203 



DISTRIBUTION OF SURFACE TEMPERATURE 



Meridional lines of observations in this region in April are only available for one year (1938). No 



warm belt was then distinguishable, and the isotherms in Plate VIII are drawn accordingly, t is no 



perhaps certain that the S-shaped bend in the isotherms is normally smoothed out in Apri bu it 



may be that the advance of winter conditions has by then obliterated any signs of a warm belt at the 



The loop in the convergence in 30° E is referred to on p. 184. It is possible that more of the adjacent 

 isotherms are involved in this disturbance than the charts suggest, but more data would be needed 



to decide this point. . , ^^^^^„t 



The course of the isotherms in the Indian Ocean and Australian regions call for no special comment. 



The charts are self-explanatory, and it will be seen that more data are needed in some months. 

 The Ross Sea is another locality in which open water is found in a high latitude while pack-ice still 



lies to the north. The isotherm for - 1° should perhaps reappear here in January, February and March, 



but there is little material to show in what shape it should be drawn, and it is therefore omitted. 



4 0- 



3-0^ 



u 2-0- 



o 



kJ 

 a: 



D 



5G S 





I 0- 



0. 



5 

 u 

 1-0-0- 



-I 0- 



, I I I I I I I I I I I I I ' ' 



SEPT OCT NOV DEC JAM FEB MAR APR MAY JU^.E JULY AUG SEPT OCT NOV DEC 



Fig. II. Seasonal rise and fall of temperature, derived from Fig. 10. 

 In the Pacific sector the northward extension of cold water between 120 and 160° W, and the com- 

 pression of S^ isotherms in the winter and spring months in the Bellingshausen Sea and Drake Strait 



^"XTs^e^^e^^^^^^^ of the temperature m different months and 



diff nf iXdes. For Fig. :o the latitude of each isotherm in each month ^^^--f-]^ 

 II-XII at every 30^^ of longitude, and the average was plotted. For example, in Plate VI (Februa^^ 

 tL avLag lal^ of the 0° isotherm was found to be 65° 40' S. Thus m Fig xo the curve f r 

 the average latitu isotherms are correct the diagram should 



'^eTrect ii; esl ^Z^^. i. ave.ge .e.pe..u.e between .he differen. .on.hs and 

 o ,he Ze of the temperature gradient south of the convergence. The slope of the gradient on the 

 cllrgence and to the north of it, and differences in the shapes of the monthly curves are perhaps 

 notve^ rel able The figure shows that surface temperatures are much the same m January and March, 

 : il Nov mber and j!ly, that south of the convergence the water is warmer „t December than m 

 T ,! b,„ if wou d OTobably not be justifiable to infer much more than this, 



'"pt ;'is de W dtom' FTg .0, the teripera.ure for each month being plotted at arbitrary m.ervals 

 of Lfitude Ttflattening of the curves at the foot of the diagram indicates that the water ,s (on the 



