REPORT ON THE c2i / 9

44
e REPORT ON THE c2i / 9.E INDUCED POLARIZATION AND RESISTIVITY SURVEY ON THE RICK CLAIM GROUP HOLBERG, BRITISH COLUMBIA FOR HOLBERG MINES LTD. ‘.T: c: ‘7 _L , ,A< ,v;lz-- BY P. G. HALLOF, PH.D. AND D. B. SUTHERLAND. M.A. NAME AND LOCATION OF PROPERTY: HOLBERG: 2 MILES WNW OF HOLBERG, 50”, 128” NE DATE STARTED - FEBRUARY 1, 1963 DATE COMPLETED - MARCH 10, 1963

Transcript of REPORT ON THE c2i / 9

Page 1: REPORT ON THE c2i / 9

e

REPORT ON THE c2i / 9.E INDUCED POLARIZATION

AND RESISTIVITY SURVEY

ON THE RICK CLAIM GROUP

HOLBERG, BRITISH COLUMBIA

FOR

HOLBERG MINES LTD. ‘.T: c: ‘7 _L

, ,A< ,v;lz--

BY

P. G. HALLOF, PH.D.

AND

D. B. SUTHERLAND. M.A.

NAME AND LOCATION OF PROPERTY:

HOLBERG: 2 MILES WNW OF HOLBERG, 50”, 128” NE

DATE STARTED - FEBRUARY 1, 1963

DATE COMPLETED - MARCH 10, 1963

Page 2: REPORT ON THE c2i / 9

TABLE OF CONTENTS

Part A: Note* on theory and field procedure

Part B: Report

1. Introduction

2. Prerantation of Result8

3. Mecumion of Results

Line 25tOOE Line ZO+OOE ~iria 15tOOE Line 1 O+,OOE Line 5t.OOE Line otoo

14 pageE PaJp

1

1

3

4 4 4 5 5 5 5 Line 5.1OOW

Line 1o+oow Line 15’OOW Line 2OtOOW

Department of ‘Z 6

Line 25tOOW Line 30tOOW

Mines and Petroleum Res,ourcos i

Line 35.tOOW ASSESSMENT ~~~~~~~ 7 Line 9E2 7 Line A NO. ..,..... . $%f:..z A*.;,P,.~ ..~ 7 Line B 7 Line C 8

4. Conclusions and Recommendations 8

5. Alrsessment Detaila 11

6. Certificate - Philip G. Hallof 12

7. Certificate - Don B. Sutherland 13

8. Summary of Cost 14

Part C: Illwtrations 22 pieces

Plan Map (in pocket) Dwg. Misc. 4015 Y9’-/

I.P. Data Plots hgs. 1.P. 2040-l to -21

Page 3: REPORT ON THE c2i / 9

McPHAR GEOPHYSICS LIMITED

NOTES ON THE THEORY OF INDUCED POLARIZATION

AND THE METHOD OF FIELD OPERATION

‘I

Induced Polarization as a geophysical,measurement refers

to the blocking action or polarization of metallic or electronic

conductors in a medium of ionic solution conduction.

This electro-chemical phenomenon occurs wherever

electrical current is passed through an area which contains metallic

minerals such as base metal sulphides. Normally, when current is

passed through the ground, as in resistivity measurements, all of the

conduction takes place through ions present in the water content of the

rock, or soil, i.e. by ionic conduction. This is because almost all

minerals have a much higher specific resistivity than ground water.

The group of minerals commonly described as “metallic”, however,

have specific resistivities much lower than ground waters. The

induced polarization effect takes place at those interfaces where the

mode of conduction changes from ionic in the solutions filling the

interstices of the rock to electronic in the metallic minerals present

in the rock,

1

The blocking action or induced polarization mentioned

above, which depends upon the chemical energies necessary to allow

the ions to give up or receive electrons from the metallic surface,

increases with the time that a d. c. current is allowed to flow through

Page 4: REPORT ON THE c2i / 9

- 2 -

1

I

the rock; i. e. as ions pile up against the metallic interface the

resistance to current flow increases. Eventually, there is enough

polarization in the form of excess ions at the interfaces to effectively

stop all current flow through the metallic particle. This polarization

takes place at each of the infinite number of solution-metal interfaces

in a mineralized rock.

