84 



in their last larval stage, the latter admirably shown in 

 Buckler's " Larvae of British Butterflies and Moths," vol. iv, 

 pi. Ivi. 



There is an aspect of larva rearing in which the adepts of 

 this Society may render most important aid to science, by 

 carrying further thosejnvestigations so ably initiated by Dr. T. 

 A. Chapman, dealing with the pupal condition of the Lepidop- 

 tera-Heterocera. That accomplished entomologist in an epoch- 

 making paper read before the Entomological Societ}^, February 

 22nd, 1893 {vide Trans. Ent. Soc. 1893, pp. 97-119), has 

 shown that the Lepidoptera-Heterocera fall naturally into 

 two quite distinct divisions, viz., {a) Obtectae, which have the 

 " Pupa smooth and rounded, externally solid, inner dissepi- 

 ments flimsy. Free segments in both sexes 5 and 6 (abdo- 

 minal). Never emerges from cocoon or progresses in any way. 

 Dehiscence by irregular fracture." {b) Incompletae, "pupa 

 less solid, and rounded, appendages often partially free. Free 

 segments maj' extend upwards to 3rd (abdominal), 7th always 

 free in male, fixed in female. Dehiscence, accompanied by 

 freeing of segments and appendages previously fixed (except 

 in Pterophorinse), pupa progresses, and emerges from the 

 cocoon." 



Here is a fundamental difference of a structural character 

 existing between the two divisions, clearly set forth ; the 

 pupae of the Incompletae are more or less capable of pro- 

 gression ; in some species, indeed, their movements are so 

 lively, e.g., Macrogaster castanece {arundinis), that care has to be 

 taken, when cutting the reed inhabited by a pupa, that it does 

 not escape by rapidly wriggling out at the open end ; on the 

 other hand, those numerous species of the Obtectae which 

 pupate under ground, at the time of the emergence of the 

 imago remain immovable ; and the forcing its way out 

 through the soil is performed by the perfect insect, although 

 it would seem that the hard chitinous pupa was better 

 adapted for such work than the soft imago. 



Time will not permit me to enter into a full discussion of 

 the subdivisions which the learned Doctor tentatively suggests 

 should be made in each division ; but there are several dis- 

 locations of the classification hitherto in use which more 

 particularly call for remark. 



The Sphingidae are relieved from their very erroneous 

 association with the Sesiidse, so called erroneously, and 

 Zygaenidse, with which they have so very little in common ; 

 the two latter families are veritable Incompletae, and the 

 former are typical Obtectae. 



