INSTINCT 223 



not assume that a substance so exceptional in its 

 •character as the web of the spider could have been 

 suddenly produced by evolution. But the glands 

 would be useless without the spinnerets. If the lat- 

 ter are homologous with legs, as has been claimed, 

 then we must assume that two or three pairs of legs 

 that were probably at one time useful for locomotion 

 became so modified that they could perform the func- 

 tion of spinnerets. In what conceivable way could 

 locomotive legs have become so modified and pierced 

 with more than a thousand apertures through which 

 the web is drawn? 



True spiders are found in the Carboniferous Age, 

 and, if they were evolved, they must have appeared 

 in the Silurian. The fact that insects and scorpions 

 iiave been found in the latter renders it the more 

 probable that spiders existed at that time. The 

 spiders of the Carboniferous had, no doubt, the web- 

 making structure. I infer this from their appearance 

 and from the fact that all living spiders make web. 

 The existence of the structure necessary to manufac- 

 ture web, at a period so remote, renders it the more 

 difficult to believe that this structure could have been 

 evolved. 



If we assume that living spiders have had a com- 

 mon origin by evolution, then the instinct to manufac- 

 ture a special form of web, such as that of the Garden 

 Spider, is very ancient, for we find other species, for 

 ■example, Acrosoma arcuta, which differ greatly in 

 structure from the garden spider, but which construct 

 a nearly similar web. According to the assumption 

 the instinct has existed during the long period neces- 

 sary to evolve the great physical differences between 

 the two species. In this instance, therefore, as in the 

 ease of birds, already discussed, evolution necessarily 

 assumes the existence of instincts through immense 