When the d. c. voltage used to create this d. c. current

flow is cut off, the Coulomb forces between the charged ions forming

the polarization cause them to return to their normal position. This

movement of charge creates a small current flow which can be

measured on the surface of the ground as a decaying potential difference.

From an alternate viewpoint it can be seen that if the

direction of the current through the system is reversed repeatedly

before the polarization occurs, the effective resistivity of the system

as a whole will change as the frequency of the switching is changed.

This is a consequence of the fact that the amount of current flowing

through each metallic interface depends upon the length of time that

current has been passing through it in one direction.

The values of the “metal factor” or “M. F. ” are a measure

of the amount of polarization present in the rock mass being surveyed.

This parameter has been found to be very successful in mapping areas

of sulphide mineralization, even those in which all other geophysical

methods have been unsuccessful. The induced polarization measurement

is more sensitive to sulphide content than other electrical measurements

Page 5: REPORT ON THE c2i / 9

- 3 -

I

I

because it is much more dependent upon the sulphide content. As the

sulphide content of a rock is increased, the “metal factor” of the rock

increases much more rapidly than the resistivity decreases.

Because of this increased sensitivity, it is possible to

locate and outline zones of less than 10% sulphides that can’t be

located by E. M. Methods. The method has been successful in locating

the disseminated “porphyry copper ” type mineralization in the South-

western United States.

Measurements and experiments also indicate that it should

be possible to locate most massive sulphide bodies at a greater depth

with induced polarization than with E. M.

Since there is no 1. P. effect from any conductor unless it

is metallic, the method is useful in checking E. M. anomalies that are

suspected of being due to water filled shear zones or other ionic

conductors. There is also no effect from conductive overburden, which

frequently confuses E. M. results. It would appear from scale model

experiments and calculations that the apparent metal factors measured

over a mineralized zone are larger if the material overlying the zone

is of low resistivity.

Apropos of this, it should be stated that the induced

polarisation measurements indicate the total amount of metallic

constituents in the rock. Thus all of the metallic minerals in the rock,,

such as pyrite, as well as the ore minerals chalcopyrite, chalcocite,

galena, etc. are responsible for the induced polarizatioreffect. Some

Page 6: REPORT ON THE c2i / 9

- 4 -

” oxides such as magnetite, pyrolusite, chromite, and sdme forms of

hematite also conduct by electrons and are metallic. All of the metallic

minerals in the rock will contribute to the induced polarization effect

measured on the surface.

In the field procedure, measurements on the surface are

made in a way that allows the effects of lateral changes in the properties

of the ground to be separated from the effects of vertical changes in the

properties. Current is applied to the ground at two points a distance

(W apart. The potentials are measured at two other points (X) feet

apart, in line with the current electrodes. The distance between the

nearest current and potential electrodes is an integer number (N) times

the basic distance (X).

The measurements are made along a surveyed line, with

a constant distance (NX) between the nearest current and potential

electrodes. In most surveys, several traverses are made with various

values of (N); i. e. (N) = 1, 2, 3, 4, etc. The kind of survey required

(detailed or reconnaissance) decides the number of values of (N) used.

In plotting the results, the values of the apparent resistivity

and the apparent metal factor measured for each set of electrode

positions are plotted at the intersection of grid lines, one from the

center point of the current electrodes and the other from the center

point of the potential electrodes. The resistivity values are plotted

above the line and the metal factor values below. The lateral displace-

ment of a given value is determined by the location along ,the survey

Page 7: REPORT ON THE c2i / 9

- 5 -

line of the center point between the current and potential electrodes.

The distance of the value from the line is determined by the distance

(NX) between the current and potential electrodes when the measure-

ment was made.

The separation between sender and receiver electrodes is

only one factor which determines the depth to which the ground is being

sampled in any particular measurement. These plots then, when

contoured, are not section maps of the electrical properties of the

ground under the survey line. The interpretation of the results from

any given survey must be carried out using the combined experience

gained from field, model and theoretical investigations. The position of

the electrodes when anomalous values are measured must be used in the

interpretation.

