HATCH ™

3PN01

U PDATE ON RESOURCES
GALORE CREEK PROJECT, BRITISH COLUMBIA

prepared for:

N OVAGOLD RESOURCES INC. AND SPECTRUMGOLD INC.
Suite 3454, Four Bentall Centre
1055 Dunsmuir Street
PO Box 49215
Vancouver, BC V7X 1K8

3PN01

U PDATE ON R ESOURCES
G ALORE C REEK P ROJECT , B RITISH C OLUMBIA

1 S UMMARY

Associated Mining Consultants Ltd. (AMCL) and Hatch were retained by NovaGold Resources Inc. (NovaGold) and SpectrumGold Inc. (Spectrum) to update mineral resource estimates for the Galore Creek deposit and complete a Technical Report summarizing the findings of the update to meet the requirements of National Instrument 43-101 (the Instrument) and Form 43-101F1. The mineral resource study was a collaborative effort by NovaGold/Spectrum, Hatch and AMCL with Mr. Peter A. Lacroix, P.Eng. of AMCL directing the work and acting as the Independent Qualified Person as defined by the Instrument. Mr. Lacroix conducted a site examination at the property during the week of October 14, 2003.

NovaGold Resources Inc., through its 56% owned subsidiary, SpectrumGold Inc., is engaged in the exploration of the Galore Creek Project in British Columbia, Canada. NovaGold has recently made an offer to acquire the remaining the shares of SpectrumGold and, subject to the satisfactory conclusion of this transaction, NovaGold will own 100% of SpectrumGold Inc and its interests in the Galore Creek Project. The Galore Creek deposit is located in a remote mountainous area approximately 1000 kilometers northwest of Vancouver, 70 kilometers northwest of the Eskay Creek Mine. The project is located 200 kilometers north of the tidewater port of Stewart, British Columbia 100 kilometers northeast of the tidewater port of Wrangell, Alaska. Current access is by helicopter, although fixed-wing aircraft as well as a combination of barge and road access have been used in the past to transport personnel and equipment to the site. Typical of northwest coastal areas of British Columbia, the area is characterized by cool summers and cold humid winters with relatively heavy snowfall.

The property covers approximately 4,700 hectares and is comprised of 253 claims and 39 fractions. The claims are wholly owned by Stikine Copper Ltd. controlled by QIT-Fer et Titane Inc. (55%) and Hudson Bay Mining and Smelting Co, Limited (45%). In August 2003, Spectrum entered into an option agreement to acquire a 100% interest in the project by completing a pre-feasibility study on the project and making payments to the parties totaling US$20.3 million within a period of 8 years. There will be no retained interests, royalties or back-in rights on the project.

Mineralization was discovered in Galore Creek in the mid-1950's and during the subsequent 20 years nearly 75,000 metres of drilling and over 800 metres of tunneling was completed under the direction of Kennco Explorations and Hudson Bay. Following an economic study prepared in 1974, the project was relatively inactive until the late

1980's when both Hudson Bay and Kennecott revisited the project. This last period of activity added another 20,000 metres of drilling and culminated with a resource estimate and economic studies from 1992 through 1994. The project has since been inactive until Spectrum resumed drilling in September 2003.

Galore Creek is a large alkalic intrusive-related gold-silver-copper system. The property is underlain by volcanic and sedimentary rocks that have been intruded by an alkali syenite complex. Mineralization is most common in volcanic rocks and occurs as replacement and disseminated chalcopyrite with locally abundant bornite.

At least 12 mineralized zones have been identified on the property. The focus of historic work has been the Central Zone, which is an area of sulphide replacement in a volcanic rock over an area 1700 metres long, 200 to 500 metres wide and extending to a known depth of 450 metres. Mineralization is exposed at surface in the southern portion and is overlain by up to 75 metres of overburden in the north. Grades in the Central Zone often exceed 1% copper with gold values ranging from 0.3 to in excess of 1 gram per tonne. The other focus of historic activity has been a near-surface pod of mineralization about 400 metres by 150 metres in area that is approximately 600 metres to the southwest of the Central Zone. Mineralization in this area, known as the Southwest Zone, is generally higher in gold content and is in part hosted in a breccia body which has been interpreted to be a diatreme breccia More recently these zones have been referred to as the Main and Southwest Deposits however to avoid confusion this report will retain the original nomenclature.

Numerous other mineralized zones have also received attention in the past, including a series of flat-lying zones located about 2 kilometers northwest of the Central Zone. These mineralized zones, which range 50 to 150 metres in width and 460 metres higher in elevation than the Central Zone are collectively known as the Junction Zones.

Historically, the Central and Southwest Zones have been viewed primarily as potential open-pit targets while the Junction Zones were envisioned as moderate-sized underground bulk mining propositions. More recent studies have focussed on exploiting resources in the Central and Southwest Zones by surface methods, with production rates ranging from 30,000 to 60,000 tonnes per day. Commensurate with this emphasis, a number of historic mineral resource estimates exist, the most recent of which was based on a model completed for the Central and Southwest Zones in 1992 by Mine Reserve Associates, Inc. (MRA) on the behalf of Kennecott. That estimate, re-classified in 2002 as indicated and inferred by Kennecott, totaled 243.2 million tonnes grading 0.75% copper and 0.45 grams per tonne gold and 70.6 million tonnes grading 0.59% copper and 0.63 grams per tonne gold, respectively. The reporting cutoff was based on an in-situ metal value of $US10 per tonne using US$0.80 per pound copper and US$320 per troy ounce of gold. The estimate is considered "Historic" under the definitions of National Instrument 43-101.

Updated estimates for mineral resources contained within the Central and Southwest Zones were completed by Hatch/AMCL in collaboration with Spectrum in April 2004.

These estimates, which incorporate information from the most recent round of drilling completed in October 2003 as well as having the benefit of a revised interpretation, are summarized in Table 1. The estimates are based on a 3-dimensional computer block model with grades interpolated into individual 25 metre by 25 metre by 12 metre high blocks within a geological framework utilizing ordinary kriging. Specific gravities used for tonnage estimates range from 2.0 for glacial till to 2.49 -2.75 for volcanics and intrusives. The resource estimate is reported at a block cutoff of 0.5% equivalent copper (Eq Cu).

The resource is reported using copper equivalent cutoff values that have been calculated by estimating the net smelter return for gold and silver and allocating the revenue as if it were generated from copper alone. The selected cutoff of 0.5% equivalent copper (Cu Eq) represents a net value of about US$10 per tonne based on the assumed prices, metallurgy and smelter terms. Copper, gold and silver prices used in calculations were US$0.90 per pound, US$375 and US$5.50 per ounce respectively. Significant quantities of higher grade resources exist within this inventory however in some cases this may not represent material which can be exploited without inclusion of some lower grade resources or internal waste. Further detail with respect to the distribution of grades and the calculation of equivalent values can be found in Section 17.10.

It is believed that the stated resource classifications conform well to the CIM Standards on Mineral Resources and Reserves. The quality and density of sample information, knowledge regarding the continuity of mineralization within various geological units as well as any limits that the structural framework has imposed on the resource have all been considered carefully in setting the guidelines. The Central Zone is drilled from surface down to about 450 metres depth on an approximate spacing of 50 to 100 metres north-south and 50 metres east-west. Below this, as well as on the extremities, the drill density is much less. This change in drilling density coincides roughly with the change in classification from indicated to inferred. While the drill spacing in Southwest Zone is substantially less, the density is still sufficient near surface and at the core of the deposit to allow approximately 60% of the tonnage to be classified as indicated. The geological and structural framework has served to limit the extent of each category further.

None of the mineral resource is classified as measured. It is believed that in addition to reducing the overall drill spacing to that recommended in Section 17.8, more extensive sampling of precious metals and confirmation through further testing will be required to warrant upgrading any portion of the resource to measured. There has been no mining experience with the deposit and it would be premature to assume that the level of confidence associated with "measured" can be applied to these mineral resources without further work. While necessary to formulate detailed mine plans for production, it is not recommended that any additional drilling be completed within the area of the indicated mineral resource to upgrade it to measured. A limited surface program designed to upgrade inferred resources along the peripheries of the existing zones as well as determine their vertical and strike extent could be considered as an addendum to the recently completed program.

2 I NTRODUCTION AND T ERMS OF R EFERENCE

NovaGold Resources Inc. (NovaGold), through its 56% owned subsidiary, SpectrumGold Inc. (Spectrum), is engaged in the exploration and advancement of the Galore Creek Project in British Columbia, Canada. Associated Mining Consultants Ltd. (AMCL) and Hatch were retained by NovaGold and Spectrum to update mineral resource estimates for the Galore Creek deposit and complete a Technical Report summarizing the findings of the update to meet the requirements of National Instrument 43-101 (the Instrument) and Form 43-101F1.

The purpose of the study is to provide a current foundation for further work on the property, including a preliminary assessment of the project's economic potential and development alternatives as well as assist in the refinement of its exploration and delineation efforts and technical programs for the 2004 field season. The Technical Report will also serve to provide regulators and investors with a formal and up-to-date reference for the property that incorporates the results of the most recent work program completed by Spectrum in October 2003.

A site visit and detailed examination of the property was conducted by Mr. Lacroix on October 15 and 16, 2003 with Mr. Greg Johnson, Vice President Corporate Communications for NovaGold, in attendance. He gratefully acknowledges the assistance of Mr. Johnson and the project site personnel in this effort, including Mr. Jim Muntzert, Project Manager. During the site visit, sufficient opportunity was available to examine logging procedures and drill core from the 2003 program as well as conduct a general overview of the property, including selected drill sites and historic core, and the condition of existing project infrastructure. Based on his experience, qualifications and review of the site and resulting data, the author of this report, Mr. Lacroix, is of the opinion that the programs have been conducted in a professional manner and the quality of data and information produced from the efforts meet or exceed acceptable industry standards. It is also believed that for the most part, the work has been directed or supervised by individuals who would fit the definition of a Qualified Person in their particular areas of responsibility as set out by the Instrument.

The resource study was a collaborative effort by Spectrum, Hatch and AMCL with Mr. Peter A. Lacroix, P.Eng. of AMCL directing the work and acting as the Independent Qualified Person as defined by the Instrument. While actively involved in the preparation of the resource estimates, Mr. Lacroix, Hatch and AMCL had no direct involvement or responsibility in the collection of the data and information or any role in the execution or direction of the work programs conducted for the project on the property or elsewhere. The updated mineral resource estimates are based on the most recent interpretations by project staff coupled with other data and reports provided by Spectrum. Much of the data, including the drill assay and geological database upon which the estimate is based, has undergone thorough scrutiny by project staff as well as certain data verification procedures by the author. In addition to the supplied data, the author has also had the benefit of the aforementioned site visit and detailed collaboration with project

staff and their consultants. Scott Petsel, Jack Cote and Robert Prevost of NovaGold/Spectrum are thanked for their contributions to this effort.

Much of the information and data upon which the estimates provided in this report are based have been obtained from the files and databases of Kennecott Canada Inc. Where appropriate, references for this information as well as other sources are provided in Section 21.

3 D ISCLAIMER

This report was prepared for Spectrum by the independent firms of Hatch and AMCL and is based in part on information not within their control. While it is believed that the information contained herein will be reliable under the conditions and subject to the limitations set forth herein, Hatch and AMCL do not guarantee the accuracy thereof. In particular, while the author has performed certain requisite data verification procedures, the volume of geological and sample data is such that 100% verification would preclude timely and cost effective completion of the study. Consequently, a certain reliance upon data, information and reports provided by others is necessary, although it is believed, for the most part, that this information has been prepared under the direction or supervision of individuals believed to be "Qualified Persons" as defined in the Instrument.

4 P ROPERTY D ESCRIPTION AND L OCATION

The Galore Creek property is located in mountainous terrain of northwestern British Columbia, approximately 1,030 km northwest of Vancouver and 90 km northeast of Wrangell, Alaska (Lat. 57.13°N; Long 131.47°W) . The claims lie at the headwaters of Galore Creek, a tributary of the Scud River, which flows into the Stikine River. The property is centered at latitude 57° 07'30"N and longitude 131°27'W. It occurs within the Liard Mining Division and straddles the boundary between NTS map sheets 104G/3 and 104G/4.

The claims that cover the property are wholly owned by Stikine Copper Ltd. This company was incorporated in 1963 and is presently controlled by QIT-Fer et Titane Inc (55%) and Hudson Bay Mining and Smelting Co, Limited (45%). The property comprises 253 claims and 39 fractions for a total of 292 two-post claims covering approximately 4,700 hectares. SpectrumGold has recently acquired additional claims in the Galore Creek property area through staking, purchase, and option agreements. A total of 16,850 hectares (42,000 acres) of claims have been staked by Spectrum in the area to the southeast and east of the Galore Creek property. SpectrumGold has also acquired 100% interest in the Jack, Sphaler Creek, Paydirt and Copper Canyon properties which cover a total area of 1,725 hectares (8,550 acres). In addition Spectrum has entered into an option agreement with Pioneer Metals Corporation. on the Grace property contiguous with the Galore Creek Claims to the north. Spectrum has also entered into an option agreement with Eagle Plains Resources Ltd. on the Copper Canyon property to the east. Collectively, these claims cover a total of 4,275 hectares (10,600 acres). Figure 2 illustrates the claim boundaries for the original claims of the Galore Creek property held by Stikine Copper Ltd.

A reclamation permit (MX-1-351) and a $100,000 bond for the Galore Creek property was previously posted by Kennecott and is presently in place. Asset Liability Management Group (ALM Group) and AMEC carried out an assessment of the existing liability at the Galore Creek property in the fall of 2003. As a result of this assessment, SpectrumGold presently has a $428,000 bond in place for the Galore Creek property that covers the reclamation costs for the property and the exploration program planned for the 2004 field season. This bond and 2004 work will be carried out under the British Columbia Mineral Exploration, Activities and Reclamation Permit (MX-1-608) issued to SpectrumGold.

