86 PROCEEDINGS OF THE ACADEMY OF [Jan.. 



pressed them and also bent her head to the floor (copulation attitude), 

 but he discontinued courtship, and she moved away. At 2.42 he 

 courted again, and she again depressed her head to the ground (her 

 fourth pair of legs stretched out behind her, her other legs flexed to 

 her sides) at a distance of half an inch from him ; he then faced her for 

 two minutes without courting, but he recommenced his courting at 2.44, 

 and at 2.45 walked upon and embraced her, in the attitude previously 

 described. The left palpus was then applied to the epigynum 6 times 

 from 2.45 to 2.46, then the right palpus (on the right side of her body) 

 18 times from 2.46 to 2.52^. On each application the palpus was 

 first rubbed against her epigynum, and when inserted a large clear 

 sac, an evaginated portion of the palpus, swelled up and slowly (in 

 20 to 25 seconds) collapsed ; after its collapse he withdrew the palpus 

 and rubbed it against his chelicera for 2 or 3 seconds, then applied 

 it again to the epigynum. At 2.52^ the female rose quickly, shook 

 him off, and stood in hostile attitude, repelling several further attempts 

 to copulate. No oviposition intervened between these two acts of 

 copulation. 



Cocooning. — This process was observed several times. In captivity 

 it takes place at night, generally in the hours between 9.00 P.M. and 

 midnight; though the earliest preliminary spinning frequently begins 

 some 24 or more hours before. The following case is a good illustra- 

 tion of the method as observed in glass cages without earth : 



9 No. 54, on May 19, at 9.20 P.M., was first seen to be spinning, 

 fastening threads (inclined at an angle of about 45°) from one vertical 

 wall to the floor, and from the opposite vertical wall to the floor; at 

 each wall was then a line of such parallel threads about a quarter of 

 an inch long. This was a rather slow process, she taking often several 

 seconds to attach each thread, and sometimes pausing to rest for a 

 quarter of a minute. At 9.50 she started spinning the base of the 

 cocoon on the scaffolding of lines placed against the right wall, not 

 again returning to that on the left wall. The circular disk constitu- 

 ting the "base" of the cocoon was made by spinning upon a circular 

 area of the scaffolding, at about the middle level of the latter; it was 

 made discoidal in form by the female pressing the tips of her palpi 

 upon one edge of the area, spinning upon the opposite edge, then 

 rotating her body on its middle point as an axis. This was first done 

 very slowly, but at 9.59 she began spinning more regularl}^, brush- 

 ing her spinnerets (without elevating them) back and forth over 

 the area, at the same time rotating her body, so that gradually a 

 well-marked white disk was laid down upon the scaffolding, its 



