Friday, May 17, 2013

The Southernmost Atlantic Seacows

Its been a while since I posted news on fossil sirenians. I've been very busy with fieldwork, manuscripts, among other things. The Spring interns have now gone back home. So, while I wait for the arrival of the next round of interns, here's the latest on fossil sirenians.

Where are sirenians found

With the exception of the now extinct Steller's seacow (Hydrodamalis gigas), all extant sirenians have tropical to subtropical distribution, with some species having a notably broad latitudinal and longitudinal distribution (Marsh et al., 2011). But, when we look at the fossil record of sirenians, we see a slightly different pattern of distribution, mostly tied to tectonic and/or climatic events. For example, during parts of the Cenozoic global temperatures were higher than today (Zachos et al., 2001), so you find fossils of sirenians far off their modern range (e.g. Belgium). These climatic variations amongst other physical drivers have played a prominent role in the distribution of seagrasses and seacows (expect more on this in the nearby future).

Nowadays, in the Western Atlantic and Caribbean (WAC) region, the most common and widespread sirenian is the West Indian Manatee (Trichechus manatus) whose range extends from as far north as the Carolina's (with some individuals reaching New England) to northeastern Brazil; another species found in the region is the Amazonian manatee (Trichechus inunguis) which lives in the Amazon basin (see map below). But, it hasn't always been like this. Throughout most of the Cenozoic, dugongids, a group of sirenians are now restricted to the Indo-Pacific region, were the predominant seacow group in the WAC, including multispecies communities in the region (Domning, 2001; Velez-Juarbe et al., 2012a; see previous post on this subject). Fossil of dugongids in the WAC are found in deposits as far north as Maryland, and as far south as Argentina. However, these southernmost dugongids, are poorly known, and have had a somewhat rocky taxonomic history.

From Metaxytherium to Dioplotherium a case of mistaken identity

The most common, and temporally and geographically widespread seacow genus known is the Halitheriine dugongid Metaxytherium. Species of this genus are known from late Oligocene through Pliocene deposits, and are found from the Eastern Pacific, Caribbean, Western and Northern Atlantic, and Western Tethys regions (e.g. Domning, 1988; Sorbi et al., 2012). Therefore it shouldn't have been much of a surprise when Roy H. Reinhart (1976) described a molar from the late Miocene Paraná Formation of Entre Ríos, Argentina as that of Metaxytherium. The importance of this find, lies in that prior to its discovery, the youngest species of Metaxytherium known from the WAC was the middle Miocene M. floridanum, which is not known outside of Florida (Domning, 1988). The Paraná molar was then, the youngest and southernmost record of the genus from the Western Atlantic.

However, species of Metaxytherium display a generally conservative morphology, and because of this, it has had a long, somewhat convoluted, taxonomic history. This is, fortunately, slowly being resolved as most species of Metaxytherium have been re-described (e.g. Domning, 1988; Domning & Pervesler, 2001; Sorbi et al., 2012) and studied in detail within the last 25 years, giving us, paleosirenologist a better idea of the valid species within the genus and variation within each species. Since Reinhart's description, several workers (Cozzuol, 1996; Cione et al., 2000; Domning, 2001) have disagreed with his interpretation regarding the affinities of the Argentinian molar. All of them referring the Paraná molar to Dioplotherium, still a dugongid, but one that belongs to the Dugonginae, a group very different from that to which Metaxytherium belongs. And indeed, the overall morphology of the tooth conforms well with what we know about Dioplotherium, it is in fact, very similar to those of Dioplotherium cf. D. allisoni from the early Miocene of Brazil (Toledo & Domning, 1991). This meant that Metaxytherium may have gone extinct in the WAC at the end of the middle Miocene (Domning, 1988), and that the genus only reached as far south as northeastern Brazil (Toledo & Domning, 1991), or did it?

Left: Map showing the distribution of Miocene seacows throughout the Americas. (ER = Entre Ríos).
Right: Map showing the distribution of extant sirenian in the Americas.
(Click on the map to view larger version.)

New fossils from the Paraná Formation

A couple of years ago I received an email from an Argentinian colleague, Jorge Noriega from CONICET in Diamante, informing me of a new discovery from the Paraná Formation in Entre Ríos. The new fossils consisted of left and right partial maxillae and most of the molars of a single individual (see figure below). At this point I was close to finishing my PhD, which meant that I had look at a lot of specimens and was well acquainted with the morphology of most, if not all Oligocene through Pliocene sirenians. Once I looked at the pictures of the new material, I quickly recognize these as most likely representing a species of Metaxytherium.
Molars of Metaxytherium from the late Miocene Paraná Formation. 1-2) left maxilla and M1-3 in occlusal view. 3-4) right maxilla and M3 (modified from Velez-Juarbe et al., 2012b)
Now, I must admit that dugongid teeth are not the most diagnostic, so figuring out if these actually belonged to Metaxytherium was not an easy and quick task. After a considerable amount of reading, and detailed observations of material from various species of Metaxytherium as well as other dugongids I was confident they belonged to that genus. And so, working together with Jorge and Brenda Ferrero (also from CONICET in Diamante) we took on the task of formally re-designating the fossil described by Reinhart (1976) as well as describing the new material which actually represented a species of Metaxytherium (Velez-Juarbe et al., 2012b). The new Parana molars are quite similar to those of the middle Miocene Metaxytherium floridanum, but, their dimensions are below the range exhibited by M. floridanum and may represents a different species. One of the positive outcomes resulting from this work, was realizing that teeth of dugongids can sometimes be of taxonomic usefulness. We noticed, that the molars of some of the more derived species of Metaxytherium often have additional cusp and/or cuspules, a derived character which is not observed in Dioplotherium or any of its kin (i.e. Dugongines). The contemporaneous presence of both, Dioplotherium and Metaxytherium is not something unheard of. This same duet, occurs in the late Oligocene of Florida, early Miocene of Brazil and possibly in the middle Miocene of California and Baja California (Domning, 2001; Velez-Juarbe et al., 2012a). This again shows that multispecies communities and niche partitioning seems to have been the norm, not the exception, throughout sirenian history.


