Historical Background

Not so long ago, Mesozoic mammals were assigned to a relatively small number of higher taxa, whose relationships seemed more or less understood. According to this

Table 4.1. Synoptic classification of Mesozoic mammals (excluding Metatheria and Eutheria)

Class MAMMALIA

fAdelobasileus, fHadrocodium1 fSinoconodontidae fKuehneotheriidae Order fMORGANUCODONTA fMorganucodontidae fMegazostrodontidae Order fDOCODONTA Order fSHUOTHERIDIA Order fEUTRICONODONTA fAmphilestidae fTriconodontidae fAustrotriconodontidae Order fGONDWANATHERIA Subclass AUSTRALOSPHENIDA

Order fAUSKTRIBOSPHENIDA Order MONOTREMATA Subclass fALLOTHERIA

fTheroteinidae fEleutherodontidae Order fHARAMIYIDA2

fHaramiyidae Order fMULTITUBERCULATA

Superfamily fPlagiaulacoidea Suborder fCIMOLODONTA Superfamily fPtilodontoidea Superfamily fTaeniolabidoidea Superfamily fDjadochtatherioidea Subclass TRECHNOTHERIA

Superorder fSYMMETRODONTA fAmphidontidae fTinodontidae fSpalacotheriidae Superorder fDRYOLESTOIDEA fVincelestidae3 Order fDRYOLESTIDA fDryolestidae fPaurodontidae Order fAMPHITHERIIDA fAmphitheriidae Superorder ZATHERIA Order fPERAMURA

fPeramuridae fArguitheriidae fArguimuridae Subclass BOREOSPHENIDA

Order fAEGIALODONTIA fAegialodontidae Infraclass METATHERIA Infraclass EUTHERIA

Notes: Modified after Kielan-Jaworowska et al. (2004). The dagger (f) denotes extinct taxa.

1 Hadrocodium appears to be closer to crown-group Mammalia than are morganu-codonts or sinoconodontids, but its precise position is uncertain.

2 Haramiyids were considered to be a possible sister group of tritylodontid cynodonts by Luo et al. (2002), as shown in Fig. 4.2, but are now generally considered to be closer to mammals. Theroteinidae and Eleutherodontidae were included in the paraphyletic Haramiyida by Butler (2000) and Butler and Hooker (2005).

3 Phylogenetic position uncertain, probably a dryolestoid or a zatherian.

view (e.g., Crompton and Jenkins, 1973, 1979), a group called morganucodonts lay at or near the base of a dichotomy between nontherian (or prototherian) and therian mammals (Fig. 4.1). Nontherians included the living monotremes, whereas therians comprised all other extant mammals. On the nontherian side, morganucodonts (then regarded as basal triconodonts) were believed to have given rise to other tri-conodonts, as well as to the docodonts, and questionably to the haramiyids, which were considered possible ancestors of the Multituberculata. Although their ancestry was unknown, monotremes were unambiguously grouped with nontherians. Therian mammals were seen as evolving from Kuehneo-therium, itself derived from a morganucodont or sharing a common ancestor with morganucodonts. From Kuehneo-therium, which was considered a basal symmetrodont, evolved the other symmetrodonts on the one hand, and eupanto-theres on the other. Eupantotheres were considered ancestral to the therians—the marsupials and placentals (Fig. 4.1B).

While parts of this appealing scenario remain essentially valid, recent discoveries and an explosion of interest in this early episode of mammalian history have led to a great expansion of known forms, and with it, the realization that the Mesozoic radiations of mammals were far more complex than previously imagined. Supposed differences in brain-case construction that were the basis of the dichotomy between nontherian and therian mammals are now known to be inaccurate, and this bipartite division of Mammalia has been largely abandoned (Kielan-Jaworowska, 1992). Consequently, there is considerable disagreement among experts concerning the sequence of divergence of the various early clades and even the definition of Mammalia itself. Especially volatile and controversial are the relationships of monotremes and multituberculates to each other and to other mammals, which vary depending on the anatomical system analyzed. Here it is important to realize that the position of monotremes directly affects the content of crown-group Mammalia (Rowe, 1988). Part of the controversy stems from the difficulty in determining which mammalian traits are synapomorphous and which ones may have evolved multiple times independently, and on this there is little agreement. This situation led Lillegraven and Krusat (1991: 43) to conclude that "parallel development of similar features was an all-pervasive phenomenon within early evolution of the Mammalia, making the unravelling of phylogenetic relationships among its basal groups a daunting, yet highly interesting, task." A current view of Mesozoic mammalian relationships is shown in Figure 4.2.

Among the most important characters that evolved early in mammalian evolution and contributed to their success are increasing brain capacity, the tribosphenic molar and associated changes in mastication, a single jaw joint between the dentary and the squamosal bones, a middle ear with three ossicles (see Fig. 3.1), and changes in limb posture related to increased activity. In the following sections, the evolution of these key mammalian features is emphasized.

At this point it should be noted that the term "therian" has been used both formally and informally with various

Polyphyletic View

Fig. 4.1. Temporal distribution and relationships of Mesozoic mammals. (A) View of a polyphyletic Mammalia, widely held in first half of the twentieth century and based mainly on Simpson (1928); (B) monophyletic Mammalia, with a dichotomy between nontherian and therian clades, based on Hopson and Crompton (1969); (C) current view of relationships, based on Luo, Crompton, and Sun (2001) and Luo et al. (2002). (From Cifelli, 2001.)

Fig. 4.1. Temporal distribution and relationships of Mesozoic mammals. (A) View of a polyphyletic Mammalia, widely held in first half of the twentieth century and based mainly on Simpson (1928); (B) monophyletic Mammalia, with a dichotomy between nontherian and therian clades, based on Hopson and Crompton (1969); (C) current view of relationships, based on Luo, Crompton, and Sun (2001) and Luo et al. (2002). (From Cifelli, 2001.)

connotations. Although the basic dichotomy between Pro-totheria (nontherians) and Theria has been largely abandoned, the name Theria is still generally used to refer to the crown-group of metatherians and eutherians and their close relatives (e.g., Rowe, 1988; Hopson, 1994; McKenna and Bell, 1997). Usage here follows this convention.

0 0

Post a comment