Finding, therefore, no red substance capable as such of increasing the sensitiveness of AgBr to green light, and, on the other hand, many colorless substances which have that effect, I am entirely confirmed in the opinion originally expressed in the pages of this Journal, that there exists no relation between the color of a substance and that of the rays to which it increases the sensitiveness of silver bromide.

Philadelphia, March 13, 1876.

ART. LX. Contributions from the Sheffield Laboratory of Yale College. No. XXXIX.-On the Chemical Composition of Durangite; by GEORGE J. BRUSH.

IN an article on this rare mineral, published in 1869, I expressed the hope to make further examination of its chemical composition whenever sufficient material could be obtained for this purpose. Several years elapsed before any new discoveries of the mineral in Durango were made. I am again indebted to Mr. Henry G. Hanks of San Francisco for a new supply of the crystals obtained in recent explorations. These crystals are much smaller than those previously examined, being from one to three millimeters in diameter, and they are of a darker shade of color. The former were loose detached crystals, while these are associated with, and in some cases attached to, rolled fragments of crystallized hematite and cassiterite. The density of the small dark colored crystals is 4.07, while that of the purest of the bright colored crystals before described is 3.937. In all other physical characters there is a perfect correspondence between the two varieties.

The chemical examination of the dark colored small crystals has been undertaken, at my request, by my assistant Mr. George W. Hawes, first to estimate the amount of fluorine in the mineral, which in two determinations he found to be 7·67 and 749 per cent, and Mr. Hawes has also placed at my disposal for this article a complete analysis of this variety of the mineral. The fluorine was determined directly by Wöhler's method as modified by Fresenius.† To determine the arsenic acid, and the bases, the mineral was decomposed by sulphuric acid, and the arsenic weighed as sulphide: the alumina, iron and manganese obtained in the analysis were carefully examined to ascertain their purity. The soda and lithia were

*This Journal, II, xlviii, 179.

+Fresenius' Quantitative Analysis (Johnson's edition), p. 406.

weighed as sulphates and then converted into chlorides and separated by ether and alcohol.

The results of the analysis are as follows:

The percentage of fluorine, 7.67, corresponds to 3.23 per cent of oxygen, which, being subtracted, the analysis foots up 99-76, and calculating the percentages of the elements we have the following:

Atomic ratio,

462

335

As Al Fe Mn Na Li 0 Fl 34.63 9.18 6:50 1:45 9.69 0.31 30.33 7·67 = 99.76 116 026 ·421 044 1.894 404

Multiplying this ratio by the quantivalence of the elements we find that the ratio of R:R: As is 1:3:5 with 9(0, Fl.) The formula may therefore be written (Na, Li)3 (Al, Fe, Mn) As2 (O, Fl2)'

The ratio of oxygen to fluorine in the mineral is as 94: 1. This is a confirmation of the conclusion drawn by me from the analysis of the lighter colored crystals described in the original paper.*

The mean of my two analyses gave:

The variety examined by Mr. Hawes contains less alumina, and considerably more iron, which accounts for its darker color and slightly higher specific gravity. His results prove the mineral to be an arseniate analogous in chemical composition to amblygonite as suggested in my previous paper.

*Loc. cit.

ART. LXIII.-The Geological Survey of Brazil. First Preliminary Report made to the Counsellor Thomaz José Coelho de Almeida, Minister and Secretary of State for Agriculture, etc.; by CH. FRED. HARTT, Chief of the Geological Commission of the Empire of Brazil. Rio de Janeiro.

1876.*

On the first of May, last year [1875] I had the honor of being appointed chief of the Commission charged with the undertaking of a geological survey of the Empire. At the same time Dr. Elias Pacheco Jordao was chosen assistant, and Messrs. Orville A. Derby and Richard Rathbun were appointed assistant geologists. The government having given me authority, I engaged, as the photographer of the Commission, Senhor Marc Ferrez, photographer of the Royal Navy.

While awaiting instructions, and with the permission of his Excellency the Minister of Agriculture, I went with Commander E. P. Wilson to the district of Sao Gonçalo, in the southwestern part of the province of Minas Geraes, for the purpose of inspecting the auriferous tract pertaining to a grant in which the latter was interested. I examined with care the region between Bella Vista, on the Don Pedro II railroad and Sao Gonçalo, including the gold mines of that locality and in the vicinity of the city of Campanha, verifying the occurrence of gold in three distinct deposits.

First, it is found in gneiss of what appears to be the upper part of the Archæan series (Laurentian), the metal being distributed in more or less irregular veins, apparently conforming in direction with the stratification of the rocks. Throughout this section the gneiss is very much decomposed, remaining, however, in situ, the decomposition at times extending to a depth of 30 meters or more. While in the solid gangue the proportion of gold is not great enough to pay the cost of working, it has been extracted with profit in many places from the decomposed portion of these same lodes. It is quite probable that this softened rock contains proportionally more gold than the solid gneiss, owing to the accumulation of the contained

* Translated and arranged for publication in this Journal by Theo. B. Comstock, Ass't Professor, in charge of Geological Department, Cornell University, Ithaca, N. Y. This document has not yet been published in Brazil. A copy of the original manuscript in the Portuguese language has been kindly furnished to the translator, with full liberty of action. The report has been shortened by the omission of not a little concerning the details of the Survey, which can have no special interest to the readers of this Journal. The mode of arrangement and construction of sentences differs somewhat from the original, but mainly in cases in which conciseness is gained without the sacrifice of perspicuity. Statements of fact or opinion will be found to correspond as closely as possible with the ideas of the author. Explanatory words within brackets are added by the translator. Words enclosed in parentheses occur thus in the original.—[T. B. C.]

