188 Proceedings of the Biological Society of Washington. 



changes to pupa and issue very shortly as adult 9 ; while one 

 that will produce a male must become in sequence: — uni- 

 oviparous paedogenetic form, — egg, — curculioid 1st larva, — 

 metrophagous larva, — short-legged prepupa, — pupa, and adult 

 d\ By this time all of his sisters will have lived their short 

 lives and died. 



A detailed account of one of the most significant series of 

 observations in which the progeny of one paedogenetic mother 

 developed to reproduce by all three methods above shown, may 

 put the matter in a clearer light to the reader. 



A colony of larvae was found in a pine log at Natural Bridge, 

 Ky., in September, 1912, by Mr. T. E. Snyder, who gave some 

 fragments of wood containing cerambicoid larvae to the writer 

 on January 3, 1913. These were placed in a plaster cell and 

 on February 8 the colony was found to be maturing into 

 paedogenetic individuals. From these latter a number of iso- 

 lations were made, and one large community cell was started 

 of the progeny found in the cell of a paedogenetic mother 

 whose shrunken body had probably been consumed by the 

 young as has since been almost proven to be the rule. These 

 twenty-one, first-stage, legged larvae were placed in a plaster 

 cell in a tin box and supplied with food in the form of frag- 

 ments of decaying wood from an oak stump in which the writer 

 had failed to find evidences of this species. In an hour the 

 larvae had all disappeared but later one was found dead with 

 the fungus disease (?) that causes these larvae to turn pink, 

 die, and then throw out fine radiating white filaments for 

 nearly a millimeter in all directions. Some others probably 

 died early. After two months some of the chips were broken 

 up and larvae of about 3 mm. in length were found. Luckily, 

 on July 1st, almost five months after starting the breeding cell 

 the contents were carefully examined. Sixteen of the original 

 twenty-one were found in the following conditions: 



Seven were full-fed larvae, in two of which the paedogenetic 



form could be seen ready to moult. 

 Four were in the paedogenetic stage but did not display 



embryos within body and subsequently died (apparently 



barren). 

 Tw r o were represented only by the cells in which the 



