The present series of notes is mainly based upon analyses executed during the years 1867-1877. but these have been supplemented in a few instances by recent work. The series may be regarded as a continuation of several papers published from time to time in the Journal of the Chemical Society, in the Chemical News, and in the Proceedings of the Royal Irish Academy. Throughout these inquiries particular attention has been paid to the degree of tenacity with which the water present in the hydrous species is held.

While examining the fine series of corundums brought from Burma by Mr. C. Barrington Brown, F.G.S., and comparing them with specimens obtained from Ceylon, Southern India, the United States. and other localities, I have been led to devote especial attention to the curious " parting-planes " which are sometimes found in crystals of this mineral, but which, in other cases, appear to be entirely wanting. In making these comparisons, I have had the great advantage of being able to refer to the fine collection of corundums in the British Museum (Natural History) at South Kensington, and have to gratefully acknowledge the advice and assistance which have been so freely afforded to me at all times by Mr. Fletcher and Mr. Miers.

In recent years, the respective provinces of Mineralogy and Petrography have become much more clearly defined than was formerly the case. We no longer include uncrystallised substances, like obsidian, taehylyto and coals among mineral species; and, on the other hand, we recognise the fact that many materials originally analysed and treated as homogeneous—such as lapis-lazuli, emery, and many similar materials—are really mineral aggregates or rocks, and thus come within the domain of the petrographer. This last advantage has been one of the most important of those which have resulted from the application of the microscope to the study of inorganic bodies.

A specimen in the British Museum, consisting mainly of brown chalybite and decomposed iron pyrites--the latter mostly converted into earthy oxides--bears upon its surfitce and in its cavities a number of thin prisms encrusted with a botryoidal yellowish substance which covers a great part of the decomposed specimen. On close inspection the needles are seen to be Cronstedtite, mostly but not entirely converted into black limonite, and the encrusting substance consists mainly of phosphate (with a little carbonate) of lime. The latter is sometimes crystalline on its surface.

An orange-coloured ochreous substance from Parys Mount, Anglesey, was recently submitted to analysis.

The observed loss of water at 100°C corresponds to 4H2O, which requires 13.24 percent.

The above investigation shows:—

• 1. The nature of homogeneity of structure, and the properties which distinguish it from structureless homogeneity. The new definition of a homogeneous structure recently put forward by the author in Groth's Zeitschrift is given.

• 2. A method of realising in a concrete form, and with great generality, the kind of repetition in space which constitutes homogeneity of structure, the models employed for this purpose each consisting of a number of similar plaster hands appropriately arranged in space.

  The total number of types of arrangement, all of which can be represented in this way, is 230, this being the number of typical point systems described by Fedorow and Schönflies, derived by their extension of Sohncke's methods. The various types of homogeneous structure, like the corresponding point-systems, all fall into the 32 classes of crystal symmetry.

• 3. What property common to all homogeneous structures whatever most nearly coyresponds to Thomson and Tait's definition of homogeneity.

• 4. Reasons for regarding as untenable the arguments put forward by Fedorow in support of his recent attempt to select from among the types of homogeneous structure those which are possible for crystals, and to determine the shapes of their ultimate units.

• 5. The possibility of so classifying all conceivable ways of symmetrically partitioning all the types of homogeneous structure as to avoid all reference to the nature of the cell-faces, whether plane or otherwise, and, in other respects also, be perfectly general. Some reasons for undertaking this classification, notwithstanding its complexity, are given, the chief one being the relation of symmetrical partitioning to some stereo-chemical and other experimental facts.

The name of Monehiquite, first applied by Rosenbusch and Hunter (Tscherm. Min. Mitth. XI. 1890, p. 445) to a peculiar group of dyke rocks tound in association with elæolite syenite in Brazil, is rapidly becoming familial" to petrologists through the discovery of similar rocks from various other localities.

The Brazilian Monehiquites were described by Rosenbusch as black to grayish-black rocks, consisting of porphyritic ohvine and pyroxene, with widely varyiug amounts of hornblende and mica, in a glassy base usually crowded with hornblende microliths.

Having heard that a vein of what was supposed to be mispickel had been found about 6 miles north of Newton Stewart, in company with Professor Heddle I visited the place a week or two since. We found the vein had been opened in two places and some of the mineral had been taken out. It proved to be mispickel. The vein runs pretty nearly north and south: it occurs at tho junction of the granite and silurian rocks, commencing about half-a-mile south of the Murray monument, crossing the Palnure Burn, running nearly due south for more than a mile, and then heading to the eastward, where, at about three-quarters of a mile further on, another opening has been made in which is found ehalcopyrite associated with mispickel. The vein, where it has been opened close to the Palnure Burn, is about 7 inches wide, and contains only mispickel.

