Marcelo R. Sánchez-Villagra

A comprehensive morphological analysis of talpid moles (Mammalia) phylogenetic relationships

Some talpid moles show one of the most specialized suites of morphological characters seen among small mammals. Fossorial and more generalized shrew‐looking moles inhabit both North America and Eurasia but these land masses share none of the same genera. One of the central questions of mole evolution has been that of how many times specialized fossorial habits evolved. We investigated the origin of mole characters with a maximum parsimony analysis of 157 characters, mostly craniodental and postcranial, of representatives of all 17 living mole genera and three shrews and one hedgehog as outgroups. The result was one most‐parsimonious tree and its most novel aspect was the position of a Japanese shrew mole clade (Urotrichus, Dymecodon), which branched off after Uropsilus and was not closely related to the American shrew mole (Neurotrichus). The desmans (Galemys and Desmana) were the next clade in the tree, followed by Neurotrichus. We confirmed the monophyly of the Eurasian fossorial mole clade Talpini (Euroscaptor, Parascaptor, Mogera, Scaptochirus and Talpa). Condylura, the star‐nosed mole from North America, was sister group to a clade consisting of the Talpini plus Scaptonyx and the Scalopini (Scalopus, Scapanus, Parascalops, and Scapanulus). Based on our results and on the assumption that moles originated in Eurasia, it is most parsimonious to infer one migration from Eurasia to North America and two back‐migrations to Eurasia. It is ambiguous if Talpini and Scalopini evolved their full fossorial habits independently or not.

Fossil snakes from the Neogene of Venezuela (Falcón state)

Synopsis The first fossil snakes from Venezuela are described from three different Formations: Colombophis cf. C. portai and Boinae indet. from the Middle Miocene Socorro Formation, Boinae indet. from the Late Miocene Urumaco Formation and Viperidae indet. from the Pleistocene gravels at Cucuruchu. Although the Socorro and Urumaco Formations Boinae cannot be unambiguously referred to any particular taxa, the Venezuelan record, combined with the fauna from La Venta in Colombia, demonstrate that Boinae had achieved much of its more Northern distributions by the middle‐late Miocene. The occurrence of a viperid in the Pleistocene gravels at Cucuruchu is consistent with a geologically young record of the lineage in South America.

Homologies of the mammalian shoulder girdle: a response to Matsuoka et al. (2005)

Matsuoka et al. (2005) have used genetic labeling to map the fate of embryonic neural crest and mesodermal stem cells in the neck region. Based on their results, and on the assumption that skeletal homologies can be determined based on muscle attachments, they have discussed fundamental and classic wisdoms about the evolutionary identity of important structures of the mammalian shoulder girdle (Liem et al. 2001). Paleontological and comparative ontogenetic data are also critical to discuss hypotheses of skeletal homology, and contradict several of Matsuoka et al.’s statements. We consider important the discovery that many muscle tendons and their skeletal insertions as well as other connective tissue cells are in parts derived from the unsegmented postotic neural crest. However, we doubt that those cells can provide a priori better homology criteria for skeletal elements than traditional comparative and functional morphology. Because the authors have offered the cleithrum as test for their model, we would like to focus on this topic and review some evidence that has not been considered by Matsuoka et al. Matsuoka et al. (2005, p. 347) suggested that the cleithrum, ‘‘the major shoulder bone of extinct land vertebrate ancestors . . . seems to survive as the scapular spine in living mammals.’’ As is well known to mammalian anatomists and paleontologists, the spine to which Matsuoka et al. (2005) refer to is that of therians (marsupials and placentals) only, absent in the most basal living mammals, the monotremes (and many fossil cynodonts and primitive mammaliamorphs or stem-group mammals; Jenkins and Weijs 1979). This spine separates and serves as origin for the mesodermal supraspinatus and infraspinatus muscles in the lateral side of the scapula, an arrangement correlated with the evolution of parasagittal limb posture and movements (Fischer 2001). A homologization of the cleithrum with the therian scapula had been already suggested in the late XIX century (Seeley 1892; Broom 1899), and faces major challenges from both paleontological and comparative ontogenetic data. The cleithrum, an important exoskeletal element of the pectoral girdle of bony fishes, still forms part of the pectoral girdle in early synapsids, but a series of well-documented fossils shows its gradual reduction in the line leading to crown-group mammals (Kemp 2005; Kielan-Jaworowska et al. 2005). Furthermore, detailed studies of monotreme and therian pectoral development fail to trace any remnants of this element in perinatal stages (Klima 1973; Sánchez-Villagra and Maier 2002). The therian scapular spine is composed of two portions, which show a different pattern of development (Gro mann et al. 2002; Sánchez-Villagra and Maier 2003). Only the ventral, acromial portion of the scapular spine, which originates from the anterior margin of the scapular blade, is preformed in cartilage. The major dorsal portion is formed at a later stage by appositional bone, which expands from the perichondral ossification of the scapula into an intermuscular aponeurosis between the supraand infraspinatus muscles, which are derived from the somatic supracoracoideus muscle (Romer 1922, 1944; Cheng 1955); the cranial and caudal portions of the trapezius muscle insert only at the free end of the aponeurosis or spine, respectively. This intermuscular aponeurosis inserts more or less in the middle of the lateral surface of the developing scapula. Thus, the floor of the supraspinous fossa is already present at the beginning of scapular development, simultaneous with the infraspinous fossa. The portion of the scapular spine that is situated dorsal to the acromial process is hypothesized to be a neomorphic structure of therians. Based on the comparative anatomical and ontogenetic evidence a neural crest ‘‘cell population ghost’’ of the cleithrum (Matsuoka et al. 2005, p. 354) is not a useful concept to establish homology. If we homologize the therian scapular spine with the cleithrum of basal gnathostomes, or for that matter, with that of early synapsids (it was lost in early therapsids; Romer 1956), we need to be able to identify this structure in synapsids leading to crown-group mammals. EVOLUTION & DEVELOPMENT 8:2, 113 –115 (2006)