240 DISCOVERY REPORTS 



require no comment. I have preferred the name ' Falkland trough ' for the broad tongue of relatively 

 deep water running north between West Falkland and the mainland, to the ' Falkland channel ' of 

 some earlier writers. In describing faunal distribution in this part of the world it is particularly 

 important to avoid such misunderstandings as could arise by indiscriminate comparison of this 

 natural feature with the numerous and straiter 'Magellan Channels' to the westward. The broad 

 term 'plain of the shelf has been used to describe the central portion of the area where the gradient 

 of the sea floor is exceptionally slight. 



The topography of the shelf has been admirably described by L. Harrison Matthews (1934, 

 pp. 177-9) ii^ ^^is valuable account of the bottom deposits sampled by the conical dredge during the 

 trawling surveys. The sections shown by him (loc. cit., Fig. i) give ample illustration of the features 

 that exert most influence in a study of the distribution of the fish. These are: 



(i) The extremely slight gradient from 80 m. right out to the shelf edge at the 200 m. contour (the 

 distance exceeds 200 miles in places). This is most marked in the north; farther south the sea floor 

 slopes a little more steeply, but it is generally true to say that most of the shelf lies below water of 

 more uniform depth than is to be found over comparable areas elsewhere. The great importance of 

 this fact in studying fish distribution lies in the difficulties that result in interpreting depth relations, 

 especially of migratory species. Elsewhere hake movements lead to clear correlation between size of 

 fish and depth of water at appropriate seasons. Here the depth gradient is so slight that evidence of 

 similar movements can only be detected when distance offshore is substituted for depth. The diffi- 

 culty was augmented by areas of very slightly shallower water offshore, notably off the Golfo San 

 Jorge and south-east of Puerto Deseado. These very slight elevations can hardly be termed 'banks'. 



(2) The slope from the shelf edge to oceanic depths is very steep. In the north it is almost pre- 

 cipitous, so that trawling in water below 200 m. was limited to within a very few miles of the shelf 

 edge. Farther south trawling between 200 and 450 m. was possible over a wider area, but was difficult 

 owing to rough ground. 



(3) The Falkland trough and the area of deeper water separating the Falkland Islands from the 

 Burdwood Bank are also well illustrated by Matthews's sections. 



It is possible (for our present purpose) to make a slight improvement on Matthews's general chart 

 of the bottom topography (1934, pi. iii), by including several more recent soundings and plotting the 

 80 m. contour. This is shown in Fig. 3, which should be compared with Fig. 2 and with the separate 

 distribution charts when depth relations are under consideration. , 



The main body of Matthews's work dealing with the grading and distribution of the deposits 

 themselves is very instructive. The distribution of the coarser grades (mostly in the south of the area 

 and along the landward and seaward margins of the shelf) has obvious practical significance : coarse 

 grades, especially large fragments (loc. cit., pi. iv), usually coincide with foul ground for the trawler. 

 Matthews's main conclusion was that deposits became finer as one proceeded northwards (fine sand 

 and silt clearly predominate to the north), and that this is due to elutriation by the prevailing north- 

 ward flowing Falkland current, which may be said to act as a natural levigator. This is one reason for 

 the better trawling conditions found to the northward. 



To extract the full benefit from Matthews's work one would need to study first the relations 

 between the distribution of the deposits and the sessile benthic fauna which is more directly affected 

 by the nature of the bottom than are most of the fishes. Unfortunately, as already stated, this colossal 

 undertaking can only be begun when taxonomic revision of the main invertebrate groups is achieved. 

 However, one very striking correlation between fish distribution and bottom deposits can already be 

 demonstrated. The abundant occurrence of the flatfish Thysanopsetta naresi was almost entirely 

 restricted to a central area of brown sand delineated by Matthews (loc. cit., pi. xii, area 'C'). 



