is general, the optical character is negative, and the elongation positive, whilst the optic axial angle is small. Most important of all, however, the extinction, whilst often nearly straight, reaches generally as high as 5° with the composition-planes of the lamellae, and in one or two crystals angles of over 15° are got with the cleavage. Were it that only one or two crystals showed the obliquity of extinction one might suspect some inaccuracy of observation, but at least eight in the slide show it. Further, even greater obliquity of extinction, with similar very finely microscopic twinning, is shown by nearly a dozen crystals of apparently identical hypersthene-like pyroxene in a noritic rock discovered at Milford, and described later in this paper (see p. 420). In some instances the lamellae are here much broader than in others, whilst the extinction angles generally range up to 15°, but in one case up to as much as 42°. As a rule, the lamellae extinguish together in the one direction, but this is by no means invariably the case. In the Milford, as in the Cleddau-Hollyford rock, there is parallel intergrowth with fine lamellae of augite. It may be thought that the crystal in which the extinction angles of the lamellae reached 42£ exemplifies a similar parallel intergrowth with augite, but this is not so, for the lamellae are very distinct and comprise the whole mineral, which certainly is not an ordinary monoclinic pyroxene, for it is optically negative and has the distinct hypersthene-like pleochroism already noted. There is, however, one peculiarity of this crystal that may have a bearing upon the matter—namely, the optic axial angle appears to be large. Only one further observational fact now remains to be mentioned in connection with this pyroxene, but it is of importance: the plane of the optic axes coincides with the composition-plane of the lamellar twin. Since the extinction is oblique, this plane must be at right angles to the plane of symmetry—that is, to the 100 plane—as in normal hypersthene. The lamellar twin appears not to follow the plane 110 frequent in the inter-growths of rhombic and monoclinic pyroxenes, for all the sections showing the lamination distinctly are approximately at right angles to the optic axial plane; nor is it the common 010 plane.* See H. Rosenbusch, Mikroshopische Physiographie der Mineralien und Gesteine, vol. 1, pt. 2, p. 149, 1905. It is much more reasonably the 100 plane observed by Bowen in artificial clino-enstatite.† N. L. Bowen, The Ternary System Diopside-Forsterite-Silica, Am. Journ. Sci. (4), vol. 38, pp. 207–64, 1914. On this supposition, however, it is difficult to understand why the lamellae often appear all to extinguish in the same direction. Enough data are available to show that a monoclinic pyroxene of unusual type is exhibited in the sections from the Cleddau-Hollyford and the Milford rocks. It approaches hypersthene, but differs from it in its oblique extinction, and it seems probable that it is a monoclinic member of a series embracing this latter. It is obvious that much of the mineral in the sections that is apparently hypersthene may really be this mono-clinic pyroxene viewed in sections lying near the orthopinacoid. Chemical Classification. An analysis of the Cleddau-Hollyford quartz-norite by the Dominion Analyst gave the results tabulated in column A below. In columns B and C the analyses of somewhat similar rocks are quoted from Iddings.‡ J. P. Iddings, Igneous Rocks, vol. 2, p. 218, 1913.