11. THE FLOW OF HEAT THROUGH THE FLOOR OF THE OCEAN 



E. C. BULLARD 



1. Method 



Th(; rate of flow of heat tlii'ough unit area of the floor of the ocean is deter- 

 mined as the product of a temperature gradient and a thermal conductivity. 

 The temperature gradient is measured by forcing a probe into the ocean floor. 

 The probe contains two or more temperature-sensing elements whose readings 

 are recorded by a recorder housed in a pressure -tight container at the top of 

 the probe. 



The thermal conductivity is determined by measurement on samples brought 

 up by a coring tube. 



2. The Temperature Probe 



Most work published up to the present time has been done with cylindrical 

 probes. Those used in the Pacific from the ships of the Scripps Institution 

 have the junctions 1.4 to 2.7 m apart and are 1.9 to 4.2 cm in diameter ; the one 

 used in Discovery // is 4.6 m long and has external and internal diameters of 2.7 

 and 1.1 cm. A detailed description of the latter probe has been given by Bullard 

 (1954). The recorder is contained in a pressure-tight case, to the bottom of 

 which the probe is attached as shown in Fig. 1. The case carries fins to ensure 

 stability when running down through the water ; instability might occur after 

 the point has entered the sediment, but records from an inclinometer in the 

 recording case show that, in practice, the probe is within a few degrees of 

 the vertical after penetration. A similar apparatus has been described by Uyeda 

 et al. (1961) and by Von Herzen et al. (1962). 



Recently, Gerard, Langseth and Ewing (1962) at the Lamont Observatory 

 have constructed an apparatus in which the temperature gradient is measured 

 by an attachment to a corer. This has the great advantage that the total time 

 taken is greatly reduced and the core is obtained at the exact point where the 

 temperature measurement is made. To avoid the temperature disturbance 

 produced by the corer the thermistors used for measuring the temperature 

 gradient are carried on fins projecting from the coring tube. The thermistors 

 are in tubes about 3 mm in diameter projecting 2.5 cm below the fins and about 

 7 cm from the coring tube. 



The temperature-sensitive element in a probe may be a tliermocoui)le or a 

 thermistor. As the temperature gradient is usually about 50°C/km, the tempera- 

 ture difference between two points 4 m apart is only 0.2°C ; a pair of copper- 

 constantan thermo- junctions gives about 30 (j,V/°C and would, therefore, give 

 6 [jlV for this temi3erature difference. Even if several junctions are used in 

 series this is inconveniently small and to get a deflection of a few millimetres 

 requires a galvanometer with a sensitivity of a few tenths of a millimetre ])er 

 microvolt. The galvanometer must be specially balanced to make it insensitive 



{Mis received August, 1960 \ 218 



