25 
As pyrite and pyrrhotite are more or less accidental and variable 
constituents of the tuff, it was considered that their presence might partly 
mask regularities that would otherwise appear. The sulphur was, there- 
fore, deducted with corresponding amounts of iron, and the analyses re- 
calculated to 100 per cent, as follows: 
— 
1 
2 
3 
4 
5 
5 a 
6 
SiC>2 
59-38 
60*05 
59*94 
60*38 
64 00 
61-64 
57*50 
AlsOa. 
16-56 
16*98 
17*78 
18-50 
17-80 
18-95 
18*43 
.Fe*Qs. ...... ... 
0-47 
2*87 
1*47 
2-19 
2-34 
2*40 
FeO... 
4-95 
4-76 
5-18 
6-07 
3-36 
3-59 
6*26 
(Fe).„. • ■ 
(3-85) 
( 4 - 04 ) 
( 6 * 12 ) 
(5-75) 
(4-14) 
(4-43) 
( 6 * 55 ) 
CaO . . . . ... 
3-30 
1-69 
1*54 
0*83 
0*42 
MgO. 
2-54 
2*84 
1*56 
1-90 
i - 9 i 
2-04 
2*00 
N ai O 
2*18 
2*77 
1*80 
1*43 
1*32 
1-41 
1*14 
KjO...... 
4*08 
3*32 
4*02 
3*80 
4*72 
5-03 
4*58 
CO, 
3-08 
2*83 
1*44 
1*76 
1-12 
1*20 
2*40 
c„ 
0-78 
1*05 
1*35 
1*45 
1*27 
1*36 
1*57 
0-02 
0*03 
0*01 
0*02 
0-02 
0-02 
TiOj. .... 
0*50 
0-23 
0*76 
0*08 
0*34 
0*36 
0*69 
HsO + 
2-48 
2*47 
1*65 
1*66 
1*84 
1*97 
2*50 
HjO — 
0*15 
0*51 
0*10 
0-05 
0*11 
0*11 
0*09 
In the tables above the attempt has been made to correct No. 5 for 
excess silica, and the result is shown in No. 5a. The original analysis, 
when pyrite and pyrrhotite were subtracted, totalled 98 * 10 per cent with 
silica 62*80 per cent. From both figures 6 * 10 per cent was taken, reducing 
the silica to 56*70 per cent and the sum of all constituents to 92 per cent. 
Each constituent was then increased proportionately to bring the sum to 
100 per cent. The result, however, is not very satisfactory, as both silica 
and alumina still appear somewhat too high, and the results are, therefore, 
omitted from the curves below. 
The recalculated analyses have been platted on Mead’s straight-line 
diagram (See Figures 8 and 9) 1 . According to this method, the percentage 
of a constituent in the fresh rock is divided by its percentage in the altered 
rock, and the quotient multiplied by 100. The result represents, for each 
constituent, the number of grammes of altered rock required to contain 
the amount of that constituent originally present in 100 grammes of the 
fresh rock. The positions of the various points, platted on the straight- 
line diagram, at once show the relative behaviour of the several constituents 
during alteration. If any constituent is assumed to have remained constant 
during alteration, all constituents whose points fall to the right have de- 
creased in absolute amount, and those whose points fail to the left have 
increased. When lack of information makes it impossible to fix any point 
as a constant, only the relative gains and losses can be seen from the posi- 
tions of the points. 
In Figure 8 the points are obtained using No. 1 as the fresh rock, 
whereas in Figure 9 No, 2 is regarded as the fresh rock. It will be seen 
that the diagrams are similar, although the absolute position of the points 
varies. 
It would seem reasonably safe to assume that free carbon, on account 
of its high degree of insolubility and inalterability, remained constant 
iMead, W. J.: Ee. Geol., vel. VII, pp. 141-4 (1912). 
Also Leith, C.K., and Mead, W. J.: Metamorphic Geology, pp. 12-13, H. Holt & Co., New York, 1915. 
