572 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 83 



breinrostris to synonymy, since that name was employed often to 

 designate other species as well, such as europaeus and species in other 

 parts of the globe. Furthermore, there is no possibility that the 

 name hippocampus will have to be changed again. Therefore, it is a 

 fortunate coincidence that sinking the name brevirostris as a synonym 

 of hippocampus will serve the triple purpose of complying with the 

 code, clearing away the existing confusion implied in the name 

 brevirostris, and fixing the name of this species with finality. 



Uncertain specimen. — A single specimen in the University of 

 Michigan Museum (111750), found in the same lot of seahorses 

 forming the basis of multiannularis (see p. 542), probably belongs to 

 hippocampus. Trmik segments 11; caudal segments 37; dorsal rays 

 18; pectoral rays 15; first caudal segment hexangular; penultimate 

 trunk segment septangular, tubercles nearly obsolescent. Length 

 102 mm, with a brood pouch; depth 18, snout 6.7, postorbital 10.5, 

 head 20.5, trunk 30.5, tail 67.5, and orbit 4 percent of length. If 

 these measurements are compared %vith table 2, it wall be noted that 

 by the length of its snout this specimen is either a europaeus or a 

 hippocampus, but its general physiognomy is more like hippocampus 

 and agrees more with the latter species in the depth of the trunk and 

 the appearance of the tubercles. The number of caudal segments and 

 dorsal rays falls just outside the frequency distribution of hippocampus 

 as determined (compare with table 1); but it seems to fit well in that 

 distribution as an extreme variant. If this specimen was one of the 

 original lot from Dagry Freres (see p. 542) and came from the Bay 

 of Biscay, it seems possible that hippocampus, like guttulatus, is repre- 

 sented on the Atlantic coast of Europe by a distinct subspecies. 

 However, that remains to be determined. It is more likely that it 

 came from the Mediterranean and represents a variant of its species 

 with respect to the number of caudal segments and dorsal rays. 



HIPPOCAMPUS KEIDI Ginsburg 



Figures 65, 66 



Hippocampus longirostris Kaup (not Schinz, 1822, a French species; not Cuvier, 

 1829, sec pp. 520 to 523 for discussion). Catalogue of the lopIiol)ranchiate 

 fish in the collection of the British Musuem, p. 12, pi. 3, figs. 2, 2a, 1856 

 (Martinique and St. Lucia; recognizal)lc figure of this species published). 



Hippocampus guttulatus Goode (not Cuvier), Amer. Journ. Sci., vol. 14, p. 291, 

 1877 (Bermuda). 



Hippocampus punctulatus Meek and Hildebrand (in part), Publ. Field Mus. 

 Nat. Hist., zool. ser., vol. 15, pt. 1, p. 255, 1923 (specimens from Porto 

 Bello only belong to present species). 



Hippocampus reidi Ginsburg, Journ. Washington Acad. Sci., vol. 23, p. 561, 1933 

 (Grenada, British West Indies; Porto Bello, Panama; Jamaica, W. I.; Haiti). 



Diagnosis. — First caudal segment hexangular (incompletely hex- 

 angular in one out of 12 specimens examined); last trunk segment 



