Zr/Hf ratios and other elemental data for 68 lunar basalt, KREEP, and glass samples were obtained by coincidence-anticoincidence counting and normal counting-techniques during instrumental neutron activation analyses. Zr/Hf ratios indicate a complementary relation between a mean Zr/Hf ratio of 41.0¿0.4 for KREEP and the mean ratios of 30--32 in Apollo 11, 12, and 17 mare basalts. The uniformity of Zr/Hf in 20 Apollo 14, 16, and 17 KREEP samples is consistent with petrogenesis of KREEP compositions from a single major source. The observed complementary ratios require greater Zr-Hf fractionation than what is predicted by lunar evolutionary models and presently known distribution coefficients. The difference requires that either (1) ilmenite and armalcolite, the most predominant phases conductive to Zr-Hf fractionation, begin to crystallize earlier and in greater amounts than is presently assumed, or (2) the relative partitioning of Hf over Zr in Ti-oxides and/or major minerals and/or spinels is greater than what is observed in natural or synthetic systems. Complementary ratios also support a primordial Zr/Hf ratio of ~35, represented by the E-C-O (enstatite, C2 and C3 carbonaceous, and ordinary) chondrites. Compiled Ta and Hf data from this laboratory indicate a well-defined Ta/Hf trend for most Apollo mare basalts and Apollo 14 KREEP. Apollo 11 B3 and Apollo 17 basalts define an alternate trend of higher Ta/Hf that we attribute to relatively higher Ta in assimilated Ti-bearing urKREEP residual phases due to the oxidation of Ta+4 to Ta+5 under elevated fO2. Apollo 15, 16, and 17 KREEP samples exhibit lower Ta/Hf ratios relative to the main trend and have a complementary relation to late-stage Ti-bearing cumulates with enhanced Ta/Hf. |