296 N. S. SHALER BROAD VALLEYS OF THE CORDILLERAS 



from the rea. It is a well known fact that no such explosions have been 

 noted in the historic period at a greater distance than about 250 miles 

 from the ocean ; 3^et many of the eruptions in tlie Cordilleran district 

 which have recently occurred are several times as far from the coast. 

 The only apparent reason for the limitation of volcanoes to the sea- 

 floors or to the margins of tlie hinds is that the development of the ten- 

 sion, principall}'^ of steam, that brings about eru[)tions is caused by the 

 rise of temperature in rocks due to the non-conducting effect of strata 

 which have been laid down on the sea-floors. Accepting this explana- 

 tion, it ap[)ears not unlikel}'^ that the outbreaks which have taken place 

 in the region we are considering may have been caused by the great de- 

 posits which have been accumulated over the wide area of the broad val- 

 leys. It is true that the rise in temperature would not in general exceed 

 one degree Fahrenheit for each 50 feet of beds laid down, and that a 

 depth of 10,000 feet would probably effect no more increase in the base- 

 ment of the section than 200 degrees Fahrenheit ; yet such again in heat 

 might at a de[)th in the crust when the tensions were alread}'^ near the 

 critical part of an explosion be efficient in bringing it about. Moreover, 

 as is indicated by the hot springs in this region, there are evidentl}^ many 

 large areas where the temperatures at no great depth are high, a condi- 

 tion due perliaps to ancient volcanic pipes or great dikes or sheets of lava 

 which are still very hot. In case these were deeply covered with sedi- 

 ments the rate of rise of the isogeothermal planes would be much more 

 rapid than wouUl ordinary conditions, and an explosive tension would 

 be more readily attained. 



It may ))e asked why the eruptions from the interior districts of the 

 Cordilleras have a})parently now ceased, while they were evidently of 

 frequent occurrence in the later Tertiary and perhaps in the early Cre- 

 taceous periods. The answer may be found in the fact that the process 

 of deposition in the broad valleys was evidently arrested during the 

 humid })eriod, of which the local glaciation of the region was an inci- 

 dent, and that for some time and in a considerable measure it was re- 

 placed by erosion. The further deposition now going on has evidently 

 not been sufficient to restore to these valle3^s what was removed during 

 that time of more abundant rainfall. The only sufficient verification of 

 the hypothesis that these inland volcanoes owed their origin to local 

 deposition wmU be had in an examination of other countries where like 

 deeply tilled broad valle3^s exist. I have essaj^ed this through records, 

 but these are so imperfect that they have not proved serviceable. 



Some of the at)parently modern eruptions of the Cordilleran region 

 appear to negative this hypothesis, as, for instance, those which have 

 occurred in the Colorado can3^on. I have not been able to examine 



