This study provides the first quantitative assessment of seasonal variation in indoor PM2.5 in the high-air pollution risk and densely populated slums in Mumbai, India and estimates the contributions of indoor and outdoor sources to it. Indoor and outdoor 24-h averaged gravimetric PM2.5 (n(ind )= 20 homes, n(out) = 10 days) was measured during summer (May- June 2016) and winter (January-March 2017). During the summer, real-time PM2.5, Lung Deposited Surface Area and Black Carbon were also measured (n(ind) = 8,n(out)= 8). Additionally, detailed questionnaire surveys on housing, household and health characteristics were conducted in similar to 500 homes of the seven Mumbai slums. More than 60% of the homes in slums had no separate kitchen or cross-ventilation, especially in low socioeconomic status homes. Respiratory and cardiovascular diseases were reported in 15% and 6% of the households, respectively, with a higher prevalence in homes burning mosquito coils. Significantly higher indoor PM2.5 was observed during winter (111 +/- 30 mu g/m(3)) than summer (36 +/- 12 mu g/m(3)). Although liquefied petroleum gas was the only indoor cooking fuel reported, the winter-time indoor levels were similar or higher than the concentrations observed in other urban slum homes using biomass fuels for cooking. This could be attributed to the alarmingly high winter-time ambient PM2.5 (192 +/- 80 mu g/m(3)) and its larger contribution to indoor PM2.5 (81%). On the other hand, the contribution of indoor and local outdoor sources was significantly higher for Lung Deposited Surface Area (33%) and Black Carbon (43%) compared to PM2.5 (19%), which are more fine and toxicity-relevant particle metrics. [GRAPHICS] .