In the field procedure, the interval over which the potential

differences are measured is the same as the interval over which the

electrodes are moved after a series of potential readings has been made.

One of the advantages of the induced polarization method is that the

same equipment can be used for both detailed and reconnaissance surveys

merely by changing the distance (X) over which the electrodes are moved

each time. In the past, intervals have been used ranging from 100 feet

to 1000 feet for (X). In each case, the decision as to the distance (X)

and the values of (N) is largely determined by the expected size of the

mineral deposit being sought, the size of the expected anomaly and the

speed with which it is desired to progress.

Page 8: REPORT ON THE c2i / 9

- 6 -

e The diagram in Figure 1 below demonstrates the method

used in plotting the results. Each value of the apparent resistivity

and the apparent “Metal factor” is plotted and identified by the position

of the four electrodes when the measurement was made. It can be seen

that the values measured for the larger values of (n) are plotted farther

from the line indicating that the thickness of the layer of the earth that

is being tested is greater than for the smaller values of (n); i. e. the

depth of the measurement is increased.

METHOD USED IN PLOTTING DIPOLE-DIPOLE

INDUCED POLARIZATION AND RESISTIVITY RESULTS

I 2 u 4 5 6 7 8 9

Stations on line x = Electrode spread length n: Electrode seporotion

n-2 MF M.S M.S MS MS 12-4.5 2.3-5.6 3.4-67 4,5-7x! 5$-8.9

111 n-3 MS M.S M.S M~S 1.25.6 2.3-67 3,4.7,8 4.5-8.9

dpp..*nl Melo, Fador

n-4- ME M.S M~F 1.2-67 2.3-7.6 3.4 E,9 .!yiJ

Page 9: REPORT ON THE c2i / 9

McPHAR GEOPHYSICS LIMITED

REPORTONTHE

INDUCED POLARIZATION

AND RESISTIVITY SURVEY

ON THE RICK CLAIM GROUP

HOLBERG, BRITISH COLUMBIA

FOR

HOLBERG MINES LTD.

1

1. INTRODUCTION

At the request of Chapman. Wood and Griswold Limited,

Conoultants for the Company, an induced polarisation eurvey he8

been carried out near H~olberg, B.C. on behalf of Ho&erg Mine8 Ltd.

The survey covered part of a grid on the Rick Claim Group; the work

was planned to locate any u&now-n zonee of mineralisation in the area

of l tveral narrow, high-grade copper vein*. It wan hoped that a

larger volume of mineralisation might be preeent at depth.

The survey was carried out uring 200’ electrode intervale.

With this separation only w44k anomalier would be expected from the

indhMuaLnarrow v4in4;howev4r. a Isrgsr volume 0fmetaIlic min4rrltaa-

tion would be detected to depths of 200-300 fed.

1

2. PRESENTATION OF RESULTS

The induced polarisation and reristivity rewlts are ehown

on the following enclosed data plots. The remlta are plotted in the

manner dencribed in the notee preceding thin report.

