The first Mesozoic mammal was discovered in 1764, although its significance was not appreciated until almost a century later (Owen, 1871). The first scientifically documented Mesozoic mammal, now known as Phasco-lotherium bucklandi, was discovered in 1812 in a masonry quarry in the Middle Jurassic Stonesfield slates near Head-ington, England. The fossil was later deposited at Oxford University, where it was recognized by Baron Georges Cuvier to be mammalian in 1818, announced as such by William Buckland (1824), and formally described by Broderip (1828). However, prior to the appearance of the definitive study by Richard Owen (1842), it was generally believed that no mammals existed before the Tertiary (see historical reviews by Desmond, 1984 and Rowe, 1999).
During the late nineteenth century, Mesozoic mammals were discovered in Europe, especially in England (Owen, 1871), and in North America (Marsh, 1879c, 1881, 1887, 1889a; Osborn, 1887a,b, 1888a,b). Knowledge of these fossils was summarized by George Gaylord Simpson in two of the most seminal paleontological monographs of the twentieth century: A Catalogue of the Mesozoic Mammalia in the Geological Department ofthe British Museum (1928a) and American Mesozoic Mammalia (1929a). In these monographs, Simpson collectively recognized 39 genera of Mesozoic mammals known from Europe and North America; of these, 12 "genera" are now regarded as junior synonyms or otherwise invalid. Fifty years later, Jason A. Lillegraven, Zofia Kielan-Jaworowska, and William A. Clemens (1979) edited the book Mesozoic Mammals: The First Two-Thirds of Mammalian History, in which they listed 116 valid genera of Mesozoic mammals worldwide, with all but a few from Europe, Asia, and North America.
The 23 years since publication of the Mesozoic mammal book by Lillegraven et al. (1979) have witnessed a truly phenomenal growth in knowledge of Mesozoic mammals. By our count, the number of known Mesozoic genera increased to 283 by the year 2000 (chapter 2). New Mesozoic mammals discovered in the past 20 years are one and one-half times more than the total of those known to science from the previous 200 years combined (figure 1.3).
Our knowledge of the earliest mammals has greatly improved in many ways apart from the total number of described taxa—notably anatomical information and geographic/temporal sampling. Many new Mesozoic mammals have been found in areas where the record was previously blank, or nearly so. Most intriguing of these discoveries are those from the Gondwanan continents. In 1979, the only evidence of Mesozoic mammals from the entire South American continent were the footprints designated Ameghinichnus patagonicus from the Callovian or Oxfordian of Santa Cruz Province, Argentina (Casa-miquela, 1964), and some fragmentary fossils, which may be of Tertiary rather than Cretaceous age, from Laguna Umayo, Peru (chapter 2). Since that time, highly diverse Cretaceous and some Jurassic mammals have been discovered at several sites in Argentina, owing to the great enterprise of South American paleontologists, especially José F. Bonaparte, Rosendo Pascual, and their respective collaborators. Mesozoic mammals from South America now include purported Docodonta, australosphenidans, eutri-conodontans, multituberculates, "symmetrodontans," dryolestoid "eupantotherians," prototribosphenidans, and enigmatic gondwanatherians (see Bonaparte and Rougier, 1987; Bonaparte, 1990, 1994; Kielan-Jaworowska and Bonaparte, 1996; Pascual et al., 2000; Pascual and Goin, 2001; Rauhut et al., 2002; and chapters 5-11 for reviews). Pascual and co-workers also discovered a toothed mono-treme from the Paleocene of Argentina (Pascual et al., 1992a, 2002).
