Palaeocarpinus is an extinct genus of plants placed in the Betulaceae subfamily Coryloideae. The genus had a wide spread circumboreal distribution, with species found in North America, Europe, and Asia. The first appearance dates to the Early Paleocene and species are spread through the Paleocene and Eocene with the youngest species occurring at the Eocene-Oligocene boundary. A total of thirteen species have been named with a small grouping of referred specimens that have not been redescribed in the modern literature. Three species have been named from Asia, in 1996, 1998, and 2000. The type species and the first described species are both from Europe, being named in 1981 and 1859, though a North American species is reported in Northern Europe as well. North America has the largest diversity of the genus, with two groups of species known and several outliers. The first species was reported from the Okanagan Highlands in Canada and two additional species were named from the Highlands in 2003. A series of species is known from sites across the Great Plains-Rocky Mountains, with species named in 1992, 1996, 2004, and 2021. An additional species was described in 2021 from Oregon, and an undescribed fossil was reported from the Yukon but subsequently lost in transport. The first two species to be described were placed into several unrelated genera or into a form taxon for fruits before being moved into the genus. Palaeocarpinus is now placed basally into Coryloideae, with an unresolved relationship to the two modern coryloid tribes Coryleae and Carpinieae. The Palaeocarpinus plant was likely wind pollinated as with other Betulaceae species, while the fruits were possibly being wind or animal dispersed at maturity.
Distribution and age Palaeocarpinus species have a Circumboreal distribution across North America, Europe, and Asia. The ages of the species are concentrated in the early Paleogene, with most species being Paleocene to Eocene aged, though the youngest is possibly placed just into the Oligocene. Two occurrences of the genus in Europe date from the Paleocene, while the third is tentatively identified as Eocene. P. laciniata are type locality in strata of the Reading Formation outcropping in the Cold Ash quarry, near Newbury, Berkshire England. While the exact age is uncertain, the general age is late Paleocene (upper Paleocene of Crane, 1981). To the south in France, fossils recovered from the Menat Formation near Menat in Puy-de-Dôme have an agreed upon age of late Paleocene based on associated fossils and volcanic deposits. The northernmost occurrence of Palaeocarpinus are from the arctic Spitzbergen in western Svalbard. The fossils are produced by strata of the Storvola Formation which Golovneva (2002) reported outcropping along Adventfjorden, however, the site was later reported being on Prins Karls Forland by Julian E. Correa-Narvaez and Steven Manchester (2021). The exact age is uncertain, but based on surrounding stratigraphy plus palynological data, its most often cited to be Eocene.
Of the three Asian occurrences divided between Russia with two species and China with one, and includes both the oldest and youngest occurrences for the genus The western most site is south of the Irtysh river in the Altay Prefecture from rocks of the Wulungu Depression. Based on the associated plant flora of the formation, a Paleocene age is applied to the formation. The older of the two Russian localities, from which P. pacifica originates, is located near the lower course of the Amur in Khabarovsk Krai in the Russian Far East. The Malo-Mikhailovka Formation which outcrops near Puer Ridge between the Amur and the Strait of Tartary, has been identified as from the Early Paleocene's Danian age. In contrast, the Sikhote-Alin site on the Strait of Tartary's Buoy Bay, south of Nelma village is the youngest of the known Palaeocarpinus sites. The locality is part of the Kizi Group, with the site included in the uppermost strata of the Buoy Bay Member. Floral and mammalian correllaitions are indestinct and give possible ages between the Late Eocene and the Early Oligocene. One volcanic rock yielded a Potassium-argon date of 34 ± 0.8 million years old, within a million years of the Eocene-Oligocene boundary.
The greatest concentration of species and localities is found across the Western regions of North America, with species groups in the Great Plains-Rocky Mountains, in the Oregon interior, and in the greater Okanagan region of British Columbia and Washington. The northern most of the great plains localities is at Munce's Hill near Joffre, Alberta where Paskapoo Formation rocks of possible Lacombe member affinity outcrop in a road cut. The age of the Lacombe member is accepted to likely be Late Paleocene in the Tiffanian. Most of the Great Plains and Rocky Mountains sites are placed into various Fort Union Group formations. The two western North Dakota occurrences, at Almont and Beicegel Creek respectively, are both in the Tiffanian Sentinel Butte Formation of the Fort Union Group, while the eastern Montana sites around Big Timber are Fort Union Formation. All species in the Wyoming region are known from Fort Union Formation fossils, with most sites in Wyoming in the P-3 pollen zone and thus Middle Paleocene.
