MIGRATIONS AND SEASONS. lix 



one of Nature's methods of preventing the extermination of the race. For 

 decreasing their size will cause them to be less sought after, and it will only 

 be when the larger fish are left for breeders that larger offspring result. In 

 short, when man or other causes afford to fish sufficient protection, then 

 Nature assists in improving the race ; when man or other destructive factors 

 greedily kill all they can, then the breed dwarfs, and so possibly prevents its 

 extermination. 



To which ever division fish belong, they are generally perceived at the 

 commencement of their breeding season migrating to localities most suitable 

 for the reception of their eggs and the bringing forth of their young. The 

 majority of marine forms seek banks, or are found nearer in shore or in 

 shallower waters than such as they inhabit at other times. This arrival of 

 gregarious kinds of mature fish occurs when they are ready for breeding, 

 while their eggs are deposited prior to their leaving, so whether they come 

 to perpetuate their race or seek food for preventing death in each individual 

 of the species, it eventuates that at these periods breeding usually occurs, as 

 may be observed in the herring or in the mackerel. Anadromous forms pass 

 up rivers, sometimes for long distances, and then deposit their ova : among 

 the most widely distributed of such is the shad, of which we possess two 

 species in the British Isles, both found in the Severn, but up which they 

 now rarely ascend in numbers to any considerable extent, due to weirs across 

 the river, deficiency of water, or else its poisonous condition. The salmon 

 similarly ascends from the sea to deposit its ova in rivers and streams, and 

 this instinct of migration or necessity for exchanging its locality to a suitable 

 breeding spot may be more or less observed among a large number of 

 members of the family. 



The season at which breeding occurs varies with the family of fish and 

 the locality. This again is susceptible of further modification in accordance 

 with the temperature and perhaps composition of the water, the amount of 

 food procurable, and many other local circumstances. Likewise there is some 

 condition in the fish itself respecting which we know but little, but which 

 plays its part. It is easy to understand that during very cold winters 

 breeding is usually late, which may be partly occasioned by the ova taking 

 longer to hatch, as well as by the parent fish being later depositing its eggs. 

 The period at which the Sahnonidce in these isles breed may be roughly 

 estimated (unless under exceptional circumstances) at from the commence- 

 ment of September until the middle of January or February. In 1865 

 some brook trout eggs were despatched from Hampshire and Buckingham- 

 shire to Tasmania, and the first young reared in the Antipodes formed their 

 redds in July, 1869, or during the coldest season of the year (see vol. ii, 

 p. 60). But easy as this theory would be in order to explain the different 

 months fishes select for breeding purposes, there exist many exceptions 



