The septomaxilla of fossil and recent synapsids and the problem of the septomaxilla of monotremes and armadillos

The septomaxilla is a paired intramembranous ossification in the external nares that occurs in Lepidosauria among Recent Sauropsida and is purported to be present in Monotremata and Dasypodidae (armadillos) among Recent Mammalia. A review of neontological and palaeontological evidence regarding this element in mammals supports the following conclusions: (1) monotremes have a true septomaxilla resembling that known for non-mammalian therapsids and some Mesozoic mammals; (2) the element in dasypodids is a neomorph; it neither resembles the septomaxilla of other synapsids nor does it exhibit the same relationship to the developing nasal-floor cartilage as the septomaxilla of lepidosaurs and monotremes; (3) a septomaxilla is lacking in all Recent therians, and there is no evidence that this bone is fused to the premaxilla in Recent therians, as has been suggested by previous authors.     

New ontogenetic evidence on the septomaxilla ofTamandua andCholoepus (Mammalia,Xenarthra), with a reevaluation of the homology of the mammalian septomaxilla

An intramembranous ossification at the anterior end of the cartilaginous nasal capsule is described for the first time in prenatal specimens of the anteaterTamandua and the slothCholoepus and redescribed in prenatal specimens of the armadillosDasypus andZaedyus. From comparisons of this bone with the septomaxilla of monotremes and various Mesozoic mammals, it is concluded that (1) the bone inTamandua andCholoepus is homologous with the central part (processus ascendens) of the bone inDasypus, Zaedyus, and other armadillos and (2) the xenarthran processus ascendens, in turn, is homologous with the central part of the septomaxilla of monotremes and various Mesozoic mammals. Therefore, the bone in question in xenarthrans is a true septomaxilla. It is further concluded that the armadillo septomaxilla has two neomorphic components: a lamina palatina beneath the cartilaginous nasal floor and a processus intrafenestralis extending rostrally into the nasal fossa.     

An X-radiographic and SEM study of the osseous inner ear of multituberculates and monotremes (Mammalia): implications for mammalian phylogeny and evolution of hearing

Multituberculate petrosals with well-preserved, three-dimensional internal anatomy from the Late Cretaceous/early Paleocene Bug Creek Anthills, Montana, U.S.A., are described from X-radiographic and SEM images, as well as from conventional visual observations, and are compared with the anatomy of the osseous inner ear in monotremes and in primitive non-therian and therian mammals. Results of this study indicate that: (1) the cochlea of at least some multituberculates retained a lagena, previously known only in monotremes among mammals; (2) an enlarged vestibule evolved in several lineages of multituberculates independently, and hence is not a synapomorphy of the order; (3) the cochlear canal lacks osseous laminae in support of the short, wide basilar membrane, which was probably inefficient in responding to high-frequency airborne vibrations; and (4) consequently, bone-conducted hearing in some multituberculate species may have been important in interpretation of their surroundings. Comparisons with the inner ear of monotremes and primitive therians indicate that curvature of the cochlea and cribriform plates for passage of vestibulocochlear nerve branches through the petrosal are unlikely homologues between monotremes and therians. From non-therian to therian mammals, there is a distinct morphological gap in the inner ear transition, characterized by acquisition of a number of neomorphs in the therian inner ear; an intermediate stage has yet to be discovered.     

The relationships of mammals

A cladistic analysis generates alternative hypotheses regarding both the origin and the interrelationships of mammals to those most widely accepted at the present time. It is proposed that the tritylodontids are more closely related to mammals than is Probainognathus ; that the non-therian mammals do not constitute a monophyletic group; and that the monotremes are related to the modern therians, the ear ossicles among other characters having evolved only once. The multituberculates may be related to the monotremes.
It is argued that the current views are variously based on an overemphasis of superficial dental similarities, misinterpretation of the structure of the mammalianbraincaseand too readyacceptance of parallel evolution amongstthe groups concerned. The hypotheses proposed here are apparently much more parsimonious.     

