MINERALOGICAL MAGAZINE, VOLUME 62A Contrasting REE geochemical features between Archaean and

described the sedimentary REE geochemical transi-tion from Archaean to Proterozoic, i.e. from positive or slight Eu-anomaly toclear Eu-depletion. Gibbs et al. (1986), Condie (1993) and other researchers have proved wrong the views of Taylor and Mclennan. They argue that the transitions have been deduced from an incorrect comparison of Proterozoic plat-form-type sediments with those from Archaean greenstone, as if comparing an orange with an apple. To verify their views, we have studied the REE geochemical characteristics ofkhondalite series of the Archaean and Proterozoic of the North China Craton (NCC) and found great differences between them. Ages and spatial distribution of the khondalite series in the North China Craton All the khondalite series of the NCC had long been regarded as Archaean. Since Chen (1990) argued that most of the khondalite series in China formed during 2300-2050 Ma, nearly all the khondalite series have been universally accepted as Palaeoproterozoic n age, except for the Fuping Group in the Taihang Mountains. The Fuping Group is distributed in the core of the NCC, the others along the its rim. Possessing less graphite and less BIF, the Fuping Group is different from the Proterozoic khondalite series. The other geochemical features of REE and trace elements are also consistent with PAAS (Taylor and Mclennan, 1985). Marbles of the Archaean khondalite series, Fuping Group, are characterized by positive Eu-anomaly. Neither felsic gneisses nor amphibole gneisses how clear Eu-depletion, their average Eu/Eu * values being 0.89 and 0.85 respectively. The REE patterns are similar to Archaean shales (Taylor and Mclennan