478 Johnston, Merwin, and Williamson — 



Analogous results were observed when solutions of cobalt salts 

 were mixed with solutions of alkali carbonates. 



The other tests proposed depend even more obviously on the 

 state of subdivision and rate of dissolution of the carbonate. 

 The latter is the only factor, so far as one can see, in Nieder- 

 stadt's test, for the difference in solubility of calcite and 

 aragonite is, as we shall see, so small that the difference in 

 concentration of the two solutions can hardly be a factor in 

 the result. It may be that the phenomenon of isomorphism 

 plays a role here, in the sense that a carbonate isomorphous 

 with that already present will tend to precipitate on the sur- 

 face of the latter and thus to diminish its surface and con- 

 sequently the rate of precipitation ; but there would seem to 

 be little likelihood of giving a definite and reasonable account 

 of this matter until more is known of the factors which deter- 

 mine the formation and growth of crystals. 



In the opinion of the authors, these chemical tests should 

 not be used alone because they are indecisive, particularly so 

 where they would be most useful ; namely, for the characteri- 

 zation of fine-grained material which may carry something that 

 interferes with the test ; this opinion is based in part upon the 

 above observations, in part upon the fact that some of the fac- 

 tors upon which these tests depend are very difficult to control. 

 These factors are: concentration of reagent, temperature, C0 2 

 pressure (hence acidity, for these are dependent variables) ; the 

 size of grain and stability of the precipitate formed ; and in 

 particular the rate of dissolution of the calcium carbonate — 

 which in turn depends upon temperature, crystalline form, 

 state of subdivision, and constitution of the solvent medium. 

 Moreover, while these tests can be so carried out as to yield 

 trustworthy indications with good crystals of calcite and ara- 

 gonite, they cannot be trusted when applied to less homo- 

 geneous material such as may be obtained from precipitation 

 experiments. Such material, even when minute irregular 

 aggregates are present, usually yields to optical methods, though 

 these methods may be uncertain when there are interlocking 

 small grains or (as in " vaterite ") inclusions of submicroscopic 

 size ; in such cases we must use as a basis all the criteria possi- 

 ble and, bearing in mind their several limitations, draw such 

 conclusions as are admissible from the evidence as a whole. 



Calcite (£-CaC0 3 ). 



Under all the conditions which have hitherto been studied, 

 with the exception noted below, calcite is the stable form of 

 CaCO s ; for it is the least soluble at ordinary temperatures, and 

 at about 400° the other forms go over rapidly into it. It is 



