Apkil 14, 1911] 



SCIENCE 



557 



ferences in the phases of eruption which 

 built up the two volcanic regions. 



On the rhyolite plateau there are no 

 evidences of violent explosive action. The 

 complete absence of true volcanic breccias 

 is a significant feature of these later flows. 

 Dikes, veins and horizontal sills, together 

 with nearly all the phenomena of deep- 

 seated intrusions, are wanting. The rhyo- 

 lite shows scarcely any indications of 

 hydrothermal activity during eruption. 

 In the abrupt escarpments made up of 

 successive sheets there are no signs of sur- 

 face flows having been exposed to long- 

 continued atmospheric agencies, no wind- 

 strewn ashes or any vestiges of vegetation. 

 On the contrary, everything clearly indi- 

 cates a relatively rapid accumulation of 

 viscous masses from the beginning to the 

 end of the rhyolite period. "What im- 

 presses one most is the absence of stages of 

 activity with intervals of quiescence, there 

 being rather a series of massive eraptions 

 piled up one upon another. In the central 

 portion of the park the rhyolites have a 

 maximum thickness of 2,000 feet, and over 

 large areas they may be assumed to meas- 

 ure 1,500 feet. 



Subsequent to the rhyolites and the 

 building up of the park plateau came a 

 few dikes and thin sheets of basalt. They 

 are the most easterly occurrence of those 

 broad basaltic flows that spread over 

 southern Idaho and the Snake River plain. 

 In the park country they are of Pliocene 

 age, that is to say, they are older than the 

 glacial ice. They make the final chapter 

 in the history of Tertiary igneous rocks. 

 As they play no recognized part in the 

 problems bearing upon thermal waters 

 they may be dismissed at the present time 

 with this brief mention. 



Unquestionably the Pleistocene age, with 

 its changed conditions, set in not long after 

 the dying out of rhyolitic eruptions, as is 



shown by the relatively slight erosion of 

 the plateau and the beginning of canyon 

 sculpturing. All geological evidence tends 

 to prove that the rhyolites belong to the 

 Pliocene age. 



DURATION OF THERMAL ACTIVITY 



That the activity of thermal waters was 

 approximately coincident with the cessa- 

 tion of rhyolite ejections is, fortunately, 

 clearly proved by the massive horizontal 

 beds of calcium carbonate laid down on the 

 summit of Terrace Mountain, where they 

 attain a maximum thickness of nearly 250 

 feet, although the average is much less. 

 Without doubt they are the oldest deposits 

 of travertine in the region of Mammoth 

 Hot Springs, and rest directly upon fresh, 

 unaltered rhyolite. Glacial ice from the 

 Gallatin Mountains moving eastward oc- 

 cupied the intervening Swan Lake Valley 

 and passed over the top of Terrace Moun- 

 tain on its way to the broad, open valley of 

 the Yellowstone. On the recession of the 

 glacier fragments of crystalline rocks, un- 

 doubtedly brought down from the Gallatin 

 country, were left strewn over the traver- 

 tine of Terrace Mountain. 



It is a fair assumption that if these 

 thermal waters were issuing through rhyo- 

 lite at one locality in pre-glacial time, 

 similar hot waters and gaseous emanations 

 should have reached the surface at other 

 points on the plateau. If any such rem- 

 nants of sinters still remain it seems im- 

 possible, from our present knowledge, to 

 discriminate between them and those of 

 post-glacial time. Erosion has carried 

 away not only every trace of these earlier 

 deposits, but has removed nearly all evi- 

 dences of pre-glacial rock decomposition. 

 Modifications in topographical relief fail 

 to indicate two distinct periods, owing 

 probably to the relatively slight deposition 

 of sinter before the ice. 