Other important discoveries have come from Australia, previously a totally blank region insofar as Mesozoic mammals were concerned. In 1985, Michael Archer and colleagues reported the discovery of the first toothed monotreme from the Cretaceous, Steropodon; two more Early Cretaceous monotremes (Kollikodon and Teino-lophos) have since been described (Flannery et al., 1995; Rich, Vickers-Rich et al., 2001; see also chapter 6). Since 1997 the able team of Thomas Rich, Patricia Vickers-Rich, and their colleagues has been recovering an assemblage of tribosphenic mammals from the Lower Cretaceous of Australia. These authors have argued that these Australian tribosphenic mammals are placentals (Rich et al., 1997; see also Rich and Vickers-Rich, 1999 and references therein). Mammals with tribosphenic molars have also been reported from the Middle Jurassic of Madagascar by Flynn et al. (1999) and from the Middle or Late Jurassic of Argentina by Rauhut et al. (2002). Our team (Luo, Kielan-Jaworowska, and Cifelli, 2001; Luo et al., 2002; see also
figure 1.3. Number of Mesozoic mammal genera known through time. This graph shows the total number of described genera (line joined by diamonds) and the number described during each 10-year interval (line joined by squares) from 1830 to 2003. Three distinct episodes of intense scientific activity in this field can be recognized. The first, in the 1870s and 1880s, corresponds to Owen's major study (1871) of British mammals from the Purbeck and Stonesfield slates, together with publication, largely by Marsh, of the first known Jurassic and Cretaceous mammals from North America (e.g.,Marsh, 1879b, 1887,1889a). The second took place in the 1920s, when Simpson (1928a, 1929a) reviewed all Mesozoic mammal fossils then known and added a number of new genera, mainly based on previously unstudied or understudied fossils from Como Quarry 9. The third, which continues unabated, began in the 1960s and early 1970s (see References). Source: original.
Kielan-Jaworowska et al., 1998, 2002) challenged the placental affinities of southern Mesozoic mammals with tribosphenic dentitions. We argued that Gondwanan tri-bosphenidans are more closely related to toothed mono-tremes than to placentals and marsupials and that the functionally adaptive features on tribosphenic molars (a protocone on upper molars opposing a basined talonid on lowers) are homoplastic. According to this hypothesis, tribosphenic mammals have dual origins on the Laurasian and Gondwanan continents. The southern tribosphenic mammals belong to a clade in their own right, designated australosphenidans, as opposed to the northern Mesozoic mammals with tribosphenic molars, designated bore-
osphenidans, and we follow this interpretation (see chapters 6,11, and 15).
Mesozoic mammals previously known from Africa include morganucodontans from the Liassic Stormberg Series of South Africa and a toothless dentary of Bran-catherulum from the Upper Jurassic of Tendaguru in Tanzania. More recently Denise Sigogneau-Russell discovered and described an important collection of isolated teeth representing diverse Early Cretaceous mammals from Morocco (Sigogneau-Russell, 1989a,b, 1991c, 1992, 1995a,b, 1999; Sigogneau-Russell and Ensom, 1998 and references therein). Almost simultaneously, Louis Jacobs, Michel Brunet, and collaborators recovered mammalian fossils from the Early Cretaceous of Cameroon (Jacobs et al., 1989). In the 1990s and 2000s, David Krause and colleagues described several isolated teeth from the Late Cretaceous of Madagascar (Krause et al., 1997; Krause, 2001). Wolf-Dieter Heinrich (1998,1999) recovered three specimens of mammals from the matrix associated with dinosaur bones (Upper Jurassic Middle Saurian Bed) obtained for the Museum für Naturkunde in Berlin by the 1909-1913 German Expeditions to Tendaguru, Tanzania. The newly found Tendaguru mammals belong to eutri-conodontans (Tendagurodon), peramurans (Tendaguru-therium), and haramiyidans (Staffia). Finally, Christopher E. Gow (1986a) published on new materials of morganucodontans from South Africa.
In the 1980s and 1990s, Mesozoic mammals were also found in the Late Triassic, Early Jurassic, and Late Cretaceous deposits of India, described by P.Yadagiri, D. P. Datta, G. V. R. Prasad, and their co-workers, adding significantly to the Mesozoic mammalian faunal lists of India (Yadagiri, 1985; Prasad and Sahni, 1988; Prasad et al., 1994; Datta and Das, 1996,2001; Prasad and Manhas, 1997,2002).
