148 CARNEGIE INSTITUTION OF WASHINGTON. 



carbonic acid and for the total CO2 present in solution; in practice they yield 

 accurate results, provided that due attention is paid to conditions discussed 

 or referred to in this paper. But tliese two determinations suffice in general to 

 characterize the solution with respect to either its content of free CO2, the 

 proportion of carbonate or bicarbonate, or the degree of alkalinity or acidity; 

 for, since we are dealing with an equilibrium capable of fairl}^ rapid readjust- 

 ment, we are justified in applying the equilibrium constants to calculate the 

 above quantities in the great majority of those cases in which a knowledge of 

 them is of real importance. 



(13) The complete solubility curve of calcium carboHate. John Johnston and E. D. 



Williamson. J. Am. Chem. Soc, 38, 975-983 (1916). 



A further development of the views discussed in an earlier paper (Year 

 Book No. 14, page 168). The graph showing the concentration of calcium 

 in the solution at equilibrium in the system CaO-H20-C02 is made up of three 

 curves, along which the stable solid phase is hydroxide, carbonate, bicarbon- 

 ate, respectively. The first extends only up to values of P, the partial pres- 

 sure of CO2, of about 10~i'* atm. at 16°; the second, starting from the transi- 

 tion-point, decreases to a minimum and then rises again, as the value of P 

 increases continuously, until P is about 15 atm.; beyond the second transition- 

 point bicarbonate is the stable^ solid phase. Along the whole course of the 

 graph, all three ions 0H"~, COj, HC07 are present at relative concentrations 

 depending upon P; so in this, as in other analogous cases, the solubility curve 

 ascertained by experiment w^ould have different forms according as one 

 determined one or other of the several molecular species in solution. Thus the 

 maximum concentration of COT occurs when the solubility — as measured by 

 the concentration of calcium in solution — is a minimum, and it is only within a 

 restricted range of P that the base associated with COy is more than a frac- 

 tional proportion of the total base in solution. 



The transition pressure at which both hydroxide and carbonate are stable 

 may be calculated either from the solubilities of hydroxide and carbonate or 

 from their thermal dissociation pressures; these two absolutely independent 

 methods yield results surprisingly concordant, a circumstance which demon- 

 strates the essential correctness of the view discussed in this paper. 



(14) The several forms of calcium carbonate. John Jolmston, H. E. Merwin, and E. D. 



Williamson. Am. J. Sci. (4), 41, 473-512 (191G). 



The prevalence of calcium carbonate as a constituent of the crust of the 

 earth has led to a vast amount of discussion of the chemistry of its formation 

 and of the stability relations of the several crystalline forms in which it occurs. 

 The value, alike to the geologist and to the chemist, of an exact knowledge of 

 the facts has also been repeatedly emphasized. The evidence has, however, 

 been incomplete and in part contradictory or wrongly interpreted, and has 

 never been presented systematically. It appeared, therefore, to be a useful 

 task to give a coherent critical statement of the facts and to discuss the 

 deductions which, in the light of present knowledge, may legitimately be drawn 

 from them. 



Under ordinary conditions, calcium carbonate appears in three crystalline 

 anhydrous forms, viz, as calcite, aragonite, and a form which we have 

 designated m-CiiCO.v The other reputed forms, including "vaterite" and 

 "amorj^hous" CaCOj, are not definite forms; their divergent properties are 

 due mainly to differences in size of grain and mode of aggregation. 



Of these three established forms calcite is, at ordinary pressure, the stable 

 one at all temperatures from 0° (or loA^er) up to 070", at which temjierature it 



