Dr. T. O. Bosworth has described some effects of wind and sun in producing topographical forms and in modifying the shapes of stones on the surface of the deserts of Tumbez and Paita. This paper describes certain forms of rocks and stone erosion, not mentioned in his work, from an area further north than that in which his more detailed observations were made.

The geological structure of the British Virgin Islands correlates them indisputably with the Greater and not with the Lesser Antilles. The latter are composed essentially of Tertiary volcanic rocks of andesitic or basaltic type, with or without development of sedimentary strata which, when present, only dip at gentle angles and never show the violent effects of such dynamic forces as have been responsible for the folding and “up-ending” of the strata in the Virgin Islands. The tremendous depth of the channel separating the British Virgin Islands from the Lesser Antilles lias been ascribed to faulting, probably initiated in Pliocene times.

The work of Cleve (1), Hill (4), Vaughan (5), and others (6) in the American Virgin Islands, Porto Rico, the Dominican Republic, Cuba, Haiti, and Jamaica, indicates that the British Virgin Islands form geologically only the eastern termination of that main group of islands and have been subjected to the same earth movements as them. The evidence for attributing a Cretaceous age to the sedimentary series has already been referred to.

With regard to the age of the folding and igneous intrusion, Vaughan considers that the folding took place between upper Eocene and middle Oligoccne times, and that the intrusion of the diorites took place at approximately the same date. (7)

Wythe Cooke considers that the igneous basal complex in the Dominican Republic certainly dates from Cretaceous time, but that part is probably older. He also considers that the stresses that folded and sheared these rocks were probably active during Eocene time or earlier, and that the intrusion of the great masses of dioritic rocks probably occurred before the deposition of the Eocene sediments.

According to Hill (4), however, “in mid-Tertiary times granitoid intrusions were pushed upward into the sediments of the Greater Antilles, the Caribbean, Costa Eican, and Panamic regions.” Frazer, on the other hand (8), considered the nuclear axis of Cuba and San Domingo, and possibly of all the Caribbean islands, to be Archaean, a view upheld also by Dr. W. Bergt. (9)

There is no doubt that the key to these problems lies in the larger islands of the Greater Antilles, for it is only there that unaltered fossiliferous sediments occur and can be studied in relation to the igneous intrusion and metamorphism. There appears, however, to the writer to be nothing either in the geological or faunal evidence necessarily indicating previous land connexion at any time between the Virgin Islands and the Lesser Antilles, ahd on this matter it is hoped to furnish further evidence at a later date.

My attention has recently been called to some very interesting comments on the problem of the delimitation of the Silurian and Devonian systems, which were made by Dr. Elles in her Presidential address to Section C of the British Association at Liverpool. I am so heartily in agreement with all the principles of the modern classification of strata, as laid down by Dr. Elles, that it seems of advantage to recall very briefly how closely these principles are adhered to in using the Ludlow Bone-bed as the local base of the Devonian in England.

During recent years the nature of species has been frequently discussed by biologists, or, at least, by students of recent animals and plants, but much less attention has been paid to the subject by palaeontologists. Nevertheless, in palaeontology, as in neontology, as animals and plants are subjected to a more intensive study it is becoming increasingly difficult to make the concept of species fit the facts of nature.

The constitution and origin of anthracite has long been a question of absorbing interest, but until very recently the theories which were brought forward dealing with this subject, owing largely to the insurmountable difficulty of preparing thin slices upon which micropetrological work could be done, have lacked substantial evidence either to support or refute them. Lately, however, a new method which will materially widen our knowledge and advance research on the anthracite problem has been evolved.

A River flood is produced whenever the supply of water is in excess of what can be accommodated by the bed alone. Such a condition can be produced either by a sufficiently large obstruction in the bed of the river, or more frequently by an increase in the volume of water coming down the valley, or a combination of these two causes.