Professional Documents
Culture Documents
Gold Metallogeny and Exploration
Gold Metallogeny and Exploration
net/publication/272820535
CITATIONS READS
13 2,364
2 authors, including:
William Shilts
University of Illinois, Urbana-Champaign
209 PUBLICATIONS 2,898 CITATIONS
SEE PROFILE
All content following this page was uploaded by William Shilts on 09 March 2015.
11.1 Introduction
In regions that were glaciated in the Quaternary, mineral exploration can be hampered
by the complexity of the surficial sediments, which are largely allochthonous in
relation to the bedrock they overlie. In the context of almost totally glaciated
landscapes, which includes most of North America north of 40 oN, Greenland and
Iceland, most of Europe north of 50 oN , extensive parts of Asia north of 60 oN, parts of
South America, and Antarctica, the sediments have particular characteristics that
influence the selection of sample media, sampling design and interpretation of data.
Foremost among these sediments is till or its recycled derivatives. Till represents the
texturally and compositionally heterogeneous debris eroded by and carried directly
within or on a glacier. Once deposited, till has distinctive provenance features: (1) it is
a first-derivative sediment; (2) it is widespread; and (3) it serves as the parent material
for most other glacial and non-glacial surficial sediments.
The nature of glacial dispersal and the resultant dispersal trains, particularly as related
to mineral exploration, have been described in general by a number of authors in the
last decade (e.g. Shilts, 1975, 1976, 1982a, 1984; Hirvas, 1977; Minell, 1978; Miller,
1984; Geological Surveys of Finland, Norway and Sweden, 1986a-f; Salonen, 1986a,
b, 1987; Strobel and Faure, 1987; Clark, 1987; and Coker and DiLabio, 1989-to cite
a few examples).
During the Quaternary, till was produced by the glacial erosion, transport, and
deposition of fresh and weathered unconsolidated sediments and bedrock. Till is,
therefore, a geologically young sediment which at any given site is not an in-situ
weathering product, but a lithological summation of source units up-ice from the site.
Debris from any size of source unit is dispersed down-ice to produce a ribbon-shaped
or fan-shaped dispersal train, comprising a body of till that is enriched in debris from
the source relative to the till surrounding the train. Shilts (1976) has shown that,
ideally, a plot of the abundance of glacially dispersed debris vs. distance down-ice
approximates a negative exponential curve (Figure 11.1), in which the concentration
of a component reaches a peak near its source (i.e. within a zone referred to as the
'head' of dispersal) and then declines exponentially to background levels down-ice
1100
UL1RABAJ
~ 2I'm FRACTION OF TILL
900 OUlCROP ....
.6410' m ·FRACTION OF TILL
E
~ 700
z:
500
300
100· -'--
.----HEAO---- - - - - - TAIL
~
i5
OUTCROpl.OF'
en COMPONENT "X"
- DIRECTION OF GLACIAL FLOW
>
""
<..)
~
?<
.....
3z:
o
~
o<..)
o 20 40 60 80
Km IN DIRECTION OF ICE FLOW GSC
Figure 11.1 Dispersal curves for nickel in till, Thetford Mines area, Quebec. Actual (top) and
idealized (bottom) curves show the relationship of the head and tail of a negative exponential
curve, (After Shilts, 1976.)
(i.e, in a much larger zone termed the 'tail' of a dispersal train). The dispersal tail is
generally many times larger than the head, and as a result it is generally the part of a
dispersal train first detected by till sampling programmes. An advantage of
prospecting in glaciated terrain is that, if it can be detected, a dispersal train of
distinctive boulders, minerals, trace and/or major elements, or radioactive components
may enhance the size of mineral exploration targets by several orders of magnitude. A
major objective of till geochemistry, then, is simply to detect the tail of a dispersal
train, trace it back to its head, and locate its source.
The composition of a till sample at a given site is truly like a fingerprint - no two
samples are exactly alike, despite the well-known propensity for glaciers to
homogenize bedrock signals through thorough mixing of drift components during
transport and deposition. The sediment composition is potentially the net result of
mixing all bedrock lithologies from the sample site along a flow line to the dispersal
centre for the glacier that deposited it. This complicated melange is further influenced
by the width of outcrop of each lithology traversed, the topography of the bed, and the
338 GOLD METALLOGENY AND EXPLORATION
changing dynamic conditions of the ice itself during entrainment, transport, and
release of the particles of which the sediment is composed. These, among many other
factors, make the likelihood remote that one sample will be strictly comparable in all
compositional aspects to another, even if they are collected in close proximity. Thus,
the 'case history' approach to developing exploration strategies in glaciated terrain
has often foundered; it is a much more useful strategy in areas of residual soil where
only easily observable local factors influence soil composition.
In glaciated terrain, the interpretation of the composition of samples and their
intercomparisons can be done most effectively by understanding glacial sedimentation
principles, history of the depositing and previous ice sheets, and nature of post-glacial
weathering, periglacial disturbance, fluvial erosion, etc. In other words, comparison
of drift prospecting data gathered on similar mineralization in contrasting glaciated
settings is likely as not to lead to false interpretations. Evaluation on the basis of
known glacial and post-glacial history and on principles of glacial sedimentation,
while considerably more difficult than direct comparison and requiring input from a
geologist trained in glacial geology and/or drift prospecting, is far more likely to
produce viable exploration targets in glaciated terrane.
In glaciated terrain, the composition of a till sample may be the composite of many
overlapping dispersal trains. Most of the individual dispersal trains are not
identifiable, however, because they are too small or are composed of rocks or minerals
that are not distinctive. The size and shape of the dispersal train are controlled by the
orientation of the source relative to ice flow, by the size and susceptibility to erosion
of the source, and by the topography of the source and dispersal areas, which can trap
trains in valleys, break them into disjointed segments in rough terrane, or even
truncate them against hills or mountains.
Dispersal can occur at a variety of scales ranging from continental (100s of
kilometres), to regional (100 to lOs of kilometres), to local « 10 km), to small-scale
(final stages of mineral exploration in the 100s to lOs of metres) (Shilts, 1984a).
