MR. R. 1. POCOCK — MIMICRY IN SPIDERS. 265 



protected by their resemblance to these Hymenoptera, can be used with equal 

 cogency to explain the ant-mimicry o£ spiders. There is no proof that ant- 

 like spiders habitually feed upon the insects they mimic ; but if this be shown 

 in the future to be the case, I must still think that it is a secondarily acquired 

 habit that has arisen subsequently to and independently of the superficial 

 similarity in shape, colour, and gait between the two groups of arthropods. 



Apart from the evidence supplied by the independent attainment of an ant- 

 like form by a large number of spiders belonging to widely divergent families, 

 strong testimony in favour of the protective usefulness of the resemblance 

 can be brought forward. 



An item of indirect evidence pointing to this conclusion is the fact that at 

 least some of the spiders thus modified are known to produce at birth a very 

 small number of young. According to Mrs. Peckham the ant-like spiders 

 of North America have lower fecundity than any other group of spiders. 

 Peckhainia picata, for example, lays only three eggs at a time. Since fertility 

 is a measure of mortality, this exceptionally low birth-rate proves that the 

 species is especially protected in some way ; and since the spider is small, 

 weak, and unprovided with special means of defence, it is justifiable to regard 

 its survival, despite its infertility, as due to its likeness to ants. On the other 

 hand, it may be asked, is there any direct evidence that ants are immune to 

 the attacks of the enemies of spiders ? A confident answer in the afiirmative 

 can be given. Ants are believed to be partially protected by an acid taste 

 and are known to be able to bite and sting. They are also protected by their 

 numbers and pugnacity, and are small, wiry, comparatively innutritions, and 

 not worth the taking where other insects are to be had for food, excejjt indeed 

 by the comparatively small number of specialized vertebrate ant-eaters which 

 are alleged to lick them up by the score *. 



But whatever may be the reason, it is nevertheless a fact that ants are not 

 preyed upon to any great extent at all events by other insects. Mrs. Peckham f 

 too, found that of the large numbers of Salticidse kept by her in captivitv 

 none would touch ants, though they quickly devoured other insects and also 

 spiders even of their own kind. 



Most important of all is the fact that the solitary digger- or mason-wasps 

 the most inveterate enemies of spiders, scarcely ever, so far as I can ascertain 

 provision their cells with ants |. Some species, indeed, have been stated to 

 evince a positive aversion to these insects. Horne§, for example, says that 



* In this connection it is very necessary to bear in mind the confusion that is commonly 

 made between ants aud white-ants (Termitidaj). I suspect that many of the so-called 

 " ant-eaters " like Myrmecojj/iaga, Manis, and On/cteropus, have been so styled largely on 

 account of their termite- eating- propensities. 



t Occ. Papers Nat. liist. Soc. Wisconsin, 1889, p. 107. 



X One of the Crabronides (Fertonius luteicollis) preys upon ants in Algeria — see D. Sharp 

 Insects, pt. ii. (Cambr. Nat. Hist.) p. 130, 1899. 



§ Trans. Zool. Soc. vii. p. 171, 1872. 



