The foundation lines which may thus have given the spider great trouble to secure, are of extreme importance to it, and may serve for several snares in succession. There is little hesitation or delay about the subsequent operations. The spokes of the wheel are readily formed by carrying lines across to opposite points of the frame-work and uniting them where they intersect. They are laid down in no special order, but more or less alternation is generally noticeable—apparently for the purpose of keeping the tension equally balanced—and the spider will occasionally desist in order to go and brace up the frame-work with additional stays, which generally have the effect of converting it to a polygon. Before long the requisite number of fairly equidistant “spokes” or radii are visible, and then the spider, starting from the centre, rapidly spins a spiral thread consisting of a few coils only, to the circumference, stepping from spoke to spoke. This is only a temporary scaffolding and will not be suffered to remain in the completed snare. If the structure is touched at this stage of the operations it does not adhere to the finger; the viscid spiral remains to be laid down. Though it does not hesitate for a moment, the spider now works with a peculiar deliberation, but the operation will be much better understood by actual observation than by any amount of description, and we shall only recommend the reader to note that the new spiral is exceedingly elastic and that at the moment of its attachment to a spoke it is stretched and let go like the string of a bow. The spider seems carefully to avoid treading on it as it proceeds, utilising the non-viscid spiral scaffolding already described. [Illustration: Fig. 3. Stretching the viscid spiral.] A little attention to the centre of the wheel, and the snare is complete. Some species of _Epeira_ entirely remove the centre, leaving a circular empty space, while others fill it with an irregular network of threads. How does the garden spider avoid getting caught in its own web? We have shown that there are many lines which are not viscid, and no doubt these are utilised as far as possible, but it can hardly happen that the spider never touches adhesive portions of the web with legs or body. Possibly some explanation is furnished by an ingenious experiment which Fabre performed. He found that a glass rod, lightly smeared with oil, did not adhere to the viscid spiral; neither did a leg freshly taken from a garden-spider unless allowed to remain in contact for a considerable time. When, however, this leg had been washed with bisulphide of carbon, which dissolves any kind of oily substance, it adhered at once. It would seem likely, therefore, that the legs and body of the spider itself are protected by some oily exudation from any danger of adherence to its own lines. CHAPTER IV MENTAL POWERS OF SPIDERS