Total body irradiation (TBI) is often used as part of the conditioning process prior to bone marrow transplants for diseases such as leukemia. By delivering radiation to the entire body, together with chemotherapy, tumour cells are killed and the patient is also immunosupressed. This reduces the risk of disease relapse and increases the chances of a successful implant respectively. TBI requires a large flat field of radiation to cover the entire body with a uniform dose. However, dose uniformity is a major challenge in TBI. (AAPM Report 17) The ICRU report 50 recommends that the dose range within the target volume remain in the range of -5% to +7%. Whilst it is generally accepted that this is not possible for TBI, it is normally clinically acceptable that ±10% of the prescribed dose to the whole body is sufficiently uniform, unless critical structures are being shielded. TBI involves complex dosimetry due to the large source to treatment axis distance (SAD), dose uniformity and flatness over the large field, bolus requirements, extra scatter from the bunker walls and floor and large field overshoot. There is also a lack of specialised treatment planning systems for TBI planning at extended SAD. TBI doses at Westmead Hospital are prescribed to midline. Corrections are made for variations in body contour and tissue density heterogeneity in the lungs using bolus material to increase dose uniformity along midline. Computed tomography (CT) data is imported into a treatment planning system. The CT gives information regarding tissue heterogeneity and patient contour. The treatment planning system uses this information to determine the dose distribution. Using the dose ratio between plans with and without heterogeneity correction the effective chest width can be calculated. The effective chest width is then used for calculating the treatment monitor units and bolus requirements. In this project the tissue heterogeneity corrections from two different treatment planning systems are compared for calculating the effective chest width. The treatment planning systems used were PinnacleTM, a 3D system that uses a convolution method to correct for tissue heterogeneity and calculate dose. The other system, RadplanTM, is a 2D algorithm that corrects for tissue heterogeneity using a modified Batho method and calculates dose using the Bentley - Milan Algorithm. Other possible differences between the treatment planning systems are also discussed. An anthropomorphic phantom was modified during this project. The chest slices were replaced with PerspexTM slices that had different sized cork and PerspexTM inserts to simulate different lung sizes. This allowed the effects of different lung size on the heterogeneity correction to be analysed. The phantom was CT scanned and the information used for the treatment plans. For each treatment planning system and each phantom plans were made with and without heterogeneity corrections. For each phantom the ratio between the plans from each system was used to calculate the effective chest width. The effective chest width was then used to calculate the number of monitor units to be delivered. The calculated dose per monitor unit at the extended TBI distance for the effective chest width from each planning system is then verified using thermoluminescent dosimeters (TLDs) in the unmodified phantom. The original phantom was used for the verification measurements as it had special slots for TLDs. The isodose distributions produced by each planning system are then verified using measurements from Kodak EDR2 radiographic film in the anthropomorphic phantom at isocentre. Further film measurements are made at the extended TBI treatment SAD. It was found that only the width of the lungs made any significant difference to the heterogeneity correction for each treatment planning system. The height and depth of the lungs will affect the dose at the calculation point from changes to the scattered radiation within the volume. However, since the dose from scattered radiation is only a fraction of that from the primary beam, the change in dose was not found to be significant. This is because the calculation point was positioned in the middle of the lungs, so the height and depth of the lungs didn't affect the dose at the calculation point. The dose per monitor unit calculated using the heterogeneity correction for each treatment planning system varied less than the accuracy of the TLD measurements. The isodose distributions measured by film showed reasonable agreement with those calculated by both treatment planning systems at isocentre and a more uniform distribution at the extended TBI treatment distance. The verification measurements showed that either treatment planning system could be used to calculate the heterogeneity correction and hence effective chest width for TBI treatment planning.