3PN01 - Update on Resources Galore Creek Project, British Columbia

Page 10

final	Hatch/Associated Mining Consultants Ltd.
	June 3, 2004

---

In 1969, two map reserve permits were granted to Stikine Copper by the B.C. government. Notation of Interest Number 85742 covers an area of approximately 1,060 hectares in the vicinity of the Anuk and Stikine Rivers and was established due to its importance as potential mill site, tailings disposal and town site. A second Notation of interest (Number 886067) was established to protect the most viable access route from Galore Creek to the Stikine River (Scud Airstrip). These reserves do not give the owner any exclusive rights to the areas but simply ensure that Stikine Copper will be informed of any other applications for land alienation. Both these permits have been renewed at various times and are presently extended until further notice from the Skeena Regional Office of Land and Water Management. A large no-staking reserve was granted by the Ministry of Energy, Mines and Petroleum Resources in 1968 to cover a tract of land extending from the Galore Creek claims to the Stikine River. This was requested by Kennco (Stikine) Mining Ltd. for prevention of nuisance staking over a proposed tunnel access route to the property. In March 1989, this was amended to a "Conditional Reserve" where staking was permitted but claim holders could not interfere with any tunnel or underground workings created by previous mineral title holders.

The project falls within the boundaries of the Cassiar Iskut-Stikine Land and Resource Management Plan (LRMP) which was finalized in May 2000. The approved plan supports further exploration and development of the areas mineral resources by providing information to be considered during the permitting and impact assessment processes. The LRMP is primarily in territory claimed by the Tahltan First Nation. The Tahltan Joint Councils, representing the Tahltan Band from Telegraph Creek and the Iskut Band, were full table members throughout the process and endorsed the LRMP. Neighbouring First Nations include the Nisga'a, Kaska, and Tlingit Nations.

The LRMP identifies 15 geographic resource management zones, covering 31% of the plan area. One of these, the Lower Stikine-Iskut Grizzly Salmon Management Zone, includes the valley of the Stikine River from the Chutine confluence to the US border, and the lower Iskut River west of the Craig River. It also includes the Scud River into which Galore Creek drains. Mineral exploration and development are accepted activities within the Coastal Grizzly/Salmon Management Zone, including road access where needed. Adjacent areas includes an operating mine (Eskay Creek), a number of past producers including the recently closed Snip mine, two projects in the Environmental Assessment Process (mine development review phase), and a number of other exploration prospects. Logging is only allowed for the purposes of mineral exploration and/or mine development and for localized use.

The Galore Creek claims fall within the broader traditional territory of the Tahltan First Nation as registered in their Statement of Intent (SOI) area through the treaty process. Mining has long been an important segment of the Tahltan economy. Men from Telegraph Creek and Iskut were trained by professional geologists to work as prospectors during the initial exploration of the area by helicopter in the late 40s and 50s. Tahltans also worked in the asbestos mining town of Cassiar, north of Dease Lake, from construction through mining to shutdown. As well, the Tahltans participated in the exploration, construction, operations and reclamation of the recently closed Golden Bear

Mine, also north of Dease Lake. The largest mining operation in the area at present is the Eskay Creek gold mine located 80 kilometres southeast of Galore Creek. Approximately 35% of the Eskay Creek work force (including contractors) are First Nations people. The Tahltan Nation has the road maintenance and catering contracts at Eskay.

In 1992 Kennecott conducted an inventory of the camp and it was cleaned up in preparation for a period of inactivity. Other work included reclamation of roads and drill pads built for the 1991 exploration program. In 1999 Kennecott retained Norecol Dames and Moore (NDM) to conduct a site inspection to assess reclamation requirements and costs. Water draining from the Central Zone adit was tested and found to contain metal concentrations well below the Contaminated Sites Regulation Aquatic Life (CSR AW) standards. The portal site was heavily overgrown and showed signs of collapse within. It was recommended that all adit entrances be sealed.

In 1999, the Ministry of Energy, Mines & Petroleum Resources requested a reclamation bond of $100,000 be placed on the property in order to permit any further exploration work. The amount of the bond was based on costs of reclaiming the camp site, airstrips and previous exploration activity.

In September 2002 Keewatin Consultants conducted a site visit to assess environmental liabilities. Items noted included leakage from bulk fuel storage tanks and fuel drums near the camp area, a scrap metal site approximately 750 metres south of the camp and numerous empty fuel drums around the Portal airstrip.

In August 2003, SpectrumGold entered into an option agreement to acquire a 100% interest in the project by completing a pre-feasibility study on the project and making payments to the parties totaling US$20.3 million within a period of 8 years. Payments of US$0.3 million in aggregate are required over the first three years of the option, with the remaining US$20 million to be paid over the following 5 years. There will be no retained interests, royalties or back-in rights on the project

5 A CCESSIBILITY , C LIMATE , L OCAL R ESOURCES , I NFRASTRUCTURE AND P HYSIOGRAPHY

The property is located approximately 200 kilometers north of Stewart B.C. and 96 kilometers northeast of Wrangell, Alaska the two closest communities with tidewater facilities. The town of Smithers, 370 kilometers southeast, is the nearest major supply centre and has an airport with regularly scheduled flights to and from Vancouver.

Galore Creek and the Scud River are part of the tributary system of the Stikine River, an international waterway that drains an area of 49,000 square kilometers. In the past the river was used by shallow draft barges and riverboats to transport goods from Wrangell, Alaska to Telegraph Creek, British Columbia, a distance of 302 kilometers. The river is navigable for this type of watercraft from mid May to October. The nearest point on the Stikine River to the property is the mouth of the Anuk River, which lies 16 kilometers west of the camp.

Helicopter is the present means of access to the Galore Creek property. A 500-meter gravel airstrip is located on the property but will require repair for use by aircraft with short take-off and landing capabilities. A second airstrip at the mouth of the Scud River has been utilized in the past for aircraft up to the size of a DC-3 and would also require repair for use.

The Bob Quinn airstrip on the Stewart-Cassiar Highway is located approximately 75 kilometers east of Galore Creek and was the staging area for the 1991 project mobilization and demobilization.

In the 1960's Kennecott constructed 48 kilometers of road from the mouth of the Scud River to the Galore Creek camp. This road would require repair along the Scud River and portions of the Galore Creek Valley for use by the project.

Galore Creek is located in the humid continental climate zone of coastal British Columbia and is characterized by cool summers and cold humid winters, with several months of snow cover. Summer temperatures may be above +20°C and minimum winter temperatures may fall to well below -20°C. Average annual precipitation is 76 cm with approximately 70% of this falling between September and February, mainly as snow.

Elevations on the property range from 500 to 2080 metres above sea level. The terrain over the central and northern portions of the property is gentle and rolling and the surrounding topography is characterized by rugged mountains. The elevation of the tree line is variable but alpine vegetation predominates above the 1100-meter level. Below that, forests are made up of Balsam fir, Sitka spruce and a few cedars. Higher up the valley, the moraines are bare to sparsely overgrown by sub-alpine vegetation.

The project is currently isolated from power and other public infrastructure. Sufficient space is available on site for the various facilities required for a mining operation,

including personnel housing, stockpiles and process facilities. Ample water supply is available from surface and subsurface sources.

Mineralization was first discovered in the upper Galore Creek valley in 1955 by M. Monson and W. Buchholz while prospecting for Hudson Bay Exploration and Development Company Limited. Staking and sampling were completed in the area in 1955. Work in 1956 included mapping, trenching and diamond drilling. No further work was undertaken and most of the claims were allowed to expire.

In 1959 reconnaissance stream silt surveys were carried out by Kennco Explorations (Western) Limited in the Stikine River area. Results from this work prompted Kennco to stake mineral claims the following year around the remaining 16 claims owned by Hudson Bay. Four of the original claims were subsequently optioned by Consolidated Mining and Smelting Company of Canada Limited from W. Buchholz. Late In 1962 the three companies agreed to participate jointly in future exploration work. As a result, Stikine Copper Limited was incorporated in 1963.

Work conducted since discovery in 1955, outlined a significant gold-silver-copper resource in the Central Zone and identified a number of satellite deposits of which the most important are the Southwest Zone and two deposits known as the North Junction and Junction Zones.

From 1960 to 1968, the property was operated by Kennco Explorations (Western) Ltd. Exploration work during this period included 53,164 metres of diamond drilling in 235 holes and 807 metres of tunneling in two adits. The Central Zone was the focus of most of this work. No work was done from 1968 to 1972. In 1972, Hudson Bay Mining and Smelting became operator and in 1972 and 1973 an additional 25,352 metres of diamond drilling was completed in 111 holes. This work focused exclusively on blocking out resources in the Central and North Junction zones. A further 5,310 metres of diamond drilling was completed in 24 holes in 1976. In 1989, Mingold Resources Inc. (an affiliated company of Hudson Bay's) operated the property in order to investigate its gold potential. Mingold drilled a further 1,225 metres in 18 holes during 1990. Kennecott resumed operatorship of the project in 1991 and completed 13,830 metres of diamond drilling in 49 holes. An airborne geophysics survey and over 90 line kilometers in an induced polarization (IP) survey were also completed. The other eleven mineral occurrences on the property are: Junction, North Junction, West Rim, Butte, Southwest, Saddle, West Fork, South Butte, South 110, Middle Creek and North Rim.

Table 2 summarizes the work performed during past exploration programs.

A mineral resource estimate was prepared by Mine Reserve Associates, Inc. (MRA) for Kennecott Corporation in 1992 and included both the Central and Southwest Zones. This study, entitled "Pre-feasibility Mining Evaluation Galore Creek Project" reported a ‘Proven and Probable reserve' of 342.5 Million Tonnes grading 0.64% Cu and 0.38 g/t Au. A further 105.7 Million tonnes grading 0.46% Cu and 0.44 g/t Au was classified as ‘Possible'. This resource was used as the basis for several subsequent pit designs with strip ratios ranging from 1.5:1 to 2.2:1. MRA's scope of work did not include check assay analysis or geological interpretation.

In November of 2002, Kennecott re-classified the mineral resource to comply with current industry reporting standards. Economic parameters used were a US$10/tonne in situ metal value as a cutoff grade based on US$0.80/lb Cu and US$320/oz Au prices. Model blocks were classified as ‘indicated' if at least 5 composites were used in the interpolation and the closest composite was within 60 metres of the block centroid. All other estimated blocks were assigned to the ‘inferred' category. Results are summarized in the following table:

Table 3 - Historic Mineral Resource Estimate
Galore Creek – Kennecott 2002

Silver was not included in the 1992 resource model but earlier resource estimates from the 1970's reported grades ranging from 6.8 to 7.5 g/t for the Central Zone at corresponding copper grades of approximately 1%.

7 G EOLOGICAL S ETTING

7.1 Regional Geology

The Galore Creek deposits lie in Stikinia Terrane, an accreted package of Mesozoic volcanic and sedimentary rocks intruded by Cretaceous to Eocene plutonic and volcanic rocks. The eastern boundary of the Coast Plutonic Complex lies about 7 kilometers to the west of the claims. The property lies within a regional transcurrent structure known as the Stikine Arch.

Stikine Terrane at this latitude can be grouped into four tectonostratigraphic successions. The first and most important one in this area is a Late Paleozoic to Middle Jurassic island arc suite represented by the Stikine assemblage of Monger (1977), the Stuhini Group (Kerr, 1948) and Hazelton Group equivalent rocks. The other successions are: Middle Jurassic to early Late Cretaceous successor-basin sediments of the Bowser Lake Group (Tipper and Richards, 1976); Late Cretaceous to Tertiary transtensional continental volcanic-arc assemblages of the Sloko Group (Aiken, 1959); and Late Tertiary to Recent post-orogenic plateau basalt bimodal volcanic rocks of the Edziza and Spectrum ranges.

The oldest stratigraphy in the area is known as the Stikine assemblage and comprises Permian and older argillites, mafic to felsic flows and tuffs. These rocks grade upward into two distinctive Mississippian limestone members separated by intercalated volcanics and clastic sediments. The topmost stratigraphy consists of two regionally extensive Permian carbonate units that suggest a stable continental shelf depositional environment. The Middle to Upper Triassic Stuhini Group unconformably overlies the Stikine assemblage. Stuhini Group rocks comprise a variety of flows, tuffs, volcanic breccias and sediments, and are important host rocks to the alkaline intrusive-related gold-silver-copper mineralization at Galore Creek. They define a volcanic edifice centered on Galore Creek and represent an emergent Upper Triassic island arc characterized by shoshonitic and leucitic volcanics (de Rosen-Spence, 1985), distal volcaniclastics and sedimentary turbidites. The succession at Galore Creek was divided by Panteleyev (1976) into a submarine basalt and andesite lower unit overlain by more differentiated, partly subaerial alkali-enriched flows and pyroclastic rocks.

A fault-bounded wedge of unnamed Jurassic sediments unconformably overlies the Stuhini Group rocks. Within this unnamed Jurassic succession is a basal purple to red polymictic boulder and cobble conglomerate with an arkosic matrix. It contains granitic clasts including distinctive K-spar porphyries that are Galore Creek syenite equivalents.

Three intrusive episodes have been recognized in the region. The earliest and most important is the Middle Triassic to Middle Jurassic Hickman plutonic suite that is coeval with Upper Triassic Stuhini Group volcanic flows. The Mount Hickman batholith comprises three plutons known as Hickman, Yehino and Nightout. The latter two are exposed north of the map area. The Schaft Creek porphyry copper deposit is associated

with the Hickman stock, and is located 39 km northeast of Galore Creek. This stock is crudely zoned with a pyroxene diorite core and biotite granodiorite margins. Alkali syenites of the Galore complex like those found at the nearby Copper Canyon deposit and the pyroxene diorite bodies of the zoned Hickman pluton have been interpreted as differentiated end members of the Stuhini volcanic - Hickman plutonic suite by Souther (1972) and Barr (1966). The alkali syenites are associated with important gold-silver-copper mineralization at Galore Creek and at Copper Canyon. These rocks are believed to be at least as old as Early Jurassic in age, based on K-Ar dating of hydrothermal biotite in the syenites intruding the sequences (Allen, 1966). An Ar-Ar age of 212 Ma (Logan et al., 1989) in syenite may give the time of crystallization of the intrusive rocks at Copper Canyon, to the east of Galore Creek. More recent U-Pb dates of Galore Creek syenites have given ages ranging from 205-210 Ma (Mortensen, 1995).

