124 . AMERICAN SPIDERS AND THEIR SPINNINGWORK. 



iind of the orb within the triangular foundation Hues. The other details 

 are only approximately accurate. 



My notes do not show the complete form of the web, but Mrs. Eigen- 

 mann's observations abundantly establish the fact that the orb of Gaster- 

 acantha has the notched zone and free space as is the case with the webs 

 of Epeira, which it closely resembles. None of the California webs, how- 

 ever, had anything like the tufted decorations which I observed in the 

 Texas individual. Whether or not other examples in the same vicinity 

 exhibit the peculiarity which I have described I am not able to say.^ 



The interest in the problem here presented is much increased, although 



the problem itself is brought little nearer solution, by facts concerning the 



snares of this genus recorded by M. Vinson. ^ He observed the 



African game characteristic noticed by me in the Texas example, in the 



m i?x J webs of Gasteracantha bourbonica of certain African islands. 

 Tufted. 



This spider spnis a vertical web a metre or more m diameter. 

 The snare is often suspended across the path in forests. The threads are 

 different from those of other spiders; they show little cottony tufts (ren- 

 flements cottoneux) distributed at intervals, but quite nearly approached. 

 The spider hangs at the centre. She is active when she moves ; but when 

 one touches her she throws herself from her place, holding on by her 

 thread, by the aid of which she is able to replace herself upon her snare. 

 In another part of his book M. Vinson records the same observation 

 in this language : The Gasteracanthse of the Isle of Reunion introduce 

 into the variously stretched lines [foundation lines], in the midst of which 

 they establish their regular nets, a finishing-up (" confection ") altogether 

 special. As these lines are isolated, quite separated one from another, 

 they are differently wrought from those of the interior net, and show at 

 short intervals little cottony puffs (" renflements ") which cause them to 

 appear as though interrupted from point to point. These puffs give the 

 web, of necessity, very great strength and elasticity.^ 



The same author, however, introduces another observation upon this 

 species, which brings us face to face with the same curious diversity, if 



not divergence, in the habit which I have shown in the Gas- 

 Webs teracanthse of our country. M. Vinson declares that the Gas- 

 rn fL. teracanthte of Madagascar, wdiich are both larger and more 



numerous than those of lieunion, do not follow the custom of 

 decorating their webs with cottony tufts as do those of the last named is- 

 land, but spin their vertical snares in a manner altogether similar to those 

 of the common Epeira. 



^ As I was at tlio time intent upon tlie study of the natural history of the agricultural 

 ants, I was i-ouiiH-llcd, often at fjreat sacnfu'e of my feelin<j;s, t(» resist the attractions every- 

 where around nie to observe and seek out tlie habits of the spider fauna. 



^ Araneides de la Reunion, Maurice, et Madagascar, page 238. 



^ Up. cit., Introduction, page xvii. 



