5 6 DISCOVERY REPORTS 



In the Drake Passage (section i, Plate III) sub-Antarctic surface water in mid-April 

 1930 (Sts. 385-388) had a surface phosphate content of 91 mg. just north of the Antarctic 

 convergence and 74 mg. on the continental shelf near Cape Horn. In the Scotia Sea, 

 south of the Falkland Islands, the surface value was 72-78 mg. in late March 1934 

 (Sts. 1323-1325) and 82 mg. in late September 1934 (Sts. 1423-1424). The value of 

 82 mg. south of the Falklands is surprisingly low for late September and is probably 

 explained by reason of a southward movement of low phosphate water in this region 

 in winter. This explanation is supported by biological evidence of the presence of a 

 temperate-water species, Chaetoceros decipiens. Immediately after the observations in 

 each month south of the Falkland Islands we have data from just north of the Antarctic 

 convergence north of the South Orkney Islands. These observations gave a surface value 

 of 83 mg. in April 1934 (St. 1338) and of 108 mg. in early October 1934 (St. 1435). 



Observations made along 30 W in the western part of the South Atlantic Ocean 

 (section 2, Plate IV) in April 1931 (Sts. 668-671) showed that the surface value in 

 sub-Antarctic water varied from 79 mg. in the south to 73 mg. at the most northerly 

 station in the zone, with a probable value of about 60 mg. just south of the subtropical 

 convergence. These values indicate that regeneration was in progress when the observa- 

 tions were taken. In late April 1933 at St. 1162 in the eastern part of the South 

 Atlantic Ocean (section 3, Plate V), a surface value of 73 mg. was recorded in the 

 southern part of the zone. The order of this value agrees very well with that found 

 previously in 30 W. 



In the sub-Antarctic zone south of Cape Town in late March 1935 (Sts. 1556-1558, 

 section 4, Plate VII), the surface phosphate value varied from 38 to 89 mg. from north 

 to south. In October 1930 in a section south-south-west from Cape Town towards Bouvet 

 Island (Sts. 449-45 1 ), the surface content in sub-Antarctic water varied southwards from 

 50 to 91 mg. Farther to the east between the Crozets and Durban (section 6, Plate X), 

 in the third week of May 1934 (Sts. 1 367-1 369), the surface phosphate content was 98 mg. 

 in the south of the zone and decreased to 74 mg. at the most northerly station, whilst 

 in early March 1935 at St. 1553 in 49 35-1' S, 30 44-4' E (section 4, Plate VII), the 

 surface value at the southern boundary of the zone was 88 mg. North of Marion Island 

 at St. 1565 in 44 047' S, 37 21-9' E, the surface content was 60 mg. in mid-April 

 1935 ; the position of this station was nearer to the subtropical convergence than to the 

 Antarctic convergence. In a section in early May 1932 across the sub-Antarctic zone 

 between Enderby Land and Fremantle (section 7, Plate XII, Sts. 866-870), the surface 

 phosphate content decreased from no mg. in the south to 72 mg. in the north. 



South of Australia in early June 1932 the phosphate content in the surface water of 

 the sub-Antarctic zone (section 9 from the ice-edge in 63 41' S, 130 07' E, south 

 of Australia, to Melbourne, Sts. 892-895) varied between 101 and 37 mg. from south 

 to north (Plate XIV). The value of 101 mg. at the southern end of the zone probably 

 represents very nearly the maximum surface value in sub-Antarctic water. South of 

 the Tasman Sea between Melbourne and the ice-edge in 6i° 25' S, 154 26' E (section 10, 

 Plate XV, Sts. 899-903), the surface value of sub- Antarctic water increased from 43 to 



