154 PROF. W. H. JACKSON ON THE 



Vanessa lo, that the part of the abdomen from the posterior edge of the eighth sternum, 

 and the spot where there is a slight fold in the pleural membrane of eitlier side 

 (PL XVI. fig. 25, X ) backwards, represents the ninth and tenth somites of the larva 

 and pupa, now indistinguishable, whilst the terminal cone is a new formation within the 

 area of the tenth somite. The eremaster of the female pupa, it may be added, undergoes 

 complete atrophy, and is not represented in the imago. 



It remains to give a summary of what is described by Ilerold in his ' Entwickelungs- 

 geschichte der Schmetterlingc ' (Cassel and Marburg, 1815), and by Suckow in his 

 ' Anatomisch-Physiologische Untersuchungen der Insecten und Krustenthiere,' Th. 1 

 (Heidelburg, 1818), so far as their accounts relate to our purpose. 



In the full-grown female larva of Pieris brassicce, the species investigated by Ilerold, 

 the larval oviducts which are continuous with the larval ovaries run backwards near the 

 dorsal middle line ; they pass between the tracheae given off from the seventh pair of 

 stigmata, and are attached to a main trachea of the bundle by a filament ; they each 

 turn towards the ventral middle line in the seventli somite behind the seventh stigmata, 

 pass beneath the main longitudinal and oblique muscles of the somite, and are attached 

 close together to the hypodermis (=Sclileimnetz) in the middle ventral line at the 

 posterior margin of the seventh somite *. From the attached ends of the larval oviducts 

 start fine longitudinal strito which traverse the eighth sternal region and connect the 

 larval oviduct to a white mass composed of two oval pieces attached to the hypodermis 

 ( = Sclileimnetz) beneath the rectum upon the intersegmental membrane immediately 

 following the eighth somite (=auf der hintersten oder letzten ringformigen Einkerbung 

 der Raupenhaut, op. cit. p. 10). During the quiescent state preceding pupation the two 

 oval pieces become approximated to the ends of tlie larval oviducts, owing to a sliortening 

 or contraction of the hypodermis (Sclileimnetz). As soon as pupation has taken place, 

 they fuse with the oviducts and acquire a soft loose consistence (eine weiche und auf- 

 gelockerte Consistenz, op. cit. p. 49). During the pupal state the two larval oviducts 

 fuse together at their point of union with the oval pieces to constitute the common {i. e. 

 azygos) oviduct, whilst the oval pieces give rise to three processes, a left larger, the 

 rudiment of the bursa copulatrix (Ilerold's " Samenbehalter "), a right smaller, the 

 rudiment of the receptaculum seminis (Hcrold's " ein-horniges Absonderungsorgan "'), 

 and a bifid posterior, the paired sebaceous gland (Ilerold's " zwei-horniges Absonderungs- 

 organ "). The mass which gives rise to these processes disappears, whilst they increase 

 in size and acquire the form they possess in the imago. Herold nowhere explains how 

 the two separate apertures to the genital duct are acquired ; nor can any information be 

 gained from his text or later plates (Taf. xxvii., xxix., xxx., xxxi., and xxxiii.). But he 

 was quite aware of their existence and describes them in his account of the genital 

 organs of tlie imago with good figures (Taf. iv. figs. 1, 2). 



Suckow's investigations were carried out on DendroUmus {Odonestis) j^ini', or, as he 

 terms it, following LinnEBus, Bomhijx piid. Ilis results agree with Herold's, but certain 

 slight difi'erences may be noted. The two filamentous oviducts are prolonged beyond 



* Herold describes the course of the ducts correctly, but the way in which he figures them in his plates (Taf. vii., 

 ix., xi.) is most misleading. 



