MECHANICAL ANALYSES 415 



events in the earth's history. Fdden^ and others have already pointed 

 out some significant features and relationships, such as the high perfec- 

 tion of sorting of wind-bknvn and beach materials and the significance of 

 position and magnitude of secondary maxima in the diagrammatic repre- 

 sentation of the analyses. Groups of analyses of the same or closely 

 related materials furnish characteristics not made clear and reliable by 

 single analyses.* 



Mechanical analyses are most easily iiscd when plotted in diagrams, of 

 wliich there are three common types — direct, cumulative, and logarith- 

 mic. Each type has its special uses, and it will probably be found desir- 

 able to use all three for most sets of analyses. The composite cumulative 

 plot or monogram for a group of closely related samples seems especially 

 likely to display diagnostic features and thereby furnish reliable data 

 concerning the origin of the deposit. 



However, what can be done with analyses plotted in any form is still 

 largely unknown. The field is, for certain definite reasons, to be looked 

 on as one of great promise. Several significant features have already 

 been discovered and the variety in analyses is so great and of such a 

 nature that it is almost certain that other significant features can be 

 discovered. We need additional and more detailed and standardized 

 analyses of material deposited under various known conditions. 



Eeference Collection 



If a central laboratory were established the need would immediately 

 arise for a representative collection of samples. Certainly a portion of 

 each of the specimens analyzed should be preserved and, if possible, a 

 portion of each fraction separated in analysis. x\lso, it would seem to be 

 highly desirable to assemble a reference collection of sedimentary rocks 

 that would be representative, both geographically and lithologically. 

 True, every deposit seems to differ from every other and a collection con- 

 taining a specimen of each variety may be impracticable, but if we had a 

 dozen well selected specimens from every county in the country, might it 

 not be well for the collection to be of the same order of magnitude as 

 collections of fossils, minerals, or igneous rocks? The specimens should, 

 of course, be large enough to supply some working material ; some speci- 

 mens should weigh two or three pounds, others need weigh only a few 

 ounces. Generally a thimbleful of fine sand or silt is sufficient for an 

 analysis. ' 



« J. A. Udden : Loc. cit. 



*E. W. Sbaw : The significance of sorting in sedimentary deposits. Bull. Geol. Soc. 

 Am., vol. 28, no. 4. pp. 925-932. 



