The Miguasha fossil site in eastern Québec was among the first major paleontological localities to have been discovered and excavated in North America. The discovery of the first fossils at Miguasha was made in 1842 by Abraham Gesner, the government geologist in New Brunswick, also known for his work on the distillation of kerosene. While surveying the northern part of New Brunswick for coal, Gesner came to Miguasha and reported “I found the remains of fish, and a small species of tortoise with foot-marks” (Gesner 1843, p. 64). Evidently, this fossil was not a tortoise but rather a placoderm fish, most likely Bothriolepis canadensis, one of the most common fish from the Escuminac Formation. Although fossil plants were found, there was no potential in terms of coal mining, and the site was forgotten for more than 30 years.

Between 1879 and 1881, the Geological Survey of Canada organized several expeditions to Miguasha lead by R. W. Ells (Fig. 1a), A. H. Foord and T. C. Weston. A few dozens of collected fossils were given to pioneer paleontologists: Joseph F. Whiteaves, a British paleontologist working at the Geological Survey of Canada, who studied the fossil fishes, and Sir J. William Dawson, a Canadian paleobotanist working at McGill University, who looked at the fossil plants. These workers were the authors of the first scientific descriptions of the Miguasha fossils (Whiteaves 1880; Dawson 1882). From the late 1880s until the 1940s, British and American paleontologists came to Miguasha (frequently referred to mistakenly as Scaumenac Bay) in order to collect new material for major museums, such as the British Museum of Natural History (London, England), the Royal Scottish Museum of Edinburgh (Scotland) and the American Museum of Natural History (New York, USA). From 1887 to 1892, M. Jex collected an impressive array of fossil fishes in Miguasha, which he sold to different museums in the United Kingdom. As a result, part and counterpart of the same fossils were sold separately to paleontological collections in Edinburgh and London. From this fossil hunting, four fish species were named by two famous British paleontologists, R. H. Traquair and A. S. Woodward.

In 1892, the American vertebrate paleontologist E. D. Cope was the first to recognize that the osteolepiform Eusthenopteron foordi from Miguasha had a fin anatomy similar to that of the limbs of stegocephalians (Cope 1892), a paraphyletic group acknowledged today to include stem tetrapods. Fossils from Miguasha thereby made their entrance to studies documenting the origin and evolution of major groups of vertebrates, a perspective fairly new at the time, considering the Darwinian revolution. Following Cope’s (1892) publication, Eusthenopteron foordi was considered a key species in the transition from fishes to tetrapods, thus promoting the focus of numerous studies on the anatomy of its paired fins, vertebral column and nostrils. Between 1905 and 1993, local collectors (Fig. 1b–c) familiar with fossil fish hunting in Miguasha were pivotal in creating reference collections now available around the world (Lemieux 1996). In the early 1920s, Swedish paleontologists started to study in great detail the anatomy of fishes from Miguasha, contributing, in part, to recognition of the famous ‘Swedish School’ of paleozoology at the Naturhistoriska Riksmusett in Stockholm (Schultze 2009). Between 1937 and 1998, Erik Jarvik wrote some thirty scientific articles on Eusthenopteron (Cloutier 1996c), while Erik Stensiö published on the detailed anatomy of the placoderm Bothriolepis (Stensiö 1948).

The history of the Miguasha biota can also be tracked through the sequence of scientific publications and the date of original descriptions for the various vertebrate species (Fig. 2). From 1880 to 1900, half of the known vertebrate diversity had been described on the basis of original collecting in Miguasha. Between 1900 and 1924, most likely as a result of global social instability and World War I, paleonto-logical research, including that on the Miguasha biota, was not a priority in the scientific community. Renewed interest in the fossil material collected in Miguasha by British, Swedish and American paleontologists stimulated a burst of taxonomic descriptions in the mid to late ‘30s, which eventually faded out with World War II.

Since 1960, a great deal of attention has been given to the evolutionary interpretation of the Miguasha fossil fishes. Reinterpretation of the fossil material in various collections resulted in the fluctuation of species-level diversity figures until a thorough revision of the Miguasha biota was put together in 1996 (Schultze and Cloutier 1996). Emphasis then shifted from a descriptive perspective to an interpretive one, where the phylogeny, the paleoenvironment, and the taphonomy became important research focuses, and it was at this time that Miguasha was perceived as a Lagerstätte. For the past 168 years, fossils from the Miguasha biota have contributed greatly to our understanding of the early evolution of vertebrates. Over the years, at least 367 scientific papers (Fig. 2) have been published by 230 authors from 25 countries. As a modern indicator of interest, more than 200,000 pages on the internet mention fossils from Miguasha.

Aware of the international interest in Miguasha fossils, the Government of Québec created a conservation park in 1978 to protect the Escuminac Formation. Starting in the 1990s, this initiative stimulated renewal of interest in fish paleontology among Canadian researchers who received their Ph.D. training either in Canada (F. Charest), the USA (R. Cloutier) or in France (M. Belles-Isles, P.-Y. Gagnier, D. Vézina), and who studied different perspectives of the Devonian Miguasha biota. In 1999, the Miguasha National Park was confirmed as a UNESCO World Heritage Site in recognition of its global status as the best representation of the Devonian ‘Age of Fishes’ (Cloutier and Lelièvre 1998).

