12 U.S. NATIONAL MUSEUM BULLETIN 259 



sider the system of numbering pereopods to include the gnathopods 

 (thus, pereopods 1-7) a more workable system. The spelling of 

 "pereopod" has been adopted in preference to "peraeopod" or "pereio- 

 pod." Although the latter two spellings are widely used, the former 

 is in keeping with the recent trend among a number of American 

 carcinologists to eliminate superfluous diphthongs wherever possible. 

 (5) "Brood plates" has been adopted in preference to either "mar- 

 supial plates" or "oostegites" to designate the four pairs of ventral 

 plates characteristic of the female. (6) Segments of the pereopods 

 are usually referred to by number with the following exceptions: 

 dactyl(s) for segment 7, propod(s) for segment 6, and basis(es) for 

 the expanded second segments of pereopods 5, 6, and 7. (7) "Sternal 

 gills" (or processes) have generally been referred to as gills (in the 

 sense of respiration) in this paper, although there may be some 

 question as to the exact function of these structures. When present, 

 sternal gills occm' as follows: one, two, or three single, slender proc- 

 esses on pereonites 2, 3, and 4; a single pair of slender processes on 

 the first pleonite; and two pairs of laterally placed, simple or bifur- 

 cate processes on pereonites 6 and 7. (8) "Abdominal side plates" 

 have been also occasionally referred to as "epimeral plates" or simply 

 "epimera" by other investigators. They occur as paired, lateral 

 plates or pleurites of pleonites 1, 2, and 3 and are fused to the body 

 dorsally but generally free posteriorly and ventrally. (9) "Rastellate 

 setae" — this term is introduced and defined here to designate the 

 conspicuous, bipectinate, clavate setae which often occur in single 

 sets on the posterior margin of segment 5 of one or both gnathopods. 

 "Rastellate" is derived from the Latin rastellus which means rake. 

 This structure is so named because of the double row of distal tines 

 which give it the overall appearance of a rake or brush. 



Taxonomy 



Fresh-water amphipods have been regarded by some workers as 

 taxonomically difficult (see especially Bousfield, 1958), but it was 

 generally concluded from working with Stygonectes, that, despite the 

 lack of taxonomically usable genitalia, more than enough diagnostic 

 characters were usually available for providing morphological dif- 

 ferences between closely allied species. A number of problems were 

 encountered, however, and these are discussed in some detail below. 



Structural differences and similarities in the gnathopods, pereopods, 

 and abdominal side plates were especially useful criteria in erecting 

 species groups. In differentiating individual species the two most 

 consistently reliable characters were the gnathopodal propods and 

 the telson. In general, however, almost all of the external body 