Terrestrial locomotion in monotremes (Mammalia: Monotremata)

Cineradiographic analysis of the limb movements of Ornithorhynchus reveals that the proximal limb bones undergo horizontal retraction, long-axis rotation and distal elevation (humerus) or depression (femur) during the propulsive phase of walking. Cinematographic records show that the locomotor movements of Ornithorhynchus differ in several respects from those of Tachyglossus and Zaglossus which are essentially similar. Comparison of the propulsive phase limb movements of Ornithorhynchus with those previously established radiographically for Tachyglossus reveals that the humerus is on average less laterally but more dorsally directed, and that the femur is on average more dorsally and slightly more laterally directed in Ornithorhynchus . Comparison of the limb orientations of monotremes with those empirically determined in therian mammals and lepidosaurian reptiles indicates that monotremes are most similar to therians. Consideration of several evidently derived features of the appendicular skeleton of monotremes leads to the conclusion that the limb orientations used by early mammals were probably similar to those used by locomotively generalized living therians, and that monotremes show modifications of this.     

Updating and Recoding Enamel Microstructure in Mesozoic Mammals: In Search of Discrete Characters for Phylogenetic Reconstruction

The previously unknown enamel microstructure of a variety of Mesozoic and Paleogene mammals ranging from monotremes and docodonts to therians is described and characterized here. The novel information is used to explore the structural diversity of enamel in early mammals and to explore the impact of the new information for systematics. It is presently unclear whether enamel prisms arose several times during mammalian evolution or arose only once with several reversals to prismless structure. At least two undisputed reversions or simplifications are known—in the monotreme clade from Obdurodon to Ornithorhynchus (via Monotrematum?), and (perhaps more than once) within the clade from archaeocete to a variety of odontocete whales. Similarly, both prismatic and nonprismatic enamel is present among docodonts. Seven discrete characters showing enough morphological diversity to be of potential importance in phylogenetic reconstructions may be identified as a more appropriate summary of enamel microstructural diversity among mammaliaforms than the single character “prismatic enamel-present/absent” employed in recent matrices. Inclusion of five of these characters in the matrix of Luo et al. (2002) modifies the original topology by collapsing several nodes involving triconodonts and other nontribosphenic taxa. There is considerable support for prismatic enamel as a synapomorphy of trithelodonts plus Mammaliamorpha, and multituberculates appear to have small or “normal” sized prisms as the ancestral condition, with some (as yet) enigmatic changes to nonprismatic structure in some basal members of the group and the appearance of “gigantoprismatic” structure as an autapomorphic state of less inclusive clades. Other potential qualitative characters and the need for attaining appropriate methods to incorporate quantitative features may be important for future analyses.     

Review of the monotreme fossil record and comparison of palaeontological and molecular data

Monotremes have traditionally been considered a remnant group of mammals descended from archaic Mesozoic stock, surviving to the present day on the relatively isolated Australian continent. Challenges to this orthodoxy have been spurred by discoveries of 'advanced' Cretaceous monotremes (Steropodon galmani, Archer, M., et al., 1985. First Mesozoic mammal from Australia-an Early Cretaceous monotreme, Nature. 318, 363-366) as well as by results from molecular data linking monotremes to therian mammals (specifically to marsupials in some studies). This paper reviews the monotreme fossil record and briefly discusses significant new information from additional Cretaceous Australian material. Mesozoic monotremes (including S. galmani) were a diverse group as evidenced by new material from the Early Cretaceous of New South Wales and Victoria currently under study. Although most of these new finds are edentulous jaws (limiting dental comparisons and determination of dietary niches), a range of sizes and forms has been determined. Some of these Cretaceous jaws exhibit archaic features-in particular evidence for the presence of a splenial bone in S. galmani-not seen in therian mammals or in post-Mesozoic (Tertiary and Quaternary) monotreme taxa. Tertiary monotremes were either archaic ornithorhynchids (toothed platypuses in the genera Monotrematum and Obdurodon) or tachyglossids (large echidnas in the genera Megalibgwilia and Zaglossus). Quaternary ornithorhynchid material is referable to the sole living platypus species Ornithorhynchus anatinus. Quaternary echidnas, however, were moderately diverse and several forms are known (Megalibgwilia species; 'Zaglossus' hacketti; Zaglossus species and Tachyglossus aculeatus).     

