N@N

African Gold Group, Inc.: African Gold Group, Inc., Kobada "Zone 1" Resources Estimate Will Increase

TORONTO, ONTARIO--(Marketwire - Sept. 23, 2009) - African Gold Group, Inc., ("AGG" or the "Company") (TSX VENTURE:AGG) is pleased to report that, the Company recently engaged Senior Geologist, Mr. Pierre Lalande, P. Geo., to conduct a comprehensive review and analysis of AGG's Kobada, Mali gold project. Mr. Lalande's priorities for the Kobada gold project include:

- Conducting an independent review and analysis of the complete Kobada data base, consisting of all historical drill and analytical data compiled by the previous owners, (the BRGM, La Source and Cominor SA), as well as, all drill and analytical data generated by AGG, since acquiring Kobada, through the course of its exploration program(s), which commenced with diamond drilling in November 2005;

- Prepare an independent report, based on the analysis of the entire Kobada data base, that detailed Lalande's findings and conclusions with respect to Kobada's potential;

Mr. Lalande is a career geologist with approximately 40 years of field experience of which the vast extent of this experience has been spent and is currently being spent in West Africa. From 1970 - 1994 Pierre was a Senior Geologist with Watts, Griffis and McOuat Ltd. From 1994 - 2001 Pierre served as Chief Geologist for IAMGOLD Corp., and from 2001 - 2005 as Technical Advisor to Orezone Resources Inc. The West African projects that Pierre contributed in adding value to include: Sadiola (Mali), Yatela (Mali), Siguri (Guinea), Kiniero (Guinea) and Samira-Libiri (Niger).

AGG filed its Initial National Instrument 43-101 compliant Mineral Resources Estimate for its Kobada, Mali gold project on SEDAR on May 15, 2008.

The Mineral Resources estimate was completed by Watts, Griffis and McOuat Limited of Toronto ("WGM"), a well respected international consulting firm.

In summary, WGM estimates that "Zone 1" of the Kobada deposit, representing approximately 10% of the overall Kobada Trend, as presently outlined, contains an Inferred Mineral Resource of between 450,000 ounces of gold and 740,000 ounces of gold as follows:

-------------------------------------------------------------------
 Inferred Mineral Resources - "Zone 1" Kobada Project, Mali
-------------------------------------------------------------------
 No Assay Cutting Assays Cut to 10 g Au/t
Cut-off Tonnage ---------------------------------------------
Grade Avg. Grade Contained Avg. Grade Contained
(g Au/t) (x 1,000) (g Au/t) Ounces (g Au/t) Ounces
-------------------------------------------------------------------
0.3 18,381 1.25 738,080 1.11 657,386
-------------------------------------------------------------------
0.5 8,482 1.99 542,804 1.72 470,461
-------------------------------------------------------------------
1.0 5,569 3.02 540,933 2.53 452,320
-------------------------------------------------------------------

The complete 43-101 report is available for public viewing at:

[ www.sedar.com/DisplayCompanyDocuments.do?lang=EN&issuerNo=00020520 ]

Mr. Lalande has drawn a number of conclusions from his extensive review of the entire Kobada data base. The following is of great significance to AGG and its shareholders:

- The Initial 43-101 compliant Mineral Resources Estimate for "Zone 1" of the Kobada deposit, completed by WGM and filed on SEDAR on May 15, 2008, excluded 100% of the Reverse Circulation ("RC") drill data and analytical results that were generated by the previous owners of Kobada (The BRGM, La Source and Cominor SA). This was not understood by AGG prior to its review of Lalande's independent review of the Kobada data base.

- Upon inquiring of WGM with respect to the omission of the historical RC drill data, it was ascertained that the historical RC drill information was excluded from the "Zone 1" 43-101 Mineral Resources Estimate for the following reasons: 1/ WGM was unable to re-check the historical samples generated by the previous owners due to the extensive passage of time and therefore sample degradation; 2/ the historical analytical results, from the previous owners RC campaigns, were from (small) 15 gram aliquot, tri-acid digestion, MIBK extraction and AAS (Atomic Absorption Spectroscopy) as compared to AGG's analysis by 50 gram fire assay (FA50) with approximately 25% of the samples re-assayed by Metallic Screened Fire Assay on 1 kg aliquot (SFA).