A Brief notice of this new mineral, which was found associated with Lewisite in the cinnabar-bearing sands of Tripuhy, near Ouro Preto, Minas Geraes, Brazil, has already been given by tile authors in Min. Mag. Vol. XI. No. 50, p. 85. Since that time a sufficient amount of material (1½ grms.) has been separated from the sand to permit a more detailed investigation of its crystallographic and chemical characters.

On November 14th, 1892, Prof. J. A. Krenner read before the Hungarian Academy of Sciences an account of a new silver ore from Felsöbánya, to which he gave the name of Andofite. It was described as orthorhombie with a : b : c = 0.97756 : 1 : 0.86995, and as having the chemical composition 2PbS.Ag2S.3Sb2S3.

The humite series has formed one of the most interesting problems to mineralogists ever since Scacchi established the three types. I may, perhaps, recall that he found the crystals to be divisible into three classes, all capable of representation by reference to a common parametral plane and three zone-axes at right angles to one another. But the three classes differ in the following respects:--(1) Each has a peculiar set of faces, and, with the exception of the pinakoids and the actual or theoretical prisms, few of the forms coincide in different types. (2) Whilst all the faces on crystals of one class can be easily represented by quite simple indices, such as (111) (123) (103) by a judicious selection of one of its faces as parametral plane, the faces on crystals of the remaining two classes give for the same parameters very high and complicated indices.

In the course of some researches on the relation between the chemical constitution and the crystalline form of substances, I found it necessary to attempt an accurate determination of the densities of crystals.

Messrs. Playfair and Joule, in 1843, in the course of similar researches, determined the specific gravities of numerous salts by the use of an instrument which showed the actual volume of the solid investigated. This method was found to be subject to many errors, and they abandoned it for one which is the prototype of that ordinarily used in accurate work at present.

In a previous paper (Min. Mag. XI. 1895, p. 80), the result was given of an analysis of the new mineral, Zirkelite, which was discovered by Dr. Hussak in association with baddeleyite in the decomposing magnetite-pyroxenite of Jaeupiranga, S. Paulo, Brazil.

In that analysis only a small amount of material (less than a quarter of a grin.) was used, and no very serious attempt was made to separate small percentages of rare earths in the main zirconia precipitate.

Argentina.--L. Brackebusch (Las Especies Min. de la Rep. Argentina), Anal. Soc. Cient. Argentina, 112, 1878; H. D. Hóskold, Mémoire général et spécial sur les mines, &c., duns la Rep. Argentina (Exposition de Paris, 1889), p. 126, analyses and maps; Catálogo oficial of the Argentine minerals at the Paris Exhibition, 58, 1889; Argentine Rep., Philadelphia Exhibition, 1876, pp. 197, 193; A. Sella, Nachr. d. k. Ges. d. Wiss. zu Göttingen, 317, 1891 (specific heat).

Augelite was one of several aluminium phosphates from the iron mine of Westanå, Skåne, Sweden, described by C. W. Blomstrancl in 1868 (Öfv. Ak. Förh. Stockh. XXV. 1868, p. 199). He states it to be a massive mineral with distinct cleavages in three directions, and names it augelite (avγη lustre) on account of the strong pearly lustre on the cleavage ṡurfaces. He determines its specific gravity as 2.77, and, as the result of his analyses, represents its chemical composition by the formula, 2Al2O3. P2O5. 3H2O. Since this paper of Blomstrand, as far as we are aware, no further reference to augelite has been made, and, since hitherto distinct crystals of the mineral have not bhen found, its position as a definite species seems to have been regarded as somewhat doubtful, for we find that Dana, in the last edition of the System of Mineralogy, includes it in a rather unsatisfactory group of more or less doubtful species appended to evansite.

Some years ago I found in the bed of the Satlej at Wangtu, in the N.W. Himalayas, about 80 miles, as the crow flies, north-east of Simla, a boulder of the rock described in the following pages, It is composed of biotite, cyanite and cordierite, with apatite and rutile as accessory minerals.

The specimen has a specific gravity of 2.92. Its structure is holocrystalline, and its minerals show no trace of parallelism in their orientation. It is, I think, a product of contact metamorphism.

This new antimonate was found in the cinnabar-bearing gravel of Tripuhy, Minas Geraes, Brazil. The minerals associated with it in the gravel, including the two new titano-antimonates lewisite and derbylite, have already been described in this Magazine (Vol. XI. No. 50, p. 80, and No. 52, p. 176). The original source of the lewisite and derbylite has been proved to be the muscovite-schists accompanying itabirite in the neighbourhood of Tripuhy, From these schists, doubtless, is also derived the tripuhyite, although as yet this new mineral has only been found in fragments loose in tbe gravel, and not intergrown with fragments of the muscovite-schists as is often the ease with both lewisite and derbylite.