PSYCHOLOGY — PORTER. 221 



With respect to the above parts of the web, counts have been made of 

 approximately 200 Argiope riparia ; of about the same number of Acrosoma 

 spiiica; some 40 Epcira hortoruui; about 100 of B. sclopetaria, and of a 

 smaller number of the triangle spiders, probably Hyptiotes cavatiis, H. trifo- 

 lium, B. insularis, and Argiope transversa. Work with the last three is still 

 being continued. Acrosoma spinea has additional interest for the reason that 

 a priori it should be less variable, having fewer accessories- Bpeira hortorum 

 is interesting from the fact that the web is as nearly in the horizontal plane as 

 the others are in the vertical, and it has a screen beneath. 



Since the counts on a single web, when all the five parts are included, may 

 be as many as 125, it is readily seen that not many can be made, as they 

 must be confined to the early part of the day. The webs are usually so torn 

 by the middle of the day that it is unsafe to attempt to count their parts 

 during the afternoon. 



2. The ontogenesis of the web is very interesting. The web of the young of 

 riparia and transversa is very different from that of the adult. 

 This change with age involves chiefly the omission of certain 

 parts. 



a. The zigzag or "winding stair" in webs of young riparia is curved 

 about the center of the web to the number, at times, of 5 or 6. 

 These are soon reduced to a single zigzag, and this is fairly 

 constant with riparia. With the closely allied species, Argiope 

 transversa, there is no zigzag at first. The lower half appears 

 some days before the upper. The entire structure is more 

 often omitted in the web of this species than in that of riparia. 

 In the web of Acrosoma spinea there is no lower half to the 

 zigzag. The upper half is probably a more constant structure 

 than with riparia. This structure is more variable in the webs 

 of riparia males than those of the females of the same species. 

 In webs of males of transversa it is seldom found. 



h. The side-screens show similar changes with age. The one on the 

 under side of the web appears first and is the more constant. 

 Transversa is less variable here, at least, when young. Both 

 screens are often absent, and, with the adult transversa, about 

 October i they are rarely present. Since the males approach 

 the web of the female by means of the lower screen, their defi- 

 niteness of approach could be observed in connection with the 

 above. If there is but a single male his approach is a fairly 

 definite one. 



c. The anchor line, or the thread from the center of the web on the 

 side opposite the spider to the upper side-screen or plant, is a 

 most variable structure. There may be at the same time as 

 rnany as six of these in the same web. It is probably suffi- 

 ciently constant for short periods to be considered a product of 

 instinct. 



On the points a, h, and c there are many hundreds of observations. All 

 observations possible have been made on those webs having a signal thread 