- The previous owners drilled 135 RC drill holes within "Zone 1" of the Kobada deposit. Drill fences were collared on 100 meter sections over 1.2 km of strike. These RC drill hole fences were focused on defining the oxide mineralization from surface to a maximum vertical depth of 110 meters and averaged 80 vertical meters, from surface.

- Mr. Lalande concludes that the 43-101 Mineral Resources Estimate excludes a considerable amount of the upper portion of the laterite/saprolite profile by not including the historical RC data that could not be verified to WGM's satisfaction. Furthermore, Mr. Lalande states his perception that the excluded mineralized blocks within the laterite/saprolite within "Zone 1" could conservatively add 20% or more to the Initial 43-101 Mineral Resources Estimate for Kobada, "Zone 1".

- WGM the author of the Kobada, "Zone 1" Initial 43-101 Mineral Resources Estimate states clearly in section 18.3 (pg. 86) of their report:

"Many of the existing cross-sections show a need to improve the untested up-dip extension of existing mineralization. Examples of existing sections with good up-dip, near surface potential are 16+00S thru 17+00S and 18+00S thru 18+50S. Examination of additional cross-sections shows that sampling data in historical reverse circulation ("RC") drill holes indicates good up-dip mineralization potential (greater than 0.5 g Au/t) on the South Zone on 100 m spaced drill sections from 20+00S to 26+00S."

WGM has reviewed Mr. Lalande's comments stated above and agreed that its estimate excluded some significant blocks of mineralization close to surface due to the placement of drill holes and gaps in coverage. WGM acknowledged that the inclusion of areas tested exclusively by RC drilling, and shown to be gold-bearing would increase the Mineral Resources, perhaps significantly. WGM also pointed to the presence of artisanal workings along strike from the area presently drilled and concluded that substantial new resources could be added with additional drilling, including drilling at depth below the resources presently defined.

This portion of the news release has been reviewed and approved for the Kobada gold project by Al Workman, P.Geo., Vice President of Watts, Griffis and McOuat Limited, and an Independent Qualified Person for the project in accordance with regulations under National Instrument 43-101.

Separately and of additional interest, Mr. Lalande notes in his report that The Inferred Resources Estimate prepared by WGM is based on nearly 100 cored holes collared at 50 meter centres. Approximately 20,000 samples were analyzed by 50-g fire assay (FA50) with approximately 25% of the identical sample intervals re-assayed by metallic screened fire assay on 1-kg aliquot (SFA). The resources estimate used the most reliable procedure (SFA) for the results, when more than one analytical procedure was used on the same sample interval.

Based on Lalande's study of the difference in analytical values derived from the different analytical procedures (FA50) vs (SFA) for the identical sample interval, he notes:

"submitting a larger sub-sample to the lab and using larger aliquot samples has the potential of increasing the resources tonnage by 30% at the 0.3 g/t cut-off while increasing grade by 11% for a combined in situ gold increase of 30% to 45%. This is based on using 2-kg leachwell cyanidation instead of 1,000-g metallics screen fire assay (SFA) which, in fact, did increase the overall grade of the 50-g fire assay (FA) by 11%.

Table 2, below, is based on pairs of results and summarizations for various FA 50 grade ranges and the corresponding grades and ranges of the larger SFA aliquot compared to FA. The table shows an overall grade increase of 11% with considerable grade increases for FA50 results below 0.3 g/t and significant grade decreases for samples above 3.0 g/t."

----------------------------------------------------------------------------
 Table 2
 COMPARISON FIRE ASSAY (FA) TO
 METALLICS SCREENED FIRE ASSAY (FSA)
----------------------------------------------------------------------------
 Abilab (Bamako) Eco-Tech
 (Kamloops)
 ------------------------------------------
 Ratio Ratio
Comparison based on SFA SFA
 ranges (g/t) sorted by SFA to SFA to
 FA 50 FA 50 1000 FA FA 50 1000 FA
----------------------------------------------------------------------------
Greater than 10.0 : arith.
 average (g/t) 22.615 17.179 0.76 25.843 29.147 1.13
 number of pairs 24 24 1% 7 7 15%
----------------------------------------------------------------------------
3.0 - 10.0 : arith. average (g/t) 4.487 4.044 0.90 5.488 8.081 1.47
 Number of pairs 157 157 4% 8 8 17%
----------------------------------------------------------------------------
1.0 - 3.0 : arith. average (g/t) 1.651 1.774 1.07 1.815 2.122 1.17
 number of pairs 427 427 10% 33 33 69%
----------------------------------------------------------------------------
0.5 - 1.0 : arith. average (g/t) 0.698 0.840 1.20
 number of pairs 526 526 12%
----------------------------------------------------------------------------
0.3 - 0.5 : arith. average (g/t) 0.387 0.523 1.35
 number of pairs 485 485 11%
----------------------------------------------------------------------------
0.1 - 0.3 : arith. average (g/t) 0.174 0.343 1.97
 Number of pairs 1274 1274 30%
----------------------------------------------------------------------------
0.0 - 0.1 : average (g/t) 0.045 0.109 2.44
 Number of pairs 1390 1390 32%
----------------------------------------------------------------------------
All pairs : arith. average (g/t) 0.652 0.722 1.11 5.931 7.056 1.19
 Number of pairs 4280 4280 100% 48 48 100%
----------------------------------------------------------------------------

