630 Prof. W. Thomson on the Voyage of the ' Challenger' 



light upon the observations taken in the West Atlantic in the year 1873. 

 These seemed at first sight to present certain anomalies ; thus in our 

 somewhat hurried run from Bahia to Tristan d'Acunha, we appeared to 

 have missed the source of supply of water at an unusually low tempera- 

 ture which we found near the island of Pernando Noronha. The result 

 has fully justified our anticipations. 



Diagram A, Plate 32, represents the vertical distribution of tempe- 

 rature at Station 112, lat. 3° 33' 8., long. 32° 16' "W., 21 miles to the 

 N.E. of Pernando ISToronha. Diagram B gives the temperatures at 

 Station 129, lat. 20° 12' S., long. 35° 19' "W., nearly halfway between A 

 and C, which represents the distribution of temperature at Station 327, 

 one of the most characteristic in the section at present under considera- 

 tion. The depth at Station 327 is 2900 fathoms, and the depths at the 

 two other stations 2150 and 2200 respectively ; and it will be seen that 

 at the latter stations the bottom-temperatures correspond almost pre- 

 cisely with the temperature at Station 327 at like depths. The isothermo- 

 bath of 2° C. is at the same height, 1500 fathoms, at the two southern 

 stations ; and at the northern station only, near the equator, it sinks to 

 1800 fathoms. The isothermobaths of 2°'5 and 3° C. correspond almost 

 exactly in level at Stations B and C ; at Station A all the isothermobathic 

 lines under that of 4° C. down to the line of 1° C. are much lower than 

 at Stations B and C ; that is to say, that at the equator between 4° O. 

 and 1° C. the water is considerably warmer than it is further south. 

 The isothermobathic lines of 4° and 5° C. seem everywhere in the 

 Atlantic to mark broadly the line of demarcation between the upper 

 zone, where the temperatures are obviously affected by the diffusion of 

 water by wind-currents, and the lower zone, where the temperatures are 

 continuous with those of the Southern Sea. In the North Atlantic they 

 are markedly lower than they are to the south of the equator ; that is to 

 say, there is a much larger body of water above them heated by con- 

 duction, convection, and mixture. 



I have given in Plate 33 a general diagrammatic scheme of the verti- 

 cal distribution of temperature in the western trough of the Atlantic, 

 constructed from our serial temperature-soundings in the years 1873 

 and 1876. The diagram extends from lat. 42° S. to lat. 42° N., and in- 

 cludes as its southern limit the banking up of the Antarctic indraught 

 against the South- American coast, and as its northern the " cold wall " 

 of the Labrador current. This Plate seems scarcely to require much 

 comment ; but there are one or two points to which I would wish to 

 direct attention. 



At Station 327, so often referred to as a typical sounding on our present 

 line, there is a mass of water 400 fathoms thick at a temperature below 

 zero. Station 131, 600 miles south of Martin Vas, is on the edge of the 

 central plateau, with a depth of 2275 fathoms and a bottom-temperature 

 of 0°*7 0.; the deep passage therefore probably lies, if it have not 



