THE CYCLE OF LIFE 459 



the deep eastern and western basins and over the Azores 

 ridge separating them. The larger laivse were all got 

 north of the Azores, and the younger stages were aU found 

 south of the Azores, which led Dr. Hjort to suggest that 

 the spawning area is probably in the southern central part 

 of the North Atlantic. No transformation-stages were 

 found in mid-ocean, and it may be that the change 

 only occurs on the Continental slope. But it must 

 always be remembered that the developing eggs have not 

 yet been discovered. 



The Salmon. — ^In British rivers, the time of salmon 

 spawning is in the late autumn or winter. The eggs are 

 laid in the gravelly bed of the stream, and they develop 

 very slowly. After three or four months the egg-envelope 

 bursts and the larva is set free, still encumbered with a 

 large yolk-sac, on the contents of which it subsists for about 

 seven weeks. About the eighth week after hatching, the 

 supply of yolk is exhausted, and the ' fry ' — about an inch 

 long- — ^begin to fend for themselves and to move energetic- 

 ally. They grow by the end of the year to be somewhat 

 trout-hke ' parr ', about four inches long. In their second 

 year, usually, the young salmon change in coloration, 

 donning a beautiful ' sea-jacket,' and are known as ' smolts ' 

 — six or seven inches in length. These go down to the sea, 

 feed voraciously, grow rapidly, accumulate stores, and 

 become grilse. After a variable period of feeding and 

 growing, which may last a year or two years or more, they 

 are ready to spawn, and return to the place of their birth 

 in the fresh waters. Such in outhne is the typical Ufe- 

 history of the salmon, but there are many variations on 

 this theme. We have described in our Biology of the Seasons 

 the journey up the rivers, the struggle against the stream 



