398 



AMERICAN SPIDERS AND THEIR SPINNINGWORK. 



belonged to the Orb weavers. (See Fig. 336.) The natural size of the insect 



is given within the circle at the side of the drawing. The insect may not 



have been directly parasitic on the spider's eggs, but on the larvre of 



Pezomachus or some other spider parasite. 



The above examples are sufficiently representative, and they indicate the 



mode in which the Hymenoptera wage war upon the Aranese. As to the 



extent of that war I may simply say that I find a large propor- 



Extent ^Jqj-^ q£ Epeiroid cocoons infested by parasites. In going among 



... the old cocoons in the spring it is often more usual to find them 



asitisni. . ^ " 



occupied by cocoons of Pezomachus than full of healthy eggs. 

 Of five large egg nests of Epeira cinerea now before me, not one is free 

 from parasitic cocoons. Often the spider bi'ood will be in part preserved; 

 frequently the parasites have full possession ; and, again, the traces of 

 a full and healthy brood are shown by the shells or first moults within 



the central bag. Of course the ratio 

 of destruction varies at different times 

 and places. 



There appears to be no special 

 proclivity on the part of parasitic hy- 

 menopters to confine their operations 

 to any species of spider. They apj^ear 

 to choose their host indiscriminately 

 from among the cocoons in which the 

 eggs are swathed. Pezomachus gra- 

 cilis will attack the hard, stiff, and 

 compact cocoon case of Argiope cophinaria, or will choose as a host the 

 eggs of a species of Epeira, or indeed of other tribes. 



This point needs further investigation, and would be a matter of some 

 importance to solve with absolute certainty. It would greatly add to our 

 resi)ect for the discriminating powers of these strange insects if 

 Parasite ^^,g ^.gj.g ^^ establish the fact that they can select a spider's eggs 

 ences "sven when they are hidden away under jjrotecting cases so widely 



differing in ajipearance, construction, and location. The cjuestion 

 would then rise, in what manner do the ichneumons determine the pres- 

 ence of the eggs ? Do they watch the spiders themselves ? Are they able 

 to detect the presence of sjjiders' eggs through the enclosing enswathment 

 by some sense so delicate that it cannot be appreciated by human beings ? 

 The wingless condition of the females doubtless greatly favors them in their 

 search for objects hidden away as spider's cocoons commonly are. 



Pig. 337. Parasitized cocoon of Epeira cinerea, 

 opened to show the infesting ichneumon cocoons. 



VIII. 



These are not all the hostile agents arrayed against the embryo life of 

 si^iders. Parasitic plants as well as parasitic insects assail them, for not 



