520 Scientific Proceedings^ Royal Dahlin Society. 



entirely ventral to the ninth. Small and narrow, it projects ventrally and 

 slightly posteriorly to assist in locomotion, for which it is capable of being 

 protruded and reti-acted. The armature consists of a transverse series of 

 eight minute bristles near the apical border, two being posterior and the 

 others anterior. The anus opens on the apex as a hollow slit. Unless in the 

 region of the bristle-bearing and other chitinised plates, the cuticle of the 

 body thronghont is soft, with the surface raised into numerous, rounded, 

 evenly distributed, wart-like tubercles (PI. XVII, fig. 22). These are most 

 apparent at the junction of two segments, i.e., in the intersegmental 

 areas. 



(vi) The Spiracles. — There are eight pairs of abdominal spiracles, all 

 similar in structure, and placed latero-ventrally to the alar lobes. The 

 circular periti'ene is chitinised and slightly larger on the thorax and eighth 

 abdominal segments. Surface examination shows each spiracle as a round 

 hole, surrounded by a yellow band of chitin. This hole leads into a cup- 

 shaped vestibule, having a small opening at the bottom. This is succeeded 

 by a short narrow tube, having unstrengthened walls, and provided at the 

 base with a long and a shorter arm, which project therefrom. Below these 

 arms is a constricted area, from which the trachea proper arises. This peculiar 

 arrangement in the spiracle was observed by Sanderson (56), but not under- 

 stood. Diebel (17) and lioving (6) have since worked out the anatomy and 

 mechanism of similarly constructed spiracles in other Chrysomelid larvae, 

 and have shown that the two conspicuous arms which Sanderson considered 

 as tracheal appendages belong to the spiracle, and, being provided with 

 muscles, serve as regulators of the air-supply to the trachea. The writer has 

 not yet had an opportunity of studying the internal anatomy of the spiracle 

 in the larva of L. parvuhis, but is satisfied that it is an essentially similar 

 organ to that described for the other Chrysomelid larvae referred to above. 



(vii) Specific description of the bristles. — In 1901 Sanderson (56) proposed 

 a system whereby each bristle-carrying area received a number. "Woods (64) 

 followed Sanderson's idea, with modifications and additions in specific de- 

 scriptions of species. The present writer has not adopted either, realizing 

 that both are imperfect, and that it still remains for some worker on the 

 larvae of many genera of Chrysomelidae to evolve a satisfactory scheme to 

 include all possible variations. For example, both systems referred to pro- 

 vide only for larvae whose segments carry two transverse rows of bristles, 

 whereas the larva of L. parvulus has three such rows, as already described. 

 Moreover, a reference to a particular area or bristle by means of a number 

 alone is both ineffective and cumbersome, and the system of MacGillivray (47), 

 whereby bristles are referred to with respect to the particular area of the 