Phylogenetic analysis of 1.5 Mbp and platypus EST data refute the Marsupionta hypothesis and unequivocally support Monotremata as sister group to Marsupialia/Placentalia

The extant mammalian groups Monotremata, Marsupialia and Placentalia are, according to the 'Theria' hypothesis, traditionally classified into two subclasses. The subclass Prototheria includes the monotremes and subclass Theria marsupials and placental mammals. Based on some morphological and molecular data, an alternative proposition, the Marsupionta hypothesis, favours a sister group relationship between monotremes and marsupials to the exclusion of placental mammals. Phylogenetic analyses of single genes and even multiple gene alignments have not yet been able to conclusively resolve this basal mammalian divergence. We have examined this problem using one data set composed of expressed sequence tags (EST) and another containing 1 510 509 nucleotide (nt) sites from 1358 inferred cDNA genomic sequences. All analyses of the concatenated sequences unambiguously supported the Theria hypothesis. The Marsupionta hypothesis was rejected with high statistical confidence from both data sets. In spite of the strong support for Theria, a non-negligible number of single genes supported either of the two alternative hypotheses. The divergence between monotremes and therian mammals was estimated to have taken place 168–178 Mya, a dating compatible with the fossil record. Considering the long common evolutionary branch of therians, it is surprising that sequence data from many thousand amino acid sites were needed to conclusively resolve their relationship to monotremes. This finding draws attention to other mammalian divergences that have been taken as unequivocally settled based on much smaller alignments. EST data provide a comprehensive random sample of protein coding sequences and an economic way to produce large amounts of data for phylogenetic analysis of species for which genomic sequences are not yet available.     

Monotreme ossification sequences and the riddle of mammalian skeletal development

The developmental differences between marsupials, placentals, and monotremes are thought to be reflected in differing patterns of postcranial development and diversity. However, developmental polarities remain obscured by the rarity of monotreme data. Here, I present the first postcranial ossification sequences of the monotreme echidna and platypus, and compare these with published data from other mammals and amniotes. Strikingly, monotreme stylopodia (humerus, femur) ossify after the more distal zeugopodia (radius/ulna, tibia/fibula), resembling only the European mole among all amniotes assessed. European moles also share extreme humeral adaptations to rotation digging and/or swimming with monotremes, suggesting a causal relationship between adaptation and ossification heterochrony. Late femoral ossification with respect to tibia/fibula in monotremes and moles points toward developmental integration of the serially homologous fore- and hindlimb bones. Monotreme cervical ribs and coracoids ossify later than in most amniotes but are similarly timed as homologous ossifications in therians, where they are lost as independent bones. This loss may have been facilitated by a developmental delay of coracoids and cervical ribs at the base of mammals. The monotreme sequence, although highly derived, resembles placentals more than marsupials. Thus, marsupial postcranial development, and potentially related diversity constraints, may not represent the ancestral mammalian condition.     

The major histocompatibility complex in monotremes: an analysis of the evolution of <Emphasis Type="Italic">Mhc</Emphasis> class I genes across all three mammalian subclasses

We report the isolation and characterization of cDNA clones of expressed, functional major histocompatibility complex class-I ( Mhc-I) genes from two species of monotremes: the duck-billed platypus and the short-beaked echidna. The cDNA clones were isolated from libraries constructed from spleen RNA, clearly establishing their expression in at least this one peripheral lymphoid organ. From the presence of conserved amino acid residues, it appears the expressed sequences encode molecules that likely function as classical Mhc-I. These clones were isolated using monotreme Mhc-I processed pseudogenes as probes. These processed pseudogenes were isolated from genomic DNA and, based on their structure, are likely independently derived in the platypus and echidna. When all the monotreme sequences were included in phylogenetic analyses, we found no apparent orthologous relationships between the platypus and echidna Mhc-I. Analyses that included a large number of Mhc-I sequences from other taxa support a separate monotreme Mhc-I clade, basal to a therian Mhc-I clade that is comprised of sequences from marsupial and placental mammals. The phylogenies also support the hypothesis that Mhc-I genes of placental mammals, marsupials, and monotremes are derived from three separate lineages of Mhc-I genes, best explained by two rounds of duplications and deletions. The first round would have occurred prior to the divergence of monotremes and therians, and the second prior to the divergence of marsupials and placental mammals. The sequences described here represent the first reported functional monotreme Mhc-I, as well as the first processed pseudogenes of any type from monotremes.     

