EXTERNAL STRUCTURE OF THE BUTTERFLY LARVA. 23 



Thaids, which are placed in a transverse slit on the upper part of the 

 prothorax. When irritated, the larvae thrust therefrom a large orange- 

 yellow, Y-shaped, fleshy, tubular process (the osmaterium), from which 

 is diffused a very appreciable odour, varying in its nature according to 

 the species, usually more or less objectionable, and, in some cases, 

 exceedingly so ; this is frequently accompanied by a drop of 

 fluid which Packard says is acid, and turns litmus paper red. The 

 mechanism has been described and figured by Klemensiewicz. Packard 

 notes that, when at rest or retracted, the osmaterium lies in the upper 

 part of the body in the three thoracic segments, and is crossed obliquely 

 by several muscular bundles attached to the walls of the body, and that 

 by the action of these muscles, the evagination of the osmaterium is 

 strongly promoted. After eversion, the tubes are slowly retracted by 

 two slender muscles inserted at the end of each fork or tube, and 

 arising from the sides of the metathorax, crossing each other in the 

 median line. Secretion takes place in an oval mass of glandular 

 cells at the base of the forks ; in the glandular mass is a furrow-like 

 depression about which the secretory calls are grouped. The secretion 

 collects in very fine drops on the side of each furrow opposite the 

 glandular cells. Its particular structure in Papilio machaon and the 

 details of secretion and method of movement have already been dealt 

 with (Sat. Hist. Brit. Lep., i., p. 95). We need only add here that the 

 osmateria are probably protruded by the muscular contractions of the 

 walls of the body, forcing the contained fluid into the tube, and thus 

 pressing out the reversed osmaterium. 



On the dorsum of the 7th abdominal segment of most (probably 

 all) Kuralid (Lycaenid) larvae is a narrow transverse slit. In this slit, 

 in many species, is a very minute eversible sac, whose function appears 

 to be directly opposite to that of the osmateria already described, for 

 the sac exudes a sweet fluid very attractive to ants, which may be 

 diffused more widely by the delicate spinulose bristles crossing the 

 summit. It is assumed that, in return for a supply of the sweet fluid, 

 the larvae are protected by the ants from predaceous enemies. Scudder 

 says that all Lycaenid larvae have the slit, though all do not possess 

 the gland ; in those that do possess it, it is found to be a vesicle of 

 somewhat tubular shape that can be thrust through the transverse 

 slit, which, when closed, looks exactly like a transverse line running 

 across the dorsum of the segment. The connection between these larvae 

 and ants has already been noticed (Nat. Hist. Brit. Lep., i., pp. 97-98). 

 In addition, Edwards states that, in several Lycaenid species, there 

 is, besides the gland on the dorsum of the 7th abdominal segment, a 

 pair of minute dorsal evagiuable tubercles. The larva of Pirochala 

 isocrates, the well-known Pomegranate butterfly, is said by Pargiter to 

 have two white spots near the anal end of the body, in each of which 

 is a small hornlike process, which the larva continually protrudes 

 and retracts. Niceville gives (Butterflies of India, vol. iii) an excellent 

 account of two tubercles with protruding flagella, found one on each 

 side of the 8th abdominal segment of Curetis thetis. These are de- 

 scribed as two diverging cylindrical rigid pillars, arising from the sub- 

 dorsal region, and of a pale green colour. When the insect is touched 

 or alarmed, a deep maroon tentacle, as long as the right pillar, bearing 

 on its end long parti-coloured hairs (the basal third black, and the 

 upper two-thirds white), is everted. The maroon tentacle, with its long 



