32 Day* etc. — Determ ination of Mineral 



Relative volumes of solid and liquid diabase. — The volume 

 of the molten diabase at 1250°, from both of our curves of 

 rising temperature and from our second curve of falling tem- 

 perature, is 390. The first curve of falling temperature points 

 to a somewhat lower value, which is probably due to incom- 

 plete fusion of the feldspar. After the recrvstallization with 

 falling temperature the two curves coincide. 



Our value 390 for the volume at 1250° is undoubtedly high, 

 since at this temperature carbon monoxide is forming slowly 

 and escaping at the sides of the fused block (see p. 29). The 

 amount of the error can be roughly estimated, from the small 

 depressions left around the edges of the recrystallized block, at 

 about one per cent. 



The value obtained from Barus' curves is 383 at 1250°. ^The 

 net effect of the errors of his method is probably to give too 

 low a value. The figure of 383 which we have obtained from 

 his data is in fact 1*8 percent lower than our maximum of 390. 

 The most probable value for the volume of a kilogram of 

 liquid diabase at 1250° is therefore about 386. 



In the absence of data on the volume at high temperatures 

 of each of the constituent minerals of crystalline diabase, its 

 true volume near its fusion temperature can not be directly 

 measured, on account of the shattering caused by unequal ex- 

 pansion of different minerals and by the escape of gases (see 

 p. 23). It will be necessary, therefore, to resort to a certain 

 amount of extrapolation in order to obtain a value for the vol- 

 ume of the crystalline rock. 



In fig. 9 is plotted the volume of Sudbury diabase up to 

 1000°, calculated from the expansion data of Wheeler* and 

 based on a specific gravity of 2'99S (volume 333'6).t The rock 

 is described as a typical fresh olivine diabase, massive, and pos- 

 sessing a typical ophitic or diabase structure.:}; Its specific 

 volume is only a little less than that of the Palisade diabase. 



The curve shows the expansion on the initial heating. The 

 permanent volume increase after this first heating amounts to 

 2*1 per cent. Extrapolation of the part of the curve below 400° 

 gives a value for the volume of the rock at 1250° of 344. Extra- 

 polation of the portion from 700° to 1000° gives 351. Extra- 

 polation of the curve of falling temperature (1000° to 20°, not 

 shown in fig. 9) also gives 351. The latter value must be 

 regarded as the maximum, representing the expansion of a 

 rock already shattered and permanently expanded. The lower 

 value of 344 is probably nearer the true volume ; it corresponds 

 to a mean coefficient of dilatation from 0° of 25'2 X 10~ 6 . In 



* Trans. Roy. Soc. Canada, iv, 19-44, 1910. 



f F. D. Adams, private communication, 1913. 



\ A.dams and Coker, Carnegie Institution, Pub. No. 46, p. 57. 