Coast Range intrusions comprise the large plutonic mass west of the map area. Three texturally and compositionally distinct intrusive phases were mapped by previous workers. From inferred oldest to youngest, they are K-spar megacrystic granite to monzonite; biotite hornblende diorite to granodiorite; and biotite granite. Small tertiary intrusive stocks and dykes are structurally controlled in their distribution. At Galore Creek young post-mineral basalt and felsite dykes are abundant as a dyke swarm in the northwest part of the property. Elsewhere, Tertiary intrusions may be important in their association with small gold occurrences.

The regional geology has been affected by polyphase deformation and four main sets of faults. The oldest phase of folding is pre-Permian to post-Mississippian and affected the Paleozoic rocks between Round Lake and Sphaler Creek. This deformation is characterized by bedding plane parallel foliation in sediments and fragment flattening in volcaniclastics. Pre-Late Triassic folding is characterized by large, upright, tight to open folds with north to northwest trend of axial plane traces and westerly fold vergence. Metamorphism accompanying the first two phases of deformation reached greenschist facies. The third phase of folding is manifested as generally upright chevron folds with fold axes pointed west northwesterly.

The oldest and longest-lived fault structures in the area have a north strike and subvertical dip. The best example occurs on the west flank of the Hickman batholith, where a major fault juxtaposes Permian limestone with a narrow belt of Stuhini Group volcanics. The second important fault type occurs at Copper Canyon as a west directed thrust fault with a north strike and east dip of 30 to 50 degrees. It juxtaposes overturned Permian limestone and Middle Triassic shale with Stuhini volcanics below. Early to Middle Jurassic syenite intrusions occupy this contact. A third important set of faults with northwest strike mark the boundary between Upper Triassic and Paleozoic rocks between Scud River and Jack Wilson Creek. The youngest faults have a northeast strike direction and are of great local importance. At Galore Creek, some of these faults show considerable post-mineral movement of up to 200 metres while others appear to control the emplacement of mineralized intrusive phases and breccia bodies.

The Galore Creek property is mainly underlain by volcanic and sedimentary rocks of the Middle to Upper Triassic Stuhini Group. They are intruded by an Alkali Syenite complex composed of multiple intrusions emplaced into volcanic rocks of similar composition. The complex is centered in the west fork of Galore Creek and is 5 km in length and 2 km in width. To date, twelve gold-silver-copper mineralized zones have been identified on the property. The mineralization is most common in highly altered volcanic rocks and to a lesser degree in syenite intrusions.

7.2.1 Volcanic Rocks

Volcanic rocks host the bulk of the mineralization at Galore Creek and have been subdivided into six units.

A heterogeneous sequence of augite bearing mafic flows, flow breccias and volcaniclastics are interbedded with pseudoleucite volcanic rocks in the northern portion of the Central Zone. Rare bedding is preserved in lapilli tuffs of this unit and graded bedding is locally preserved. These rocks generally host only weak to moderate mineralization in comparison to the pseudoleucite–bearing rocks.

Augite-bearing flows contain porphyritic and, infrequently, amygdaloidal textures. Augite phenocrysts vary in size from 2-5 mm and are generally euhedral to subhedral, stubby and dark green to black. They comprise up to 30% of the rock and are supported in a medium to dark green, aphanitic groundmass. The augite phenocrysts are usually altered to biotite, epidote and chlorite. Locally, strong garnet-biotite-orthoclase alteration is also observed. Interbedded with the augite bearing flows are augite-bearing volcaniclastics in the form of fine and coarse lapilli tuffs, tuff breccias and flow breccias containing subangular to subrounded fragments of augite porphyry. These volcaniclastics are generally matrix supported.

7.2.1.2 Pseudoleucite-Bearing Volcanics (V2)

Pseudoleucite-bearing trachytes occur as moderately west dipping sequences interbedded with augite bearing units, intermediate and lesser mafic volcanic rocks. The original textures are often obliterated by intense orthoclase and sericite alteration. Copper/gold mineralization appears to occur preferentially in these rocks.

In unaltered areas, euhedral and broken pseudoleucite phenocrysts up to 1.5 cm occur within a bluish grey to salmon pink groundmass. These phenocrysts often exhibit orthoclase-sericite altered cores. Rims are sometimes altered to sericite, magnetite and chlorite.

In the Central Zone, fragments of pseudoleucite bearing volcanic rocks are seen as fragments in mineralized hydrothermal breccias containing abundant garnet.

Orthoclase-bearing volcanics are predominantly fine and coarse crystal lithic tuffs with possible subordinate flows and are common in the southern part of the Central Zone, where they crop out on surface and are often seen in drill core. In this area, they are often strongly mineralized with disseminated bornite, chalcopyrite and gold. They appear to be cogenetic and coeval with dark syenite porphyry intrusives, which may be their subvolcanic equivalents.

The crystal fragments in the tuffs are broken orthoclase shards up to 7 mm across supported by a highly altered biotite-orthoclase +/- garnet-anhydrite matrix. Rare bedding is preserved locally.

7.2.1.4 Undifferentiated Volcanics (V4,V5,V6)

In some areas, intense alteration has obliterated original textures resulting in the more vague classification of "undifferentiated volcanics". Such rocks have been classified on the basis of colour and association.

Mafic volcanic rocks (V4) are dark green, chloritic flows and tuffs common in the north part of the Central Zone. These are interbedded, and may, in part, be correlated with unit V1 (augite-bearing volcanics). Porphyritic and amygdaloidal flow textures have been preserved locally and volcanic clasts are sometimes preserved in pyroclastic rocks.

Intermediate volcanic rocks (V5) are very common in the Central Zone. These rocks are medium greenish grey volcaniclastics and flows, and may be aphyric equivalents of the pseudoleucite bearing volcanic units. Included in this unit are possible trachy-andesites containing subrounded orthoclase phyric fragments. Aphanitic volcanic clasts up to 3 cm across have also been observed within a fine grained to aphanitic matrix. Secondary biotite occurs both as a spotted to patchy alteration and as coarse aggregates an veins.

Intense orthoclase flooding has resulted in pale grey, felsic volcanic rocks (V6) which are fine to medium grained volcaniclastics and flows. V6 rocks are present in the north and central part of the Central Zone, often interbedded with pseudoleucite volcanic rocks that may be their equivalent.

7.2.2 Sedimentary Rocks

Sedimentary rocks consist mainly of interbedded siltstone, argillite, greywacke and conglomerate and are most common immediately to the north of the Central Zone. The large amounts of volcanic and subordinate intrusive material in these rocks indicate a volcanic-intrusive source.

Sedimentary structures such as graded bedding, flame structures and channel scour features have been observed in drill core north of the Central Zone, from outcrop in North Rim Creek, and more rarely in drill core from the Central Zone. Where observed in outcrop, tops indicators show the sequence is right side up.

Conglomerates are common north of the Central Zone, in North Rim Creek and North Rim Zone, and in the North Junction Zone. The unit is heterolithic and unsorted. Fragments are subrounded to rounded, matrix supported by sand and silt sized grains. Fragments of volcanic and syenitic rocks are present and comprise up to 30% of the rock. Conglomerate contains local intercalations of argillite and greywacke. Channel scours and load casts are common.

Grey-green, poorly sorted, medium to coarse grained greywackes are common north of the Central Zone, in North Rim Creek. They also appear rarely in drill core within the Central Zone as intercalations with lapilli tuffs. This unit is locally very well bedded and graded. Fragments of argillite and volcanic material are subangular to subrounded.

7.2.2.3 Siltstone (S3)

Siltstone is fine to medium grained, grey, massive to well bedded and locally contains graded bedding.

7.2.2.4 Argillite (S4)

Argillite occurs as alternating medium to dark grey and black, aphanitic, well bedded sequences. Beds vary in thickness from 0.5 to 1 cm. Local flame structures have been observed, particularly in North Rim Creek and in the 2070 Adit in the Central Zone, where it is interbedded with siltstone (S2).

Multiple intrusive phases are present in the complex and divided into pre, inter, late and post mineralization phases. This classification is based on crosscutting relationships along with the degree of alteration and mineralization characteristics. Based on recent U-Pb age dating (Mortensen et al, 1995), the probable crystallization ages range from 210 ± 1 Ma for a syn-mineral intrusion (I4) immediately south of the Central zone, to 205.0 ± 2.3 and 205.0± 1.8 for post-mineral intrusions (I9b). These results indicate that the magmatic activity continued for a period of at least 1.6 Ma. and that mineralization occurred at an early stage.

Petrologic examination (Enns et al, 1995) has shown that the Galore Creek intrusive rocks contain variable proportions of orthoclase, plagioclase (oligoclase or albite), pseudoleucite, melanite, clinopyroxene, biotite and hornblende phenocrysts in a matrix of pilotaxitic K-feldspar, disseminated magnetite, apatite and titanite.

Early intrusive units (I1 through I5) consist of K-feldspar and pseudoleucite porphyritic dikes and sills. These are followed by relatively equigranular intrusions (I6 and I8), K-feldspar porphyritic and megaporphyritic units (19b-I11), and a relatively equigranular intrusion (I12). This apparent oscillation between porphyritic and equigranular textures may reflect variations in the volatile fugacities of the melts.

The modal change in the primary mineral assemblage of the Galore Creek intrusions, from syenitic to monzonitic, back to syenite and finally to quartz syenite, suggests differences in the compositions of the parent melts.

7.2.3.1 Pre-Mineralization Intrusions (Units I1 to I3)

These early intrusives are pre-mineral and include thin pseudoleucite porphyry dikes (I2), and syenite porphyry. Medium-grained syenite porphyry with prominent pseudoleucite megacrysts is present as an elongate body centered on the Central Zone as well as within the North Junction Zone. The Central Zone intrusion extends for at least 1500m north-south, 500 metres east-west and constitutes a west-dipping body. It is closely associated with the most intense hydrothermal alteration and may represent the causative intrusion for the main stage Cu-Au mineralization.

7.2.3.2 Inter-Mineralization Intrusions (Units I4a & I4b)

Relatively small bodies of inter-mineralization porphyry are present in association with the Central and Junction-North Junction Zones in close proximity to the pre-mineralization porphyry intrusions. These porphyries have been termed "dark syenite porphyry" due to the dark colour of the groundmass resulting from the presence of magmatic and hydrothermal biotite. Several inter-mineralization porphyry phases are

present, the earlier ones carrying significant gold-silver-copper mineralization. Orthoclase megacrysts are ubiquitous, occasionally accompanied by pseudoleucite.

7.2.3.3 Late-Mineralization Porphyries (Units I5 to I12)

Dikes and/or sills of late-mineralization porphyries intrude the Central and Junction/North Junction zones and may also constitute a large stock between these zones. These phases are easily recognized because of the lack of mineralization and the propylitic alteration assemblage. The most common phases are grey equigranular to porphyritic, medium-grained syenite (I8), megaporphyry (I9a & b), grey medium grained syenite porphyry (I11a) and lavender porphyry (I12). Minor phases include dikes and small stocks of fine-grained syenite and syenite porphyry (I6 & I7), plagioclase syenite porphyry (I10) and medium-grained syenite porphyry (I11). Most of these late minor phases were defined from drill core in the Central zone. Minor mineralization associated locally with these dikes suggests that they are either syn-mineral, or minor late mineralization was introduced with them.

The lavender porphyry (I12) is a quartz-bearing syenite that occurs as a small body near the centre of the complex. It intrudes a late-mineral breccia exposed in Dry Creek and is locally mineralized with chalcopyrite.

7.2.3.4 Post mineral dykes

Fine grained, mafic to felsic post mineral dykes occur mainly as steep to vertical, north-south and east-west trending bodies. Rarely, they have sub-horizontal attitude. The ages of these dykes are undetermined but thought to be related to the Coast intrusions of mid-Mesozoic to Tertiary age.

7.2.4 Breccias

Both hydrothermal (orthomagmatic) and diatreme (phreatomagmatic) breccias occur in the deposit areas. They are distinguished mainly by the fragment shapes and types, nature of matrix, alteration assemblages and the presence or absence of altered and mineralized clasts.

The Southwest Zone is hosted by an unbedded, polylithologic, matrix-supported breccia that is interpreted tentatively as the fill of a phreatomagmatic diatreme. Clasts include a variety of altered and/or mineralized lithologies including some that are copper bearing, indicating that some mineralization events preceded brecciation. The breccia matrix is

rich in rock flour but also appears to contain comminuted crystals of possible tuffaceous origin.

A dark-coloured, unbedded breccia lacking obvious mineralized clasts occurs at several locations in the Central and Junction-North Junction zones. There is a possibility that this rock of volcaniclastic origin but the most likely interpretation is a facies of the diatreme breccia.

Several, relatively small bodies of orthomagmatic (magmatic-hydrothermal) breccia carrying gold-silver-copper mineralization were emplaced during the latter intrusive phases of the system. Breccia bodies south and west of the Central Zone carry clasts of propylitized, late-mineralization mega-syenite porphyry. The breccias are mainly clast-supported and characterized by a matrix of magnetite accompanied by one or more of epidote, garnet, pyroxene and biotite.

One of the most prominent breccias at Galore Creek is exposed in lower Dendritic Creek on the west side of the Central Zone. This breccia is pervasively altered and mineralized and grades into the highly altered I3 intrusion and adjacent pseudoleucite-phyric volcanic rock. This close spatial association with the early I3 intrusion suggests that this breccia formed early in the main stage of mineralization.

Other orthomagmatic breccias include the mineralized Saddle Zone breccia and the West fork breccia. The latter was found to contain clasts of unaltered intrusive phases including I11, suggesting that it formed late in the intrusive sequence.

7.2.5 Alteration Assemblages

7.2.5.1 K-Silicate Alteration

The dominant alteration assemblage at Galore Creek is K-silicate, which affects the volcanics, pre- and inter- mineralization porphyries, and to a minor extent, post-mineralization porphyries. The most intense K-silicate alteration is found in the pre-mineralization porphyry stock and comprises K-feldspar, biotite, magnetite, anhydrite, bornite and chalcopyrite. K-silicate alteration is notably weaker in the inter-mineralization porphyries although biotitization of the diatreme breccia is intense.

Within the core region of the Central Zone, an associated "Ca-K-silicate" assemblage is developed, characterized by the addition of a dark brown garnet with locally occurring diopside, epidote and plagioclase. This alteration likely occurred when the same K-silicate altering fluids encountered more calcic mafic rocks, derived excess Ca and precipitated garnet. The decreasing Ca component in the alteration from the core region

to the north and south regions of the Central Zone is accompanied by generally increasing magnetite and early hematite.

