closed, in a straight line over the middle of the hole, as shown in fig. 18, 

 Plate V. Every night (I observed one in my room for three months) 

 the spider would throw open its doors as shown in fig 17, Plate V and 

 in the morning close them. If I opened the door during the daytime 

 the spider would invariably come up and close it, but never immediately, 

 and never did it offer to catch the door and resist its opening. When 

 members of this genus build their nest under stones, as they sometimes 

 do, if the stone is elevated from the earth in one place the spider builds 

 the tube above ground to the under surface of the stone, and here makes 

 some sort of door, I have never been able to see just what, but probably 

 very much like the ones made in captivity. A longitudinal section of 

 one of these nests which I found under a stone is shown in fig. 1 1, Plate, 

 IV, the surface of the ground at a and the mouth of the tube at b. On 

 turning over the stone I saw what I thought was the cocoon of a moth, 

 from which the insect had escaped at the upper end. I picked it up, 

 and discovered the tube which I found about 12 inches deep, with the 

 spider 1 at the bottom. This nest, perhaps inappropriately, I have called 

 the hall door nest, because of the enlargement just within the entrance. 

 This I consider would be of advantage to the spider in managing the 

 door while catching an ant. The portion above ground was made of 

 loam and leaf mould cemented with viscid liquid and lined on the inside 

 with silk. So far as I have oberved, the members of this genus line only 

 that portion of the nest with silk which is liable to cave in near the up- 

 per portion of the tube. 2 



Nest of Myrmekiiphila foliata. 



The nest of this species is not constant in type, and shows wider va- 

 riation in different individuals than any of those thus far described; but 

 when taken in connection with the food habit of the species there does 

 not seem to be a very great departure from a common type. 



The first individual was collected March 13, '86 by myself. In 

 collecting specimens of ants and their root feeding "cows'' (Aphides) 

 which they were protecting through the winter, I found a trap-door on 

 turning over a stone. The tube I traced down about 14 inches, when 

 I came upon and struck the spider off at one sidej In doing so I broke 

 off a leg and palpus so that the spider died the next day. At the time I 

 thought the spider must have underground galleries in which it hunted 

 for ants similar to the galleries in the moss of the nest of Nidivalvata 

 marxii. But since finding the nests of other individuals I think there 



1 Nidivalvata angustaia. 



2 I have observed the same thing with Myrmekiaphila foliata. 

 Entomologica Americana. 16 August 1886. 



