Much of the world's modern vertebrate diversity originated in a rapid surge of diversification in the early Paleogene, as survivors of theCretaceous–Paleogene extinction event took advantage of empty ecological niches left behind by the extinction of the non-avian dinosaurs, pterosaurs, marine reptiles, and primitive fish groups.Mammals continued to diversify from relatively small, simple forms into a highly diverse group ranging from small-bodied forms to very large ones, radiating into multipleorders and colonizing theair andmarine ecosystems by theEocene.[8]Birds, the only surviving group of dinosaurs,quickly diversified from the very fewneognath andpaleognath clades that survived the extinction event, also radiating into multiple orders, colonizing different ecosystems and achieving an extreme level of morphological diversity.[9]Percomorph fish, the most diverse group of vertebrates today, first appeared near the end of the Cretaceous but saw a very rapid radiation into their modern order and family-level diversity during the Paleogene, achieving a diverse array of morphologies.[10]
The Paleogene is marked by considerable changes in climate from thePaleocene–Eocene Thermal Maximum, through global cooling during the Eocene to the first appearance of permanent ice sheets in the Antarctic at the beginning of the Oligocene.[11]
The Paleocene is the first series/epoch of the Paleogene and lasted from 66.0 Ma to 56.0 Ma. It is divided into three stages: theDanian 66.0 - 61.6 Ma;Selandian 61.6 - 59.2 Ma; and,Thanetian 59.2 - 56.0 Ma.[13] The GSSP for the base of the Cenozoic, Paleogene and Paleocene is at Oued Djerfane, west ofEl Kef,Tunisia. It is marked by aniridium anomaly produced by anasteroid impact, and is associated with the Cretaceous–Paleogene extinction event. The boundary is defined as the rusty colored base of a 50 cm thickclay, which would have been deposited over only a few days. Similar layers are seen in marine and continental deposits worldwide. These layers include the iridium anomaly,microtektites,nickel-richspinel crystals andshocked quartz, all indicators of a major extraterrestrial impact. The remains of the crater are found atChicxulub on theYucatan Peninsula inMexico. The extinction of thenon-avian dinosaurs,ammonites and dramatic changes inmarine plankton and many other groups of organisms, are also used for correlation purposes.[13]
The Eocene is the second series/epoch of the Paleogene, and lasted from 56.0 Ma to 33.9 Ma. It is divided into four stages: theYpresian 56.0 Ma to 47.8 Ma;Lutetian 47.8 Ma to 41.2 Ma;Bartonian 41.2 Ma to 37.71 Ma; and,Priabonian 37.71 Ma to 33.9 Ma. The GSSP for the base of the Eocene is at Dababiya, nearLuxor,Egypt and is marked by the start of a significant variation in globalcarbon isotope ratios, produced by a major period of global warming. The change in climate was due to a rapid release of frozenmethane clathrates from seafloor sediments at the beginning of the Paleocene-Eocene thermal maximum (PETM).[13]
The Oligocene is the third and youngest series/epoch of the Paleogene, and lasted from 33.9 Ma to 23.03 Ma. It is divided into two stages: theRupelian 33.9 Ma to 27.82 Ma; and,Chattian 27.82 - 23.03 Ma. The GSSP for the base of the Oligocene is atMassignano, nearAncona,Italy. The extinction of thehantkeninid planktonicforaminifera is the key marker for the Eocene-Oligocene boundary, which was a time of climate cooling that led to widespread changes in fauna and flora.[13]
From the Late Cretaceous into the early Paleocene, Africa began to converge with Eurasia. The irregular outlines of the continental margins, including theAdriatic promontory (Adria) that extended north from the African plate, led to the development of several shortsubduction zones, rather than one long system.[15] In the western Mediterranean, the European plate was subducted southwards beneath the African plate, whilst in the eastern Mediterranean, Africa was subducted beneath Eurasia along a northward dipping subduction zone.[14][16] Convergence between theIberian and European plates led to thePyrenean orogeny[17] and, as Adria pushed northwards the Alps and Carpathian orogens began to develop.[18][16]
Present day tectonic map of southern Europe, North Africa and the Middle East, showing structures of the western Alpine-Himalayan orogenic belt.
