GASTEROSTEIDiE. 237 



■waste of space to follow out the views of tlie various Britisli and of some French 

 ichthyologists respecting the fresh water forms of Gasterosteus, as the marine 

 G. spinachia, looked on as a distinct genus by some, is not subject to the variations 

 perceived in such sticklebacks as reside entirely or mostly in brackish or inland 

 ■waters. Pennant, 1776, Shaw, 1803, Turton, 1807, Donovan, 1808, and Fleming 

 in 1828, merely recognized two fresh -water species, the three-spined and the 

 ten-spined. In 1829 the fourth volume of Cuv. and Val. Histoire Nature He des 

 Poissons, was published, in this many forms heretofore looked upon as varieties were 

 considered to be species. Yarrell's British Fishes, published in 1836, contains : — 



1. G. trachurus, Cuv. and Val. 2. G. semiarmatus, Cuv. and Val. 3. G. leiurus, 

 Cuv. aud Val. 4. G. brachycentrus, Cnv. and Val. ; and 5. G. spinulosus, Jenyns 

 and Yarrell ; also 6. G. pungitius, Linnsens. Jenyns, 1835, restricted these forms 

 to three : — 1. G. aculeatus, including trachurus, semiarmatus, leiurus, and hrachy- 

 cerdrus. 2. G. spinulosus, and 3. G. pungitius. Parnell, 1838, gave 1. G. leiurus, 



2. G. semiarmatus, 3. G. trachurus, 4. G. spinulosus, and 5. G. pungitius. Thompson 

 in the Annals and Magazine of Natural History, 1841, and Natural History of 

 Ireland, 1856, restricts these numbers to 1. G. aculeatus, and 2. G. pungitius. 

 White, 1851, again simply enumerated 1. G. aculeatus, 2. G. spinulosus, and 



3. G. pungitius. Giinther, 1859, recognized 1. G. aculeatus, 2. G. brachycentrus, 

 3. G. spinulosus, 4. G. ptmgitius, while the latter author* in 1880 again reduced 

 his British forms by omitting number 2. Canestrini (An. Soc. Nat. Mod. ii, 1867, 

 p. 9) disputed the accuracy of the varieties of G. aculeatus, as found in Italy, 

 being considered as species. 



M. Blanchard in his elegant work on the Fishes of the Fresh Waters of France, 

 1866, describes eight sticklebacks irrespective of G. pungitius and allied forms, 

 ■which latter he divides into five species. , Dr. Sanvage, 1874, recognizes three or 

 four more out of a recorded total of forty-five : while Dr. Giinther subsequently 

 remarks in 1880 that " about ten species are satisfactorily known," and Dr. Moreau 

 in 1881 reverts to the original two forms as existing in the fresh -waters of France, 

 placing the others as varieties. 



It appears remarkable how many species have been named of sticklebacks, 

 outnumbering even those of the SalmonidsD of the fresh waters, and it becomes 

 a first consideration whether any general principles are perceptible in the 

 distribution of these species or varieties. It is in the ocean more than in fresh 

 •waters that we must seek the spiny-rayed fishes : and similarly it is on the 

 sea-boards or skirts of the ocean that we must look for sticklebacks in which the 

 armature of the side is most developed (as in the variety trachurus), -while such as 

 have the free portion of the tail unarmed are furthest inland or on elevated 

 plateaus : while in the centre of Ireland I have captured examples of G. pungitius 

 in -which the armature had so decreased that the ventral spine was entirely absent. 

 It has been pointed out (Phil. Mag. 1834, v, p. 299) that the variety on the 

 continent with the shortest spine or the most defenceless form, comes from 

 Tuscany, and is peculiar to still waters, where it would have the fewest enemies, 

 and here it attains to a great size. Taking large numbers of Irish specimens I 

 found considerable differences in the length of the ventral spines and pubic plates, 

 conclusively showing that such characters afEord no reliable data.f 



Heckel and Kner, in their account of the Fishes of Austria; did not admit 

 the foreo'oing to be more than varieties differentiated by the development of the 

 lateral scutes or plates, ■which they found varied in number between three and 

 t-wenty- eight. 



It appears to me that our British Gasterostei consist of merely three species, 

 ■which are subject to an almost endless variation in colour and form. They may 

 be divided as follows : 



1. Gasterosteus aculeatus, D. 3-4/10-12. Sides of the body with or destitute 

 of vertical flat plates. 



2. Gasterosteus pungitius, D. 8-10/9-10. Sides of the body smooth. 



3. Gasterosteus spinachia, D. 15/6-7. Snout produced : body very elongate. 



* Introduction to the Study of Fishes, p. SO.";. 



f Linn. Soc. Journal, Zoology, vol. xiii, " On some Irish Gasterostei," by Francia Day. 



