REVIEW OF HIPPOCAMPUS — GINSBURG 511 



gives tliem such a strildng appearance that it is hard to conceive 

 of it as not being of importance in classification. Some species, 

 indeed, have been based in large part on this character, as ramulosus 

 (p, 518) and kincaidi (p. 569). My study has shown unmistakably, 

 however, that among seahorses this character is largely due to individual 

 variation, a conclusion to which other investigators arrived at pre- 

 viously (see p. 518 for definite examples cited). To some extent it is 

 an age character — that is, filaments are oftener present in the smaller 

 size groups, at least in the specimens examined; but in either young 

 or full-grown specimens they may be indiscriminately present or 

 absent. To a certain limited extent it possibly is a species character 

 in the sense that in some species specimens with filaments or with a 

 profusion of filaments apparently are oftener present, while in other 

 species specimens mth filaments are comparatively infrequent or the 

 number of filaments, when present, is few. However, this slight 

 specific difference may be discerned only after the species have been 

 separated by other means. Taken by itself this character is of little 

 value in classification. Indeed, in the manner in which it has been 

 used, it has had a rather negative value. 



In the species I examined, filaments on the tubercles as an individual 

 variation were found in all except H. europaeus, but the eight available 

 specimens of this species are in indifferent condition. In hildehrandi 

 and obtusus there are a few short chunky appendages instead of 

 slender filaments, and such appendages were observed in occasional 

 specimens of other species also. As a result of my studies of the 

 seahorses and a consideration of descriptions and figures in the litera- 

 ture, it is evident that the presence or absence of filaments or other 

 appendages on the tubercles is highly variable. They are probably 

 present in all or nearly all the species, in some specimens at least. 



In addition to the filaments or fleshy appendages on the tubercles, 

 seahorses are often found with pimplelike excrescences scattered over 

 the skin, or with many very short filaments on the surface of the 

 skin. These structures apparently also differ with age, with the 

 individual, and to a limited extent with the species, in the same 

 manner as do the filaments on the tubercles. The apparent develop- 

 ment of these small excrescences probably depends also on the state of 

 preservation of the specimens. (See further discussion under the 

 account of hudsonius, p. 555.) 



GEOGRAPHICAL DISTRIBUTION 



In regard to the geograpliical distribution of the seahorses, Giin- 

 ther '^ makes the follo\ving statement: "They are pelagic fishes wliich 



IB Catalogue of the fishes of the British Museum, vol. 8, p. 198, 1870. 



