occurrinf/ on tJie Xorioegian Coas-ffi. H89 



to this genus a new species, (Edicerus nomzealandi<x^ and 

 tliouglit that some of the characters Avhich Kroyer ascribed 

 to the genus were of little value as generic characters. To 

 these species Bruzelius has added two, and Sars one, to which 

 a sixth species is now added. Thus so many species are 

 now brought together that the proposed generic characters 

 require a fresh investigation. These Sjjecies may be divided 

 into two distinct sections, — one of which includes the species 

 soginatus^ Kr., affinis^ Bruz., lynceus^ Sars, and norvegicuSj 

 mihi, the other novizealandu^^ Dana, and obtusus, Bruz. 

 Dana, iindino- that his CE. novizealayidice acrreed in some re- 

 spects with LE. saginatus, Kr., especially in the considerable 

 length of the last pair of thoracic legs, by which it was sepa- 

 rated from the genus Ipliimedia as he conceived it, assumed 

 therefore that it must belong to the genus (Edicerus^ Kr. ; but 

 it, as Avell as (E. obtusus^ Bruz., of which I have had examples 

 from Finmark to examine, differs much from the typical spe- 

 cies. Thus : — In these two species the superior antenna are 

 elongated, whereas in the other they are very short, only a 

 little or scarcely longer than the peduncle of the inferior 

 antennre. The head does not project in a long and strong 

 rosti'um in Avhich the eyes are placed so close together as to 

 look like a single organ, but the rostrum is wanting, and the 

 eyes are, as usual, placed on the sides of the head. The apices 

 of the mandibles are not toothed ; and their palpi, which in 

 species of the same genus are of the same structiu-e, differ in 

 the form of the second joint from those of (E. saginatus. The 

 inner lamella of the first pair of maxilla3 is large and furnished 

 with several strongly ciliated hairs, whilst in those of GE. sagi- 

 natus &c. there is only a single seta. There is therefore suffi- 

 cient ground for separating these two forms from each other, 

 especially as each of them includes several species. I have 

 therefore set up CE. obtusus, Bruz., as the type of a new genus, 

 Aceros, mihi, to which I also refer CE. novizealandice, Dana. 

 These tAvo species differ from each other in the length of the 

 peduncle of the superior antennae ; for this, in Dana's species, 

 is short, Avith a long liagellum, and in CE". obtusus lone;, Avith 

 the ilagellum short. Lastly, Avith regard to the place of this 

 genus in the system, Dana has placed it in the family Gam- 

 maridffi, as he only kncAv his species Avith long antenna? ; but 

 I think that it must go with the preceding genera, as the form 

 of the ovigerous lamella? and their relation to the respiratory 

 plates are the same as in these, Avhereas those of the genera 

 resembling Gammarus are of a different form. In the stnic- 

 ture of tlie hands on the first tAvo pairs of legs the genus is 

 related to the subfamily Leucothoinae. 



