Swimming patterns of Malagasyan Triassic fishes and environment

Fig.1 : Map of Malagasy and location of outcrops.

Laurent BARBIERI :Laboratoire de Paléontologie des Vertébrés. Case 106. Tour 15. 3° étage. Université Pierre et Marie Curie (Paris 6). 4, Place Jussieu. 75252 PARIS cedex 05
Michel MARTIN : Musée d’Histoire Naturelle. 115 bd Eurvin. F 62200 Boulogne/mer.

The basis of this study is a sample of Early Triassic fishes from Malagasy (Northwestern area) collected by the team of the Muséo Civico di Storia Naturale di Milano. The dentition and the general morphology have been used in order to deduce the swimming patterns of the different genera on the basis of Webb’s study (Webb 1984). The environment of fossil forms is deduced from comparison to living forms of similar swimming patterns and dentition. We accept that similarities in locomotion and feeding imply similar environments. Of course, it is possible to conduct this study only on the basis of well preserved and consequently neither long tranported nor reworked material. Moreover the accumulation of fishes in the same level does not mean that the different species were living in the same environment, especially when the faunas of different horizontal sections of water are mixed in the same assemblage.

The paleoecological conclusions of this kind of study must be tested on the basis of associated invertebrates and geological data. In this paper we intend to study the ecological change in four Malagasyan localities (Fig.1) (two collected by the Italian team and two after Nielsen 1961) and in four fossiliferous levels from a single locality in Greenland (Cape Stosch. Central East Greenland after Nielsen 1961). The assemblages from Britsh Columbia (Vega-Phroso after Schaeffer and Mangus 1976) and from Spitzbergen (after Stensiö 1921 and 1925) have been also studied

Principles and definition

We count the number of individuals in every genus and then calculate the percentage of every genus in each locality. The distribution of percentage being indicative of the ecology. The four swimming types defined by Webb (1984) are present in the Malagasyan faunas(Fig.2). The other assemblages are very similar from the systematic point of view.

Fig. 2 : Comparison of Early Triassic genera to living Actinopterygians A. Saurichthys I. Serranus hepatus B. Belone belone J. Pteronisculus C. Australosomus K. Paracentrophorus D. Scomber scombrus L. Birgeria E. Bobasatrania M. Salmo trutta F. Stromateus fiatola N. Boreosomus G. *Perleidus* O. Sardina pilchardus H. Parasemionotus A-D maitained swimming E-F manoeuvrers G-I accelerators J-M accelerator generalists N-O generalists.

– the manoeuvrers for example Bobasatrania
– the “acceleration swimmers” like *Perleidus*, Pteronisculus, Paracentrophorus and the Parasemionotidae and Birgeria (we used *Perleidus* since Lombardo 1995 concluded that the Early Triassic fishes previously refered to this genus are different from the type species on the base of the caudal skeleton)
– the specialists of “maintained swimming” like Saurichthys and Australosomus
– a single generalist swimmer is recorded : Boreosomus

Swimming patterns, morphology and environment.

The “manoeuvrable swimming” is used by fishes which need quick accelerations for capturing their preys. Theses fishes possess a deep caudal fin and backwardly displaced impaires fins. They usually inhabit midwater or near the bottom on irregular substrates in wait for a prey, making a short final dash at them.

In maintained swimming (used by the scombridae among the living forms) the speed is high and constant. Theses fishes possess a narrow caudal peduncle and the caudal fin exhibits a semilunar form. They live in open sea.

The generalist swimmers are able to use the three patterns according to the changing conditions and for this reason, they cannot be indicative of any environment.

At first, we note that Australosomus and Saurichthys are never present at a high percentage together in the same locality or level. Australosomus was probably microphageous meanwhile Saurichthys was a predator. We assume that Australosomus tried to avoid the dangerous Saurichthys which feed partly on it.



In Malagasy, four localities are studied from North to South : Bobasatrana, Ankitokazo, Iraro and Anaborano (Fig. 1, 3). In Bobasatrana, there are only a few generalists and the accelerators and the manoeuvrers are hugely dominant. This assemblage is indicative of very irregular, probably rocky bottom. A high percentage of coelacanths is recorded in this locality which could have the same significance according to Belles-Isles (1992). However the percentage of Australosomus is indicative of a moderate depth. From the North to the South, the percentage of accelerators and manoeuvrers decreases suggesting again rocky bottom but associated to an increased depth. Finally, in Anaborano, only about 30% of accelerators and almost no manoeuvrers are present suggesting a more regular bottom may be associated to a larger depth corresponding to the 20% of Australosomus and Saurichthys.

In the same vertical section, in Greenland (Fig. 4), the fauna has recorded chronological variation in the environment. The manoeuvrers and accelerators suggest the presence of an irregular rocky bottom associated to changes in depth or connections with open sea. Theses conclusions are supported by the data based on the ammonites. The bottom was much more irregular in zone II than in zone V but the depth was probably the most important in zone IV.

In British Columbia, the assemblage is only indicative of an irregular substrate. This conclusion is supported by the lithological data.

In Spitzbergen, the fishes collected in many localities are indicative of an open sea.

In Malagasy, the ecological conclusions based on the swimming patterns are both supported by the data provided by the invertebrate fauna (Garassino and Teruzzi 1993) the Stegocephalians (Lehman 1961 and 1966) and the lithology (Besairie 1972).


BELLES-ISLES, M. (1992): The modes of swimming of Sarcopterygians. In Fossil Fishes as living animals. Mark-Kwik E. (Ed.). Academia, 1. p.117-130. Academy of Sciences of Estonia. Tallinn

BESAIRIE, H. (1972): Géologie de Madagascar, I. Les terrains sédimentaires. Ann. Géol. Madagascar, 35:p.27-32.

GARASSINO A. and TERUZZI G. (1993) : Studies on Permo-Trias of Madagascar. 3. The decapod crustaceans of the Ambilobè region (NW Madagascar). Atti Soc. it. Sci. nat Museo civ. Stor. nat. Milano.,Milano, 134(1) p.85-113.

LEHMAN J.P. (1961) : Les Stégocéphales du Trias de Madagascar. Ann. Paléont., Paris, XLVII. p.111-154

LEHMAN J.P. (1966) : Nouveaux Stégocéphales de Madagascar. Ann. Paléont., Paris, LII(2) : p.117-139

LOMBARDO C. (1995): Perliedus altolepis (Actinopterygii, Perlei diformes) from the kalkoschieffezone of Ca del Frate (N. Italy). Geobios H.S., Lyon. 19 : p.211-219.

NIELSEN, E. (1961): Geology of the Arctic. Vol I. On the Eotriassic fish faunas of Central East Greenland. In Raasch, G.O. (Ed), Toronto, Univ. Toronto Press: p.255-257.

SCHAEFFER, B. and MANGUS, M. (1976): An Early Triassic fish assemblage from British Columbia. Bull. Am. Mus. Nat. Hist., 156(5): p.515-564.

STENSIÖ, E.A. (1921): Triassic fishes from Spitzbergen. Part. I. Adolf Holzhausen (Ed), Vienne, 307 p.

STENSIÖ, E.A. (1925): Triassic fishes from Spitzbergen. Part. II. Kungl. Sven. Vet. Akad. Handl., Stockolm, 3(2): 261 p.

WEBB, P.W. (1984): Body form, locomotion and foraging in aquatic vertebrates. Amer. Zool., 24 : p.107-120.


Fig. 3


Fig. 4

Tilmelding til DGF Nyhedsbrev