Continental-scale dispersal trains are found throughout the area originally covered by
the Laurentide Ice Sheet in Canada and the United States of America. Among the most
typical and prominent of these is a train of red till and matrix that extends eastwards
from Dubawnt Group rocks into and across northern Hudson Bay (Shilts et al., 1979;
Shilts, 1982b; Figure 11.2). Continental-scale trains can be detected only when a
characteristic lithological component of the train is present in adequate amounts or is
particularly distinctive against background rock types in the dispersal area. For drift
prospecting purposes, these large trains are significant in that the exotic lithology of
the till can mask the lithology and geochemistry of mineralized debris eroded from
local sources, for example the Palaeozoic carbonate trains that extend across
greenstone belts south-west of Hudson Bay (Geddes and Kristjansson, 1986; Gleeson
and Sheehan, 1987). Large trains such as these can be detected by sampling at
densities as low as one till sample per 100 km 2 , the sort of sampling carried out during
surficial geology or bedrock geology mapping in Canada.
Regional-scale dispersal trains are more likely to be detected in the preliminary
stages of mineral exploration programmes and may reflect mineralization or bedrock
environments suitable for mineralization. At this stage of exploration, one till sample
per km 2 will define which parts of a favourable bedrock unit are most metalliferous
and may even detect the tails of dispersal trains derived from small sources. This
sampling density may provide geochemical targets that should be sampled at a
GEOCHEMICAL EXPLORATION FOR GOLD IN GLACIATED TERRAIN 339
'00·
~-
92'
---<-- ... " SQ'
Figure 11.2 Major dispersal trains around Hudson Bay. (Redrawn and adapted from Shilts,
1982b, and Kaszycki and DiLabio, 1986.)
detailed scale to differentiate areas of high background metal levels from those
resulting from several overlapping small trains derived from areas of mineralized
bedrock.
'Detailed' sampling, in which sample spacing is on the order of lOs to 100s of
metres, is designed to locate heads of dispersal trains. This sample spacing is designed
to trace dispersal trains up-ice or to test geophysical anomalies and/or favourable
geological structures and contacts. An idealized glacial dispersal model (Miller, 1984)
serves to illustrate some of the characteristic features of 'normal' dispersal trains
(Figure 11.3): they are generally ribbon-, fan- or flame-shaped in outline; they have
abrupt lateral edges with the surrounding barren till; and the concentration of the
distinctive component within a train decays rapidly down-ice. 'Normal' dispersal
trains are formed near or at the base of glaciers by repeated cycles of erosion,
entrainment, abrasion, deposition and re-entrainment. At the detailed scale of
sampling, post-glacial mobilization of trace elements in groundwater and soil water
may spread the dispersal train downslope, partially obscuring its original shape.
Recently, another type of dispersal train, formed by late glacial ice streams, has
been recognized. Trains formed by ice-stream mechanisms are often long and have
abrupt lateral and distal terminations. Component concentrations may decay
340 GOLD METALLOGENY AND EXPLORATION
.. B'
I
c'
I
0'
I
G I I
I
I
!~~
I
&
SURFACE ANOMALY I MINE~AlIZED
~ SUBOUTCROP
C)
I
,
B 0
CROSS SECTIONS
LONGITUDINAL SECTION
A B c o
Figure 11.3 Idealized glacial dispersal model. (Redrawn and adapted from Miller, 1984.)
insignificantly down-ice, and they may not obey the exponential curve criterion
(Hicock, 1988; Thorliefson and Kristjansson, 1988).
Only in recent times has use been made of samples collected stratigraphically in
areas of deeper overburden, using various drilling techniques to produce
three-dimensional data sets (i.e. Thompson, 1979; Averill and Zimmerman, 1986;
Bird and Coker, 1987; Sauerbrei et al., 1987; Harron et al., 1987; Smith and Shilts,
1987; Brereton et al., 1988; Coker et al., 1988). Variations in these data reflect vertical
changes in stratigraphy, sedimentary facies, and diagenesis - particularly weathering
during non-glacial intervals such as the present.
Till can be thought of as the first derivative of bedrock (Shilts, 1976). Sediments
resulting from the reworking of till or other unconsolidated sediments (i.e. stratified
drift) are second-derivative sediments; they have been subjected to sorting and have
undergone an episode of transport in air or water, often along a different path from the
direction of the original ice movement. In this way, glaciofluvial gravel and sand
represent the coarse fractions, and glaciolacustrine silt and clay represent the fine
fractions derived directly from the texturally heterogeneous debris at the base of the
glacier or entrained in ice near its base. Indirectly they may be formed by meltwater
erosion of already-deposited till. It is difficult to interpret the provenance of these
sediments because they have travelled along transport paths consisting of at least two
vectors, transported first by ice, then by water. Till is clearly the optimum glacial
GEOCHEMICAL EXPLORATION FOR GOLD IN GLACIATED TERRAIN 341
sediment type to use in mineral exploration, because it has the least complicated
source-transport-deposition history.
Data on ice-movement directions may be obtained from a variety of glacial features
including: striations; glacially moulded, streamlined landforms; dispersal trains; and
fabric of glacial diamictons (till) or palaeocurrent measurements of associated ice-
contact fluvial sediments. Ice-flow directions, estimated by measurement of striae, are
not always the most significant flow directions in terms of drift transport (Shilts,
1984a). It has been noted at several sites that the bulk of the till was deposited by
movement of ice in a direction different from the ice-flow direction indicated by the
youngest set of striae (Veillette, 1986; Kaszycki and DiLabio, 1986a).
The past decade has seen a substantial increase in our knowledge of the relationship
of glacial sedimentation, dispersal patterns, stratigraphy and ice-flow directions to
practical problems in mineral exploration. Several examples of the integration of the
principles of glacial geology and mineral exploration can be cited:
(i) The Nordkalott Project, which included a regional surficial geochemistry and
mapping component, was carried out by the Geological Surveys of Finland,
Norway and Sweden north of latitude 66°N (250000 km 2 ) on the
Fennoscandian peninsula (Geological Surveys of Finland, Norway and
Sweden, 1986a-f, 1987). Ice-movement directions were systematically
recorded, revealing areas of simple unidirectional ice flow in coastal areas and
multiple directions of ice flow inland. Trenching and drilling, using
techniques developed in major regional sampling projects carried out by the
Geological Survey of Finland in the late 1960s and 1970s, provided data on
the Quaternary stratigraphy. Data on glacial geology were used to interpret the
regional geochemical patterns that emerged.
(ii) In Labrador, Klassen and Thompson (1987) identified ice-flow patterns that
are simple near the coast, becoming complex inland, reflecting the
complicated ice-flow history of the shifting Labrador-Nouveau Quebec Ice
Divide. These ice-flow patterns cause dispersal trains to be ribbon-shaped
near the coast and fan-shaped or amoeboid inland (Figure 11.4; Klassen and
Thompson, 1989).