Based on the analysis of the above data, Mr. Lalande also concluded that the large number (approximately 4,500 samples) of metallic screened fire assay (SFA) results lends credibility to the potential of utilizing a low capital cost gravity recovery plant to treat the open pitable saprolite mineralization. In addition to the relatively low capital cost of constructing a gravity plant for gold recovery one would also anticipate a relatively low operating cost for extracting gold from the open pitable saprolite mineralization, utilizing a gravity plant.

This assertion is based on the fact that metallic screened fire assay (SFA) results indicate that 35% of the gold reports to the less than 4% of the samples retained on 106 microns sieved fractions of the screened fire assay. This is a clear indication that a scrubbing (washing) / desliming plant could eliminate from 80% to 95% of the clay from the feed, leaving a washed sandy gravel to be treated in a much smaller conventional gravity plant with or without a CIL (Carbon in Leach) or CIP (Carbon in Pulp) plant.

Mr. Lalande's assertions are partially based on his extensive historical field testing on similar deposits in Mali, Guinea, Niger and Burkina Faso, more specifically, that the washed, sandy gravel, could contain in excess of 85% of the gold that is contained within the total sample. A desliming plus (+) gravity plant has the potential to reduce both capital and operating costs by 60 to 80% when compared to the capital and operating cost of a CIL or CIP plant, while losing less than 20% of the gold.

Separately, AGG is pleased to report that the Company has signed a number of Confidentiality Agreements with distinct parties that have expressed interest in the co-development of AGG's Kobada, Mali and Asankrangwa, Ghana gold projects. The Company is focused on creating shareholder value through the future development of both gold projects.

Corporate Update

AGG is pleased to report that, as of July 29, 2009, warrant holders exercised a total of 5,540,200 warrants at an exercise price of $0.10 per warrant for gross proceeds to the Company of $554,020. The warrants were issued as part of the Unit Offering that closed on January 29, 2009 for gross proceeds of $1,331,050.

The Company is also pleased to report that Mr. Jean-Jacques ("JJ") Lefebvre has agreed to join AGG's Board of Directors. Mr. Lefebvre is a Belgium national, born and raised in the Democratic Republic of Congo and has work experience in virtually every country on the African continent. JJ is the former Chief Geologist for Union Miniere, Central Africa and formerly served in the capacity of Senior Geologist with PBK Engineering Ltd., as Director of Geology in Congo for the SODEMIZA project. Mr. Lefebvre was involved in the negotiation and acquisition of SOMINKI for Banro Resource Corp. (1995) and was instrumental in negotiating the acquisition of AGG's Kobada, Mali gold project with Cominor S.A. in July 2005.

African Gold Group, Inc., based in Toronto, Canada, is engaged in the identification, acquisition and exploration of prospective gold projects that are situated along significant gold trends within West Africa. To date, the Company controls a total of twelve gold concessions that are consolidated in five distinct standalone exploration projects, of which three projects are located in Ghana and the remaining two are located in Mali, West Africa.

Additional Information is available on the Company's website at [ www.africangoldgroup.com ] and on [ www.sedar.com ] and through the Company's offices at: BCE Place, Canada Trust Tower, 27th Floor, 161 Bay Street, Toronto, Canada M5J 2S1

On Behalf of the Board:

Michael A. J. Nikiforuk, President, Director

The TSX Venture Exchange has not reviewed and does not accept responsibility for the adequacy or accuracy of this release.