234 NATURAL HISTORY OF SELBORNE. 



of the water with its belly uppermost. The reason why fishes, 

 when dead, swim in that manner is very obvious ; because, when 

 the body is no longer balanced by the fins of the belly, the broad 

 muscular back preponderates by its own gravity, and turns the 

 belly uppermost, as lighter from its being a cavity, and because 

 it contains the swimming-bladders, which contribute to render it 

 buoyant.* Some that delight in gold and silver fishes have 

 adopted a notion that they need no aliment. True it is that they 

 will subsist for a long time without any apparent food but what 

 they can collect from pure water frequently changed ; yet they 

 must draw some support from animalcula, and other nourish- 

 ment supplied by the water ; because, though they seem to eat 

 nothing, yet the consequences of eating often drop from them. 

 That they are best pleased with such jejune diet may easily be 

 confuted, since if you toss them crumbs they will seize them 

 with great readiness, not to say greediness : however, bread 

 should be given sparingly, lest, turning sour, it corrupt the 

 water. They will also feed on the water-plant called lemna 

 (duck's meat), and also on small fry. 



When they want to move a little they gently protrude them- 

 selves with their pinna pectorales; but it is with their strong 

 muscular tails only that they and all fishes shoot along with such 

 inconceivable rapidity. It has been said that the eyes of fishes 

 are immoveable: but these apparently turn them forward or 

 backward in their sockets as their occasions require. They take 

 little notice of a lighted candle, though applied close to their 

 heads, but flounce and seem much frightened by a sudden stroke 

 of the hand against the support whereon the bowl is hung ; 

 especially when they have been motionless, and are perhaps 

 asleep. As fishes have no eyelids, it is not easy to discern when 

 they are sleeping or not, because their eyes are always open. 



Nothing can be more amusing than a glass bowl containing 

 such fishes : the double refractions of the glass and water repre- 

 sent them, when moving, in a shifting and changeable variety of 

 dimensions, shades, and colours ; while the two mediums, as- 

 sisted by the concavo-convex shape of the vessel, magnify and 

 distort them vastly; not to mention that the introduction ol 



* The uses of the swimming bladder of fish are at present nut in the least understood, though 

 many naturalists have taken great pains to investigate the subject. It is wanting, or rather (to 

 express it better) it is not present, in many species, even in some which are closely allied to those 

 which have it. Thus, it is not found in the common mackarel, while it exists in its near congener 

 the Spanish mackarel, a species comparatively rare in the British seas. ED. 