(iii) In Nova Scotia, Stea et al. (1988) found, as in Labrador, that complex
dispersal is recorded in areas of shifting centres of glacial outflow. These
authors were able to classify different areas of Nova Scotia as to their
expected sequence of ice-flow events based on dispersal of various indicators
from lithologically distinctive source outcrops.
(iv) Veillette (1986, 1989) identified three ice-flow events in the Abitibi region of
western Quebec (Figure 11.5), and showed that the intermediate one was
responsible for the bulk of the drift transport. In an area of active exploration
by drift prospecting, this interpretation was immediately useful in exploration.
The most important aspect of data collection, and the resultant sample treatment and
geochemical analyses of glacial overburden, starts in the field or at the drill site, where
it is essential to make the best possible identification of the type of glacial sediment
342 GOLD METALLOGENY AND EXPLORATION
"'cr---_____________---------r",------,"l7"
10" ')9"
-75
1\
Quebec
Labrador
/ 'i!,Y7f
d.. "/
~fo/
~
4
':~'~~~'\
.
...
':". '
e. :
'.'
'
••
_/
/
:/
/
100 krn
~--'----"
-
'J;~oo _ _ _ _ _ _ _ _ _ _ _ _ _- - - - - - ; ' ; ,
Figure 11.4 Dispersal trains in Labrador (from Klassen and Thompson, 1989). Broad fan shapes
reflect transport in two or more phases of ice flow.
eo'l
'YJO •
Detour lake Mine
ONTARIO
.
lebel· sur· Ouevlllon
• Sennetcu re
o km 'YJ
Figure 11.5 The three ice-flow directions recorded in the Abitibi-Timiskaming area of Quebec
and Ontario. (After Veillette. 1986.)
drilling. To date, however, no cost-effective drilling system for recovering large till
samples at intermediate depths, from 10 to 20 m, has been devised.
In reverse-circulation drilling, water, sometimes used in conjunction with com-
pressed air, is pumped down the outer tube of a system of dual tube rods. The water
mixes with the cuttings at a tricone bit (tungsten-carbide buttons) and the slurry is
forced to the surface through the inner tube. The sample slurry discharges into a
cyclone, to reduce the velocity of the discharge material, and empties through a 2 mm
(10 mesh) sieve into a series of sample buckets. Logging is carried out by the
Quaternary geologist on the drill. The geologist sees a washed and disturbed sample,
which makes accurate logging difficult for qualified personnel and impossible for
unqualified personnel. There is only one chance to log the sample material as it goes
by. In addition, the fine fraction of the material, such as the ore minerals and including
some forms of gold, is generally lost (Shelp and Nichol, 1987) and becomes
cross-contaminated by the recirculated water.
The rotasonic drill uses high-frequency resonant vibration produced by a rotating
eccentric cam (averaging 5000 vpm) and rotation of the drill bit to obtain continuous
solid cores. Sediments are cored with tungsten-carbide-tipped bits. The cores are
extruded, in 5 ft lengths, into plastic sleeves and placed in core boxes. Logging and
sampling of the cores can be carried out on-site or at a later stage. Rotasonic drills may
produce cores that are longer, shorter, or equal in length to the interval sampled. These
variations appear to be due mainly to the manner in which different sediment types
react to the drilling stresses. Deformation due to the drilling also has been observed
in rotasonic drill core (Smith and Rainbird, 1987), and can interfere with interpreta-
tions of sediment origin, particularly where deformations may be confused with struc-
tures caused by glacial overriding or proglacial, subaerial, or sublacustrine mud flows.
For gold analysis, sand-sized heavy mineral concentrates (HMCs) are the only
reliable media that can be prepared routinely from samples obtained by reverse-
circulation drilling. Although any grain size can be analysed selectively from
rotasonic core, the material must first be disaggregated, which, in the case of
overconsolidated tills, can be difficult and time consuming.
346 GOLD METALLOGENY AND EXPLORATION
The heavier materials are separated from the lighter materials by agitation in water
on a shaker table (see Sivamohan and Forssberg, (1985) and Stewart (1986) on the
principles of tabling). Once again, fine materials, probably fine gold, are lost (Shelp
and Nichol, 1987). The tabled heavies are dried and the magnetic fraction removed
(using a magnet) and stored. For gold mineralization associated with iron formation,
the usefulness of the removal of the magnetic fraction should be carefully considered.
The non-magnetic fraction is further concentrated using a heavy liquid (e.g.
methylene iodide, SG = 3.3) separation technique. The samples are cleaned and dried,
and in some instances crushed and ground to homogenize them, before being
analysed. In many projects, the pre concentration on a shaker table is omitted, the
sand-silt-sized material being fractionated directly by submersion in a heavy liquid.
U sing this method, little is lost but concentrating samples large enough to avoid the
'nugget effect' in gold analysis is time consuming.
With rotasonic-core or surface-till samples, the fine fractions (e.g. clay-sized
material « 2 /lm) and/or silt and clay sizes combined (-250 mesh; < 63 /lm» as well
as the sand-sized heavy minerals, can be separated and analysed. If a till is sufficiently
cohesive to come up the reverse-circulation drill in balls or lumps, it is also possible
to collect, in the sieve above the collection bucket, a relatively undisaggregated
sample from which to obtain the fine fraction.
HMCs and the other size fractions of tills are commonly analysed for a wide but
varied suite of elements, determined by the type of mineralization and nature of
deposit being sought. Whether heavy minerals are used is determined by the oxidation
state of the till and by the presumed resistance to weathering of the minerals being
sought (Shilts, 1975, 1984; Shilts and Kettles, in press). All analytical work should be
quality- controlled using reference control and duplicate sample analyses. Analytical
techniques generally include various combinations of fire assay (FA) - dissolution -
atomic absorption spectrometry (AAS)/direct current plasma or inductively coupled
plasma emission spectrometry (DCP-ES or ICP-ES) or mass spectrometry (MS)
methods as discussed by Hall and Bonham-Carter (1989). This series of analytical
techniques usually involves sample-splitting, which, in the case of analysis for gold,
particularly in HMCs, may lead to spurious or non-comparable results, due to the
'nugget effect' (Clifton et al., 1969). These techniques also commonly involve
destruction of the sample. Neutron activation allows non-destructive analysis of the
whole sample, but one must be aware of the type of irradiation used, because some
samples, depending on their matrices and/or chemistry, will be rendered permanently
too radioactive to handle. Non-destructive analysis and subsequent recovery of the
whole HMC sample facilitates later mineralogical work on anomalous samples to gain
some insight on the nature of the mineralization, its geological environment, and the
distance of transport of ore fragments.
