DISTRIBUTION OF SURFACE TEMPERATURE i97 



the mean isotherms cannot always be made to correspond with the obser^'ed temperatures so well as 



they do in this chart. • . tv/t * 



Of the published records from other sources, perhaps the most important are those m the Meteor 

 Reports (Bohnecke, 1936). Here a very large number of surface temperature readings are tabulated, 

 but they are from all parts of the Atlantic Ocean and only a small percentage he south of the Antarctic 

 convergence Bohnecke gives two principal tables. One shows the temperatures (also S°/oo, etc.) 

 recorded by the 'Meteor', with date and position (pp. 41-64). Only a small number of these are m 



Fig 8 A. Part of a track chart for October 1930. showing hourly positions of the ^Discovery II ' during a voyage from Cape 

 Town to South Georgia. See Fig. 5 A for the thermograph record of part of this track. 



12 n X- Ship's position at 1200 hours on 13th October. 



20^00 (451): Stopped on Station 451 from c. 2000 to 0000 hours. 



B. P^Jof^rCperature chart for October, including ^Je t.J in A w^^^^^^^^ 



rrtte-^s^trorie^^^^^^^^^ -P-- -' -'«- --''-' 



from Plate^I, and the 2° isotherm lies on the mean pos.t.on of the convergence^^^ 



A.C. (No. 33): Observed position of the Antarctic convergence (No. 33 m lable 9, p. 20OJ. 

 ('25): Meteor observations, 1925. 

 ('26): 'Wi.lliam Scoresby' station, 1926. 



Antarctic water, but they arc of some assistarrce although they fill no important gaps in *e Discovery 

 material For tkc other tabic (pp. 65-186) the Atlantic Ocean ts div.dcd into areas enclosed by ro 

 Tnati udc and longitude, and these are subdivided into 1° areas. Data from many sources are mcluded 

 aid o far as the data permit, the average temperature is given for each f area for each momh. Some 

 of A e a e n Antarctk waters, bu, it seems that such averages, compiled from m.seellancous sources, 

 1 u d be t Led with caution. B6hnecke drew charts of the Atlantic Ocean showmg the surface 



