REVIEW OF HIPPOCAMPUS — GINSBURG 571 



of head, often very dark all over and definite spots hardly evident; 

 minute white dots present or absent, often coalescing to form irregular 

 lines or a fine network, especially marked on head and to a lesser 

 extent on trunk, often fine white lines radiating from eye. Dorsal 

 with a whitish margmal band, underlaid by a dark brown submarginal 

 band, basal part more or less dusky, sometimes nearly uniformly dark 

 below marginal whitish band. (See tables 1 and 2 for counts and 

 measurements.) 



Distinctive characters and relationships. — H. hippocampus is appar- 

 ently related both to europaeus and to reidi, as discussed under the 

 accounts of those species. It has a distinctive appearance, owing to 

 its very low or obsolescent tubercles, short snout, and rather deep 

 body. In the low or nearly obsolescent tubercles it somewhat re- 

 sembles reidi but differs markedly in its conspicuously deeper trunk 

 and shorter snout and in having fewer pectoral rays, although there 

 is a small degree of intergradation in the latter character. It may 

 be sharply distinguished from guttulatus, its congener occurring in 

 the same region with it, by a number of characters, as pointed out 

 on page 544. 



A fair percentage of the specimens tend to have the first caudal 

 segment quadrangular. This deviation occurs less frequently in 

 hudsonius and punctulatus, while in the subgenus Jamsus (see p. 584) 

 it becomes the dominant condition. In hippocampus tliis variation 

 is apparently correlated with a novemangular antepenultimate trunk 

 segment. 



Material examined and geographic distribution. — Bay of Naples, 

 S. E. Meek, April 1897, four specimens (48325). Also seven speci- 

 mens from the collection of the American Museum of Natural History, 

 as follows: Two from the Zoological Station, Naples, Dr. Hovey 

 (1082), and five purchased from the Zoological Station, Naples (5042) 

 without further data. All these no doubt belong to the same species. 



Total number of specimens studied, 11, 55 to 104 mm long (one 

 specimen with the tip of the tail broken possibly somewhat larger 

 than the longest specimen recorded here) . All the specimens, except 

 the smallest one, have a brood pouch or at least a rudiment of one. 

 According to Rauther most of the females of this species have a 

 brood pouch more or less developed; apparently the sexes cannot be 

 distinguished by that character. 



Nomenclature and synonymy. — This species has been designated 

 most generally by the name of hrevirostris, but the review of the 

 literature (pp. 520 to 522) shows that that name has been proposed as a 

 substitute for the earlier name hippocampus, of which it consequently 

 becomes a synonym. In this case there is greater advantage in 

 following the rules rather than general usage and sinking the name 



