1903.] NATURAL SCIENCES OF PHILADELPHIA. 67 



the basG; thereby leaving the egg mass exposed. In all these cases 

 the first part of the cocooning and the oviposition proceeded normally; 

 the mistake was in making the cover or in tearing the cocoon loose. 

 And once a mistake in the progress was started none of these indi- 

 viduals w^ere able to rectify it, though one tried for a few moments to 

 spin over the rent in the cocoon, and each in conclusion chewed up 

 either the imperfect cocoon or the eggs; when the cover was torn off 

 the eggs, the spider would very frequently chance to break some of 

 the latter, and once she tasted their contents she would eat them. 

 They would seem perplexed after a cover had separated from the base, 

 would remain still a long while, would feel with the palpi first the cover 

 and then the egg mass on the base ; usually they would take the isolated 

 cover beneath the body and spin upon it there, as they do with a 

 normal cocoon, but did so irregularly. An isolated cocoon cover was 

 never carried attached to the spinnerets more than a short time. One 

 spider had made her cocoon perfect except for a break at one point 

 between the cover and the base; this resulted in an elongated, irregu- 

 larly conical cocoon, which she dropped from her spinnerets after a 

 few days. 



Now the reason for such abnormal cocoons seems to be in a mistake 

 made in tearing the cocoon loose; but there is still a reason back of this 

 one, at least for some of the cases. It was my custom with these 

 spiders, as with the other Lycosids, to put a drop of water upon the 

 floor of their cages every day or two, for them to drink. Three of 

 the miscarried cocoons whose making was observed, were commenced 

 by spinning over the surface of this drop of water. The water in some 

 way mingled with the silk of the cocoon, and at least contributed to 

 making it a failure. Why they should spin over the water I cannot 

 say, except that perhaps the feeling to the touch of the surface of the 

 water may simulate the feeling of a silken layer, and so awaken reflexes 

 to spin at that place. In any case, all the Lycosids and one Philo- 

 dromid, which chanced to lay the base of their cocoons upon a water 

 surface, made a failure of these cocoons. But in no other species have 

 I experienced such a high percentage of failures as in Lycosa sionei. 



The number of cocoons, and the dates, were as follows: 9 No. 36, 

 June 6, a failure; ? No. 165, June 25, a failure; ? No. 184, June 29, a 

 failure; ? No. 227, June 27, July 16, both failures; $ No. 228, June 25, 

 July 14, the second one a failure; ? No. 229, June 29, a perfect cocoon, 

 but cut open too soon by the mother, so that the eggs rolled out; 

 ? No. 230, June 27, July 14, both failures; 9 No. 231, June];22, a 

 perfect cocoon, but the mother destroyed it. 



