Lunar thermal history revisited.
New information is used to demonstrate that better models for the thermal history of the moon are required. As a first step, account is taken of (1) a nonuniform initial composition in terms of fraction of low melting to high melting phase present, and for variation in the uranium, potassium, and thorium contents as a function of depth, (2) partitioning of the radioactive elements between the melt and the solid phases, and (3) a cutoff value of melt which must be exceeded before magma can move to the surface. The results of several attempts to determine whether reasonable conditions, composition, and thermal properties can be expected to give rise to two separate periods of volcanism are discussed. Two models with somewhat different distributions of radioactive heat sources and different conductivities are examined.