GEOPHYSICAL LABORATORY. 149 



inverts reversibly to a-CaCOs; under these conditions aragonite and the 

 yu-form are always unstable with respect to calcite, though there is an indica- 

 tion that aragonite has a stable field of existence at about —100° or lower. 

 Under all ordinary conditions, therefore, pure aragonite tends to go over into 

 calcite; in how far it actually does so depends upon the rate of this process 

 under the particular circumstances. There is therefore no definite transition- 

 point; the interval required for the transformation at 100° in presence of water 

 and calcite is measured in days, and for its inversion at or about 400° is meas- 

 ured in hours. By reason of this instability, one can not specify the factor or 

 factors which determine the precipitation of CaCOs as aragonite; indeed, its 

 appearance would seem to be a matter of chance — in other words, it depends 

 upon factors which can not be controlled at the present time. In so far as 

 we have been able to ascertain, natural aragonite is formed (a) through 

 organic agencies; (b) by deposition from hot springs; (c) when an isomor- 

 phous carbonate is present to serve as nucleus; (d) in salt waters containing 

 sulphate even at ordinary temperature. Pure aragonite can persist as such 

 only when dry; but aragonite containing other substances in solid solution 

 may thereby be enabled to persist in presence of certain solutions. There 

 are indications that the /x-form often occurs as an intermediate step in the 

 precipitation of the other forms ; but it soon transforms in presence of water, a 

 fact which, combined with the fact that it can not be differentiated from 

 aragonite bj^ means of the usual color-tests, is responsible for the circumstance 

 that it has not been recognized more often as occurring in nature. 



The properties of these three forms, and of the hexahydrate CaC03.6H20 

 (which, moreover, is also unstable vnth respect to calcite even at 0°) are 

 described. The several forms can be differentiated from one another by 

 optical, crystallographic, and chemical tests, by observation of density and of 

 the stability of the material when heated; but it is not safe to trust to a 

 single one of these tests, especially with fine-grained material, as the result 

 may be ambiguous. Much of the work in the past is vitiated by the fact that 

 the criteria employed were insufficient to differentiate one form from another; 

 and thus many of the statements to be found in the literature require revision. 

 In this connection, we might point out that a determination, no matter how 

 carefully carried out, of a single property of a substance is of much less value 

 than two or more characteristics of the same specimen; and make a plea for 

 the determination on an analyzed specimen of as many diverse properties as 

 possible, for it is only in this way that it will be possible to coordinate the 

 various observations in a satisfactory way. As an illustration of this point, 

 one can not state the manner in which the properties of so well-defined and 

 common a mineral as calcite vary with the presence of other substances in it, 

 in spite of the enormous number of isolated observations which have been 

 made; and it is obvious that such a coordination, once established, would save 

 an immense amount of labor thereafter. 



(1.5) The composition of natural bornite. E. T. Allen. Am. J. Sci. (4), 41, 409-41.3 (1916). 



Analyses of a number of bornites from very widely separated localities, the 

 essential homogeneity of which had been proved by a careful microscopic 

 examination of several polished surfaces, were found to agree closely with the 

 formula Cu3FeS4. The conclusion is supported by the constancy in the 

 color and density of the specimens. A critique of previous work on the 

 subject is given, and in particular it is shown why the evidence against 

 this view presented by Kraus and Goldsberry (Am. J. Sci. (4), 37, 539, 1914) 

 is inconclusive. The conclusion reached confirms that of Harrington (Am. J. 