Contamination by elements such as Wand Co, due to fragments from tungsten-
carbide bits, also can be detected during examination of HMCs. Although some
laboratories offer to analyse grain morphology, caution must be exercised in utilizing
the shape of gold grains, or any other mineral grain for that matter, as an indication of
distance of transport. Variability in the original morphology of the grains and the
manner in which they were glacially transported are far more important in determining
grain shape than distance of transport. In the case of style of transport, the grain shape
has radically different implications, depending on whether the particle was
GEOCHEMICAL EXPLORATION FOR GOLD IN GLACIATED TERRAIN 347
transported in the active basal zone or over long distances in the passive englacial
zone of the ice.
All too often larger clasts (i.e. those > 1mm) are routinely discarded from samples.
These clasts are rock fragments, and when counted can give a useful picture of the
petrology of the lithologies across which the depositing glacier flowed (see Bird and
Coker, 1987). In addition, sand-sized or finer labile minerals, which can be destroyed
by weathering, may be enclosed within siliceous, coarser rock fragments from which
they can release a geochemical signature on crushing and analysis.
11.5 Occurrence of gold in till and soil and the effects of weathering
In recent years, more attention has been placed on trying to understand the
comminution behaviour of ore minerals during glacial erosion and deposition, the
residence sites of metals in tills, and the effects of weathering on trace metal contents
(Shilts, 1975, 1984). Studies have shown that the mineralogy, petrography and major
element chemistry of tills are clearly dependent on till-forming processes as well as on
bedrock variations (Haldorsen, 1977, 1983; Taipale et al., 1986).
Fractionation experiments on gold-bearing till (DiLabio, 1982a, 1985, 1988;
Guindon and Nichol, 1983; Nichol, 1986; Shelp and Nichol, 1987; Coker et al., 1988)
I) (b)
0000
Waverley Till
150m 10000
Beaver Dam
,..., ...
/
/
/
'\ /
1000 1000 / \ I
/ \ /
/ \ /
/ ~
:cc. J
100 c. 100
:>
<C
10 t:. ",2000
10
[:J <2000 ~180
o <180~ 63
• <63 ~ 2
o <2
• HMC
__ Au Zone
Figure 11.6 Abundance of gold vs. grain size of analysed fraction of till at (a) different distances
down-ice from a gold deposit at Waverley, Nova Scotia (after DiLabio, 1982a, 1985, 1988), and
(b) along a sample line traversing the Beaver Dam gold deposit, Nova Scotia. (Coker et ai., 1988.)
348 GOLD METALLOGENY AND EXPLORATION
indicate that gold distributions in various particle sizes of till are complex, because of
the combined effects of (a) glacially comminuted detrital particulate gold; (b) the
grain size of re-precipitated gold released during the weathering of sulphides; (c) the
grain size of precipitated or adsorbed gold; and (d) the original grain size of native
gold at its source. In general, till is richest in gold in its finer size ranges (Figure 11.6),
although coarse fractions may also be auriferous (Figure 11.6).
At the base of the thick (> 30 m) Quaternary sequence at the Owl Creek gold deposit
near Timmins, a green till and its oxidized equivalent are preserved. The unoxidized
(a)
5 N
OWL CREEK
" •
® >10
>20 ppb
o < 10
lK
Tw/~
GF
iA
n
LK - Lacustrine sediments
~Om
GF - Glaciofluvial sediments
(b)
100,000
10,000
Au
ppb 1,000
Key'
• Gold minerali:lotion
100 • N M H fraction
... ·6Jum fraction
10
Figure 11.7 Comparison of the gold distribution in the heavy mineral concentrate (HMC) (NMH
= non-magnetic heavy mineral concentrate) and the < 63 Ilm fractions of till at (a) the Owl Creek
gold mine, Timmins, Ontario, and (b) associated with gold mineralization at Hemlo. Ontario.
(From Shelp and Nichol, 1987.)
GEOCHEMICAL EXPLORATION FOR GOLD IN GLACIATED TERRAIN 349
green till contains most gold and abundant fresh pyrite in the fine sand sizes (DiLabio,
1985, 1988). In the overlying oxidized layer, pyrite has been altered to earthy
limonite-goethite that ranges in size and habit from sand-sized pseudomorphs after
pyrite to amorphous silt- and clay-sized grains. As the oxidized till is much more
auriferous in all size fractions than the unoxidized till, it was concluded that gold had
been added to the limonite-goethite grains, perhaps from groundwater.
Work by Shelp and Nichol (1987) at the Owl Creek gold mine shows that both the
gold content of the HMCs and of the < 63 !lm (clay and silt) material clearly depict
anomalous glacial dispersal trains (Figure 11. 7(a». The lengths of the anomalous
dispersal trains are similar in both cases, although the contrast and absolute levels of
gold in the < 63!lm fraction are lower. For Hemlo-type mineralization
(Figure 11.7(b », they found that HMCs anomalously enriched in gold are restricted to
drift lying directly over the deposit and show little evidence of dispersal. The
< 63!lm fraction shows significant dispersal. Thus, analyses of fine fractions and
HMCs can provide different, often complementary, information.
It should be apparent that the suitability of the use of the conventional -80 mesh
material for till geochemical surveys should be carefully evaluated on a project by
project basis. If possible, orientation surveys should be conducted for each project
area to determine the optimum size fraction to use. If it is not possible to carry out a
meaningful orientation survey, fine-grained materials « 63 !lm) should be used,
because they are easiest to obtain, and in many cases they are the best size fraction for
reflecting gold mineralization. The nature of the weathering history of the material
also determines the appropriateness of the grain size to be used. In tills that show
evidence of weathering, the use of data on the geochemistry of the fine and HMC
fractions to complement each other may well increase the effectiveness and reliability
of gold exploration.
Drift prospecting is the use of data on the geochemistry and lithology of glacial
sediments (mainly till) to identify economically significant components in the
sediments and to trace them up-ice to their bedrock source (DiLabio, 1989). The
concept of predictable patterns within dispersal trains (e.g. exponential decay, as
shown in Figure 11.1, and the shape of trains, as shown in Figure 11.3), when
considered during the design of a geochemical exploration programme, will influence
the choice of sample types, the sampling plan, the analytical scheme, and the
interpretation of the data. Compared to geochemical anomalies in unglaciated areas,
once a part of a glacial dispersal train has been detected, it can be traced more easily
up-ice to its source because simple clastic dispersal is the main mechanism involved
in the formation of a train. This assumes, in the case of gold, that the gold observed
visually is from a bedrock or discrete source and is not formed in situ by the
weathering of sulphides carrying trace amounts of gold.
