iiKsrKHiDi: i;i'AK(;vi{i:rs invKUS. 1407 



more oflcii ijiiii iIk- trcts ;iiiil ctiiistnict in some secure place a cocoon of 

 leaves or fragments of .stiiMilc, tlif iiittriitr of wliicli is lined with ii loose 

 welj of silk." Sjn'ciniiMis rcarid in i'a[itivil_v made their cocoons by draw- 

 ing togetiier lea\ es of locust so as to form a smootii oval cavity within ; 

 tiie leaves were attaciied by short threads so as to present no opening large 

 enouifh to look witliiii : tiie wliolc interior was lined with a thin web of 

 silk .similar to tiic carpet wiiich tiic larva of Euphoeadcs troilus stretches 

 acrot^s tiie leaf of a sassafras tree. Witiiin the cavity the caterpillar sjiun 

 txvn Y-shaiK(i threads wlicicon liic clMysaiis was eventually suspended ; 

 that at the hinder extremity iiad tiie stem about (!..') mm. long, the forks each 

 4..'> nun. long and diverging at an angle of about 75'"\ their tips being (J mm. 

 apart ; at the junction of the tiireads the booklets of the tail of the chrysalis 

 were plunged into the silk ; the other Y struck the mesothorax of the chrysalis, 

 which rested with its i)ack in the looj) ; the angle of the forks was slightly 

 less but the forks themselves were longer, l)eing 11-12 mm. in length and 

 \2 mm. apart at their tips. Measurement of the shrouds of another cocoon 

 formed in leaves gave the following result : the stem of the thoracic Y was 

 5 mm. long, the two limbs !• mm. and 11.5 mm. ; the stem of the anal Y 

 was 8. .5 mm. long ; the two limbs 3. .5 and 4.5 mm. 



Mr. .Vngus found large numbers of the cocoons of tiiis insect underneath 

 boards, shingles and rubbish lying on and aliout a lieap of saw dust beneath 

 some large locust trees, and in tiie place of the particles of earth, straw, 

 leaves, etc., with which cocoons formed on the ground are usually covered, 

 they were coated with saw dust. "The greater portion of them," says Mr. 

 Angus, "were found between shingles and the boards to which they were 

 nailed : we cut some shingle roofs up for tire wood and I was astonished at 

 the quantity of pupae which were found ; some of the crevices in which they 

 had hidden were scarcely larger than the chrysalis and in these instances, 

 shingles above and boards'beneath served for two sides of the cocoon. They 

 are frequently found under boards that are left lying flat on the ground and 

 in this case the cocoon is generally very slim — sometimes barely enough to 

 secure the chrysalis to the board." Mr. Angus has always found them in 

 "some secluded crevice, either in a stone wall, board fence, under boards 

 or such situation ; they are generally found in a iiorizontal position, but 

 this probably depends somewhat on the position of the crevice." 



These saw-dust cocoons (82: 9) form a rather regular ovate mass about 

 35 mm. long and from 15 to 18 mm. broad ; the cavity within is barely 

 more than large enough to contain the chrysalis and is lined with a thin 

 tissue of brownish silk, delicate as a sj)i(ler's wef) but very tough and 

 strong ; a strand of dirty whitisii silk, parted above, is stretched up and 

 down one extremity of the cavity where it is 8 mm. in diameter, and in 

 this the tail of the chrysalis is ])lunged ; the Y-shaped thread for the 

 support of the thorax has a stem 4 nun. long and less than half as thick as 



