510 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 83 



The sexes may be distinguished in most species by the presence of 

 a brood pouch in the male. Young males have at least the rudiments 

 of a brood pouch indicated by an elliptical fold of sldn or sometimes 

 by an elUptical pigmented area on the underside of the anterior caudal 

 segments, later developing into a brood pouch. The rudimentary 

 brood pouch appears when the fish is quite small, the size probably 

 depending on the species, at about 40 mm in hudsonius. 



The brood-pouch criterion was used to separate the sexes and segre- 

 gate the measurements presented in tables 2 and 3. This sex dis- 

 tinction does not always hold, since Kauther '^ found that the majority 

 of the females of brevirostris (= hippocampus) and a fair percentage of 

 the females of guitulatus also have the brood pouch developed, at 

 least in rudimentary form. The sex in europaeus as well probably 

 cannot be distinguished readily by the development of the brood 

 pouch (see p. 550). It yet remains to be determined to what extent, 

 if any, this condition occurs in other species. On account of the 

 failure of this criterion to distinguish the sexes in all cases, it would 

 have been desirable to separate the sexes more definitely by a his- 

 tological examination of the gonads. However, by a close inspec- 

 tion of my rough data, I concluded that the probable greater accuracy 

 to be attained by such an examination would result merely in showdng 

 a somewhat greater divergence or, in most cases, a lesser degree of 

 intergradation of average differences. It seems evident also that no 

 absolute distinction between the sexes may be made on the basis of 

 proportional measurements. Since the available material, when 

 sorted according to sex, size, and locahty, is not sufficient for satis- 

 factory statistical studies anyway, it was not deemed expedient to 

 spend more time on histological studies of the gonads for the mere 

 purpose of showing more accurately the average differences between 

 the sexes, or drawing more finely the lines of distinction between the 

 species. 



It is significant that in nearly every sj^ecies in which a fair number of 

 specimens were exaixdned the largest individual had a brood pouch. 

 This would indicate that in seahorses the male attains the larger size.^* 



FILAMENTS AND OTHER APPENDAGES OF THE SKIN 



The use of the presence or absence of filaments in specimens of 

 seahorses as a character in classification has caused considerable con- 

 fusion in the systematics of Hippocampus and has resulted in some 

 unnecessary synonyms. 



Specimens of seahorses are found now and then that have a pro- 

 fusion of long and branched filaments (see fig. 64). This character 



" Die Syngnathiden des Oolfos von Neapel, pp. 212-213, 1925. 



'* Gill, Proc. U. S. Nat. Mus., vol. 28, p. 809, 1905, states that the male is somewhat smaller than the 

 female. 