New discoveries over the past quarter century on the Laurasian continents have been numerous and, in some cases, spectacular. The vast territory of China has provided very well-preserved fossils. The classic vertebrate sites in the Early Jurassic Lower Lufeng Formation, Yunnan Province, have long been important for our understanding of basal mammals (Young, 1947; Patterson and Olson, 1961). The late British paleontologist Kenneth A. Kermack (1919-2000) and his co-workers published an exhaustive description of the Chinese Early Jurassic mammal Morganucodon (Kermack et al., 1973,1981). New field explorations in the 1980s and 1990s yielded important new specimens (Sun et al., 1985; Luo and Wu, 1994). Recent studies make it clear that Sinoconodon retains many "reptilian" characteristics in the dentition (Crompton and Sun, 1985; Crompton and Luo, 1993; Zhang et al., 1998). Even more astonishing was the discovery in the Early Jurassic Lufeng Formation of a tiny skull dubbed Hadro-
codium, described by Luo, Crompton, and Sun (2001). Hadrocodium has an enlarged brain cavity and no postdentary trough in the dentary, indicating separation of the middle ear bones from the mandible.
In 1982, the late Chinese paleontologist Minchen Chow (1918-1996), in cooperation with Thomas Rich from Australia, described an entirely new kind of mammal from the Middle Jurassic of China: the enigmatic Shuo-therium, which has "pseudotribosphenic" lower molars. Their largely hypothetic model of occlusion received welcome corroboration by subsequent discoveries of upper teeth belonging to Shuotherium (Sigogneau-Russell, 1998; Wang, Clemens et al., 1998). We now interpret shuothe-riids as the sister taxon of Australosphenida (Kielan-Jaworowska, Cifelli, and Luo, 2002; see chapter 6).
The Early Cretaceous Yixian Formation of Liaoning Province, China, has yielded a trove of new mammals. Most remarkable about the Yixian taxa is that most of the specimens are preserved as exceptionally complete skeletons, filling in many blank areas of the postcranial anatomy for some major clades of early mammals. Noteworthy discoveries include an almost complete skeleton of the "symmetro-dont" Zhangheotherium, described by Chinese paleontologist Yao-Ming Hu and colleagues (Hu et al., 1997); a complete eutriconodontan skeleton designated Jeholodens and a nearly complete skeleton of an early eutherian known as Eomaia, described by Qiang Ji and colleagues (1999,2002), and a new multituberculate named Sinobaatar (Hu and Wang, 2002a,b). The newly discovered gobiconodontid Re-penomamus is represented by many skulls and skeletons (Li et al., 2000; Wang, Hu, Meng et al., 2001). In the 1980s and 1990s, the international teams of Zhi-Ming Dong, Philip Currie, and Pascual Godefroit collected well-preserved mammals from the Late Cretaceous Bayan Mandahu sites in Inner Mongolia (Dong, 1993; Smith et al., 2001).
The Gobi Desert of Mongolia is renowned for its unusually well-preserved Cretaceous mammals. Collections of Mongolian Mesozoic mammals assembled by the Polish-Mongolian Expeditions between 1963 and 1971 were partly reported by Lillegraven et al. (1979). Zofia Kielan-Jaworowska subsequently studied these materials, sometimes in cooperation with Boris A. Trofimov, Demberlyin Dashzeveg, Percy M. Butler, Alfred W. Crompton, Cecile Poplin, Robert Presley, P. P. Gambaryan, and J0rn H. Hurum.
Russian-Mongolian Expeditions (1969-1996) also assembled an important collection of Cretaceous mammals, described mostly by B. A. Trofimov, often in cooperation with Z. Kielan-Jaworowska. A major contribution from the Russian-Mongolian Expeditions was a rich collection of Early Cretaceous (Aptian-Albian) mammals (see, e.g., Kielan-Jaworowska, Dashzeveg, and Trofimov, 1987). These expeditions also discovered the first marsupials from the
Late Cretaceous of Asia, described by B. A. Trofimov in cooperation with American paleontologist Frederick S. Szalay (Trofimov and Szalay, 1994; Szalay and Trofimov, 1996). The Mongolian paleontologist Demberlyin Dashzeveg, working by himself, assembled a notable collection of Early and Late Cretaceous mammals (e.g., Dashzeveg and Kielan-Jaworowska, 1984; Kielan-Jaworowska and Dashzeveg, 1989; Sigogneau-Russell et al., 1992).