A calc-silicate assemblage dominated by diopside and pale brown garnet replaces one or more fragmental horizons in the footwall of the pre-mineralization porphyry stock. This assemblage is considered to constitute prograde skarn and is metal-poor unless overprinted by hydrothermal biotite. In the Central Zone, cross cutting relationships with the K-Silicate assemblage and lack of mineralization indicates that this alteration is very late in the sequence.

Intermediate argillic alteration appears to be most intense in the upper exposed parts of the pre-mineralization intrusion and often overprints K-silicate alteration with green sericite, carbonate and minor chlorite. Pre-existing magnetite is altered to hematite and much of the sericite is developed at the expense of hydrothermal biotite. Sericite alteration affects all the volcanic units, the pre- and, to a lesser extent, post-mineral porphyries. The paragenetic relationship of the alteration assemblages suggests that the intermediate argillic alteration was largely post mineral but prior to the intrusion of the post mineral porphyries.

7.2.5.4 Sericite-Anhydrite-Carbonate (SAC)

Sericite-anhydrite-carbonate (SAC) alteration overprints the dominant early alteration phases and is locally extensive. In the southern part of the Central Zone it is accompanied by late hematite. In the northern portion it is marked by increasing pyrite ± hematite and decreasing bornite-chalcopyrite, suggesting that copper was remobilized during SAC alteration. In the core of the Central Zone, SAC alteration is patchy and more common on the periphery. The later hydration of anhydrite to gypsum by groundwater action and consequent volume increase has resulted in intense sheet fracture development in parts of the Central Zone to depths of as much as 213 metres below surface. The gypsum has been leached out to depths ranging from 30 to 122 metres leaving loose crumbly sheets of rock. The sheet fractures are best developed in volcanic rocks and weak to absent in the porphyries.

An epidote-chlorite assemblage, locally accompanied by garnet, is present in the late-mineralization mega-porphyries and penetrated for a few metres into their wall rocks.

The Galore Creek intrusive complex is believed to represent an eroded volcanic centre. Syenite sheets are subvolcanic intrusions, the extrusive equivalents of which may be the orthoclase-bearing tuffs and pseudoleucite lavas.

The two main structural elements are faults, which offset and segment intrusive bodies, and a subhorizontal fracture cleavage. Faults are abundant in the Central Zone, but displacements on most of them appear to have been small. An exception is a major post-mineralization west-dipping fault east of the Central Zone that has an apparent normal displacement of up to 300 metres. Significant displacement might have occurred on faults in the northern part of this zone but it has been difficult to correlate marker beds and dykes between drill holes.

A north-trending mylonite zone at least 100m in thickness is exposed in volcanic rocks along the west margin of the intrusive complex. The zone dips 42° to the west and predates mineralization and alteration in upper Butte Creek.

A major east-west regional structure, marked by a prominent aeromagnetic linear, lies parallel to lower Dendritic Creek. This feature cuts through the Central Zone where it is occupied by a late-mineral 19b-megaporphyry dike at section 6773N.

A series of northeast-striking faults from extend from east of the Southwest zone, northward beyond the boundary of the claims. The long axes of both early porphyry I3 and the breccia in lower Dendritic Creek are aligned parallel to this trend suggesting a common regional structural control. This fault system intersects the east trending linear described above near the core of the Central Zone.

Post-mineralization, sub-vertical dikes and felsic dike swarms occupy east-trending structures of probable Tertiary age.

The terms "sheet fractures" and "fracture cleavage" are applied to a set of near-surface, closely spaced subhorizontal fractures. These are attributed to expansion resulting from the hydration of anhydrite by meteoric waters.

The Galore Creek deposit is classified as an alkalic intrusive related gold-silver-copper deposit. These deposits occur throughout the length of Stikinia Terrane in Upper Triassic Nicola-Takla-Stuhini volcanic rocks and comagmatic alkaline plutons. Deposits of this class include Afton, Mount Polley, Copper Mountain and Lorraine. They typically occupy brecciated and faulted subvolcanic zones in the intrusions and country rocks that are overprinted by extensive potassium, propylitic and pyrometasomatic alteration zones. International examples of this type of deposit include the OK Tedi mine in Papua New Guinea, Candelaria in Chile and the deposits of the Cadia Hill district in Eastern Australia.

Mineralization at Galore Creek occurs in upper Triassic felsic to intermediate volcanic flows and fragmental rocks. It is associated most closely with intense, pervasive K-silicate alteration as replacement, disseminated and fracture-controlled chalcopyrite with locally abundant bornite. Higher gold values are normally associated with bornite mineralization.

The Central Zone is the largest and most extensively explored of all the deposits and is characterized by fairly complex geology. Mineralization is exposed in the southern part of the zone, but elsewhere it is covered by up to 75 m. of glacial overburden. Between 80% and 90% of the gold-silver-copper occurs as sulfide replacement of the host volcanic rocks. The grade of the mineralization commonly exceeds 1% Cu decreasing rapidly at the margin of the zone.

The Central Zone deposit has an orientation of 015° and dips steeply to the west. It is 1700 metres. long, 200 to 500 metres wide and has been traced to a depth of 450 m and remains open. The eastern boundary of the Central Zone mineralization lies near the surface projection of a major, steeply west dipping, brittle normal fault. In the west and south, mineralization is truncated by post-mineral megaporphyry dykes. Intense SAC alteration has obliterated mineralization in the northwestern part of the Central Zone. In the north, mineralized volcanic rocks end abruptly against a thick sequence of weakly to unmineralized epiclastic sedimentary and volcanic rocks.

The Central zone exhibits considerable internal variations in both mineralization and alteration. Hydrothermal alteration changes from Ca-K-silicate in the core region to intense K-silicate alteration toward the north and south parts of the zone. In terms of gold-silver-copper replacement mineralization, the most favourable volcanic lithologies are the pseudoleucite-bearing volcanic rocks in the north and the dark crystal tuffs in the

south. Augite-bearing units in the north are low to moderate in copper content and the core of the deposit hosts a mineralized orthomagmatic breccia. Gold values are highest in the northern and southern portions of the Central Zone where significant disseminated bornite, magnetite and hematite are present. Lower gold grades correlate with the intense Ca-K-Silicate altered core region. Chalcopyrite is the most important copper mineral and occurs as replacements, disseminations and fracture fillings throughout the zone. Supergene copper mineralization is minor and occurs primarily as malachite, azurite and chrysocolla on fractures within 60 metres of surface. Pyrite increases in abundance to the east of the Central Zone reaching concentrations of up to 5%.

The Southwest Zone is located about 600 metres southwest of the south end of the Central Zone and contains some of the highest grade near surface gold mineralization. Kennecott envisioned the southwest area as a potential high-grade starter pit. Drilling has outlined an elongate pod shaped body that trends roughly east-west and dips approximately 60° to the south. The zone is up to 400 metres long and may be as wide as 140 metres; however, the 1991 drilling suggests that the zone narrows at both the eastern and western ends of the deposit. The Southwest Zone is still open at depth. Primary hosts for the Southwest mineralization are a diatreme breccia and an early syenite phase intrusive. Localization of high-grade gold-silver-copper mineralization within the diatreme appears to relate to a combination of structural traps and mineralizing faults.

The Junction and North Junction deposits lie about 2 kilometers northwest of the Central Zone and about 460 metres higher in elevation. They are a series of irregular, flat-lying manto-shaped bodies plunging about 20° to the northeast. Together these two zones are interpreted to have been a single deposit that was originally 1400 metres in length before it was faulted. Width of the zones varies from 50 to 150 metres. The mineralization, consisting of disseminated chalcopyrite and bornite, is hosted by fine to coarse lapilli tuff and feldspar phyric flows. Higher gold and copper grades correlate with the presence of bornite in the North Junction zone. K-silicate alteration consisting of pervasive hydrothermal biotite and K-feldspar flooding is associated with the mineralization. A large mass of late-mineral I9b megaporphyry truncates the zone on the west.

9.4 Other Zones

A number of smaller gold-silver-copper showings occur in the claim group but have had only limited exploration work.

The Butte and West Rim deposits lie along the western margin of the Galore Creek complex. Both host NE-striking, west-dipping zones of disseminated chalcopyrite and

bornite. At the Butte zone, mineralization is hosted by K-silicate altered pseudoleucite-bearing volcanics. At surface, mineralization grades up to 2.18% copper and 0.5 grams gold per tonne across 96.6 metres. The zone is truncated at depth by post-mineral (Ill) intrusions. Drilling in 1964 on the West Rim showing intersected 48.8 metres grading 0.8% copper. This zone is hosted by volcanic tuffs with intense K-Silicate alteration and is truncated at depth by I9b megaporphyry intrusions.

Between 1963 and 1991, ten holes were drilled in the North Rim Zone located about 1600 metres north northwest from the Central Zone. Here, a broad area of widespread weak copper mineralization was discovered in outcrop and subsequent exploration revealed coincident copper-gold geochemical and chargeability anomalies. Two holes intersected zones of low grade copper (0.3 -0.4%) over widths of 21-24 metres with accompanying gold grades up to 0.5 grams per tonne. The last two holes intersected many narrower (<10m) intervals grading 0.3 -1.0 grams gold per tonne and one as high as 6.34 grams per tonne over 3 metres. Spotty copper grades above 0.3% were also present. The area is underlain mainly by intrusive phases cut by small orthomagmatic breccia bodies with magnetite-rich matrix and Ca-K-Silicate alteration.

Field mapping in the Middle Creek area in 1991 re-located mineralization reported by previous prospectors in the mid 1960's. This area lies about 1 kilometer west of the Central Zone. Mineralization consisted of finely disseminated bornite, chalcopyrite and magnetite associated with pervasive fine-grained biotite and garnet in an altered, intermediate tuffaceous unit. Only one drill hole was completed and it intersected 63 metres grading 0.42% copper and 0.07 grams gold per tonne. The zone is apparently constrained by late megaporphyry intrusions (I9b), lavender porphyry (I12), and the large diatreme breccia in upper Dendritic Creek.

The Saddle Zone lies 2.6 kilometers southeast of the Central Zone and is hosted by a breccia dominated by a matrix of magnetite. Four holes were drilled here between 1954 and 1963 and two more added in 1990. The later holes intersected two 12 metre zones of mineralization grading 0.55% copper and 2.19 grams gold per tonne, and 2.49% copper and 3.98 grams gold per tonne, respectively. Three of the older holes also intersected low to moderate copper and high gold grades (between 1.0 and 6.9 g/t Au) but gold analyses were incomplete. An adjacent showing named the South 110 zone is possibly a continuation of the Saddle Zone to the north.

10 E XPLORATION

10.1 Geochemistry

Regional stream silt geochemistry was instrumental in the discovery of the mineralization at Galore Creek and more detailed silt sampling programs were carried out in 1960-61 and 1989.

A significant area of the property lacks sufficient soil development for soil geochemistry to be of any practical use. Soil grids were established in the areas around the North Rim and Southwest zones. A few reconnaissance traverse soil lines were also sampled along contour between the Saddle Zone and the Central Zone.

In 1991, 600 soil samples were collected from a grid established in the North Junction / North Rim area. Samples were taken on 20-meter stations along lines spaced 100 metres apart. A coincident Cu-Au soil anomaly with peak values of 9060 ppm copper and 550 ppb gold was located over the North Rim showing area. A total of 63 surface rock chip samples were also collected from various showings on the property.

Various ground and airborne geophysical programs have been carried out on the property. In 1961 H.W. Fleming of Toronto conducted a vertical magnetic field survey over 22 square kilometres on a line spacing of 800 feet (244m). The same year an airborne magnetic survey was carried out by Aero Surveys Limited over an area of 64 square kilometres using 270-meter line spacing.

Between 1961 and 1963, McPhar Geophysics Ltd. carried out pole-dipole resistivity/IP, VLF and AFMAG surveys over an area of 20 square kilometers with line spacing of 122 and 244 metres. Dipole lengths used were 30, 60 and 120 metres.

In 1991, Aerodat Limited conducted an airborne geophysical survey including magnetics, EM, radiometrics and VLF over an area of 35 square kilometers using a line spacing of 100 metres east-west and 150 metres north-south. Also in 1991, Lloyd Geophysics was engaged to carry out a combination of pole-dipole resistivity/IP, VLF and Total Magnetic surveys over an area of 5.5 square kilometers on lines spaced 100 metres apart with a 60 metre dipole spacing.

Both of the airborne surveys mapped three principal features; two large magnetic highs separated by a broad magnetic low. The largest magnetic high occupies the northwest quarter of the claim block and is coincident with a broad area of syenitic intrusive rocks. This feature is elongated in a north-northeast direction with dimensions of 4.5 by 2.0 kilometers. Although the 1991 survey showed a singular large high, it also revealed

numerous magnetic highs and lows which probably reflect the complex poly-phase nature of the intrusives. The second major magnetic high occurs in the southeast portion of the claim block roughly coincident with the intrusive rocks around the Saddle Zone. The magnetic low defined by both surveys lies within the central portion of the claim block and is coincident with an area of volcanic rocks. The low is elongate along the same north-northeast trend as the intrusive rocks.

The 1991 airborne survey revealed a set of prominent magnetic lineaments oriented N20°E. This direction is roughly parallel to the mineralization in the Central, Butte, Junction, West Rim, and North Junction zones. This set of lineaments also parallels two mapped faults associated with the east side of the southern Central Zone and one of these lineaments coincides with the west boundary of the Central Zone. Other significant magnetic lineaments are oriented east-west and N60°W. The former set appears to represent early structures based on observed segmentation and offset by later structures and obliteration by intrusive activity.

Airborne EM data reveal clusters of negative in-phase responses with a small out-of-phase component that correlate directly with mineralization at the Southwest Zone and in the southern half of the Central Zone. Similar responses were found lateral to mineralization at the Junction zones, the North Rim zone and the northern half of the Central zone. Nearly all the discrete conductors identified with the airborne EM occur within the Galore Creek drainage and are not likely related to massive sulfide mineralization. The survey also identified many narrow dike-like zones which are interpreted to be magnetite bearing structures.

The airborne radiometrics showed enhanced potassium levels over most known deposits, which was not unexpected due to the prevalence of potassic alteration associated with the copper mineralization. Inman (1992) recommended further processing and modeling of element ratios in order to remove the effect of vegetation and soil moisture content.

