Directors:

A J Wright (Chairman), I D Cockerill

†

(Chief Executive Officer), K Ansah

#

, A Grigorian°, N J Holland

†

(Chief Financial Officer), J G

Hopwood,
G Marcus, J M McMahon

†

, D M J Ncube, R L Pennant-Rea

†

, P J Ryan, T M G Sexwale, C I von Christierson

†

British,

#

Ghanaian,

°

Russian

Corporate Secretary:

C Farrel

Gold Fields Limited

Reg. 1968/004880/06
24 St Andrews Road
Parktown, 2193

Postnet Suite 252
Private Bag X30500
Houghton, 2041
South Africa

Tel +27 11 644-2400
Dir +27 11 644-2502
Fax +27 11 484-0590

www.goldfields.co.za

EXECUTIVE

Karl Hiller
Branch Chief
United States Securities and Exchange Commission
Division of Corporation Finance
100 F Street, N.E.
Washington, D.C. 20549-7010
United States of America

31 July 2007

By EDGAR

Dear Mr. Hiller:

Re: Gold Fields Limited

Form 20-F for the year ended June 30, 2006

Filed on November 24, 2006

File No. 001-31318

We refer to the comment letter (the “Comment Letter”) dated July 3, 2007 of the staff (“Staff”) of the
Division of Corporation Finance of the Securities and Exchange Commission (the “Commission”) on
the above referenced Form 20-F (the “2006 Form 20-F”) of Gold Fields Limited (“Gold Fields” or the
“Company”). For your convenience, each comment is repeated below, prior to the response.
Management has discussed the content of this letter (the “Response Letter”) with its auditors.

Form 20-F for the Fiscal Year Ended June 30, 2006

Results of Operations – Years Ended June 30, 2006 and 2005, page 145

Depreciation and Amortization, page 150

1.

We note your June 12, 2007 letter response to our prior comment 2, indicating that you will be
filing an amendment to update your reserve information through your fiscal year end of June
30, 2006. As conveyed to your external legal counsel Jennifer Schneck in a telephone call on
June 14, 2007, we will need your proposed mineral reserve disclosures as of June 30, 2006,
mentioned in your response, including a description of the methodology applied to calculate
your reserve changes and encompassing the reasons for the changes over the entire fiscal
period, in order to continue with our review. Please submit a draft amendment to your Form 20-
F, marked to show all changes necessary to comply with the comments issued during this
review. We may have further comments after reviewing your proposed disclosures.

Response

The proposed disclosure which the Company expects to include in the Company’s amended
annual report on Form 20-F for the fiscal year ended June 30, 2006 has been included as an
Appendix to this Response Letter.

2

2.

On a related point, in your response to prior comment 2, you explain that in calculating
depreciation, depletion and amortization for the periods presented, you “…already followed the
approach suggested by the SEC by monitoring and assessing the implications of significant
changes in the Company’s reserves that it became aware of during the relevant period…,” for
the purpose of updating the reserve estimates utilized in your formulas to reflect material
information as it arises during the period. However, you then state that for some of your
properties, such calculations are “adjusted half-yearly,” while the calculation for other properties
are adjusted either in conjunction with a three phase process resulting in your annual reserve
updates, or possibly from operational data received from the mine operations during your
monthly meetings.

Further ambiguity surfaces in your proposed disclosure, stating that in calculating amortization
and depreciation, “…account is taken of any major changes in the proven and probable
reserves between formal reserve estimations resulting from acquisitions, dispositions or
discoveries or changes in assumptions that have a material impact on the reserves that the
Company becomes aware of.” We do not see how the examples you provide result in timely
updating of the formulas utilized in calculating DD&A expense, as would be necessary to reflect
new information arising during the period that would have a material effect on the DD&A
expense that is otherwise being calculated. Please further revise your proposed disclosure to
address the following points.

(a)

Clarify the manner by which “account is taken” so that it is clear how this compares to
adjusting your DD&A formulas when necessary to reflect new information, and also
differentiate between major changes and non-major changes that would have a
material effect on the DD&A expense.

(b)

Indicate the amount of time that generally passes between an event in which you may
have informal information about reserve changes, and the formal reserve estimations
that may later be prepared and considered in conjunction with one of your monthly, bi-
annual, or yearly events.

(c)

Discuss the extent to which you have evaluated changes which would have a
material effect on DD&A recorded during the period, in addition to those changes
having a material impact on reserves.

If you are not able to show that your DD&A formulas have been updated during the periods
presented on a timely basis, to reflect new information that would have a material effect on your
DD&A expense, it may be necessary to revise your financial statements. Please describe the
nature of the documentation you have supporting the application of your policy.

Response

The Company acknowledges the Staff’s observations and wishes to advise the Staff on a
supplemental basis that the manner by which (a) “account is taken” and how it differentiates
between major changes and non-major changes that would have a material effect on DD&A
expense; and, as a consequence, (b) the amount of time that generally passes between an
event about which the Company may have informal information relating to the possibility of
reserve changes, and the formal reserve estimations that may be later prepared, differs based
on the nature of its mining properties.