Reexamination of the morphological evidence for the cohort Epitheria (Mammalia,Eutheria)

Novacek and co-workers recognized a monophyletic clade Epitheria, comprising all eutherians except edentates and the extinct palaeoryctoids, on the basis of two synapomorphies: a stirrupshaped stapes and a foramen ovale enclosed within the alisphenoid. To evaluate this phylogenetic hypothesis, we reexamined the distributions of stapedial morphologies and positions of the foramen ovale across Recent and extinct mammals and nonmammalian cynodonts. The states and distributions of the stapes and forament ovale characters used by Novacek and coworkers were modified by recognizing two stapedial characters (one relating to shape of the crura, the other to the nature of the foramen) and a single, multistate foramen ovale character (within, behind, and lateral to the alisphenoid). The taxon-character matrix used by Novacek (1989, 1992b), substituting our amended stapedial and foramen ovale characters and adding several previously unscored extinct taxa and three new characters, was subjected to a series of PAUP manipulations. Identified among the most parsimonious trees were three major topologies for the base of Eutheria: (1) a polytomy including an Edentata/Ungulata clade, (2) a polytomy with Edentata and Ungulata as separate clades, and (3) Edentata and (when included) Palaeoryctoidea as the successive outgroups to a monophyletic Epitheria. We conclude that topology 2 best reflects the current state of knowledge. An edentate/ungulate clade is supported by three characters (from the mastoid region and subarcuate fossa); however, other morphological studies require modification of the distributions of these characters in xenarthrans and bassal ungulates, thereby eliminating support for this clade. In nearly all manipulations, obtaining a monophyletic Epitheria required that one or two steps be added to the most parsimonious trees. When a monophyletic Epitheria was obtained, it was supported by a triangular stapes and, in some trees, the reappearance of a stapedial artery (lost earlier at the level of Recent therians) and a transpromontorial internal carotid artery. In the most parsimonious trees, a foramen ovale within the alisphenoid was an equivocal synapomorphy of Recent therians or cutherians, and a stapes with strongly convex crura (our state closest to the stirrup-shaped state of Novacek and co-workers) appeared independently within various eutherian lineages. The reduction or loss of the stapedial foramen was identified as an independent event in monotremes and within marsupials and various eutherian lineages.To whom correspondence should be addressed.     

Septomaxilla of nonmammalian synapsids: soft-tissue correlates and a new functional interpretation. DELETE

The function of the septomaxilla of nonmammalian synapsids has long been problematic. Distinctive features of this bone, including a prominent intranarial process and a septomaxillary canal and foramen, are characteristic of pelycosaurs and nonmammalian therapsids, but are lost in their mammalian descendants. Numerous contradictory reconstructions have been proposed for the soft anatomy associated with the septomaxilla of nonmammalian synapsids. This review supports the following conclusions: 1) No particular correlation exists between the septomaxilla and the vomeronasal organ (VNO), and the most likely location for the VNO is on the dorsal surface of the palatal process of the vomer; 2) The most likely occupant of the septomaxillary canal is the nasolacrimal duct, which opened either anterior or medial to the intranarial process, near the opening of the VNO duct; and 3) The occupant of the septomaxillary foramen remains uncertain. These conclusions suggest that the functional significance of the septomaxilla in the nonmammalian synapsids is tied to that of the nasolacrimal duct. The association of this duct and the VNO in these animals resembles the condition in Recent amphibians and lepidosaurs, in which the nasolacrimal duct supplies orbital fluids to the VNO, apparently to enhance vomeronasal function. The peculiar shape of the synapsid septomaxilla may have served to collect vomeronasal odor molecules. The changes of the septomaxilla in early mammals, and its nearly complete loss in extant mammals, are probably correlated with a dissociation of the nasolacrimal duct and VNO, and functional changes in both structures.     

Miniaturisation and the origin of mammals

In this paper, the hypothesis of miniaturisation to explain the origin of mammals (Rowe 1993, Mammals phylogeny: mesozoic differentiation, multituberculates, monotremes, early therians, and marsupials. New York: Springer-Verlag, p. 129–145) is discussed, based on three lines of evidence resulting from new discoveries of eucynodonts in the Late Triassic of Southern Brazil (Bonaparte et al. 2003, Rev Bras Paleont 5:5–27; 2005, Rev Bras Paleont 8:25–46; 2006, New Mexico Museum Nat Hist Sci Bull 37:1–8; 2010, Rev Bras Paleont) that are: (1) the incomplete fossil record of eucynodonts known until 2003; (2) the structure of the primary palate rejects the ancestral condition of thrinaxodontids, probainognathids, chiniquodontids and cynognathids to the earliest mammals; and (3) the relatively large postdentary bones of the Middle Triassic brasilodontids that are otherwise very small in size (skull 44 mm long) suggest that small size per se did not help to improve the middle ear or other sophisticated organs present in the earliest mammals (Rowe 1993; Kemp 2005, The origin and evolution of mammals. Oxford University Press, p. 1–391). Small size possibly was not a secondary character, but a persistent primitive one. This new interpretation has resulted from comparative study of non-mammalian eucynodonts discovered in the Middle and Late Triassic of Brazil and those known previously. The general acceptance of the hypothesis of miniaturisation is thus a consequence of the poor fossil record of Middle and Late Triassic eucynodonts before 2003.     

