236 LUVAKUS IMPERIALIS. 



Mr. Regan, in a paper appearing in the Annals of Natural 

 History for October, 1902, states that it was placed by 

 Grunther in the Coryphcenidce, the typical fish of which 

 is the Bramu Rail or Ray's Bream. There are some points 

 of similarity in the external appearances of these fishes. 



Cuvier gives it a place in this family also, but Regan, 

 owing to several important agreements, arrived at the con- 

 clusion that the Luvaridce were closely allied to the Acan- 

 thuridce ; but you will see later on that he has not adhered, 

 owing to his study of our specimen, to this conclusion. The 

 points of agreement between the Ac ant I lur idee and Luvarida 

 he gives as follows : — " In both families the body is oblong 

 and compressed, the dorsal and anal fins are long, the caudal 

 peduncle slender, the caudal fin deep, the scales small, 

 rounded and usually rough, covering the head and body, 

 the lateral line is concurrent with the dorsal profile. In both 

 also the gill-membranes are broadly united to the isthmus. 

 There are four gills, with a slit behind the fourth, five branchio 

 stegals, well developed pseudo-branchioe and short gill rakers. 

 The mouth is small, the premaxillaries are not protractile 

 and the maxillaries are attached to them and not indepen- 

 dently moveable. The toothless palate, the palatine arch 

 attached only to the pre-ethmoid, the coalescent pelvic bones, 

 are further points of agreement between the two families." He 

 goes on to say : " Luvarus, like the Acanthuridce, is a vege- 

 table feeder and exactly resembles them in its visceral anatomy. 

 The stomach is large and thick-walled, the pyloric appendages 

 short, simple and few in number, and the intestine very long 

 and much coiled ; the air bladder is large." Mr. Regan says 

 that " the skeleton of Luvarus resembles the AcanthuridcB in 

 many features." 



One point of disagreement Mr. Regan speaks of. He 

 says : " Apparently the remarkable feature of the loss of the 

 anterior rays of the dorsal and anal fins during growth is 

 peculiar to Luvarus and not paralleled in the Acanthuridce" 



In a second note on the " Skeleton and systematic position 

 of Luvarus" published in the same magazine for April, 1903, 

 and after a study of the specimen now in our Museum, Mr. 

 Regan gives some interesting details for which see the original 

 papers. 



I have called attention to Mr. Regan's notes because it 

 is only by reference to them that an accurate knowledge of 

 the classification of the fish can be obtained, and of the reasons 

 why it is that, although Mr. Regan and others had placed 

 the fish in the families named, it is now moved to the Scomhridce. 



