26 THE ROYAL SOCIETY OF CANADA

the magma differentiated to form peridotite with dunite inclusions
at the centre and pyroxenite and gabbro at the borders; (3) that
since the heavier peridotite masses, which presumably crystallize first,
occur in the centres of the intrusive bodies and not to one side, the
bodies are not sills or laccoliths, but batholithic in nature; and (4) that
some of the gabbros are younger than the ultrabasics.

Although it is often stated that ultrabasic rocks are derived from
a parental basaltic magma by magmatic differentiation, in recent
years several workers have suggested that certain bodies were formed
by the intrusion of a primary ultrabasic magma. The field evidence
suggests the latter origin for the ultrabasic rocks of the Fort Fraser
area. The facts may be summarized as follows:

(1) The two largest ultrabasic bodies are 65 and 80 square miles
in area respectively. If they represent differentiates of a primary
basaltic magma from which the much larger feldspathic complement
had been squeezed out, one would expect to find these more acid
components abundantly represented in the surrounding rocks. One
would also expect to find gradational phases such as feldspathic
peridotites. These features are not present in the area. The soda
granite body near Tsitsutl! Mountain cuts the nearby ultrabasic
body. Evidence indicates that this granite may be early Tertiary
in age and hence much younger than the Jurassic ultrabasics. Al-
though the contact between the ultrabasic body south of Stuart Lake
and the hornblende diorite to the south-east was nowhere actually
observed, the strike of it would indicate that the diorite is the younger.
The small stock of augite diorite south of Redrock Lake definitely
cuts and contains inclusions of the adjacent peridotite. The above
facts indicate the acid intrusives in the area are younger than the
ultrabasics, and probably in part much younger.

(2) As previously pointed out, the two large ultrabasic bodies
consist of a central mass of peridotite bordered by pyroxenite and
gabbro. These relations are characteristic of magmas that differen-
tiate in place.

(3) No crushing of the crystals was observed and such might be
expected if the ultrabasics originated by a process of squeezing
residual magma from a mass of already formed crystals as postulated
by magmatic differentiation of a basaltic magma.

The Palaeozoic rocks in the vicinity of Trembleur Lake are folded
into a major anticlinorium whose axis strikes north-west crossing the
lake about the centre. The Mount Williams ultrabasic body occurs
along the axis of this anticlinorium. The batholith is elongated
parallel to the strike of the invaded strata. The Middle River valley