The collision of Adria with Eurasia in the early Palaeocene was followed by a c.10 million year pause in the convergence of Africa and Eurasia, connected with the onset of the opening of the North Atlantic Ocean asGreenland rifted from the Eurasian plate in the Palaeocene.[18] Convergence rates between Africa and Eurasia increased again in the early Eocene and the remaining oceanic basins between Adria and Europe closed.[15][19]
Between about 40 and 30 Ma, subduction began along the western Mediterranean arc of the Tell, Rif, Betic and Apennine mountain chains. The rate of convergence was less than the subduction rate of the denselithosphere of the western Mediterranean androll-back of the subducting slab led to the arcuate structure of these mountain ranges.[15][17]
In the eastern Mediterranean, c. 35 Ma, the Anatolide-Tauride platform (northern part of Adria) began to enter thetrench leading to the development of the Dinarides, Hellenides and Tauride mountain chains as thepassive marginsediments of Adria were scrapped off onto the Eurasiacrust during subduction.[15][20]
TheZagros mountain belt stretches for c. 2000 km from the eastern border ofIraq to theMakran coast in southernIran. It formed as a result of the convergence and collision of theArabian and Eurasian plates as the Neotethys Ocean closed and is composed sediments scrapped from the descending Arabian Plate.[21][22]
From the Late Cretaceous, avolcanic arc developed on the Eurasia margin as the Neotethys crust was subducted beneath it. A separate intra-oceanic subduction zone in the Neotethys resulted in theobuction of ocean crust onto the Arabian margin in the Late Cretaceous to Paleocene, with break-off of the subducted oceanic plate close to the Arabian margin occurring during the Eocene.[21][22] Continental collision began during the Eocene c. 35 Ma and continued into the Oligocene to c. 26 Ma.[21][22]
Map showing the northwards drift of the Indian continent between 71 and 0 Ma. The leading edge of Greater India (not shown on the map) collided with the Eurasian plate c. 55 Ma, whilst India itself still lay to the south. (From: Dèzes, 1999)
The Indian continent rifted fromMadagascar at c. 83 Ma and drifted rapidly (c. 18 cm/yr in the Paleocene) northwards towards the southern margin of Eurasia. A rapid decrease in velocity to c. 5 cm/yr in the early Eocene records the collision of the Tethyan (Tibetan)Himalayas, the leading edge of Greater India, with theLhasa terrane ofTibet (southern Eurasian margin), along theIndus-Yarling-Zangbo suture zone.[14][23] To the south of this zone, the Himalaya are composed ofmetasedimentary rocks scraped off the now subducted Indian continental crust andmantle lithosphere as the collision progressed.[14]
Palaeomagnetic data place the present day Indian continent further south at the time of collision and decrease in plate velocity, indicating the presence of a large region to the north of India that has now been subducted beneath the Eurasian plate or incorporated into the mountain belt. This region, known as Greater India, formed byextension along the northern margin of India during the opening of the Neotethys. The Tethyan Himalaya block lay along its northern edge, with the Neotethys Ocean lying between it and southern Eurasia.[14][24]
Debate about the amount of deformation seen in the geological record in the India–Eurasia collision zone versus the size of Greater India, the timing and nature of the collision relative to the decrease in plate velocity, and explanations for the unusually high velocity of the Indian plate have led to several models for Greater India: 1) A Late Cretaceous to early Paleocene subduction zone may have lain between India and Eurasia in the Neotethys, dividing the region into two plates, subduction was followed by collision of India with Eurasia in the middle Eocene. In this model Greater India would have been less than 900 km wide;[24] 2) Greater India may have formed a single plate, several thousand kilometres wide, with the Tethyan Himalaya microcontinent separated from the Indian continent by anoceanic basin. The microcontinent collided with southern Eurasia c. 58 Ma (late Paleocene), whilst the velocity of the plate did not decrease until c. 50 Ma when subduction rates dropped as young, oceanic crust entered the subduction zone;[25] 3) This model assigns older dates to parts of Greater India, which changes its paleogeographic position relative to Eurasia and creates a Greater India formed of extended continental crust 2000–3000 km wide.[26]
During the Late Cretaceous to Paleogene, the northward movement of the Indian plate led to the highly oblique subduction of the Neotethys along the edge of the West Burma block and the development of a major north-southtransform fault along the margin of Southeast Asia to the south.[28][27] Between c. 60 and 50 Ma, the leading northeastern edge of Greater India collided with the West Burma block resulting indeformation andmetamorphism.[28] During the middle Eocene, north-dipping subduction resumed along the southern edge of Southeast Asia, from west Sumatra to West Sulawesi, as the Australian plate drifted slowly northwards.[27]
