M.Phil. / The combustion of fuels in poorly designed cookstoves is a major anthropogenic source of atmospheric emissions with severe environmental and health implications. It is widely acknowledged that these challenges are best addressed with the development and dissemination of clean cookstoves. Widely used stove testing protocols (UCB Water Boiling Test and variants) are often single task-based and not representative of real-world uses or likely combinations of the manner in which fuels, stoves and pots may be used. The hypothesis of this study is that a stove testing procedure that provides for testing of stove/fuel/pot combinations, in a sequence of heterogeneous tests, provides a better representation of thermal performance and emissions than existing protocols based on prescribed fuels and fuel loads, and single tasks. The study aimed to develop and evaluate a set of testing protocols for determining thermal efficiency and emissions performance of domestic fuels and cooking devices to satisfy the rigorous performance specifications expected for claims under the Clean Development Mechanism (CDM) carbon trading market. The Heterogeneous stove Testing Protocol (HTP) was developed and documented as a complete set of standard operating procedures (SOPs), using a template derived from the Desert Research Institute (DRI), Reno, Nevada, and used for performance evaluation of fuel/stove combinations. The effect of pot size on the performance of two paraffin wick stoves and a pressurised paraffin stove was assessed and was found not to be a major factor, which affected thermal efficiency only at the high power setting. Power setting was found to influence the thermal efficiency and combustion performance of all stoves tested, indicating the need for assessment of the devices across the full range of power settings (where feasible). The HTP was also employed in characterising the combustion performance of coal stoves, using three different ignition methods, giving qualitative and quantitative results. Compared to the bottom-lit up-draft (BLUD) ignition method, the Basa njengo Magogo, also referred to as the top-lit up draft (TLUD) method, proved to be a better method of coal fires ignition, in terms of reduced CO:CO2 ratio and less smoke generation than in conventional braziers. The bottom-lit down-draft (BLDD) ignition method, incorporated in the SeTAR prototype coal stove, was found to be effective in fuel utilisation and improved combustion efficiency compared to the TLUD and BLUD methods, with CO:CO2 emission factors below 1% for 230 minutes. A number of parameters employed by the Water Boiling Test (WBT) were examined and compared with the HTP (e.g. turn-down ratio; simmer process; hot-start phase; use of standardised fuels and test pots). The HTP was found to provide more representative performance data over a wide range of use scenarios, the equivalent of providing performance curves rather than the minimum and maximum performance points provided by the WBT. The findings of this study have shown that the Heterogeneous stove Testing Protocol is consistent, robust, and transportable; making it a valuable tool for stove design improvements, and for the assessment of stoves under voluntary and compulsory carbon markets.