XXIII. MEMBRAS 57 



86. Membras martinicus Cuvier and Valenciennes, 1835. 



Atherina martinica Cuvier and Valenciennes, Hist. Nat. Poiss., 10, 459, 1835. 

 Menidia martinica Jordan, Proc. U. S. Nat. Mus. 9, 530, 1886 (redescription 



of type). 

 KiRTLANDiA MARTINICA Jordan and Evermann, Bull. U. S. Nat. Mus., 47, pt. 1, 



795, 1896. 

 Type-locality. — Martinique. 

 Range. — Known only from the type locality. 



This species is known only from the original types. 



87. Membras vagrans vagrans (Goode and Bean), 1879.^*- 

 • (Plate IV, Fig. 16) 

 Chirostoma vagrans Goode and Bean, Proc. U. S. Nat. Mus., 2, 148, 1879, 



Menidia vagrans Jordan and Gilbert, ibid., 5, 267, 1882; Bull. U. S. Nat. Mus., 



16, 407. 1883 (in part). 

 Kirtlandia vagrans Jordan and Evermann, Bull. U. S. Nat. Mus., 47, pt. 1, 



794; 4, pi. 124, fig. 336, 1900 (figure inaccurate in certain details). 

 Type-locality. — Pensacola, Florida. 

 Range. — Gulf Coast of the United States, south to Tampico, Mexico. 



Soft rays of second dorsal, 7 to 9 ; pectoral fins a little longer or 

 a little shorter than head. 



87a. Membras vagrans laciniatus (Swain), 1883.-^'- 



Chirostoma menidium Jordan and Gilbert, Proc. U. S. Nat. Mus., 1, 383, 1878 



(in part). 

 Menidia bosci Jordan and Gilbert, Bull. U. S. Nat. Mus., 16, 407, 1883 (exclusive 



of references). 

 Menidia vagrans Jordan and Gilbert, /. c, 407 (in part). 



Kirtlandia vagrans Smith, N. Car. Geol. E<on. Surv., 2, 178, 1907. 

 Menidia vagrans laciniata Swain, in Jordan and Gilbert, Bull. U. S. Nat. Mus., 

 16, 908, 1883. 

 Menidia laciniata Jordan and Gilbert, Proc. U. S. Nat. Mus., 5, 589, 1882 (1883). 

 Kirtlandia laciniata Jordan and Evermann, Bull. U. S. Nat. Mus., 47, pt. 1, 



795, 1896; Cockerell, Proc. Biol. Soc. IVash., 23, 47, 1910. 

 Kirtlandia vagrans laciniata Fowler, Ann. Rept. New Jersey State Mus., 



213, 1905 (1906) ; 392, pi. 95, 1906 (1907). 

 Type-locality. — Coast of North Carolina (fixed as Beaufort, North Carolina, by 

 Jordan and Evermann). 

 Range. — Atlantic coast of United States, from New York to South Carolina. 



Dr. Hugh M. Smith in 1907 questioned the validity of laciniata, 

 referring that form to the synonymy of vagrans. He does not state the 

 evidence which led him to this conclusion, and we are unable to agree 

 with him. In several specimens from the Rio Panuco at Tampico, 

 Mexico, the scale rows vary from 42 to 45 ; in five from Cedar Keys, 

 Florida, they vary from 45 to 47; in nine from Ocean View, Virginia, 

 from 48 to 50; in five from Carson's Inlet, New Jersey, from 47 to 

 50. In the same lots the soft anal rays vary as follows : 



