Mammalia constitute one of the five major classes of vertebrates (the others being Pisces, Amphibia, Reptilia, and Aves). There are, at latest count, some 4,170 species of mammals living today, classified in about 1010 genera, 130–140 families, and 18 orders (Corbet and Hill, 1980; Honacki et al., 1982). In addition, sixteen orders and uncounted genera and species of mammals are known to be extinct (see synoptic classification in Table 1). While most living mammals are terrestrial like their Mesozoic ancestors, one large progressive group has invaded the air (bats, represented by about 150 living genera and 920 species). Two other smaller groups are fully aquatic (whales, including some 40 genera and 80 species; and sirenians, 3 genera and 5 species). Mammals occupy all of the continents, and all of the larger oceanic islands. Faunal diversity is greatest in lowland tropical forests, woodlands, and savannas, decreasing rapidly at high elevations and latitudes.

Commonly chapters by paleontologists are included in books about a particular group otherwise written by neontologists; it is much rarer that the situation is reversed, and I am conscious of the heavy responsibility imposed by my charge to suggest why paleontologists should pay some attention to living mammals. Most of us accept that the past is the key to the present, but here I must justify inverting the aphorism. A few years ago such an attempt, particularly as it related to the phylogeny of mammals, would have seemed arrogant; and while a number of molecular evolutionists might privately have held the view expressed in the epigraph above, I think few paleontologists would have agreed.

The origin of mammals, often treated as a discrete but obscure event that took place sometime between the Late Triassic and Early Jurassic, can also be viewed as the product of many transitions - from the early synapsid radiations during the Pennsylvanian and Early Permian through the emergence of placentals and marsupials in the Cretaceous (see Hopson, 1969). A fundamental dichotomy appeared early in the evolution of amniotes; sauropsids (represented today by living reptiles and birds) constituted one lineage, synapsids (which includes the mammal-like reptiles- pelycosaurs and therapsids- and their mammalian descendants) the other. Thus, mammalian ancestry may be traced to pelycosaurs that first appear in the fossil record as part of the earliest known reptilian fauna (Carroll, 1982). And mammalian bony and dental structures continued to undergo substantial modification throughout the Mesozoic, long after the appearance of forms technically classified as Mammalia. Given these phylogenetic changes, the suggestion that there occurred a point when mammals “originated” seems simplistic. Yet the major evolutionary stages may still be evaluated with the purpose of identifying the inception of features and functions basic to the radiation of modern mammals. Accordingly, this survey summarizes our current understanding of the evolution of mammals with full acknowledgment that an account of mammalian origins has neither a definitive beginning nor a climactic end.

There are three living infraclasses of Mammalia: Prototheria (Monotremes), Metatheria (Marsupials), and Eutheria (Placentals). The egg-laying, poison-spurred and in other ways morphologically reptile-like monotremes now inhabit Australia and New Guinea. Whereas discussion of their past history (Miocene and Pleistocene of Australia; Woodburne and Tedford, 1975) would be a fascinating topic, that group will not be discussed further here.

Living mammals are divided into three groups: (1) monotremes (“prototherians”, three genera of egg-layers of the Australian realm); (2) marsupials (“metatherians”, the diverse “pouched mammals” of the Australian realm and the Americas); and (3) eutherians (“placentals”, the rest of us, with the greatest taxonomic and adaptive complexity, and having nearly cosmopolitan distribution). The marsupials plus eutherians are referred to as “therians,” a subclass that has a rich history of diversity of extinct varieties in addition to the living orders (Cassiliano and Clemens, 1979; Kraus, 1979; Kielan-Jaworowska, Eaton and Bown, 1979). The phylogenetic relationship between monotremes and modern therians is contested. Traditional wisdom had it that monotremes have been distinct from therians since virtually the beginnings of mammalian history (e.g., Marshall, 1979) in the Late Triassic or Early Jurassic. Kemp (1983), however, favored a closer relationship between monotremes and therians, with their presumed divergence at some later, but paleontologically undefined, time.

The Paleocene (approximately 65 to 55 million years ago) is the first epoch of the Cenozoic Era; it is also the beginning of an era in which mammals underwent a major adaptive radiation and became dominant land animals. It was a fascinating and exciting era of transition in mammalian evolution between the dark ages of the Cretaceous, when mammals were few and small and lived in the shadow of the great, “ruling reptiles”, and the explosive ages of the Eocene, when mammals were diverse and in full blossom. In the Late Cretaceous there were approximately eight families of marsupials and placentals; by the early Eocene over 70 had come into being. Were mammals from the Paleocene totally unknown, creationists would surely have plenty of grist for their missing-link mills. But Paleocene mammals are now known from five continents and they give evidence of archaic groups of mammals that were unleashed from the constraining suppression of dinosaurs. Paleocene mammals evolved rapidly and included “evolutionary experiments” that failed soon thereafter but also groups that were at the base of enormously successful adaptive radiations later in the Cenozoic. In fact, more than one-third of the known orders of the Cenozoic had their beginnings, or origins, in the Paleocene. Origins are captivating. At one time or another every one of us has pondered the origin of the universe, the origin of life, the origin of humankind, etc. Without knowledge of Paleocene mammals, many basic questions concerning the origin and evolution of mammalian adaptations would not be answerable.