Results of the early IP surveys suggested that the mineralized zones did not respond well to IP. However, reprocessing of the data by Inman (1992) indicated that the core of the Central Zone exhibited a metal and resistivity high signature while the peripheral clay/argillic alteration zones showed a metal and resistivity low signature. The important geophysical characteristic of known mineralization within the Galore Creek area were summarized by Inman as follows (in order of importance):

Based on these criteria it was concluded that the Butte and North Rim zone area had the highest potential for extensions of known mineralization (Figure 7)

In order to extract a 50-ton bulk sample for pilot plant testing, an adit was driven into the Central Zone and samples were collected from four crosscuts. The work was carried out by Haste Mine Development between August 1966 and January 1967 and totaled 799 metres of underground drifting (2mx2m). The rock quality in this part of in the Central Zone was found to be generally weak and intensely fractured in gypsum-free areas, but tough and competent in zones of gypsum cementation.

Seven underground diamond drill holes were collared from the 2070 adit. Severe recovery problems were encountered because the holes were of small diameter and drilled sub parallel to the flat-lying, sheet fractures. No assay data was located for these holes.

Sampling of the adit and drift walls was carried out over continuous horizontal 3 metre (10 ft) intervals plus vertical channels alongside the traces of diamond drill holes. Although commonly referred to as "channel" samples, one internal memo described them as "contentious (sic) chip samples". The vertical samples taken adjacent to the drill hole traces correlated within 0.1% copper. When compared to horizontal samples on the opposite side of the drift, significant variation was found in higher-grade areas (>1.5% Cu) where massive blebs of chalcopyrite were encountered. In these areas variations often exceeded 0.4% copper for opposing walls. Subsequent check sampling along some of the same channels confirmed this variation.

At the North Junction Zone, a smaller adit (1.2mx2.1m or 4'x7') was collared in badly fractured and altered tuff. After driving through 26 metres of material grading about 0.5% copper, a low-grade dike was encountered. Total length of the adit was 51 metres.

11 D RILLING

11.1 Past Drilling Programs

Since initial discovery in 1960, there have been 439 diamond core holes drilled on the property. As shown in the following figure and table, most of this work has focused on the Central Zone, with lesser amounts of work on eleven other areas. Some zones have received only reconnaissance drilling.

During the 1970's drilling was principally confined to the Central Zone with nine holes drilled on the North Junction Zone. In the Central Zone average core recovery was between 75% and 85% with the poorest recovery at depths between 60 and 90 metres, where the open sheet fractures were encountered. At depths below 90 metres core recovery approached 100%. In the North Junction zone recovery averaged around 60%

due to shattered and sheared sections encountered both near surface and at intervals throughout the holes.

In 1989/1990, Mingold drilled holes on the Southwest Zone (8 holes, 1026 m.), the North Rim showing (6 holes, 546 m.), the Saddle Zone (2 holes, 226 m.) zones and 2 reconnaissance holes.

The 1991 program was mainly directed at areas peripheral to the Central Zone as well as exploration holes located in the Southwest, Butte, North Rim and Dry Creek zones. Only 6 holes were drilled within the Central Zone itself.

During 2003, Spectrum completed an 8-hole 2,800-meter drill program to verify previous results and better understand deposit variability, zonation and controls for mineralization. Particular emphasis was directed at understanding the gold variability in the deposit. The number, length and location of the holes completed during the 2003 program are listed in Table 5. In addition, 2 holes were lost in overburden and not completed.

All 8 diamond core holes were angle drilled in order to intersect the mineralized structures as close to right angles as possible. The 6 holes within the Central Zone were all drilled toward the east and focused on the gold rich lenses at the north and south peripheries of the zone, and the higher-grade central copper replacement zone. The 2 remaining holes were drilled in the Southwest Zone. These holes were angled towards the north.

Drill hole 441 was targeted to test both an upper mineralized horizon and the potential for a lower mineralized zone. The hole successfully intersected the upper horizon as well also encountering a new mineralized horizon not tested in previous drilling. This lowermost horizon totaled 65 metres in width before the hole ended. The hole ended in well mineralized material. Table 6 provides a summary of the most significant mineralized intercepts encountered in the 2003 program.

12 S AMPLING M ETHOD AND A PPROACH

Drill core in mineralized zones was generally sampled in 3-meter (~10 foot) intervals. In some areas where lithologic and/or mineralogical changes occurred, narrower assay intervals were employed. Intervals not sampled were mainly comprised of late stage, post-mineral dykes. The samples were tagged then split in half using a mechanical splitter. One half of the core was returned to the core box and the other either processed on site or shipped to an outside laboratory for analysis. The core returned to the boxes remains on site as a record of the hole. Pulps and rejects were stored either on site, in a warehouse in Smithers or in Vancouver. Those pulps and rejects remaining on site are in degraded sample bags and are not considered worth salvaging.

Core has been stored in either plastic, galvanized steel or wooden boxes. All have been marked with metal tags inscribed with the hole number and interval. An estimate 1,500 metres. of core was spilled in 1972 due to the collapse of a core storage rack. In the winter of 1976 one core shed collapsed and although most of the core was rescued, a number of intervals were not salvageable. Core from the Central Zone was largely re-logged as part of the 1991 exploration program. It is now stacked in pallets exposed to the elements and the top layers have suffered deterioration from weathering. Several intervals have also been removed in the past for the purposes of metallurgical testing. Other intervals have been quartered for check assaying.

No site-specific standards, blanks or field duplicate samples were used in any of the previous exploration programs. During the most 1991 program, every twentieth sample was re-assayed by an umpire laboratory while the main assay laboratory performed internal checks.

All drill core from the 2003 program, excepting intervals of overburden and till material, were sampled to create a complete geochemical profile of the drill hole. Drill core sampling occurred within a minimum of 1-metre (3.2 feet) and a maximum of 3-metre (10-feet) intervals. Drill core in mineralized intervals was generally sampled in 2-metre (~6.5 feet) intervals. In obviously unmineralized sections, core was sampled in 3-metre intervals. Variation from the prescribed sample length resulted when sample breaks were marked at observed lithologic, mineralogic and major structural changes in the core.

All core samples were tagged by the geologist that logged the hole. Photographs of the core were taken prior to splitting for analysis to provide a permanent record. All the drill core samples were split using a rock saw. One half of the core was returned to its original box (5-foot long wooden box) for long term storage. The remaining half was sealed in a polyethylene bag for direct shipment to the laboratory for analysis. The core that was returned to the box remains on-site as a permanent record of the hole. For long term storage of the core, the boxes have been marked with aluminum tags inscribed with the hole number, box number and sample interval. The boxes have been stacked on wooden supports and covered with plywood as temporary protection from the elements. An arrangement for longer-term on-site storage is planned for the 2004 program.

Samples were also taken for metallurgical purposes during the 2003 program. Sections of core were quartered and bagged and submitted for testing in Vancouver. These were: GC03-437 (24.39m - 47.3m), GC03-437 (47.3m - 74m), GC03-437 (74m - 91.36m), GC03-43 (258.17m - 306.93m), GC03-441 (99.67m - 142m), and GC03-442 (288m - 329.1m) .

Sampling protocol called for the geologist to insert three control samples; a blank, a standard and a duplicate for every 20 samples to be submitted to the lab for analysis. The 20-sample size equates with the size of the sample batch grouped for analysis at the lab. The placement of all control samples was essentially random within the 20-sample batch. Blanks, which consisted of non-metalliferous marble, were inserted and tagged in the sequence determined by the geologist and bagged during the core splitting and sampling collection. Duplicate sample locations were marked by the geologist. Empty bags with duplicate tags were submitted to the lab to indicate the sample was to be split for duplicate analysis.

Quality control analytical standards were purchased from WCM Sales of Vancouver BC for the 2003 program. The standards consisted of three types. There were two gold ore standards consisting of a high-grade (PM-152) and a typical or average grade (PM-184). The grades of these gold standards were 2.83 grams per tonne and 0.52 grams per tonne respectively. The copper standard (Cu 108) was 0.66% copper and 18.3 grams silver per tonne. Standards were stored on site in 30 gram sachets that were bagged in the sequence chosen by the geologist and submitted to the lab for analysis. The assay lab submitted 10% of the samples for internal checks. Only limited third party analysis of the assay results has been completed.

13 S AMPLE P REPARATION , A NALYSES AND S ECURITY

Sample preparation has gone through several transitions since the early drilling in the 1960's. Prior to 1964, drill core was split in 3-meter (10-foot) lengths and half of the core shipped to Coast Eldridge laboratory in Vancouver for copper assay. Some 30-meter (100 ft.) composites were assayed for gold during this period.

In 1964, a small assay laboratory was constructed on site and during the first season of operation, 3,747 samples were processed. Half of the split core was crushed on site to minus quarter inch then a 340 gram (0.75 lbs.) split was separated using a Jones splitter. At the lab the sample was split and crushed to minus10 mesh then 95% of the crushed material was pulverized to minus100 mesh and assayed for copper using a double digestion with titration and colorimetric determinations. Intervals that met or exceeded a threshold copper grade of 0.4% over intervals of 12 to 18 metres (40-60 ft.) were composited and shipped to Coast Edridge for gold and silver assays. It is not known if these pulps were re-homogenized before compositing. Some gaps in the original sampling exist within distinctly anomalous gold intervals. Early gold assays performed on 30.5 -meter (100 ft.) composites differ significantly from the 12-18 metre (40-60 ft.) composites assayed in later campaigns. These composites were assayed by commercial lab then checked by the Kennecott Research Centre lab and are considered to be reasonably accurate.

During 1964, cross checking of Galore Creek laboratory copper analyses were carried out on a routine basis by Kennco Explorations laboratory in North Vancouver and at Coast Eldridge laboratory. Several samples were also checked at Hawley and Hawley Assayers of Tucson and by Bear Creek Laboratory in Denver.

In 1966, composites from the Central Zone were reportedly re-assayed for gold and silver after discrepancies were found between the values obtained in 1964 and 1965. The re-assay was carried out at the Kennco Laboratory in North Vancouver. The results showed marginal increases in gold and silver content. Assay certificates could not be located for either original or re-assayed material.

In 1967, the pulps from 140 samples were split and a portion analyzed by five separate laboratories; including Coast Edridge, Sudbury and three of Kennecott's labs. A standard was included with the samples in order to check the reproducibility of the method being used. Comparison of standard assays showed that the Coast Eldridge laboratory (using the titration method) was the least reliable. According to R. Simpson, an Independent Qualified Person and author of the previous Technical Report on Galore Creek prepared for Spectrum, the field lab assays compared well with atomic absorption analyses at other labs (Figure 8).

In the 1970's programs, the split half of the core samples was still crushed on site to minus 0.5 inches and split to obtain a 0.75 pound sample. This was further crushed in a cone crusher then placed in Kraft paper bags and shipped by air in locked metal boxes to either the Kennco Exploration Lab in North Vancouver (1972/73), or Chemex Lab (1974) for assay. Assaying for gold and silver was only performed on composite samples (up to 15m) which averaged over 0.4% copper. No information on check assays or quality control from the 1970's drill programs could be located. All coarse rejects from the 1970's were stored on the property.

During the 1990 drill program carried out by Mingold, half of the split core was crushed on site to ¼ inch (6.35 mm) and a 300-325 gram split was taken and shipped to Min-en Laboratory in Smithers for further processing and assaying. For gold analysis, a 30 gram sample split underwent fire assay fusion with analysis by atomic absorption spectroscopy (AAS finish). If greater than 1,000 ppb gold was present then a fire re-assay was carried out. If high copper content was noted in drill logs the sample was directly fire-assayed. If gold content reached or exceeded 3.11 g/t (0.1 oz/ton) then the reject portion of the sample was shipped to the Min-En lab in Vancouver for metallic screening. For this process, the entire reject was pulverized to -102 mesh, recombined with the previous pulp portion and metallic screened for +120 mesh gold. Two 30 gram assays were then done on the -120 mesh fraction and the results averaged. The values from both fractions

(+120/-120) were then mathematically combined to produce a net gold value. Copper and silver analyses were done on a 2-gram sample split from the initial pulp. No check assays were documented and rejects were stored on site.

In 1991 sample preparation was modified on recommendations from Min-En after they undertook a number of tests on coarse reject core samples. The raw core was crushed to 3 mm (1/8") and a 500-gram split taken, pulverized to 95% -120 mesh then rolled and bagged for analysis. The remaining reject was bagged and stored in Smithers. Samples were fire assayed using one assay ton sample weight. For each gold analysis batch of 24 samples they would insert a blank and a standard sample and when the value of the standard fell outside a 95% confidence limit the entire batch was re-run. Internal monitoring of copper assays was routinely conducted on 50 sample batches. The top 10% of all gold assays per page were rechecked and reported in duplicate along with the standard and blank. Every 20th sample was shipped to Eco-Tech laboratories of Kamloops for check assay.. Figure 9 shows the gold check assay comparison for the 571 samples analyzed in 1991.

The check assays showed reasonable correlation for copper and fairly good correlation with gold at grades exceeding 0.25 g/t, although Eco-Tech assays tended to be marginally higher. Gold grades below 0.25 g/t, showed considerable variation

Shipment of core samples from the Galore Creek camp occurred on a hole by hole basis. Rice bags containing bagged core samples each were marked and labeled with the sample numbers and the Vancouver laboratory address. Rice bags were assembled into sling loads for transport by helicopter to the Bob Quinn airstrip and transported by freight forwarders (Banstra) that delivered the samples directly to the laboratory by truck.

All laboratory analysis for the 2003 program was carried out by ALS Chemex of Vancouver. Upon arrival to the lab the samples were logged in a tracking system and the weight was recorded. Each sample was crushed and dried and a 250 gram split pulverized to greater than 85% passing 75 microns. Sample analysis for gold content was conducted by preparing a 30 gram sample by fire assay fusion followed by analysis with atomic absorption spectroscopy (AAS finish). The lower and upper limits for this method were 0.005 ppm and 10 ppm respectively. Overlimit analysis was provided by nitric acid aqua regia digestion of a 0.4 to 2.0 gram sample followed with an AAS finish.