3

South African Operations

The majority of Gold Fields’ operations, based on gold production and therefore having the
most significant impact on DD&A expense, are located in South Africa. These South African
operations are long-life mines (ranging from 13 to 28 years), characterized by geological and
grade continuity proven over many years. Due to the relatively long life of these operations,
their proven and probable reserves are generally less sensitive to changes than the Company’s
shorter life, surface operations, such as those in Australia. In addition, and perhaps more
relevant to the discussion of DD&A, is the fact that the majority of the changes in period-on-
period proven and probable reserves at these South African operations occur in below-
infrastructure reserves (DD&A is calculated based on only above-infrastructure reserves, as
discussed with the SEC in previous comment letter). As an example, the Kloof Extension Area
(the “KEA”) drilling program was completed in fiscal 2005, which resulted in revised geological
and evaluation models used to create a new mine design and schedule. Similar revisions were
made to the Eastern Boundary Area (the “EBA”) geological and evaluation models after
completion of the surface drilling program, surveys and analysis in fiscal 2005. These revisions
resulted in ounces being removed from both the EBA and KEA in the reserve statement for
fiscal 2005. However, the revisions only impacted below-infrastructure reserves, resulting in a
decrease in below-infrastructure reserves at Kloof from 14.25 million ounces as of June 30,
2004 to 1.9 million ounces as of June 30, 2005 and therefore did not impact DD&A calculations.
In contrast, above-infrastructure reserves increased marginally during the same period.

As a general guideline at its South African operations, it is the Company’s policy to update its
DD&A calculations only once the new ore reserve declarations have been approved by the
Company’s Board and Executive Committee. The Company considers this policy reasonable in
light of the fact that, as confirmed with the Staff in previous comment letters, the Company
calculates DD&A at its South African underground operations, which are less sensitive to
changes in reserve assumptions as described above, using the units-of-production method
based on above-infrastructure proven and probable reserves only. However, if it becomes
aware of significant changes in its above-infrastructure reserves ahead of the scheduled
updates, the Company would not hesitate to update its DD&A calculations in advance of
receiving approval by the Company’s Board and Executive Committee.

Therefore, if the Company becomes aware of a significant change in its above-infrastructure
reserve assumptions, including through new geological information (which is gathered and
loaded into the geological mapping and analysis systems on a continuous basis), it would
update its DD&A calculations as soon as the information has been verified. The Beatrix mine,
as an example, is a lower grade mine than Driefontein or Kloof, and therefore generally more
sensitive to changes in it cost profile than the Company’s other South African operations. As a
result, changes in the cost profile, which affect pay limits and therefore reserves, are monitored
particularly closely. During the May and June 2005 monthly operations meetings, it became
apparent that there were cost increases at Beatrix that resulted in an increase in the pay limit.
Due to the increase in the pay limit, certain reserves (above and below infrastructure) at Shaft
No. 2 (now part of the South section) and Vlakpan included in the fiscal 2004 reserve statement
became uneconomical to mine and were therefore excluded from the 2005 life of mine profile.
In addition, due to the restructuring at the South section of the mine, certain areas were closed
which further impacted the life of mine plan. The Company incorporated this new information it
its DD&A calculations and year-end impairment analysis in May 2005, three months before

4

approval by the Company’s Board and Executive Committee. In particular, the Company
confirms that it updated its reserve information as of June 30, 2005 to appropriately conduct
impairment testing, following the guidance in paragraphs 16 through 21 of SFAS 144. As a
result, the Company recorded an impairment of $211.1 million. Similar procedures are in place
at the Company’s other South African operations to monitor relevant new information, and to
assess whether the new information might also represent a “major change” to be incorporated
in DD&A calculations ahead of approval of the new reserve numbers by the Board and
Executive Committee.

International Operations

Gold Fields’ international operations include:

•

Australia: The Australian operations consist of the St. Ives and Agnew mines which are
located in the state of Western Australia. These two mines together accounted for
approximately 17.6% of attributable gold production in fiscal 2006.

•

Ghana: These operations consist of: (1) the Tarkwa mine, which comprises several open
pit operations with two heap leach recovery facilities and a SAG mill and CIL plant and (2)
the Damang mine, which consists of a number of open pit operations with a CIL plant. The
Ghana operations accounted for approximately 16.5% of attributable gold production in
fiscal 2006.

•

Venezuela: The Venezuela operation consists of the Choco 10 mine, located in the El
Callao district of Guayana, Venezuela, and accounted for approximately 0.6% of
attributable gold production in fiscal 2006.

The lives of the Company’s operations in Australia are significantly shorter (two to four years)
than the Company’s South African underground operations and, as a consequence, more
susceptible to changes in reserve estimates. This has resulted in the Company adapting its
accounting practices to ensure that these operations’ DD&A calculations are updated on a more
regular basis for all known changes in proven and probable reserves (regardless of whether
they represent a “major” or “non-major” change). These updates occur at least quarterly, and at
times as frequently as monthly. The nature of the orebodies, and the on-going information being
gathered in connection with the orebodies, facilitates the Company’s quarterly updates.

