8 DEPARTMENT BULLETIN 1267, XJ. S. DEPT. OF AGRICULTURE 



occurs in' nature. The experience of the writers, however, both in 

 the field and in experiments, indicates that an excess of moisture 

 is almost as unfavorable for the eggs as a deficiency. The places 

 in which the species normally occurs and in which it breeds most 

 abundantly are so situated with reference to local drainage condi- 

 tions that, although the ground retains a constant supply of avail- 

 able moisture, it is never saturated with water for any considerable 

 period of time during summer. It is doubtless these moisture re- 

 quirements which underlie the marked tendency of the species to 

 congregate in the lowlands bordering marshes and drainways and 

 to be limited on the higher groimd to local sunken areas in which 

 the soil is rather slowly drained. 



INCUBATION 



Experiments on the duration of the period of incubation were 

 conducted at Charlottesville. Owing to conditions attending the 

 work, it was rarely possible to determine the precise time at which 

 an egg had been deposited, and, for this reason, the results obtained 

 are at best but approximations. On one occasion (July 26) a female 

 was found in the act of depositing an egg. On August 9 this egg 

 hatched, giving an incubation period of 14 days. 



There may be considerable variation in the time required for in- 

 cubation. This may be accounted for by variations of temperatures, 

 high temperatures accelerating and low ones retarding development. 

 As a rule, the period of incubation under favorable midsummer con- 

 ditions varies from two to three weeks. In the fall this time is 

 greatly extended, extremes of from 35 to 50 days being reached in 

 October and November. All eggs which had failed to hatch by 

 the middle of November were buried in their containers in the ground 

 and there kept over winter. In early April they were dug up and 

 examined, but in all instances the eggs were dead. 



HATCHING 



The chitinized larval jaws may be seen through the translucent 

 egg membrane (PI. II, D) several days before hatching takes place. 

 When hatching occurs, the egg membrane appears to collapse and 

 to split at a point close to the dorsal surface of the larva. In one 

 instance the rupture of the membrane occurred in the vicinity of the 

 thoracic region ; in another at about the level of the third abdominal 

 segment. It evidently results from the contortions of the inclosed 

 larva in an effort to free itself. After the membrane has split the 

 larva continues its efforts, bending and extending its body at fre- 

 quent intervals until it has finally managed to extricate itself, though 

 occasionally portions of the membrane may adhere to the larva for 

 a considerable time after hatching. In no instance did the larva? 

 make any attempt to devour the egg membrane. 



The time consumed in the process of hatching was in one instance 

 7 minutes, in another 35 minutes. Plate II, F, shows the newly 

 hatched larva and Figure 3 shows the relative size of the head and 

 body immediately after hatching. 



