SECT. 1] THE FLOW OF HEAT THROUGH THE FLOOR OF THE OCEAN 231 



as the places where the current sinks again. The horizontal limbs joining the 

 rising and sinking currents would then produce a horizontal force on the crust 

 and perhaps tensional cracks, along which the dykes postulated to explain the 

 high heat flow could rise. The irregularity of the heat flow on the ridges is 

 consistent with its being associated with a swarm of dykes. 



The reality of this picture and its connection with a possible widening of the 

 oceans and with continental drift must await a more thorough survey of many 

 features of the ridges. 



The low heat flows observed in the Acapulco and South American trenches 

 have no obvious explanation. A large thickness of poorly conducting sediments 

 in the bottom of a trench would present a large thermal resistance and would 

 tend to divert the heat flow and produce a reduced flow in the trench and an 

 increased flow on each side. However, seismic studies do not show an excep- 

 tional thickness of sediments in these trenches and it seems that this explana- 

 tion cannot be the correct one. If South America were overriding the Pacific, 

 cold surface rocks from the ocean floor would be forced beneath the continental 

 margin and the heat flow would be reduced. With a rate of drift of 0.2° of arc 

 in 10^ years (2.2 cm/year), which is of the order suggested by palaeomagnetism, 

 the width of the trench (60 km) would be traversed in 3x10^ years. The 

 thermal time constant for a sheet of rock 5 km thick is of the order of 2 x 10^ 

 years, so that this rather fanciful explanation cannot be immediately rejected 

 as thermaUy impossible. The high value of 2.72 {jical/cm^ sec shown in Fig. 7 at 

 the station on the landward slope of the trench in latitude 13°S would be 

 difficult to explain on this theory. 



5. Future Work 



If the distortion of the rocks of the earth represents the working of a heat 

 engine, the measurement of the variations of heat flow from place to place are 

 of great importance. The present methods are tedious and do not always yield 

 satisfactory results. More reliable and more easily used equipment should be 

 developed. 



The most urgent programme of measurement is to determine how generally 

 high heat flows occur along ridges and to make measurements on closely 

 spaced sections across them. Measurements in the Red Sea and along its junc- 

 tion with the ridges of the Indian Ocean would also be particularly interesting. 



References 



Bullard, E. C, 1954. The flow of heat through the floor of the Atlantic Ocean. Proc. Roy. 



Soc. London, A222, 408-429. 

 Bullard, E. C. and A. A. Day, 1961. The flow of heat through the floor of the Atlantic 



Ocean. GeopJnjs. J., 4 {The Earth Today), 282-292. 

 Bullard, E. C, A. E. Maxwell and R. Revelle, 1956. Heat flow through the deep sea floor. 



Advances in Qeophys., 3, 153-181. 

 Cox, R. A, and N. D. Smith, 1959. The specific heat of sea water. Proc. Roy. Soc. London, 



A252, 51-62. 