The Company’s operations in Ghana, although surface, are treated similar to the Company’s
South African underground operations – mainly as a result of the longer life of the orebody. As a
general guideline, the Company only updates its DD&A calculations once the new ore reserve
declarations have been approved by the Company’s Board and Executive Committee.
However, it will update its DD&A calculations ahead of these approvals being obtained if it
becomes aware of significant changes in its proven and probable reserves ahead of the
scheduled updates.The Company’s assessments of asset recoverability will, consistent with its
other locations, always incorporate all information known about its reserves on the balance
sheet date.

All Operations

Finally, in response to the Staff’s request in (c) above, the Company wishes to confirm that all
changes expected to have a material impact on DD&A in addition to those changes having a
material impact on reserves, such as an acquisition, sale, closure or fire at a shaft, are

5

incorporated in its DD&A calculations as soon as they become known. During fiscal 2004 for
example, Gold Fields sold a portion of the Driefontein mining area (Block 1C11) to AngloGold
(which could be accessed from the adjacent TauTona mining operation of AngloGold). The
Company immediately updated its DD&A calculations in January 2004 upon obtaining approval
by the South African Competition Commission.

In response to the Staff’s comments, the Company proposes to amend its annual report on
Form 20-F for the fiscal year ended June 30, 2006 to expand its discussion of critical
accounting policies under Item 5 as follows:

“Depreciation, depletion and amortization of mining assets

Depreciation, depletion and amortization charges are calculated using the units of production method
and are based on Gold Fields’ current gold production as a percentage of total expected gold production
over the lives of Gold Fields’ mines. An item is considered to be produced at the time it is removed from
the mine. The lives of the mines are estimated by Gold Fields’ geology department using interpretations
of mineral reserves, as determined in accordance with the SEC’s industry guide number 7.

Depreciation, depletion and amortization at Gold Fields’ South African operations (which are long-life
mines ranging from 13 to 28 years), are calculated using above-infrastructure proven and probable
reserves only, which because of their reserve base and respective long lives, are less sensitive to
change in reserve assumptions. Accordingly, at these locations, it is the Company’s policy to update its
depreciation, depletion and amortization calculations only once the new ore reserve declarations have
been approved by Gold Field’s Board and Executive Committee. However, if Gold Fields becomes
aware of significant changes in its above-infrastructure reserves ahead of the scheduled updates, the
Company would not hesitate to update its DD&A calculations in advance of receiving approval by the
Company’s Board and Executive Committee.

A similar approach is followed at Gold Fields’ operations in Ghana, due to the longer-life of the primary
orebody. At Gold Fields’ Australian operations, where mine-life ranges from two to four years, proven
and probable reserves used for the calculation of depreciation, depletion and amortization are more
susceptible to changes in reserve estimates. At these locations, Gold Fields’ depreciation, depletion and
amortization calculations are updated on a more regular basis (at least quarterly) for all known changes
in proven and probable reserves. The nature of the orebody, and the on-going information being
gathered in connection with the orebody, facilitates these updates.

The estimates of the total expected future lives of Gold Fields’ mines could be different from the actual
amount of gold mined in the future and the actual lives of the mines due to changes in the factors used
in determining Gold Fields’ mineral reserves. Changes in management’s estimates of the total expected
future lives of Gold Fields’ mines would therefore impact the depreciation, depletion and amortization
charge recorded in Gold Fields’ consolidated financial statements.”

6

Financial Statements

Note 2 – Significant Accounting Policies, page F-7

(d) Property, Plant and Equipment, page F-8

3.

We note your disclosure under point (i) indicating that you capitalize costs incurred to further
develop or define mineralization in an orebody, or which establish or expand productive
capacity. Please disclose how you determine whether a cost establishes or expands productive
capacity, while addressing the uncertainty that exists at the time of incurring the cost, as to
whether you will be successful in this regard.

Additionally, explain how you account for costs of drilling and other activities necessary to
convert measured, indicated and inferred resources to proven or probable reserves, and for
upgrading resources from one category to another, in close proximity to your development and
production stage properties.

Response

The determination of whether the costs incurred will establish or expand productive capacity
(and therefore embody a probable future benefit), will depend upon the data that has previously
been gathered in connection with the orebody through soil sampling, geo-physics, mapping and
drilling. The uncertainty, at the time of incurring the cost, will generally be diminished as a result
of previously established proven (where the size, shape, depth and mineral content of reserves
are well-established) and probable (where the degree of assurance, although lower than that
for proven reserves, is high enough to assume continuity between points of observation)
reserves. Therefore, where proven and probable reserves have already been established, there
is generally a high degree of confidence (“probable”) that the costs will establish or expand
productive capacity.

For the South African operations, as an example, all of the surface drilling and long inclined
underground boreholes for periods included in the Company’s 2006 Form 20-F have been “in-
fill” drilling or reserve definition drilling, which is designed to provide better resolution of the
structural and sedimentological features, in turn increasing the local knowledge and hence
confidence in the mine planning. The Company considers these costs to be development costs
eligible for capitalization since the costs are expected to provide future economic benefits as a
result of previously-established proven and probable reserves.

