METHODS USED IN THE 'WILLIAM SCORESBV 135 



divergence line and producing in the coastal waters a distribution of density such that a current will 

 form parallel to the coast. Seawards of the divergence line, and in the upper part of the vertical eddy, 

 the water will be transported offshore. 



Studying the general biology of the region, Hentschel (1936) found it necessary to adopt some 

 arbitrary spatial definitions of the several sea areas he discussed. The 'ten degree field' extending 

 westwards as far as St Helena which he took as his Benguela current region has unfortunately masked 

 the great richness of the coastal current inshore. Consequently most of the generalizations he felt able 

 to make apply rather to the subtropical oceanic surface water. Of the existence of the rich coastal belt 

 he was, of course, very well aware. This was the only part of the region where large quantities of 

 diatoms were to be found among the microplankton. He did not, however, regard the coastal current 

 as distinct from the oceanic circulation, whereas our present results indicate that the natural boundary 

 between the two types of surface-water is more significant than any of the boundaries of his arbitrarily 

 chosen areas. No doubt he was unable to demonstrate the effect of this natural boundary on account 

 of the wide spacing of the ' Meteor ' stations, and the fact that the physical data were only partly 

 digested by the time his work was published. His reference to the characterization of the region by 

 the steep east-west gradient in microplankton quantities rather than by an abrupt transition such as 

 our results indicate, his inclusion of many warm-water forms among the characteristic dominants of 

 the region, and his insistence on the dominance of Dinophyceae over Diatoms among the protophyta, 

 all arise from the same causes. Provided this is all borne in mind it can be seen that his generalizations 

 are in very good agreement with our own findings — that is, apart from his insistence on the importance 

 of the ' South-west Africa Tongue ' which will be discussed later. 



While extremely valuable for interpreting the large-scale picture, the 'Meteor's' observations 

 unavoidably had the disadvantage of being widely spaced both in time and geographically, and the 

 'William Scoresby' surveys can be considered as the next logical step in making a more localized and 

 intensive study of the Benguela region in particular. 



METHODS USED IN THE 'WILLIAM SCORESBY' 



Observations and collections 

 Throughout the two surveys the watch-keeping officers maintained regular meteorological observa- 

 tions. In addition to those kept at four-hourly intervals, further records were taken while the ship 

 was occupied on ' station '. The observations included personal estimates of the force and direction of 

 wind, sea and swell, and instrumental records of barometric height, and dry- and wet-bulb temperature. 



Continuous echo-soundings were made whenever possible, but these were restricted to the conti- 

 nental shelf and slope, as the depth range of the machine (720 fathoms, 1317 m.) did not permit any 

 deep-water sounding. The latter were eventually obtained by a Lucas wire-sounding machine, but 

 unfortunately this was not functional during the first survey. On the second survey, however, the 

 Lucas machine made it possible not only to obtain deep soundings but also to delimit the extent of 

 a region of reducing mud on the sea floor. For this the sampling technique depended upon the nature 

 of the sediment, and both Baillie rods and snapper leads were found effective. The samples were 

 preserved in alcohol after a preliminary examination. 



The 'William Scoresby' was fitted with a distant-reading thermograph which provided a con- 

 tinuous record of the sea temperature. Since the bulb of this instrument was installed in the engine 

 condenser intake, the record represents the temperature at a depth of about 4 m., but this of course 

 varied considerably with the degree of loading of the ship. Checks on the accuracy of the instrument 