Page 10: REPORT ON THE c2i / 9

-2-

Line 25+00 Ea6t

Line 2OlOO E66t

Line lSiO0 Eant

Line 10*00 Ea6t

Line 5+00 E66t

LinQ O-100

Line 5400 Wed

Line lO+ 00 We6t

Line 15too west

Line 20100 W66t

Line 25too West

Lina 30100 Weat

Litlo 35r,oo Wsrt

Line BE2

Line A

Line B

Line c

200’ Spread8

200’ Spread8

200’ Spra8d6

200’ SpP66d6

200’ Spr68d6

200’ Spreads

200’ Spr*ad6

200’ Spr6rd6

200’ gproad6

200’ spread8

200’ Spread6

200’ Spread6

100’ Spread6

50’ Spreads

200' Spr66d6

100’ Spreada

50’ Spreada

200’ Sprsadr

200’ spread6

200’ Spread6

200’ gprard6

Dwg. I.P. 2040-I

Dwg. 1-P. tQ40-2

Dwg. I.P. 2040-3

Dwg. I.P. 2040-4

Dwg. 1-P. 2040-5

Jhg. I.P. 2040-b

Dwg. I.P. 2040-7

Dng. I.P. 2040-g

Dwg. I.P. 2040-9

Dwg. I.P. 2040-10

Dwg. I.F. 2040-11

Dwg. I.P. 2040-12

Drsg. I.P. 2040-13

Dwg. I.F. 2040-14

Dwg. I.P. 2040-15

Dwg. I.F. 2040-16

Lhvg. I.P. 2040-17

Dwg. I.P. 2040-18

Dap. I.P. 2040-19

Dwg. I.P. 2040-20

Dwg. I.P. 2040-21

AI60 sntilosed with tbi6 report i6 I.hvg. Mi6c. 4015, a pIan

map of tbc grid at 6 6caIe of 1” = 200’. The liner covered by the rurvay

Page 11: REPORT ON THE c2i / 9

- 3 -

are shown on thie map. Line 9E2, Line A. Line B. and Line C were

eurvoyed to cover other areas of interest, and their poeitione are not

ehown on thie mop. The definite end poeeibla induced polarisation

anomeliee are indicated by eolid and broken bare reepectiveiy on tkie

plan map a8 well a8 the data plate. Theee bare rapreeent the ourSace

projection of the anomaloue eonee as interpreted from the location of

the transmitter end recsivar electrodee when the anomalous value8

were measured.

Since the Induced palarization meaeuroment ie eeeentially

an averaging proceee, ae are aA1 potential methode, it ie frequently

difficult to exactly pinpoint the l ouree of an anomaly. Certainly, no

anomaly can be located with more accuracy than the eproad length; i. a.

when ueing 200’ epreade the poeition of a narrow sulphide body can only

be determined to lie between two etatioae 200’ apart. In order to locate

eourcee at rome depth, larger l preade muet be ueed, with a correepond-

ing increase in the uncertaintise of location. Therefore, while the centar

of the indicated anomaly probably correeponde fairly well with source,

the length of the indtcatad anomaly along the line ehould not be teken to

repreeent the exact edgee of the anomaloue material.

3. DISCUSSION OF RESULTS-

The known minerelisation on the Rick Claim Croup coneiete

of eeveral emall aonee of high-grade copper mineralisation. Three

some are shown on Dwg. Misc. 4015. The liner l hewn on Dwg. Misc.

Page 12: REPORT ON THE c2i / 9

-4-

1 4015 were rurveyed in an attempt to locate other mineralised xonee that

might be larger md therefore of more commercial interoet.

To the we&. the northern parta of the Unoe extoaded into

the low, ewampy ground north of the fault. The surface rosietivitiea

were low in this area and the voltages to be measured in the I.P. survey

were much reduced in magnitude. These lines ore also the cloweet to

the Nolberg RCAF Base. and the electrical noiee created by the inrtella-

tion at the Baee warn large. Due to the low voltagee on there linem.

meaeuremente were difficult, or impoeaiblt; on many of the liner only

the mearurameate for n=l and a=2 could be made. Others of the meaaure-

merits are doubtful due to the aoiea.

I Line 2si-OOE

This is the taeternmort line surveyed. Two moderate magni-

tude, but definite, I.P. anomaliae art indicatsd by thaws rewlte. One

anomaly ia centsred at 16N at the fault, but the other ie at 22N ta 24N,

where swamp covers the surface.

Line 20+00E

Only a l hort part of We line was eorveyed. Although the

background I.P. effecta are somewhat larger then ueual, they are not

really anomalous .

Line 15tOOE

Tbie line is aleo abort. The I. P. beokground 10 high, and

Page 13: REPORT ON THE c2i / 9

- 5 -

there ie a poorly defined anomaly at 16N.

There am no d4fiaitely ammalous I. P. 4ffeetm on this

line.

Line 5?OOE

There ir a shallav, narrcm anomaly that should be checkad

with shorter spreads between 1 ON and 12N.

