366 Dr. J. W. Mallet on the Gases occluded in [May 30, 



from any crust or scale. The shavings were subjected to the purification 

 practised by Graham, namely, washing with a hot solution of potassic 

 hydrate, followed by washing with distilled water and thorough drying. 

 The solid strip of iron was not so treated, care having been taken to use no 

 oil upon the tool employed in cutting it. Both specimens gave off gas 

 readily when heated in the Sprengel vacuum, the amount in each case 

 being larger in proportion to the bulk of the iron than in the experiment of 

 Graham ; and analysis showed that the same gases were present as those 

 found by him, with the addition of carbonic anhydride in not inconsiderable 

 amount. 



The final experiment was made as follows, with great care, and with all 

 precautions which could be thought of to avoid error. 



A parallelopiped of iron was cut upon a planing-machine from the 

 largest of the three masses found (that spoken of as No. 1* in the paper 

 above referred to), the work being done with special care, to avoid the 

 least trace of grease being derived from the machine. 



Not only was the cutting-tool itself made red-hot in the blacksmith's fire, 

 hardened in clean water, and tempered and ground without contact with 

 any thing greasy, but every part of the machine-bed, set-screws, and frame, 

 from which any risk was to be feared, was carefully cleansed, and paper 

 used to cover the whole of the iron, except where actually borne upon by 

 the tool. The piece of iron measured about 75 millims. long, 16 millims. 

 wide, and 12 millims. thick. It was cut from as solid a portion of the 

 mass as could be found, and was quite bright upon the surface and free 

 from crust, though traces of a very minute crack or fissure were barely 

 perceptible at one end. The piece weighed 124*589 grammes ; and as the 

 specific gravity of the iron had been found to be 7*853, the volume was 

 15*87 cub. centims. A new and perfectly clean porcelain tube, with sound 

 glaze, was used, heated by a small upright fire-clay furnace with good 

 draught, through holes in the opposite sides of which the tube was passed. 

 The fuel was charcoal, in pieces a little larger than a walnut. The 

 Sprengl pump had a fall-tube of about 1*34 metre long ; its connexions 

 were made with great care, and were protected by outer casings of india- 



88-706 

 10-163 

 •396 

 •003 

 •002 

 trace. 

 •341 

 •019 

 •003 

 •172 

 •067 



The results of ordinary analysis were : 



Iron 



Nickel 



Cobalt 



Copper 



Tin 



Phosphorus 



Sulphur 



Chlorine 



Carbon 



Silica 



99-872 



