1886.] Larva of Smerinthus ocellatus and its Food-plants. 155 



suggested, is similar in its effects to 8. rubra. The effects of various 

 plants hitherto untried have also been observed as a result of the 

 experiments and work in the field. 



As to the occurrence of individual variation in larvae from the same 

 batch of eggs and fed upon the same food-plants, it is now quite 

 certain that such variation may take place, but any considerable 

 divergence is very exceptional. 



Thus in the twenty- three larvae bred in 1884, there was practi- 

 cally no individual variation, while in 1885 there were only eight 

 instances out of seventy-five larvae, and in none of these instances 

 did the variation amount to more than a remove of one place from that 

 which contained the largest number of larvae, and which therefore 

 represented the normal result of the food-plant for each particular 

 experiment. In such a calculation the differences between the 

 larval colours are arranged in five classes, i.e., white, whitish inter- 

 mediate, intermediate, yellowish intermediate, and yellow. The 

 difference between any two of these is very small, and hence it is 

 seen how entirely insignificant was the amount of individual varia- 

 tion even in the few cases in which it occurred. In one instance only 

 was there a variation on both sides of the normal result, i.e., in 

 Series III, Nos. 11, 12, 13, where seven larvae fed upon S. cinerea 

 became intermediate in one case, yellowish intermediate in five 

 cases, and yellow in one case. Here there is a difference of two 

 places between the extremes, but one larva only varied in each direc- 

 tion, while five remained normal. Thus, although this is by far the 

 most striking instance of individual variation met with in about a 

 hundred bred larvae in 1884 — 1885, it is by no means extreme, and 

 cannot alone explain such excessive variations as have been met 

 with in the field out of about an equal number of instances. I refer 

 especially to the instance recorded in my last paper (p. 302), in which 

 a bright yellowish variety was found upon apple, the divergence from 

 the normal result being as wide as possible (five places). Another 

 almost equally striking instance was met with this year (as will be 

 recorded) upon 8. cinerea in the field, one larva being whitish inter- 

 mediate and four others yellowish. Here the divergence amounts to 

 four places, and compares in an interesting way with the lesser 

 divergence in the larger number of larvae bred upon the same plant. 

 A divergence equal to that upon 8. cinerea was recorded in my last 

 paper (p. 301) upon 8. ferruginea, one larva being yellowish and three 

 whitish intermediate. It is possible or even likely that considerable 

 divergence is occasionally caused by individual variation, but that 

 this is not the only or indeed the chief explanation of the few in- 

 stances of extreme divergence recorded, is proved by the fact that 

 such variation only occurs when the probabilities are greatly in 

 favour of correspondingly different hereditary tendencies, and that 



