264 W. A. BROWNE AND W. A. GREIG. 
The hornblende deserves more than passing notice. 
Although at first sight it would be classed as uralite, a 
closer scrutiny must cause the revision of this opinion. 
Certainly it has many of the attributes of secondary horn- 
blende. For example it is quite evidently in many cases 
being formed at the expense of pyroxene, the junction with 
the latter being irregular and ill-defined, and often marked 
by a discharge of secondary magnetite dust. Further, it 
appears in parallel circumgrowth to the pyroxene, and 
twinning in the latter is continued out into the hornblende, 
phenomena which are both mentioned by Rosenbusch! as 
characteristic of uralite. 
But on the other hand certain features point to the 
mineral being primary. Though formed partly at the 
expense of the pyroxene it has not been wholly produced 
in this way, but has formed outgrowths from the latter 
mineral, the boundaries of which are clearly those of horn- 
blende, a fact that is most evident in the case of cross- 
sections. Then again the mineral is not fibrous, as uralite 
should be, but compact, and it is sometimes twinned when 
there is no corresponding twinning in the pyroxene around 
which it is grown. Rosenbusch has pointed out that as 
amphibole is generally poorer than pyroxene in lime, during 
the process of uralitization the excess of lime may appear 
in epidote or some other lime-bearing mineral; nothing of 
this sort is found in the present instance. It is to be noted, 
too, that the hornblende outgrowths are generally moulded 
on plagioclase but idiomorphic towards biotite, orthoclase 
and quartz. Finally, inside the rim of hornblende the clino- 
hypersthene is frequently seen to be altering to a felt of 
very pale green actinolitic fibres with comparatively weak 
birefringence. All these indications would go to show that 

2 Micrescopical Physiography of the Rock-Formiug Minerals. Iddings’ 
Translation, 4th Edition, 1905, page 271. 

: 
: 