With regards to the last portion of the Staff’s comment, the Company wishes to advise the Staff
that the costs of drilling and other activities necessary to convert “Measured”, “Indicated” and
“Inferred” resources to proven or probable reserves, as well as for upgrading resources from
one category to another, in close proximity to its development and production stage properties,
and prior to the establishment of final feasibility study, are expensed as incurred. This approach
is in accordance with its accounting policy for mining exploration on page F-19 of the
Company’s 2006 Form 20-F.

7

Supplemental information

Pursuant to a request from the Staff, Gold Fields hereby acknowledges (i) it is responsible for the
adequacy and accuracy of the disclosures in the filing, (ii) Staff comments or changes to disclosure in
response to Staff comments do not foreclose the Commission from taking any action with respect to
the filing and (iii) it may not assert Staff comments as a defense in any proceeding initiated by the
Commission or any person under the federal securities laws of the United States.

* * * * *

Should you or the Staff have any questions or require any additional information, please contact the
undersigned at +27 11 644 2502 or via e-mail at nholland@goldfields.co.za.

Yours sincerely,

/s/ Nicholas Holland

Nicholas J. Holland
Chief Financial Officer

Gold Fields Limited

cc:

Lily Dang, Securities and Exchange Commission

Jenifer Gallagher, Securities and Exchange Commission
Ken Schuler, Securities and Exchange Commission
Michael Fleischer, Gold Fields Limited
Paul Schmidt, Gold Fields Limited
Jennifer Schneck, Linklaters

8

APPENDIX

The following text will replace the similarly titled text on pages 39-44:

Reserves of Gold Fields as of June 30, 2006

Methodology

The June 30, 2006 ore reserve statement presented below formed the basis of the signed Competent
Persons Report, requested by Gold Fields and as required by the Johannesburg Stock Exchange,
that underpinned its bid for the South Deep Mine during 2006. The mining industry consultant used in
the process was previously involved with auditing Gold Fields’ ore reserve statements and based
these June 30, 2006 ore reserves on site audits, extensive reference to internal reports,
documentation prepared by Gold Fields, production details, financial valuations and metallurgical and
tailings data for each operation.

The methodology used by Gold Fields’ internally to produce its ore reserve statements as at
December 31, 2005 and that used by the mining industry consultant to re-state the ore reserves to
June 30, 2006 were similar in nature and involved for each mining asset reviews and assessments of
(i) the mining

asset, including title, rights and applicable laws; (ii) the geology; (iii) mineralised material

from which ore reserves are derived; (iv) the mine plan, schedule and ore reserves; (v) the processing
method; (vi) tailings management; (vii) the engineering infrastructure, expected overhead costs and
planned capital projects; (viii) human resources; (ix) safety and health issues; (x) any environmental
issues, including legislation and liabilities; (xi) valuation, including financial models and resultant net
present values; and (xii) risk assessment, including general risks, specific risks and remediation
measures.

The main difference in the process for preparing the ore reserve statement for June 30, 2006 was in
the shorter timeframes involved, as only half the time had passed since the preparation of the
immediately prior ore reserve statements, resulting in less available data, reviews and processes to
which the resulting figures were exposed. The re-statement therefore focused on a review of all
available new information, updates, and any other material issues apparent since the last full audit of
Gold Fields’ ore reserve statements as at June 30, 2005, while also considering Gold Fields’ ore
reserve declaration as at December 31, 2005. In arriving at the final ore reserves for the June 30,
2006 declaration, following and based on the reviews and assessments outlined above, the ore
reserve estimates were updated where material changes were apparent (such as increases to the
below infrastructure projects at Kloof and Driefontein), other figures associated with the items outlined
above were also updated, and finally mining depletions were applied at the various operations.

While there are some differences between the definition of the South African Code for Reporting of
Mineral Resources and Mineral Reserves, or SAMREC Code, and that of the SEC’s industry guide
number 7, only reserves at each of Gold Fields’ operations and exploration projects as June 30, 2006
which qualify as reserves for purposes of the SEC’s industry guide number 7 are presented in the
table below. See “—Glossary of Mining Terms.” In accordance with the requirements imposed by the
JSE, Gold Fields reports its reserves using the terms and definitions of the SAMREC Code. Mineral or
ore reserves, as defined under the SAMREC Code, are divided into categories of proven and
probable reserves and are expressed in terms of tons to be processed at mill feed head grades,
allowing for estimated mining dilution and recovery factors.

Gold Fields reports reserves using cut-off grades (mainly for open pit operations) and pay limits to
ensure the reserves realistically reflect both the cost structures and required margins relevant to each

9

mining operation. Cut-off grade is the grade that distinguishes the material within an orebody that is to
be extracted and treated from the remaining material. The pay limit is the grade at which an orebody
can be mined without profit or loss calculated using an appropriate gold or copper price and working
costs, plus modifying factors. Modifying factors used to calculate the pay limit grades include
adjustments to mill delivered amounts, due to dilution incurred in the course of mining. Modifying
factors applied in estimating reserves are primarily historical, but commonly incorporate adjustments
for planned operational improvements such as those described below under “—Description of Mining
Business—Productivity Initiatives.” Tonnage and grade may include some mineralization below the
selected pay limit and cut-off grade to ensure that the reserve comprises blocks of adequate size and
continuity. Reserves also take into account cost levels at each operation and are supported by mine
plans.

