LIFE-HISTORIES OF THE COD AND WHITING. 3 



conform to a more or less definite sequence, whilst others, tailing* off at either end of the 

 average, will constitute the transitional forms. 



For mere convenience' sake, we have in prior papers # divided the habitat into — 

 Bottom, in moderate depths, indicated by (B) ; surface-water in offshore and the water 

 near the surface, by (S) ; midwater, or the regions of the water extending from about 

 two fathoms to the bottom, in moderate depths, by (M) ; and lastly, the littoral district 

 and the sea floor therein as (L). 



It is evident that these are purely arbitrary divisions, and, as has been indicated 

 elsewhere,t it is not unlikely that physical changes may cause a temporary migration 

 from surface to midwater, and even to the bottom. Such changes are known to take 

 place in the case of other pelagic forms, and there is apparently no reason why the 

 pelagic eggs and larvse should form an exception. This phenomenon, however, does 

 not invalidate the demonstration of the true ontogenetic migration, and the eggs, for 

 example, of the cod, will still be acknowledged as normally of pelagic surface-habitat 

 in spite of the fact that under certain physical conditions some may be found at or 

 near the bottom. 



Again, it must be remembered that the division of two fathoms or otherwise between 

 the midwater and the surface is a purely arbitrary one, and that hence there are no 

 doubt several post-larval forms placed amongst the midwater, which might be more 

 accurately located amongst the surface section. 



If the surface-forms, as indicated in Table I.,j be arranged as in Table II., with their 

 lengths in mm., forming the abscissae, and the number in each haul the co-ordinates, 

 then a distinct curve results. It will be seen that the early post-larval stages, at 4 and 

 5 mm., swarm in great numbers in the surface-habitat, and that as growth proceeds, 

 there is a steady diminution in numbers till a length of 13 mm. is reached. The 

 occurrence of solitary individuals carries the curve on to a length of 17 mm., after 

 which the young cod no longer appear at the surface. 



This marked and rapid decrease in number of young cod may be due to several factors. 

 Firstly, a great number of them, as development proceeds, forsake the surface-water 

 and move downwards through the midwater, where they appear in the diagram under a 

 separate curve. Secondly, as development proceeds, the young cod gradually become 

 more and more dispersed over the inshore water, so that although the same net was 

 employed throughout, and approximately the same districts were worked over, yet the 

 number of fishes in each haul steadily decreases, even if we make a combined curve of 

 all the four here depicted. At the same time, one can scarcely explain the great 

 decrease as being entirely due to a distribution or scattering of the fry, and one is 



* Annals and Mag. Nat. Hist., vol. xvi., Oct. 1895. 



t 14th Annual Scottish Fishery Board Report, 1896 ; loth Annual S. F. B. Report, 1897. 



X The tables were exhibited at the reading of the paper, but have unfortunately disappeared since the work was 

 handed in for publication. — P. G. T., Sec. R.S.E. 



Table II. here given is a reproduction of the original to show this graphic method of illustrating the ontogenetic 

 migration of food-fishes, and as an explanation of the text. The figures on this table are not absolutely accurate, 

 except in so far as they have been taken from the text. — A. T. M., March 1900. 