Thin line passee near the known copper ahowbga. Thera

arc mvaral weak 1.F. anomalie* just north of the baeoline, over tha

showing@. The44 affects would probably increase in magnitude if the

mearurementr w4fe repeated using shorter elactrede intorvale. There

ir also a deflnlte increare in the effects at depth at 6N to SN.

Line 5toow

Ths reeult8 on thie line indicate an anomaly that 18 @hallow

at 4N to 6N. This is somewhet north of the knewa mineralisation. Thsro

ir some indication that the son4 increosem in width at depth.

Lb%4 1o+oow

The shallow, narrow anomaly at 0.4 00 to 2N on thim line cot-

rolatee with known miaerelieation. There lr alro the anomaly at depth

at 4N to 6N, that is at the fault, and correlates with the anomalies to the

ea6t.

Page 14: REPORT ON THE c2i / 9

-6-

Line 15+OOW

The anomaly on tbie liae ie very similar to that on Liae 5+OOW.

The enomely correlates with the fault, cad the effects increase slightly

with depth.

Liue ZO+OOW

There is no defiaite eaomely at the fault on tbie line. H~w4ver.

there ir a feult at depth at 1tN to IQN in the area coverod by the l wemp.

The reeietivitiae at depth in thb area arc) low, md th4 1-P. effect8 are

moderetely high.

Line 25+00W

0

There ie a narrow aaomaiy at 0 to ZS on thir line that eorrefrtcs

apprcudmately with the fault. However, it would have to be detailed with

shorter electrode intervale before it caa be fully evaluated.

Ta the north, the rerlrtivities are low rr oa Lhs 2O+OOW, but

the I. P. effects are not as large. Ssv4ral of tb.e measurements could not

be made because of the noise, but th4 I.P. values seem to indicate a high

background rather tbaa an eaomaly.

The reeulte uoiag 200’ electrode intervals oa tbie liae suggeet

a broad aoae of slightly anmtaaloue 1.X’. effocto north of 6N. H4w4ver,

the meaeuromeats with 100’ interval. and 50’ iatsrvele indicate two narrow

anomali48 within this moae .

Page 15: REPORT ON THE c2i / 9

-7 -

‘I

The narrow anomaly at 4N to SN IS definite with 100’

spreads, but only the top iii evident with 50’ epreode. Therefore the

top of the mouree ie probably 50’ or more deep.

The anomaly at 12N to 13N appeare to be eomewhat broader,

and the 50’ spread raoultr do not indicate the enomaly 81 clearly.

Line 35+OOW

The 200’ l prerd remlta on thio line show an anomaly at

depth at 2N, and a elngle high value at depth at SN to 10N. The 100’

spread rem&s ehow two definite anomalieai they are cantered at 9N to

10N end 13N to 14N. Both anomrlfefi suggeet rome depth to the top of

the source .

When the rerultr were repeated ueing SO’ electrode intervals,

the anomaliee can &ill be eeen. but the pattern0 are not ae definite, eug-

getting that there ia mome depth to the eource of moat of the aaomaly.

Line 9EZ

There arc no I.P. effecte on this line.

Line A

There are no definite anomaliee on thie line.

Line B

Several weak anomeliee are indicated on thie line, but further

meaouremento would be required to completely evaluate their importance.

Page 16: REPORT ON THE c2i / 9

- 8 -

I

Line c

The anomaly at depth at 0 to 4s on this line correlates with

a hnilar anody OA Line B.

4. CONCLUSIONS AND RECOMMENDATIONS

The induced polarieetion meaeuremento on the recoanaiseancs

grid on the Rick Claim Group gave only week anomalies over tbe known

mineralizatloa. The known veine are quite aarmw, and the electrode

l eparatione ueed are much greater then the width of mineralisation.

However, other aaomalie8 were lo-ted by the eurvey and \

theee & be due to larger volumse of mineralisation. The follotig

drill programme, consisting of 3,000 to 4,000 feet of drilling hae been

planned to teet the meet important of theee anomalies.

