THE LIFE-HISTORY OF THE SALMON. 291 



they rush after each other in their fight for the females, and 

 here and there the surface is broken by the splash of a combat. 

 The sight from the Bridge at Galway as the fishes pass up to 

 their spawning-ground is also striking, but at a given moment 

 they are far fewer than those collected on such a spawning- 

 ground as that just mentioned. There the female stirs up the 

 gravel with her tail, sheds the eggs in batches, probably during 

 several days, and covers them over, the area being, as it were, 

 sown with them, whilst the attendant male emits milt for 

 fertilization. The newly extruded ova are covered with viscid 

 mucus, which to some extent causes them to roll less readily on 

 the stones, and prevents the current carrying them down- 

 stream. Their specific gravity, moreover, is greater than that 

 of the water, the reverse being the case in the pelagic eggs of 

 the Cod and many others. Moreover, the eggs of the Salmon 

 are shed into the ccelom, and thence out by the genital pore 

 behind the vent, whereas the Cod has oviducts. 



The developing egg remains in the gravel for 120 to 130 

 days or more, according to temperature, when the young fish is 

 extruded. In a private apparatus in a bedroom (fig. 3) the 

 young were hatched in sixty days. 



The larval Salmon keeps to its gravelly bed for a month or 

 six weeks, until its store of yolk is more or less exhausted ; and 

 then it swims freely as a little fish about an inch in length 

 (called a Parr), though at first it keeps to hollows in the bed of 

 the stream where the currents are less strong. 



The destruction of ova in their natural spawning-beds is 

 considerable by floods, Trout, Salmon-Trout, larvae, beetles, 

 birds, and other forms. Hence the basis on which hatcheries 

 were advocated, and there is no doubt great benefit in many 

 cases has resulted from the artificial rearing of Salmon to a 

 certain stage. Yet if the adults are sufficiently protected on 

 their spawning-beds (which unfortunately they are not in cer- 

 tain cases, e.g. Ireland), the need for hatcheries would be 

 restricted. It is the serious obstruction by weirs or impassable 

 falls and the too-severe netting, together with the interference 

 with the fishes on their spawning-beds, which reduces the 

 numbers in many fine rivers. 



As the breeding-ponds at Stormontfield-on-the-Tay were 



