April, '13] JONES: FALL ARMY WORM 235 



were kept supplied with fresh grass and the earth in the jar was kept 

 moist. Until the third molt the larvae fed only on one epidermis of 

 the grass, but after that upon the entire blade. 



With the exception of one larva, which for some reason did not 

 develop as quickly as the others, all the larvse entered the soil on May 

 21. Seven consecutive molts of the head were found in four of the 

 vials and six in the remaining four, the time elapsing between molts 

 ranging from two to four days, and the last molt occurring, on an aver- 

 age, five days before the larva entered the soil. Four of the seven 

 moths obtained from the larvse originally placed in the vials issued 

 twelve days after the larvse had entered the soil to pupate. Two re- 

 mained in the soil eleven days and one thirteen days. On June 3, a 

 day after the majority of these issued, two adults came from the soil 

 in the battery jar. No other specimens were obtained from the jar. 



Summarizing, we have, under the breeding-house conditions, and 

 the climatic conditions prevailing at the time, twenty- one days for 

 the larval period and an average of twelve days for the prepupul and 

 pupal stages combined. 



From observations made on this insect the writer believes that the 

 time necessary for the completion of its life cycle in Porto Rico varies 

 considerably, this variation depending upon the temperature and 

 moisture conditions. This difference might, perhaps, not be expected, 

 since the temperature conditions, especially on the north coast, are 

 quite equable. Four adults, from larvse that were collected when 

 very small and cared for in the same way as were those on which the 

 above life history notes were made, issued from the soil in April, six- 

 teen days after the larvse had gone below the surface. This is a period 

 of four days longer duration than was necessary for the prepupal and 

 pupal stages during the latter part of May of the same year. Other 

 notes show a difference in the length of time consumed in the various 

 stages of development, but unfortunately we have no thermographic 

 or hygrographic records to accompany them. 



The breeding records of the station show that adults, from eggs or 

 larvae collected in the field, have issued at various times during the 

 months of January, February, March, April, July, October and De- 

 cember. Further observation throughout the year will probably show 

 that the larvse are to be found throughout the twelve months. 



Enemies. Three species of Tachinid flies have been bred from 

 Laphygma frugiperda at the station at Rio Piedras. These have been 

 determined by Mr. W. R. Walton, of the U. S. Bureau of Entomology, 

 as Frontina archippivora Will., Gonia crassicornis Fabr., and Archytas 

 piliventris V. d. W. 



Adults of a Carabid beetle, Calosoma aliernans Fabr. (determined by 