In 1990, the American Museum of Natural History and the Institute of Geology of the Mongolian Academy of Sciences began to carry out expeditions in Mongolia, amassing spectacular collections (several hundred skulls and several dozen skeletons). Some of these have been prepared and studied by Demberlyin Dashzeveg, Inez Horovitz, Malcolm C. McKenna, Michael J. Novacek, Guillermo W. Rougier, John R. Wible, and others (see review by Kielan-Jaworowska et al., 2000). The preliminary studies published thus far include the discovery of the marsupial (epipubic) bone in Late Cretaceous eutherian mammals (Novacek et al., 1997), a new eutherian (Ukhaatherium), and new materials of Deltatheridium that support the hypothesis of deltatheroidan-marsupial affinities (Rougier et al., 1998). Several papers on the cranial anatomy of multituberculates and the ankle of eu-therian mammals have also been published (Horovitz, 2000; Wible and Rougier, 2000).
Exercising remarkable enterprise and drive, the late Russian paleontologist Lev A. Nessov (1947-1995) organized a series of expeditions to Uzbekistan, Kazakhstan, Tadzhikistan, Kirgizstan, and adjacent areas. Overcoming very harsh conditions, these expeditions amassed a significant collection of fossils, including a number of mammals from a hitherto almost unknown time interval, the early Late Cretaceous (see posthumously published review by Nessov, 1997; and Averianov, 2000). After Lev Nessov's death, Alexander Averianov continued exploration in these areas, often in collaboration with J. David Archibald and other colleagues from both Russia and elsewhere. The mammalian faunas are dominated by ungulate-like "zhelestids," and hence are quite different from the multituberculate-dominated assemblages of the Gobi Desert. Russian paleontologists have also discovered promising new Cretaceous mammal localities in Siberia (Maschenko and Lopatin, 1998; Averianov and Skutschas, 1999b, 2001; Averianov, 2000; Gambaryan and Averianov, 2001).
The last quarter-century saw both new discoveries and publication of important, previously collected early mammals from Western Europe. Significant among the latter is a huge, Late Jurassic assemblage from the famous Guimarota coal mines in Portugal (see Martin and Krebs, 2000), discovered in 1961 by the legendary German fossil hunter Walter G. Kühne (1911-1991). Gerhard Hahn (often accompanied by his wife Renate Hahn) completed a long series of monographs and descriptive papers on the multituberculates from Guimarota, as well as other Jurassic and Early Cretaceous localities on the Iberian Peninsula. The late Bernard Krebs (1934-2001) worked on dry-olestoids from Guimarota and elsewhere, a project that is now being continued by Thomas Martin and his students. The late Georg Krusat (1938-1998), accompanied by Jason A. Lillegraven, described the docodontans from Guimarota.
William A. Clemens (1980a) published a review of existing collections of Late Triassic mammals from the Keuper of Switzerland and Germany. Most significant among newly collected fossils from the Keuper are those from Saint-Nicolas-de-Port, France, published in a series of contributions by Denise Sigogneau-Russell (e.g., 1983b, 1989b). Late Cretaceous mammals (multituber-culates) were found in Romania (e.g., Radulescu and Samson, 1996; Csiki and Grigorescu, 2000), France, and Spain (Gheerbrant et al., 1997; Gheerbrant and Astibia, 1999).
In 1986, Paul Ensom began assembling an impressively large, diverse collection of mammals from the Purbeck Limestone Group (now regarded as of Early Cretaceous age) near Dorset, England. Ensom studied the multi-tuberculates with Z. Kielan-Jaworowska and worked on other mammals with D. Sigogneau-Russell. Among British paleontologists, Kenneth A. Kermack and coauthors continued their work on various Mesozoic mammals; the last paper by this team was published in 1998 (Kermack et al., 1998). An experienced researcher of early mammals, Percy M. Butler, continued to publish on the function of teeth in Mesozoic mammals and their relationships (e.g., Butler, 2000).
