No sooner had I gently touched the door with the point of a penknife than it was drawn slowly downwards, with a movement which reminded me of the tightening of a limpet on a sea-rock, so that the crown which at first projected a little way above, finally lay a little below the surface of the soil. I then contrived to raise the door very gradually, despite the strenuous efforts of the occupant, till at length I was just able to see into the nest, and to distinguish the spider holding on to the door with all her might, lying back downwards, with her fangs and all her claws driven into the silk lining of the under surface of the door. The body of the spider was placed across, and filled up, the tube, the head being away from the hinge, and she obtained an additional purchase in this way by blocking up the entrance. I did not force the spider to release her hold, but, by a rapid stroke with a long-bladed knife, cut out the upper part of the tube with the surrounding mass of soil, and thus secured the trap-door and its owner. This specimen is represented at fig. C, Plate VIII., where the pin-point holes made by the claws may be seen in pairs round the whole circumference of the flatter portion of the lower surface of the door except on the side next to the hinge. Whenever a spider resists in this way she must make these holes, but I have very rarely seen them in other nests; this may perhaps be accounted for by their having been effaced by the action of moisture which would stretch the silk. However this may be, this specimen showed the claw marks quite distinctly on my return to England after the lapse of several weeks. Much has been written about these marks, which are frequently spoken of as holes purposely made in the silk in order to give the spider a better purchase. It has also been stated that two holes may be seen in the silk of the tube near the mouth on the side away from the hinge, but these I have never been able to find. The door of nest A in Plate VIII. is rather abnormal, as it is made up of two doors, the smaller one being spun into the top of the one now in use. This is, I believe, an abnormal and rather clumsy example of the ordinary way of enlarging the nest, but of this we shall see more when we come to speak of the construction and repairing of these nests generally. Fig. B in this plate represents a moss-covered sod pierced by the tube of a nest, the door of which is entirely concealed from view, and only discovered when opened as at B 1.