1901.] CRUSTACEA. OF THE " SKEAT EXPEDITION'." 561 



but that Tale Sap, though quite or nearly sweet at its head, increases 

 gradually in salinity towards the mouth, where it is quite salt. In 

 view of the differences between this species and Car. gracilirostris 

 it is peculiarly unfortunate that the exact record or" the locality 

 should have become, by mischance, obliterated. 



The only distinction between this species and Oar. grn>ciUrostris 

 is found in the rostrum (and, iu a small degree, in the telson and 

 1st two pairs of legs). The structure and relations of antennules, 

 antenna?, maxillipedes, and legs are the same as in C. gracilirostris, 

 save only that the 1st pair reach, not to the end of the penultimate 

 joint of the antenna?, but barely to the middle of that joint, and 

 similarly the 2nd pair reach barely beyond the end of the penulti- 

 mate joint instead of nearly to the end of the last joint. The 

 telson again, though exactly similar in all other points, presents 

 only four pairs of teeth on the upper surface, not five or six. 



The rostrum, however, is markedly different. Dr. Xobili has 

 recently described (Ann. Mus. Civ. Stor. Nat. Genov. xx. p. 477, 

 1900) a new species, 0. modiglianii, and the rostrum in our species 

 appears intermediate between those of this latter species and 

 0. gracilirostris. The number of the teeth on the upper margin 

 is nearly the same as iu 0. gracilirostris, viz. 6-9; these teeth are, 

 however, not nearly so widely separate distally, though the most 

 distal tooth is sometimes placed at a considerable distance from 

 the rest. At the same time the interspaces do increase distally, 

 probably more than in 0. modiglianii ( ii sono Jitti e avvicinati nella 

 i :iia. nuova specie''). The number of the teeth on the lower edge 

 approaches that of Dr. Nobili's species, ranging from 12-23 as 

 extremes, 16-18, however, being the most common number; in 

 Dr. Nobili's single specimen they numbered 17. As in his species, 

 too, the proximal teeth are better defined than the distal. The ros- 

 trum also agrees with those in both these species in being slender 

 and upwardly curved towards the tip ; the amount of this curve, 

 however, varies from nearly straight to a strong upward bend. 

 The tip bears mostly a small tooth that gives it a bifid appearance ; 

 in a few cases, however, there are two such teeth, making it trifid. 



But it is in the length of the rostrum that this variety shows its 

 distinctness, as the following table will show : — 



Rostrum. Carapace. 



nun. mm. 



( '. gracilirostris 8 4 



0. modiglianii 7 4 



C. gracillima 3-6 4 



That is, although in a few cases it exceed.- the carapace by half 

 its lengtb only (instead of being double, or nearly so, tho length 

 of the carapace), it generally exceeds il by only one-quarter, or 

 equals it; while iii a few cases yei it even falls short of it. I give 

 measurements (p. 562), taken from twenty-one individuals picked 

 nut at random; together with the number of rostral teeth. 



These resemblances and differences are, within the limits of 



