38o DISCOVERY REPORTS 



captured being only 14I cm. long. As forage for larger species, of greater potential economic value, 

 Thysanopsetta is evidently of considerable local importance, for, as we have already seen, it figured 

 largely in the stomach contents of hake [Merluccius hubbsi) and Thyrsites atun, as well as in more 

 typical bottom-feeders such as Cottoperca gobio and some of the rays. 



The distribution of Thysanopsetta is very interesting. It seems to be peculiar among the fish fauna 

 of the shelf in being influenced more by the nature of the bottom deposit than by any other distri- 

 butional factor. 



Some seasonal movement there is, for when the catches are charted (Fig. 53) it can be seen that the 

 fish were farthest south in autumn, farthest north in winter, and nearest to the mainland coast in 

 summer. These trends agree in time and direction with those followed by most other fish in the region 

 that show seasonal movements, but were much smaller in extent, so that Thysanopsetta was rarely 

 found far from its ' metropolis ' on the central area of brown sand that occupies most of the plain of 

 the shelf in the intermediate region. This was delineated by Matthews (1934, pi. xiii), when he was 

 working on the bottom deposits obtained with the dredge during the trawling surveys, long before 

 the fish data had been worked up. The close correlation between this deposit and the distribution of 

 Thysanopsetta was first seen by E. R. Gunther, and was very clearly described by him in a preliminary 

 (verbal) account of these surveys. He had subsequently altered the manuscript so drastically that I 

 have had to attempt an independent presentation of the data (Table 39) and can only hope that this 

 will prove adequate, in conjunction with the charts. 



Table 39. The relation betzveen a localised type of bottom deposit and catches of the flatfish 



Thysanopsetta naresi 



Figures in heavy type relate to observations made within the central area of brown sand, delineated by Matthews 

 (1934, pi. xiii); figures in italics to observations made elsewhere, on other deposits. 



It also seems clear that Thysanopsetta is most concentrated (i.e. most given to shoaling) in summer. 

 Its absence from our few spring hauls should not be considered as evidence of the converse (greatest 

 dispersal at that season) for they were nearly all worked to the north of the normal limits of distri- 

 bution of the species. It is more likely that Thysanopsetta is most widely dispersed during winter. 



Thysanopsetta naresi is evidently a true bottom-dweller, and even more so than some other species 

 of flatfish inhabiting similar depths, for the latter often show by the nature of their food that they 

 must spend a considerable part of their time swimming clear of the bottom. We know that Thysanop- 

 setta must lie exceptionally close, because although it was so frequently captured in the accessory nets 

 attached to the back of the trawl, it was only once taken in the Russell bottom net. This is designed 

 to fish a few inches clear of the bottom, and very many hauls were made with it in regions where 

 Thysanopsetta was known to be plentiful. It is the more remarkable that such (normally) semi-pelagic 

 feeders as hake and Thyrsites were found to have fed extensively upon Thysanopsetta. I believe that 

 the explanation partly lies in the uniformly moderate depths over the plain of the shelf, for this brings 

 the bottom fauna well within reach of larger predators which are given to diurnal vertical migration. 



