The presence of pathogens in various organic wastes requires careful attention to the composting process to minimise the chance of their survival. An optimised composting mixture from previous in-vessel composting studies (50% food waste, 40% manure, and 10% bulking agent) was evaluated using windrow composting. Conventional layering and mixing methods were used for constructing the windrows. Temperature was monitored as the indication of the efficiency of composting. Initial and final physico-chemical characteristics (moisture content, C/N ratio, pH, and volatile solids) and microbial survival were also determined. There was no significant difference between the observed peak temperatures in spreader and mixer windrows (probability P > 0.05). Leaching problems and structural instability were observed within the first 15-20 days. In spreader windrows, populations of faecal coliform, faecal streptococcus, Salmonella, and Escherichia coli O157:H7, in most probable number (MPN), were reduced from 418-432 MPN/g[dry compost] to 5.2-8.4, 78.8-396.0, 18.0-59.4, 5.2-5.7MPN/g[dry compost], whereas for mixer windrows were decreased from initial values of 431-447 MPN/g[dry compost] to final values of 7.8-8.0, 413.4-425.5, 0.5-33.6, 6.1-29.2 MPN/g[dry compost], respectively. The results from this study suggested the use of faecal coliforms as indicators for Salmonella (coefficient of determination r(2) = 0.94-0.99) and E coli O157:117 (r(2) = 0.89-0.98), but did not recommend faecal streptococcus as an indicator microorganism for Salmonella (r(2) = 0.37-0.39) and E coli O157:H7 (r(2) = 0.11-0.48). In conclusion, the optimum compost mixture determined from in-vessel system was composted more successfully in windrows than the in-vessel system due to higher temperatures maintained for a longer period of time and extended processing time in windrows. Further study is needed to eliminate the leaching and structural instability problems. (c) 2005 Silsoe Research Institute. All rights reserved Published by Elsevier Ltd.