them to a cylinder, the two colors were wound separately from the living spider. Thus far, however, I have failed to ascertain their distinction with this species. So my figure and description may not be correct in assigning to the anterior pair the duty of supplying the interradial lines. However, it seems probable that the process is as follows: One or more pairs of the spinners are first pressed together and then separated. This draws out the silk as a band connecting their tips. By keeping them apart, and repeatedly carrying the calamistrum backward across their tips, the lines from each of the two mammulæ in one pair are kept separate until thoroughly dried. When the line is completed and drawn taut they remain distinct, but very near together.
We are now ready to observe the way in which the spider employs the organs above described. Let us suppose that the framework of the net is completed, and that the first or longest interradial line (Fig. 8, I' ) has also been made. Instead of beginning the second interradial at S'''', she begins at 4; and instead of climbing up the interradial or the strong and convenient base-line (B B' ), she runs to a point (2) near the apex, crosses the two intermediate radii, and passes along the upper radius to the attachment of the first interradial (S' ). On reaching this, she turns and moves for about her own length toward the apex. Contrary to the usual habit of spiders, during this roundabout passage from 1 to 4 she spins no thread. She now spreads her spinners a little, and presses them upon the radius, keeping them so while she advances again about her own length. This forms the attachment of the second interradial. The spider then lets her abdomen fall somewhat, supporting her body and advancing upon the line by means of her first, second, and third pairs of legs. The fourth pair are applied together to the spinners with great rapidity, at least five times in a second or three hundred times in a minute, and in so doing they draw out a double line.
The spider moves slowly along the radius until she reaches a point (5) where she can step across to the next radius (R''). While so doing, she ceases to draw out the double line, and carefully keeps it from
- From a notice in the American Naturalist for February, 1875 (page 125), it appears that Mr. A. J. M. Underbill has lately published, in Science Gossip, some observations upon the employment of the different pairs of spinners. He assigns to the third (middle?) pair the production of a line which is either viscid or curled.
- I must here admit an error in a previously-published account of the net ("Proceedings of the American Association for the Advancement of Science," 1873, pages 264-274). The interradial lines were there described as viscid. The fact is, that I had never thought it necessary to examine them under the microscope, since the interradial lines of all the Epeiridiæ are viscid; that is, consist of a slender axis enveloped by a viscid coating, which, soon after the net is completed, runs spontaneously into minute globules. Finding that the interradial lines of the "triangle-spider" were elastic, and that they readily adhered to the prey, or to any other body, I not unnaturally, but most unscientifically, drew the inference that with this spider the lines were likewise viscid. During the summer of 1874, while examining the manner of attachment of these lines to the radii, I saw that the interradial lines were neither viscid, like those of the Epeiridiæ, nor provided