Atomic Volume and Specific Gravity. 631 



as near each other as the estimation of specific gravities by 

 any two different experimenters. We do not rest the claims 

 of our theories on our own experiments, but are willing to 

 admit the accuracy of other experimenters, especially of 

 Karsten, Hassenfratz, Kopp, and others who have preceded 

 us on this subject* ; while at the same time we believe that 

 our methods of taking specific gravities have enabled us to 

 introduce more uniformity into the results. The simplicity 

 of the methods themselves is due to Bishop Watson, who was 

 the first to take specific gravities by the increase in the stem 

 of an instrument; and to Holker the suggestion is due of 

 using a saturated solution instead of the water employed by 

 Watson. :')'j 



We conceive that the primitive volume 9*8 is transformable 

 into the primitive volume 11, and vice versa, and for this reason 

 we sometimes see sulphate of ammonia 9'8 x 4, at other times, 

 in combination as in bisulphate of ammonia or the anhydrous 

 double sulphates, it is 1 1 x 4 ; and numerous other instances 

 of transformation are presented in the previous sections. 



The liquid volume being to the solid volume either as 



9 : 11 or as 9 : 9*8, these numbers, used as the divisor for 



the liquid and solid volumes respectively, usually yield the 



same quotient. Thus the liquid volume of sulphate of cop- 



45 55 



per is 45, its solid volume is 55. — - = 5, and — - = 5 j 



so that we say the salt affects the same number of volumes 



in the liquid and in the solid state. In the same manner 



subphosphate of soda has a volume of 216 in solution and of 



2] 6 235 



235 in the state of salt. Now -— - = 24 and ^^ = 24, so 



that the number of volumes affected in solution and in the 



solid state are exactly the same. This is a general rule and a 



powerful argument of the accuracy of our position. The rule 



has exceptions in salts of potash, in which the volumes are 



increased by one volume on becoming solid ; thus KO, SOg 



18 33 



— = 2 in solution, and -— r = 3 in the solid state. This is 

 9 ' 11 



* The only decided difference which we found from other experimenters 

 is in the case of hydrated salts. Thus our determination of the volumes 

 of the double magnesian sulphates and sulphate of potash (Table VI.) dif- 

 fers from Kopp's experiments as 99 : 103. These salts contain from 3 to 

 4 per cent, of mechanical water, as Graham long ago pointed out (Trans. 

 R. S. E. vol. xiii. p. 12), and the neglect of this in Kopp's experiments 

 has probably caused the difference. We take this opportunity of stating 

 that when more than one specific gravity is given by us, the salts have been 

 prepared at different times ; in many instances this is not the case, but in 

 much the largest proportion it is so. 



