Why pterosaurs weren't replaced by birds

Copyrighted from Richard Butler, Paul Barrett, Steven Nowbath & Paul Upchurch: "Estimating the effects of the rock record on pterosaur diversity patterns: implications forhypotheses of bird/pterosaur competitive replacement, SVP meeting abstracts2008, 59A":

"Pterosaurs formed an important component of terrestrial and marginal marine ecosystems during the Mesozoic, and were the first flying vertebrates. The fossil record appears to indicate a decline in pterosaur taxic diversity [ =species count] in the Late Cretaceous, followed by extinction at theCretaceous/Paleogene boundary. This reduction in species-richness has beenlinked to the ecological radiation of birds in the Early Cretaceous -- it has been proposed that early birds competitively excluded pterosaurs from many key niches. Hypotheses of competitive replacement are frequently posited based on the fossil record, but few have been tested in detail. Here we present a detailed examination of pterosaur diversity through time based upon a new comprehensive database of the spatial and temporal distribution of pterosaurs. We use this database to calculate taxic and phylogenetically corrected diversity estimates, and compare these estimates with a model describing temporal variation in rock availability. We use numbers of pterosaur-bearing formations (PBFs) as a proxy for rock availability;temporal variation in the number of PBFs is then used to generate a model in which rock availability is a perfect predictor of diversity. Both taxic and phylogenetic diversity curves are strongly correlated with numbers of PBFs, suggesting that a significant part of the signal contained within pterosaur diversity patterns may be controlled by geological and taphonomic megabiases rather than macroevolutionary processes. Moreover, significant differences between observed diversity and the diversity predicted by the rock availability model coincide with the occurrence of sites of exceptional preservation (Lagerstätten), again indicating major biases in the pterosaur fossil record. There is no evidence for a long-term decline in pterosaur diversity during the Cretaceous, although a reduction in species-richness might have occurred in the Late Cretaceous. Available data provide little support for the long-term competitive replacement of pterosaurs by birds."