Cione, A. L., M. M. Azpelicueta, M. Bond, A. Carlini, J. Casciotta, M. A. Cozzuol, M. de la Fuente, Z. Gasparini, F. Goin, J. Noriega, G. Scilato-Yané, L. Soibelzon, E. Tonni, D. Verzi, and M. G. Vucetich. 2000. Miocene vertebrates from Entre Ríos Province, Argentina. INSUGEO, Serie Correlación Geológica 14:191-238.

Cozzuol, M. A. 1996. The record of the aquatic mammals in southern South America. Münchner Geowissenschaftliche Abhandlungen A30:321-342.

Domning, D. P. 1988. Fossil Sirenia of the West Atlantic and Caribbean region. I. Metaxytherium floridanum Hay, 1922. Journal of Vertebrate Paleontology 8:295-426.

Domning, D. P. 2001. Sirenians, seagrasses, and Cenozoic ecological change in the Caribbean. Palaeogeography, Palaeoclimatology, Palaeoecology 1:27-50.

Domning, D. P., and P. Pervesler. 2001. The osteology and relationships of Metaxytherium krahuletzi Depéret, 1895 (Mammalia: Sirenia). Abhandlungen der Senckenbergischen Naturforschenden Gessellschaft 553:1-89.

Marsh, H. D., T. J. O'Shea, and J. E. REynolds, III. 2011. Ecology and conservation of the Sirenia: dugongs and manatees. Cambridge University Press, 521p.

Reinhart, R. H. 1976. Fossil sirenians and desmostylids from Florida and elsewhere. Bulletin of the Florida State Museum, Biological Sciences 20:187-300.

Sorbi, S., D. P. Domning, S. C. Vaiani, and G. Bianucci. 2012. Metaxytherium subapenninun (Bruno, 1839) (Mammalia, Dugongidae), the latest sirenian of the Mediterranean Basin. Journal of Vertebrate Paleontology 32:686-707.

Toledo, P. M., and D. P. Domning. 1991. Fossil Sirenia (Mammalia: Dugongidae) from the Pirabas Formation (Early Miocene), northern Brazil. Boletim do Museu Paraense Emílio Goeldi, Série Ciencias da Terra 1:119-146.

Velez-Juarbe, J., D. P. Domning, and N. D. Pyenson. 2012a. Iterative evolution of sympatric seacow (Dugongidae, Sirenia) assemblages during the past ~26 million years. PLoS ONE 7(2):e31294.

Velez-Juarbe, J., J. I. Noriega, and B. S. Ferrero. 2012b. Fossil Dugongidae (Mammalia, Sirenia) from the Paraná Formation (late Miocene) of Entre Ríos Province, Argentina. Ameghiniana 49:585-593.

Zachos, J., M. Pagani, L. Sloan, E. Thomas, and K. Billups. 2001. Trends, rhythms, and aberrations in global climate 65 Ma to present. Science 292:686-693.


JORDI B B said...

Un artículo interesante, pero hay una comentario que me ha generado una duda. Mencionas que la morfología dentaria en dugungidos no es diagnóstica, y con ello entiendo que no hallamos un patrón que defina, por ejemplo, a Metaxytherium y lo diferencie de otros géneros. Por los artículos que he leído, no te falta razón y no es habitual encontrar diagnosis de especies o géneros basadas en morfologías dentarias como sí ocurre en otros órdenes de mamíferos.

En cambio, en artículos como el de D.P.Domning (2001) se establece una relación entre la dieta (tipos de fanerógamas, tamaño de rizomas, consumo de hojas) y determinadas características morfológicas (ángulo de deflexión rostral, tamaño del primer par de incisivos superiores). Es curioso que el tipo de dieta no implique cambios específicos en los molares. Pienso que no sería raro pensar que dos dugóngidos que se alimenten de plantas con rizomas muy distintos mostrarían patrones distintos (distinto número de cúspides, presencia de cúspides accesorias, cíngulos y cosas por el estilo).

¿Cual es tu opinión?

J. Velez-Juarbe said...

Hola Jordi,

Pues si, basado en las diferencias dietéticas, uno esperaría ver más diferencias en la morfología de los molares, y si se ven, solo que en las especies relativamente recientes o en grupos muy derivados. Los molares de Dugong dugon, los Trichechinae y los Hydrodamalinae son bastante distintos (de hecho en Hydrodamalis spp. están ausentes). Yo escribí un poco sobre eso en una entrada anterior.

El problema es que en muchas de las especies fósiles estas diferencias son un tanto más sutiles o no han sido estudiadas en detalle, como en el caso de los Metaxytherium, y por eso fue que pude determinar a que género pertenecían los molares de Entre Ríos. El otro problema, es que hay que tener en cuenta que los molares puede ser muy variables en algunas especies (como por ejemplo M. floridanum [Domning, 1988]) y que no siempre tenemos más de un ejemplar de cada especie para saber si varían o no, o para ver como varían a medida que se van desgastando, lo cual dificulta las comparaciones.

En fin, todavía queda mucho por hacer para determinar bien cuan útil pueden ser los molares de los sirenios para identificar las distintas especies. Como en todo, hay mucho trabajo por delante, y no tantos dispuesto a hacerlo.