ATP activated [Ca2+](i) oscillations were measured in single rat megakaryocytes using fluorescence ratio microscopy. With increasing ATP concentration the duration of the [Ca2+](i) oscillations increased, however, there was considerable variation from cell to cell in the absolute value of the peak [Ca2+], and the frequency and duration of the oscillations. This variation depended, in part, on the level of Fura-2 loading suggesting that megakaryocytes are sensitive to buffering of [Ca2+](i) by Fura-2. Agents, that increase the level of intracellular cGMP (sodium nitroprusside and 8-pCPT-cGMP) or cAMP (prostacyclin, IBMX, forskolin and 8-bromo-cAMP) inhibited [Ca2+](i) oscillations. Despite the large cell to cell variation in the patterns of [Ca2+](i) oscillations, reapplication of the agents that elevated cAMP or cGMP inhibited the oscillations similarly. Using video rate fluorescence ratio imaging we found that the agonist induced [Ca2+](i) oscillations were the result of a well-defined [Ca2+](i) wave, which spread across the cell with an average speed of about 35 mu m/s, during the rising phase of each oscillatory spike. After reaching a peak, [Ca2+], decreased uniformly across the whole cell during the falling phase of the spike. Analysis of the temperature dependence of [Ca2+](i) waves showed that the rate of [Ca2+](i) decay exhibited a strong temperature dependence (Q(10) similar to 4), whereas, the rate of rise exhibited a weak temperature dependence (Q(10) similar to 1.3), suggesting, that the rate limiting process for [Ca2+](i) wave propagation in rat megakaryocytes is the rate of [Ca2+](i) diffusion.