200 TRANSACTIONS OF THE- [aPR. 16, 
conditioned by that of the interspace which they fill. The 
spinel granules are separated from the feldspar by a sharp line; 
on the other side they usually become gradually less numerous 
till they disappear. The evidence seems strongly in favor of 
the secondary nature of the outer zone, and I believe that in the 
ease of the St. John gabbros, at least, it is formed by the same 
influences that caused the uralitization of the rock. Supposing 
those conditions to be present which favor the formation of 
uralitic hornblende, it would form first and most easily on the 
contact of a magnesia-iron silicate with a lime-alumina silicate, 
because here all the materials necessary to make an ayerage am- 
phibole would be furnished with a minimum of transportation. 
In a less basic rock the diallage might be of the required compo- 
sition, and thus be uralitized before any reaction rims formed ; 
but in the St. John gabbro, apparently, this was not so. In the 
formation of the uralite, the surplus magnesia and iron from the 
olivine and hypersthene would combine with the surplus alumina 
from the feldspar to make spinel, the position of the spinel being 
perhaps determined by the relative solubility of its acid and 
base, the more soluble magnesia and iron being more easily 
transported. 
With regard to the inner zone, I have seen no evidence that it 
is secondary; on the contrary its continuity with the large 
hypersthene grains, its refusal to follow the outer zone into 
secondary cracks, and its fresh and granular character, lead one 
to believe that like most other occurrences of hypersthene it is 
an original mineral, which collected around the olivine crystals 
on account of their attractive influence, partly chemical, partly 
physical, on the molten magma. 
Dr. Adams * has described rims very similar to these sur- 
rounding the olivine in the saguenay anorthosites, and also 
mentions them as occurring in the Dolin’s Lake gabbro. He 
regards them as original, and the same view is taken by Dr. G. 
H. Williams + as to the pyroxene-actinolite rims in the perido- 
tites of the Cortland series. On the other hand Dr. Vogt { and 
Dr. Patton § speak of. hypersthene-actinolite rims as produced 
by metamorphic action while Dr. Becke || and Prof. Teall § 
consider the somewhat similar anthophyllite-actinolite rims as 
* Amer, Nat., Nov., 1885, p. 1087; Neues Jahrb., B. B. VIII., p. 466. 
+ Amer, Jour. Sci., Jan., 1886, p. 35. 
{ Zeit. f. prakt. Geol., April, 1893, p. 132. These rims, as well as some of those 
described by Lacroix (soc. Franc. de Min., 1889, p. 223),are very like ours in com- 
position and structure. 
§ Rep. Mich. State Board of Geol. Survey for 1891-2, p. 185. 
|| Min u Pet. Mitth. IV., p. 450. 
{ Min. Mag., 1888, p. 116. 
