The investigation of the Launton, Warbreccan, and Cronstad meteorites was undertaken with the object of placing on record descriptions of these stones. Launton and Warbreccan were found to conform very well with the type of Baroti and Wittekrantz, containing from 7-9 per cent. of nickel-iron, described in a previous paper. Cronstad, however, which contained as much as 18 per cent. of nickel-iron, showed so much variation from that type as to suggest a passage towards the rare Hvittis class of chrondritic stone containing even larger amounts of nickel-iron.

In a previous paper attention was drawn to the close similarity in mineral and chemical composition of chondritic meteoric stones. In the case of about forty chondrites it was shown that, apart from variations in the amount of nickel-iron, not only are the constituent minerals present in very similar amounts, but their chemical compositions vary to no very great extent. In that paper sufficient importance was not attached to the variations in the amount of nickel-iron.

There are two distinct kinds of image which are the subject of optical investigation: object-images, in which the form of the object is reproduced, and directions-images, in which every point represents a direction followed by light before reaching the objective. An instrument that shows a directions-image may be described as a hodoscope. It is focused not on the object but on infinity, so that parallel rays are collected at the same point in the image.

The determination of the direction in which the light illuminating a point in a directions-image (the so-called image in convergent light) was travelling when it reached the objective has hitherto been carried out by first measuring the position of the point in the image and thence calculating the angular direction it represents. It is usually assumed that the distance of a point in the image from the centre of the field is in proportion to the sine of the angle between the path of the light that illuminates the point and the axis of the microscope.

Last year, while working through the geographical collection of rocks in the Geological Department of the Imperial College, I found an interesting series of volcanic rocks from the neighbourhood of the Lucalla River in the district of Cambambe, Angola, which hitherto has not been described. The specimens were collected about the year 1860 by J. J. Monteiro, a former student of the Royal School of Mines, who devoted many years to the exploration of Angola. The collection, though consisting of only ten specimens z comprises porphyritic basalts, alkalitrachytes, a nephelinite allied to monchiquite, and an augite-andesite.

The mineral described in this paper occurs at tile Hudson Bay zinc mine about five miles east of Saline, near Nelson in the West Kootenay district of British Columbia, I regret that I have no data as to the exact character of the ore-body except that it occurs in a region where palaeozoic rocks are intruded by biotite-granite and that it is probably genetically connected with the intrusion of the latter. The zinc occurs ill the form of the silicate and tb a less extent the carbonate. There is also a small amount of oxidized lead-ore, principally cerussite.

The specimen here described was acquired in 1913 for the mineral collection of the British Museum. It came from one of the numerous sulphur mines in the province of Girgenti, Sicily. Being of some size (15 cm. = 6 inches across) and with the two individuals well separated and symmetrically disposed, it affords a striking illustration of a twin growth. The crystal is colourless and perfectly transparent; it encloses one or two small patches of native sulphur and compact limestone. A portion, probably about one-third, of the original specimen has been broken away along the perfect cleavage of gypsum ; and this fact, together with the amall point of attachment, increases the resemblance of the twin to tbe butterfly form, which is here more striking than in the so-called butterfly-twins of calcite.