200 MR. E. B. POIJLTON ON THE EXTERNAL 



or pronounced, as in Plate XXI. fig. 34, or accompanied by marked deformity, as in 

 Plate XX. fig. 26. 



I have already stated that these organs appear to the naked eye to belong to the ninth 

 abdominal. They are represented in this position in all figures which are of the natural 

 size. "When moderate powers are employed, together with the best means of illumination, 

 and when the most favourable individuals are selected for examination, the organs appear 

 to be surrounded by a furrow or line which is continuous on either side with the 

 boundary between the ninth and tenth abdominal segments. This interjiretation is 

 supported by Plate XX. figs. 17, 18, 19, 21, and 23, and Plate XXI. fig. 2 ; that is to say, 

 it is supported by all the most carefully drawn and highly magnified figures. In some 

 of these, and especially in Plate XXI. figs. 12, 13, the organs appear to belong to an 

 anterior median extension of the tenth abdominal. 



I have not described the differences between these organs in the various species 

 examined, nor the sculpture of tlieir surface as compared with that of the adjacent area, 

 because all this can be sufliciently made out by studying Plates XX. and XXL, and by 

 reading the description of the figures. 



3. The Female External Reproductive Organs. — These organs will also be described 

 briefly, and the reader is referred to Plates XX. and XXL for the details of specific and 

 individual differences. 



When the female pupa is examined with the naked eye, or with a low magnifying- 

 power, the first character which arrests attention is an anterior median ventral wedge- 

 shaped extension of the tenth into the ninth and sometimes into the eighth abdominal 

 segment. In the majority of pupa? this feature is certainly more conspicuous than the 

 generative apertures, and I observed and figured it long before I had detected the latter. 

 It is interesting to note that Lyonet figured it in the pupa of Bomhijx rubi, although 

 without any other feature characteristic of the external reproductive organs. (See 

 Lyonet, plate xxiv. fig. 12.) The median prolongation is especially conspicuous in the 

 pupa3 of Heterocera, but distinct traces of it can be often made out in Pthopalocera. The 

 anterior median ventral margin of the ninth abdominal is also prolonged into the eighth 

 for a variable distance, thus permitting the ventral prolongation of the tenth abdominal, 

 or the generative aperture in connexion with its apex, to extend beyond the zone of the 

 ninth abdominal (see page 199). The development of both these median processes varies 

 extremely, not only in diflPerent species but in different individuals of the same species. 

 Sometimes there is a median line which traverses the prolongation of the tenth abdo- 

 minal, as in Macroylossa stellatarum (Plate XX. figs. 24 & 25). Sometimes the base of 

 the prolongation appears to be separated from the rest of the tenth abdominal as in Cossus 

 and Zeuzera (Plate XX. figs. 27 and 29; Plate XXI. fig. 1.) Any discussion as to the 

 possible meaning of this feature is better deferred until after the consideration of the 

 generative apertures. 



There are two distinct generative openings in female pupse — an anterior for the bursa 

 copwlatrix, and a posterior for the oviducts. The anterior is probably always associated 

 with the eighth abdominal ; it is more distinct than the other, and is usually in'ovided 

 with prominent lips. It is often slit-like, and extends from the posterior to the ante- 



