44 THE ROMAN COMAGMATIC REGION. 
From the figures of the norm and the ratios it is seen that the rock falls well 
within the persalane class and the domalkalic rang, and fairly well within the sodi- 
potassic subrang, though somewhat close to the dopotassic border. But it is almost 
exactly on the line between the perfelic and lendofelic orders, falling in order 5 or 6 
according to whether the small amount of apatite is calculated in the norm or not. 
It is therefore eminently a transitional one belonging strictly in putaskose, I. 5. 2. 3, 
but very close to the border of the subrang I. 6. 2. 3, as yet unnamed, which may 
be called procenose, from the locality of the specimen analyzed. The magmatic 
name then will be procenose-pulaskose. 
The analysis of the Astroni rock in II resembles the other very closely, except 
in containing 2.5 per cent more silica. This larger amount, by increasing the quan- 
tity of normative albite and diminishing that of normative nephelite, removes the 
type well away from the border of order 6, while, like the preceding, it falls well 
within class I and rang 2. On the other hand, the ratio of soda to potash brings it 
very near the line of dopotassic subrang 2, so that this type must be regarded as a 
vulsinose-pulaskose. It will be remembered that the other rocks of the Astroni 
Volcano fall in vulsinose, one, indeed, being strictly a pulaskose-vulsinose. 
The analysis by Ricciardi in III must be of a rock almost, if not quite, the 
same as that analyzed by me, as Klein's description agrees with mine in all essen- 
tial respects, and the town of Proceno is built upon but one thick lava flow, from 
which the two specimens must have come. It will be seen that here, as in so many 
other of the same chemist's analyses, the amounts of the alkalis reported are quite 
irreconcilable with the mode of the rock. Thus the potash reported would 
allow the formation of but 18.35 f orthoclase or 14.39 of leucite, and the silica is so 
high that it can not be satisfied by the bases present, and the rock would necessarily 
contain considerable amounts of quartz. Indeed, the discrepancy is so marked 
that Klein comments on the apparently low figure for alkalis. 
Mode. Estimation of the mode of the paglial procenose-pulaskose could be 
carried out by Rosiwal's method only as far as the phenocrysts and the dark minerals 
were concerned. The feldspar laths of the groundmass were far too fine and con- 
fused in arrangement to allow of separate estimation of the orthoclase and the labra- 
dorite. The amount of the large leucite phenocrysts (10.4 per cent by volume) was 
determined in the hand specimen, while the thin section was used for those of the 
groundmass and the other constituents. The results may be thus tabulated: 
Orthoclase phenocrysts. . 
Labradorite phenocrysts. 
Leucite 
Vol. %. S P . gr. 
i.i X 2.6 
2.9 X 2.7 
. 16.6 X 2.5 
= 2.9 
= 7.8 
41 . 5 
Wt.%. 
1. 1 
2.9 
15.6 
Augite 
4.0 X i . i 
16 2 
6 i 
Magnetite 
14. X 5.2 
7 3 
2 8 
Feldspar, groundmass. . . 
. 73.1 X 2.6 
= 190.0 
71-5 
ioo. o 265.7 ioo. o 
In calculating the mode from the norm it will be found that if all the nephelite 
is assumed to take up silica to form modal albite there must result 36.2 per cent of 