Collision between India and the West Burma block was complete by the late Oligocene. As the India-Eurasia collision continued, movement of material away from the collision zone was accommodated along, and extended, the already existing majorstrike slip systems of the region.[28]
During the Paleocene, seafloor spreading along theMid-Atlantic Ridge propagated from the Central Atlantic northwards between North America and Greenland in theLabrador Sea (c. 62 Ma) andBaffin Bay (c. 57 Ma), and, by the early Eocene (c. 54 Ma), into the northeastern Atlantic between Greenland and Eurasia.[14][29] Extension between North America and Eurasia, also in the early Eocene, led to the opening of the Eurasian Basin across the Arctic, which was linked to the Baffin Bay Ridge and Mid-Atlantic Ridge to the south via major strike slip faults.[14][30]
From the Eocene and into the early Oligocene, Greenland acted as an independent plate moving northwards and rotating anticlockwise. This led to compression across theCanadian Arctic Archipelago,Svalbard and northern Greenland resulting in theEureka orogeny.[14][30] From c. 47 Ma, the eastern margin of Greenland was cut by the Reykjanes Ridge (the northeastern branch of the Mid-Atlantic Ridge) propagating northwards and splitting off theJan Mayen microcontinent.[14]
After c. 33 Ma seafloor spreading in Labrador Sea and Baffin Bay gradually ceased and seafloor spreading focused along the northeast Atlantic. By the late Oligocene, the plate boundary between North America and Eurasia was established along the Mid-Atlantic Ridge, with Greenland attached to the North American plate again, and the Jan Mayen microcontinent part of the Eurasian plate, where its remains now lie to the east and possibly beneath the southeast of Iceland.[14][30]
A Paleogene-aged basaltic lava flow on the Isle ofStaffa, Scotland (person standing on cliff top for scale). The bottom section of this cliff isvolcaniclastic rock. The middle and top sections are two parts of a single basaltic lava flow; each part of the lava flow cooled differently, forming rock with different characteristics. The middle layer shows spectacularcolumnar jointing resulting from relatively slow cooling; the top layer has very irregular closely-spaced joints caused by more rapid cooling.[31]
TheNorth Atlantic Igneous Province stretches across the Greenland and northwest European margins and is associated with the proto-Icelandicmantle plume, which rose beneath the Greenland lithosphere at c. 65 Ma.[30] There were two main phases of volcanic activity with peaks at c. 60 Ma and c. 55 Ma.Magmatism in the British and Northwest Atlantic volcanic provinces occurred mainly in the early Palaeocene, the latter associated with an increased spreading rate in the Labrador Sea, whilst northeast Atlantic magmatism occurred mainly during the early Eocene and is associated with a change in the spreading direction in the Labrador Sea and the northward drift of Greenland. The locations of the magmatism coincide with the intersection of propagating the rifts and large-scale, pre-existing lithospheric structures, which acted as channels to the surface for themagma.[30][32]
The arrival of the proto-Iceland plume has been considered the driving mechanism for rifting in the North Atlantic. However, that rifting and initial seafloor spreading occurred prior to the arrival of the plume, large scale magmatism occurred at a distance to rifting, and that rifting propagated towards, rather than away from the plume, has led to the suggestion the plume and associated magmatism may have been a result, rather than a cause, of the plate tectonic forces that led to the propagation of rifting from the Central to the North Atlantic.[30][32]
Mountain building continued along theNorth America Cordillera in response to subduction of theFarallon plate beneath the North American Plate. Along the central section of the North American margin, crustal shortening of the Cretaceous to PaleoceneSevier orogen lessened and deformation moved eastward. The decreasing dip of the subducting Farallon plate led to aflat-slab segment that increased friction between this and the base of the North American Plate. The resultingLaramide orogeny, which began the development of theRocky Mountains, was a broad zone ofthick-skinned deformation, withfaults extending to mid-crustal depths and the uplift ofbasement rocks that lay to the east of the Sevier belt, and more than 700km from the trench.[33][34] With the Laramide uplift theWestern Interior Seaway was divided and then retreated.[33]
During the mid to late Eocene (50–35 Ma), plate convergence rates decreased and the dip of the Farallon slab began to steepen. Uplift ceased and the region largely levelled byerosion. By the Oligocene, convergence gave way to extension, rifting and widespread volcanism across the Laramide belt.[33][34]
