116 The Petrography and Genesis of Sediments 



analyses did not have a reliable accuracy of more than 5 rag., which is 

 the smallest unit to which weights were then recorded. The sample was 

 next washed into an 8 oz. milk-sterilizing bottle with water and generally 

 a little ammonia to help disintegrate the clay. The bottle was then shaken 

 on a rotary shaker. This is simply an axis to which a board is fastened. 

 The bottle is attached to the board at right angles to the axis with the 

 middle of the bottle over the axis, so that when the axis is rotated the 

 sediment and water, which should less than half fill the bottle, flop from 

 one end of the bottle to the other twice in one revolution and by this 

 jarring the sample disintegrates. The method seemed fairly effective; 

 just how effective it is hard to say. Certainly in some of the samples 

 there was a perceptible amount of clay granules, a little in all; they are 

 mentioned in some of the analyses that follow, though they are not con- 

 sistently recorded. But it is a question whether, in some cases at least, 

 these clay granules are not an essential part of the sediment representing 

 some kind of growth or concretion in the clay. That there is a possibility 

 of the existence of such concretions is indicated by the round, clay-like 

 granules with faint aggregate polarization that were found in a few of 

 the samples (see sample 4) and believed to represent a stage intermediate 

 between clay and glauconite. The uncertainty prevailing in the whole 

 matter appears from the difference of opinion concerning the best method 

 of disintegrating clay, Mitscherlich, e. g., recommending that the sample 

 be boiled some fifteen minutes, while others say that only luke-warm 

 water should be used, because hot water coagulates the clays. 



After being shaken ten to thirty hours, according to the apparent 

 amount of clay in the sample, the material is washed out of the bottle into 

 an evaporating dish of 12 cm. diameter. Here it is allowed to settle for a 

 while, the mud decanted into a 1500 c. c. separating funnel, hot wash 

 water added in the evaporating dish, the settling and decantation repeated, 

 etc., several times. The length of time during which the material was 

 allowed to settle varied for different samples and decreased for each suc- 

 cessive deeantaion. If the sample was muddy the writer started with 

 fifteen minutes, allowed ten minutes on the second settling and so on down, 

 depending somewhat on the observed rate of clearing of the upper part of 



