266 U.S. NATIONAL MUSEUM BULLETIN 2 02 Vol. 2 



no accessory pelvic scale; scales cycloid; vomer with a triangular 

 patch of teeth, a few tiny teeth on front of palatines in largest specimen, 

 absent in smallest; a band of villiform teeth in upper jaw, broadest 

 forward, with an outer row of conical caninelike teeth, and each side 

 of symphysis an enlarged canine tooth; lower jaw with a narrow band 

 of villiform teeth forward only, an outer row of enlarged but short 

 canine teeth; pelvic insertion behind pectoral base, opposite base of 

 fourth or fifth dorsal spine; bases of soft doreal, anal, caudal, and 

 pectoral fins with scales; third lower simple pectoral ray longest; 

 rear margin of caudal fin truncate or slightly rounded; preopercular 

 margin finely denticulate, becoming rough in largest specimen; 

 interorbital space flattish in j^oung, a little concave in adults, and 

 scaled. 



Color in alcohol. — Background color whitish or pale, upper half of 

 body a little darker than lower half; along midaxis of body from 

 behind head a wide white baud, becoming faint and disappearing on 

 caudal peduncle; upper half of body with numerous roundish black 

 spots and below white band only a few indistinct ones; caudal white, 

 edged above and below by a black line; mid ventral part of gill mem- 

 branes blacldsh; dark spots lacking on head; lower half of head 

 white, upper park dusky. 



Color when alive. — Upper half of body and of head light brownish, 

 more dusky forward; spots black; lower half of body wdth about 5 

 orange-yellow lengthwise streaks; all fins orange-yellow; snout and 

 pale streak below eye pinkish; lower lip pink; pale lateral band pink. 



Ecology. — Tliis species occured among coral heads where there was 

 considerable wave action. 



Remarks. — Marsliall (Bull. Raffles Mus., Singapore, No. 22, pp. 

 183-4, 1950) suggested that P. hemistictus and P. polystictus are the 

 same species. I have been able to determine the sex of 5 of our 

 specimens; of these, the 3 with the color pattern of hemistictus are 

 females, and the 2 with the color pattern of polystictus, as illustrated 

 here (plate 111,C), are males. I therefore conclude that polystictus 

 is the male of he7nistictus. 



Family MUGILOIDIDAE 

 By Leonard P. Schultz 



I have spent considerable time investigating what genera of fishes 

 should be referred to this family and have concluded that it is not 

 possible to determine with certainty until further osteological investi- 

 gations are made. The cliaracter by which the Mugiloididac are 

 distinguished from the Branchiostegidae (= Latilidae) is the position 

 of the pelvic fins. In the Mugiloididac the pelvics are supposed to 



