125 



a "subgenus" of Paratelphusa, or rather, haviog regard to the law of priority, 

 that Paratelphusa is one of the subgenera of Gecarcinucus. 



However, accepting Gecarcinucus as a genus of equal weight with Para- 

 telphusa, then the species here under consideration- — the type of the genus 

 Ciilindrotelphusa — must be attached to Gecarcinucus rather than to Para- 

 telphusa, on account of its long, narrow, much-deflexed front, the width of 

 which only slightly exceeds that of the orbit. There are certain Indian species 

 of Paratelphusa in the very old males of which the breadth of the front falls 

 well below a fourth the gi^eatest breadth of the carapace, but even in these 

 abnormal specimens the front is very much wider than the orbit. 



In Cylindrotelphusa, as in Gecarcinucus, the epigastric and post-orbital 

 crests are low and not at all conspicuous, the post-orbital crests being approxi- 

 mated to the upper border of the orbit and following its curve ; the carapace is 

 convex and subcylindrical ; the cervical groove is very deep and broad, and 

 forms a long comparatively narrow loop, the trend of which is markedly 

 antero-posterior with a minimum of lateral deflection ; the free edge of the 

 epistome is everted — but not to such a marked extent- — and the exhalant 

 branchial orifice is subtubular — though the channelled plate of the endopodite 

 of the 1st maxillipeds is, perhaps, not quite so much seen ; and the mandibular 

 palp is identical. 



The differences that mark the subgenus Cylindrotelphusa are the following: — 



(1) The antero-lateral borders of the carapace, though thick and full, are 

 crenate, and one of the crenations is a lateral epibranchial denticle ; 



(2) The orbits, though not well seen in a dorsal view, are of the usual 

 Paratelphusa shape, and the external orbital angle is quite apparent ; 



(3) The abdomen of the adult female is broad, and ends broadly, as is 

 usual among Potamonidw. 



Another difference, to which little importance need be attached, is that 

 the carapace is not so broad ; hence the breadth of the front relative to that 

 of the carapace is greater than in typical Gecarcinucus ; but the breadth of the 

 front relative to that of the orbit is not difi"erent. 



3. Gecarcinucus (Cylindrotelphusa) steniops, Wood-Mason. (Fig. 36.) 



Telphusa steniops and microps, Wood-Mason (names only). 



Cai'apace strongly convex, subcylindrical ; its length varies very considerably 

 — from about three-fourths to nearly four-fifths its greatest breadth ; its 

 depth also varies a good deal — from a little more than half to nearly two-thirds 

 its length ; its surftice is broken near the lateral borders by some short, fine, 

 oblique ridges, which may (rarely) spread some way on to the epibranchial 

 regions ; the frontal region, again, may be rugulose ; otherwise the surface is 

 smooth. 



