On the Absorption of Water in Hot Glass. 461 



9. Summing up the results obtained we see that :— 



(1) It is possible by the observation of Michelson's interfer- 

 ence-fringes to separate a double point-source, or detect 

 breadth and ellipticity in a slightly extended source. 



(2) But the distance between the two points, or the dimen- 

 sions of the extended source, must lie within certain limits 

 depending on the length and breadth of the slits *. 



(3) The dimensions of the slits also considerably affect the 

 general theory, the formulae obtained not being identical with 

 Michelson's. The law of appearance and disappearance of 

 the fringes depends very largely on the distance between 

 the points or the dimensions of the extended source. 



XLII. The Absorption of Water in Hot Glass. Second Paper. 

 By Carl Barus f. 



1. A FTER finishing my account X of the action of hot 

 a\. water on glass, observed in fine - bore capillary 

 tubes, it seemed to me that the experiments made several 

 years ago (1891) left questions of considerable interest out- 

 standing. I refer in particular to a further examination of 

 the contents of the capillary tubes. Certain evidence was to 

 be obtained as to the occurrence of syrupy glass at 185°, 

 solidifying to a firm glass when cold, the composition 

 remaining unchanged except as to the water absorbed. 

 Again, as it was improbable that volume-contraction would 

 continue at the same rate indefinitely, the conditions of sub- 

 sidence were to be determined, together with the effect of the 

 elastic and viscous constants of the tube itself on the apparent 

 volume contractions and compressibilities observed. Ques- 

 tions relative to the acceleration of the reaction at higher 

 temperatures were to be held in view. 



Finally, if the inferences drawn from data for capillary 

 tubes are correct, it should be possible to obtain the fusible 

 glass on a larger scale. Experiments in progress in this 

 direction have proved quite successful, and will be described 

 in a later paper. The present remarks are restricted to the 

 experiences with capillary tubes. 



2. The apparatus used was the same as that heretofore 



* Since the above was read, a paper has appeared in the Comptes 

 Rendus de VAcademie des Sciences for Nov. 28, 1898, dealing with the 

 modifications in Michelson's formulae when we take into account the 

 breadth of the slits. The author, M. Hatny, follows Michelson in not 

 considering variations of intensity parallel to the slits. This, I think, 

 accounts for his results not quite agreeing- with mine. 



+ Communicated by the Author. 



X Phil. Mag. (5) xlvii. p. 104, Jan. 1899. 



