N0.1235. LIST OF JAPANESE FISHES— JORDAN AND SNYDER. 753 



From near the original locality (Okinawa) avc have also a single 

 specimen received from the Imperial University of Tokyo. 



These specimens differ in color and in the depth of the body. 



The Okinawa specimen (in spirits) has the g-reater part of the body 

 l)right chocolate brown, without bands or stripes; lighter below and in 

 the region of the pectoral tins. The Shimoda specimens have the body 

 of a pale yellowish brown color, with three light lateral bands extend- 

 ing along the sides; wider apart and broader anteriorly, converging 

 and becoming narrower on the caudal peduncle. Many of the scales 

 of the body have each a small light spot. In each case the fore |)art 

 of the head is of the same general color as the body. The width of 

 the vertical l)and of l)lue varies somewhat in each indi\idual. 



The depth of the tirst-mentioned specimen is .53 of the total length; 

 the scales between the lateral line and insertion of the dorsal are in 

 6 series; between the lateral line and the anal. IH. In the larger 

 of the Shimoda specimens, which is of equal length with the one 

 from Okinawa, the depth measures .oO; the scales number 7-20. The 

 smaller one measures, depth, .60; scales, 7-20. The scales in the 

 lateral line of the three number, respectiyely, 40, is. aiid 47 scales. 

 The tin ra3\s are as follows: Okinawa specimen D. IX. !'.>; A. II, 15; 

 Shimoda examples D. IX, 19; A. II, 14, and D. IX, 17; A. II, 14. 



Believing that these differences, though considerable, are of such a 

 nature that a large series of specimens would show them to be merely 

 individual variations, we do not deem it advisable to record the 

 examples at hand as belonging to two ditferent species. 



41. POMACENTRUS TRILINEATUS Bleeker. 



{Pu)iinri'ntrns tlorKdIw (.Till.) 



Shimoda, J. Morrow: the original tvpe of PnnnKuitf rn^ durxdlix 

 Gill. 



Dr. Bleeker regards I^oinuccDtrKK dorxaHxii^ probal)ly identical with 

 Pomacentrus trilmeatus from the East Indies. We are indeed unable 

 to detect any difference between Gill's type from Shimoda and those 

 two of Dr. Bleeker ;s figures' which correspond nearest to it in stage 

 of development, showing two white bands on the anal, the blue dots 

 on the head and the black dt)rsal ocellus, preceded by white, except 

 that the body in the Shimoda specimen is a very little deeper, the 

 depth 2 in length. This species belongs with the preceding and the 

 next to the fauna of the rock pools flooded b}^ the Kuro Shiwo. 

 Except GilPs type no second specimen has been taken in Japan. This 

 species having the teeth angulate at the tip, and in a single row. is a 

 Parapomacentrns in Bleeker's arrangement. Bleeker says that the 

 teeth are biserial, which would place it in his division Pomacentrus. 

 We find but one row. 



' Atlan Ichth., pi. ccrvi, liyrs. 1 ami 2. 

 Proc. N, M. vol. xxiii 4<s 



