THE SMITHSONIAN INSTITUTION. 279 



east ; by which circumstance, if my supposition, suggested chiefi}' by the 

 comparison of distances on both sides of the mountain chain, should prove 

 just, the sierra, even keeping its relative situation, would be brought 

 nearer to the hne of direction of the Rocky Mountains than it comes in 

 realit}^ and by the combinations of the two errors the disfigurations of 

 our maps appear to have been doubled. Thus, while the southern 

 terminal ranges of the Rock}^ Mountains have been laid down too far 

 west, the northern terminal ranges of the Sierra Madre have been laid 

 down too far east, and both have been brought nearer to each other 

 than they really are. 



6. Of the latter ranges, the extreme northern spurs, situated south of 

 the middle and lower Gila, are passed by Cook's route on the trail 

 between the Guadalupe pass and Fort Yuma. Near the latter place, 

 or the junction of the Gila and Colorado, the Coast Range of Sonora 

 and Sinaloa, which forms the western foot of the whole Sierra Madre 

 system — a system which, throughout its whole extension, is formed by 

 parallel ranges — has its northern termination. Beyond the Gila and 

 Colorado, however, its direction is continued by a chain of mountains 

 which the traveller on his way through the desert, between the latter 

 river and Carizo creek, has at some distance to his right hand. At 

 a very acute angle it converges with the chain which comes from the 

 peninsula of Lower California, till at last it falls in with it, the San 

 Bernardino peak forming, as I have been assured by persons who have 

 been on the spot, the point of junction. Thus the extreme northwestern 

 spur of the Sierra Madre constitutes what has been called by geologists 

 the San Bernardino range, but has been known to the old Californians 

 under that same name of Sierra Madre, as I have already stated. If, 

 therefore, the Sierra Madre has a northern equivalent, we have to look 

 for it not in the Rocky Mountains but in the Sierra Nevada system. 

 But the real meaning of all these relations will receive more light from 

 their connexion with the more general structure of the w^estern halt of 

 our continent, of which, therefore, 1 shall try to give a few outlines. 



This western half is known to be composed of a great longitudinal 

 basin, extending, in a direction corresponding to the Pacific coast, from 

 the Isthmus of Tehuantepec to the polar region. Through the greater 

 part of its extent it is confined between an eastern and w^estern marginal 

 chain of mountains. The greater part of its surface has an elevation 

 which gives it the character of a table land, and by its marginal 

 chains it is separated from an eastern and a western lateral terrace. 



In California and Oregon, Utah and New Mexico, and in the coun- 

 tries farther to the north, the two marginal chains are clearly and con- 

 spicuously marked by nature. The eastern one is formed by the 

 Rocky Mountains, the western one by the Sierra Nevada, Cascade 

 Mountains, and their more northern equivalents. In Mexico, the w^estern 

 chain is constituted by the Sierra Nevada, and is likewise clearly traced 

 by nature ; but the eastern one, composed of that line of detached and 

 irregular groups and ridges which crosses the Rio Grande from east to 

 west at the narrows and rapids of San Carlos, is less conspicuous, and 

 may be entirely overlooked by those who are not sufficiently informed 

 about the matter. Nevertheless, as already stated, if the Rocky Moun- 

 tains have a southern equivalent, it must be recognised in the mountains 

 of western Texas, Coahuila, Nuevo Leon, San Luis Potosi, and Vera 



