144 Mr. E. B. Poulton. Colour-relation Letween the [Feb. 4, 



nearly full fed, and slightly upon the yellowish side of intermediate, 

 this being the last note, and giving the final result. 



12. Salix cinerea. — Three larvae (hatched July 11th) were tolerably 

 full fed when they were examined on August 16th. One was on the 

 white side and one on the yellow side of intermediate, while the third 

 was a yellowish variety (although not strongly yellowish). It was 

 extremely interesting to note that the latter — the only undoubtedly 

 yellowish larva yet obtained in my breeding experiments in 1884 and 

 1885 — possessed traces of the red spots that occur commonly on the 

 yellowish varieties of 8. ocellatus. On the first five abdominal 

 segments there was a little local darkening of ihe green borders to 

 the oblique stripes occupying the position of the upper row of red 

 spots, and in the centre of each dark spot there was an extremely 

 faint tinge of red. There was also a very slight tendency towards 

 the suffusion of the ground colour round the spiracles with a tinge 

 of red. On August 20th the larvae were as they have been described 

 (except that the whitish intermediate larva was now intermediate), 

 and were practically full fed. The yellowest one was a bright yellow 

 variety (although there was but little yellow on the under surface, so 

 that the larva was not one of the strongest varieties). On August 22nd 

 the yellow variety and the intermediate larva had ceased feeding, 

 while the yellowish intermediate larva became adult about August 25th. 

 There was no further change in the colour of any of the larvae. 



13. Salix cinerea. — Three larvae (hatched July 12th) were well in 

 the last stage when they were examined on August 16th. One was 

 on the white side of intermediate, and two intermediate or slightly on 

 the yellowish side. On August 20th and 27th the larvae were again 

 examined and had progressed in the direction of the yellowish variety, 

 so that on the latter date — when two had ceased feeding, and the 

 other, though still feeding, was mature — they were all on the yellowish 

 side of intermediate, although only slightly so in one case. 



14. Populus nigra. — Five larvae (hatched July 10th) were examined 

 on August 12th, when four were in the last stage and one smaller. 

 They were all whitish, but looked as though they were progressing in 

 the direction of intermediate. On August 16th only two remained 

 alive, one being well in the last stage and a whitish intermediate 

 variety, while the other was whitish, being much smaller and not 

 thriving. On August 27th the large larva was the only one alive and 

 was advanced in the last stage, and a distinct intermediate variety 

 without any tendency in either direction. This represents the final 

 result, as the larva subsequently died without further change. 



15. Salix triandra. — Eleven larvae (hatched July 9 — 12th) — of 

 which three were small, but eight were advanced in the last stage — 

 were examined on August 12th, and were all intermediate varieties, 

 as different as possible from those from the same batch of eggs which 



