Page:Harvesting ants and trap-door spiders. Notes and observations on their habits and dwellings (IA harvestingantstr00mogg).pdf/181

 such a manner that even when the outer earthy part has become cracked, or been torn away by the action of the rain, they remain firm and fit to conceal their inhabitant. I have often found the tubes of web thus left exposed, as they are represented in Plate I., Fig. 4, situated in the cement of a wall, and among Lycopodium denticulatum, Adiantum Capillus-Veneris, Marchantia polymorpha, and other small plants. And it seems that the animal, perceiving the nature of the soil, takes care to reinforce the silken case, so much the more as she finds the earth less firm, and vice versâ. So that in burrows excavated in solid ground, with the exception of a little space close to the aperture, the nest is merely smoothed and daubed; while sometimes the spider constructs a tube so strong that it supports itself even when deprived of all the earth, the animal having had the foresight to attach it along the course of the clefts of the rock, or to the cement of the pieces of tufa in the wall, as represented in Plate I. They have often also a double aperture, and the upper portions of the burrows converging, meet and anastomose at about two inches distance. The aperture is closed by a little door or valve (a), which, having its hinge in the upper part and a little on one side, falls by its own weight, and fits itself exactly to the opening. The outer surface of the wicket is covered with earth, cemented by the glue of the spider, so that it is rendered imperceptible to common eyes, and the industrious little creature takes care to leave around the aperture a kind of rim, to which the door fitting closely, leaves no passage for any animal, nor does it show its edges. At the bottom of its tube the creature keeps her numerous offspring, and always stands herself as sentinel at the door, holding the wicket raised by means of the four anterior feet, and the palpi, curved extremities of which she inserts between the rim of the tube and of the door, as represented in a' f. Sometimes, however, they do not appear, but she leaves only the chink for observation, as one sees in a of the same figure. Fig. 2, at c, represents the aperture of an abandoned burrow, and at d the raised door of another burrow, with its almost funnel-shaped aperture. That which Sauvage, Olivier, and Latreille relate of her is not true—namely, that she remains at the bottom of