It is difficult for anyone standing on the brink of the Chasm, after having seen the placid flow of the Zambesi above the Falls, to believe that the fissure into which the river is so suddenly precipitated has been formed gradually by the action of the river itself, and not by some great convulsion during which the very crust of the earth was rent. The narrowness of the abyss, the strange zigzags along which the tumultuous waters rush after their first great plunge, the mystery which has long surrounded the further course of the river after it swings away out of sight among its forbidding precipices, and the knowledge that the rocks across which it plunges are of volcanic origin, are all factors that have aided the illusion. Hence it is not surprising to find that the explanation given by David Livingstone half a century ago, that the majestic Zambesi has here been intercepted by a rent due to some earth movement in the solid rocks, has been adopted without question in all the later descriptions of this wonderful spectacle.

Among the geological specimens obtained by Mr. H. T. Ferrar, while the “Discovery” was visiting New Zealand, are seven rock-fragments containing tubular shells, and an eighth with some obscure markings. They are labelled as follows:—

There are several kinds of acid volcanic rocks in Wales. In the X neighbourhood of Snowdon there are at least five kinds, as follows:

1. There are undoubted lavas, showing lines of viscous flow and sometimes vesicular, and weathering into cubical blocks. Such may be seen on Crib Goch and on Crib Yddysgl, on Cerrig Cochion, and elsewhere. These rocks are rarely cleaved.

2. There are rocks like the last as to fracture and mode of weathering, but which are neither viscous nor vesicular. They are probably masses of felsitic dust or mud. Such may be seen in Cwm Llan.

3. There are also massive felstones, rudely cleaved, such as form Moel Meirch. and Clogwyn Llwyd.

4. There are the so-called ashes of various kinds, usually wellbedded and often highly cleaved.

5. Lastly, there are highly cleaved felsitic rocks, which show no lines of viscous flow, are generally unbedded, and which are in many places fragmentary.

In the first part of this series, p. 313, I referred to the recent invasion of the Eastern Baltic by the Mya arenaria, as first pointed out by Dr. Nathorst. Dr. Petersen has called my attention to an important paper by A. S. Jensen, which has apparently been overlooked by English conchologists, and I propose to condense his results, as they are very noteworthy from their geological lessons.

The shell has hitherto been treated as a typical Arctic shell. Thus Crosse and Debeaux write of its original home: “De l'océan Glacial arctique, qui parait être sa véritable patrie, elle est descendue dans les mers du nord de l'Europe, jusques et y compris la Manche etune partie de nos côtes de l'Océan” (Journ. de Conchologie, ser. III, iii, 254, 1863). Gwyn Jeffreys, in describing the shell, says, “The occurrence of this circumpolar shell-fish so near the tropic of Cancer probably indicates the most southern limit in space of the glacial epoch” (Brit, Conch., iii, 65, 66).

Detailed examination of the remains of fossil mammals from the fluvio-marine beds (Upper Eocene) of the Fayûm has, in several cases, shown that the number of species of the various genera has been underestimated. This is especially the case with Palœmastodon, of which originally only one species, P. beadnelli, was recognised. Subsequently another form, P. minus, was named, and now it has become necessary to establish at least two other species, to one of which probably by far the greater number of the specimens collected must be referred.