The estimation of reserves at the South African underground operations is based on surface drilling,
underground drilling, surface three-dimensional reflection seismics, orebody facies, structural
modeling, underground channel sampling and geostatistical estimation. The reefs are initially explored
by drilling from the surface on an approximately 500 meter to 2,000 meter grid. Once underground
access is available, drilling is undertaken on an approximately 30 meter by 60 meter grid.
Underground channel sampling perpendicular to the reef is undertaken at three meter intervals in
development areas and five meter intervals at stope faces.

The following sets out the reserve estimation methodologies for the different categories of reserves at
the underground operations of each of the South African mines.

Driefontein

Reserve Classification

Sample Spacing Range

Min/Max

(meters)

Maximum Distance Data is Projected

(meters)

Proven

3 to 180

110

Probable (AI)

(1)

3 to 1,140

570

Probable (BI)

(1)

3 to 2,840

1,420

Note:
(1)

AI is above infrastructure; BI is below infrastructure.

For proven reserves, the orebody is opened-up and sampled on a 3 meter spacing for development
(such as raises), and a 5 meter grid for stoping, together with underground borehole spacings ranging
from tens to hundred of meters. Blocks classified as proven are therefore generally adjacent to close
spaced sampling and generally pierced by a relatively dense irregular pattern of boreholes. Estimation
is constrained within both geologically homogenous structural and facies zones, and is generally
derived from either ordinary or simple kriged small-scale grids, ranging from 10 meter to 20 meter
block sizes.

For above infrastructure probable reserves, the estimates access the significant numbers of samples
on a 3 meter spacing for development, and a 5 meter grid for stoping bordering these areas. In
addition underground borehole spacings ranging from tens to hundred of meters are used together
with surface drillholes and seismic surveys. Blocks classified as probable (AI) are generally adjacent
to blocks classified as proven. Estimation is constrained within homogenous structural and facies
zones, and is generally derived from either ordinary or simple kriged medium to macro scale sized

10

grids ranging from 40 meter to 420 meter sizes, or through declustered averaging or Sichel “t”
techniques. For planning purposes these blocks are further evaluated to facilitate the selection of
blocks above the cut-off grade.

For below infrastructure probable reserves, the estimates access the significant numbers of samples
on a 3 meter spacing for development, and a 5 meter grid for stoping above these areas. In addition
underground borehole spacings ranging from tens to hundred of meters are used together with
surface drillholes and seismic surveys. Blocks classified as probable (BI) are generally below blocks
classified as proven or probable (AI). Estimation is constrained within homogenous structural and
facies zones, and is generally derived from either ordinary or simple kriged medium to macro scale
sized grids ranging from 40 meters to 420 meter sizes, or through declustered averaging or Sichel “t”
techniques. For planning purposes these blocks are further evaluated to facilitate the selection of
blocks above the cut-off grade.

Kloof

Reserve
Classification

Sample Spacing Range Min/Max

(meters)

Maximum Distance Data is Projected

(meters)

Proven

3 to 150

150

Probable (AI)

(1)

3 to 718

360

Probable (BI)

(1)

3 to 1,390

890

Note:
(1)

AI is above infrastructure; BI is below infrastructure.

Estimations for proven reserves are made on the same basis as at Driefontein.

Estimations for above infrastructure probable reserves are made on the same basis as at Driefontein,
but with medium sized kriged grids starting from 40 meters to macro blocks of 400 meters. For
planning purposes these blocks are further evaluated to facilitate the selection of blocks above the
cut-off grade.

Estimations for below infrastructure probable reserves are made on the same basis as at Driefontein,
but with medium sized kriged grids starting from 40 meters to macro blocks of 400 meters. The
distinction between estimation techniques for above infrastructure and below infrastructure probable
reserves is the same as at Driefontein. For planning purposes these blocks are further evaluated to
facilitate the selection of blocks above the cut-off grade.

Beatrix

Reserve Classification

Sample Spacing Range Min/Max

(meters)

Maximum Distance Data is Projected

(meters)

Proven

3 to 120

120

Probable (AI)

(1)

3 to 940

750

Probable (BI)

(1)

540 to 610

740

Note:
(1)

AI is above infrastructure; BI is below infrastructure.

Estimations for proven reserves are made on the same basis as at Driefontein but with kriging blocks
ranging from 16 meters to 32 meters.

11

Estimations for above infrastructure probable reserves are made on the same basis as at Driefontein
but with medium-sized kriged blocks of 32 meters, and macro geological zone estimates being made
through declustered averaging or Sichel “t” techniques. For planning purposes these blocks are
further evaluated to facilitate the selection of blocks above the cut-off grade.

Estimations for below infrastructure probable reserves are made on the same basis as at Driefontein
but with medium-sized kriged blocks being 32 meters, to macro geological zone estimates through
declustered averaging or Sichel “t” techniques. The distinction between estimation techniques for
above infrastructure and below infrastructure probable reserves is the same as at Driefontein. For
planning purposes these blocks are further evaluated to facilitate the selection of blocks above the
cut-off grade.

