CLASSIFICATION OF SAND GRAINS 627 



thrown together under one heading. Similarly aqneoiis, seolian, oolitic 

 and miTch volcanic sand would have to be grouped together because of 

 general similarity in the shape of the grannies. Chemical composition 

 is almost as unsatisfactory as a basis of sand classification for similar 

 reasons. In the classification adopted it was found that the main varie- 

 ties of sand commonly encountered may be distributed under seven dis- 

 tinct types; but since the sand-producing agencies are in operation more 

 or less continuously, these types may be still further acted on by one 

 or more of the other agencies, and intermediate siih types be produced. 

 To distinguish these subtypes a system of compound terms is proposed, in 

 which the last term designates the type form and the first term indicates 

 the agency by which the type has begun to be modified. In this way a 

 large number of intermediate varieties of sand, many of them of much 

 geological importance, may be designated very simply as soon as their 

 mode of production can be determined. To illustrate: An aqueo- 

 residual sand is one in which the granules have been produced by the 

 various residual agencies and have been subsequently more or less modi- 

 fied by water action. A residuo-aqueons sand, on the other hand, is to 

 be understood as one in which water-rounded granules have been sub- 

 jected "to the agencies of weathering, and give more or less evidence of 

 such action. Through these subtypes the aqueous and residual types of 

 sand are connected, as they are found to be in nature, to the extent that 

 the secondary agency has operated on the original type. If the water 

 action is complete enough to obliterate all traces of the residual action, 

 or if the residual agencies have shattered and destroyed the aqueous sand 

 beyond recognition, then the subtype passes into one of the other types. 

 This same system of classification will be found to apply to pehhles on 

 one hand and the variously formed t}^es of dust on the other, thus 

 bringing out the intimate relation between sand and all such clastic 

 material.* The distinction between these three classes of materials is, of 

 course, entirely arbitrary, and would be made differently by different 

 investigators. In describing a residual sand from IMedford, Massachu- 

 setts, MerrilP makes the following distinctions : Gravel, above 2 milli- 

 meters; fine gravel, 2 to 1 millimeter; coarse sand, 1 to .5 millimeter; 



* As presented in this paper the classification is not absolutely complete, several geo- 

 logically unimportant varieties of sand not heing included, in order to secure the de- 

 sired simplicity. In the manufacture of talus, in avalanches, rock slides, rock and 

 mud flows and earth movements along joint-planes, sand is produced along with coarser 

 and finer material, of a nature quite similar to that due to glacial action. Meteoritlc 

 matter of sand-like texture may also reach the earth and he recognized hy form and 

 composition. It seems quite possible that rounded sand grains might originate by con- 

 cretionary action of the clay-ironstone variety and he segregated by some agency. 



^U. S. Geological Survey, Bulletin No. 150, 1898, p. 380. 



