8 Annals Entomological Society of America [Mole 'V, 
* 
firm supports. The orb-weavers stretch their webs in mid-air 
between distant supports. Webs so exposed must be replaced 
frequently in part at least. It is important therefore that there 
should be an economical use of the dry silk. 
It is only the orbicular part of the web that is renewed at 
frequent intervals. The outer foundation lines are a permanent 
investment and are carefully saved. 
In the two families of orb-weavers the same type of founda- 
tion has been attained. The web of Uloborus, so far as its 
foundation is concerned, closely resembles an argiopid web; but 
the nature of the viscid silk is very different. The Uloborids 
‘possess a cribellum and calamistrum and spin a true hackled 
band. This, however, is of a comparatively generalized type, 
the warp consisting of only two straight, elastic threads. The 
viscid silk is evidently less fluid than that of other families and 
consequently retains the form given it by the combing strokes 
of the calamistrum, that is a regular series of overlapping lobes. 
A remarkable variation of the uloborid type of web is the 
triangular web of /Hyptiotes, which consists of only a sector 
of a web. 
Passing to the webs of the Argiopide, we find in the different 
subfamilies striking variations in the details of the structure of 
the orb, only a few of which can be mentioned here. But first 
let us examine the viscid thread. 
In these webs where is found the most highly specialized 
frame-work of dry silk, there is found the simplest type of viscid 
thread. This consists of a double thread of elastic silk, upon 
which is poured the viscid silk. This viscid silk, is fluid, and 
almost immediately the surface tension of this fluid causes it to 
collect in drops, which are distributed along the elastic support 
in a very regular manner. 
The simplest known orb-web is that of Theridiosoma, one 
of the Argiopide. This has few radii and no hub; it is used, 
however, in a peculiar way, described long ago by Dr. McCook. 
Before examining more perfect argiopid webs, I wish to 
review briefly the steps in the building of an orb-web. These 
are so well-known that it is only necessary to enumerate them 
without a detailed description. First the outer frame-work is 
made; this is a permanent part of the web. In the open space 
surrounded by this frame-work the radii are then stretched. 
Upon the radii in the vicinity of the point where they converge 
