REVIEW OF HIPPOCAMPUS — GINSBURG 509 



In taking measurements, and also in counting the caudal segments, 

 it is important to note whether the tail is broken off at the tip. Some- 

 times one or more caudal segments have been broken off at the tip. 

 Usuali}^ a broken tail may be discerned readily, but in infrequent 

 cases it has been almost perfectly regenerated and may be detected 

 only by close scrutiny, by the unnatural stumpy appearance of the 

 tip of the tail. In such cases proportional measurements, with the 

 length used as the standard, are evidently of no value or may be 

 even misleading. Consequently, measurements and the caudal 

 segment count of such specimens have been omitted from the data 

 here presented. 



Fi7i rays. — An absolutely accurate count of rays in the dorsal and 

 pectoral fins is essential, but this count is not likely to vary with 

 different observers. Sometimes the last two, and less often the first 

 two, dorsal rays are closely approximated. For this study such rays 

 were enumerated separately. The pectoral rays were always enu- 

 merated on the right side, unless the fin was broken on that side, and 

 no account was taken of any occasional probable differences in the 

 numbers of both sides. 



DIFFERENCES DUE TO SEX AND AGE 



In studying seahorses it is very important to take account of the 

 size and sex of the specimens, especially when one attempts to estab- 

 lish a new species. In general, younger fish of either sex differ from 

 older individuals in having better-developed tubercles or spines, a 

 liigher coronet, a longer snout, and a slenderer trunk. The tail is 

 somewhat shorter and the trunk longer, but these two differences 

 are not so well marked as the preceding ones. The females differ 

 from the males in the same way as the young of either sex differ from 

 grown specimens, but the differences between the sexes are generally 

 not so pronounced as those due to size. A study of tables 2 and 3 

 (pp. 531-533) shows these differences to hold, except in a few cases 

 possibly due to the small number of specimens measured, in con- 

 sequence of which extreme variants are likely to have a greater effect 

 on the averages. The exceptions may also be due to the sexes not 

 having been completely separated, as hereafter discussed, or to the 

 inherent difficulty of taking very acciirate measurements of seahorses. 



Three of these differences — the relative development of the tuber- 

 cles, the length of the snout, and the depth of the trunk — are also of 

 importance in separating some of the species. It is evident, therefore, 

 that size especially must be considered when identifying and dis- 

 tinguishing seahorses. It is also of considerable, though lesser, im- 

 portance to compare specimens of the same sex. 



