1887.] NATURAL SCIENCES OF PHILADELPHIA. 375 



attendance upon the same female. Something of loss may be attri- 

 buted to this cause, but I am satisfied that in a much larger degree 

 losses and malformations are due to the accidents of moulting. One 

 example I may cite, the loss of two limbs experienced by a large 

 tarantula which I had kept under observation ; for during the last 

 few years I have had a number of these large creatures in artificial 

 nests. This spider lay upon its back during part and on its side 

 during the remainder of the time of moulting. The skin was cast 

 by a succession of movements of the body or parts of the body re- 

 curring at regular intervals reminding one of labor pains among 

 mammals. For some reason two of the legs refused to separate 

 from the skin and after a prolonged struggle they Avere broken off 

 at the coxae, and remained within the moult. One foot of another 

 leg shared the same fate. This? moult occurred in the spring; during 

 the latter aprt of August of the same year the spider again moulted. 

 The moult was a perfect cast of the animal, the skin, spines, claws 

 and the most delicate hairs all showing, and their corresponding origi- 

 nals appeared bright and clean upon the spider. When the cast off 

 skin was removed the dissevered members were lacking thereon, but 

 on the spider itself new limbs had appeared, perfect in shape but 

 smaller than the corresponding ones on the opposite side of the body. 

 The dissevered foot was also restored. The rudimentary leffs had 

 evidently been folded up within the coxae, and appeared at once 

 after the moult, rapidly filling out in a manner perhaps somewhat 

 analogous to the expansion of wings in insects after emerging.^ 



It is possible that my tarantula "Leidy" was too much exhausted 

 by long previous fasting to endure the severe strain which evidently 

 is laid upon the organism in the act of moulting, although judging 

 from the disjecta membra of the skin recovered from the burrow it 

 had succeeded in casting them all off without any mutilation. The 

 Spring of 1887 was a backward one, and I experienced great diffi- 

 culty in procuring insects for food from the immediate neighborhood. 

 The annual supply of grass-hoppers and locusts upon which I had 

 relied came very late. Perhaps had the spider been strengthened 

 by a few week's generous feeding previous to its last moult, it might 

 still have been alive. 



2. Hoiu to keep spiders alive. I may sa}' here that my experience 

 in keeping other large spiders is that there is quite as much danger 

 from over-feeding as under-feeding. I have found the best success 



1 See Proceedings Academy of Natural Sciences Philadelphia, 1883. p. 190. 



