BY ARTHUR M. LEA. 205 



than prothorax, each with nine or ten rows of more or less 

 elongate punctures placed in grooves ; the interstices usually 

 raised and costate, sometimes only on apical and lateral parts. 

 Pectoral canal wide and shallow, the anterior coxte more or less 

 excavated to receive rostrum, but seldom distinctly separated at 

 their bases. Mesosternal i^late more or less depressed, apex usually 

 emarginate, anterior angles more or less raised, sides incurved or 

 oblique, base usually much narrower than apex. Metastermim 

 with depressed disc and jDrecipitous sides; episterna rather narrow, 

 each with one row of punctures sometimes placed in a groove. 

 Ahdotnen with distinct sutures, two basal segments large, 1st 

 frequently slightly concave, intermediates combined usually 

 slightly longer than apical. Legs of moderate size ; femora 

 grooved beneath, strongly dentate, posterior not extending beyond 

 and seldom reaching apical abdominal segment ; tibise more or 

 less compressed, frequently with punctures running in rows and 

 giving the parts affected a grooved appearance; tarsi the length 

 of or shorter than tibiae, 1st joint moderately long, 3rd wide, 

 deeply bilobed; claw-joint long, thin, feebly or not at all pubes- 

 cent, claw\s feeble, feebly or moderately separated. Ovate or 

 elliptic, convex, sparsel}^ clothed, densely punctate, winged. 



There are many points of interest about this genus. Certain 

 species have characters which were they supported by other 

 characters might be considered as worthy of generic rank, yet as 

 there are so man}^ connecting links between species and species I 

 think that it is inadvisable to erect genera which in all probability 

 would only be degraded as their relationships became better 

 known. It would be easy to erect a dozen genera from Melan- 

 terius all founded (on paper) on sufficiently strong characters, but 

 I do not think that technical characters, however much use they 

 may be in defining genera, should be allowed to ride roughshod 

 over very obvious affinities. The peculiar abdomen of ventralis, 

 did it exist in a species in another part of the subfamily, would, 

 in all probability, cause me to think that the species was generically 

 distinct; but in Melanterius, which appears to be intermediate in 

 position between the Erirhinides and the true Cryptorhynchides, 



