22 THE VOYAGE OF H.M.S. CHALLENGER. 





Sources of the large effect obtained in the Press. 



Now comes the question (no longer important to the Challenger work, but of great 

 scientific interest), "What are these various sources, and how much of the effect is clue to 

 each ? First of all we have seen that the water in the press is heated when pressure is 

 applied. Using Sir William Thomson's formula I found the amount of that heating- 

 should be about 0°-05 F. at 43° F., 0°'16 at 50°, and only 0°-3 at 59°, for one ton of 

 pressure. [These numbers, as will be seen in Appendix B, are rather too small. AVe 

 do not yet know to what extent the temperature of the maximum density point of water 

 is lowered by pressure.] These cannot be expected to be fully shown under the circum- 

 stances of the experiments, and even if they were fully shown the greatest of them 

 represents only about one-half of the whole of Captain Davis' result; there must therefore 

 be some other cause. 



I next thought of the heat produced by pumping water into the pressure vessel. 

 That vessel, as is shown in the cut, is connected to the pump by means of a long narrow 

 tube of copper, leading to another long narrow tube in the lower end of the vessel. 

 These tubes were from ^th to -j^th inch in bore, and nearly three feet in length. To 

 estimate roughly the heat developed by the pumping, I calculated that about 300 strokes 

 of the pump had to be made in getting up a pressure of about three tons, each stroke 

 though about 2 feet, and with a mean or average pressure of about 20 lbs. weight. If the 

 whole work done in that way had been expended in heating the water, the temperature 

 effect would have been about 0° - 5 F., as it Avas due to 12,000 foot lbs. of work done 

 on about 30 lbs. of water. But this is an overestimate, and besides a very large portion 

 of the heat actually developed is given to the pump and the connecting tubes, and much 

 of the rest is at once conducted away by the walls of the massive cylinder, and thus the 

 rise of temperature due to this cause is exceedingly small. 



The direct compression of the thermometer tubes already referred to accounts for on 

 the average about 0°"25 F. per ton pressure, so that there is still a considerable part of 

 the whole result to be accounted for. I saw at once that it must be due in part at least 

 to the glass protecting bulb, and perhaps also in part to the vulcanite on which the 

 thermometers are mounted. In order to verify the latter hypothesis I took one of the 

 Challenger thermometers and embedded the bulb of it (protecting case and all) in a mass 

 of lard. I was sure the lard would act as a perfectly plastic body under these great 

 pressures, and so could do no harm to the bulb. The result far more than answered my 

 expectations, because by a pressure of not more than 3 tons the effect on the lard-covered 

 thermometer was over 5° F. 



This showed me at once that we were working in a new sort of world, where all things 

 had properties very different from those they show under ordinary pressures, and therefore 

 I began to think it possible that the vulcanite might have a large share in the residual effects. 



