Mauna Kea lava flows cored in the HilIo hole range in age from <200 ka to about 400 ka based on 40Ar/39Ar incremental heating and K-Ar analyses of 16 groundmass samples and one coexisting plagioclase. The lavas, all subaerially deposited, include a lower section consisting only of tholeiitic basalts and an upper section of interbedded alkalic, transitional tholeiitic, and tholeiitic basalts. The lower section has yielded predominantly complex, discordant 40Ar/39Ar age spectra that result from mobility of 40Ar and perhaps K, the presence of excess 40Ar, and redistribution of 39Ar by recoil. Comparison of K-Ar ages with 40Ar/39Ar integrated ages indicates that some of these samples have also lost 39Ar. Nevertheless, two plateau ages of 391 ?? 40 and 400 ?? 26 ka from deep in the hole, combined with data from the upper section, show that the tholeiitic section accumulated at an average rate of about 7 to 8 m/kyr and has an mean recurrence interval of 0.5 kyr/flow unit. Samples from the upper section yield relatively precise 40Ar/39Ar plateau and isotope correlation ages of 326 ?? 23, 241 ?? 5, 232 ?? 4, and 199 ?? 9 ka for depths of -415.7 m to -299.2 m. Within their uncertainty, these ages define a linear relationship with depth, with an average accumulation rate of 0.9 m/kyr and an average recurrence interval of 4.8 kyr/flow unit. The top of the Mauna Kea sequence at -280 m must be older than the plateau age of 132 ?? 32 ka, obtained for the basal Mauna Loa flow in the corehole. The upward decrease in lava accumulation rate is a consequence of the decreasing magma supply available to Mauna Kea as it rode the Pacific plate away from its magma source, the Hawaiian mantle plume. The age-depth relation in the core hole may be used to test and refine models that relate the growth of Mauna Kea to the thermal and compositional structure of the mantle plume.