The primary assumptions of continuity of the geologically homogenous zones are driven by the
geological model, which is updated only if new information arises. Any changes to the model are
subject to peer, internal technical corporate consultant and independent consultant review.
Historically, mining at South African deep level gold mines has shown significant geological continuity,
so that new mines were started based on limited surface borehole information. Customarily,
geological facies are primarily based on the definition of different facies within each conglomerate
horizon. These facies are extrapolated into new, undeveloped areas taking into account any surface
borehole data in those areas. Normally these facies are continuous, supported by extensive historical
sample databases, and can be incorporated in the macro kriging of large blocks.

For the Tarkwa open pit operation, estimation of reserves is based on a combination of an initial 100
or 200 meter grid of diamond drilling and in certain areas a 12.5 meter to 25.0 meter grid of reverse
circulation drilling. For the Damang open pit operation, estimation of reserves is based on a 20 meter
to 80 meter grid of diamond drilling and in certain areas reverse circulation drilling.

At the Australian operations, the estimation of reserves for both underground and open pit operations
is based on exploration, sampling and testing information gathered through appropriate techniques,
primarily from drill holes and mine development. The locations of sample points are spaced closely
enough to deduce or confirm geological and grade continuity. Generally, drilling is undertaken on
grids, which range between 20 meters by 20 meters to 40 meters by 40 meters, although this may
vary depending on the continuity of the orebody. Due to the variety and diversity of resources at St.
Ives and Agnew, sample spacing may also vary depending on each particular ore type. For Choco 10
and the Cerro Corona Project, estimation is based on diamond drill and reverse circulation holes. The
spacing of holes at Cerro Corona is generally around 50 meters, with some areas approximating a 25
meter grid. The drill spacing at Choco 10 is varied, depending on geological and grade continuity, with
a general spacing of 50 meters by 25 meters to 25 meters by 25 meters.

12

Reserve Statement

As of June 30, 2006, Gold Fields had aggregated attributable proven and probable gold reserves of
approximately 61.8 million ounces as set out in the following table.

Gold ore reserve statement as of June 30, 2006

(1)

Tons

Proven

reserves

Head Grade

Gold

Tons

Probable

reserves

Head Grade

Gold

Tons

Total

reserves

Head Grade

Gold

Attributable

gold

production

in the 12

months

ended June

30, 2006

(2)

(million)

(g/t)

(‘000 oz)

(million)

(g/t)

(‘000 oz)

(million)

(g/t)

(‘000 oz)

(‘000 oz)

Underground

Driefontein (total) ...........

26.0

7.8

6,491

61.3

8.1

15,918

87.3

8.0

22,408

1,011

Above infrastructure

(3)

....

26.0

7.8

6,491

27.8

8.0

7,120

53.8

7.9

13,611

1,011

Below infrastructure

(3)

....

—

—

—

33.5

8.2

8,798

33.5

8.2

8,798

—

Kloof (total)....................

13.7

9.7

4,286

29.3

10.2

9,590

43.0

10.0

13,877

897

Above infrastructure

(3)

...

13.7

9.7

4,286

24.1

9.8

7,568

37.8

9.8

11,854

897

Below infrastructure

(3)

...............

—

—

—

5.2

12.1

2,022

5.2

12.1

2,022

—

Beatrix (total) .................

15.9

4.9

2,517

28.7

5.9

5,410

44.6

5.5

7,927

596

Above infrastructure

(3)

...

15.9

4.9

2,517

28.7

5.9

5,410

44.6

5.5

7,927

596

Australia

St. Ives ........................

0.2

5.4

27

5.0

5.4

868

5.1

5.4

895

243

Agnew...........................

0.6

12.1

237

0.9

7.2

211

1.5

9.2

448

166

Total Underground ......

56.4

7.5

13,558

125.2

7.9

31,997

181.5

7.8

45,555

2,913

Surface (Rock Dumps)

Driefontein ....................

—

—

—

6.2

1.1

225

6.2

1.1

225

139

Kloof ............................

—

—

—

9.9

0.7

216

9.9

0.7

216

17

Beatrix...........................

—

—

—

0

—

0

0

—

0

—

Surface (Production
Stockpile)

Ghana

Tarkwa ..........................

4.1

0.8

102

—

—

—

4.1

0.8

102

—

Damang ........................

5.4

1.2

210

—

—

—

5.4

1.2

210

—

Australia

St. Ives ........................

7.1

1.1

257

—

—

—

7.1

1.1

257

—

Agnew...........................

0.4

1.4

19

—

—

—

0.4

1.4

19

—

Surface (Open Pit)

Ghana

Tarkwa .........................

117.9

1.3

5,021

120.2

1.2

4,758

238.0

1.3

9,779

504

(4)

Damang

(5)

..................

3.2

2.4

247

8.6

1.7

471

11.7

1.9

717

167

(4)

Australia

St. Ives

(5)

...................

1.4

2.8

123

10.4

1.9

647

11.8

2.0

770

253

(4)

Agnew

(5)

.....................

0.6

2.5

47

1.7

3.1

172

2.2

3.1

218

56

(4)

Venezuela

Choco 10 ...............