Ocean-continent convergence accommodated by east dipping subduction zone of the Farallon plate beneath the western edge of South America continued from the Mesozoic.[35]
Over the Paleogene, changes in plate motion and episodes of regional slab shallowing and steepening resulted in variations in the magnitude of crustal shortening and amounts of magmatism along the length of theAndes.[35] In the Northern Andes, anoceanic plateau with volcanic arc was accreted during the latest Cretaceous and Paleocene, whilst the Central Andes were dominated by the subduction of oceanic crust and the Southern Andes were impacted by the subduction of the Farallon-East Antarctic ocean ridge.[36][37]
TheCaribbean plate is largely composed of oceanic crust of theCaribbean Large Igneous Province that formed during the Late Cretaceous.[37] During the Late Cretaceous to Paleocene, subduction of Atlantic crust was established along its northern margin, whilst to the southwest, an island arc collided with the northern Andes forming an east dipping subduction zone where Caribbean lithosphere was subducted beneath the South American margin.[38]
During the Eocene (c. 45 Ma), subduction of the Farallon plate along the Central American subduction zone was (re)established.[37] Subduction along the northern section of the Caribbean volcanic arc ceased as the Bahamas carbonate platform collided with Cuba and was replaced by strike-slip movements as a transform fault, extending from the Mid-Atlantic Ridge, connected with the northern boundary of the Caribbean Plate. Subduction now focused along the southern Caribbean arc (Lesser Antilles).[37][39]
At the beginning of the Paleogene, the Pacific Ocean consisted of the Pacific, Farallon,Kula andIzanagi plates. The central Pacific plate grew by seafloor spreading as the other three plates were subducted and broken up. In the southern Pacific, seafloor spreading continued from the Late Cretaceous across the Pacific–Antarctic, Pacific-Farallon and Farallon–Antarctic mid ocean ridges.[14]
The Izanagi-Pacific spreading ridge lay nearly parallel to the East Asian subduction zone and between 60–50 Ma the spreading ridge began to be subducted. By c. 50 Ma, the Pacific plate was no longer surrounded by spreading ridges, but had a subduction zone along its western edge. This changed the forces acting on the Pacific plate and led to a major reorganisation of plate motions across the entire Pacific region.[40] The resulting changes in stress between the Pacific andPhilippine Sea plates initiated subduction along theIzu-Bonin-Mariana andTonga-Kermadec arcs.[40][37]
Subduction of the Farallon plate beneath the American plates continued from the Late Cretaceous.[14] The Kula-Farallon spreading ridge lay to its north until the Eocene (c. 55 Ma), when the northern section of the plate split forming theVancouver/Juan de Fuca plate.[37] In the Oligocene (c. 28 Ma), the first segment of the Pacific–Farallon spreading ridge entered the North American subduction zone nearBaja California[41] leading to major strike-slip movements and the formation of theSan Andreas Fault.[14] At the Paleogene-Neogene boundary, spreading ceased between the Pacific and Farallon plates and the Farallon plate split again forming the present dateNazca andCocos plates.[37][41]
The Kula plate lay between Pacific plate and North America. To the north and northwest it was being subducted beneath theAleutian trench.[14][37] Spreading between the Kula and Pacific and Farallon plates ceased c. 40 Ma and the Kula plate became part of the Pacific Plate.[14][37]
TheHawaiian-Emperor seamount chain formed above theHawaiian hotspot. Originally thought to be stationary within the mantle, the hotspot is now considered to have drifted south during the Paleocene to early Eocene, as the Pacific plate moved north. At c. 47 Ma, movement of the hotspot ceased and the Pacific plate motion changed from northward to northwestward in response to the onset of subduction along its western margin. This resulted in a 60 degree bend in the seamount chain. Other seamount chains related to hotspots in the South Pacific show a similar change in orientation at this time.[42]
Slow seafloor spreading continued between Australia and East Antarctica. Shallow water channels probably developed south of Tasmania opening theTasmanian Passage in the Eocene and deep ocean routes opening from the mid Oligocene. Rifting between the Antarctic Peninsula and the southern tip of South America formed theDrake Passage and opened the Southern Ocean also during this time, completing the breakup of Gondwana. The opening of these passages and the creation of the Southern Ocean established theAntarctic Circumpolar Current.Glaciers began to build across the Antarctica continent that now lay isolated in the south polar region and surrounded by cold ocean waters. These changes contributed to the fall in global temperatures and the beginning of icehouse conditions.[33]
Paleogene flood basalts on the Ethiopian Plateau with the Afar Depression in the background.
