142 THE JOURNAL OF GEOLOGY. 
that the plagioclase grains are in reality products of the altera- 
tion of the hornblende and not particles derived from the ground- 
mass which have penetrated between the augite and opacite 
grains. Inthe largest and most altered crystals this opacite core 
is surrounded by a border or frame of almost pure augite in the 
form of colorless prismatic grains, giving extinction angles of 36°, 
and their long axes parallel to the ¢ axis of the hornblende, so 
that the ends of many of these hornblende pseudomorphs have a_ 
fringed or bearded appearance. In the less fresh specimens this 
augite assumes a greenish yellow color, when it resembles more 
than ever uralitic hornblende. This prismatic augite does not 
only occupy the border of the hornblende pseudomorphs, but in 
many cases penetrates the opacitic aggregate in irregular patches. 
Many of these altered hornblendes show signs of great mechani- 
cal disintegration since their alteration, patches and prisms of the 
secondary augite being seen to have been torn away from their 
original positions and scattered through the groundmass of the 
segregation. Kiich (pp. 46, 176) observed apparently similar 
occurrences, though not in segregations, which he refers to a par- 
allel growth of augite and hornblende. This does not seem to be 
the correct explanation, at least for the present case, and it is 
most probable that these borders of prismatic augite are due to 
causes similar to those which produced the opacitic and ‘“‘augite- 
opacite aggregate” alterations of hornblende, forming a later and 
probably the final stage of these. 
The occurrence of these alterations in the case of segregations 
in rocks where they also occur, while they are not found in the 
segregations in rocks containing fresh green hornblende, is very 
interesting, and contrary to what Kitch and Belowsky observed 
in their otherwise very similar occurrences. The observation 
often previously made that only the brown hornblende is liable 
to this alteration is again confirmed, and this fact seems to be a 
fundamental one to adopt in any explanation of the set of 
phenomena. Judging from the present cases it seems certain that 
the alteration took place before the solidification of the ground- 
mass, which would place the period of alteration of the hornblende 