Non-coding RNAs and the acquisition of genomic imprinting in mammals

Genomic imprinting, representing parent-specific expression of alleles at a locus, is mainly evident in flowering plants and placental mammals. Most imprinted genes, including numerous non-coding RNAs, are located in clusters regulated by imprinting control regions (ICRs). The acquisition and evolution of genomic imprinting is among the most fundamental genetic questions. Discoveries about the transition of mammalian imprinted gene domains from their non-imprinted ancestors, especially recent studies undertaken on the most ancient mammalian clades — the marsupials and monotremes from which model species genomes have recently been sequenced, are of high value. By reviewing and analyzing these studies, a close connection between non-coding RNAs and the acquisition of genomic imprinting in mammals is demonstrated. The evidence comes from two observations accompanied with the acquisition of the imprinting: (i) many novel non-coding RNA genes emerged in imprinted regions; (ii) the expressions of some conserved non-coding RNAs have changed dramatically. Furthermore, a systematical analysis of imprinted snoRNA (small nucleolar RNA) genes from 15 vertebrates suggests that the origination of imprinted snoRNAs occurred after the divergence between eutherians and marsupials, followed by a rapid expansion leading to the fixation of major gene families in the eutherian ancestor prior to the radiation of modern placental mammals. Involved in the regulation of imprinted silencing and mediating the chromatins epigenetic modification may be the major roles that non-coding RNAs play during the acquisition of genomic imprinting in mammals. Supported by National Natural Science Foundation of China (Grant No. 30830066), the Ministry of Education of China and Natural Science Foundation of Guangdong Province (Grant No. IRT0447, NSF-05200303) and National Key Basic Research and Development Program of China (Grant No. 2005CB724600)     

Resolving the evolution of the mammalian middle ear using Bayesian inference

Background

The minute, finely-tuned ear ossicles of mammals arose through a spectacular evolutionary transformation from their origins as a load-bearing jaw joint. This involved detachment from the postdentary trough of the mandible, and final separation from the dentary through resorption of Meckel’s cartilage. Recent parsimony analyses of modern and fossil mammals imply up to seven independent postdentary trough losses or even reversals, which is unexpected given the complexity of these transformations. Here we employ the first model-based, probabilistic analysis of the evolution of the definitive mammalian middle ear, supported by virtual 3D erosion simulations to assess for potential fossil preservation artifacts.

Results

Our results support a simple, biologically plausible scenario without reversals. The middle ear bones detach from the postdentary trough only twice among mammals, once each in the ancestors of therians and monotremes. Disappearance of Meckel’s cartilage occurred independently in numerous lineages from the Late Jurassic to the Late Cretaceous. This final separation is recapitulated during early development of extant mammals, while the earlier-occurring disappearance of a postdentary trough is not.

Conclusions

Our results therefore suggest a developmentally congruent and directional two-step scenario, in which the parallel uncoupling of the auditory and feeding systems in northern and southern hemisphere mammals underpinned further specialization in both lineages. Until ~168 Ma, all known mammals retained attached middle ear bones, yet all groups that diversified from ~163 Ma onwards had lost the postdentary trough, emphasizing the adaptive significance of this transformation.

