84 ACANTHOPTERYGII. 



ventral spines, and in the lateral plates. In some individuals these do 

 not occupy more than the central portion of the sides, in others the whole 

 sides, and again are intermediate. 



In the absence either of a specimen for comparison, or a figure to refer 

 to, it may perhaps be considered that certainty cannot be arrived at re- 

 specting G. semiloricatus. This fish is stated to ditier from G. traclmrus 

 in having only 22 or 23 plates on each side to the origin of the caudal 

 keel instead oi' its 25 or 26, and in the shoulder-plate (plaque de I'epaule) 

 being larger. It has been seen that some of my specimens, and of these 

 some of the largest size, possess only the number of lateral plates attri- 

 buted to G. semiloricatus. In examples of equal length, and from the 

 same as well as from different localities, I find the size of the shoulder- 

 plate to vary like other characters. Hence I am disposed to regard some 

 of the examples under consideration as this fish. 



In the Hist, des Poiss. it is remarked of G. semiloricatus, " Nous n'avons pu 

 trouver aux environs de Paris que des epinoclies a queue nue ; il nous en est 

 Venn de pareilles des departemens de la Somme et de I'Oise, de la Rochelle et 

 de quelques autres lieux : nous avons observe celle a queue cuirassee dans les 

 ruisseaux des cotes de Normandie, et encore recemment M. Deslongchamps 

 nous I'a envoyee de Caen, et M. Baillon en a pris dans le Hable-d'Ault, lac 

 saumatre de I'embouchure de la Somme, pres du Treport. C'est la seule qui 

 se trouve dans les etangs des environs de Berlin, et elle y est en quantite iimom- 

 brable. Peut-etre est-ce I'espece qui habite plus frequemment pres des bords 

 de la mer, et qui pent entrer dans I'eau salee. Des observations ulterieures 

 nous apprendront sans doute bientot ce qui en est." — t. iv. p. 494. 



This accords generally with my own observation, as in seven out of 

 the nine localities whence my specimens mailed tlu'oughout the sides 

 were derived — Avhether they be called G. trachurus or G. semiloricatus — 

 they were taken either in the sea or estuary. The exceptions are the 

 largest specimens, which were procured in a " pool of brackish water ac- 

 cessible to the sea ; " and those from Rathlin, obtained in fresh water. 

 From the passage just quoted, we learn that the G. semiloricatus inhabits 

 the pools about Berlin. It has always seemed to me not improbable, that 

 in the sea, Avhere the enemies of this diminutive fish are more numerous 

 than in the fresh Avater, the protecting hand of Nature had as a deience 

 armed its body with these lateral plates. That some fishes have the 

 power of accommodating their colour to that of the ground or bottom of 

 the water they frequent, and are thus rendered comparatively inconspicu- 

 ous to their enemies, is well established. 



A third species of 3-spined stickleback, armed throughout the sides 

 like those here treated of, is the G. Norehoracensis, which, as its name 

 denotes, is found at New York. Judging from the description and figure 

 of this fish in the Hist, des Poiss., I should not consider it distinct 

 from G. trachurus or G. semiloricatus. The specimens which have come 

 under my examination differ much in the few characters which are said to 

 distinguish this fish from G. trachurus. The high position of the lateral 

 line is the chief character of G. Norehoracensis ; in some specimens be- 

 fore me this line is so near the back that three-fourths of the body of the 

 fish are below it. Our G. Puuf/itius is admitted as an American species 

 by Dr. Storer in his interesting work on the Fishes, &c., of Massachu- 

 setts (p. 32), and for a copy of which I am indebted to his kindness. 



The descri])tions and figure of the G. oholarius, Cuv. and Val. — a 3- 

 spined stickleback armed throughout the sides, and found in the North 

 Pacific Ocean and the Gulf of Kamtschatka — are said in the Hist, des 



