Spiders J roiii (lie HhIkuiki IsJtmds. 331 



Gasteracantha Kochii, A. G. Butlor. (PI. VII. fig. 10.) 



Gasteracattt/ia Kochiifhuth'T. Type in Coll. R. M. I/af>. PtitL Nom. 

 uov. for he.iacantha, C. K., 'I'lnns. Knt. Soc. Loud. 1873, p. KiU. 

 no. 60. 



This is obviously identical with Q. hexacantha^ C. K. (non 

 hc.racantlia, Fabr.). 



There are many examples of this form from various parts 

 of Anicriea in the Mu.^cum collection. In the Keyserling 

 collection from Taguava, Mexico, Bogota, and Kio Grande do 

 Sul. Also from Upj)er Surinam, San Josd (Costa Rica), 

 Ecuador {Rosenberg). Colombia, Lower Amazons, Par^, 

 collected by the author ; I have taken it also in the forest near 

 Santarem. 



It may be recognized by the shape of the abdomen being 

 rectangular, much broader than long ; the anterior and lateral 

 (or median) spines are very short and of equal length, the 

 j)Osterior spines larger. 



The examples from Ecuador are very stout and convex 

 below, others from other regions are much more compressed. 

 They vary also very much in coloration. Tliese great differ- 

 ences, however, in examples of the same species will be quite 

 familiar to those who have ever collected a large series of any 

 species of the genus in their natural habitat. I have often 

 myself, in picking them with the fingers out of their webs, 

 felt some of them nearly flat and others quite convex, and been 

 surprised that they were not different species. 



Gasteracantha insitlana, Thor. 

 Gasteracantha iiisulana, Thor. — Galapagos Islands. 



If the example from this locality in the British Museum 

 collection be rightly identified, this form is probably identical 

 with Kochii, Butler, but one cainiot be sure without a laro-er 

 series of specimens. 



It is possible that G. sexserrata (Wlk.), Ins. Apt. ii. p. 157 

 Cayenne; G. quinque-serrata (Wlk.), ii. p. 157, Guyana- 

 G. triserrata (Wlk.), ii. p. 158; and G. Servillii (Wlk.), ii! 

 p. \i)\), Brazil, are all G. Kochii^ Butler; and if so, the last 

 name will go as a synonym of sexserrata. Possibly G. mam- 

 viosa, C. K., xi. p. 57, Brazil, is cancrifoDnis^ Linn. : but 

 what G. lata (Wlk.), ii. p. 1G5, from Guadaluupe, may be, 

 is very doubtful. 