Action of

metal of the lower portion in this part of the vein. this nature was first demonstrated by Dr. James E. Mills in the auriferous district of Rio Grande do Sul and has been confirmed by the observations of Viscount de Barbaçena in his mineralogical studies of the auriferous gneiss of Minas Geraes.

Secondly, the gold occurs in the lower part of a deposit of red earth, which covers the whole surface of this region. This is a mechanical mixture of the ingredients of the decomposed gneiss, the local origin of which is not perfectly evident. All parts of the earth are not equally rich, and the gold is probably distributed in some relation to the outcrops, as has been shown by Dr. Mills in the eastern part of the province.

Thirdly, some gold is found in the alluvial deposits of the river-valleys, and in the vicinity of the ancient excavations of Sao Gonçalo and Campanha; the abundant "waste" also contains a considerable quantity of the metal.

Notwithstanding the small amount of gold and the fact that the richer deposits have already been extensively washed, this section appears to be worthy of scientific study, and, probably, by the application of methods similar to those employed in the California placers, the region may yet become productive. To this end are required the detailed exploration of this part of the country, the limitation of claims and the introduction of a cheap method of extraction. A survey of the character mentioned will probably be soon made.

Last year, Dr. Mills spent about six months in the study of the auriferous region of Minas Geraes, where he succeeded in discovering and applying the law of the distribution of gold, especially of its occurrence in the series of rocks characterized by the presence of ilacolumite,-a law extremely important in its bearing upon the future development of the resources of that province. This gentleman has promised to give me a complete report, in which he will not only enunciate and illustrate this law, but will also include all the scientific results of his interesting explorations.

For several years past I have been engaged in the systematic examination of the border of the Brazilian plateau for the purpose of discovering and critically studying the different geological formations, which are there better developed and more fossiliferous than in the interior, being convinced that I could thus most quickly obtain a sure basis for the study of the general geology of the country. Having received instructions for continuing those studies by beginning in the northern provinces and proceeding southward, I judged it best to inaugurate the labors of the Commission by the exploration of the coast of Pernambuco. On the 10th of July [1875] I left Rio for that province, accompanied by Drs. Jordao and Freitas, and Senhor

Ferrez. Opening a provisional laboratory [in the City of Pernambuco], I commenced the examination of the vicinity, soon discovering limestone beds of the Cretaceous formation containing many species of fossils. The explorations were continued northward as far as Catuáma. At Maria Farinha, Cretaceous rocks were found so rich in fossils that several weeks were spent in carefully studying the formation, making a map. of the locality, determining accurately the position and sequence of the strata, and collecting enormous quantities of fossils, among which are very many new species. Other collections were obtained from the vicinity of Olinda, from Iguarassú, from the island of Itamaracá, and at Catuáma.

I studied with especial care the limestones which are used for lime-making, analyses of which will be given in my report, and I am also preparing for the same report a chapter on the manufacture of lime and the construction of calcining furnaces.

The examination of the coast was then continued as far as Santo Agostinho, including a reconnoissance along the line of the railroad to Una, with excursions from several points upon both sides. The geology of this portion of the country is, however, extremely monotonous and of little interest.

At the same time I examined carefully the reefs along the coast,-whose geological features are of the highest importance to the country. As I have already indicated to your Excellency, these are of two classes: 1. Coral Reefs, composed of calcareous material derived from the débris of certain species of Polyps, Acalephs, and calcareous plants, and 2. Consolidated Praias [Beaches], made up of compacted siliceous sand. The reef of Pernambuco is a representative of the latter class This I examined very minutely, and, with the aid of Drs. Jordao and Freitas, made a map of a portion of it. Senhor Ferrez, under my direction, procured a fine series of photographs exhibiting its structure and appearance.

In the study of these reefs a magnificent collection of corals. was obtained, including hundreds of large specimens, with many new species. Along with these, and with very little trouble, we gathered large numbers of marine animals, such as fishes, Echinoderms, Crustaceans, etc., etc., among which there is an extraordinary variety of unknown forms.

In September I went with the members of the Commission to make a reconnoissance along the Sao Francisco River to a point a little above the rapids of Paulo Affonso. We ascended the river in a sailing-boat as far as Piranhas, the limit of navigation. At that place, thanks to Senhor Ventura, we found prompt conveyance to the falls, where we remained eight days. Senhor Ferrez took a series of views of the most characteristic points. This was an extremely laborious task, and I cannot but com