La dalle qui porte les empreintes est orientée NE–SW, avec un pendage de 80 à 85° vers le sud. Nous décrivons ici la dalle principale ; une seconde dalle, plus petite, située à une trentaine de mètres au sud-ouest, porte également de nombreuses empreintes, sans qu'aucune piste soit reconnaissable. Plus de 170 empreintes de pieds et de mains ont été relevées sur la dalle principale, longue de 28 m et haute de 5 à 6 m (Fig. 2). Les pistes mesurent de 3 à 9 m de long. La plupart des traces sont des hyporeliefs concaves, avec des bourrelets. Les pistes les mieux préservées (T1, T2) présentent des traces de pied ovales et de petites empreintes de mains en forme de croissant. Nous n'avons décelé aucune trace de griffe. Toutes les voies sont étroites, avec des traces de pied proches de l'axe central des voies. Ces pistes sont rapportées à l'ichnogenre Parabrontopodus, généralement attribué à des sauropodes primitifs ou à des diplodocoïdes. Dans l'hypothèse où ces dinosaures auraient les mêmes proportions qu'un Diplodocus, leur taille est estimée entre 27 et 32 m, pour une hauteur au bassin variant de 338 à 408 cm.

The first sauropod footprints discovered in Europe were described by Kaever and Lapparent in 1974 [13] in the Kimmeridgian of Lower Saxony (Germany; see Lockley and Meyer, 2000 [16] for a reinterpretation of this tracksite); sauropod footprints were described later from Spain [11], [16] and [20]. Since 1990 new discoveries were reported from the Late Jurassic of Portugal (see Lockley and Meyer [16] for a synthesis), Southern France [14], Spain [1] and [12] and Switzerland [19], [21] and [22]. In 2003, Mezga et al. [23] reported narrow-gauge trackways from the Upper Tithonian of Croatia. Most of these trackways have been referred by Lockley and Meyer [16] to the ichnogenera Brontopodus (wide-gauge sauropods) and Parabrontopodus (narrow-gauge sauropods); other ichnotaxa have been used by former authors (such as Elephantopoides Kaever and Lapparent, 1974 or Gigantosauropus Mensink and Mertmann, 1984), but were shown by Lockley et al. (1994) [17] to be nomina dubia. It is thus convenient when dealing with Late Jurassic sauropod footprints to use the widely employed terms Brontopodus Farlow et al., 1989 and Parabrontopodus Lockley et al., 1994. In their 2000 review of European dinosaur footprints, Lockley and Meyer [10] and [16] noticed that Kimmeridgian sauropod footprints are abundant (Portugal, France, Switzerland, Germany) and show a mixture of wide-gauge (Brontopodus type) and narrow-gauge (Parabrontopodus type) animals, while in the Tithonian there is a dominance of wide-gauge types (although, as noted above, narrow-gauge trackways are now reported from the Tithonian of Croatia and France).

The tracksites are located near the village of Coisia, in the southern part of the Jura department, 10 km northwest of Oyonnax (Fig. 1). The two exposures occur along road CD 60 and they were discovered by one of us (C. Gourrat) in April 2004; both exposures are subvertical bedding planes showing sauropod footprints impressed in a limestone referred to the ‘Couches du Chailley’ Formation by geologists. The ‘Couches du Chailley’ are dated from the Uppermost Kimmeridgian to the Tithonian by ammonites [18]. The track-bearing levels are situated well above the Late Kimmeridgian G. irius levels, and can safely be referred to the Tithonian. The Couches du Chailley are bioturbated by Thalassinoides burrows and have yielded, besides ammonites, gastropods (Nerinea, Pterocera), bivalves (Pholadomya, Trigonia), algae and foraminifers. They represent a subtidal environment such as a lagoon separated from the open sea by a coral reef. Laminites of algal origin and now dinosaur trackways indicate periods of emersion (supratidal environment). It is interesting to note that in 1984 Bernier [2] had reported circular depressions in the same levels, one kilometre to the southwest of the cliff we describe here; he briefly suggested a possible footprint site. This outcrop is now badly eroded but undoubtedly shows sauropod footprints.

