GENERAL COCOOKING HABITS OP SPIDERS. 137 



webs of Lineweavers; is attached above to a strong thread, and is stayed 

 and balanced by various guy lines along the entire length. On opening 

 the several cocoons of one of these strings I found (in one of twelve co- 

 coons) the first seven contained only the first moults or shells of tlie escai)cd 

 spiders ; the next three, young spiders in successive degrees of advanced 

 growtli ; and the last two, eggs alone. 



The exterior case of the cocoons is a light straw color or creamy white. 

 It is made of two saucer shaped pieces well woven together at the edges, 

 and is about tliree-eighths of an inch in diameter. Fig. 166 gives a side 

 view of a cocoon string, showing the waj' in which the cocoons overlap one 

 another. 



Dictyna usually makes several cocoons, small flattened globes of pure 

 white, about one-eighth inch diameter, which are placed within the snare, 

 usually grouped near one of circular doors on which the web 

 lines converge. (See Vol. I., page 349.) When she spins her 

 web along a brick wall or like surface, the cocoons are fastened to the wall, 

 arranged along the angle or clustered together loosely. When the spider 

 makes her snare within a leaf, as she frequently does, the cocoons are 

 placed upon the leaf, protected, of course, by the enclosing cross lines. 

 The mother is found near her cocoons, though apparently not exercising 

 any special vigil upon them. She simply la3's her eggs in the position 

 most convenient to herself. The edges of the leaf are sometimes drawn 

 well together (Fig. 167) and sewed in the prevailing aranead style; but 

 more frequently the edges of the leaves are simply bent over by silken 

 lashings as in Fig. 168. This cut is drawn from a sketch made on the 

 grounds of the Smithsonian Institution, Washington. 



III. 



Of the typical cocoons of the Territolaria> we may speak with some 

 positiveness ; but the number of species whose cocoons are known is small. 

 „ , However, it is higldv i>robable that the variety of form and 



telarife naethod of suspension and care is not great, and we may per- 

 haps conclude that we possess a good knowledge of the general 

 cocooning habit of the tribe. 



Mr. Enock determined the jjosition in which the mother Atypus piceus 

 spins her cocoon. In a tube ten inches long and from a half to five- 

 eighths inch in diameter he found that about six and a half inches below 

 the surface the tunnel widened into a sort of pouch. On opening this he 

 saw the mother's cocoon suspended in a beautiful hammock of silk one 

 inch long, the flat ends of which were about three-sixteenths of an inch 

 wide, and were attached to the top and bottom of the pouch. ^ 



This description entirely corresponds with that previously recorded by 



^ Life History of Atypus piceus, Trans. Ent. Soc. Lend., 1885, page 394. 



