these of irregular outline is crossed by the Rhine at Andernach. Its western and more important part is about three miles long and two broad. The Laacher See is well known as a lake occupying an old crater in this part of the country, not far from Andernach. This district was described and illustrated by Dr. S. Hibbert, in 1853, in his 'History of the Extinct Volcanos of the Basin of Neuwied,' etc. The other district, characterized by volcanic rocks and craters dispersed over an area of about four miles by three, is at a short distance to the south-west, and contains several large lake-craters, such as the Gemunder Maar, the Pulver Maar, the Meerfelder Maar, etc. One of the old lava-streams in this area is met with at Bertrich, on the Ees, a small river running into the Moselle half-way between Treves and Coblentz. Consisting of columnar basalt, and being perforated by a natural aperture, this mass of volcanic rock presents the aspect of a basaltic colonnade, and has always attracted the attention of travellers, especially as the joints of the basalt, instead of taking a regular polygonal or angular shape, are more or less spheroidal, "so that a pillar is made up of a pile of balls, usually flattened;"[1] hence the grotto at Bertrich is called the Käsegrotte, or Cheese-cave. "The basalt is part of a lava-stream, from thirty to forty feet thick, which has proceeded from one of several volcanic craters still extant on the neighbouring hills;"[2] and, having run in the valley, it has since been partially destroyed and excavated by long-continued water-action. Mr. J. E. Lee has favoured us with a pencil sketch of this interesting Cheese-grotto, from which Plate VIII. has been engraved; and, although the grotto is well known to geological students by the woodcut in Sir C. Lyell's 'Manual,' p. 386, yet we think that as the subglobular character of the basalt is very much better shown by our correspondent's sketch than in the little woodcut alluded to, we shall be doing good service by producing it here.
In connection with the Cheese-grotto, Sir Charles alludes to the occurrence and characters of globular lavas and trap-rocks, adducing particularly the globiform pitchstone of Chiaja di Luna, described and figured by Mr. Poulett Scrope, in his account of the Ponza Isles (Geol. Trans., 2nd ser. vol. ii.). This pitchstone has the globiform structure near its junction with prismatic trachyte; and itself shows a tendency to the columnar division; the columns, however, separating into large globes or ellipsoids, placed one upon another, and, when weathered, readily desquamating at a touch into numerous concentric coats, having a kind of onion-peel structure. Different degrees of the prismatic or columnar condition, passing into the concentric and nodular, are observable in many basaltic and trachytic lavas, as well as in older trap-rocks (diorites, etc.); and indeed granite not unfrequently shows a tendency to split and exfoliate in a similar manner.
The explanation of the columnar and nuclear structure is well given on Mr. P. Scrope's 'Considerations on Volcanos,' etc., 1825 (an admirable work, now out of print, but about to be revived, we hope). In chapter 6, p. 134, etc., the divisionary structure assumed
