160 Condensation of Gases at the Surface of Glass. [Feb. 12. 



in a tube made for the purpose, which fitted on over the up-turned 

 ends of the five fall-tubes. At the end of this time the vacuum was 

 again fairly good, though not so good as it was before the heating 

 commenced. The McLeod gauge indicated 1*2 M. It was seen that 

 very little more air was being carried down, and I did not wish to 

 push the vacuum farther than, or quite so far as, the vacuum which 

 had been obtained before the liberation by heat of the condensed gas. 



The collecting tube was now removed, and the gas obtained was 

 measured and analysed, so far as it was possible to analyse a quantity 

 so small. 



The total amount of gas collected was calculated to be, at 15° C. 

 and a pressure of 760 mm., 0"45 of a cubic centimetre. To this a small 

 quantity of strong solution of caustic potash was added, and time was 

 given for absorption. A small quantity of pyrogallic acid was next 

 added, and the further absorption observed. The residue was so small 

 that I could do nothing farther. 



The result of the analysis showed 8 '24 per cent, of the whole to be 

 carbonic acid gas (absorbable by caustic potash). Of what remained 

 24*8 per cent, was oxygen (absorbable by pyrogallic acid and caustic 

 potash mixed). The residue 75'2 per cent, was, I presume, mainly if 

 not wholly nitrogen. I ought to remark that my pump was furnished, 

 as is usual, with the phosphoric acid drying tube. The gas, therefore, 

 which I collected was perfectly dry, and I have no way at present of 

 ascertaining how much moisture adheres to the spun glass. In stating 

 the results of the analysis I have made no correction for moisture 

 introduced with the potash solution. 



In order to make an estimate of the amount of surface exposed by 

 the spun glass, I measured, with a screw micrometer gauge, the diame- 

 ters of 200 of the fine glass fibres taken at random. I found them, as 

 I expected from the care with which they had been prepared, fairly 

 uniform, and the average diameter was 7*06 hundredths of a milli- 

 metre. Weighing also the 200, and then the whole quantity, I found 

 the whole number of the fibres to be 6370. The average length was 

 10*25 cm. The surface was thus 1448 sq. cm., or equal to that of a 

 square 38 cm. in the edge. 



I am preparing for further experiments on this subject, and hope 

 soon to be able to add to it observations on the amount and on the 

 electric conductivity of the film of moisture condensed upon the sur- 

 face of glass. 



Additional Note. Received February 12. 



Since the writing of my former communication on this subject, I 

 have made some further experiments on it, and I beg leave to give an 

 account of the results of one of these experiments. 