Among the most important and exotic discoveries of early mammals in recent years was that made by Farish A. Jenkins, Jr., and colleagues in the Late Triassic of northwestern Greenland. Among others (Jenkins et al., 1994), they recovered associated dentary, cranial, and postcranial remains of a haramiyidan, a group represented until then only by isolated teeth (Jenkins et al., 1997).
A virtual flood of new fossil mammals has poured in from the Mesozoic of North America, and space constraints permit us to mention only a small sample here (see chapter 2 for a fuller account). Some of these discoveries have helped to fill in knowledge for periods of geologic time that were hitherto not represented by the fossil record of this continent. Geologically oldest of these are Dinnetherium (a morganucodontan) and several other taxa recovered from the Early Jurassic of Arizona by Jenkins et al. (1983). Somewhat younger mammals, mainly
?eutriconodontans, were discovered in the ?Middle Jurassic of Mexico by James M. Clark (see summary by Mon-tellano et al., 1998). Significant inroads have been made into the record for the Early Cretaceous and early Late Cretaceous Aptian-Albian mammals, including a reasonably well-represented eutherian (Cifelli, 1999b),have been recovered from the Cloverly Formation by Richard L. Cifelli, W. Desmond Maxwell, and colleagues. Additional Early Cretaceous mammals have been collected in the Trinity Group, Texas and Oklahoma (Jacobs et al., 1989; Winkler et al., 1990; Cifelli, 1997a); and Thomas R. Lipka made the seemingly improbable discovery of relatively well-preserved Aptian mammals along the East Coast, in Maryland (Cifelli, Lipka, et al. 1999; Rose et al., 2001). Mammalian faunas, in some cases diverse and well-represented, have been recovered in southern and central Utah by Cifelli and Jeffrey G. Eaton in rocks spanning much of the first half of the Late Cretaceous (see Cifelli, Kirkland, et al., 1997; Eaton, Cifelli, et al., 1999). Finally, work in northwestern Colorado by J. David Archibald and his students has shed new light on the poorly understood "Edmontonian" (Diem, 1999).
New discoveries have also significantly expanded geographic sampling of early mammals from North America. Several sites have yielded Late Cretaceous mammals from the Eastern Seaboard; of these, the Ellisdale locality, New Jersey, under investigation by Barbara Grandstaff and her colleagues, has shown most promise to date (e.g., Grand-staff et al., 1992). In the South and Southwest, Timothy B. Rowe, Anne Weil, Cifelli, and collaborators have recovered a diverse assemblage of Judithian age from the Aguja Formation, Texas (see Rowe et al., 1992); and Cifelli, Jeffrey G. Eaton and others have amassed significant Judithian- and Aquilan-aged faunas from southwestern Utah (Eaton, Cifelli et al., 1999). Without doubt, the greatest geographic range extension for North American Cretaceous mammals was made by William A. Clemens and his students, who collected a small faunule from the North Slope of Alaska (Clemens, 1995).
Finally, the North American record of Mesozoic mammals has benefited from sampling (involving large numbers of specimens in some cases) new local faunas of similar age to those already known. Such additions to knowledge are important because they can be placed within the context of an enormous body of existing data. George Engelmann, Scott Madsen, and co-workers have collected significant new mammalian fossils from the Morrison Formation of Dinosaur National Monument, Utah (Engelmann and Callison, 1998). Important new local faunas from the Cretaceous include: Judith River and Hell Creek formations, Montana, collected by William A. Clemens and students (e.g., Archibald, 1982; Montellano,
1992); "Mesaverde," Lance, and Ferris formations, Wyoming, by Jason A. Lillegraven and students (e.g., Lil-legraven and McKenna, 1986; Eberle and Lillegraven, 1998a,b; Webb, 2001); and Frenchman and/or Ravenscrag formations, Saskatchewan, by Richard C. Fox, John E. Storer, and associates (e.g., Johnston and Fox, 1984; Fox, 1989; Storer, 1991).
Was this article helpful?