of the fxed Alkalis. 105 



combines with sulphur, and forms a triple compound of a 

 dark gray colour. 



VI. On the Proportions of the peculiar Bases and Oxygen in 

 Potash and Soda. 

 The facility of combustion of the bases of the alkalis, and 

 the readiness with which they decomposed water, offered 

 means fully adequate for determining the proportions of their 

 ponderable constituent parts. 



I shall mention the general methods of the experiments, 

 and the results obtained by the different series, which ap- 

 proach as near to each other as can be expected in operations 

 performed on such small quantities of materials. 



For the process in oxygen gas, I employed glass tubes 

 containing small trays made of thin leaves of silver or other 

 noble metals, on which the substance to be burnt, after 

 being accurately weighed or compared with a globule of 

 mercury, equal in size*, was placed : the tube was small at 

 one end, curved, and brought to a fine point, but suffered 

 to remain open ; and the other end was fitted to a tube com- 

 municating with a gasometer, from which the oxygen gas 

 was introduced, for neither water nor mercury could be 

 used for filling the apparatus. The oxygen gas was carried 

 through the tube till it was found that the whole of the com- 

 mon air was expelled. The degree of its purity was ascer- 

 tained by suffering a small quantity to pass into the mer- 

 curial apparatus. The lower orifice was (hen hermetically 

 sealed by a spirit lamp, and the upper part drawn out and 

 finally closed, when the aperture was so small, as to render 

 the temperature employed incapable of materially influencing 

 the volume of the gas ; and when the whole arrangement 

 was made, the combination was effected by applying heat to 

 the glass in contact with the metallic tray. 



In performing these experiments many difficulties oc- 



* When the globules were very small, the comparison with mercury^ 

 v.hich may be quickly made by means of a micrometer, was generally em- 

 ployed as the means of ascertaining the weight : for jn this case the globule 

 could be immediately introduced into the tube, and the weight of mercury 

 ascertained at leisure. 



, curred. 



