part are either wholly unfossiliferous, or they contain the remains of plants or the bones of reptiles, such as may easily have been drifted from some neighbouring shore.  The few fossils which may be considered as properly belonging to these deposits are chiefly Crustaceans (Estheria) or Fishes, which may well have lived in the waters of estuaries or vast inland seas.  We may therefore conclude, with considerable probability, that the barren sandy and marly accumulations of the Bunter Sandstein and Lower Keuper were not laid down in an open sea, but are probably brackish-water deposits, formed in estuaries or land-locked bodies of salt water.  This at any rate would appear to be the case as regards these members of the series as developed in Britain and in their typical areas on the continent of Europe; and the origin of most of the North American Trias would appear to be much the same.  Whether this view be correct or not, it is certain that the beds in question were laid down in shallow water, and in the immediate vicinity of land, as shown by the numerous drifted plants which they contain and the common occurrence in them of the footprints of air-breathing animals (Birds, Reptiles, and Amphibians).  On the other hand, the middle and highest members of the Trias are largely calcareous, and are replete with the remains of undoubted marine animals.  There cannot, therefore, be the smallest doubt but that the Muschelkalk and the Rhaetic or Koessen beds were slowly accumulated in an open sea, of at least a moderate depth; and they have preserved for us a very considerable selection from the marine fauna of the Triassic period.

[Illustration:  Fig. 140.—­Zamia spiralis, a living Cycad.  Australia.]

The plants of the Trias are, on the whole, as distinctively Mesozoic in their aspect as those of the Permian are Palaeozoic.  In spite, therefore, of the great difficulty which is experienced in effecting a satisfactory stratigraphical separation between the Permian and the Trias, we have in this fact a proof that the two formations were divided by an interval of time sufficient to allow of enormous changes in the terrestrial vegetation of the world.  The Lepidodendroids, Asterophyllites, and Annularioe, of the Coal and Permian formations, have now apparently wholly disappeared:  and the Triassic flora consists mainly of Ferns, Cycads, and Conifers, of which only the two last need special notice.  The Cycads (fig. 140) are true exogenous plants, which in general form and habit of growth present considerable resemblance to young Palms, but which in reality are most nearly related to the Pines and Firs (Coniferoe).  The trunk is unbranched, often much shortened, and bears a crown of feathery pinnate fronds.  The leaves are usually “circinate”—­they unroll in expanding, like the fronds of ferns.  The seeds are not protected by a seed-vessel, but are borne upon the edge of altered leaves, or are