INSTINCTS OF TilE SPIIEX. 301 



imposed necessity, instinctively choosing a number of minute 

 points in the uniformly soft body of its prey, with an appa- 

 rently very precise knowledge tliat it is only at these par- 

 ticular points that the peculiar paralyzing influence of its 

 sting can be exercised. After duly considering this case, I 

 must candidly say that I feel it to be the most perplexing 

 which has yet been brought to light, and the one whicli is 

 most difficult of explanation upon the principles of tlie fore- 

 going theory. It is, however, most desirable that tlie facts 

 should be more thoroughly investigated, for it might then 

 appear that some clue would be given as to tlie origin and 

 development of this instinct. So far as our information at 

 present extends, I can only suggest that this origin must have 

 been of a purely secondary kind, although its subsequent 

 development may probably hav^e been assisted by natural 

 selection. In other words, so far as we have any means of 

 judging, I can see no alternative but to conclude that these 

 w^asp-like animals owe their present instincts to the high 

 intelligence of their ancestors, who found from experience 

 the effects of stinging caterpillars between the segments of 

 their bodies, and consequently practised the art of so stingmcr 

 them till it became an instinct. 



During the last year of his life I had some conversation 

 with Mr. Darwin upon this matter, and, after deliberating 

 upon it for some time, he eventually came to the conclusion 

 which I have just stated — as will be at once apparent from 

 the following letter wdiich he wrote to me, and which will 

 serve in a few words to indicate what appears, I think, to be 

 the most probable steps by which these singular instincts were 

 acquired. 



" I have been thinking about Pompilius and its allies. — 

 Please take the trouble to read on perforation of the corolla 

 by Bees, p. 425 of my "Cross-fertilization," to end of chapter. 

 Bees show so much intelligence in their acts, that it seems not 

 improbable to me that the progenitors of Pompilius originally 

 stung caterpillars and spiders, &c., in any part of their bodies, 

 and then observed by their intelligence that if they stung 

 them in one particular place, as between certain segments on 

 the lower side, their prey was at once paralyzed. It does not 

 seem to me at all incredible that this action should then 

 become instinctive, i.e., memory transmitted from one genera- 

 tion to another. It does not seem uesessary to suppose that 



