222 BULLA.RD [chap. 11 



that would be attained if the initial frictional heat spread instantaneously 

 through the probe ; in fact, the thermocouple never reaches this temperature 

 since some heat is lost to the sediment before it has had time to penetrate to 

 the centre of the probe. For example, in Fig. 3, the observed maximum tem- 

 perature difference is O.SrC and occurs after 3 min, whilst (^i+^o) is 0.5rC. 

 The equilibrium temperature difference, di, is 0.251°C, whilst the last observed 

 temperature difference was 0.266°C; the correction is, therefore, 6%. Experi- 

 ments with a probe in which the temperature gradient could be measured both 

 over the whole length and over the bottom half have shown the latter to be 8% 

 greater than the former (Bullard and Day, 1961). The reason for this is un- 

 known ; bacterial action in the surface layers and the increase of conductivity 

 with depth would produce a difference in the opposite direction. It may be that 

 in the upper 2 m some of the heat is carried by movement of water. In view of 

 the large variation in the observed heat flow from place to place these un- 

 certainties in measuring the gradient and in the extrapolation to find the 

 equilibrium gradient are relatively unimportant. 



The miniature probes attached to a corer reach equilibrium in under a 

 minute and no extrapolation is needed to find the steady state; their thermal 

 properties have been studied in the laboratory by Uyeda et al. (1961). 



The reliability of temperature measurements made with probes has recently 

 been confirmed by measurements made in the 600-ft trial boring for the Mohole. 



3. The Thermal Properties of Sediments 



It would be possible to measure the thermal conductivity of the sediments 

 of the ocean floor by generating a known amount of heat in a probe and follow- 

 ing the subsequent change in temperature; this experiment has never been 

 carried out and aU measurements have been made on samples obtained with a 

 coring tube. 



Two methods of measurement have been used to obtain the thermal con- 

 ductivity of samples. Butler and Ratcliflfe at the National Physical Laboratory 

 have used a static method with disc-shaped specimens (Ratcliffe, 1960). This 

 method is the one normally employed for precise absolute measurements of the 

 conductivity of poor conductors and is believed to be free from serious syste- 

 matic error. It is, however, tedious as it is necessary to wait for the apjjaratus 

 to attain temperature equilibrium. A transient method has, therefore, been 

 developed (Von Herzen and Maxwell, 1959) in which a hypodermic needle 

 0.086 cm in diameter and 6 cm long is inserted into a piece of core and its 

 temperature measured as a function of time while heat is dissipated at a constant 

 rate, Q, by a heater within the needle. If r is the radius of the needle and k 

 the conductivity of the sediment, the temperature, d, at time t is 



Q , 2.246/c^ 

 4:7Tk r- 



M'here k is the thermometric conductivity of the sediment. This expression is a 