In the last decade, many studies of gold dispersal trains have been documented
within Fennoscandia and Canada. Details of most of these studies/and case histories
are summarized and referenced in Coker and Dilabio (1989). It is impractical to go
into much depth of discussion for all of these examples. The focus has been on studies
which are most familiar and which best make the points we wish to emphasize, such
350 GOLD METALLOGENY AND EXPLORATION
.----
10 .. <10.
50 .. 2:10<50 ..
100. :::50<100.
Ore zone. ~ :::100 ..
Ice movement ..... ?
C ><c
/
.,u I
I
I
I
><c /
/
L -_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ~
.----
10 <10. 500. < 500.
Figure 11.8 Beaver Dam gold deposit, Nova Scotia. The distribution of gold in (a) B-horizon
soil; (b) whole till; and (c) heavy mineral concentrates. (From Coker et al., 1988.)
A reverse circulation overburden drilling programme carried out at the Golden Pond
gold deposit, Casa Berardi, Quebec, by Sauerbrei et ai. (1987) identified a thin lower
till containing anomalous geochemical concentrations of gold and abundant gold
grains. Of note was the discovery that glacial dispersal of gold was very limited
(200-400 m). This was because the ice that deposited the lower till moved subparallel
to the strike of both the mineralized structure, itself recessive, and a bedrock trough,
which further confined the dispersal train.
GEOCHEMICAL EXPLORATION FOR GOLD IN GLACIATED TERRAIN 351
In Timmins, Ontario, a reverse-circulation and rotasonic drilling programme was
carried out by Bird and Coker (1987) at the Owl Creek gold mine. This revealed a
deep and complex overburden stratigraphy with up to four glacial sediment packages,
each with different ice-movement directions (Figure 11.9). In the lowest (Older) till,
on bedrock, dispersal is very local being truncated against a bedrock ridge. The
highest gold concentrations in till are adjacent to the subcropping gold mineralization.
Although the overlying Matheson till has not been in contact with mineralization or
bedrock, it contains gold reworked from the lower (Older) till. Matheson dispersal is
longer, approximately 600 m, and the area of maximum gold concentration in till is
displaced some 300 m down-ice from the subcropping gold mineralization.
STRATIGRAPHY
COCHRANE
~ ~
M.ATHESON
~
40lD£A"
~ ~ ·OLOEST-
GOLD (ppb)
?~ "",B,,,,
IEDROCK GOLD (ppb)
'OLDER' TILL MATHESON TILL
HEAVY MINERALS HEAVY MINERALS
Figure 11.9 Gold contents in heavy mineral concentrates from the 'Older' and Matheson Tills at
the Owl Creek gold mine. Timmins. Ontario. (Redrawn and adapted from Bird and Coker. 1987.)
In the Hemlo area, at the Page-Williams gold deposit 'A' zone, Gleeson and
Sheehan (1987) sampled till using percussion drills with 'flow through' bits. They
found that the upper 'exotic calcareous till gave little indication of the gold
mineralization (Figure 11.10). This till is probably analogous to the ice-stream tills
(Hicock, 1988; Thorliefson and Kristjansson, 1988) of the Beardmore-Geraldton
area. In areas underlain by this type of till, the geochemical and mineralogical
signature of the underlying bedrock is almost totally masked. The underlying,
locally-derived limonite-rich till gave good response to the gold mineralization
(Figure 11.10) in all size fractions and HMCs, using gold, arsenic. antimony,
molybdenum, mercury, tungsten, and barium. Once again, dispersal was short (i.e. 200
m), partly because the deposit lies in the lee of a bedrock-high and is protected, and
partly because dispersal is truncated against a bedrock ridge down-ice.
352 GOLD METALLOGENY AND EXPLORATION
0 c
-
...'"
QI
QI
~
2
-
.5
s:; 4
Q.
QI 5
0 ORE ZONE
100m horizontal
Figure 11.10 Page-Williams gold deposit, 'A' zone. Hemlo, Ontario: bedrock topographic
profile and geochemistry across the mineralization. (From Gleeson and Sheehan, 1987.)
Perhaps one of the most frustrating and commonly unresolved problems of gold
exploration in glaciated terrain is determining whether high gold concentrations in
drift or preserved pre-glacial regolith or alluvium are the result of clastic dispersal
from a discrete bedrock source ('mother lode') or whether the gold was formed and
concentrated by secondary, low-temperature, geochemical processes acting on a
diffuse, non-economic source. In the latter case, further complications arise in
determining whether low-temperature processes might have been associated with
prolonged weathering over a broad area or whether hydrothermal action caused gold
to be precipitated in discrete zones. Concentration of geochemically immobile gold
formed by these processes could have been by placer-forming processes or by intense
weathering over a time period sufficient to remove much of the silicate part of the soil.
GEOCHEMICAL EXPLORATION FOR GOLD IN GLACIATED TERRAIN 353
LEGEND
Greenstone / aru:lesde
,;-----I
~nyollte
••
Gt3nod l oJlliI!!!
O.Orile
'\..
r' •
/, .. /
"
••
O ri ental I on 01
. /' \ /./' -,
\1' •
Su belQP 01 E P lone
i ./~
"*
Gold pay :Uf .... 1t. .n t i l l l , • •••
2·~
t'8In! \ \ .. .'i·· .~.I'·· , ..
/'. \
c=3 Low grade d'5pe J$IOn train
!'imI 3
••
High gr.illde dlsp.rslon
P~ .. t.. 1 50nu:; 0 ,111 hoi.
Follow up IoniC dlil l ho le
\. ~'4 .~ •
.... .."""".',
2 ! .;-'--":l). "
\ •• .,-. .:......,
. .'\. .-------~'".\
••
•
.. . . .....
• "II • ••• • • ".
. .. e.. .. ..r.,
,
/\~ \
---'-
....
. ::iJ~
. .. GOmbl:.~. \
e. . '
2
:~-..........
""
.:;,1'-:•• /'(!)i"• •
•• ,
\
\
\
(
".
