138 Mr. Poulton's 7iotes in 1885 upon 



stage, and I was very interested to see that the two 

 apical head tubercles still retained traces of a red colora- 

 tion, thus proving that these tubercles are red in earlier 

 stages, as in S. ocellatus. The way in which the colour 

 changed at the beginning of the last stage was exactly 

 as in S. ocellatus, the red fading into the cleft between 

 the yellowing tubercles, and soon disappearing entirely. 

 The prothoracic spiracle appears to be white, the others 

 red ; the effect being produced by a distinct red line 

 which encircles the latter, while careful examination 

 shows that the former is also surrounded by a very fine 

 red line, which, however, does not play any part in the 

 general appearance of the larva. I expect that the 

 anterior spiracle is concealed during rest, as in S. 

 ocellatus. During the past summer (1885) I also found 

 two adult larvae of this species at Oxford, with the red 

 borders to the stripes spoken of by Weismann. There 

 could not be the least doubt that the red borders repre- 

 sented the violet ones of Sphinx lignstri, and the coloured 

 borders of the oblique stripes in other Sphingidce, and 

 that Weismann's identification is perfectly correct. But, 

 as I have before stated, I cannot think that in S. ocellatus 

 and S. populi we have the traces of a similar marking, 

 for there is never any tendency towards the elongation 

 of the spots into borders in these species, and yet the 

 spot system is developed to a much greater extent than 

 in S. tilicB. And yet in a larva of S. ocellatus, which 

 possessed this character to a very slight extent, there 

 was a very strong resemblance to Weismann's description 

 of the faintest trace of the red borders in S. tilice. 

 Weismann says of the latter larvae in the fourth stage, 

 on p. 235 of the English translation of his essay on 

 * The Origin of the Markings of Caterpillars ' : — " Many 

 of them possess a blood-red spot on the anterior 

 side of the stripes, this spot showing all gradations in 

 size and depth of colour between maximum development 

 and a mere trace." The larva of *S'. ocellatus, to which 

 I allude, was one of two which possessed the spots out 

 of about seventy-five which had been bred fr'om the egg. 

 The spots were only present in the upper row (three 

 rows appearing in extreme varieties) on the first five 

 abdominal segments. The spots, which could only be 

 detected in the last stage, were indicated by a little local 

 darkening of the green anterior border to the stripes, 



