very common.

Sternberg (1932) described only a single specimen of Amblydactylus, and did not refer to any trackways. This leaves the reader with the impression that it must be a rare genus. However, 50 /o of the trackways discovered and 90 /o of the isolated footprints observed are attributable to Amblydactylus.

Amblydactylus footprints of the Peace River Canyon were preserved in several types of depositional environments. At Site 1 (Currie 1980) the tracks are found in a grey, fine-grained sandstone. Rootlets and biological turbulence are evident in overlaying layers, and the variety of footprint types and sizes suggest that the dinosaurs were walking on the muddy margin of a quiet body of fresh water. The major footprint bearing level at Site 3 is a grey, ripple-marked sandstone with rootlets and bioturbation. There are many footprints of small animals, and it seems probable that the site was a soft, flat expanse of mud at the edge of a body of quiet water, or was possibly even covered with a couple of centimetres of. water. At Site 4, a trackway of Amblydactylus is found in a ripple-marked, ferruginous sandstone. The pattern of the tracks at this site suggests that the animal was partially floating when the trackway was made, and that the tracks were made in the muddy bottom of a quiet body of water that was at least two metres in depth. Numerous other sites in, the canyon appear to have been underwater when the footprints were made. A large block was found on the talus slope near Site 5 with three large negative footprints of a single Amblydactylus. On the other side of the slab were worm burrows of a type known as Rhizocoralia, indicating that shortly after the dinosaur walked over the mud, additional sediments accumulat- ed in a body of brackish water. The seashore must have been very close at that time. The hadrosaurs that left their footprints at the main level of Site 8 (fig. 2) were walking across a splay crevace. A single footprint was recovered from Site 10 from a coarse-grained, cross-bedded channel sandstone, showing that Amblydactylus was not restricted to quiet waters. Footprints at Site 15, including the type specimen of Amblydactylus gethingi, were made in the water covered mud on the edge of a sandbar. Another footprint recovered from the talus slope above Site 15 appears

HADROSAUR TRACKWAVS 67

to have been made in a very organic layer of mud. In summary, the hadrosaurs of the ancient Gething Delta appeared to live in proximity to almost all of the available depositional environments.

Amblydactylus trackways are well represented. In most cases, neither handprints nor tail drag marks are preserved with footprints of Ambly- dactylus. This is expected if the animal is floating or swimming in the water. Handprints are associated with the footprints in at least six of the Amblydactylus trackways, indicating that the animals were facultative and not obligatory bipeds. The juveniles do not show any signs of having used their hands in supporting their bodies even though they were found at Site 3 where there appears to have been little or no water to support them. The tail was used as a counterbalance as the animal walked, and was never dragged as it was in many of the dinosaurs of the Lower Jurassic of the Connecticut River valley (Hitchcock 1858).

The trackway at Site 4 appears to indicate that hadrosaurs were efficient swimmers. Here an animal was walking on the muddy bottom of a quiet body of water. As the water became deeper, its stride decreased, and it appears to have been pushing off the bottom with its toes because the mark for the heel pad is very shallow and poorly defined. At one point, the midline of the trackway shifts more than a metre to the right, and several steps later, it shifts to the left again. It would difficult to explain these shifts unless the three or four tonne weight of the body was buoyed up by water. Furthermore, there are other isolated but clearly defined tracks at Site 4 which suggest that animals were swimming over the muddy bottom, but were occasionaly putting a foot down to push off the bottom. There are many other sites in the canyon where clearly defined Amblydactylus footprints are not associated in trackways, even though individual tracks cover the entire bedding plane. The best explanation for the lack of continuous trackways is that the animals were in water deep enough to swim. Finally, it; should be pointed out that at many of the sites, only the footprints of large animals were found, suggesting that the water was too deep for small animals, or that the smaller dinosaurs were floating at the surface.

Juvenile Amblydactylus kortmeyeri footprints were found at Site 7 (Currie and Sarjeant 1979) and Site 3. In both cases, two animals of approximately the same size were moving in the same direction, suggesting