.......... '--;-!#
• •• I 4
1:;- • 2 \
'-., '-.. J
\
'.
' -. 1 '. . ' \ (.
\ ~'dd l. TI'· , \. 2
'\
\ \.
". '
(.
"
3
". \
\
.
"
',\
\ , '\
.~". ....1 ,
KoSI. "'-. 2 ( I ,/
C,"""'"
,
n il ,
,-.~ ".
,
(,,-.-._j
.........
-.
Figure 11,11 Glacial dispersal train from the EP gold zone, Waddy Lake, Saskatchewan. (After
".
'.
Averill and Zimmerman, 1986.)
For over a century, in both the Cordillera and south-eastern Quebec, these questions
have intrigued and frustrated prospectors working in placer deposits. The voluminous
literature on the problem of placer gold in the Yukon and British Columbia is beyond
an adequate discussion in this chapter, but the interested reader is referred to recent
work by Morrison and Hein (1987) and Eyles and Kocsis (1988).
354 GOLD METALLOGENY AND EXPLORATION
Th.o,.Uc:al
SU.U9'.~ltolc Column
D so"ed "nd end gea'01 ~ bedeock w"h leesh selph'des and quae" ,e,,'5
~ pood, socled d,am,clon angela' clasls D wea,heeed ,one en bed,ock and o'e,bccden
Figure 11.12 Interpretation of borehole records across Riviere Gilbert Valley, Quebec, in which
gold-bearing regolith has been protected from glacial erosion and buried under a complex glacial
sedimentary sequence. Periods of pre-glacial and inter-glacial weathering, indicated by shading,
were times of gold concentration by chemical destruction of pyrite and concomitant precipitation
of gold. Pre-glacial, inter-glacial and post-glacial streams concentrated chemically precipitated
and glacially reworked gold into placers. (A) Development of pre-glacial, gold-bearing regolith
and alluvial placers by prolonged weathering; (B) inter-glacial weathering and accumulation of
gold after first glacial cycle (note that glaciers block drainage, forming lakes when they enter or
leave this region); (C) post-glacial weathering, fluvial erosion, and accumulation of gold after
several glacial cycles. (Modified from sketches by Sharon Smith in Shilts and Smith, 1988.)
356 GOLD METALLOGENY AND EXPLORATION
more than 1100 g (Figure 11.13). Repeated glaciations (at least three) removed most
of this weathered regolith and diluted it with fresh, glacially ground bedrock. In the
steeper reaches of some valleys oriented perpendicular to the direction of glacial flow,
pockets of palaeoplacers and the regolith were preserved and mined in the nineteenth
and twentieth centuries.
Gold, which is now widespread in modern alluvium and soil from the same area,
comes from at least two sources: (a) gold reworked through at least three glaciations
by glacial mixing of the pre-glacially-widespread regolith with glacially eroded
detritus derived from local bedrock; and (b) gold which appears to be forming at
present in the solum on till, and presumably also during inter-glacial or inter-stadial
intervals as a result of low-temperature geochemical processes associated with
weathering (Figure 11.12). This is a particularly rapid process compared to pre-glacial
rates of gold formation, because glaciers have crushed, on average, one or more
metres of bedrock to sand and finer sizes, liberating sulphides as discrete grains, and
increasing the chemical reaction surface by several orders of magnitude over that
available on the exposed and fracture surfaces of consolidated bedrock. From
experience, this can translate to bulk, near-surface till samples with 6-10 grains of
visible gold per kilogram in a till containing pyrite averaging less than 50 ppb gold
(Table 11.2).
Table 11.2 Gold concentrations in pyrite from the Eastern Townships, Quebec (from Shilts and
Smith, 1988)
Locality Gold (ppb) Details
Samson River 105 Streambed placer, 20 g, 1 mm-l em pyrite cubes (Locat, pers.
comm.)
Samson River 94 Streambed placer, 20 g, 1 mm-l cm pyrite cleaned of surface
tarnish
Mining Brook 33 Panned, oxidized pyrite, <10 g
Mining Brook 123 Panned, unoxidized pyrite, <10 g
Various 5-65 Unoxidized pyrite from till samples (range from 11 samples)
The studies described above illustrate the special knowledge of glacial geology
required to address fundamental problems in gold exploration in glaciated terrain.
Although glacial geological principles have not resolved centuries-old controversies
and hypotheses of gold genesis in glaciated terrain, they have relatively recently
begun to bring some order to the chaos that the peculiar fever or enthusiasm of gold
discovery provokes.
A number of points have been emphasized in this chapter. Many are areas in much
need of further work and research. These include the need for:
(i) Regional studies of till provenance, including the effects of exotic drift on
local geochemistry.
(ii) Correlation of tills in areas of complex stratigraphy and assignment of
ice-flow directions to tills.
(iii) Investigation of the comminution and weathering behaviour of ore and
ore-generating (sulphide) minerals, particularly gold and platinum group
elements, in order to design better sampling and analytical schemes.
GEOCHEMICAL EXPLORATION FOR GOLD IN GLACIATED TERRAIN 357
(iv) Development of more cost-effective drilling systems, particularly in areas of
intermediate overburden thickness (i.e. 10-20 m).
(v) Education and training of geologists in geochemistry and Quaternary geology
and a commitment by senior explorationists and exploration managers to the
use of qualified people to carry out surficial geochemical and overburden
drilling programmes in till- covered terrains.
References
Averill, S.A. and Zimmennan, 1.R. (1986) The riddle resolved: the discovery of the Partridge Gold
Zone using sonic drilling in glacial overburden at Waddy Lake, Saskatchewan. Can Geol. J.
Can. Inst. Mining Metall. 1, 14-20.
Bird, D.J. and Coker. W.B. (1987) Quaternary stratigraphy and geochemistry at the Owl Creek
gold mine, Timmins, Ontario, Canada. In Geochemical Exploration 1985, part 1, ed. Garrett,
R.G. J. Geochem. Explor. 280/3),267-284.
Brereton, W.E., Briggs, D.N. and Rollinson, J.P. (1988) Till prospecting in the area of the Farley
Lake gold deposits, Northwestern Manitoba, Canada. In Prospecting in Areas of Glaciated
Terrain- 1988, ed. MacDonald, D.R. Can. Inst. of Mining and Metallurgy, 225-240.
Clark, P. U. (1987) Subglacial sediment dispersal and till composition. J. Geol. 95, 527-541.
