In Pursuit of Early Mammals

15. Aspects of Paleobiology of Mammaliaforms and Early Mammals

209 15 Aspects of Paleobiology of Mammaliaforms and Early Mammals The last two decades have brought an unusual blooming of papers that clearly show that Mesozoic mammaliaforms and mammals were much more diversified than had been previously thought. Luo (2007a, b), in his excellent reviews of Mesozoic mammal diversification, recognized five “experiments.” In the evolutionary path of early mammals , Luo demonstrated the convergence of new “experiments” with modern mammal morphotypes. Luo challenges the traditional assumption that Mesozoic mammals were small animals with generalized feeding and terrestrial habits and had few of the diverse ecomorphotypes of Cenozoic mammals. This hypothesis . . . is now contradicted by recent discoveries of a great range of ecological specializations, such as: a) the swimming and fish-feeding docodont Castorocauda (Ji et al. 2006), and semi-aquatic habits of Haldanodon (Martin 2005, 2006); b) ambulatory carnivory or scavenging (predation or feeding on other vertebrates in large gobiconodontids, see Hu et al. 2005b) and large individuals of Sinoconodon (Luo, Crompton, and Sun 2001b); c) scratch-digging and feeding on colonial insects in Fruitafossor (Luo and Wible 2005); d) scansorial (climbing) limbs characteristic of basal eutherians and metatherians, and their near relatives (Krebs 1991; Ji et al. 2002; Luo et al. 2003); e) volant (gliding) adaptation in Volaticotherium (Meng et al. 2006). (Abridged from Luo 2007a) To the five types recognized by Luo (2007a) one may add at least two: f) some multituberculates, resembling some modern rodents in the structure of dentary and arrangement of dentition and, apparently, in the mode of life (see chapter 10); and g) zalambdalestids, resembling present-day Macroscelididae that are fast-moving, often jumping animals, living in the steppes and savannahs of middle-north Africa. As demonstrated by Kielan-Jaworowska (1978), apparently zalambdalestids had a lifestyle similar to that of macroscelidids (see chapter 14). As noted in chapter 4, the animals that arose from the therapsids during the Late Triassic, previously assigned to Mammalia, are now more often assigned to the two high-rank taxa, non-mammalian Mammaliaformes In Pursuit of Early Mammals 210 and Mammalia. The main skeletal difference between early mammaliaforms and mammals relates to the structure of the lower jaw and its suspensorium . It is not clear if these differences in anatomy had an effect on their external appearance and mode of life. In the text below I examine some aspects of the supposed physiology of early mammaliaforms and early mammals that might shed light on this question. The well-known South African paleontologist Professor A. S. Brink (1924– 1991) spent part of his scientific life studying cynodonts and other therapsid reptiles from the Karoo Formation of southern Africa. He described several excellently preserved specimens of the common Early Triassic cynodont Thrinaxodon liorhinus including juveniles (Brink 1956, 1980). Brink argued that Thrinaxodon probably had a diaphragm, characteristic of mammals, related to the more efficient breathing needed for homeothermic animals. The existence of a diaphragm may be deciphered from the structure of the vertebral column. In the majority of the therapsid reptiles the lumbar ribs exist. However, in some genera of cynodonts, for example, Thrinaxodon, the lumbar ribs have been notably shortened (Jenkins 1971), which was regarded as evidence of the presence of a mammalian diaphragm. The lack of lumbar ribs in placental mammals has been preceded by the shortening of them in some cynodonts and mammaliaforms. The mobile and reduced lumbar ribs were found in the following mammaliaforms: in a “symmetrodontan” Akidolestes and in a docodontan Castorocauda. In addition to the taxa mentioned above, the reduced and mobile lumbar ribs have been also found in many primitive early mammals. The development of respiratory turbinals, such as those occurring in present-day mammals, also relates to respiration. The complete ossified respiratory turbinals have not been found in mammaliaforms. Hillenius (1994) and Hillenius and Ruben (2004) argued that bony ridges that occur within the nasal cavity of non-mammalian therapsids should be interpreted as attachment sites for turbinals. Kielan-Jaworowska and Trofimov (1980: 175) described the sinuses frontales in the Late Cretaceous eutherian genus Barunlestes and stated: “These sinuses evidently contained complicated turbinals,” but these structures have not been shown in the illustrations. Lillegraven and Krusat (1991: 67) mentioned the ridge that occurs on the ventral side of the nasal bone of a docodontan Haldanodon and stated: “The ridge probably represents the broken base of a nasoturbinal scroll.” Presumably the best evidence of the presence of ossified turbinals was found in multituberculates. Hurum (1994), when studying the serial sections of Late Cretaceous multituberculate skulls...