CORRELATION OF FORMATIONS 473 



to the lower Dolores valley, with but two gaps of a few miles each for 

 the two formations below the Dakota. The following general statement 

 of Spencer's observations is made with his permission. 



The La Plata formation was found to increase in the thickness of its 

 sandstone members going down the San Miguel valley, while the calca- 

 reous medial member disappears as such, but seems to be replaced by 

 sandstones of thin bedding. In Paradox valley the La Plata formation 

 has three parts. The upper is 400 to 450 feet in thickness, white in its 

 upper portion, and stained red or orange below. It is mainly massive, 

 but is often friable and crumbling in its lower portion. 



The middle member is reddish sandstone, shaly at the top with a few 

 chert bands. Thickness, 250 feet. 



The lower sandstone is from 300 to 350 feet thick, of the typical massive 

 and cross-bedded texture common in this part of the formation. In 

 some places this sandstone is white, but it is more frequently reddish or 

 orange-colored through weathering, while nearly white within. The La 

 Plata thus becomes a formation of about 1,000 feet in thickness, and its 

 massive portions form abrupt cliffs of red, orange, yellow, gray, or white 

 colors in different places. 



Above the La Plata Spencer found the McElmo formation to be about 

 500 feet in thickness, with variable relations of sandstone and shale. It 

 has a decided red color locally. 



Below the La Plata, distinguished by Spencer, there occurs a much 

 deeper red sandstone formation, vermilion or brick-red in hue, and cor- 

 responding to the upper division of the Dolores as it has been described 

 in this paper. Apparently the lower part, characterized near the San 

 Juan mountains by the variable strata carrying fossiliferous limestone 

 conglomerate, is not, on the Dolores, markedly different from the upper 

 in texture or color. Spencer found limestone conglomerate rich in bone 

 fragments in several places, and on La Sal creek this horizon was but 

 about 100 feet below the La Plata sandstone. The total thickness of the 

 nearly uniform Dolores formation is about 1,000 feet in the neighbor- 

 hood of Sinbad valley, while much less in other localities. 



Below the Dolores beds Spencer found coarser Red beds, often con- 

 glomeratic, with pebbles 3 inches or more in diameter, and several hun- 

 dred feet of such strata were noted. No opportunity was found to 

 measure a section showing the full thickness of these coarser Red beds, 

 but, as observed by Peale, they are underlain by fossiliferous Pennsyl- 

 vanian Carboniferous in Sinbad valley, where there is also much struc- 

 tural complexity obscuring the relations. 



Spencer's observations seem to show that the section of the lower 

 Dolores valley embraces strata to be correlated with the Cutler, Dolores, 

 La Plata, and McElmo formations of the San Juan region. 



