100 



one upon the other, as fed in succession to the press, in likewise thin 

 and equable layers, until the chambers were quite filled and filtration 

 ceased. With oscillatory pressures and with substances of widely differ- 

 ing specific gravities, such as whiting, brown coal, red lead, wood char, 

 and ultramarine, one following upon the other, the various lamina 

 proved most irregular in their deposition upon the filter-bed, being com- 

 paratively of excessive thickness in parts while running out altogether 

 in others, the plane of contact being besides often obliterated or scarcely 

 defined, because of partial intermingling between the different sub- 

 stances employed. The same effects, also, found their cause in the use 

 of any given substance fed alternately in fine and coarse division, or at 

 first in high followed by low percents of the matrix. 



There can be little doubt that for the best results in general filter- 

 press work, this indicates, as afterwards substantiated for sugar liquors 

 by the use of hydrostatic columns on the one hand and intermittency 

 secured through means of a by-pass valve on the other, the first im 

 portance of constant pressures, freed especially from the vibratory pul- 

 sations of ordinary pumps, and a liquid so agitated while awaiting the 

 process as to carry to the press, at all stages of this, a reasonably uni- 

 form percentage of whatever matrix is employed, the laws of hydraulics, 

 as illustrated in silt-bearing streams, here again exhibiting themselves in 

 complete application. 



Satisfied that the mechanical arrangement of the large apparatus was 

 appropriate to the intervention of a matrix and that the small answered 

 to all the essential conditions of the large, systematic work with brown 

 coal, under what is known as the Kleemann process, began on Novem- 

 ber 29. Five long tons of this article had been imported by Mr. Dan'l 

 Thompson, through the Sangerhausen Maschinenfabrick, Germany, 

 which, however, was so superlatively unfit for its destined duty, by 

 reason of uneven and inadequate pulverization, as to have required pre- 

 vious and, of course, laborious hand -sifting. 



It was first sought to learn what relation varying quantities of this 

 article bore to speed in the filtration of defecated but unskimmed juices. 

 With this intent different percentages, based upon the estimated weight 

 of the contained sucrose, as the most convenient, although not, assuredly, 

 the most rational standard of reference, were employed with the-results 

 which follow: 



The average juice per press and per square foot of filtering surface, 

 per twenty-four hours, stand calculated on the basis of a 60-day con- 

 tinuous run. Here, taking the average weight of the juice at 8.85 

 pounds per gallon, and its sucrose at 13J per cent. for percents of lig- 



