TO INDIAA' CAliCmOLOGY. 377 



hand is pei't'ectly siuootli aucl glabrous. Tlie posterior surface of the three pairs of 

 ambulatory legs is smooth. Tlie peuultiuiate segment of the male abdomen is not 

 specially dilated. A SaccuUiia is frequently attached to the abdomen, and Portaiticepon 

 Heiidersouu, Giard and Bonnier, occurs in the l)ranchial chamber. 



G. A-.VRIEGATUM, var. CALLIAXASSA (Herl)st). 



In the specimens which I refer to tliis variety, which also is common at l^^adras, 

 the median frontal projections are less rounded, and the transverse ridges of the carapace 

 are more strongly developed, especially two on the cardiac area, and there are two 

 parallel lines on the branchial area. Tiui merus of the chelipedes has usually three 

 spines on the anterior margin ; there are four spines on the upper surface of the hand, 

 and strongly granulated ridges on both the inner and the outer surfaces of the hand. 

 The whole surface of the chelipedes is more or less strigose, but more particularly the 

 under surface of the liand. Longitudinal pubescent lines are met with on the posterior 

 surface of the ambulatory legs. The penultimate joint of the male abdomen is so dilated 

 as almost to form part of a circle in outline. The specimens are almost certainly 

 identical with that figured by Herbst (Naturgesch. Krabben u. Krebse, pi. liv. fig. 7) as 

 Cancer caUianassa ; at the same time they are probably the same as that figured by 

 De Haan as PoHunus {CJiarybdis) variegatns, Eabr. (Crust. Japon, tab. i. fig. 2). 



I have examined a large series of both forms from Madras, and as a rule any specimen 

 can be determined at once by the characters I have enumerated for each variety. 

 I have met Avith a few specimens, however, in connexion with which some diificulty 

 is experienced, and in Avliich there appears to be an admixture of the two sets of 

 characters. 



The var. bimacidutum, Miers, taken by the ' Challenger ' at Japan, is, I think, perhaps 

 a distinct species ; its fi'ontal teeth are quite different in form and very obtuse ; if it is 

 really a variety of G. carienatiini,, there is an extraordinary range of variation in this 

 species. 



Bistrihutioii. Malay Archipelago, China, Japan, N. Australia. (It is impossible to 

 say whicli variety is referred to in regard to the localities assigned 1o this species.) 



11!). GOXIOSOMA ROSTRATUM, A. Miliic-Edw. 



(r. rostral inn, A. Mihie-Edwards, Nouv. Arcli. .Mas. Hist. Nat. t. x. p. ;37!), pi. xxw. fig. .'i (1861). 



Sunderbunds and Calcutta, several specimens {Day); Gulf of Martabau, eight 

 specimens {Gales). 



This species is distinguished by the general form of its front, and especially by tlie 

 great prominence of the median frontal teeth, which are obtusely rounded. In all the 

 above-recorded specimens the last lateral spine of the carapace is considerably larger 

 than those preceding it; but, judging Irom a larger and apparently full-grown specimen 

 from the Hoogly, in tlie British Museum, this s])ine diminishes witli age, for in this 

 example it is scarcely larger than those in front of it. 



