MATKKKAL INDI'STJIY : COCOONS OF OKl'.WKA VKKiS. 



103 



Fi<i. 92. Section of Cau- 

 data's snare, showing 

 manner of suspend- 

 ing cocoons. (Nat- 

 ural size.) 



lapping one aiiother like tiles. Sometimes nodules of flossy silk, or of silk 

 mixed with the debris of cai)ture(l and devoured insects, are irregularly 

 interposed between the cocoons. Tliis is, indeed, a fixed 

 and most interesting habit of the species, which will be 

 described in a succeeding chapter. 



During a temporary stay in Florida, April, 1886, 1 

 found nested u[)on the porch of Dr. Wittfeld's place, 

 Fairjdand, Merrit's Island, on the Indian River a little 

 way below Ilockledge, a new spider, which I named Cyr- 

 tophora bifurca. Its snare resembles that of Cyclosa 

 caudata. It also resembles that spider in the manner of 

 hanging its cocoon string in the vertical axis of its orb 

 just above the hub. The character of the cocoon, how- 

 ever, differs entirely from that of Caudata. It is, in 

 shape, a somewhat irregular octagon, and is of a dark 

 green color. I have found as many as fourteen cocoons 

 in one string, overlapping one another in the manner of 

 cocoons of the Labyrinth spider, and which may also be 

 seen at times with the cocoons of Caudata, although for 



the most part, the latter are arranged at intervals 

 along the string. (See Figs. 96, 97.) 



The cocoon strings collected varied in the 

 number of cocoons attached thereto, probably ac- 

 (§J ^ it\ cording to the period of advancement in the proc- 



ess of ovipositing on the part of the mother. 

 Of the specimens collected one string contained 

 fourteen, another twelve, and another ten cocoons. 

 They are Ijound together, along one side, by con- 

 tinuous series of thick white thread.s, which ex- 

 tend from the top to the bottom of the string. 

 Each cocoon consists of two parts, which have 

 evidently been fastened together by a .selvage. 

 These parts present tlie appearance of twci dishes 

 placed together edge to edge. They are woven 

 of a soft, but rather tough, texture. A very .slight 

 tuft of flossy white silk is found inside, and witli- 

 in this the eggs are deposited. In one cocoon of 

 a string of thirteen, twenty-five minute dead spi- 

 ders were counted, which had passed their first 

 moult. In another cocoon, taken from a string 

 of five only, there were twenty-six. The num- 

 ber varies a good deal, however. The cocooning 

 period appears to extend into May; at least I have received from Miss 



Fi(^. 93. Fig. 95. Fig. 96. 



Fi<;. 93. Cocoon string of Caudata, 

 with sillc nodules interposed. (Nat- 

 ural size.) Fii;. 94 (upper). Cocoons. 

 (Natural size.) Fig. 95 (lower). En- 

 larged. Fig. %. Cocoon string of 

 Epeira bifurca. showing shape and 

 superposition. (Natural size.) 



Anna Wittfeld, as late as the middle of June, a string, in which were 



