Of late years a great deal has been written on the genesis of ore-deposits, and in particular much space has been devoted to more or less controversial discussions of the possibility of a magmatic origin for many important types, both in individual instances and by way of generalizations. The widespread acceptance in many quarters of a belief in the derivation of such deposits on a wholesale scale from igneous magmas may perhaps to some extent be regarded as a natural reaction from some of the theories of markedly contrasted character that prevailed almost universally in past times. The “swing of the pendulum” is such a well-known psychological phenomenon that it is unnecessary to enlarge on it here. It is applicable to nearly all branches of knowledge, and of ignorance, and was long ago summarized in the law that action and reaction are equal and opposite. If a theory is pushed too far, it is inevitable that in a while it will be regarded as wholly, or for the most part, erroneous, although it may nevertheless contain a good deal of truth. Most of these difficulties and controversies arise from the ineradicable habit of generalizing too broadly, and the rest of them can generally be traced to misunderstandings as to the precise meaning of the terms employed. In the history of the scientific study of ore deposits both of these sources of error have undoubtedly played an important part.

F. P. Paul described in 1906 (T.M.P.M., vol. 25, p. 309) a nepheline-eudialite-basalt from Shannon Tier, Tasmania, containing an unknown mineral referred by him to Ca2SiO4. Subsequently Bowen (Amer. J. Sci., 1922, p. 30) pointed out the close correspondence between the optical properties of Paul's mineral and of the artificial βCa2SiO4 prepared by Day, Shepherd and Wright. Still later (Geol. Mag., 1927, p. 43) the writer of this note ventured to apply the name shannonite to the mineral of this singular occurrence. Since that time through the kindness of Mr. P. B. Nye, Government Geologist of Tasmania, a specimen of the rock described by Paul from Shannon Tier has been obtained and examined. That the specimen examined is identical with the rock described by Paul is clearly indicated by his detailed description and the photomicrograph which accompanies his paper. Unfortunately my own determinations are not in agreement with those of Paul, and the purpose of the present note isto state the nature of these discrepancies. The rock is a felspar-free nepheline-basalt of which the chief constituents are idiomorphic nepheline, a titaniferous augite, olivine (chrysolite), the unknown mineral of Paul, biotite, apatite, magnetite, perofskite, and minor amounts of interstitial analcime. The described eudialite has not been discovered, and from the description attached to the photomicrograph this mineral appears to have been confused with nepheline. I find further that the properties of the unknown mineral agree, not with those of any of the artificial calcium orthosilicates, but with those of monticellite. The two olivines are readily distinguished in thin section, the chrysolite occurring as chadacrysts and the monticellite as oikocrysts, the latter appearing as shapeless grains often surrounding chrysolite and enclosing numerous idiomorphic crystals of nepheline and magnetite, and thus belonging to a late stage in the crystallization history of the rock. The chrysolite moreover is readily distinguished—apart from these textural differences—by its higher birefringence and larger optic axial angle (near 90°).

Of the more recent accounts of the Middle and Upper Chalk of Cambridge, only that of Jukes-Browne employs the definitions of zonal boundaries now usually accepted in England; for the most part, the horizons to which he refers the exposures are correct.

The Faeröe Islands, 180 miles north-west of the Shetlands, are a part of that far-flung region known geologically as the Brito-Arctic or Thulean province of Tertiary volcanic rocks, here represented mainly by basalt lavas. Various investigators have visited them, and written about them from time to time, but as yet, no complete, detailed geological description of them has appeared. My acquaintance with them is limited to the islands of Stromöe and Suderöe, and my visit was of too short duration to allow much detailed field-work being done. These brief notes and analyses are presented now in the hope that they may be of use to other geologists.

The island of Dominica is situated in lat. 15° 20′ N., and long. 62° 30′ W., and lies almost midway between the French islands of Guadeloupe (on the north) and Martinique (on the south). In the northern channel the Antillean plateau is submerged to a depth of about 2,100 feet, but in the southern a deep embayment entering from the west reaches a maximum depth of 6,600 feet. Some miles to the south-east of Dominica, however, the sea floor rises to within 270 feet of the surface, while the Madiana banks, east of Martinique and north of Barbados, are submerged only to about the dame depth. Lying as they do between the sedimentary parts of Antigua, Guadeloupe, and Barbados it would be interesting to ascertain, if possible, by boring, whether these banks are of a sedimentary nature in contrast to the volcanic nature of the main range of islands. On both the windward and leeward sides Dominica rises precipitously out of the sea, but this is more accnetuated on the leeward side where within six or seven miles of the shore the sea bottom reaches a depth of over a thousand fathoms. The preciepitous nature of the sea floor and the absence, of any sort of harbour at Roseau—the capital—render anchorage even at favourable times a very precarious matter, and one of great difficulty and danger in rough weather.

IN the course of an investigation into the distribution of millerite by Dr. F. J. North and the writer, the presence of linnaeite (sulphide of cobalt Co3S4) in the clay-ironstone nodules of the Coal-measures of East Glamorgan was confirmed. That sulphide of cobalt should accompany sulphide of nickel is not surprising, but the only record of its occurrence in Wales so far as the writer is aware, takes the form of a note (without title) by A. L. des Cloizeaux in a French publication. This record was based on a specimen from the “Rhonda” (Rhondda) discovered by a Mr. Terrill, of Swansea, which does not seem to have been preserved. It is possible, too, that the mineral may occur in the Coal-measures of Belgium, as nickel and cobalt were detected in the soot from flues under which coal from Beyne (near Liége, Belgium) had been burned.

The mineral custerite [Ca(F,OH)]2SiO3 was first described from a limestone contact rock from Custer, Co. Idaho, the associated minerals being diopside, garnet, and magnetite. There appears to be no further record of this mineral from contact zones, though the related cuspidine [Ca(F,OH)]2SiO3 is known from Vesuvius and Franklin Furnace. In examining some metamorphosed limestones from the well-known mineral locality of Crestmore, California, the writer has found that custerite occurs in a vesuvianite rock in some abundance. As the mineral associations of this custerite are different from that of the original custerite, and owing to the very rare occurrence of this mineral, it is deemed of sufficient interest to describe this association briefly. The rock consists largely of idioblastic vesuvianite, these crystals being cemented together by a light brown base which proves to consist almost entirely of custerite, monticellite, and calcite. The custerite forms comparatively large shapeless areas enclosing subidioblastic crystals of monticellite. Under the microscope the custerite is seen to be colourless with a distinct cleavage, parallel to which lamellar twinning is developed. Sections cut perpendicular to the twin lamellae show symmetrical extinction of 6°.