82 PEOCEEDINGS OF UNITED STATES NATIONAL MUSEUM. 



and thirteen m the markets of Havana. Five appear to be still new to 

 science, or at least not yet recognizably described. 



Onr materials are not yet snfficient for a general revision of the Amer- 

 ican Scari. Very few of the many species can yet be said to be well 

 known, and the present paper is offered as a contribution toward an ex- 

 act knowledge. We give descriptions and life coloration of each of the 

 species included in this collection, with the synonymy of each, and a key 

 by which these species may be distinguished. To species not included 

 in this collection we make in general no reference. 



In our descriptions of the species we have generally avoided the repeti- 

 tion of characters common to all, as well as those which are dependent 

 on the age of the specimen and not ui^on specific difference. JNIost of 

 the differences in form and proportions of the body belong to this latter 

 class. Generally speaking, it is only the canine teeth, the scales of the 

 head, the scales of the lateral line, the caudal fin, and the coloration 

 which afford specific characters. The dorsal fin, the lower pharyngeals, 

 and the isthmus afford generic distinctions, but are essentially similar 

 in all the species of the same genus. In all the species the numbers are 

 usually D. IX, 10. A. II, 9. Scales 2^—24—6. 



Three well-marked generic groups are represented in this collection. 

 For these we should, in accordance with generally received laws of 

 nomenclature, adopt the names Bcarus, Sparisonia, and Grypiotomus. 

 These correspond respectively to Pseudoscanifi, Scarus, and Galliodon of 

 Bleeker and Giinther. 



Of these genera, Cryptotomus approaches most nearly to the typical 

 Labroids ; Sparisoma is closely allied to it, while Icarus represents a 

 considerable difterentiation. Most writers, with Dr. Giinther, have dis- 

 tinguished >S'cflrH.9and Sparisoma chiefly by the number of rows of scales 

 on the cheek, a character in itself of no systematic importance. On ac- 

 count ot the trifling value of this character, several writers (Steindach- 

 uer, Gope, Jordan, and Gilbert) have preferred to consider the groups 

 thus defined as subgenera merely. An examination of manj' species 

 leaves, however, no doubt that Sparisoma and Scarus are fully worthy 

 of generic distinction, and, unless intermediate forms occur, Cryptotomus 

 should be separated from Sparisoma. 



ANALYSIS OF GENERA. 



fl. Lower pharyngeal spoon-sliaped, mucli longer than broad; teeth of jaws fully 

 coalesced ; each jaw divided by a distinct median suture; gill-membranes 

 forming a fold across the isthmus; dorsal spines flexible; lateral line 

 interrupted, &c. ; scales about head rather numerous, lower jaw in- 

 cluded SCAUL'S, A. 



aa. Lower pharyngeals broader than long, flattish, or basin-shaped ; gill membranes 

 broadly joined to the isthmus, not forming a fold across it ; lateral line 

 subcontinuous; scales on head rather few. 

 h. Teeth chiefly coalescent, the jaws divided by a rather indistinct median suture ; 



dorsal spines pungent; lower jaw inojecting Spahisoma, B. 



blj. Teeth coalescent laterally ; the anterior chiefly separate ; the median snture not 

 evident ; dorsal spines flexible Cryptotomus, C. 



