1886.] Larva of Smerinthus ooellatus and its Food-plants. 159 



thoracic segment in front of and parallel with the more distinct and 

 larger " eighth stripe " upon the first abdominal segment. 



5. Experiments upon captured Larvas. 



Being engaged in the extensive breeding experiments already 

 described, I did not attempt much with the captured larvae, especially 

 as nearly all of the latter were full-grown when found. The strongly 

 yellowish larva in the fourth stage, found August 2nd on Salix rubra, 

 was put upon apple on August 3rd, when it was 24 mm. long. On 

 August 27th it was 51 mm. long, and was much affected by the 

 change of food, being an intermediate variety, or perhaps slightly on 

 the yellowish side of intermediate. It was interesting to note that 

 the change of colour affected the shagreen dots, which became white, 

 having been formerly yellow as in all strongly yellowish varieties. 



6. Conclusions arrived at by the Consideration of the captured Larvae : 

 The Reconciliation of conflicting Evidence. 



The colours of the captured larvae were wonderfully uniform for 

 their respective food-plants. Salix rubra produced a large number 

 of yellow larvae, and others which were but little removed from 

 yellow. The larva upon 8. babylonica resembled these latter. 8. 

 triandra also produced yellow larvae, and so with 8. cinerea (with one 

 exception). There was but little confliction in the results of 8. vimi- 

 nalis, and 8. Smithiana produced a normal larva, and the colours of 

 the Swiss larvae (with the exception of that upon 8. incand) might 

 have been almost exactly anticipated by investigating the colour of 

 the under sides of the leaves. Thus there are fewer exceptions than in 

 the larvae captured in 1884, and yet among them was one instance 

 which suggested to me the explanation of those conflicting results 

 which have been the chief obstacle to the complete acceptance of my 

 theory of the colour- relation between food-plant and larva ; I mean 

 especially the immense difference between Mr. Boscher's experience 

 (quoted by Mr. Meldola as above referred to) and my own with regard 

 to 8. viminalis. The larva which suggested the interpretation was 

 the yellowish intermediate variety found August 9th upon Salix 

 linearis. I had much wished to find a larva upon this foreign species 

 of Salix, of which there are many fine specimens in the Oxford 

 University Parks ; for I had noticed for over a year that the under 

 sides of the leaves were more densely covered with down and whiter 

 than any species of Salix I had ever seen, and even more so than 

 apple. The upper sides of the leaves were dark-green and glossy, and 

 the leaves were narrow, pointed, and very small, and extended at right 

 angles to the twigs. The leaves resembled those of 8. viminalis, only 

 they were much smaller and whiter underneath. I thought that such 

 a tree must produce the most extreme white varieties, and I was 



