THE FORMATION OF SANDSTONE OONORETIONS. 



By George ]\ Merrill, 



Curator of the Department of Geology. 



I 



Many an interesting and instructive lesson in geological processes is 

 frequently to be gained by observation of what is going on almost at 

 our doors, but which is overlooked by the .unateur because his atten- 

 tion has never been properly directed to it, and perhaps by the profes- 

 sional as well, because, as is so frequently the case, he is more inter- 

 ested in larger problems at a distance. 



Such a lesson may be learned from the study of the globular and 

 irregular rounded masses or concretions of ferruginous sand, sometimes 

 quite hollow, or again partially filled with loose sand which falls out 

 when the concretion is broken, leaving but the empty, deeply convex 

 shells. As to what these are and what their method of formation may 

 be, one may consult his geology long and in vain for a satisfactory 

 solution. 



The abandoned reservoir for the waterworks extension near Howard 

 University, in Washington, D. C, furnishes in all its details so plain 

 and interesting an explanation that he who runs may read, and I am 

 temi)ted to describe it in detail even at the risk of wearying those to 

 whom the illustration is neither new or needed. 



The excavation above noted was made in the so-called Potomac divis- 

 ion of the Cretaceous, consisting here of rather loose beds of sand and 

 gravel, containing not infre([uently fossilized logs of considerable size, 

 both silicifled and in tlie partially carbonized state known as lignite. 

 It is with the last, only, that we have to do here. 



In close contact with these lignites, either in the form of rounded 

 and irregular nodular masses or as veins in the mass itself, are numer- 

 ous globular aggregates of siliceous sand and iron di sulphide in the 

 form known as marcasite. (See tigs. 1, 2, 3, and 4 of PI. vi). So 

 long as protected from atmospheric intiuences, such seem to have 

 preserved their mineralogical identity fairly well. When disturbed, 

 however, either in the work of excavation or through other means, so as 



Proceedings National Museum, Vol. XVII— No. 987. 



87 



