1 84 FAUNA HAWAIIENSIS 



has undergone total loss of thoracic setae, and but very few that have lost even one of 

 the pairs. 



I have made use of the characters of prothoracic chaetotaxy for the purposes of 

 arranging the species in genera. As I have examined about 6500 individuals and as 

 this is the first time that these structures have been used to any great extent for taxo- 

 nomical purposes 1 shall be pardoned for discussing this subject at some length. These 

 setae are very easily removed by rough handling, and specimens in collections are often 

 deprived of them. Each seta springs however from a pit extending to some depth in 

 the chitin and this always remains. When such a pit exists we are justified in treating 

 the specimen as one that had a seta there^. The pits are however rendered obscure in 

 certain cases by both artificial and natural causes. Sometimes they are filled up by dirt 

 — and this dirt may be either extraneous, or an excretion from the insect itself. The 

 natural punctuation, or sculpture, of the surface is sometimes of a nature that makes 

 it difficult to perceive the pit. In some cases, too, a decision as to the existence of 

 the pit is rendered difficult by its situation ; it may be on the side of a raised margin, 

 and this may prevent the light from falling on it. With a little care and experience all 

 these difficulties disappear. 



There are however facts that to some extent invalidate the utility of this 

 character. In some cases the pit and seta are present on one side of the thorax and 

 not on the other. This however occurs but rarely. A similar anomaly occurs in some 

 details of the nervuration of the wings of other insects, but it is found not to seriously 

 affect the value of the character for ta.xonomic purposes, for it is believed that the anomaly 

 only occurs on both sides as a very rare exception. This is certainly the case as regards 

 the pit and seta. I have borne this in mind and am convinced that only in two or three 

 cases is it probable that the complete anomaly occurs. These I shall subsequently 

 mention as they are of extreme interest. 



When, in the Hawaiian Carabidae, the pit and seta are present on one side only, 

 the individual may be safely treated as one of a species in which the seta is present. 

 This departure from bilateral symmetry is in fact a deficiency ; it is not a sport, by 

 excess, of a species having no seta". It is recorded that in certain cases a seta may be 

 in duplicate ; that is to say, instead of there being one pit and one seta in a certain 

 situation, there are two in close proximity. I have only observed one such anomaly in 

 our Hawaiian Carabidae ; it occurs in a specimen of Thriscothorax undus, one individual 

 of which has the setae doubled on one side of the thorax. 



Under these circumstances I have found it safe to rely on the seta for discriminative 

 purposes. 



In the cases of Mccyclothorax robustiis and Atrachycnemis sharpi I have acted 



This may not be absolutely true, as will be seen by my subsequent remarks as to Atelothnts fransiens. 

 This, again, may not be absolutely true; cf. what is said as to Mdromeniis pavidus, and Mesothrisais 

 inincatus. 



