562 



SCIENCE 



[N. S. Vol. XXXIII. No. 850 



limits of the rhyolite at Soda Butte 

 Spring, and again near the western 

 border. 



On the other hand, wherever the heated 

 waters issue from the rhyolite of the table- 

 land they are characterized by a high per- 

 centage of silica. These waters occur 

 distributed over a wide area and furnish 

 the great volume of water discharged from 

 the geysers and hot springs, and for this 

 reason have excited more general interest 

 than the smaller springs. They have sup- 

 plied the silica for the many square miles 

 of glittering white sinter plains. For the 

 most part they are siliceous alkaline waters, 

 as in the Upper Geyser Basin and the Fire- 

 hole Basin. They may, however, be slightly 

 acid or neutral, as in the case of many of 

 them in Norris Basin. 



The silica occurs in solution as hydrated 

 silica associated with carbonates and 

 chlorides of the alkalies, together with 

 small quantities of sulphates. Arsenic and 

 boron have been determined in nearly all 

 geyser waters, probably combined with 

 soda as arsenates and borates. Traces of 

 bromine, phosphoric acid, soda, man- 

 ganese, lithium, cEesium and rubidium 

 were detected in several instances, but 

 lithium and bromine are the only elements 

 present in sufficient quantities to allow of 

 estimation. Tests were made for titanic 

 acid, nitric acid, iodine, fluorine, barium, 

 and strontium, but none of them were 

 found. Special examinations were made 

 in concentrated solution for tin, copper, 

 and lead, but no one of them was present. 

 In this connection it may be pointed out 

 that while veins carrying lead, copper, and 

 silver are found associated with Eocene 

 and Miocene igneous rocks, these metals 

 have never been detected in either the 

 rhyolite or waters of the park. 



A study of these chemical analyses brings 

 out clearly the marked differences in per- 



centages of substances held in solution, espe- ■ 

 cially silica, even in adjoining geysers. ■ 

 This holds equally well for the siliceous 

 alkaline waters from the same geyser basin 

 as from those collected from different lo- 

 calities. Waters examined the same year 

 show as great variations as those collected 

 one or two years apart. The silica, as de- 

 termined by analysis, ranges from .22 to 

 .67 gram per kilogram of water, the 

 former being the amount found in the 

 cauldron of the Excelsior, having the larg- 

 est outflow of any pool in the park, and the 

 latter from Opal Spring in Norris Basin, 

 with but slight run-off, and without any 

 apparent inflow. The cause of these differ- 

 ences is, I believe, to be sought in the 

 varying amounts of infiltrating surface 

 water. 



Dr. W. H. Hallock has shown conclu- 

 sively, by experimental tests with self- 

 registering thermometers, that the thermal 

 waters stored in underground geyser reser- 

 voirs possess a temperature far in excess 

 of the boiling point at the surface, due to 

 increased pressure of the overlying column 

 of water in the geyser tube. The results 

 were in accord with the theoretical boiling 

 point. It can not be affirmed positively 

 that these superheated waters maintain 

 the same composition after being thrown 

 out as the underground waters at greater 

 or less depth or even with those of the 

 geyser reservoirs. On the other hand, 

 there exists no evidence of chemical 

 changes due to relief of pressure before 

 the waters reach the surface through geyser 

 orifices. 



Upon the relief of pressure, hydrated 

 silica, associated with traces of an equally 

 insoluble silicate of alumina and lime, is 

 deposited upon the broad plains of the 

 geyser basins. Nearly all remaining con- 

 stituents are carried away in solution by 

 surface streams. Although the composi- 



