1T8 TINKER. 



each of them has a small triangular fin membrane at its 

 jiosterior base; all of them are erectile at the i)leasurc of the 

 fish, and wlien erected it is seen that they are not seated 

 \'xactly on the median line of the back, but on two lines, 

 each removed, almost imperceptibly, to the right and left of 

 a median line; on one line there are five spines, on the other 

 four, and they are seated alternately. Every spine, moreover, 

 on the right line has a most decided inclination to the right, 

 and every spine on the left line to the left, so that the series 

 are well represented by the teeth of a saw recently set, when 

 they are alternately and purposely bent to the right and left. 

 The carinated scales on the sides of the tail, which Cuvier 

 makes a character of G. pimgitius, are entirely wanting. The 

 colour is very uniform, as compared with that of our ordinary 

 Sticklebacks; it is a somewhat metallic yellow green on the 

 back, gradually becoming paler, and almost white on the belly, 

 the whole being irrorated with minute black dots; the fins are 

 very pale, almost colourless. The male resembles the female 

 until the month of March, when he begins to assume his 

 nuptial livery: the median line of the breast and belly then 

 becomes black, a colour which day by day extends on either 

 side until all the lower parts of the fish become of the most 

 intense velvety black; this eventually extends almost over the 

 whole body, the back only retaining slight indications of the 

 normal colourine;. 



It is a fearless and ferocious little fish, instantly reconciled 

 to captivity, and attacking with fury any prior inhabitant of 

 the vessel in which it is placed. It will frequently seize a 

 fellow-prisoner by the gill, the tail, or a fin, and retain its 

 grip with the firmness of a bull-dog; in the same way it will 

 instantly seize a worm when presented to it, and allow itself 

 to be drawn out of the water without relinquishing its hold. 

 The females become very much distended with ova, and deposit 

 them simultaneously; they are very large, generally eight or 

 ten in number, and are immediately devoured if found by 

 fishes of the same or other species: its nest is not known to 

 me. Unlike other species of Gasterostcus it will not exist 

 when confined in salt-water, however diluted. Mr. Newman's 

 doubts of the identity of this species, arising from the difference 

 of colour, would have ceased if he had recollected how 