Extensional stresses from the subduction zone along the northern Neotethys resulted in rifting between Africa and Arabia, forming theGulf of Aden in the late Eocene.[43] To the west, in the early Oligocene,flood basalts erupted acrossEthiopia, northeastSudan and southwestYemen as theAfar mantle plume began to impact the base of the African lithosphere.[14][43] Rifting across the southernRed Sea began in the mid Oligocene, and across the central and northern Red Sea regions in the late Oligocene and early Miocene.[43]
Climatic conditions varied considerably during the Paleogene. After the disruption of theChicxulub impact settled, a period of cool and dry conditions continued from the Late Cretaceous. At the Paleocene-Eocene boundary global temperatures rose rapidly with the onset of thePaleocene-Eocene Thermal Maximum (PETM).[14] By the middle Eocene, temperatures began to drop again and by the late Eocene (c. 37 Ma) had decreased sufficiently for ice sheets to form in Antarctica. The global climate entered icehouse conditions at the Eocene-Oligocene boundary and the present dayLate Cenozoic ice age began.[33]
The Paleogene began with the brief but intense "impact winter" caused by theChicxulub impact, which was followed by an abrupt period of warming. After temperatures stabilised, the steady cooling and drying of the Late Cretaceous-Early Paleogene Cool Interval that had spanned the last twoages of theLate Cretaceous continued,[11] with only the brief interruption of theLatest Danian Event (c. 62.2 Ma) when global temperatures rose.[44][45][46] There is no evidence for ice sheets at the poles during the Paleocene.[14]
The relatively cool conditions were brought to an end by the Thanetian Thermal Event, and the beginning of the PETM.[11] This was one of the warmest times of the Phanerozoic eon, during which global mean surface temperatures increased to 31.6 °C.[47] According to a study published in 2018, from about 56 to 48 Ma, annual air temperatures over land and at mid-latitude averaged about 23–29 °C (± 4.7 °C).[48][49][50] For comparison, this was 10 to 15 °C higher than the current annual mean temperatures in these areas.[50]
This rapid rise in global temperatures and intense greenhouse conditions were due to a sudden increase in levels of atmosphericcarbon dioxide (CO2) and othergreenhouse gases.[33] An accompanying rise in humidity is reflected in an increase inkaolinite in sediments, which forms bychemical weathering in hot, humid conditions.[14] Tropical and subtropical forests flourished and extended into polar regions. Water vapour (a greenhouse gas) associated with these forests also contributed to the greenhouse conditions.[33]
The initial rise in global temperatures was related to the intrusion of magmaticsills into organic-rich sediments during volcanic activity in the North Atlantic Igneous Province, between about 56 and 54 Ma, which rapidly released large amounts of greenhouse gases into the atmosphere.[14] This warming led to melting of frozenmethane hydrates oncontinental slopes adding further greenhouses gases. It also reduced the rate of burial of organic matter as higher temperatures accelerated the rate of bacterialdecomposition which released CO2 back into the oceans.[33]
The (relatively) sudden climatic changes associated with the PETM resulted in the extinction of some groups of fauna and flora and the rise of others. For example, with the warming of the Arctic Ocean, around 70% of deep seaforaminifera species went extinct,[33] whilst on land many modern mammals, includingprimates, appeared.[51] Fluctuating sea levels meant, during low stands, a land bridge formed across theBering Straits between North America and Eurasia allowing the movement of land animals between the two continents.[14]
The PETM was followed by the less severeEocene Thermal Maximum 2 (c. 53.69 Ma),[52] and the Eocene Thermal Maximum 3 (c. 53 Ma). The early Eocene warm conditions were brought to an end by theAzolla event. This change of climate at about 48.5 Ma, is believed to have been caused by a proliferation of aquatic ferns from the genusAzolla, resulting in the sequestering of large amounts of CO2 from the atmosphere by the plants. From this time until about 34 Ma, there was a slow cooling trend known as the Middle-Late Eocene Cooling.[11] As temperatures dropped at high latitudes the presence of cold waterdiatoms suggests sea ice was able to form in winter in the Arctic Ocean,[33] and by the late Eocene (c. 37 Ma) there is evidence of glaciation in Antarctica.[14]