1.3

3.7

162

17.8

1.7

972

19.1

1.8

1,133

24

Peru

Cerro Corona .........

22.0

1.2

816

57.0

1.0

1,761

79.1

1.0

2,578

—

Total Surface.........

163.4

1.3

7,004

231.7

1.2

9,222

395.0

1.3

16,224

1,160

Total .....................

219.8

2.9

20,562

356.9

3.6

41,219

576.6

3.3

61,778

4,073

Notes:

(1) (a) Quoted as mill delivered tons and Run of Mine, or RoM, grades, inclusive of all mining dilutions and gold losses

13

except mill recovery. Metallurgical recovery factors have not been applied to the reserve figures. The approximate
metallurgical factors are as follows: (1) Driefontein 97%; (2) Kloof 97%; (3) Beatrix 96%; (4) Tarkwa 95% for milling, 64%
for heap leach; (5) Damang 90.9% to 93.5%; (6) St. Ives 85% to 95% for milling, 55% to 60% for heap leach; and
(7) Agnew 93% to 95%. The metallurgical recovery is the ratio, expressed as a percentage, of the mass of the specific
mineral product actually recovered from ore treated at the plant to its total specific mineral content before treatment. The
South African operations have a fairly consistent metallurgical recovery, while the recoveries on the International
operations vary according to the mix of the source material and method of treatment.

(b) For Driefontein, Kloof and Beatrix, a gold price of Rand 92,000 per kilogram ($400 per ounce at an exchange rate of
Rand 7.15 per $1.00) was applied in calculating ore reserve figures. For the Tarkwa and Damang operations and the
Cerro Corona Project, ore reserve figures are based on an optimized pit at a gold price of $400 per ounce. For the
Australian operations ore reserve figures are based on a gold price of A$560 per ounce ($400 per ounce at an exchange
rate of A$1.40 per $1.00). Open pit ore reserves at the Australian operations are similarly based on optimized pits. The
gold price used for reserves is the three-year average, calculated on a monthly basis, of the London afternoon fixing
price of gold.

(c) For the South African operations, mine dilution relates to the difference between the mill tonnage and the stope face
tonnage and includes other sources stoping (which is waste that is broken on the mining horizon, other than on the
stope face), development to mill and tonnage discrepancy (which is the difference between the tonnage expected on the
basis of the mine’s measuring methods and the tonnage accounted for by the plant). For the International operations,
dilution relates to unplanned waste and/or low-grade material being mined and delivered to the mill. Ranges are given
for those operations that have multiple orebody styles and mining methodologies. The mine dilution factors are as
follows: (i) Driefontein 24%; (ii) Kloof 22%; (iii) Beatrix 23%; (iv) Tarkwa 10%; (v) Damang 11% to 15%; (vi) St. Ives 16%
to 18%; and (vii) Agnew 10% to 20%.

(d) The mining recovery factor relates to the proportion or percentage of ore mined from the defined orebody at the gold
price used for the declaration of reserves. This percentage will vary from mining area to mining area. This percentage
reflects planned and scheduled reserves against total potentially available reserves (at the gold price used for the
declaration of reserves), with all modifying factors, mining constraints and pillar discounts applied. The mining recovery
factors are as follows: (i) Driefontein 86.5%; (ii) Kloof 61%; (iii) Beatrix 65%; (iv) Tarkwa 95%; (v) Damang 91.6%; (vi) St.
Ives 81%; and (vii) Agnew 96%.

(e) The pay limit (South African operations) and cut-off grade (International operations) vary per shaft, open pit or
underground mine, depending on the respective costs, depletion schedule, ore type and dilution. The following are the
average or range of values applied in the planning process: (i) Driefontein 1,520 cm.g/t; (ii) Kloof 1,500 cm.g/t;
(iii) Beatrix 970 cm.g/t; (iv) Tarkwa 0.33 g/t for heap leach and 0.54 g/t for mill feed; (v) Damang 0.89 g/t for fresh ore
and 0.59 g/t for oxide ore; (vi) St. Ives 0.8 g/t for heap leach, 1.0 g/t for mill feed—open pit, and 2.2—5.1 g/t for mill
feed—underground; (vii) Agnew 0.71-0.89 g/t for mill feed—open pit, and 3.1—6.1 g/t for mill feed—underground.

(f) Totals may not sum due to rounding. Where this occurs it is not deemed significant

(2) Actual gold produced after metallurgical recovery.

(3) Above infrastructure reserves relate to mineralization which is located at a level at which an operation currently has

infrastructure sufficient to allow mining operations to occur. Below infrastructure reserves relate to mineralization which
is located at a level at which an operation currently does not have infrastructure sufficient to allow mining operations to
occur, but where the operation has made plans to install additional infrastructure in the future which will allow mining to
occur at that level.

(4) Includes some gold produced from stockpile material, which cannot be separately measured.

(5) Excludes inferred material within the pit design.

The following table sets forth the proven and probable copper reserves of the Cerro Corona Project as
of June 30, 2006, that are attributable to Gold Fields.