Over the past 20 years, the focus on the biota has shifted from detailed morphological studies, to the phylogenetic positioning of the Miguasha taxa, to the paleoenviron-mental reinterpretation of the Escuminac Formation, to studies of exceptional preservation, including documentation of numerous fossilized ontogenies. In combination, these studies make the Miguasha biota one of the most intensively investigated paleontological sites anywhere.

Miguasha National Park is located in eastern Québec on the south coast of the Gaspé Peninsula along the estuary of the Restigouche River (Fig. 3a). The Miguasha biota is contained in the Late Devonian Escuminac Formation (119 m thick), which outcrops primarily as a cliff (from three to 30 m high) along the Restigouche River. Additional satellite outcrops, located between 3 and 40 km away from the main cliffs, have been discovered over the past ten years (Fig. 3c). In 2007, the main Miguasha exposures (Fig. 3b) were named the ‘René Bureau Cliffs’ to honour a self-taught geologist and paleontologist (Fig. 1c) whose work in the 1930s and 1950–1970 was pivotal in establishing this site as a Québec conservation park.

The Escuminac Formation conformably overlies the Fleurant Formation (conglomerate containing sandstone lenses); together, these units constitute the Miguasha Group. Rust et al. (1989) interpreted the Fleurant Formation as proximal alluvium deposits. The Lower Carboniferous Bonaventure Formation (alternating conglomerate and coarse sandstone) rests disconformably on the Escuminac Formation and are interpreted to represent proximal alluvial fan deposition (Rust et al. 1989).

The Escuminac Formation is considered to be middle Frasnian in age, or 385 to 374 Ma according to the timescale of Walker and Geissman (2009), based on miospore content and the fish assemblage (Cloutier et al. 1996; Elliott et al. 2000). Although a precise and absolute dating of the duration of the Escuminac Formation is not possible, the timespan is estimated to lie between 59.5 ka and 2500 ka (Cloutier et al. 2011).

Numerous sedimentological descriptions of the Escuminac Formation show the importance of the alternation between siltstone – sandstone and shale (Alcock 1935; Russell 1939; Dineley and Williams 1968a, b; Carroll et al. 1972; Hesse and Sawh 1982, 1992; Vézina and Cloutier 1991; Prichonnet et al. 1996; Cloutier et al. 2011), and suggest that the sequence comprises lacustrine or estuarine turbidite deposits (Hesse and Sawh 1992; Prichonnet et al. 1996). The clear alternation of lithology allows recognition of 394 individually numbered horizons from the base to the top of the Escuminac Formation at René Bureau’s Cliffs (Fig. 3d). As a result of such precise stratigraphic positioning, systematic collecting has enabled documentation of the stratigraphic distribution of various Escuminac species (Cloutier et al. 1996; Cloutier et al. 2011) and of taphonomic modes (Parent and Cloutier 1996; Cloutier et al. 2011).

The depositional environment of the Escuminac Formation has been variously considered as lacustrine, estuarine, coastal marine or marine. An estuarine interpretation best accommodates the different lines of evidence provided by the fauna (Schultze and Cloutier 1996), the palynofacies (Cloutier et al. 1996), the trace fossil assemblage (Maples 1996), the sedimentological and stratigraphic setting of the formation (Hesse and Sawh 1992; Cloutier et al. 2011), and the geochemistry of the sedimentary rocks and bones (Schmitz et al. 1991; Vézina 1991; El Albani et al. 2002; Matton et al. 2012).

Over the past 15 years, we have accumulated evidence that the Miguasha biota inhabited an estuarine paleoenvironment, starting with the discovery of acritarchs in association with a large part of the fish assemblage (Cloutier et al. 1996). Furthermore, characterization of the organic matter by kerogen pyrolysis has revealed a predominance of Type II organic matter, which suggests a mixing from continental and marine sources (El Albani et al. 2002). Geochemical analyses of both sediment and bony elements from the Escuminac Formation also suggest a transitional environment. The strontium isotope ratio of Escuminac bony fragments reveals a Frasnian seawater signature, suggesting a slight diagenetic contamination, although fitting within the transitional range (Matton et al. 2012). Some facies contain evidence of tidal influence, such as thin rhythmites showing daily and lunar cycles (Cloutier et al. 2011). These were originally interpreted as lacustrine varves. Thinning of the rhythmites is associated with neap tides, whereas a thickening is associated with spring tides. Such rapid cyclic deposition, characteristic of the lower part of the Escuminac Formation, facilitated rapid burial of fish carcasses and allowed for the exceptional preservation typical of this facies. Tidal rhythmites provide a perfect host for Konservat– and Konzentrat–Lagerstätte horizons within the Escuminac Formation.