     

Molecular Phylogeny and Evolution of the Neurotrophins from Monotremes and Marsupials

We have investigated the phylogenetic relationships of monotremes and marsupials using nucleotide sequence data from the neurotrophins; nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), and neurotrophin-3 (NT-3). The study included species representing monotremes, Australasian marsupials and placentals, as well as species representing birds, reptiles, and fish. PCR was used to amplify fragments encoding parts of the neurotrophin genes from echidna, platypus, and eight marsupials from four different orders. Phylogenetic trees were generated using parsimony analysis, and support for the different tree structures was evaluated by bootstrapping. The analysis was performed with NGF, BDNF, or NT-3 sequence data used individually as well as with the three neurotrophins in a combined matrix, thereby simultaneously considering phylogenetic information from three separate genes. The results showed that the monotreme neurotrophin sequences associate to either therian or bird neurotrophin sequences and suggests that the monotremes are not necessarily related closer to therians than to birds. Furthermore, the results confirmed the present classification of four Australasian marsupial orders based on morphological characters, and suggested a phylogenetic relationship where Dasyuromorphia is related closest to Peramelemorphia followed by Notoryctemorphia and Diprotodontia. These studies show that sequence data from neurotrophins are well suited for phylogenetic analysis of mammals and that neurotrophins can resolve basal relationships in the evolutionary tree. Received: 27 January 1997 / Accepted: 20 March 1997     

Ear ossicle morphology of the Jurassic euharamiyidan Arboroharamiya and evolution of mammalian middle ear

The middle ear bones of Mesozoic mammals are rarely preserved as fossils and the morphology of these ossicles in the earliest mammals remains poorly known. Here, we report the stapes and incus of the euharamiyidan Arboroharamiya from the lower Upper Jurassic (～160 Ma) of northern China, which represent the earliest known mammalian middle ear ossicles. Both bones are miniscule in relation to those in non‐mammalian cynodonts. The skull length/stapedial footplate diameter ratio is estimated as 51.74 and the stapes length as the percentage of the skull length is 4%; both numbers fall into the stapes size ranges of mammals. The stapes is “rod‐like” and has a large stapedial foramen. It is unique among mammaliaforms in having a distinct posterior process that is interpreted as for insertion of the stapedius muscle and homologized to the ossified proximal (stapedial) end of the interhyal, on which the stapedius muscle attached. The incus differs from the quadrate of non‐mammalian cynodonts such as morganucodontids in having small size and a slim short process. Along with lack of the postdentary trough and Meckelian groove on the medial surface of the dentary, the ossicles suggest development of the definitive mammalian middle ear (DMME) in Arboroharamiya. Among various higher‐level phylogenetic hypotheses of mammals, the one we preferred places “haramiyidans” within Mammalia. Given this phylogeny, development of the DMME took place once in the allotherian clade containing euharamiyidans and multituberculates, probably independent to those of monotremes and therians. Thus, the DMME has evolved at least three times independently in mammals. Alternative hypothesis that placed “haramiyidans” outside of Mammalia would require independent acquisition of the DMME in multituberculates and euharamiyidans as well as parallel evolution of numerous derived similarities in the dentition, occlusion pattern, mandibles, cranium, and postcranium between the two groups and between “haramiyidans” and other mammals. J. Morphol. 279:441–457, 2018. © 2016 Wiley Periodicals, Inc.     

Cranial structure and relationships of the Liassic mammal Sinoconodon*

The skull of the 'Rhaeto-Liassic' mammal Sinoconodon changchiawaensis (Young) from the lower Lufeng Series of China is described. It is characterized by a relatively larger and more robust dentary condyle and a greater reduction of the post-dentary bones than are present in the Morganucodontidae, Kuehneotheriidae, and Dinnetherium. In other aspects Sinoconodon is more primitive; precise post-canine occlusion is lacking, the mandibular symphysis is deep, the jaw articulation lies below a line projected through the apices of the teeth, the pterygoparoccipital foramen is large and the post-canine teeth cannot be divided into molars and premolars. The jaw articulation and braincase of Sinoconodon are compared with those of the two cynodont therapsids Probainognathus and Thrinaxodon. It is concluded that in the transition from therapsid to mammal the medial surface of the groove in the squamosal housing the quadrate was lost and, as a result, in Sinoconodon, Morganucodon and Dinnetherium the hollow medial surface of the quadrate abutted directly against the paroccipital process. A definition for the Class Mammalia is given. It is suggested that the three-boned middle ear was present in Cretaceous triconodonts, and that it probably arose independently in the lines leading to multituberculates and Cretaceous therians. The structure of recently discovered 'Rhaeto-Liassic' mammals and that of Sinoconodon indicates that there was greater diversity among the earliest known mammals than was previously thought.