PEALE.] buxsen's geysee theory. 419 



jected, but the temperature depends on the time that has elapsed since 

 the last eruption. As a great eruption comes near it approaches the 

 boiling point. 



(3) At a depth of about 45 feet the difference between the tempera- 

 ture of the water and the calculated boiling point for that i)ressure is 

 the least. 



Mr. Walker, in 1874, verified Bunsen's results. His temperatures are 

 given in Fig. 22, on the left. He found the smallest diiference at a 

 depth of about 39 feet. 



In Fig. 31 these temperatures are shown in curves. The tempera- 

 tures are represented on A B, and the depth of the tube is measured 

 on B D. The curve a h rei)resents the calculated boiling points at 

 the difi'erent depths. The following are the other curves (the letters 

 referring to those in Fig. 31): 



7i- Iz shows the temperatures recorded by Walker. 



The following are the curves showing Bunsen and Descloizeau's tem- 

 peratures : 



c m d, July 6, 1846, 0^. 30™. a. m., three hours after a great eruption 

 and eleven hours' before the next one. 



c e, July 6, 1846, 8.20 a. m., nine hours after a great eruption and 

 twenty-three hours before the next one. 



c/, July 7, 1846, 2.55 p. m., four hours before a great eruption. 



G g, July 7, 1846, 6.58 p. m., ten minutes before a great eruption. 



e s d, July 7, 1846, 9.45 p. m., two hours after a great eruption. 



Of these curves, c e is farthest from the curve of the boiling point, 

 and it shows the temperatures at twenty-three hours before an eruption ; 

 c g approaches the curve of the boiling point most closely, and records 

 the state of the tube at ten minutes before the erui^tion takes place, 



Kow, this being the state just before the eruption takes place, there 

 can be no doubt that at the time it does occur, it must actually reach 

 the boiling point, L e., the curve c g must touch the curve a h. Bun- 

 sen accounts for this as follows: He supposes that the column in the 

 central tube communicates by a long and sinuous channel with some 

 space, be it what it may, which is subjected to the action of the direct 

 source of the subterranean heat. The temperature gets raised above 

 the boiling ])oint, due to the pressure, and a sudden generation of steam 

 is the result. This steam rises in the column of water, which, being 

 cooler, causes it to condense. Gradually the heat of the water is 

 raised until the water of the channel must boil, and the steam therefore 

 cannot condense, but must accumulate and acquire a gradually increas- 

 ing tension. The condensation of the bubbles possesses a periodic 

 character, and to this is due the uplifting of the water in what Bunsen 

 calls conical water-hills, which are accompanied by the subterranean 

 explosions. The water in the tube, by the influx of heated water, is 

 gradually brought so near the respective boiling points for the pres- 

 sures under which it is that a A^ery slight shock is sulhcient to carry a 

 certain layer to a point of the tube where its temperature would be in 

 excess of that required to vaporize it, and the result would be a sud- 

 den and violent generation of steam. A rise of only a few feet near 

 the middle of the column would be sufficient to cause the instantaneous 

 generation of an additional volume of steam, which would rush ui)wards 

 and thus relieve the pressure of the strata beneath and thereby contrib- 

 ute to the production of more steam. The enormous force (calculated 

 to be equal to that produced by an engine of 700-horse i>ower) is not 

 expended in one effort, but there are several successive eruptions, the 



