearly that they are 

 able to withstand cold as well as desiccation. 