Clifton, H.E., Hunter, R.E., Swanson, F.J. and Phillips, R.L. (1969) Sample size and meaningful
gold analysis. US Geol. Surv., Prof. Paper 625-C. 17pp.
Coker, W.B., Sexton, A., Lawyer, I. and Duncan, D. (\988) Bedrock, till and soil geochemical
signatures at the Beaver Dam gold deposit, Nova Scotia, Canada. In Prospecting in Areas of
Glaciated Terrain - 1988, ed. MacDonald, D.R Can. Inst. of Mining and Metallurgy, 241-254.
Coker, W.B. and DiLabio, R.N.W. (1989) Paper 30, Geochemical exploration in glaciated terrain:
geochemical responses. In Proc. Exploration '87: Third Decennial Internat. Conf. on
Geophysical and Geochemical Exploration for Minerals and Groundwater, ed. Garland, G.D.
Ontario Geol. Surv., Special Vol. 3, 336-383. 960pp.
DiLabio, RN.W. (1982) Gold and tungsten abundance vs. grain size in till at Waverley, Nova
Scotia. In Current Research, Part B. Geol. Surv. of Can., Paper 82-IB, 57-62.
DiLabio, RN.W. (1985) Gold abundances vs. grain size in weathered and unweathered till. In
Current Research, Part A. Geol. Surv. of Can. Paper 85-IA, 117-122.
DiLabio, R.N.W. (1988) Residence sites of gold, PGE and rare lithophile elements in till. In
Prospecting in Areas of Glaciated Terrain - 1988, eds. MacDonald, D.R. Can. Inst. of Mining
and Metallurgy 121-140.
DiLabio, RN.W. (1989) Terrain geochemistry. Chapter 10 in Quaternary Geology of Canada and
Greenland, eds. Fulton, R.J., Heginbottom, J,A, and Funder, S, GeoL Surv, of Can, Geology of
Canada, VoL I,
Eyles, N, and Kocsis, S,P. (1988) Gold placers in Pleistocene glacial deposits; Barkerville, British
Columbia, Can. Insf. Mining and Metall. Bull. 81 (916), 71-79,
Geddes, R.S. and Kristjansson, F.J, (1986) Quaternary geology of the Hemlo area: constraints on
mineral exploration. Can. Geol. J. Can. Inst. Mining Metall. 1 (1),5-8,
Geological Surveys of Finland. Norway and Sweden (1986a) Map of Quaternary Geology, Sheet
2: Glacial Geomorphology and Paleohydrography, Northern Fennoscandia, ISBN-91-7158-378-5.
Geological Surveys of Finland, Norway and Sweden (l986b) Map of Quaternary Geology, Sheet
3: Ice Flow Indicators, Northern Fennoscandia. ISBN-91-7158-379-3.
Geological Surveys of Finland. Norway and Sweden (l986c) Map of Quaternary Geology, Sheet
4: Quaternary Stratigraphy, Northern Fennoscandia, ISBN-91-7158-380-7,
Geological Surveys of Finland, Norway and Sweden (1986d) Map of Quaternary Geology, Sheet
5: Ice Flow Directions, Northern Fennoscandia, ISBN-91-7158-381-5.
Geological Surveys of Finland, Norway and Sweden (l986e) Geochemical Interpretation Map,
Northern Fennoscandia, ISBN-9J- 7158-382-3,
Geological Surveys of Finland, Norway and Sweden (l986f) Geochemical Atlas of Northern
Fennoscandia, ISBN-91- 7158-383-1.
358 GOLD METALLOGENY AND EXPLORATION
Geological Surveys of Finland. Norway and Sweden (1987) Map of Quaternary Geology, Sheet I:
Quaternary Deposits, Northern Fennoscandia. ISBN -9-7158-377 -7.
Gleeson, C.F and Sheehan, D.G. (1987) Humus and till geochemistry over the Doyon, Bousquet
and Williams gold deposits. Can. Inst. Mining Metall. Bull. 80 (898),58-66.
Guindon, D.L. and Nichol, 1. (1983) Speciation of free gold in glacial overburden. Ontario Geol.
Surv., Open-File Report 5444. 15pp.
Haldorsen, S. (1977) The petrology of tills - a study from Ringsaker, South-Eastern Norway.
Norges Geologiske Undersokelse 336, 1-36.
Haldorsen, S. (1983) Mineralogy and geochemistry of basal till and their relationship to till
forming processes. Norsk Geologisk Tidsskrift 63, 15-25.
Hall, G.E.M. and Bonham-Carter, G.F (1988) Review of methods to determine gold, platinum and
palladium in production-oriented laboratories, with application of a statistical procedure to test
for bias. J. Geochem. Explor. 30, 255-286.
Harron, G.A., Middleton, R.S., Durham, R.B. and Philipps, A. (1987) Geochemical and
geophysical gold exploration in the Timmins area, Ontario; a case history. Can. Inst. Mining
Metall. Bull. 80 (898),52-57.
Hicock, S.R. (1988) Calcareous till facies north of Lake Superior, Ontario: Implications for
Laurentide ice streaming. Geog. Phys. Quater. 42 (2), 120-135.
Hirvas, H. (1977) Glacial transport in Finnish Lapland. In Prospecting in Areas of Glaciated
Terrain - 1977, ed. Davis, G.R. Inst. Mining and Metall., London, 128-137.
Kaszycki, C.A. and DiLabio, R.N.W. (1986) Surficial geology and till geochemistry, Lynn
Lake-Leaf Rapids Region, Manitoba. In Current Research, Part B. Geo!. Surv. of Can. Paper
86-lB,245-256.
Klassen, R.A. and Thompson, FJ. (1987). Ice flow history and glacial dispersal in the Labrador
Trough. In Current Research, Part A. Geo!. Surv. of Can., Paper 87-1A, 61-71.
Klassen, R.A. and Thompson, FJ. (1989) Glacial history and drift prospecting in Labrador. In
Drift Prospecting - 1987, eds. DiLabio, R.N.W. and Coker, W.B. Geo!. Surv. of Can., Paper
89-20 21-40.
Miller, J.K. (1984) Model for clastic indicator trains in till. In Prospecting in Areas of Glaciated
Terrain - 1984, Inst. of Mining and Metallurgy, London, 69-77.
Minell, H. (1978) Glaciological interpretations of boulder trains for the purpose of prospecting in
till. Sveriges Geologiska Undersokning, Series C, Number 743. 51pp.
