Mr D. Ellis on the Natural History of the Salmon, ^()5 



their destined changes. These peri(x3s may be varied, to a cer- 

 tain extent, by accidental circumstances, or the purposes of na- 

 ture be in some instances entirely frustrated ; but such acci- 

 dental occurrences only partially disturb, but do not counteract, 

 the operation of general laws. In certain seasons, for example, 

 a deficiency of water in any particular river may, in the first 

 instance, prevent the parent fishes from ascending to deposit 

 their spawn, when, by nature, they are prepared to do so ; and 

 the proper season for spawning may thus be delayed, or some- 

 times entirely lost. In other instances, obstacles, either natural 

 or artificial, may oppose the ascent to the spawning grounds ; 

 and the female be constrained, as she sometimes is, to discharge 

 the ova in the deep water of rivers, or in the sea, where they are 

 wholly lost. Even when she gains the upper parts of rivers, 

 some time may elapse before she finds a suitable place to depo- 

 sit the spawn, or a male to impregnate the ova : or the bed, in 

 which the impregnated ova may have been duly deposited, may 

 not retain a suitable quantity of water : or the water itself may 

 become contaminated, and not furnished with air fitted to carry 

 on the evolution of the o\a : or, though the water and air be 

 duly supplied, a difference of temperature, arising from season, 

 from elevation above the sea, or from the prevalence of winds, 

 may check the progress of developement, and proportionably re- 

 tard, in particular rivers and situations, the appearance of the 

 young fry, or even prevent it altogether. Even when the evo- 

 lution of the ova may have been accomplished in due time and 

 manner, the want of water in rivers, during very dry seasons, 

 may retard their descent to the sea until a later period than 

 usual, or sometimes altogether prevent it. 



Making, however, all due allowance for these varying cir- 

 cumstances and their corresponding results, there seem to be 

 some rivers in which the breeding period of the salmon is uni- 

 formly earlier than in others. Thus, says Mr Little, the rivers 

 Annan, Esk and Nith, do not afford salmon in perfection until 

 a full month after the Dee, which is adjacent to them ; and the 

 salmon caught in the Dee are in bad condition nearly a month 

 sooner than in the other rivers ; they are full bellied, and in 

 worse condition. So, likewise, the salmon taken in the river 

 Shannon in Ireland, are in greatest perfection in February, 



s2 



