REVIEW OF HIPPOCAMPUS — GINSBURG 523 



now had a species with a still longer snout from "les deux Indes"; 

 and he consequently introduces a new name, guttulatus, for the 

 French species. Although he does not definitely state so, his guttu- 

 latus must be regarded as a substitute for Schinz's longirostris on the 

 basis of available evidence, both of those names having been based 

 on the sam.e account, in the first edition of Cuvier's "Le Regne 

 Animal." Since two subspecies of long-snouted seahorses exist on the 

 coasts of France, one in the Atlantic and another in the Mediterranean, 

 it becomes necessary to restrict the name guttulatus also. It seems 

 that no previous author made this restriction, although I do not have 

 all the literature readily available for consultation. Since the name 

 guttulatus was evidently proposed as a substitute for longirostris 

 Schinz, the two names must go together. Anyway, guttulatus is 

 herewith formally restricted to the population of the common long- 

 snouted species, which occurs on the northern Mediterranean coast. 

 Cuvier's statement "museau plus long" also applies more nearly to 

 the Mediterranean seahorse, which averages a longer snout than 

 its Atlantic close relative designated below as multiannularis (see 

 table 2). Furthermore, the best and most adequate current accounts 

 of guttulatus are based largely on Mediterranean specimens, I follow 

 general usage and continue to employ Cuvier's name guttulatus for 

 that subspecies rather than Schinz's earlier navae longirostris (p. 546). 

 Cuvier's longirostris from "les deux Indes" was evidently not 

 intended to be the same as the longirostris of Schinz, although both 

 refer to Willughby's figure 4. That figure was previously restricted 

 by Cuvier (1817) to a French species for wliich Scliinz subsequentlj?" 

 proposed the name longirostris. Cuvier's later (1829) assignment of 

 the same figure to a species from "les deux Indes", therefore, must 

 be held nomenclatorially untenable, although zoologically it was an 

 appropriate emendation, the long-snouted seahorses from the Indo- 

 Pacific region having their snout more nearly as shown in Willughby's 

 figure 4. Consequently, the longirostris of Cuvier is a composite of 

 two things: (1) A figure, nomenclatorially at least, belonging to a 

 French species, and (2) a locahty belonging to a different species. 

 If we exclude the figure, longirostris of Cuvier becomes a nomen 

 nudum, and if the locality is excluded, it must be regarded nomencla- 

 torially to be the same as longirostris Schinz. Moreover, it is pre- 

 occupied by longirostris Schinz. In any case, therefore, it is unten- 

 able. The name H. longirostris Cuvier was later used for two dis- 

 tinct species of seahorses in different parts of the world, first by 

 Schlegel ^^ for a Japanese species and later by Kaup ^^ for a West 

 Indian species. The West Indian species has been renamed as a 



32 In Siebold's Fauna Japonica, Pisoes, p. 274, 1842. 



33 Catalogue of the lophobrancbiate fish in the collection of the British Museum, p. 12, 1856. 