14

Copper ore reserve statement as of June 30, 2006

Tons

Proven

Reserves

Grade Cu

Cu

Tons

Probable

Reserves

Grade Cu

Cu

Tons

Total

Reserves

Grade Cu

Cu

Attributable

Copper

production

in the 12

months

ended June

30, 2006

(million)

(%)

(million lbs) (million)

(%)

(million lbs) (million)

(%)

(million lbs) (million lbs)

Surface (Open Pit)
Peru

Cerro Corona...........

20.7

0.6

288

55.1

0.5

591

75.9

0.5

879

—

Gold and copper price sensitivity

The amount of gold mineralization that Gold Fields can economically extract, and therefore can
classify as reserves, is very sensitive to fluctuations in the price of gold. At gold prices different from
the gold price of $400 per ounce used to estimate Gold Fields’ attributable reserves of 61.8 million
ounces of gold as of June 30, 2006 listed above, Gold Fields’ operations would have had significantly
different reserves. Based on the same methodology and assumptions as were used to estimate Gold
Fields’ reserves as of June 30, 2006 listed above, but applying different gold prices that are 10%
above and below the $400 per ounce gold price used to estimate Gold Fields’ attributable reserves,
the attributable gold reserves of Gold Fields’ operations would have been as follows (excluding
attributable reserves from Choco 10 as explained below):

$ 360/oz

$ 400/oz

$ 440/oz

(‘000 oz)

Driefontein

(1)

................................................................

14,145

(2)

22,633

22,741

Kloof

(1)

...........................................................................

10,432

(3)

14,092

15,526

Beatrix

(1)

.......................................................................

6,402

7,927

9,362

Tarkwa ..........................................................................

8,404

9,881

12,441

Damang .......................................................................

814

928

984

St. Ives ..........................................................................

1,622

1,921

2,100

Agnew ...........................................................................

538

685

753

Cerro Corona ..............................................................

2,419

2,578

2,578

(4)

Total

(1)(5)

.......................................................................

44,776

60,645

66,485

Notes:

(1)

South African operations’ reserves include run-of-mine ore stockpiles.

(2)

Excludes Shaft No. 5 below infrastructure material that would not be economical to mine, and thus would not be a
reserve, at this lower gold price.

(3)

Excludes revised Kloof Extension Area, or KEA, below infrastructure material that would not be economical to mine,
and thus would not be a reserve, at this lower gold price.

(4)

The tailings management facility, or TMF, has a finite capacity of 90Mt and the ore reserves therefore cannot change
once the TMF has been filled.

(5)

The sensitivity analyses are calculated as 10% above and below the base price in the local currency of the respective
operation, with Ghana and Cerro Corona in US$, at an exchange rate of Rand 7.15 per $1.00 for the South African
operations and A$1.40 per $1.00 for the Australian operation. Choco 10 was acquired by Gold Fields in early 2006,
and, because equivalent sensitivities to those declared above were not available, they have been excluded from the
above table.

The London afternoon fixing price for gold on November 15, 2006 was $617.75 per ounce.

15

Gold Fields’ attributable gold reserves decreased from 62.8 million ounces at June 30, 2005 to 61.8
million ounces at June 30, 2006, primarily as a result of depletion due to mining, and offset in part by
the inclusion of Choco 10.

The amount of copper mineralization that Gold Fields can economically extract, and therefore can
classify as reserves, is sensitive to fluctuations in the price of copper. At copper prices different from
the copper price of $1.00 per pound used to estimate Gold Fields’ attributable copper reserves of 879
million pounds as of June 30, 2006 listed above, Gold Fields’ operations would have had significantly
different copper reserves. Based on the same methodology and assumptions as were used to
estimate Gold Fields’ copper reserves as of June 30, 2006 listed above, but applying different copper
prices that are 10% above and below the copper price of $1.00 per pound used to estimate Gold
Fields’ attributable copper reserves, the attributable copper reserves of Gold Fields’ operations would
have been as follows:

$ 0.90/lb

$ 1.00/lb

$ 1.10/lb

Copper (million lbs)

Cerro Corona................................................................

861

879

879

(1)

Gold Fields’ methodology for determining its reserves is subject to change and is based upon
estimates and assumptions made by management regarding a number of factors as noted above
under “—Methodology.” Accordingly, the sensitivity analysis of Gold Fields’ reserves provided above
should not be relied upon as indicative of what the estimate of Gold Fields’ reserves would actually be
or have been at the gold or copper prices indicated, or at any other gold or copper price, nor should it
be relied upon as a basis for estimating Gold Fields’ ore reserves based on the current gold or copper
price or what Gold Fields’ reserves will be at any time in the future. See “Risk Factors—Gold Fields’
reserves are estimates based on a number of assumptions, any changes to which may require Gold
Fields to lower its estimated reserves.

The following additional terms will be included in the Glossary of Mining Terms, which starts on page
119:

Declustered averaging: an estimation technique used in the evaluation of ore reserves.

Kriging: an estimation technique used in the evaluation of ore reserves.

Sichel “t”: an estimation technique used in the evaluation of ore reserves.

Note:

(1)

Under the current tailings dam design at the Cerro Corona Project, reserves would not respond to an upward
movement of the copper price because of current capacity constraints at the tailings storage facility for the Cerro
Corona Project.