The Eocene was an especially interesting and important time in the history of mammals. Whereas the Paleocene witnessed the rise of mammals from their previously subordinate position among terrestrial vertebrates, it was during the Eocene that the second major radiation of eutherian mammals took place. Many of the modern orders diversified and achieved dominance on land, and mammals even began to invade the seas and take to the air. By the late Eocene, this explosive radiation had produced the greatest ordinal diversity of mammals known at any time in their history (Lillegraven, 1972). If we could travel back to the Paleocene, we would see few familiar mammals; most belonged to archaic groups that long ago became extinct. But a visitor to the Eocene (given wide latitude in space and time, of course) might recognize many of the denizens of Eocene forests and floodplains—bats, prosimian primates, squirrel-like rodents, rabbits, moles, armadillos, and miniature hoofed animals resembling mouse deer. Along the shore, our hypothetical Eocene naturalist might spot a whale or a primitive sea cow. A closer look would reveal that these animals differed in many ways from living forms, of course, but the resemblances are real and, in some cases, striking. In fact, at least 10 orders of mammals alive today appeared in the fossil record for the first time, or experienced major radiations of modern groups, in the Eocene. Not all Eocene mammals belonged to modern orders, however. There was still a considerable proportion of archaic mammals, many representing lines that had begun in the Paleocene or, in some instances, before. And some of them were highly successful too.

The second half of the Age of Mammals yields many stories worth telling. The one that I find grandest and most central, however is the “Savanna Story.” While featuring it as the principle theme, I hope to weave in the following secondary subjects:

1) the adaptive radiation of four representative groups, namely hypsodont horses, camels, elephants and cricetid rodents;

2) intercontinental dispersal patterns across Beringia and between the American continents;

3) and faunal turnover patterns.

The order Cetacea was the earliest, and is the most diverse mammalian group to adapt to a marine existence. The first fossil whales appear in the fossil record in the Middle Eocene. Interestingly, cetaceans developed in some ways that are parallel to earlier marine reptiles and appear to have filled some of their niches. Some cetaceans later re-adapted to fresh water.

The defining characteristic of the mammalian order Rodentia is a pair of continually growing incisors (Figure 1). There are rodents that scamper, hop, burrow, swim, and glide; that eat seeds, grass, nuts, insects, and worms; that live in the tropics, arctic, and temperate regions, in, on, or above the surface of the earth; in desert, grassland, savannah, and forest, but all are easily recognized by the incisors and correlated features. The lower incisors grow as helices approximating arcs. The upper incisor represents a relatively larger arc of a smaller circle than the lower incisor, which is a smaller arc of a larger circle. The root of each incisor is far back in the jaw. Enamel is essentially restricted to the labial surface of the incisors and comprises two layers. The wear of lower incisors against uppers, the distribution of enamel, the relative hardness of enamel versus dentine, and continual growth results in the maintenance of a sharp chisel edge. Thus, the gnawing incisors are primary food-gathering organs for rodents, but can also function in a number of other ways, such as in digging or as forceps.

The mammalian order Primates includes living tarsiers, lemurs, monkeys, apes and humans. The name Primates itself, derived from Latin primus, first or foremost, betrays a characteristically immodest self-appraisal of our place in nature! When proposed by Linnaeus, in the definitive 1758 edition of his Systema Naturae, nature's center stage was indeed defined as that on which we play. Now, knowing that we are but one of 200 or so living primate species, knowing that primates are but one of 18 or so living mammalian orders, nature's stage appears much grander – claiming the center is clearly more difficult. Central or not, primates we are – human interest in human origins being what it is, primates will always hold a special fascination.

The evolutionary history of humans is well understood in outline, compared to that of many other groups of mammals. But human evolution remains enigmatic in its details, and these are compelling both scientifically and personally because they relate to the biological uniqueness of humans. Humans are placed in the primate family Hominidae, which, in traditional classifications, contains a single living species, Homo sapiens. The closest living relatives of humans are great apes: the chimpanzees Pan paniscus and Pan troglodytes, the gorilla Gorilla gorilla, and the orangutan Pongo pygmaeus. These apes have traditionally been placed in the family Pongidae as the sister group of Hominidae. Living Hominidae and Pongidae, together with Hylobatidae (gibbons) comprise the modern representatives of the primate suborder Hominoidea.