Changes in deep ocean currents, as Australia and South America moved away from Antarctica opening the Drake and Tasmanian passages, were responsible for the drop in global temperatures. The warm waters of the South Atlantic, Indian and South Pacific oceans extended southward into the opening Southern Ocean and became part of the cold circumpolar current. Dense polar waters sank into the deep oceans and moved northwards, reducing global ocean temperatures. This cooling may have occurred over less than 100,000 years and resulted in a widespread extinction in marine life. By the Eocene-Oligocene boundary, sediments deposited in the ocean from glaciers indicate the presence of an ice sheet in western Antarctica that extended to the ocean.[33]
The development of the circumpolar current led to changes in the oceans, which in turn reduced atmospheric CO2 further. Increasing upwellings of cold water stimulated the productivity ofphytoplankton, and the cooler waters reduced the rate of bacterial decay of organic matter and promoted the growth of methane hydrates in marine sediments. This created a positive feedback cycle where global cooling reduced atmospheric CO2 and this reduction in CO2 lead to changes which further lowered global temperatures. The decrease inevaporation from the cooler oceans also reduced moisture in the atmosphere and increased aridity. By the early Oligocene, the North American and Eurasian tropical and subtropical forests were replaced by dry woodlands and widespread grasslands.[33]
The Early Oligocene Glacial Maximum lasted for about 200,000 years,[53] and the global mean surface temperature continued to decrease gradually during theRupelian.[11] A drop in global sea levels during the mid Oligocene indicates major growth of the Antarctic glacial ice sheet.[33] In theLate Oligocene, global temperatures began to warm slightly, though they continued to be significantly lower than during the previousepochs of the Paleogene and polar ice remained.[11]
Restoration ofPalaeotherium, which diversified in warmer climates
Tropical taxa diversified faster than those at higher latitudes after the Cretaceous–Paleogene extinction event, resulting in the development of a significant latitudinal diversity gradient.[54]
Mammals began a rapiddiversification during this period. After the Cretaceous–Paleogene extinction event, which saw the demise of the non-aviandinosaurs, mammals began to evolve from a few small and generalized forms into most of the modern varieties we see presently. Some of these mammals evolved into large forms that dominated the land, while others became capable of living inmarine, specialized terrestrial, and airborne environments. Those that adapted to the oceans became moderncetaceans andsirenians, while those that adapted to trees becameprimates, the group to which humans belong.
Myctophids first appeared in the Late Palaeocene or Early Eocene, and during the Eocene and most of the Oligocene were restricted to shelf seas before expanding their range into the open ocean during the warm climatic interval at the end of the Oligocene.[55]
Pronounced cooling in theOligocene resulted in a massive floral shift, and many extant modern plants arose during this time.Grasses and herbs, such asArtemisia, began to proliferate, at the expense of tropical plants, which began to decrease.Conifer forests developed in mountainous areas. This cooling trend continued, with major fluctuation, until the end of thePleistocene period.[56] This evidence for this floral shift is found in thepalynological record.[57]
^Meredith, R. W.; Janecka, J. E.; Gatesy, J.; Ryder, O. A.; Fisher, C. A.; Teeling, E. C.; Goodbla, A.; Eizirik, E.; Simao, T. L. L.; Stadler, T.; Rabosky, D. L.; Honeycutt, R. L.; Flynn, J. J.; Ingram, C. M.; Steiner, C.; Williams, T. L.; Robinson, T. J.; Burk-Herrick, A.; Westerman, M.; Ayoub, N. A.; Springer, M. S.; Murphy, W. J. (28 October 2011). "Impacts of the Cretaceous Terrestrial Revolution and KPg Extinction on Mammal Diversification".Science.334 (6055):521–524.Bibcode:2011Sci...334..521M.doi:10.1126/science.1211028.PMID21940861.S2CID38120449.
^abcdVandenberghe, N.; Hilgen, F. J.; Speijer, R. P.; Ogg, J. G.; Gradstein, F. M.; Hammer, O.; Hollis, C. J.; Hooker, J. J. (2012-01-01), Gradstein, Felix M.; Ogg, James G.; Schmitz, Mark D.; Ogg, Gabi M. (eds.),"Chapter 28 - The Paleogene Period",The Geologic Time Scale, Boston: Elsevier, pp. 855–921,ISBN978-0-444-59425-9,archived from the original on 2015-09-24, retrieved2024-07-15
^abcdefghijklmnopqrstuvwTorsvik, Trond H.; Cocks, Leonard Robert Morrison (2017).Earth history and palaeogeography. Cambridge: Cambridge university press.ISBN978-1-107-10532-4.
.png)
.jpg)
