CONSTRUCTION OF AN ORBWEB. 77 



the spider uses her limbs as a measure to regulate the distances of her 

 radii or "wheel spokes," and the circular meshes interweaved into tliem.^ 

 The above method of alternate apposition shows that tlie dis- 

 Legs no tances between radii, at least, are not determined by any such men- 

 ^^^ ° suration, which of course would only apply on the suj^position that 

 the lines were spun consecutively. Moreover, having frec[uently 

 measured the distances between the radii at their circumference or attach- 

 ment to the foundation lines, I have found that there is often great irreg- 

 ularity therein ; the interspaces sometimes vary in the proportion of three 

 to one on the same orb. The fact that the number of radii is not con- 

 stant in the successive webs of any individual spider, but varies from day 

 to day (although within a narrow limit), is also against this hypothesis. 



It has already been intimated that the notched zone serves an important 

 end in bracing the radii while they are yet in outline. This appears dis- 

 tinctly while observing the above described behavior. The first 

 Forming padii that are inserted bend and swaj^ under the weight of the 

 „ spider, which, as she clambers over them, suggests the idea of a 



carpenter engaged upon a scaffolding in its first crude state. 

 In some cases the aranead stops at the guidon, after having placed the first 

 few radii, and swings her spinnerets around their bases as though to 

 strengthen them. This act maybe repeated several times; and in fact the 

 spider -a'henever she comes to the centre is apt to make two or three of 

 these gyrations. However, when all the radii are inserted she proceeds to 

 complete the notched zone, laying in the spirals thereof from the centre 

 outwardly. These vary in number from four or five to ten or even more. 

 This variation holds in webs of the same species ; for example, in seven webs 

 of Argyroepeira hortorum, the spirals in the notched zone numbered, suc- 

 cessively, 8, 10, 10, 10, 6, 8, 5. I have counted as many as twenty in the 

 web of Epeira gibberosa. In the group of Orbweavers, of which Epeira 

 insularis^ is the type, the number of notched spirals is quite persistently less, 

 commonly five or six. The width of the notched zone is about equal to, or 

 •a little greater than that of the free zone, and about twice that of the hub. 

 In at least one example noted (an orb of Tetragnatha extensa) three wide 

 notched spirals were first spun before the others were laid in. These ap- 

 peared to be the analogue of the spiral scaffold, referred to in the next 

 section ; but tliis is not the ordinary rule of construction. 



Mr. Romanes^ quotes Dr. Leach as giving, on the authority of Sir J. 

 Banks, a case of a web-spinning spider which had lost five of its legs, 



' Insect Architecture, page 313. 



- This familiar and beautiful spider Dr. Thorell declares quite identical with the Epeira 

 marmorea of Europe. I have no specimens of E. marmorea with which to compare, Ijut 

 have no doubt of Dr. Thorell's identification. However, I retain Hentz's name in the two 

 volumes on Habits, and will endeavor to adjust the nomenclature of this and other species 

 in the final volume. 



^ Mental Evolution in Animals, page 209. 



