i897. SPAWNING HABIT OF TELEOSTEAN FISHES. 387 



A (the spawning-area) would only be limited by the necessity for being 

 a certain distance offshore, and beyond this the adults could spawn 

 with impunity in whatever environment they might happen to be. A 

 spawning-migration would then be eliminated, except in the case of 

 individuals living in very close proximity to land. It is probable that 

 the dab is a species largely conforming to this type of spawning. 



The first factor in the determination of the spawning-area of a 

 given species is, then, the adaptability or otherwise of the young forms. 

 The second is the nature of the inshore nursery, whether it slopes 

 rapidly or shelves gently. 



Let us now consider the other factors determining the length of 

 A B, assuming that B is to some extent fixed, as a mean between B, 

 and B,,. Any agents which tend to lengthen A B will then necessitate 

 a spawning-area (A) further out to sea and vice vevsa. 



An egg, or larva, floating inshore remains, as already said, in the 

 pelagic-habitat, till the migration downwards is effected ; and this 

 takes place at a definite stage in development. The time taken to 

 reach this stage in ontogeny may be termed the ' pelagic-period,' and 

 it is evident that, for maximum results in the way of successful 

 development, this 'pelagic-period' must correspond exactly with the 

 time taken for an egg or any object to drift from A to B. Suppose 

 that the rate of drift be too rapid, then by the time the egg has 

 reached B the pelagic-period will not have come to an end, and the 

 egg or larva will be washed ashore and perish. Some observers {e.g. 

 G. O. Sars, loc. cit.) have recorded cases in which myriads of cod's 

 eggs have been thus ' beached,' forming a long glistening line at high- 

 tide-mark. On the other hand, if the rate of drift be too slow, the 

 pelagic period will come to an end prematurely, and the young forms 

 will migrate downwards into deeper water than they are suited for, 

 and, at least in large part, pay the penalty by extermination. There 

 must, therefore, be a close agreement between the pelagic period and 

 the time of drift. 



The rate of drift in a given area depends upon the average 

 strength of the prevalent winds ; and this on its part depends largely 

 upon the season of the year. Further, as the rate of drift is simply 



~^ , the position of the spawning-area must be regulated by 



time of drift 



the season of the year at which spawning takes place, in such a 



way that the stronger the winds, the further seawards it must be 



removed. 



Again, the duration of the pelagic period has to be taken into 



account, for as seen above, it must correspond with that of pelagic drift. 



The two known factors which determine the length of the pelagic period 



are temperature and bulk. It has been found that the eggs of fishes, 



unHke those of higher animals {e.g. birds) have not a fixed and 



definite period of ' incubation,' but that the hatching and development 



of a young fish can be markedly hastened or retarded by alterations 



