good instances of ridges in the marvellous ramifications round Encke, Kepler, and Marius, and in the region north of Aristarchus and Herodotus.  Perhaps the most perfect examples of surface swellings are those in the Mare Tranquilitatis, a little east of the ring-plain Arago, where there are two nearly equal circular mounds, at least ten miles in diameter, resembling tumuli seen from above.  Similar, but more irregular, objects of a like kind are very plentiful in many other quarters.

It is a suggestive peculiarity of many of the lunar ridges, both on the Maria and elsewhere, that they are very generally found in association with craters of every size.  Illustrations of this fact occur almost everywhere.  Frequently small craters are found on the summits of these elevations, but more often on their flanks and near their base.  Where a ridge suddenly changes its direction, a crater of some prominence generally marks the point, often forming a node, or crossing-place of other ridges, which thus appear to radiate from it as a centre.  Sometimes they intrude within the smaller ring-mountains, passing through gaps in their walls as, for example, in the cases of Madler, Lassell, &c.  Various hypotheses have been advanced to account for them.  The late Professor Phillips, the geologist, who devoted much attention to the telescopic examination of the physical features of the moon, compared the lunar ridges to long, low, undulating mounds, of somewhat doubtful origin, called “kames” in Scotland, and “eskers” in Ireland, where on the low central plain they are commonly found in the form of extended banks (mainly of gravel), with more or less steep sides, rising to heights of from 20 to 70 feet.  They are sometimes only a few yards wide at the top, while in other places they spread out into large humps, having circular or oval cavities on their summits, 50 or 60 yards across, and as much as 40 feet deep.  Like the lunar ridges, they throw out branches and exhibit many breaches of continuity.  By some geologists they are supposed to represent old submarine banks formed by tidal currents, like harbour bars, and by others to be glacial deposits; in either case, to be either directly or indirectly due to alluvial action.  Their outward resemblance to some of the ridges on the moon is unquestionable; and if we could believe that the Maria, as we now see them, are dried-up sea-beds, it might be concluded that these ridges had a similar origin; but their close connection with centres of volcanic disturbance, and the numbers of little craters on or near their track, point to the supposition that they consist rather of material exuded from long-extending fissures in the crust of the “seas,” and in other surfaces where they are superimposed.  This conjecture is rendered still more probable by the fact that we sometimes find the direction of clefts (which are undoubted surface cracks) prolonged in the form of long narrow ridges or of rows of little hillocks.  We are,