EvoWiki is now a project of the RationalMedia Foundation.
We are moving all content to RationalWiki.
See the EvoWiki project page for details!

A new gruiform from Messel

From EvoWiki

Jump to: navigation, search

Salmila robusta: A New Gruiform Mosaic from Messel and Comments on Avian Evolution



The Messel Oil Shales of Germany, of Middle Eocene age, are among the most famed deposists in the world, from a country renowned for its interest to paleornithologists. The Messel shales have yielded among the richest avifauna to be found anywhere in continental Europe, matched only by that of the Phosphorites du Quercy, in France. Representatives of no less than 17 extant and extinct orders of birds have been collected from Messel (Peters 1991, 1992b, Feduccia 1996). Much to the fortune of paleornitholgical science the fossils of the Messel shales are not only abundant, but they are preserved with astonishing fidelity.

The recent discovery of an additional specimen of the gruiform mosaic Salmila robusta, initially described by Mayr (2000b) from relatively incomplete remains, has shed new light on the Tertiary gruiform radiation, and demonstrated the cariamid affinity of the South American Psophiidae (Trumpeters), as well as the holophyly of Gruiformes, a topic of recent contention (see Mayr & Clarke 2003). Considering the vast adaptive radiation of the gruiform birds in the lower Paleogene, the importance on this new mosaic for resolving the tangled phylogenetic web of basal gruiforms, cannot be overestimated.


Salmilidae exhibit an amalgamation of psophiid and cariamid characters, upon the basis of which Mayr (2002) extricated Salmila from Cariamidae and placed it as an outgroup to Cariamae. While the osteology of Salmila has been reviwed elsewhere (Mayr 2000b), a revised diagonsis of the species and discussion of the traits which demonstrate its status as a gruiform mosaic, are in order.

Mayr (2002) delineated Salmila and Salmilidae (diagnosis same as for species) on the basis of the following characters:

  1. Furcula exceptionally robust
  2. Coracoid lacks foramen N. supracoracoidei
  3. Humerus robust, proximal end inflated
  4. Costal margin of sternum short, cadaul margins with one pair of deep incisions
  5. Ulna not exceeding length of humerus, and with a reduced olecranon process
  6. Trochlea carpalis between minor metacarpal and pisiform distinctly raised
  7. Tarsometatarsus shorter than ulna
  8. Hypotarsus with medial and lateral processes separated by a distinct sulcus
  9. Proximal end of first phalanx of fourth pedal digit displays a large and medially directed projection

Salmila displays a confusing mosaic of psophiid and cariamid traits, as indicated by those with which Mayr (2002) delineated the taxon. The osteology of the skull is broadly congruent with those of Psophiidae and Cariamidae, and in particular the postorbital process closely matches in morphology that of the trumpeters. There has been some debate regarding whether the beak is schizorhinal or holorhinal, as observed in Psophiidae and extant Cariamidae. Given that the nasal process of the premaxilla is broader than that of extant schizorhinal gruiforms such as the Gruidae, the latter intepretation seems most accurate.

The vertebral column lacks fusion of the thoracics to form a notarium as in trumpeters, cranes, sunbitterns, and kagus. The vertebral centra bear prominent pneumatic foramina, a trait consistent with Psophiidae but unknown in Cariamae. There are seven unfused caudals proximal to the pygostyle, and this is the plesiomorphic character state for Gruiformes (Livezey 1998). The pygostyle itself matches in general form that of Cariamae, while being distinctly different from the reduced pygostyle of Psophiidae.

The structure of the coracoid, ever useful in avian phylogenetics, is broadly congruent with that of Psophiidae, although there are distinct differences. The procoracoid is reduced, unlike the prominent process observed in psophiids, and the foramen nervi supracoracoidei is not present, a synapomorphic trait limited to Cariamae. The furcula is extremely robust, far more so than in most gruiforms and amongst extant gruiform taxa is matched only by the condition in Otididae. In both the Psophiidae and Cariamae the furcula is greatly reduced comparative to that in Salmilidae, and thus we have every reason to believe that Salmila was a far more talented aerialist than its living relatives. The sternum displays a strange mosaic of characters including those not particularly expected in gruiforms, and the carina resembles that of Anatidae in that it protudes strongly in cranial aspect. The costal margin of the sternum is shorter than that of extant Psophiidae, with fewer costal articulations. It most closely matches Cariamae in which only five costal ribs articulate with the sterunm, as in Salmila, although in cariamids the costal margin is longer than that in Salmilidae. Interestingly, the sternum of Salmila with its incised caudal margins, is at least superficially similar to that of Galliformes and Tinamiformes, though the incisures are not as deeply inscribed on the sternal blade as they are in the latter two taxa (see Fig. 5, Mayr 2002).

The osteology of the humerus, previously poorly understood, has been greatly clarified with the addition of the new Salmila material to existing collections. The humerus, particularly proximally, is extremely robust, with an inflated deltopectoral crest. It most closely is congruent in overall morphometric proportions with that Elaprohcnemus, a basal gruiform of uncertain affinity, but the humeri of Psophia and Cariama are broadly similar, though less robust. Again the only extant gruiform displaying as developed a humeral head as Salmila are the bustards (Otididae). Contrary to the character state in both Cariamae and Psophiidae, the transverse sulcus is distinctly incised, and readily delineated. The bicipital crest, in contrast to the deltopectoral, is small and not particularly well-developed. Distally, the ventral condyle is globular unlike that of Cariamae, and the dorsal condyle is in contrast elongate and narrowed, more closely matching the condition observed in Psophiidae. The ulna is little differentiated from that of Psophiidae and Cariamae, but is most closely congruent with that of Psophia. In a similar vein the carpometacarpus resembles that of both the Idiornithidae and the Psophiidae. Synapomorphic characters of Psophiidae observed in the carpometacarpus of Salmila include the presence of a distinct ventral tubercle on the minor metacarpal, and the elevated face of the trochlea carpalis between the pisiform and minor metacarpal.

