PHYSICAL COMPOSITION OF SOILS. 



93 



the muddy water containing the soil in suspension ; and the rate 

 of its flow, together with the velocity of rotation, determines 

 the size of the sediment-granules that will be deposited in the 

 slack-water below the mouth of the delivery tube. The muddy 

 soil-water is kept agitated in a funnel-shaped reservoir by air- 

 bubbles from a constant-pressure chamber. 



While the principle of this instrument is good, it is quite 

 complicated and the results obtainable from it in practice have 

 not as yet been made public. The inventor claims that an 

 analysis may by its means be completed in less than three 

 "hours. 



In all hydraulic elutriators a provision for constant press- 

 ure in the reservoir supplying the current of water is needed ; 

 although in Schone's and some other instruments a gradually 

 decreasing pressure in a plain reservoir is employed. A large 

 glass bottle or carboy fitted with the proper tubes so as to con- 

 stitute a Mariotte's bottle ( in which the air enters near the 

 bottom of the vessel), is a very convenient arrangement. 



Number of Sediments. The number of grain-sizes or sedi- 

 ments into which the soil mass is to be segregated is of course 

 entirely within the option of the operator. Experience has 

 shown that it is unnecessary to discriminate very closely be- 

 tween the several sizes of the coarser portion of the sand, such 

 as those lying between one-fourth and one-half of a millimeter. 

 But below this point, and especially between one-tenth of a 

 millimeter and the clay, a proper discrimination becomes very 

 important. The series first devised by the writer in 1872 is 

 based upon a consecutive doubling of the velocities of the cur- 

 rent from a quarter of a millimeter per second to thirty-two 

 millimeters per second ; the sediment of sixty-four millimeter- 

 velocity corresponding to a diameter of one-half of a milli- 

 meter, will remain in the elutriator. Above this, as before 

 remarked, the sieve (especially when aided by a jet of water) 

 effects a satisfactory segregation. 



The table below shows the elements of these series both as 

 regards current-velocities and maximum quartz-grain diame- 

 ters carried off by each. In a great many cases, however, it is 

 altogether unnecessary to go into such detail, and a subdivision 

 into six or seven divisions is quite sufficient. Such a sub- 



