388 



Part III. — Twentieilt Aunual lieport 



It is evident, however, that the range and the averages in many of 

 these cases require correction, owing to the fact that the smaller fishes of 

 the series are not, especially iii the earlier months, properly represented 

 in the collections. This is caused partly by the meshes of the net being 

 too large to retain them, and chiefly by the fishes at this stage not living 

 at the bottom but being dispersed in the water between bottom and 

 surface. And it is clear that the numbers that thus escape capture 

 may be considerable, although it would be difficult to assign the propor- 

 tion for any given month. Small whitings much below the mininmm 

 sizes stated are procured in tow-nets late in the season. Thus, at the 

 end of July specimens from 17 to 30 mm. were procured in the Moray 

 Firth, and on 9th August others from 22 to 65 mm. ; on 29th August 

 four caught in the Forth measured 24, 26, and 59 mm. ; one on 1st 

 October was 19 mm., and on 3rd October four, measuring 41, 47, 70, and 

 72 mm., were taken in Aberdeen Bay in mid-water, and one of 56 mm. 

 on 6th October. Clearly, therefore, the lower limit at the end of summer 

 and in autumn is below what is represented by the catches of the small- 

 meshed net. In January, however, we get, I believe, a fairly accurate 

 idea as to the minimum size of the series at that time, because the hauls 

 were then made with a .smaller-meshed net (fig. 1, b), from which few, if 

 any, whitings below the size of those taken could escape, and the limit 

 then was 88 mm,, or 3| inches. In July, when spawning has scarcely 

 quite terminated, it is obvious that a considerable proportion of the 

 young whitings must be small. In the Firth of Forth, as stated below, 

 they ranged from 54 to 109 mm. on the 23rd of that month — a range of 

 55 mm. ; and, looking to the fact that the ascertained range of the series 

 late in autumn and in winter exceeds 100 mm., and may be 115 mm., I 

 think the ranges and averages may be tentatively amended in the 

 preceding mc nths as follows : — 



