410 REPORT—1905. 
been produced by the refusion of once solid material. The evidence only points 
to granite resulting from the effective mixing of heterogeneous materials rather than 
to its being a mere residuum left after the abstraction from a magma of its basic 
material. 
(iii) Original Constitution of the Magmas.—There is nothing to show that the 
earth was ever homogeneous in composition, or even composed of anything 
approaching homogeneous layers. We must consequently accept even the oldest 
plutonic rocks as the result of the fusing of very diverse materials, the diversity 
increasing as more and more sediment came under the influence of retusion, from 
limestones with little or no silica to sandstones with up to 100 per cent. By the 
intimate mixture of all classes granite is evidently produced; but where the mixing 
is dependent to any considerable extent on diffusion heterogeneity would prevail 
for long periods of even geological time, and segregation might even locally increase 
it. All the usual features of a petrographical province, including the predomi- 
nance of basic lavas but of acid plutonic rocks, are readily deducible from such 
conditions. 
(iv) Circulation of Material—oOn the above basis it will be seen that there 
must be circulation of material between the igneous and sedimentary rocks 
Starting on the first-formed rocks of the cooling earth, the agencies of denudation 
have been constantly at work sorting the materials subjected to their action. Free 
silica, unaltered from its original form as quartz, has been separated from its 
associated minerals and formed into sandstones ; felspar, altered indeed but not out 
of all recognition, has given rise to clays and shales; while the ferro-magnesian 
minerals, far more profoundly changed, have provided the material for limestones 
and dolomites, as well as colouring and cementing matter for rocks of all classes. 
. Sedimentation indeed constitutes a very real analysis of the rocks on which it 
has to work. The accretion of the plutonic igneous masses, as we now know them, is 
precisely the reverse. In its course the varied materials which have been so 
laboriously sorted out by mechanical and chemical agencies are melted down, as it 
were, in one vast crucible. They are subjected to a very effective process of 
synthesis, and eventually cool down to become once more available for fresh sorting 
operations when a new cycle begins in its turn. 
8. A Consideration of the Archean Period of the Continents of North 
America and South Africa, with reference to Mineral Occurrences. 
By E. F. Henwace. 
The study of Archean rocks, though it may lack the peculiar interest attach- 
ing to that of fossiliferous strata, and is beset with more difficulties in respect of 
the decipherment of their history, is nevertheless one of much interest and of very 
great importance, for in the right reading of their history lies a key to the secret 
of the formation of the deposits of mineral wealth, 
The author considers first the continent of North America, and the connection 
of its Archzean rocks with its mineral wealth. Using the word ‘ Archian ’ (so far 
as regards America) as including Logan’s Laurentian and Huronian systems, the 
Archean rocks cover an area of about 2,000,000 square miles in Canada and 
Labrador, and large areas from Newfoundland to Georgia. They form the 
basis of the chief mountain ranges, and occur in many isolated parts of the con- 
tinents. - The Huronian rocks are characterised by vast beds of iron ore, of which 
the most noted are those of the Marquette district of Lake Superior, where the 
hematite is closely associated with jasper and intermingled with quartzite, chert, 
gneiss, and diorite, as it is also in the Menominee district and in the Vermilion 
Lake district. The origin of these deposits has been a matter of great controversy. 
In New York are extensive beds of magnetite ore, as well as in the Adirondack 
region and in Virginia and North Carolina. 
Titaniferous magnetite ore, occurring in labradorite rocks in Canada and the 
Adirondacks, is also of much scientific interest, and seems to favour the hypothesis 