D. D. H. I1 - Line 5W, 715ON, drill -45’ couth for 550 feet

Tbie anomaly lice immediately north of the known mineralised

wine. The anomaly ie not rtrong, but it ie definite and the hole hae

been #potted to te8t the upper portion of the eource. However, the 1-P.

effect* do increase with depth and if the source in of economic interest.

a hole to teat the source at greater depth ehould be drilied.

D.D.H. 112 - Line lOW, 8tOON. drill -45O south for 600 feet

The eouree of thie anomaly doee AOt come ae close to the

surface am the source on Line 5W end consequently a deeper hole has

bean #potted.

Page 17: REPORT ON THE c2i / 9

- 9 -

D.D.H. 13 - Line 35W, 10+50N, drill -45* 6outh for 220 feet

Thim strong, relatively narrow anomaly ir well out in

the volley and therefore there ir no geologic information,

D. D.H. #4 - Line 3OW, S+5ON, drill -45’ 6outh for 250 feet

This anomaly ia similar but ir not aa strong. Tba

r86utt6 with 50 foot spread6 BUg@Bt that the 6ource in romewhat

deeper and therefore a 1onSer hole i6 required.

D. D.H. #5 - Line 35W, 14tOON. drill -45’ 6outh for 250 feet

Thi6 hole ha6 been 6potted to determine the eau6e of the

high background I.P. effects mea6ured north of tke narrow sources

on Lines SOW and 35W.

D. D. H. #6 - Line 25E, 1&50N, drill -45. south for 500 feet

This anomaly ir er6t of the known mlnerali6ation. The

az10mabu6 pattern is not definite, but the vahe8 are of moderate

magnitude. Therefore a 6ingle long hole ha6 heen Bpotbd to t86t the

6ource of the anomaly at depth.

D, D.H. 17 - Line C, 1~50s. drill vertical for 400 feet

Thl6 weak anomaly occurs north of the knwns mineraba-

tiou in an area where the ba6ie rock6 are believed to underlie an up-

certain thiekne66 of lime6tone. TbS 6OUfC8 Of thfr Wd UtOXll~y i6

indicated to be at depth and the vertical hole ha6 been Ipotted to determine

Page 18: REPORT ON THE c2i / 9

- 10 *

‘I

I

its source and pommibly the thickneaa of the limeatone.

The drill holes dewribed above have been apotted to teat

the mont important anomalier. If any ancourrgemcant is obtained then

the following drill holea should alro be considered.

D. D.H. #S - Line ISE, 18t5ON. drill -45’ couth for 500 feet

Thim anomaly is similar to the anomaly on Line 25E and

only the eaetern edge of the anomaly is indicated.

D.D.H. 19 - Line lSW, 7+5ON, drill -45’ @out& for 450 feet --

This anomaly ie very similar to that on Line 5W but is of

lower magnitude.

McPHAR GEOFHYSICS LIMITED

Philip G. Hallof, Geophyaiciat .

D. B. Sutherland, Geophysiciet.

Dated: April 4, 1963.

Page 19: REPORT ON THE c2i / 9

- 11 -

ASSESSMENT DETAILS

PROPERTY: Rick Claim Group

SPONSOR: Holbcrg Minea Ltd.

LOCATION: Holberg. Vancouver L~lawl

TYPE OF SURVEY: Induced Polarisation

OPERATING MAN DAYS: 42.50

EQUIVALENT 8 HR. MAN DAYS: 73.75

CONSULTING MAN DAYS: 4.0

DRAUGHTING MAN DAYS: 5.0

TOTAL MAN DAYS: 82.75

CONSULTANTS:

MINING DIVISION: Nanaimo

PROVINCE: Britieh Columbia

DATE STARTED: February 1, 1963

DATE FINISHED: March 10. 1963

NUMBER OF STATIONS OCCUPIED: 250

NUMBER OF READINGS TAKEN: 1,752

MILES OF LINE SURVEYED: 7.24

P. G. Hallof, 5 Minorca Plree, Don Mill*, Ontario D. B. Sutherland, 412 Eglinton Avenue, Eart. Toronto 12, Ontario

FIELD TECHNICIANS: P. Bauden, c/o Forest Raaprr’a School. Dorset, Ontrrio Five helwrr supplied by client

DRAUGHTSMEN: F. R. Peer, 38 Torreno Avenue. Toronto 6, Ontario R. MacKsnsis, 55 Shannon Drive, Scarborough, Ontario

McPHAR GEOI?WYSICS LIMITED

D. B. Sutbwland, Goophysicimt.

Dated: April 4, 1963.

