152 Transactions. 



Pentarthriwi may be accepted as the typical genus of our Cossoniclce, 

 and to it, by European entomologists as well as by myself, over forty of our 

 endemic species have been ascribed within the past thirty or forty years. 

 These I have, so far as specimens were available, carefully studied, separated 

 into groups or series, and detached first one and then another as representa- 

 tives of new or distinct genera, until at length I had instituted seventeen 

 new genera for their systematic location, irrespective of thirteen I had 

 previously described. Only fifteen species have been left under the name 

 Pcntarthrum, and these are divided into three sections, so that the identifica- 

 tion of the older species has been much simplified. 



Few are aware of the trouble involved in that process. Every specimen 

 in my collection was carefully set out and gummed to strips of cardboard 

 or mica slabs, and after the superficial examination of generic characters 

 had been completed it was necessary to remove each one from its original 

 position, so that the structure and sculpture of the under-surface might be 

 as carefully studied as that of the upper before the description of any genus 

 could be properly prepared. This removal from the cardboard, and the 

 subsequent replacement of the specimens in my cabinet, was a very tedious, 

 delicate, and risky operation, as the least carelessness would result in the 

 damage or loss of the type of some species which may not be found again 

 during my lifetime. Notwithstanding all the care bestowed, I managed 

 to lose one, much to my regret. 



On reference to the following list, it will be seen that our CossonidcB 

 now consist of forty-seven genera, comprising 141 species, the descriptions 

 of which are attached hereto. 



It must, however, be understood that the object in view was not to 

 monograph the species, although each one has been more or less critically 

 reviewed, and placed, in accordance with my own views, in its proper posi- 

 tion. Every genus and its typical species have been fully described by 

 myself, and two or three diagnostic characters of the species attached to each 

 genus have been given as an aid to identification, together with references 

 to the original descriptions. The maximum measurements are given ; 

 that of the length invariably includes the rostrum of all properly mounted 

 specimens. 



During the progress of the work I found that, for the convenience of 

 colonial entomologists, it would be expedient to adopt some fixed standard 

 for purposes of comparison. Pentarthrum zealandicum, var. canaliculatiwi, 

 has been selected, and placed at the head of the list, because it is our typical 

 exponent of the Pentarthrides, and also on account of its being readily pro- 

 curable in the South as well as the North Island. Its generic description 

 has been renewed and somewhat restricted, the term " more or less elon'gate,'' 

 appUed to the rostrum in the original diagnosis, having been deleted, so that 

 only such characters as are exhibited by the endemic species itself are 

 given. 



Those who have no special knowledge of our Cossonidce should, before 

 attempting to classify and name their collections, carefully study the general 

 form, appearance, and structural details of the insect 1 have chosen as the 

 standard, the female as well as the male. To do this thoroughly four speci- 

 mens should be obtained, and one of each sex neatly set out and mounted 

 on cardboard with the antennae, legs, and tarsi properly displayed, and 

 another of each sex mounted on its back, so that differences of the under- 

 side can be examined. Euophryum or Torostoma may be shidied next, 

 noting and comparing the peculiar modifications of the rostrum and elytral 