Morrison, S.R. and Hein, FJ. (1987) Sedimentology of the White Channel gravels, Klondike area,
Yukon Territory: Fluvial deposits of a confined Valley. Soc. of Econ. Paleontol. and Mineral.
Special Pub. 39, 205-216.
Nichol, 1. (1986) Geochemical exploration for gold deposits in areas of glaciated overburden:
Problems and new developments. In Prospecting in Areas of Glaciated Terrain - 1986, Inst. of
Mining and Metallurgy, London, 7-16.
Salonen, v.-P. (1986a) Glacial transport distance distributions of surface boulders in Finland.
Geo!. Surv. of Finland, Bull. 338. 57pp.
Salonen, V-Po (l986b) Length of boulder transport in Finland. In Prospecting in Areas of Glaciated
Terrain - 1986, Inst. of Mining and Metallurgy London, 211-215.
Salonen, V-PO (1987) Observations on boulder transport in Finland. In INQUA Till Symposium,
Finland 1985, eds. Kujansuu, R. and Saarnisto, M. Geo!. Surv. of Finland, Special Paper 3,
103-110.
Sauerbrei, J.A., Pattison, E.F. and Averill, S.A. (1987) Till sampling in the Casa-Berardi area,
Quebec: A case history in orientation and discovery. In Geochemical Exploration 1985, Part 1,
ed. Garrett, R.G. J. Geochem. Explor. 28,297-314.
Shelp, G.S. and Nichol, 1. (1987) Distribution and dispersion of gold in glacial till associated with
gold mineralization in the Canadian Shield. In Geochemical Exploration 1985, ed. Garrett,
R.G. J. Geochem. Explor. 28, 315-336.
Shilts, W.W. (1975) Principles of geochemical exploration for sulphide deposits using shallow
samples of glacial drift. Can. Inst. Mining Metall. Bull. 68 (757), 73-80.
Shilts, W.W. (1976) Glacial till and mineral exploration. In Glacial Till: an Interdisciplinary
Study, ed. Legget, R.F Roy. Soc. of Can. Special Pub. 12,205-224.
Shilts, W.W. (1981) Surficial geology of the Lac Megantic area, Quebec. Geo!. Surv. of Can.,
Memoir 397. 102pp.
GEOCHEMICAL EXPLORATION FOR GOLD IN GLACIATED TERRAIN 359
Shilts, W.W. (l982a) Glacial dispersal- principles and practical applications. Geoscience Canada
9 (1), 42- 47.
Shilts, W.W. (1982b) Quaternary evolution of the Hudson/James Bay region. Le Naturaliste
Canadien 109, 95-117.
Shilts, W.w. (1984) Till geochemistry in Finland and Canada. 1. Geochem. Explor. 21, 95-117.
Shilts, w.w. and Smith, S.L. (1986a) Stratigraphy of placer gold deposits; overburden drilling in
Chaudiere Valley, Quebec. In Current Research, Part A. Geol. Surv. of Can. Paper 86-1 A,
703-712.
Shilts, W.w. and Smith, S.L. (1986b) Stratigraphic setting of buried gold-bearing sediments,
Beauceville area, Quebec. In Current Research, Part B. Geol. Surv. of Can., Paper 86- IB,
271-278.
Shilts, w.w. and Smith, S.L. (1988) Glacial geology and overburden drilling in prospecting for
buried gold placer deposits, Southern Quebec. In Prospecting in Areas of Glaciated Terrain -
1988, ed. MacDonald, D.R. Can. Inst. of Mining and Metallurgy, 141-169.
Shilts, W.W. and Kettles, LM. (in press) Geochemical/mineralogical profiles through fresh and
weathered till. In Handbook of Glacial Indicator Tracing, eds. Kujansuu, R. and Saarnisto, M.
Balkema, Rotterdam.
Shilts, W.W., Cunningham, C.M. and Kaszycki, c.A. (1979) Keewatin Ice Sheet - Re-evaluation
of the traditional concept of the Laurentide Ice Sheet. Geology 7, 537 -541.
Sivamohan, R. and Forssberg, E. (1985) Principles of tabling. 1. Mineral Proc. 15,281-295.
Smith, S.L. and Rainbird, R.H. (1987) Soft-sediment deformation structures in overburden drill
core, Quebec. In Current Research, Part A. Geol. Surv. of Can., Paper 87 -lA, 53-60.
Smith, S.L. and Shilts, W.W. (1987) Quaternary stratigraphy of Noire River cores, Beauceville
area, Quebec. In Current Research, Part A. Geol. Surv. of Can., Paper 87-IA, 159-164.
Stea, R.R., Turner, R.G., Finck, P.W. and Graves, R.M. (1988) Glacial dispersal in Nova Scotia: a
zonal concept. In Prospecting in Areas of Glaciated Terrain -1988, ed. MacDonald, D.R. Can.
Inst. of Mining and Metallurgy, 57-79.
Stewart, R.A. (1986) Routine heavy mineral analysis using a concentrating table. 1. of Sediment.
Petrol. 56, 555-556.
Strobel, M.L. and Faure, G. (1987) Transport of indicator clasts by ice sheets and transport
half-distance. A contribution to Prospecting for Ore Deposits, 1. Geology 95, 687-697.
Taipale, K., Nevalainen, R. and Saarnisto, M. (1986) Silicate analyses and normative compo-
sitions of the fine fraction of till: examples from eastern Finland. 1. Sediment. Petrol. 56,
370-378.
Thompson, LS. (1979) Till prospecting for sulphide ores in the Abitibi Clay Belt of Ontario. Can.
Inst. Mining Metall. Bull. 72 (807), 65-72.
Thorliefson, L.H. and Kristjansson, F.J. (1988) Stratigraphy and visible gold content of till in the
Beardmore-Geraldton area, Northern Ontario. In Current Research, Part C. Geol. Surv. of
Can., Paper 88-IC, 217-221.
Veillette, J.J. (1986) Former southwesterly ice flows in the Abitibi-Timiskaming region:
implications for the configuration of the Late Wisconsinan ice sheet. Can. 1. Earth Sci. 23,
1724-1741.
Veillette, J.J. (1989) Ice movements, till sheets and glacial transport in Abitibi-Timiskaming,
Quebec and Ontario. In Drift Prospecting - 1987, eds. DiLabio, R.N.W. and Coker, W.B. Geol.
Surv. of Can., Paper 89-20139-154.
Gold Metallogeny
and Exploration
Edited by
R.P. FOSTER
Senior Lecturer in Economic Geology
Department of Geology
University of Southampton