14 1HE JOURNAL OF GEOLOGY. 
Nearly all of the hornblende that appears in these sections is 
quite evidently an alteration product of the augite. It isin large 
plates of a dirty greenish or a greenish brown color, according to 
the position above the lower nicol. The outlines of these plates 
are always ragged, and they bristle with projections that extend 
beyond the main mass of the plate far between the neighboring 
plagioclase grains. As the augite changes into hornblende, the 
new substance produced orientates itself uniformly until the 
products of a dozen or so grains become merged into a single 
plate, which naturally includes a large number of feldspar and 
magnetite grains, and of biotite flakes, provided any of these 
happen to have been formed before the hornblende. The structure 
of the plates is exactly analogous in its origin to that of Salomon’s 
contact minerals, as it is here produced by selection of augite 
grains and the enclosure of those minerals that were not suitable 
for alteration into amphibole. When the small grains of augite 
are too widely scattered to affect the hornblende produced by 
their alteration, this substance occurs in isolated particles with 
independent orientations. That these little isolated grains are 
secondary in their origin is learned from the fact that many of 
the granulitic augites are partially changed into the same min- 
eral, some only on their edges, and others throughout nearly 
their entire masses. In many other sections the hornblende is 
evidently original, in which case it has the same properties as 
the least decomposed of the secondary amphibole. In rock 
No. 7061 for instance the hornblende is more irregular in 
its outline than the diallage, though it usually possesses right- 
lined contours rather than sinuous ones. In color it is brownish 
green with absorption as follows: t—b=brownish-green>@= 
pale yellow. Its grains occur between those of plagioclase, and 
they include magnetite, diallage and round or elliptical colorless 
inclosures like those observed within the pyroxene. Cross 
sections of crystals with the characteristic contours and cleavage 
of hornblende are not very uncommon. Although this horn- 
blende is not unlike the secondary amphibole in others of the 
granulitic rocks it is not possible to regard it as due to the alter- 
