678 PROCEEDINGS OF THE AMERICAN ACADEMY. 



ing distilled water until the solution was almost saturated. The hot 

 solution was filtered through asbestos in a platinum Gooch crucible 

 and allowed to cool in low crystallizing dishes. After inoculating the 

 solution with several minute sugar crystals, the substance began to 

 crystallize slowly in fine granular crystals. Several days afterwards 

 the sugar was separated from the syrup by filtration through a Gooch 

 crucible with suction, being washed with distilled water several times 

 during the process. "When the greater part of the solution had been 

 drawn off, the sugar was put into short funnels fitted with platinum 

 cones and subjected to centrifugal action. During the whirling it was 

 several times washed and stirred with distilled water. The sample 

 was desiccated, powdered in an agate mortar, and preserved in a 

 desiccator. 



Because this sample, A, might be not completely free from included 

 water, another sample of sugar was crystallized from its solution in a 

 mixture of water and ethyl alcohol which had per gram the same heat 

 of combustion as sugar. In this case, if the sugar included some of the 

 liquid from which it was crystallized, the weighed mother-liquor would 

 give the same heat of combustion as an equal weight of sugar. This 

 preparation was conducted in the following way. A concentrated solu- 

 tion was made fi'om powdered " rock-candy " and a known quantity of 

 hot distilled water ; and while the solution was still hot, it was filtered 

 through asbestos in a platinum Gooch crucible. To this solution was 

 added such a weight of absolute alcohol that the mixture of water and 

 alcohol should have the same heat of combustion as sugar. After 

 standing a short time, the sugar began to separate out as fine granular 

 crystals. The dishes with the sugar solution were covered, and at the 

 end of several days the sugar was filtered off and treated in the same 

 manner as sample A, except that it was washed with a mixture of 

 water and alcohol having the same heat of combustion as sugar. 



The third sample, C, was prepared at a different time and with differ- 

 ent material, but in a manner similar to the latter method. 



Benzol, CgHg. Two samples of benzol were prepared as follows. 

 For the first, sample A, 500 grams of Merck's benzol, freed from 

 thiophene, were agitated in a flask submerged in a bath of ice and water 

 until about two thirds had crystallized. The crystals were allowed to 

 drain slowly, so that any adhering benzol might be washed off by the 

 liquid formed from the melting crystals. The crystals were then 

 allowed to melt and the above process repeated three more times. 



For the second, sample B, the same quantity of Kahlbaum's similar 

 preparation was agitated in a flask submerged in a bath of ice and 

 water until about one third of the benzol had crystallized. These crys- 



