198 ME. E. B. POULTON ON THE EXTERNAL 



the eighth abdominal segment, and indicate thus the normal position of the outlet of the 

 generative glands." It is very difficult to see how a morphologist could have come so 

 near the discovery of the external organs of reproduction and yet have failed to find 

 them. The only possible explanation can he that, in the examination from which the 

 above description was written, only a single pupa was made use of, or perhaps a small 

 number, all of which happened to be of the same sex. It is quite clear that Professor 

 RoUeston was speaking of a female pupa, for the organs are so obvious as distinct struc- 

 tiu'es in the males that they could not liave been passed over. Besides, the descrij)tion 

 quoted above is under no circumstances applicable to the male sex, while it does form an 

 imperfect account of the appearance in certain female pupse. 



On the other hand, almost all entomologists who have carefully figured large pupse 

 exhibit in their drawings ti*aces of the sexual characters which are sometimes accurately 

 rendered in the case of the males. Thus, Lyonet figures (plate 39. fig. 3) a pupa of Cossus 

 ligniperda with distinct male organs ; Moore (' Lepidoptera of Ceylon ') represents some 

 large pupse with indications of the generative structures. Burmeister, in his beautiful 

 illustrations of the Lepidoptera of the Argentine Republic, also represents these parts. 

 Thus his plate 18. fig. 11 represents a distinct male pupa of Attacus hesperus ; while 

 his plate 20. fig. 5 B is an equally distinct female pupa of Ceratocamjja imperialis. In 

 the description of these figures the position of the generative aperture is pointed out, 

 l)ut the sexual differences are not observed. Mr. W. F. Kirby has pointed out to me 

 that there is a brief description of the male characters in Berge's ' Schmetterlings- 

 buch ' (5th ed. 1876, p. viii) : — " Bei dem mannlichen Gesclilechte auf dem vorletzten 

 Ringe zwei durch einen Eindruck getrennte Hockerchen." 



The morphology of the ^exual structures is never attempted in these descriptions, and 

 the figures are not sufficiently accurate or detailed to be of any value. This is especially 

 true of the female sex, and I believe that an accurate figure of the female characters has 

 never been publislied until now. 



I first noticed the male organs in a pupa of Sjihinx ligustri in the autumn of 1883, and 

 at once began to examine a number of pupa? in order to find the characters of the oppo- 

 site sex. This led me to undertake a careful comparison of the external generative 

 organs in a large number of species, and finally to investigate all the features which make 

 up the external morphology of this stage in Lepidoptera. This investigation has been 

 intermittently continued up to the present date. Many of the figures on Plates XX. and 

 XXI. were drawn during the autumn and winter of 1883 and early in 1884, namely, 

 Plate XX. figs. 3, 4, 5, 6, 7, 12, 13, 14, 16, 18, 19, 22, 30, 31, 32, Plate XXI. figs. 4, 5, 6, 7. 

 When I showed these figures to Professor Moseley, who took a very great interest in the 

 work, and helped me with many kind suggestions, and with the results of his own 

 observations upon pupae, he expressed the opinion that the figures should be made on a 

 much larger scale. The size of the other figures on the two Plates was adopted as the 

 result of this advice. The smaller figures are of the natural size, and are useful in 

 showing how much can be made out with the naked eye. 



I have also to thank my friends Professor Meldola and Mr. W. White for kindly pro- 

 viding me with much of the material which has been made the subject of investigation. 



