124 PROCEEDINGS OF THE AMERICAN ACADEMY 



IX. 



ON THE PROCESS OF REVERSE FILTERING AND ITS 

 APPLICATION TO LARGE MASSES OF MATERIAL. 



By Josiah p. Cooke, Jr., 



Erving Professor of Chemistry and Mineralogy in Harvard College. 



Presented May 30, 1876. 



Bt reverse filtering is meant a process of filtration in which the liquid 

 to be filtered is drawn upwards instead of flowing downwards in the 

 usual way. Such a system is often used in the arts, as when a porous 

 septum is attached to the mouth of a suction pipe ; or as in the small 

 portable filters so useful to travellers, by which clear water may be 

 sucked up from a muddy pool or turbid stream. These last suggested 

 the application of the same principle in chemical analysis to the treat- 

 ment of those precipitates which are usually weighed on a dried 

 filter. In such cases, it is of course essential that the weight of the 

 paper disk used as a filter should remain invariable ; and this constancy 

 can be best secured by making the disk as small as possible. If the 

 filter is large, it is impossible to have any confident assurance of the 

 constancy of its weight, however great the care that may be taken to 

 secure a similarity of hygrometric conditions at the two weighings ; and 

 hence it has not hitherto been practicable to determine on a dried filter 

 the weight of any considerable quantity of a precipitate with accuracy. 

 But, in the process of reverse filtering, we can both wash and collect 

 very large masses of precipitates with a filter not more than an inch iu 

 diameter ; and if, before drying, these little disks of paper are soaked in 

 dilute hydrochloric acid, and afterwards thoroughly washed in water, 

 their weight remains practically invariable. Indeed, it is not necessary 

 to enclose the filter in a weighing tube, or to pay any special regard to 

 its hygrometric conditions other than to keep the usual drying materials 

 in the balance case. The only liability to alteration of weight would 

 arise from the dissolving of soluble material in the paper, and this may 

 be wholly prevented by previously washing the disks as just described. 

 In the early part of 1873, having occasion to determine large quan- 



