His attention was at first arrested by the sight of "something hanging down which looked like the cocoon of some moth;" but, on closer examination, the silk case proved to be empty, and was continued as a tube into the ground to a depth of 9 inches, where he came upon the spider lying at the bottom. Further research revealed the existence of a number of these nests in the same locality, but the length of the different tubes varied much; they were usually about 9 inches long, but some were much longer, often baffling his attempts to follow them; the longest which he was able to secure entire measured 11 inches. All the nests were, however, alike in having a tubular silk lining, about 3/4 of an inch in diameter, a part of which protruded from the ground for about 2 inches, and was pendent, inflated, and covered with particles of sand, assimilating it to the surrounding surface; it was closed at the upper extremity, leaving no exit to the open air.
Mr. Brown took home some of these tubes in a collapsed state with the spider at the bottom. In one case, on opening the box in which the nest was placed, he perceived a movement throughout the tube, as if it were being inflated; this however soon subsided, but the following morning he was surprised to see that the whole tube was inflated, especially at the end which had lain exposed on the bank. He failed to find any aperture by which the spider could enter or leave her nest, and his captives, though passing backwards and forwards in their tubes, never came out at either end. He never saw flies or any fragments of insects in the nests; but, on drawing out one of the tubes, he observed a worm at the lower end, partially within it, partially outside, and he perceived that the spider had evidently been eating a considerable portion of its anterior extremity.
It will readily be seen that there are some discrepancies between the different accounts which have been given of the nests of _Atypus_ found in England and France,[127] and I think it quite probable that some at least of the nests described may really differ, and be the work of distinct species belonging to this genus. Mr. Brown describes his nests as having by far the greater part of their length under ground, while in those observed by M. Simon, as shown in my figure, Plate XIII. fig. A, the exposed portion of the tube equalled or exceeded the subterranean.