EUMORPHA ELPENOR. 69 



at Reading (Butler), larvae fullfed August 20th, 1889, at Aylsham 

 (Freeman), July 29th, 1891, larvae at Horning (Bowles), August, 

 1892, at Folkestone (Byrne), larva August 3rd, 1892, at Guild- 

 ford (Grover), larva August 18th, 1892, at Harrow (Rhoades- 

 Smith), July i5th-24th, 1893, at Wicken (Mitchell), August 25th- 

 September 2nd, 1894, August 19th, 1895, August i5th-27th, 

 1897, August 3rd-28th, 1898, July i9th-24th, 1 8gg, July 16th- 

 September 3rd, 1900, at Sudbury, July 22nd-24th, 1896, August 

 nth-i7th, 1897, May 25th, 1898, July 3ist-August 1 8th, 1899, 

 July i8th-August 12th, 1900, July 14th- August 1 8th, 1901, at 

 Henny, August 21st, 1898, at Borley (Ransom), larvae August 

 30th, 1894, fullfed September 1st (Glenny), fullfed larvae on July 21st, 

 1896, at Church Stretton (Newnham), larvae at Wicken from August 

 i8th-25th, 1895 (Brady), larvae common at Warminster in August, 

 1896, August 4th, 1898, a single larva at Tullylagan (Greer), 

 larva fullgrown August 4th, 1896, larvae fullfed August 17th- 

 18th, 1899, at Chelmsford (Miller), larvae September, 1896, at 

 Penmaenmaur (Bland 1 , larvae August 8th-22nd, 1897, on the banks 

 of the Stour (Wilson), August 12th, 1898, at Epping (Image), 

 larva on willow-herb, August 24th, 1898, at Stalham (Edelsten), 

 September 9th, 1898, fullfed larva, September 15th, 1899, larvae 

 at Lakeside, Windermere (Moss), 50 larvae during the first three 

 weeks of August, 1900, and August ist-20th, 1901, larvae common at 

 Rye (Henderson), larvae August and September, 1900, at South- 

 ampton and Winchester (Moberly), fullfed August 24th, 1900, in 

 the Frensham dist. (Bingham-Newland). 



The terrifying attitude of Eumorpha elpenor. — Weismann 

 deals at length (Studies in the Theory of Descent, pp. 327 et sea. J 

 with the formation of the ocellated spots in Eumorphid larvae 

 and their possible significance. He shows that the primary ocelli 

 originate on the 1st and 2nd abdominal segments by the detach- 

 ment of a curved portion of the subdorsal line, this fragment becoming 

 the " mirror," and acquiring a dark encircling zone, "the ground 

 area," the nucleus or " pupil " being added subsequently. He 

 points out that these spots make the larvae more easy of detection 

 than otherwise, and that they appear to possess a biological value 

 as a means of terrifying their enemies. He observes that the 

 ocellated Eumorphid larvae remain quiet on being attacked, that 

 they merely withdraw the head and thoracic segments into the 

 swollen 1st and 2nd abdominal segments, and that the large ocellated 

 spots being situated on these swollen segments, the larva assumes 

 an aspect which he shows by experiment (loc. cit., pp. 331-334) is 

 most alarming to certain small birds. He concludes that the 

 advantage of the protective coloration of such larvae remains as strong 

 as ever, and aids the concealment of the larvae, that these markings are 

 of service to the larvae when attacked, and that they thus serve as a 

 second means of defence, resorted to when the protective coloration has 

 failed. Poulton observes (Trans. Ent. Soc. London, 1886, p. 154) that 

 it is specially to be noticed that the terrifying appearance due to these 

 markings, would be effective only against an enemy approaching from 

 the side or from above, but would produce very little, if any, effect 

 upon an enemy advancing from the front. He points out as a fact that 

 seems to have escaped the notice of Weismann, although it is really 