We mapped the site on 31 July and 1 August 2004 after the removal of overlying beds on the southwestern part of the slab. The track-bearing slab is orientated NE–SW (N 35°) with a dip of 80 to 85° southward. We describe here the main slab; a second slab, about 30 m to the southwest, also shows sauropod footprints, but no trackways can be recognized. More than 170 pes and manus prints are exposed on the main slab, which is 28-m long and 5- to 6-m high (Fig. 2); there are at least nine different sauropod trackways (Fig. 2) between 3- and 9-m long. Most of the tracks are preserved as imprints (concave hyporeliefs) with sediment displacement rims. Some, on the northeastern part of the slab, are more likely underprints. The best preserved trackways (Fig. 3, Table 1) show a large oval pes print and a small crescentic manus print just in front of the pes print. There are no claw impressions.

Mean pes print lengths range from 57 (T5) to 69 cm (T3); mean pes widths range from 42 (T9) to 56 cm (T3). Mean manus print lengths range from 25 (T2) to 29 cm (T1); mean manus widths range from 36 (T9) to 52 cm (T1). All measurements were taken inside the rims (Table 1). The pes are thus longer than wide and the manus are wider than long.

Pace length ranges from 101 cm (T3) to 122 cm (T1); stride length ranges from 162 (T7) to 193 cm (T2). Pace angulation (pes) varies from 82° (T5) to 126° (T2).

All the trackways are narrow-gauge with footprints situated close to the central axis of the trackways. The heteropody (manus-pes size ratio) is high for trackways 1 and 2 (about 1/6). The manus prints are outwardly rotated.

Three trackways (T1, T2, T7) are oriented between the southwestern top of the slab to the northeastern bottom; one is oriented in the opposite direction (T3). Three are oriented almost vertically from the bottom to the top of the slab (T4, T5, T6). Another (T8) is oriented from the southwestern bottom to the northeastern top. Many footprints and manus prints could not be resolved into trackways, even though some alignment of footprints suggests partial trackways.

The narrow-gauge trackways are attributed to the ichnogenus Parabrontopodus Lockley et al., 1994. Parabrontopodus footprints are usually referred to basal sauropods or diplodocoids (which were narrow-hipped sauropods), while the wide gauge Brontopodus are commonly referred to brachiosaurids or titanosaurids. The dimensions of the track makers were estimated using Thulborn's [24] formula: hip height (= 5.9 × pes length) can be estimated between 338 and 408 cm; given that a 24-m-long Diplodocus is about 300-cm high at the hip, the total length of the Coisia track makers can be roughly estimated between 27 and 32 m.

Skeletal remains of sauropods are not abundant in the Late Jurassic of the French Jura. A notable exception is the partial brachiosaurid skeleton from the Oxfordian of Damparis (Jura). This skeleton was found in association with isolated theropod teeth, indicating scavenging in situ [5] and [7]. Both the Coisia footprints and the Damparis skeleton point to temporary emersion of the Late Jurassic carbonate platform in what is now the French Jura. The French Tithonian sauropod record is scanty, with a few camarasaurid teeth known as Neosodon praecursor in the Boulonnais [8], and a few camarasaurid caudal vertebrae from Haute-Marne [6]. A large caudal vertebra from Cherves, near Cognac, was referred by Le Loeuff et al. [15] to a camarasaurid, but this locality is now considered as Berriasian in age [9]. Outside France, the Late Jurassic European sauropod record comprises diplodocoids [4], brachiosaurids, and camarasaurids [25].

The sauropod footprints from Coisia are referred to the ichnogenus Parabrontopodus, attributed to basal sauropods or diplodocoids. Other footprints are known in the same level about 1 km to the southwest of the site here reported: further investigations in the French Jura might reveal a potential megatracksite comparable to the Swiss megatracksite described by Meyer [22] in the northern Jura Mountains. From a palaeogeographical point of view, the discovery of the footprints confirms that periods of emersion affected these subtidal environments [3], [5] and [18]. The Coisia tracksite represents the most important sauropod tracksite in France with a good potential for future studies. The removal of overlying beds could reveal several thousand square metres of the ‘dinoturbated’ surface.