Integral Weights in Chemistry. 321 



calculating the specific gravity of steam from that of hydrogen 

 gas at 100°, as determined by Regnault, we find (assuming 

 = 16) for the weight of a litre of steam at 100° and 760 mil- 

 lim., 0*590148 grm. ; and dividing by this number the weight 

 of a litre of water at 4° ( = 1000 grins.) get 1694*49. If, how- 

 ever, we take = 15*9633, as deduced by Stas, we get instead, 

 0*58894 for the weight of a litre of steam ; and with this as 

 the divisor find a ratio 1 : 1697"97. Correcting this for the 

 density of water at 0°, we have 1 : 1697*74 ; an approximation 

 to the experimental figure which shows the revised integral 

 weight of oxygen (15*9633) to be very near the truth. But 

 to fix the integral weight of water, we must consider its den- 

 sity at 100°, the temperature of its formation under a pressure 

 of 760 millim., which, being 0*95878, gives for the number of 

 volumes of steam at the above temperature and pressure, 

 which are condensed into one volume of water at the same 

 temperature, 1628"04. The formula of water is thus 1628 

 (H 2 0), of which the coefficient multiplied by 17*9633 gives, 

 for the integral weight of water, 29244. In close confirma- 

 tion of this figure, if we take the ordinarily received weight 

 for the litre of hydrogen at 0° and 760 millim., namely 0*896 

 grm., we find for the litre of water-vapour (H 2 = 17*9633) 

 at 100° and 760 millim., 0*589088 grm. Comparing this 

 with the weight of a litre of water at 100°, we have 



0*589088 : 958*78 : : 2 : ^ = 29236*4. 



This value of as, giving the relation of weight between equal 

 volumes of H 2 and of liquid water, both at 100°, is very 

 near that calculated above for 1628 (H 2 0) =29244, which we 

 may regard as the integral weight of water. 



7. From this value thus deduced for water by comparison 

 with hydrogen, we proceed to calculate for liquid and solid 

 species their integral weights. But the integral weight of 

 water, the received unit of specific gravity for such species, is 

 that of a body a litre of which, though weighing 1000 grm. 

 at 4°, weighs but 958*78 grm. at 100° ; so that, in calculating 

 specific gravities by comparing the weight of bodies with that 

 of an equal volume of water, it is with water at the lesser 

 density that the comparison should be made. Since most 

 liquid and solid species expand, like water, by heat, it is theo- 

 retically desirable to take the volume of these at the highest 

 temperature which they can sustain without change of state. 

 This point in many cases is hard to fix. The ready transfor- 

 mation, by expansion, of arragonite into calcite by heat ; the 

 condensation, on the contrary, which goes on when silver 

 iodide is heated from 0° to 116° ; that of Rose's fusible alloy 



