ng the line thus laid down—still spinning—to the starting point, thus doubling the strength of the cable, and indeed a large spider will often repeat this operation several times. Now the upper boundary of the future web is secured. It is next necessary to find points of attachment for the lower boundary, and the spider either drops or climbs down—always carrying a line—from one of the ends of the upper line till it reaches a spot suitable for its purpose, and the previous performance is repeated. If there is any difficulty about a fourth attachment it is always open to the spider to climb back along the two lines already laid down, and by carrying a loose line with it, to secure at all events a triangular frame-work. This frame-work, whether trapezoid or triangular, will be reinforced several times and made thoroughly trustworthy before the work of making the actual snare is proceeded with.

Now the foregoing operation is obviously perfectly simple in certain cases, as, for instance when a spider has chosen lattice work, or the mouth of an empty barrel as its “pitch,” but snares may easily be found[Pg 17] in situations where such a mode of procedure seems impossible. In a pine forest, for example, one may see huge webs stretched at a great height from the ground between boles ten feet apart; or one may find such a snare spread across a stream at a spot where the trees on either side do not intermingle their boughs. How in such cases does the spider accomplish its purpose?

There is little doubt that, wherever practicable, the spider walks round, sometimes crawling quite an astounding distance, but that it can at need, resort to another method, is easily proved by a very simple experiment in the house. Fill any vessel—a basin or a bath—with water and arrange an upright post in the middle, placing a spider