Five transgressive – regressive sequences have been recognized in the Escuminac Formation; these sequences are consistent with an inner wave-dominated estuary showing a shift towards continentalization (Cloutier et al. 2011). Throughout the formation, changes in richness, abundance and composition of the fish assemblages show that the transgressive phases are more diverse and better structured in terms of species composition than the regressive phases (Cloutier et al. 2011). In addition, rare and sporadic taxa, such as Levesquaspis, Plourdosteus, Miguashaia and Elpistostege, are found solely in the transgressive phases, whereas the most abundant species (Fig. 6) are found in both transgressive and regressive phases (Cloutier et al. 2011).

Fossils encountered in the Escuminac Formation show a wide range of preservational modes (Parent and Cloutier 1996; Cloutier et al. 2011), from specimens with no apparent sign of decay down to isolated bony elements. Contrary to popular perception, not all levels qualify as Lagerstätte horizons. Where exceptional quality of preservation predominates, the term Konservat–Lagerstätte can be applied; where diversity and abundance are outstanding, the horizon represents a Konzentrat–Lagerstätte.

The exceptional condition of preservation is indicated by the great number of fish specimens for which completely articulated specimens were found, such as multi-element skeletons of osteostracans and acanthodians (Fig. 8k–n). Although most specimens are compressed laterally, three-dimensional preservation is fairly common (Figs. 4d–e, 8a–c, e, 10a; Parent and Cloutier 1996). Not only is the external morphology well preserved, but detailed bone histology is also present either as hard tissues (e.g. enamel, dentine, cellular bone, cartilage) or as cell spaces (e.g. osteocytes, chondrocytes); examples include cartilages in Euphanerops (Janvier and Arsenault 2002), bone tissues in Bothriolepis (Downs and Donoghue 2009) and Eusthenopteron (Laurin et al. 2007; Zylberberg et al. 2010), and dental tissues of Scaumenacia (Thomson 1972; Smith et al. 1987) and Eusthenopteron (Schultze 1969). Recent investigations have focused on exceptional cases of fossilization where preservation of soft tissues such as digestive tracts, stomachs, spiral intestines, lungs, gill filaments, blood vessels (Fig. 10b) or muscles is involved (Arsenault et al. 2004; Janvier et al. 2006; Janvier et al. 2007; Cloutier 2009; Janvier and Arsenault 2009; Arsenault and Janvier 2010). In addition, preliminary studies on nonmineralized parts of the Bothriolepis body suggest that ‘skin preservation’ results from the presence of a bacterial mat outlining the surface of the carcass (Fig. 10a, c); this phenomenon is known as pseudomorphing (Briggs 2003).

Fossilized ontogenies are well represented in the Escuminac Formation. Exceptionally preserved larval and juvenile specimens (as small as 6 mm in total length) have been identified for 14 out of the 20 species of fishes (Cloutier et al. 2009), providing the opportunity to study growth and developmental change (e.g. Fig. 11a, f–g). One of the best examples of a fossilized growth series is that of the placoderm Bothriolepis, where specimens as small as 5 mm long to approximately 22 cm of shield length are known (Cloutier 2010; Fig. 11a). With such material, size and shape changes can be studied (Thomson and Hahn 1968; Schultze 1984; Werdelin and Long 1986; Cloutier 1997; Cloutier et al. 2009), as well as the process of ossification of some of these species (Cote et al. 2002; Cloutier 2010; Béchard and Cloutier 2011). These data can be compared among fossil and living taxa in order to study the evolution of development (Cote et al. 2002; Cloutier et al. 2009; Cloutier 2010).

Concentration of fossils is not limited to fish fragments, but extends to complete specimens. Konservat–and Konzentrat–Lagerstätte horizons occur throughout the Escuminac Formation. Certain horizons are best considered as Konzentrat–Lagerstätten because of their richness and abundance. One of the most outstanding horizons is bed 8 (see lower * in Fig. 3d; Fig. 11e), which includes 11 species. Many years of systematic collecting in bed 8 (Fig. 1d) provided the opportunity to create a GIS (geographical information system) plot in which more than 700 specimens are mapped within a section of a bed measuring approximately 25 m² in area and 35 cm in thickness. A few years ago, a new Konzentrat– and Konservat–Lagerstätte (see upper * in Fig. 3d; Fig. 11b) was discovered, most likely corresponding to a fish nursery or an effective juvenile habitat (Béchard and Cloutier 2011). Over 1000 larval and juvenile specimens belonging to at least five species were found (Fig. 11b–d); numerous taphonomic, paleoecological, morphological and developmental aspects of this occurrence are currently under investigation at the Université du Québec à Rimouski.

Recognition of the Miguasha fossil locality as an exemplar of the Devon-ian ‘Age of Fishes’ and its designation as a UNESCO World Heritage Site (Cloutier and Lelièvre 1998) is based on: (1) its faunal representativity of major groups of sarcopterygians, (2) the representativity of vertebrate evolutionary events, (3) the floristic and faunal representativity of aquatic and continental assemblages, (4) the paleo-biological representativity (e.g. presence of ingested prey, presence of fossilized ontogenies), (5) the quality of preservation in terms of anatomical completeness, (6) the quality of preservation in terms of exceptional fossilization, and (7) the abundance of specimens. The Devonian Miguasha biota stands as a primary Fossil–Lagerstätte, a true time capsule in the early history of vertebrates.