Mechanical Science. 161 



cific gravity of a compound is the sum of the specific gravities of its 

 elements, or it is different in consequence of contraction or dilata- 

 tion. In by far the greater number of cases it proves to be different : 

 thus, in the sulphurets of mercury, lead, arsenic, antimony, tin, and 

 iron, the specific gravity is increased; in the iodide of potassium it 

 is also increased ; in those of silver, mercury, and lead, it is dimi- 

 nished. Then endeavouring to determine whether the contraction 

 was the same for bodies having similar atomic composition, no 

 analogy was found ; so that, though many sulphurets and iodides 

 have been examined carefully, nothing can be deduced from them 

 relative to other sulphurets and iodides : even constancy of contrac- 

 tion or expansion cannot be deduced, for the iodides present cases 

 of both. 



These results on the sulphurets and iodides appear to M. Boullay 

 important, not only as adding facts to our knowledge, but as marking 

 and destroying an error into which many philosophers, occupied with 

 the same question, have fallen. They have endeavoured to deter- 

 mine the specific gravity of bodies brought to the same condition, 

 (the solid state, for instance,) but have been stopped by those sub- 

 stances which cannot be brought into that form. Assuming the 

 hypothesis, however, that in the union of two bodies in the solid 

 state there was neither expansion nor contraction, or else that the 

 negative element only was altered, they have thought themselves 

 justified in deducing from the specific gravity of a binary compound 

 and one of its elements, the specific gravity of the other : thus the 

 densities of oxygen and chlorine have been calculated from the 

 metallic oxides and chlorides. This assumption is entirely done 

 away by the facts quoted. 



Even admitting for a moment the hypothesis as good, calculation 

 from it proves its own fallacy : thus the density of oxygen derived 

 in this way from the oxides varies from 1.25 to 5.88, which, without 

 experiment, would prove great modifications by expansion and con- 

 traction. The chlorides gave still more striking results for chlorine ; 

 and from the specific gravity of the chloride of potassium it would 

 appear that a volume of this binary compound contains more than 

 its volume of metal only, indicating an enormous contraction between 

 that substance arid the chlorine*. 



5. APPARENT HYDROSTATIC ANOMALY WITH LAUREL-OIL. 



Dr. Hancock has remarked a curious apparent anomaly in the 

 hydrostatic pressure of two fluids, the lighter of which, upon mix- 

 ture, passed to the bottom, and the heavier to the top. One of the 

 fluids is laurel-oil ; the other a mixture of pure ether (i. e. free from 

 alcohol) and proof spirit in equal proportions, or with a slight excess 

 of ether. Such a mixture is lighter than the essential oil, but when 

 the latter is poured upon the former it floats, and indeed whichever 



* Annales de Chimie, xliii. 266. 

 VOL. I. OCT. 1830. M 