Additional analysis was conducted for 34 elements, including copper and silver by nitric acid aqua regia digestion followed with analysis by analyzed by inductively coupled plasma-atomic emission spectrometry (ICP_AES). The analytical results were corrected for inter-element spectral interferences.

As indicated in Section 12.2, only limited third party analysis of the assay results has been completed. A review of duplicate sample analytical results by this author indicated that differences in mean values for duplicate samples were not statistically significant at a 95% confidence level. A scatter chart of original versus duplicate samples is depicted in Figure 10 while Table 7 summarize the results of t tests for paired duplicate samples from the ALS Chemex results for copper and gold. Where the t statistic exceeds the critical value for a 5% level of significance, the difference is said to be statistically significant at that level. (ie. there is at least a 95% probability that the difference is significant).

While no statistically significant difference was detected at the stated confidence level, the database is relatively small (only 8 holes) and no firm conclusions should be drawn from the results of the analysis. Differences in mean values were relatively high for gold (9%) and a larger database should be available from the 2004 program to provide a more definitive analysis. A more comprehensive check assay program including checks of duplicates by an alternate laboratory is recommended.

All laboratory pulps and rejects from the 2003 program are stored in a company-leased warehouse in Vancouver.

Computer databases for the Galore Creek project have evolved over time, having benefited from both in-house and external services for data management. Up until 1990, data was entered manually. Subsequent to this, assay values were transferred digitally from data files supplied by the assay laboratory.

In 1992 Kennecott conducted an assay database check on 375 assay files representing 7500 samples. The most common mistakes found were typing errors or data that had missed being input into the database. There was also some confusion because of missing prefixes in check samples from ECO Tech Lab. Consequently all previous data was merged into a single database, audited and converted from Imperial to metric units prior to the final resource estimation.

In 2003, prior to the commencement of the drilling program, R. Simpson, an Independent Qualified Person and author of the previous Technical Report on Galore Creek prepared for Spectrum, conducted spot checks on the 1991 re-assay program. Values from the original assay certificates and digital assay files were compared to the digital database used in the MRA resource model and a significant number of discrepancies were discovered, however, none of them were in intervals containing significant mineralization. Out of 1329 samples audited (from 94 assay certificates), 9 were flagged as missing in the database (value=-1) and 14 were apparent data entry errors.

The data is currently maintained as an Microsoft Access ^® database. It is accompanied by a Visual Basic ^® interface to facilitate entry, viewing and export of data. For that data which could be cross-referenced to a assay certificate, the assay intervals and sample numbers have been confirmed digitally. Only about 10% of the copper and 15% of the silver assay certificates from the pre-2003 drilling could be located. Certificate data for gold was more extensive due to the comprehensive re-sampling program undertaken in 1991. Including the 2003 season, approximately 14% of the copper values, 78% of the gold values and 21% of the silver values could be cross-referenced to assay certificates. To address the limited availability of certificate data, assay and sample data from the hand-written logs were also re-entered digitally and compared with the database for verification. In 2004, prior to completion of the update on resource estimates, the author also conducted spot checks on about 15% of the data. As well, 100% of the data was screened automatically by the modeling software (Medsystem ^® /Minesight ^® ) for missing intervals and values outside the normal range of data. A number of typographical errors and inconsistencies in the treatment of unsampled and/or low-grade intervals were discovered and corrected.

Collars from the pre-2003 drilling were surveyed in 1966 and 1991. Nearly all holes in the Central and Southwest Zones have been surveyed but many in peripheral areas have only approximate coordinates. Holes drilled prior to 1991 used only acid dip tests for directional information. The only exceptions were GC218, GC228 and GC235 where a Tropari instrument was utilized. A Sperry Sun single-shot instrument was used in 1991

on angle holes but 24 vertical holes had no down-hole survey readings taken. A recent survey by Eagle Mapping has highlighted a significant discrepancy between the elevations recorded for the drill holes and the newly generated topographic surface. The average difference in elevation between the reported collar elevations and the new surface is approximately -24 metres. As an interim measure, the topographic surface has been lowered 24.2 metres to accommodate the difference. This discrepancy should be investigated and the appropriate corrections undertaken to bring the two surveys in line before the next update on resource estimates.

The Copper Canyon claims cover 2100 acres adjacent (due east) of the Galore Creek claim blocks. Spectrum entered into an agreement with Eagle Plains Resources in September 2003 to earn up to an 80% interest in these claims through a combination of shares, payments and work programs. To the extent that economic mineralization is confirmed on the Copper Canyon claims, it could be developed and processed as part of the Galore Creek operation under the terms of Spectrum's agreement with Eagle Plains.

Under the terms of the agreement, SpectrumGold has an option to acquire a 60% interest in the project from Eagle Plains by completing C$3 million in exploration expenditures over the next 4 years, issuing 400,000 shares of SpectrumGold and making payments totaling C$250,000. SpectrumGold may earn an additional 20% interest in the project for a total of 80% by paying Eagle Plains C$1 million and completing a Feasibility Study on the project by no later than September 2011.

The Copper Canyon property has seen intermittent exploration work from the mid 1950's up through 1991. This historic exploration work at Copper Canyon indicates the presence of widespread gold, silver and copper mineralization similar to the Galore Creek deposit. To date mineralization has been identified in three separate areas on the property. Highlights from drilling on the project in 1990 are summarized in the table below.

16 M INERAL P ROCESSING AND M ETALLURGICAL T ESTING

A total of 96 specific gravity measurements were made on core samples in 1991/1992. This data was sorted by material type and utilized to develop regression curves for use in tonnage determinations for the 2004 resource estimate. This data should be further supplemented by in-situ measurements of bulk density.

Underground development for bulk sampling was carried out in the 1966 and 1967. Most of the work was on the Central Zone where 756 metres of tunnels were driven from one adit. A total of 51 metres of underground development was also conducted in the North Junction Zone. Pilot plant metallurgical testing was carried out in 1967 by Wright Engineers Ltd. on a 50-ton bulk sample from the Central Zone. The head grade of the bulk sample assayed 1.28% copper. The overall metallurgical balance from the pilot plant test was as follows:

In 1992, further bench flotation tests were carried out on drill core from five 1991 holes by Dawson Metallurgical Laboratories in Salt Lake City. The object of this study was to determine the amenability of the composites to a standard flow sheet developed previously and to determine if gold recovery could be significantly improved. The study used 4 composites from the Southwest Zone and two from the Central Zone. It was found that both gold recoveries and copper concentrate grades for the Southwest Zone were lower than those indicated for the Central Zone. This was attributed to the higher pyrite content in the Southwest Zone and the association of at least part of the gold with pyrite. Overall copper and gold recoveries in a copper concentrate grading 25% copper were estimated at 90.3% and 58% respectively based on constant tailing grades of 0.065% copper. Concentrator tail grades for gold also tended to remain fairly constant at 0.137 grams per tonne for the Central Zone and 0.274 grams per tonne for the Southwest Zone. Gold recovery was projected based on head assay and rougher tail residue. A nugget effect was observed in tests from many of the higher-grade composites. Gold recoveries were not optimized as part of these studies.

It was also reported that several composites were not upgraded to 25% copper in concentrate after two stages of cleaning with regrinding.

During 2003 a metallurgical scoping test program was completed on four individual composites of drill core selected by NovaGold. The test work was carried out by G&T Laboratories Ltd. in Kelowna. The drill core, from the 2003 drilling campaign, have been identified as DDH 437, DDH-438, DDH-441 and DDH-442. The DDH 437 sample is from the Southwest Zone while the remainder are from the Central Zone. These samples were selected to investigate the metallurgy of higher-grade mineralization.

The Southwest Zone sample was the highest in grade and contained the most pyrite. Chalcopyrite was the dominant copper mineral in all the samples except DDH 441 where bornite occurrence was significant.

High copper and gold recoveries and copper concentrate grades were achieved at relatively coarse primary grinds using standard grinding and flotation procedures and reagents. Rougher concentrates upgraded well following a regrind. The recoveries and concentrate grades were marginally higher than those achieved in the 1966 and 1992 flotation test programs. Copper, gold and silver recoveries attained were of the order of 90%, 70% and 80% respectively for the samples tested. A summary of the 2003 test results is Table 13. Highlights from the previous years' testing are provided in Table 10 through Table 12.

Gravity concentration of ground samples gave variable and low recovery of gold. The gold grains were generally fine, indicating that gravity concentration might not be effective.

The key observations from the 2003 test work specific to the four samples are as follows:

17 M INERAL R ESOURCE AND M INERAL R ESERVE E STIMATES

The current mineral resource inventory for the Galore Creek Project contains estimates for the Central and Southwest Zones only. Other mineral occurrences exist within the property boundaries, however, except for the Central and North Junction Zones, which received additional focus by Hudson Bay during the 1970's, only the above-mentioned areas have been delineated sufficiently to make estimates of mineral resources. Both the Central and Southwest Zones have benefited from additional exploration activity and revised and/or updated interpretations of geological and structural data since the early 1990's. The update for these zones follows the recently completed 8-hole, 2,800 metre drilling program designed to verify previous results and better understand deposit variability, zonation and controls for mineralization.

Table 14 provides summaries of the current resource estimates for the Central and Southwest Zones. Resources have been reported using equivalent copper values which have been calculated by estimating the net smelter return for gold and silver and back-calculating what the copper grade would have to be for the revenue to be generated from copper alone. A brief description of the 2 resource areas is provided below while details on each zone as well as further information on equivalent value calculations can be found in the following subsections.

The Central Zone, which strikes north-northeast, is modeled as two primary structural domains separated by a moderate west-dipping fault along which significant displacement has occurred subsequent to the mineralizing events. A number of post-mineral and weakly mineralized intrusives have also been modeled as sub-domains

within the two. The Southwest Zone has been modeled as a single domain striking east-southeast and dipping 60 degrees south-southwest.

A total of 326 drill holes were used in the resource estimate, with approximately 21,700 assays (copper) averaging 2.94 metres in length contained within the reported zones. These were composited in 6-meter intervals down hole within each zone to form approximately 12,100 composites averaging 5.76 metres in length.

The grades were estimated by ordinary kriging using unique 3D anisotropic variograms for each grade and zone. A minimum of 6 composites from at least 2 drill holes were used to estimate grades in each block, with a maximum of 3 composites per hole and 18 total. Only those composites with the matching codes were used in the interpolation of blocks within each zone of the model. The block size was 25mx25mx12m high. The average distance from each block centre to the closest composite (adjusted to the long axis based on anisotropy) was 48 metres for indicated and 122 metres for inferred.

The mineral resource estimates for the Central and Southwest Zones are based primarily on information from core drilling, supplemented in part outcrop data and underground sampling. A total of 326 core holes have been completed in the area of the Central and Southwest Zone to date, including the 2003 program (8 core holes). The resource model was restricted to the use of sample data from the 326 core holes. Outcrop and underground data was used to assist in the 3D modeling of the geological and structural boundaries however assay data was not utilized for the resource estimate.

The Central Zone deposit is drilled from surface (630 to 850 m. elevation) down to about 350 to 500 metres elevation on an approximate spacing of 50 to 100 metres north-south and 50 metres east-west. Below this, as well as on the extremities, the drill density is much less. The deepest hole is GC-0224, which is drilled to about 160 metres elevation. The majority of the holes have be drilled close to vertical however several, including those drilled in 2003, have been drilled at angles to the east so as to intersect the structure at close to right angles as possible. A number of holes have also been drilled at angles to the west. Overall, drilling has covered an area about 2,500 metres north-south and 1,000 metres east-west.

The Southwest Zone, located about 600 metres to the southwest of the Central Zone at surface elevations from 850 to 1050 metres, is drilled at a much lower density. The majority of the holes have been angle drilled toward the north at a spacing of about 100 metres east-west and 100 to 200 metres north-south. The drilling covers an area of about 800 metres east-west and 500 metres north-south.

Further detail on drilling can be found in Section 11.

The combined database for the two zones (core holes only) consists of 21,700 sample intervals for copper averaging 2.94 metres in length. Approximately 86% of the samples fall within the interpolation boundaries of the Central Zone while the remainder are located within the Southwest Zone boundaries. A number of the early gold and silver assays were derived from composite samples taken over intervals of 15 to 30 metres subsequent to the initial sampling for copper. Some of these intervals were re-sampled over shorter intervals during the early 1990's. Overall, the database contains 21,200 assays for gold and 16,000 for silver within the reported zones. The gold and silver database is more complete within the Southwest Zone due to the emphasis on gold. Assay data distributions for copper, gold and silver are positively skewed and approximate normal distributions when transformed to logarithmic values (lognormal distributions). Brief statistical summaries are provided in Table 15 through Table 17.

The Central Zone is divided structurally into two main mineralized zones by a moderate west-dipping north-northeast trending fault known as the East Fault. It has been estimated that displacement of up to 200 metres has occurred along this fault subsequent to the main mineralizing events. In the north end of the deposit the fault may be better characterized as a fault zone up to 60 metres wide with variable displacement across multiple surfaces. For the purpose of the resource estimate, the Central Zone has been modeled as two primary structural domains (hangingwall and footwall zones) separated by the interpreted 3D location of the fault. These domains have been further subdivided by modeling the more important late and post-mineral intrusives as separate lithological sub-domains (I9 and I11) within the two. Additional detail is available with respect to structure, lithology and alteration, however it is believed that further subdivision based on these units would have resulted in a number of smaller disjointed domains, making the modeling efforts more difficult and overly complex. Future models may incorporate some of these features as well as separate modeling of distinct gold-rich zones located within the north, central and southern areas of the Central Zone.

The Southwest Zone has been modeled as a single domain striking east-southeast and dipping 60 degrees south-southwest. 3D solid models of a locally mineralized breccia unit (Southwest Breccia) as well as grade shells (SW Footwall and SW Main) representing the approximate boundaries beyond which mineralization in drill samples fall below 0.5% copper and/or 0.5 grams gold per tonne were also constructed but only used as aids to facilitate variogram modeling and interpolation strategies.

The various domains within the Central and Southwest Zones were constrained horizontally by interpolation boundaries set at reasonable proximity (200m) to the peripheries of the drilling; on top by the overburden/bedrock contact surface; and at depth by a horizontal plane set at 40 metres elevation. To aid in specific gravity determinations the interface between broken and solid rock, known as the "stick-rock" surface was also modeled, but not used for composite selection nor to limit or control grade interpolation.