The pelvis is less derived than that of Psophiidae and is thus among the most strikingly generalized structures in the samilid skeleton. The femora are more gracile than those of Cariamae, though they match in morphometric detail those of Idiornis as well as they do those of Psophia. The tibiotarsus is not particularly well preserved, but from what is known of it, it seems most similar to the trumpeters. The structure of the hypotarsus has been critical in shedding new light on the phylogenetic affinities of Salmila within Gruiformes. The hypotarsus is unlike that of Cariamae, and is elongate with a deep sulcus separating the crista lateralis and crista medialis. The pes is typically gruiform in construction. The projection of the proximal end of the first phalanx of pedal digit IV is a trait more congruent with that of Psophiidae, and was considered by Mayr (2002) to be an autapomorphy of the species.

Phylogenetic Perambulations

Salmila robusta displays a mosaic of synapomorphic characters with Psophiidae and both extant and extinct Cariamae, and is at the same time a primitive gruiform as indicated by the presence of multiple gruiform symplesiormophies in its skeleton (e.g., limb proportions). It is thus an ideal transitional fossil between two grades of organization in gruiform evolution, with implications for our understanding of gruiform phylogeny as a whole.

Mayr (2000b) initially classified Salmila as a basal cariamid, but with the addition of new material reported in Mayr (2002) it has become clear that Salmilidae more closely resemble in most aspects of their morphology the extant Psophiidae, while at the same time possessing a higher level of organization which precludes their placement within Psophiidae. Mayr (2002) argued that Salmilidae, Psophiidae, and Cariamae collectively constitute a clade with either Elaphrocnemus or Gruidae as sister clades. Synapomorphies of (Salmilidae+Psophiidae+Cariamae) include:

  1. Beak hooked
  2. Ulna with reduced olecranon
  3. Trochlea carpalis elevated between pisiform and minor metacarpal
  4. Minor metacarpal bowed, with a proximally located, robust tubercle in ventral aspect.
  5. Ischiopubic fenestra extremely narrow

Despite its close similarity to the trumpeters, Salmila possesses autapomorphic characters of the Cariamae, indicating that it may be near to the origin of this gruiform assemblage. Such characters include the structure of the coracoid and the loss of the foramen nervi supracoracoidei. The discovery of a psophiid/cariamid mosaic is broadly congruent with century old observations that trumpeters and seriemas are clearly similar, and constitute each others closest living relatives (e.g., Beddard 1890). This new fossil data also stands in stark contrast to several recent studies which have argued for the holophyly of Psophiidae + Rallidae (Houde et al. 1997), and genetic distance data which proposed a Psophiidae + Gruidae + Aramidae clade (Sibley & Ahlquist 1990). Considering recent proposals that Gruiformes is not holophyletic, (e.g., Mayr & Clarke 2003), and that Cariamidae are more closely related to the Opisthocomidae, the osteology and phylogenetic affinities of Salmila are all the more interesting. Given that this taxon convincingly demonstrates a close relationship between the Psophiidae and Cariamae, the conclusion that Gruiformes is polyphyletic must be regarded with skepticism.


  1. Beddard, F. E. 1890. On the structure of Psophia and on its relations to other Birds. Proceedings of the Zoological Society of London 1890: 329-341.
  2. Feduccia, A. 1996. The Origin and Evolution of Birds, First Edition. Yale University Press, New Haven.
  3. Houde, P., Cooper, A., Leslie, E., Strand, A., & Montano, G. 1997. Phylogeny and evolution of 12S rDNA in Gruiformes (Aves). In: Mindell, D. P. (ed.), Avian Molecular Evolution and Systematics: 121-158.
  4. Livezey, B. C. 1998. A phylogenetic analysis of the Gruiformes (Aves) based on morphological characters, with an emphasis on the rails (Rallidae). Philosophical Transactions of the Royal Society of London (B) 353: 2077-2151.
  5. Mayr, G. 2000b. A remarkable new "gruiform" bird from the Middle Eocene of Messel (Hessen, Germany). Paleontologische Zeitschrift 74(1/2): 187-194.
  6. Mayr, G. 2002. A new specimen of Salmila robusta (Aves: Gruiformes: Salmilidae n. fam.) from the Middle Eocene of Messel. Paleontologische Zeitschrift 76(2/2): 305-316.
  7. Mayr, G. & Clarke, J. 2003. The deep divergences of neornithine birds: a phylogenetic analysis of morphological characters. Cladistics 19: 527-553.
  8. Peters, D. S. 1991. Zoogeographical relationships of the Eocene avifauna from Messel (Germany). Proceedings of the Twentieth International Ornithological Congress, 572-577.
  9. Peters, D. S. 1992b. Messel birds: a land-based assemblage. In: Schaal, S. & Ziegler, W. (eds.), Messel: An Insight into the History of Life and of the Earth, 135-151.
  10. Sibley, C. & Ahlquist, J. 1990. Phylogeny and Classification of Birds: A Study in Molecular Evolution. Yale University Press, New Haven.
Personal tools