258 MR. R. I. POCOCK— MIBIIORY IN SPIDERS. 



and thus leading in appearance to the partial obliteration o£ the narrow waist- 

 like constriction between these regions which is so characteristic a feature of 

 most sj)iders. Modifications of the same kind, accompanied by the acquisition 

 of a yellow and black pattern, are instrumental in producing the likeness 

 to Coccinellidse, above referred to. 



Mimicry of Ants and Altdillas hy Spiders, 



Apart from beetles, the only insects known to me to be mimicked by 

 spiders are ants ; but so numerous and perfect are the instances of ant- 

 imitation that all other examples of mimicry amongst spiders sink into 

 insignificance beside them. It is in this category, indeed, that occur some of 

 the most complete illustrations of mimicry yet discovered. 



The structural variations that effect this end take quite a different direction 

 from those presented by the spiders that mimic beetles. Ants as a group are 

 very uniform in build. They are characterized by a peculiar elegance of 

 shape, marking them off from all other animals except their imitators. The 

 head is large, the neck slender and mobile ; the thorax and abdomen, often 

 themselves constricted, are connected by a narrow waist ; the legs are 

 slender ; the antennae, projecting from the front of the head, long and bent. 

 The movements of these insects are stamped by a certain boldness and 

 intelligence which impart an unmistakable individuality to them. The chief 

 difference in outward form between a spider and an ant lies in the absence of 

 antennee and the possession of four pairs of legs, instead of three, in the 

 spider, and in the fact that the fore part of the body is covered by a con- 

 tinuous shield or carapace and is not differentiated into "head," " neck/' and 

 " thorax.''^ In the more perfect instances of mimicry the last-mentioned 

 difficulty has been met by the formation of a distinct constriction on each side 

 of the spider's carapace, dividing it into an anterior part representing the 

 head, a narrow intermediate part representing the neck, and a posterior part 

 representing the thorax of the insect. In many cases the appearance of 

 slenderness about the neck is increased by a strip of white hairs on each side 

 of the constriction, which has the optical effect of cutting out an extra piece 

 of the integument. The waist of the ant is reproduced by the conversion of 

 ^he end of the carapace, and often of the anterior end of the abdomen, into a 

 narrow stalk. In some cases the abdomen is itself shallowly constricted, and 

 even the abdominal segments of the insect may be represented by transverse 

 bands of hairs. The legs are always slender, like tliose of an ant, and one of 

 the anterior pairs is held up in front of the head as a substitute for antennae. 

 Finally, in many cases where observations of the living animal have been 

 recorded, it has been found that the spiders carry the deception to the extent 

 of copying the manners and gait of the insects. Several observers have 

 testified to this fact. In 1879, the Rev. 0. P. Cambridge * pointed out that 



* ' Spiders of Dorset/ p. 13; see also II. Donistliorpe, 'Tlie Zoologist,' 1908, p. 424. 