In order to reduce the influence of statistically anomalous sample data on resource calculations a number of higher-grade assay values were capped at levels determined through examination of probability distribution data (Appendix 23.1) . Table 18 provides a summary of capping levels applied to the Galore Creek data. All data above these levels were set back to the chosen levels prior to compositing.

Significant inflexion points were evident in both copper and silver curves, indicating that several sample populations may exist within the data. Distribution curves for gold were relatively straight however, as with copper and silver, breaks were also noted at the upper end of the of the distributions curves. Assay values were capped at levels approximating the start of these upper breaks in the curves. As indicated previously, assay data for all

three metals are positively skewed (mean to the right of mode) and approximate normal distributions curves when transformed to logarithmic values. The capping levels represent points on the distribution curves at distances ranging from 2.1 to 4 standard deviations from the mean values after natural log transformations have been applied. In all, 8 copper, 6 gold and 6 silver values, representing only0.03% of the assay database, were capped.

17.6 Composites

Assays intervals have been composited down the hole from collar to final depth in length-weighted averages each representing 6.0 metres of continuous sample. Other composite lengths were examined, however the selected length is believed to represent the best compromise between shorter lengths where variability was too high to expect reasonable prediction of grade at a distance and longer composites, which can result in excessive smoothing of values to impart a reasonable degree of block model resolution. Assays intervals contained within the structural and lithological domains discussed earlier have been composited separately and assigned unique codes to differentiate them from the surrounding material. A summary of composite statistics can be found in Table 19. As expected, the composite populations within the mineralized zones also exhibit lognormal distributions.

In total, 6,248 intervals within the zones of interest were missing silver assays. In comparison, the database only contained about 503 intervals missing copper and 993 intervals missing gold. All missing assay intervals were assigned values of zero grade before compositing. This was necessary because the database did not differentiate between intervals not sampled because it was believed to be low grade or waste and intervals not sampled for other reasons such as poor drilling recovery. As can be observed from the differences between assay (4.9 g/t) and composite grades (3.2 g/t), this may have impacted the silver grades significantly, however it cannot be determined without further sampling whether this represents a downward bias in the estimates but this is highly probable.

Because all drilling has been collared at surface, composites begin at the upper contact and end at the lower boundary of each zone. Since it is rare that a sample interval falling within an interpreted zone is an exact multiple of the composite length, a number of composites were generated along the lower boundary of each domain which were less than the full composite length. This results in a number of short composites or "orphans" which may present a bias when used to estimate the grade model. Since the zones at Galore Creek are relatively wide, it is believed that the influence of these orphans is minimal. The average composite length is 5.76 metres however the impact is most apparent in the post/weak mineral intrusive zone where the average length is only 5.5 metres. A number of comparisons of simple arithmetic averages with length-weighted averages of composite grades were completed, with results always indicating that the use of "orphans" represented a slightly conservative bias (lower by 1.1% to 1.7% relative) with respect to grade. This means that the short composites are lower in grade than the full length ones. The difference would decrease as higher cutoffs were applied.

In order to assess the spatial relationship between composite sample points, a series of 3D relative semi-variograms (correlograms) were generated for those composites that fell within the interpreted zone boundaries. Plots of sample variograms for copper, gold and silver composites can be found in Appendix 23.2. A summary of modeling results is provided in Table 20.

These variograms formed the basis for subsequent grade interpolations as well as provide assistance in development of guidelines for resource classifications and future drilling. The post/weak mineral intrusives were not modeled due to the paucity of data. The Central Footwall Zone models were used for these zones in grade interpolations. Variograms for the Central Hangingwall Zone were the best formed due to the high density of data and credible models were readily developed. Interpretation of variogram models for the Southwest Zone was more difficult due to the lower drilling density. Most variograms demonstrated long ranges along the major and intermediate axes although nuggets, while still relatively low, were significantly higher in the Southwest Zone.

Essentially, a variogram examines the increase in variance between samples as a function of distance in a given orientation. The distance beyond which the relationship between samples ceases to be anything but random is known as the range. Based on the copper variograms developed for Galore Creek, this distance for the long axis is approximately 190 to 210 metres in the Central Zone and 180 metres in the Southwest Zone. In general, the target drill spacing should be about one-half to two-thirds that of the range in order to maintain some interrelationship between samples used to make grade estimates.

For the Central Zone, this would imply a final staggered pattern of about 100 to 130 metres along strike and 45 to 60 metres across the zone. In most areas of the Central Zone, the drill density already exceeds this level. At depth and along the extremities, this would have to be supplemented by more holes on the existing drill sections. For the Southwest Zone, the drilling program should be designed to intercept the dipping structure on a 90 to 120 metre (dip direction) by 70 to 100 metre (strike) staggered grid in the plane of the structure. This can be achieved by holes collared on surface 45 to 60

metres apart north-south and 70 to 100 metres east-west and drilled -60 degrees toward N30E..

None of these guidelines are intended to take the place of quality geologic information that may suggest alternate delineation criteria.

17.9 Block Model and Grade Estimation Procedures

The current resource estimate for Galore Creek is based on a 3-D computer block model oriented along grid north. Block dimensions are 25 metres north-south (N-S), 25 metres east-west (E-W) and 12 metres vertical. The selected block dimensions are believed to represent a reasonable compromise between deposit geometry and sample density. It is believed that a smaller block size would impart a degree of resolution not possible with the current sample density, while larger dimensions would not suit the limited thickness and variability inherent with the geometry of the zones. This is particularly important in the Central Zone, where the complex geometry of the internal zones of post/weak mineral intrusives is best modeled with smaller blocks. The block height and width are also convenient dimensions for subsequent pit optimization and planning efforts.

Prior to interpolation of grade values into each block, 3-D solid models were constructed for each of the interpreted structural, lithological and interpolation domains discussed in Section 17.4. Each block or portion thereof which fell within the modeled solids was assigned an "zone" code and percentage based on the proportion of the block contained within each domain or zone. Blocks completely internal or external to the zone were assigned a block percentage of 100 or 0 respectively. Blocks were also coded for the percentage of the block that resided below the topographic, bedrock (broken) and intact rock surfaces. In the Central Zone area it was possible for a block to reside on the interface between the two structural domains (HW & FW) as well as contain a portion of the post/weak mineral intrusive. To address this possibility, a second set of zone codes, block percentages and grade variables were carried for each block. The primary zone code was assigned on the basis of majority block ownership between the hangingwall and footwall units and the percentage determined by the combined content for the two domains. If the block also contained a portion of the post/weak mineral intrusives, the secondary block code was also assigned accordingly and block percentage determined based on the proportion of the block located within the intrusives.

Composites were also assigned a matching zone code, however this was determined on the basis of where the midpoint of the composite was located. It should be noted that because the most recent topographic survey did not match drill hole collar elevations exactly, the surface was adjusted (lowered by 24.2m) to account for the average difference in elevation between the older collar surveys and the new topographic survey. Consequently, elevations for individual collars as well as down hole positions of assays, composites and logged lithological units may be slightly above or below the adjusted surfaces. To insure that the drill hole composite codes matched the lithological units in the drill logs, the composites were coded for the bedrock and surface interfaces according

to the original surveys. Prior to completing any future updates, the discrepancies between the new topographic surface and the old surveys should be investigated further and adjustments made to the collar surveys and/or the new surface so both match. Table 21 provides a list of the zone codes utilized for blocks and composites.

The shorter search utilized a maximum search distance from each block centroid to any given matching grade composite within the zone being interpolated equal to two-thirds the modeled copper variogram ranges along each axis. The maximum distance along the major and intermediate axes to accept data was doubled in the broader search, although the distance to the closest sample was limited to 100% of the variogram range. The distance along the minor axis, which is usually cross-structure, was kept at two-thirds the variogram range in the broader search to prevent interpolation of block grades too far below the bottom of the drill holes. No values were assigned to blocks located at a distance greater than the full variogram range from the closest composite sample or when the search failed to find a second drill hole with matching composites at a distance less than four-thirds the range from the block centroid.

Search parameters used in the final grade model are summarized in Table 22. A minimum of 6 composites was used to interpolate block grades while the upper limit was set at 18. The maximum number of composites per drill hole was set at 3, insuring that at least 2 drill holes and potentially 6 or 7 maximum were used to interpolate grades into each block. The former was set in order to prevent undue influence from a single drill hole in areas were drilling was sparse, while the latter limited model smear, a feature

inherent to all weighting methods. Placing a limit on the minimum number of composites can lead to situations where blocks between drill holes are interpolated while blocks adjacent to and within very close proximity of the same intercepts are left uninterpolated. The model was reviewed in detail subsequent to final interpolation to ensure that this was not the case with Galore Creek. In the denser drilled areas of the Central Zone, the search envelope was sufficient in most cases to incorporate the maximum number of composites allowable in the grade interpolations.

Copper, gold and silver grades were interpolated by ordinary kriging, utilizing the variogram models discussed in Section 17.7. While the weighting of samples was based on individual variograms for each metal and zone, the search logic was based on the copper variograms to insure that grades were assigned to each of the three metals if a block was interpolated. Where a block which fell within the interpolation boundaries of a zone also contained a portion of the post/weak mineral intrusives, grades were estimated for both based on their corresponding composites. Other items stored or interpolated included kriging variance, distance to the closest composite, number of composites and number of drill holes used in each interpolation. Only those blocks contained at least partly within the zone boundaries received a grade estimate and only those composites with matching zone codes were used in the interpolation.

Estimates of indicated and inferred resources resulting from the application of the procedures outlined in the previous sections are contained in the following tables. Table 23 and Table 24 show a breakdown of the resource by zone while Table 20 and Table 21 provide a tonnage/grade distribution at increasing cutoff values. Further detail can be found in Appendix 23.3.

Cu Eq %= Cu% + (NSR Au US$ + NSR Ag US$)/(NSR Cu US$ per %Cu)

Assumed process recoveries and smelter terms used in these calculations are summarized in Table 25 and Table 26.

Although a distribution of tonnage and grade at increasing equivalent cutoff values is provided in Table 27 and Table 28, caution is advised with respect to the use of this information. At the selected cutoff of 0.5% equivalent copper, the resource is, for the most part, continuous and contiguous within the limits of the search parameters set. In applying higher cutoff values the reported blocks can become separate and disperse and may not represent an inventory that is potentially mineable.

In the reporting of blocks residing partly within the boundary, a block percentage is applied to the block tonnes. Tonnage calculations are based on a specific gravity of 2.49 for broken rock and a linear formula for massive rock based on the copper content. The latter was derived from regression analyses of a database of density measurements compiled in the early 1990's.

SG=2.62+0.0883*Cu%; minimum 2.65, maximum 2.75

For example:

Based on these calculations, a 25x25x12 metre block with a block percentage of 100 would contain 18,675 tonnes (2.49 SG) for broken rock and 19,875 (2.65 SG) to 20,625 tonnes (2.75 SG) for massive rock.

While a direct comparison with previous historic estimates is not possible, Table 29 and Table 30 provide estimates of current grades and tonnes using the net copper equivalent cutoff values as calculated by Mine Reserves Associates, Inc. (MRA) in their 1992 estimates (see Section 6.1 Historical Resource Estimates). Table 29, which is stated at a net equivalent copper cutoff of 0.25%, is roughly equal to MRA's estimate referred to in Section 6.1 in terms of overall tonnage but grades in the indicated category are higher in the current estimate. This is, in part, due to the smaller composite and block size, which imparts a higher degree of resolution than the previous estimates. Gold grades in the inferred category are lower as the Southwest Zone represents a smaller proportion of the total than previous estimates.

Table 30 provides estimates of current resources using a 0.4% "MRA" equivalent, which approximates the tonnes contained in the Kennecott "in-pit" resource also cited in Section 6.1. The current inventory is about 7% higher in contained copper than the "Kennecott" resource for the indicated category however the current estimates are not restricted to the those resources within the pit outline.

The distribution of resources between the indicated and inferred categories is summarized in Table 27 and Table 28 while further detail can be found in Appendix 23.3 . It is believed that this classification conforms well to the CIM Standards on Mineral Resources and Reserves. The division between indicated and inferred resources is based on the observed geologic continuity of the Galore Creek mineral zones coupled with variogram data which demonstrated that a relationship between grade samples exists up to distances exceeding 200 metres along the major axis. Those resources classified as indicated generally fall within areas of the deposit where drill spacing meets the criteria outlined in Table 31. This coincides with a maximum distance from each block centroid to any given grade composite within the zone being interpolated equal to two-thirds the variogram ranges for copper along each axis. For the inferred resource, these distances have been increased to the full variogram range for the closest sample although data has been accepted up to double the distances used for indicated along the major and intermediate axes. The range along the minor axis, which is usually cross-structure, has been left the same as for indicated to prevent inclusion of blocks too far below the bottom of the drill holes. These parameters are essentially identical to the multiple pass interpolation search strategy discussed in Section 17.9. Those blocks interpolated during the more-restricted second search have been classified as indicated while those blocks outside this area which were interpolated during the initial larger search have been classified as inferred.

Consequently, the indicated category is co-incident with that area of the deposit that is drilled on a spacing equal to or less than double the search distances outlined in Table 31 and incorporates those blocks located along the periphery of the drilling which were assigned grade values at distances equal to or less than the search range (two-thirds the variogram range). The inferred category contains primarily fringe blocks that are located equal to or less than the stipulated search distances from the closest and farthest samples but greater than the maximum allowed for indicated. This is equivalent to a drill spacing twice that for indicated along the major and intermediate axis however in practice, most blocks classified as inferred are located external to the drilled area within the maximum search ellipsoid rather than between samples.

Although the resource classification is based primarily on a statistical accumulation using composite distance as a cut-off between indicated and inferred, the distribution of indicated and inferred blocks was carefully reviewed to ensure that this procedure did not result in the mixing of the two categories within a given area. 3D shells of the boundaries for indicated and inferred were generated from block data and the shells edited to exclude isolated blocks of indicated and/or inferred outside the margins of the main envelopes as well as repatriate a few isolated blocks of inferred located inside indicated shells. For the most part, the deposit is drilled on a spacing less than the maximum set for the indicated category and as such, the area within the bounds of the drilling is classified mostly as indicated, while the margins have been designated as inferred. Correspondingly, those areas of the model classified as indicated utilized data from which at least some statistical interrelationship existed between sample points. Outside or below this boundary, data is too widely spaced to assume continuity of grade and structure. The average distance from each block centre to the closest composite (adjusted to the long axis based on anisotropy) was 48 metres for indicated and 122 metres for inferred.