Page 20: REPORT ON THE c2i / 9

- 12 -

CERTIFICATE

I. Philip George Hallof, of the City of Toronto, Pwvince Of

Ontario, do hereby eartify that:

1. I am a geophyaiciat residing at 5 Minorca Place, Don Mills

(Toronto). Ontario.

2. I am a graduate of the M,assachusetts Institute of Technology

with a B.S. Degree (19S2) in Geology and Gaophyrics. and a Ph.D. Degree

(1957) in Geophyoico.

3. I am a member of the Society of Exploration Gsophysicimtr

and the European Aooociation of Exploration Gaophyricists.

4. I bave beon pr8ctiaing my prof*snlon for ten years.

9. I have no direct or tndiract inters& nor do I expect to receive

any iat4reet. direct or indir4ct. in the property or aocuritiem of Holbrg

Mines Ltd.

6. The statements made in this report are bared On a rtudy Of

publirhed literature and unpublished private report6 and geopbysieti data.

Dated at Toronto

This 4th day Of April 1963 Philip G. Hallof, Ph. D.

Page 21: REPORT ON THE c2i / 9

- 13 -

CERTIFXCATE

I, Don Benjamin Sutherland of the City of Tom&a, Province

of Ontario, do hereby certify that :

1. I am a geophymicirt residing at 412 Eglinton Avenue, Eset,

Toronto 12, Ontario.

2. I am a graduate of the University of Toronto in Phyaice

and Geology with the degree of Bachelor of Arte (1954)l end 8 graduate

of the University af Toronto in Phyrice with the degree of Master of

Arta (1985).

3, I am a member of the Society of Exploration Geophysicists

and a member of the European Association of Exploration Geophyeiciete.

4. I have be- practistng my profe*doa for over swen year@.

5. I have no direct or indirect interelt, nor do I exprot to

receive any iatereet dimctly or indirectly, in the property or eecuritiee

of Holberg Mines Ltd.

6. The statements mad8 in thin report are based on a etudy of

published geological literature and uapubliahod private report&

Dated at Toronto

Thie 4th day of April 1963

Page 22: REPORT ON THE c2i / 9

- 14 -

SUMtvfARY OF COST

27-l/2 dsyr 0perating @ $165.OO/day lo-112 day8 Trawl. Bad Weather

b Standby @ $ 60. OOlday

Airfme Airf might Telephone and Telegraph Meals md Accommodation Supplies

$218.00 205.19

57.80 74.50 74.02

$4‘537.50

630.00

629.51

$2,

McPHAR GEOPHYSICS LIMITED

D. B. Satherland. Gaophykist.

Dated: April 4, 1963.

Page 23: REPORT ON THE c2i / 9
Page 24: REPORT ON THE c2i / 9
Page 25: REPORT ON THE c2i / 9
Page 26: REPORT ON THE c2i / 9
Page 27: REPORT ON THE c2i / 9
Page 28: REPORT ON THE c2i / 9
Page 29: REPORT ON THE c2i / 9
Page 30: REPORT ON THE c2i / 9
Page 31: REPORT ON THE c2i / 9
Page 32: REPORT ON THE c2i / 9
Page 33: REPORT ON THE c2i / 9
Page 34: REPORT ON THE c2i / 9
Page 35: REPORT ON THE c2i / 9
Page 36: REPORT ON THE c2i / 9
Page 37: REPORT ON THE c2i / 9
Page 38: REPORT ON THE c2i / 9
Page 39: REPORT ON THE c2i / 9
Page 40: REPORT ON THE c2i / 9
Page 41: REPORT ON THE c2i / 9
Page 42: REPORT ON THE c2i / 9
Page 43: REPORT ON THE c2i / 9
Page 44: REPORT ON THE c2i / 9