on colour-relation, 245 



while it afforded "no evidence" for Mr. Poulton's sug- 

 gestion "that a Hmit to the success of this method of 

 defence would result from the hunger which the success 

 itself tends to produce." 



My experiment affords strong support to Mr. Poulton's 

 suggestion, since the larva was tasted twice, and that 

 when the bird was not especially hungry, which points 

 to its being eaten if he had been excessively hungry. 



Mr. Poulton says that Mr. Jenner Weir also experi- 

 mented with Ciicidlia verbasci, on "many species of birds 

 and lizards," and that the larvae were " disregarded." 

 He refers to this as "strong support" to Prof. Wallace's 

 suggestion, and as "no evidence" for his own. 



My experiment afforded some support to Mr. Poulton's 

 suggestion, since the larva was tasted once, and tasting 

 would put a limit to the success of the method of defence 

 as well as eating, because it would be fatal to the larva. 

 I do not think, however, that this species would have 

 been eaten in any degree of hunger. 



Both my experiments supported Prof. Wallace's sug- 

 gestion also, as the larvae were in both cases refused at 

 first ; and, if the bird had not been very hungry, I do 

 not think he would even have tasted C. verbasci. 



In September, 1892, I also made some experiments 

 with larvae of Acronycta psi and Bombtjx rubi on three 

 slowworms {A. fragilis), and one lizard {Z. vivipara). 

 Both species were entirely disregarded by all the animals, 

 though they were kept very hungry, and the larvae left 

 with them for days. 



It may not be out of place to note here that during 

 this experiment I had incontestable evidence of the 

 nature of the food taken by A.frafiilis in the natural 

 state. The slowworms received rather rough handling 

 when captured, and immediately afterwards cast up a 

 quantity of half-digested food, among which I found a 

 perfectly uninjured adult shell of Zo)iites radiatulus. 

 Kemains of slugs were also recognizable. It is rather 

 remarkable that A. fragilis should be able to swallow so 

 large and hard an object as the snail-shell. 



