Age measurements on glauconite by the K-Ar and Rb-Sr methods show a consistent variation with geological age with moderately small scatter. The results appear to fall 10-20% short of ages measured on micas associated with dated igneous rocks. K-Ar and Rb-Sr age values are closely concordant in the majority of samples. In no samples measured by the writers have glauconites yielded ages that are definitely higher than expected limits in the time scale. Glauconites have interlayered structures. The interlayering involves mainly nonexpandable 10 Aa layers and expandable (montmorillonitic) layers. The K content of glauconites is inversely proportional to the percent expandable layers. Furthermore, the percentage of expandable layers is greater in younger glauconites than in older. This suggests that the glauconite pellets continue to develop into purer mineral grains over long periods of time, with a decrease from 30% or more of expandable layers in young glauconite to about 10% in glauconite of early Paleozoic age. Almost without exception lower Paleozoic glauconite was found to contain more than 5% K. Common Sr is adsorbed on basal plane surfaces of the expandable layers and is easily removed by exchange. It is commonly roughly proportional to the abundance of these layers. It does not affect the Rb-Sr age measurements, except in an increase of precision error in analysis. The only K-bearing mineral phase of importance that would be expected to be interlayered would be 1M d muscovite. A study of the K-Ar ages in illitic shales suggests that the effect of minor amounts of interlayered authigenic muscovite phases in glauconite would not be significant because the age values shown by these phases are very similar to those given by glauconite. Detrital 2M 1 muscovite occluded in the grains would presumably increase the observed age. It is expectable that some K will be contained in the expandable layers, and that Ar generated in these may be lost. If the glauconite undergoes a slow purification with time, in which the expandable layers gradually become well ordered glauconite, up to about 20% of the mineral in lower Paleozoic samples may not be truly authigenic. It is believed that this is the cause of the apparently lower ages in the Paleozoic. Superimposed on this effect is an expectable diffusion loss of the Ar in materials that have been deeply buried, giving rise to some unusually low values. However, in the normal sample of glauconite from thin sedimentary sections of the typical platform facies, the estimated diffusion loss of Ar is too small to account for the observed decrease in age. The agreement of the Rb-Sr ages also bears out this conclusion. The writers conclude that glauconite may have a limited use in the dating of unknown sediments because of the consistency indicated in world-wide comparisons, on an empirical glauconite age-scale.