None of the mineral resource is classified as measured. It is believed that in addition to reducing the overall drill spacing to that recommended in Section 17.8, more extensive sampling of precious metals and confirmation through more bulk sample testing will be required to warrant upgrading any portion of the resource to measured. There has been no mining experience with the deposit and it would be premature to assume that the level of confidence associated with "measured" can be applied to these mineral resources without further work.

While necessary to formulate detailed mine plans for production, it is not advised that any additional drilling be completed in the area of the indicated resource at this time to upgrade it to measured. A limited program designed to upgrade inferred resources along the peripheries of the existing zones as well as determine their full vertical and lateral extent could be considered as an addendum to the recently completed program. In particular, the full down-plunge and lateral extent of the Southwest Zone has yet to be defined and several holes would probably help in this regard. Efforts to advance the project would be better served by investigating the identified mineral potential along strike and down dip from known mineralization and in other areas within close proximity.

18 O THER R ELEVANT D ATA AND I NFORMATION

During 1967 a series of geotechnical studies on overburden, bedrock, massive (intact) rock and reservoir slope stability were completed by Golder, Brawner & Associates (GBA). In summary, the findings indicated initial design parameters for pit slopes in the range of 30 to 34 degrees for overburden, 40 degrees for bedrock and 42 to 56 degrees for massive rock In areas where the structure dipped into the potential pit area, such as the moderate west-dipping post-mineral intrusives, slopes would have to be designed so as to not to exceed the indicated dip angles.

In 1991 a proposal for a 7-hole triple-tube oriented core drilling program was made for the potential pit area, but never implemented. However a geotechnical study was carried out by T. Heah the same year. A total of 381 structural measurements were collected from cliff faces in the northwestern part of the Central Zone. Particular attention was paid to fractures steeper than 30° (the assumed dry friction angle), which dip easterly. In addition, exposed, easterly-dipping fractures were also noted. Lastly, the presence of faults, possible release surfaces, and major groundwater seeps were recorded. Rock hardness was also tested.

Assuming dry conditions, a friction angle of 30° and totally planar joint surfaces, it was estimated that the maximum stable angle for a pit slope trending north-south was between 28° and 40°. Probable hazards mentioned included toppling failure along steep, northerly and northeast-trending fractures and sub-horizontal release surfaces. Rock fall hazard was considered to be moderate to high.

Norecol Environmental Consultants Ltd. was engaged to conduct baseline environmental studies in 1991/1992. The program involved three site visits to the property by Norecol, water quality sampling, climate and hydrology data collection, wildlife observations and acid base accounting. Camp personnel collected baseline hydrology information at West Fork Creek and climate information at the camp during the exploration program (June through October 1991).

Galore Creek acid-base accounting was conducted on the coarse reject samples provided by Min-En in Smithers. The Galore Creek rocks contain a significant amount of gypsum which shows up as sulphur or sulphate. As a result, total sulphur is very misleading as an indicator of potential acidity because gypsum is not acid generating. Therefore, total sulphur and sulphur in the form of sulphide were used to estimate potential acidity. Sulphur in the form of sulphide is expected to give a better indication of acid generation potential.

The NPR(s) (ratio of neutralization potential to potential acidity due to sulphur as sulphide minerals) is preferred for predicted acid generation potential. If NPR(s) are less than 1, the rock is a potential acid producer. If NPR(s) are greater than 1 and less than 3, the rock is a marginal potential acid producer with the region between 1 and 2 considered the most uncertain.

The syenite megaporphyry from the central zone appears to be a potential acid consumer. Sulphide content was low (<0.4% sulphur). NPR(s) was 2.68 in one sample, but greater than 3 in the remaining samples. One sample of syenite porphyry was similar to the megaporphyry.

Results for other units suggest a greater and varied potential for net acid generation (several examples of NPR(s)<2). Where copper grade exceeded 0.4% (potential ore), the NPR(s) tended to be greater than 1 (3 out of 4 samples), implying a low potential for acid generation. The highest acid generating potential (NPR(s)<1) appears to be in the poorly mineralized volcanic rocks.

In early 2004, SpectrumGold retained Rescan Environmental Consultants (Vancouver, BC) to begin a comprehensive program of baseline environmental studies for the Galore Creek site, regional waterways, and potential access corridors. The scope of work is extensive and will include the characterization of the acid generation and metals leaching characteristics of the rock in and around the Central and Southwest Zones. In addition, SpectrumGold has initiated its Environmental Assessment Process in British Columbia and is currently consulting with Provincial and Federal agencies regarding the Terms of Reference for its Environmental Assessment report.

18.3 Proposed Power and Access Routes

Road access and power will be key design components in mine planning for the Galore Creek deposit. Numerous possible access routes and power scenarios have been proposed for the project in previous studies. These included on-site power, an intertie to the BC Hydro grid and a several different approaches using road or tunnel access. In Kennco's last studies, road access and power accounted for just over half of the total capital cost for the project of US$387 million for a mine processing 29,000 tonnes per day.

An Economic Assessment Report focused on the development of the Central and Southwest Zones is expected to be complete by mid-2004. This study will provide updated development concepts, capital and operating costs, and will consider two access alternatives. Both options contemplate an intertie with the BC Hydro grid and concentrate shipment via the BC port of Stewart. The Northernmost access corridor being studied would extend from the Galore Creek valley almost due east along the More Creek drainage to intersect BC Highway 37 near Bob Quinn. A new road distance of approximately 80 kilometres as well as a tunnel for ore conveyance and access into the

Galore Valley would have to be constructed should this option be selected. The Southern access corridor being examined would commence at the Galore Creek Valley, parallel the Stikine River south to the mouth of the Iskut River, and follow the Iskut in an easterly direction to connect with the existing Eskay Creek Mine road. This alternative would require approximately 150 kilometres of new road construction. Extensive field environmental and geotechnical investigations for both access options is planned for the 2004 field season.

19 I NTERPRETATION AND C ONCLUSIONS

Over the years, a number of resource estimates for the Galore Creek deposits have been produced by various authors. More recent studies have focussed on exploiting resources in the Central and Southwest Zones by surface methods, with production rates ranging from 30,000 to 60,000 tonnes per day. Commensurate with this emphasis, a number of historic mineral resource estimates exist, the most recent of which was based on a model completed for the Central and Southwest Zones in 1992 by Mine Reserve Associates, Inc. (MRA) on the behalf of Kennecott.

Updated estimates for mineral resources contained within the Central and Southwest Zones were completed by Hatch/AMCL in collaboration with NovaGold/Spectrum in April 2004. These estimates incorporate information from the most recent round of drilling completed in October 2003 as well as having the benefit of a revised interpretation. The following interpretation and conclusions are made regarding these updated estimates:

This report, 3PN01r1 - Update on Resources, Galore Creek Project, British Columbia, has been prepared by:

A SSOCIATED M INING C ONSULTANTS L TD .

Peter A. Lacroix, P. Eng.
Principal Mining Consultant

21 R EFERENCES

Aiken, J.D. (1959); Atlin Map-area, British Columbia, Geological Survey of Canada , Memoir 307, 89 Pages.

Allen, D.G. (1966); Mineralogy of Stikine Copper's Galore Creek Deposits, Unpublished MSc Thesis UBC, 38 Pages.

Allen, D.G., Panteleyev, A. and Armstrong, A.T. (1976); Galore Creek, in Porphyry Deposits of The Canadian Cordillera , A Sutherland Brown, Editor, CIM Special Volume 15, Pages 402-414.

Anderson, R.G. (1984) ; Late Triassic and Jurassic Magmatism Along The Stikine Arch and the Geology of the Stikine Batholith, North Central British Columbia, Geological Survey of Canada , Paper 84-1A, Pages 67-73 Barr, D.A. (1966); The Galore Creek Copper Deposits, CIM Bulletin , Vol.59, Pages 841-853

Barr, D.A. (1998); Galore Creek Access Routes, Kennecott Corporation internal report

Beane, R.E. (1982); Hydrothermal Alteration in Silicate Rocks: Southwestern North America, in Advances in Geology of the Porphyry Copper Deposits , Southwestern North America, Spencer R. Titley, Editor, Pages 117-138

Best, M.G. (1982); Igneous and Metamorphic Petrology Pages 37-38

Bradshaw, B.A. (1968); 1966-1967 Geology and Ore Reserves, Galore Creek Project, Liard M.D., B.C., Internal Report .

de Rosen-Spence, A. (1985); Shoshonites and Associated Rock of Central British Columbia, B.C. Ministry of Mines and Petroleum Resources , Paper 1985-1, Pages 426-442.

Einaudi, M.T. (1982); General Features and Origin of Skarns Associated With Porphyry Copper Plutons: Southwestern North America, in Advances in Geology of the Porphyry Copper Deposits , Southwestern North America, Spencer R. Titley, Editor, Pages 185-210

Enns, S.G., Thompson, J.F.H, Stanley, C.R. and Yarrow, E.W (1995); The Galore Creek porphyry copper-gold deposits, Northwestern British Columbia, in 'Porphyry Copper Deposits of the Northern Cordillera' . ed. by Schroeter, T., Canadian Institute of Mining and Metallurgy Special Volume 46, Paper No. 46, Pages 630-644.

Fluor Daniel Wright Ltd.(1994); Project Review Galore Creek Property Oct. 1994, Kennecott Corporation internal report

Golder, Brawner & Associates; February to October, 1967; Preliminary Stability Appraisal: Overburden; Bedrock; Massive Rock; Resevoir . Reports (4) to Kennco (Stikine) Limited

Heah, T. (1991); 1991 Preliminary Surface geotechnical data, Kennecott Corporation internal report

Inman, J. (1992); Geophysics at Galore Creek, Internal Kennecott Memorandum

McAusland, J.H., (1967); Underground Development and sampling at Galore Creek during the fall and winter of 1966-67, Kennecott Corporation internal report

Mortensen, J.K., Ghosh, D. and Ferri, F., 1995. U-Pb age constraints of intrusive rocks associated with copper-gold porphyry deposits in the Canadian Cordillera in Porphyry Copper (± Au) Deposits of the Alkalic Suite – Paper 46 , CIM Special Volume 46, Pages 142-158.

Jeffery, W.G. (1965); Galore Creek (Stikine Copper Limited), B.C. Minister of Mines, Annual Report , 1964, Pages 19-40.

Kerr, F.A. (1948); Lower Stikine and Western Iskut River Areas, B.C.; Geological Survey of Canada , Memoir 246.

LeMaitre, R.W. (1976); The Chemical Variability of Some Common Igneous Rocks, Journal of Petrology Vol.17.

Logan, J.M. and V.M. Koyanagi (1988); Geology and Mineral Deposits of The Galore Creek Area, North Western B.C. (104G/3,4), Pages 269-283, British Columbia Ministry of Energy, Mines and Petroleum Resources , Geological Fieldwork, 1988 Paper 1989-1.

Logan, J.M., Victor, M., Koyanagi and Rhys (1989); Geology and Mineral Occurrences of The Galore Creek Area, NTS 104G/03 and 04, Province of British Columbia, Ministry of Energy, Mines and Petroleum Resources , Mineral Resources Division, Geological Survey Branch, Open File 1989-8 (2 sheets).

Malmqvist, L. (1978); Some Applications of the IP Technique for Different Geophysical Prospecting Purposes, Geophysical Prospecting , Vol.26 Pages 97-121

Mine Reserve Associates, Inc. (1992), Pre-feasibility Mining Evaluation Galore Creek Project. Kennecott Corporation internal report

Monger, J.W.H.(1970); Upper Palaeozoic Rocks of Western Cordillera and Their Bearing on Cordillera Evolution. Canadian Journal of Earth Sciences , Vol. 14, Pages 1832-1859.

Moroney, M.J.(1969) Facts From Figures, Page 422

Panteleyev, A (1975) Galore Creek Map-Area, B.C. Dept. of Mines and Petroleum Resources , Geological Field Work, 1975, In press. Sillitoe, R.H.(1985); Ore Related Breccias in Volcanoplutonic Arcs., Economic Geology , Vol 80, 1985, Pages 1467-1514.

Sillitoe, R.H.(1991a); Geological Reassessment of the Galore Creek Porphyry copper-gold Deposit, British Columbia, A report prepared for Kennecott Canada Inc .

Sillitoe, R.H.(1991b); Further Comments on Geology and Exploration of the Galore Creek copper-gold Deposit, British Columbia, A report Prepared for Kennecott Canada Inc.

Souther, J.G.(1972); Telegraph Creek Map Area, British Columbia, Geological Survey of Canada , Paper 71-44, 38 Pages.

Tipper, H.W., Richard, T.A.(1976); Jurassic Stratigraphy and History of North-Central British Columbia, Geological Survey of Canada , Bulletin 270, 73 Pages.

White, W.H., Harakal, J.E. and Carter, N.C. (1968) Potassium-Argon Ages of Some Ore Deposits in British Columbia, CIM Bulletin , Vol. 61, Pages 1326-1334.

Winn, G.C. (1988); Titanium Geophysics: The Application of Induced Polarization to Sea-Floor Mineral Exploration, Geophysics , Vol.53, Pages 386-401.

Wright Engineers Limited (1974), Stikine Copper Project - Technical and Economic Study for Hudson Bay Mining and Smelting Co. Limited, Kennecott Corporation internal report

Yarrow, E.W., Enns, S.G. (April 1992) Progress Report 1991 Galore Creek Project,

Kennecott Corporation internal report

Zurowski, M.T. (1988); Gold Potential of the Galore Creek Deposit, Stikine River, Liard M.D.,B.C., Kennecott Corporation internal report

22 A DDITIONAL R EQUIREMENTS FOR T ECHNICAL R EPORTS ON D EVELOPMENT P ROPERTIES AND P RODUCTION P ROPERTIES

Although advanced technical and economic studies have been carried out in the past (Wright Engineers, 1974; MRA, 1992; Fluor Daniel, 1994), the project is currently considered to be in an advanced exploration stage.

23   A PPENDICES

23.1 Appendix – Probability Distribution Data

23.2 Appendix – Sample Variogram Models