We revisited questions about changes in the incidences of functional wood anatomical traits through geologic time and compared the incidences of these traits in the fossil record with modern wood anatomical diversity patterns in order to test classical (“Baileyan”) and more recent ecophyletic hypotheses of xylem evolution. We contrast patterns through time for tropical and higher (paleo)latitudes. Data are from the InsideWood database. There are striking differences between woods from high and mid latitudes versus tropical (paleo)-latitudes. At temperate and subtropical latitudes (Laurasia and high latitude Gondwana), the epoch by epoch time series supports the Baileyan transformation series of vessel-bearing woody angiosperms (basal woody angiosperms and eudicots): “primitive” features such as scalariform perforations, exclusively solitary vessels, apotracheal diffuse parenchyma and heterocellular rays abound in the Cretaceous and become much less frequent in the Cenozoic, especially post-Eocene. In contrast, in the paleotropics hardly any changes occur in the incidences - each epoch has an equally “modern” spectrum of wood anatomical attributes. Although climatic gradients from the poles to the equator were less steep in the Cretaceous than in the late Cenozoic, the wood anatomical differences between mid-high latitude woods and tropical woods were much more pronounced in the Cretaceous than in later epochs. This seeming paradox is discussed but we cannot resolve it.
We suggest that tropical conditions have accelerated xylem evolution towards greater hydraulic efficiency (simple perforations), biological defense and hydraulic repair (elaborate paratracheal parenchyma patterns) as evidenced by late Cretaceous tropical latitude woods having near modern incidences of almost all traits. At higher paleolatitudes of both the Northern and Southern Hemisphere “ancestral” features such as scalariform perforations were retained in cooler and frost-prone regions where they were not selected against in mesic habitats because of lower demands on conductive efficiency, and could even be advantageous in trapping freeze-thaw embolisms. The fossil wood record remains too incomplete for testing hypotheses on the wood anatomy of the earliest angiosperms. The low incidence of so-called “xerophobic” woods sensu Feild with scalariform perforations with numerous (over 40) closely spaced bars in the Cretaceous tropical fossil record is puzzling. At higher paleolatitudes such woods are common in the Cretaceous.
Ring porosity, an indicator of seasonal climates and deciduousness, occurs at low levels in the Cretaceous and Paleogene at higher paleolatitudes only, and reaches modern levels in the Miocene. In Neogene and Recent temperate Northern Hemisphere, wide vessels are virtually restricted to ring-porous woods. In the tropics, there is a low incidence of ring porosity throughout all epochs.
The fossil record indicates that ecophysiological adaptation to tropical or temperate conditions was already evident in the Cretaceous with considerable latitudinal differences.