THE GENUS CALOSOMA. 103 



The average number of caterpillars eaten by each larva in 1909 

 was 20, and in 1910, 13. Sixth-stao;e caterpillars only were supplied 

 in the latter series, while several fifth-stage caterpillars were offered 

 in the former. Twenty large caterpillars (the same or similar to 

 Porthetria dwpar, or Malacosoma americana) is about a fair average 

 for each larva to consume. 



The larvae in Table 24 were active and fed over an average period 

 of 21 days after hatching, and then entered the ground for pupation. 

 Larva? of C. scrutator are active the same number of days but 

 have a somewhat greater capacity for food. 



Five larvae were fed from hatching to maturity in 1911 on pupae 

 of J/, americana from wliich the cocoons had been removed. From 

 6 to 10 pupae were devom'ed by each larva, or an average of 9. Larvae 

 of this species did not attack and devour as many pupae as caterpil- 

 lars, and the same is true of other species of Calosoma. When a 

 Calosoma larva attacks and devours a pupa, it usually feeds until 

 notliing remains but the empty case, but when a caterpillar is at- 

 tacked part of the fluids and body contents are lost in the struggle 

 of the caterpillar to free itself from the Calosoma. 



HABITS OF LARV^. 



May 29, 1911, an experiment was begun to test the climbing of 

 these larvae. (PI. III.) Six newly hatched larvae were placed in- 

 side the tin circle and caterpillars of Malacosoma americana were lib- 

 erated upon the tree and on the ground. Later in the day. two of the 

 larvae were noted feeding upon caterpillars at the base of the tree. 

 May 30 the larvae were again noted feeding upon caterpillars on the 

 ground, on which date five were found buried in the earth. These 

 first-stage larvae remained in the circle until about June 4 before 

 transfomiing to the second stage and were deprived of food some of 

 the time until they were on the verge of starvation. Later on, July 

 17 and 21, other first-stage larvae were added and kept mider ob- 

 servation for a few days. They were placed upon the tree fre- 

 quently, but each time fell to the ground almost immediately. Sec- 

 ond-stage larvae were tested but did not chmb except on one occa- 

 sion when one was seen crawhng up and down the trunk about a foot 

 from the ground. The others died of starvation and did not attempt 

 to climb so far as could be observed, although they were under 

 observation for several days. One third-stage larva was left in the 

 experiment for a few days and it made no attempt to chmb for food 

 but remained on and in the earth. 



In aU cases starvation resulted unless food was placed on the ground. 

 These facts show conclusively that this species is teiTestrial in habit 

 during the larval stages and the same is usually true of the adults. 



Pupae of this species averaged as foUows: The females, 17.5 mm. 

 long and 7.7 mm. wide; the males, 1 .5.7 mm. long and 6.8 mm. wide. 



close observations were madf^ upon a series of larvtie iod in jars in 

 1911 to determine the actual tim(^ required to pass the pupal stage. 

 It wtus found that 1 mah^ pupa matured in 11 days, wliile th(^ females 

 matured in and 10. An averagj; of 10 days was required for each, 

 including both sex(« in the; sc'ries. The adults from the pupae in 

 question issiir'd from July 4 to 10. 