Palaeocarpinus specimens have been recovered from at least twelve individual sites in the Eocene Okanagan Highlands, with named species from the Coldwater Beds near Quilchena and Allenby Formation near Princeton, British Columbia, plus the Klondike Mountain Formation in Republic, Washington, northern Ferry County, Washington. The Okanagan highlands are aged between for the Quilchena site to for the Klondike Mountain Formations Tom Thumb Tuff member. Additional Palaeocarpinus sp. fossils were reported from Okanagan highlands sites of the Tranquille Formation near Cache Creek, British Columbia and the Horsefly Shales near Horsefly, British Columbia in 1908. Pigg, Manchester, and Wehr added the report of unplaced fruits in the Driftiwood Shales near Smithers, British Columbia, the Golden Promise site outside Republic, and Corkscrew Mountain near Toroda, Washington.
A tentative report of Palaeocarpinus fossils from the Yukon Paleogene was published by Matthew Vavrek et al. (2012). During field work in 2008 and 2009 fossils of latest Paleocene to possibly earliest Eocene were collected from an outcrop of the Bonnet Plume Formation southeast of Eagle Plains on the Peel River. Field identification of the specimens concluded them to belong to Palaeocarpinus but they were never able to be studied, being lost or destroyed in transport from the field site.
All four of the fossil sites in Oregon are placed within the Clarno Formation outcropping in the Blue Mountains area in the northeastern part of the state. Several methods of radiometric dating have been applied to that unit of the Clarno, with coinciding results. Fission track data from zircon crystals and 40Ar/39Ar data from plagioclase crystals have both yielded an Eocene age between , placing the fossils as from the Lutetian. The Clarno formation nut beds flora is also the type and only locality for the locule cast morphotaxon Kardiasperma parvum, to which the compression species Palaeocarpinus parva is notably similar to.
The genus Palaeocarpinus was first described and typified by paleobotanist Peter Crane in 1981 based on English fossils. Working on English material collected in riverine deposits of the Reading Formation exposed near Newbury, Berkshire, he encountered a group of nutlets reminiscent of both Carpinus and Corylus nuts. Given that the characters of the new material did not conform to either genus, but were a mix of the two, Crane opted to erect a new genus and species for them Palaeocarpinus laciniata. He chose the genus name as a combination of Palaeo- from the Greek "palaiós" meaning old, plus Carpinus. He coined the species name from the Latin word laciniata which means "lacerated or slashed", in reference to the deeply lobed appearance of the bracts. While the type series only encompasses the holotype, V.5982 and six paratype specimens, a total of 54 nutlets, 25 with attached bracts, were examined for the description. Coming from sites A and B in Cold Ash Quarry, all the examined specimens were deposited into the Department of Palaeontology, Natural History Museum, London.
The earliest reported fossils now attributed to Palaeocarpinus were from the Menat Formation, with Oswald Heer (1859) publishing a group of fossils as Anchietea borealis. Nearly 20 years later, another fossil from Menat was given the name Corylus lamottii by Gaston de Saporta (1877), but no description was supplied with the name so it was considered nomen nudum. Heers' Menat fossils were later redescribed and moved to Atriplex borealis by Louis Laurent. While this placement was unchanged until the 1981 publication of the genus Palaeocarpinus by Crane, the affinity of the fossils to Betulaceae was first suggested by Saporta and Antoine-Fortuné Marion (1885), who put forward the connection between the fruits and leaves called Corylus macquarii. Crane noted in his work on the English fossils that the Menat specimens were also from the same genus, but did not present a formal reassignment. His change was accepted by later workers, but an official redescription was not presented until Julian E. Correa-Narvaez and Steven Manchester (2021) in their monograph on the genus who officially moved the species to Palaeocarpinus borealis. Over the preceding century Heers original specimen, which he did not illustrate or note as holotype, was lost. As such, Correa-Narvaez & Manchester chose to designate "MEN 19" in the National Museum of Natural History, France as neotype.
Palaeocarpinus dentata is the first North American species described, being first discussed by David P. Penhallow in 1890. The original fruit was collected in 1888 by George Mercer Dawson during summer field work in the Okanagan Highlands region. Found near Stump Lake in the Quilchena area Dawson very briefly detailed the specimen and noted that it was superficially similar to Carpinus nutlets, but not referable to the genus. The bracts of the specimen are notably faint on the fruit and Dawson attributed it to being displaced pericarp, outer fleshy fruit tissues. Penhallow chose to name the species Carpolithus dentatus, the genus placement being a form genus for plant fruits and seeds of uncertain affinity. The specimen was deposited in the Redpath Museum by the Geological Survey of Canada as specimen 2.2420. The fossils from Stump Lake are part of the Coldwater Beds, and the fossils are related to the near by Quilchena flora. The specimen was later designated the lectotype for the species by Kathleen Pigg, Steven Manchester, and Wesley Wehr (2003), who moved the species from Carpolithus to Palaeocarpinus dentatus. They noted that the specific epithet dentata was likely a reference to the toothed outline of the bracts. The species name was further corrected by Correa and Manchester who noted Palaeocarpinus is a feminine Latin noun in construction, so the species name should have been shifted to P. dentata when moved from Carpolithus which is a neuter noun.
All the additional species of the genus have been described after 1990, though there are undescribed specimens which are from the early 1890's or which have been lost and never described.