43 UNIVERSITY OP VIRGINIA PUBLICATIONS 



The materials named underwent a preliminary drying at 100° C, were 

 then rubbed to a fine powder in separate warm porcelain mortars, were 

 again kept at 100° for an hour or two, and were then cautiously introduced 

 from opposite ends into a straight piece of glass tube, of about a centimetre 

 in diameter and 55 or 60 centimetres in length, haviug a clear interval of 

 about 10 centimetres between the two powders. This tube was in the case of 

 the lead nitrate experiment heated throughout the greater part of its 

 length to 100°, and a slow stream of air was drawn through by means of a 

 jet pump, this air passing first through a train of drying vessels containing 

 in succession sulphuric acid, calcium chloride and phosphorus pentoxide, 

 while a like train of dryers in inverse order was iaterposed between the tube 

 enclosing the powders and the jet pump. In a first trial of the mercuric 

 chloride it was soon found that at 100° enough vapour was given off from 

 this salt before long to redden perceptibly the surface of the potassium 

 iodide, so that in this case the experiment had to be made at the tempera- 

 ture of the room — about 25° C. In all other respects the drying arrange- 

 ments were the same for both the lead and mercury experiments. The slow 

 stream of air was kept up in each case for many hours, generally for about 

 36 hours. At the end of this time the tube was disconnected from the 

 drying trains, its ends instantly closed by corks taken directly from a 

 hot-air drying-bath, and both ends with the least possible delay drawn off 

 and sealed in the blast-lamp flame, slightly slackening the second cork to 

 avoid blowing out of the glass in the final sealing. Great care was taken ia 

 handling the tube to keep it in a horizontal position, so that there should 

 be no disturbance of the two parcels of material, and the tubes were laid 

 aside in this position in a locked cupboard for a little more than four 

 months, during which time the atmospheric temperature ranged from 

 about 20° to a little over 32° C. 



At the end of this time there was no apparent development of colour, 

 the two parcels of powder in each being quite white as at first. This indi- 

 cated that change due to volatilization had not occurred. The tubes were 

 then shaken up so as to mix the materials together. On shaking for several 

 minutes a slight reddish tinge became perceptible in the mixture of mer- 

 curic chloride and potassium iodide, this colour becoming much more de- 

 cided on allowdng the tube to lie at rest for several hours. No yellow colour 

 was observable in the mixture of lead nitrate and potassium iodide on first 

 shaking, and only after a subsequent rest of several hours could the colour 

 be clearly recognized; after 20 hours the change was very perceptible. In 



