440 M. F. L. Ekman on the Electrical Insulating -power and 



action of air free from carbonic acid, but saturated with moisture 

 at the ordinary temperature, and determined the increase of 

 weight after a certain time. The vessel in which the specimens 

 were exposed to the action of moisture was of silvered metal with 

 the surface polished. The same quantity (1 gramme) of each 

 specimen was weighed out ; for in this as in all other respects 

 the greatest possible uniformity was observed in these experi- 

 ments. The following Table shows how many milligrammes of 

 water each specimen absorbed in the time stated : — 







2hrs. 



4 hrs. 



6 hrs. 



12 hrs. 



18 hrs. 



24 hrs. 



Specimen 



No. 1. 



. 4-9 



7'7 



9-8 



121 



. . 



17*4 



)} 



No. 2. 



. 3-6 



4-8 



6-5 







152 



)) 



No. 4. 



. 6-9 



12-2 



. . 





31*4 



33-5 



a 



No. 5. 



. 8-0 



. .- 



. . 



23-3 



33-5 





With No. 3 the observation failed ; I can say, however, with 

 certainty that it absorbed less water than either No. 4 or 5. 



We see, then, that the kinds of glass which were unfit for 

 electrical apparatus were much more hygroscopic than the others, 

 and that the glass which contained much lime was least hygro- 

 scopic, whereas that which was relatively deficient in lime and 

 silica was most so. When, after being dried in air (as free as 

 possible from carbonic acid) at 160° C, the specimens thus tested 

 were reweighed, it was found that Nos. 1 and 2 had almost ex- 

 actly resumed their original weight, No. 4, on the contrary, 

 weighed 0*0014, and No. 5 0*0022 grm. more than before; and 

 both these specimens appeared in a striking degree agglomerated, 

 evidently in consequence of the powder being glued together by 

 the hydrosilicated alkali formed. 



I now placed the specimens over water in a closed vessel con- 

 taining, instead of air, carbonic acid gas. After a certain time 

 they were placed over lime and chloride of calcium till constancy 

 of weight had been attained, after which they were dried at 160° 

 Cels. and again weighed. The affinity of the specimens for 

 carbonic acid in the presence of humidity is shown in the fol- 

 lowing Table : — 





Time of exposure 



Increase of 



weight, in milli- 





to action of car- 



grammesj 



aftei 



drying. 





hnnip Qpin 





A 







hrs. 



At ordinary temp. 



At 160° Cels 



Glass No. 1. 



. . 6 



7'7 





2-7 



„ No. 2. 



. 12 







1-4 



„ No. 3. 



. . 6 



13-6 





4*5 



„ No. 4. 



. 6 



19*4 





6-7 



„ No. 5. 



. 4 



15-8 





50 



Although the time of exposure was not the same for all the 



