514 A TREATISE ON METAMORPHISM. 



decomposed rocks the processes of decomposition have gone to different 

 stages In but few of the rocks analyzed has the process nearly approached 

 completion. Also the rock which is spoken of as disintegrated has been 

 partially decomposed. This is conclusively shown by the fact that there is 

 great variability in the calculated losses for the various elements. If the 

 rock had simply been mechanically disintegrated the composition of the 

 altered rock would have been the same as that of the original rock. 



In the tables the various elements — iron, aluminum, potassium, and 

 silicon — are chosen in different cases as constant in amount. This shows 

 that the supposition that any element is fixed in amount is erroneous. 

 Evidently in each case in making the calculations the method has been to 

 choose the element in which the loss has been the least, and by this to 

 gauge the loss of the other elements. For any one of these elements it is 

 highly probable that the loss in most instances is as great as or greater 

 than the loss of this element where other elements are chosen as the 

 standard. 



For instance, when aluminum is taken as fixed the loss of the potassium 

 ranges from 29 to 83 per cent. It is highly probable that in the case of the 

 phonolite, where potassium is taken as fixed, the loss of this element is at 

 least as great as the minimum, and it may equal the maximum. Therefore 

 the percentages of loss of the other elements of the phonolite, given on page 

 512, are greater than estimated. On the average aluminum is the least 

 soluble of the elements, and in those eases where it can be used as the 

 standard the underestimates for the other elements may not be large; but 

 the occasional very considerable loss of alumina in some instances — for 

 example, a limestone (Table VII, p. 513), where the loss is over 11 per cent — 

 shows that even where this element is taken as fixed there may be very 

 considerable losses of it, and consequently corresponding underestimates 

 of the losses of the other elements. For instance, in a diabase (column 1 of 

 Table III, p. 509) the aluminum is taken as fixed and the loss of the iron 

 oxide is calculated at 18.1 per cent. In column 2 of the same table the iron 

 oxide is supposed to be fixed and the loss of the aluminum is calculated at 

 21.38 per cent. The facts probably are that there were heavy losses of 

 iron and aluminum in both rocks, and that in each case the calculated loss 

 is much too small. 



