THE INSTINCT OF SPIDERS 143 



with a clean glass rod. The radii, the hub, the 

 temporary spiral, the foundation-lines were non-viscid ; 

 the viscid spiral alone adhered to the glass rod. I 

 then smeared the rod with a thin layer of oil and found 

 that the viscid spiral adhered more feebly to the rod 

 when smeared with oil than when it was quite dry. I 

 used a spider's leg in place of the glass rod. It did 

 not adhere to the viscid spiral, but after moistening it 

 thoroughly with benzene so as to dissolve away any 

 oily coating, the leg immediately adhered to the viscid 

 spiral. From experiments similar to these M. Fabre 

 has discovered that it is the possession of an oily 

 coating which prevents the geometrical spider from 

 becoming entangled in its own web. 



A spider, I think, uses as far as possible the non- 

 viscid radii during its movements in the snare, but this 

 is not due to the fact that it would stick to the spiral, 

 for I covered all the radii of a perfect snare with a 

 thick coat of adhesive gum and the spider was well 

 able to run backwards and forwards along" them. 



Spiders other than geometrical spiders become 

 entangled in the circular snare. They do not possess 

 the oily coat so essential to the Epeiridce. I took a 

 species of Hippasa which constructs a non-viscid plat- 

 form on which insects become entangled but to which 

 they do not adhere. I placed the Hippasa on a 

 circular snare. It was quite as helpless as any fly ; it 

 did not attempt to creep over the snare, but was 

 immediately entrapped. Moreover, its surface could 

 not have possessed any fine coating of oil, for its legs 

 adhered to the viscid spiral. Thus it is clear that 

 geometrical spiders are perfectly adapted to move over 

 the glutinous filaments of their webs. 



