DID ALPHA DIVERSITY TRIPLE BETWEEN THE PALEOZOIC AND CENOZOIC?

Sampling-standardized analyses using Paleobiology Database data have shown that Cenozoic global diversity was never clearly greater than early Paleozoic diversity. However, other sampling-standardized analyses claim to show a tripling or even quadrupling of alpha diversity. The discrepancy is very problematic because global diversity is just alpha times beta diversity, and there is no reason to think that beta was three times higher in the Paleozoic. A possible solution is taphonomic. Lithification makes collection of small and fragile shells difficult, and may co-occur with diagenetic removal of aragonitic shells. Samples from lithified sediments are extraordinarily common in the Paleozoic, but less so in the Cenozoic. Eighty-three unlithified Cenozoic samples with abundance data have a mean rarefied richness at the 100 specimen level of 20.0 species, much like previously published estimates. Nine apparently lithified carbonate Cenozoic samples have a mean of 10.1 despite also preserving non-molluscan taxa and being subtropical. Furthermore, the mean unrarefied richness of all 239 unlithified Cenozoic samples in the Database is 31.4, compared with just 13.4 for all 379 lithified Cenozoic samples. Thus, diversity is inflated by a factor of two, the same as the actual ratio between Cenozoic and Paleozoic samples in other sampling-standardized studies. The bias is comparable to the possible 1.7 ratio generated by a recent simulation of an aragonitic dissolution effect. Meanwhile, it has been argued that observed Cenozoic values should be doubled to correct for high diversity in the tropics. However, only about 37% of modern shelf area is tropical, and many "temperate" Cenozoic samples are from diverse subtropical regions like the Gulf Coast, so a downwards correction for temperate environments also should be made. The net effect is therefore likely to be quite small, and hardly enough to cause a serious problem for estimates of global diversity.