•22 DR. W. T. CALMAN ON A COLLECTION OF 



Two males, 7'5 and 11 mm. long respectively, agi*ee witli Miers's type specimen in the 

 shape of the frontal lobes and in having only one spine on the posterior edge of the arm. 

 The antcro-lateral teeth, however, are not distinctly more spiniform, nor is the last tooth 

 longer than in specimens of A. grmiulatus of similar size. The second spine of the 

 posterior edge of the arm is represented by a slight rudiment, as it is, indeed, in the type 

 specimen. I do not think that the form can be ranked as more than a variety of 

 A. granulafus, as it was originally regarded by Miers. 



Localities. " Sabai Channel " ; " Murray Island.'' 



Thalamita prtmna (Herbst). 



The forms of Thalamita in which the front is divided into eight lobes were distributed 

 by A. Milne-Edwards among seven species and reunited by Kossmann into one, while 

 more recent writers haA^e expressed various views intermediate between these two 

 extremes. Alcock has recently affirmed his belief in the correctness of Kossmann's view, 

 while retaining, for the sake of convenience, sei^arate specific names for some of the 

 forms. Ten specimens collected by Prof. Haddon belong to this section of the genus and 

 fall into three groups, not one of w^hich agrees in all points with any of the described 

 species : — 



{a) A large male, the carapace of which measures 37 mm. in length by 59 mm. in 

 breadth, agrees best on the whole wdth the descriptions of the typical Th. prymna, but 

 presents certain points of diff'erence. Comparing the frontal lobes wdth the figure given 

 by DeMan (Journ. Linn. Soc, Zool. xxii. pi. iv. fig. 5), the outer or fourth pair are much 

 more strongly arcuate and resemble the figure of Th. spinimana (I. c. fig. 7) ; the third 

 pair of frontal lobes are separated by an open fissure from the second or submedian, as 

 in Dana's figure of Th. crassimana (U.S. Expl. Exp., Crust, pi. xvii. fig. 9«) ; the sub- 

 median are slightly less prominent than the median lobes, which they distinctly ovei'lap 

 above, an arrangement which, according to A. Milne-Edwards (Arch. Mus. Paris, x. 

 p. 362), "ne se voit jamais chez le Th. prymna." De Man's description and figure of 

 Th. Cfpruleipes (Zool. Jahrb. Syst. viii. p. 568, pi. xiv. fig. 12 a) fits this specimen well as 

 regards the third and fourth frontal lobes, but the median pair are stated to be wider 

 than the submedian, while in the present instance the reverse is the case. The basal 

 antennal joint carries a row of about four sharp spines besides some smaller granules. 

 Milne-Edwards assigns to it only two or three spines, while Dana figures an irregularly 

 toothed crest. The fourth antero-lateral tooth is very small, and the greatest breadth of 

 the carapace is measured between the third pair of teeth. The cheliped differs from all 

 descriptions of Th. prymna in having three spines instead of two on the upper margin of 

 the hand, the additional spine being smaller than the other two and close to the 

 proximal end. 



{h) A male specimen, 21 mm. long by 33 mm. broad, has a row of granules on the 

 basal antennal joint and a minute fourth antero-lateral tooth, and would therefore be 

 referred by Milne-Edwards's table (/. c. p. 367) to Th. Stimpsoni, which De Man regards 

 as a variety of Th. Bancv (Journ. Linn. Soc, Zool. xxii. p. 78). With De Man's figure 

 of Th. Dante {I. c. pi. iv. fig. 8) our specimen agrees in the nearly strtiight anterior edge 



