104 THE PLIOCENE AND PLEISTOCENE DEPOSITS OF MARYLAND 



boulders in a deposit does not indicate that it should be correlated with 

 the Lafayette rather than with the Columbia formation. For not only 

 is the greater mass of the Columbia deposits sealed to the geologist from 

 lack of erosion, but also the boulders are not equally distributed through- 

 out it. They are confined mostly to the greater drainage lines, while in 

 many other regions they are absent. If then outliers should be found, 

 which on examination fail to reveal boulders, it could not for that reason 

 alone be separated from the Columbia and correlated with the Lafayette. 

 In regard to employing the kind of materials occurring in any one of 

 the surficial formations to separate it from another, it may be added that 

 this method has been tried and found inadequate. The most that can be 

 said in favor of it is, that the gravels in the Lafayette formations are 

 composed almost entirely of sandstone, quartz, and quartzite pebbles 

 while the coarser materials in the other terraces frequently contain much 

 gneiss, gabbro, granite, etc., although these are by no means everywhere 

 present. It would not be surprising, however, to find somewhere in the 

 Lafayette a considerable admixture of gravel having a complex mineralog- 

 ical composition. Among the various formations of the Columbia group 

 no means of separation founded on the mineralogical composition of the 

 deposits has been found to hold in Maryland and the same is true in 

 Pennsylvania and Few Jersey. 



STAGE OP DECOMPOSITION. 



Another method of correlating the surficial deposits is by comparing 

 the stage of decomposition exhibited in each. The value of this method 

 rests on the assumption that the oldest formation- or the one first de- 

 posited, has been exposed longest to the chemical action of surface waters 

 and has consequently reached a further stage of decomposition than any 

 of the other terrace deposits, each one of which should show a more ad- 

 vanced stage of decay than the one immediately succeeding it. In sub- 

 jecting this method to a practical lest in the field, it appears that the 

 assumption on which it is founded is too sweeping. There are no doubt 

 certain classes of deposits where discriminations by the stage of decom- 

 position can be applied with confidence. In many glacial deposits for 



