i88 Memoirs of the Indian Museum. [Voi,. Ill, 



importance, are the mesothoracic episterna, which are also triangular. A band of 

 strong punctures often extends obliquely across the polished surface of these plates, 

 from about the middle of the upper margin to the middle of the anterior margin, 

 which it follows down to the lower angle. The upper angle in front of this band 

 is smooth and polished, and the posterior angle behind it matt. This pattern 

 seems to form the basis of the patterns of all species, such variations as occur 

 being confined above to the extension of the punctured band forwards to cover the 

 upper angles, and below to changes in its width and in the extent of the dulled area 

 in the posterior angles. These variations show a considerable degree of constancy 

 in individual species, provided that the pronotum can be pressed forwards sufficiently 

 to ensure that the whole of the episternum is properly exposed ; but in rigid speci- 

 mens they are apt to be misleading. 



The mesothoracic epimera are small plates which do not appear to be of any 

 taxonomic importance. 



The mesQsternum lies between the two episterna of the mesothorax. It is 

 bounded behind by a transverse groove, which extends right across the ventral surface 

 of the body a Httle in front of the place at which the middle pair of legs is inserted; 

 it is roughly triangular, but the apex of the triangle is flattened, and forms the 

 short anterior margin, which is finely striato-punctate and pilose to an extent fully as 

 variable as that of the scutellum. The surface in the lateral angles is often matt; 

 that of the middle part of the plate is usually polished. Either or both of these 

 parts of the plate may be punctured, and such puncturing often affords a useful 

 guide to the identification of a specimen. The central part often bears strongly 

 marked grooves, keels or hollows, to which considerable importance has been attached 

 by previous authors. I have never found these characters to be of any taxono- 

 mic value, and in some species (e.g. Macrolinus andamanensis) the series before me 

 proves conclusively that they have absolutely none. The mesostevnal scars are a pair 

 of depressions situated close to the lateral margins of the plate. These depressions 

 are also more variable than has been supposed, but they sometimes exhibit 

 unusually well-marked characters of undoubted value (e.g. in Leptaulax anipunctus). 

 They are present in nearly all species of the second section of the family, but are not 

 found in the first section 



The homologies of the T-shaped or I-shaped structure, situated behind the trans- 

 verse groove which I have taken to mark the posterior margin of the mesosternum, are 

 somewhat obscure. The antero-lateral parts seem to be as fully divided from the 

 median part as they are from the mesosternum and perhaps represent the trochan- 

 tine or subcoxa, though they are more firmly fused to the adjoining thoracic plates 

 than to the base of the leg. They are of no taxonomic importance. The median part is 

 probably a superficial portion of the mesosternellum or medifurca. In the Aulacocy- 

 clinae it usually appears to be I-shaped, with a complete suture across the middle-line 

 behind it and a more or less obscure transverse roughened patch across the narrow 

 piece between the bases of the legs. In the genus Cevacupes, however, the transverse 

 piece behind this patch is usually found to be more or less completely fused to the 



