128 PROCEEDINGS OF THE AMERICAN ACADEMY 



of precipitated antimonious sulphide, we shall require a crucible hav- 

 ing a capacity of 250 cubic centimetres. As much of the water as can 

 readily be decanted from the precipitate is first poured into the cru- 

 cible and drawn off with the filter, and then the precipitate is washed 

 in with as little additional water as possible. Now the filter is 

 plunged into the semi-fluid mass, and must not afterwards be removed 

 until the process is completed. The rose ought not even to be raised 

 however slightly, although additional material may be poured in 

 around it. As the mass contracts in the crucible, the filter must be 

 made to follow, always keeping it immersed ; and during this time the 

 precipitate which has collected around the j^latiuum rose may be 

 washed down by a stream from a wash bottle. In this way the greater 

 part of the water can be removed, leaving the precipitate nearly as 

 compact as it is left on a common filter, when dried by Bunsen's 

 pump. When the precipitate is in this condition, the pressure is 

 relieved by opening the nipper tap, and the rose raised, which leaves 

 the filter behind. If any of the precipitate has clung to the platinum, 

 this must now be washed into the crucible with a few drops of water, 

 the rose and tube having first been detached from the connector for 

 the purpose. It only now remains to dry the precipitate with the 

 •little filter, and weigh it. If it is important to dry the precipitate at 

 a temperature above 150° C, or even to ignite it, the mass should 

 first be thoroughly dried at 100°. The little disk of paper can then 

 be removed and weighed sei^arately, while the rest of the mass is 

 heated to a higher temperature. The amount of material which 

 remains adhering to the paper under these circumstances is exceed- 

 ingly small, not usually exceeding a few milligrammes ; and allowance 

 can be made for it in the final result, without sensible error. 



Like other analytical processes, this method has its limits ; and any 

 attempt to extend it beyond the sphere of its usefulness will lead to 

 unsatisfactory results. It is of no use for filtering turbid liquids, since 

 the small filters are rapidly clogged, and the process becomes propor- 

 tionally slow. Whenever, however, a precipitate settles clearly, this 

 method enables us to wash, collect, and weigh very large quantities of 

 precipitates in a very short time with wonderful accuracy. We have 

 used it chiefly to determine sulphide of antimony and chloride bromide 

 or iodide of silver ; and the results of the following analyses, extracted 

 from our forthcoming paper on the Atomic Weights of Antimony, 

 will show how great accuracy can be attained with it : — 



