will say nothing: partly because to do so would transgress the limits which I proposed to myself at the outset, and partly because after reading and re-reading several times I find it impossible to hold any clear image of the new theory in my mind, and I fear therefore that I might do the author injustice. Berkeley, California, April 1, 1878.

ART. XI.-On the occurrence of a Solid Hydrocarbon in the Eruptive Rocks of New Jersey; by I. C. RUSSELL.

(Read before the New York Academy of Sciences, April 29, 1878.) IN an article by T. Sterry Hunt, published in this Journal in 1863,* mention is made of an interesting locality at Cape Gaspé where a trap dike intersects the sedimentary rocks. The cavities in the trap are frequently lined with chalcedony, or with crystals of calcite or quartz, and filled with petroleum which in some cases has assumed the hardness of pitch. Recently our attention was called to a newspaper account of the occurrence of mineral oil in the lava of Mt. Etna. numerous round or irregular cavities contained in the lava are described as being coated with aragonite and filled with mineral oil. An analogous instance in our own country has been familiar to me for some time, which, taken in connection with the occurrences mentioned above, seems to be of sufficient interest to be worth recording.

The

Associated with the sheet of trap rock known as the First Newark Mountain, which traverses the central portion of the Triassic formation of New Jersey, there occurs near Plainfield, at an abandoned copper mine on the western slope of the mountain the upper surface of the trap sheet-an amygdaloid trap passing into a metamorphosed shale. In this region it is frequently impossible to distinguish in small exposures, the genuine trap from the metamorphosed shales that rest in contact with it. Many of the cavities in the amygdaloidal rock are filled with a brilliant jet black carbonaceous mineral resembling very closely the albertite of New Brunswick.t These cavities are frequently tubular in shape, having a length of three or four inches and usually a diameter of about a quarter of an inch. Sometimes these tubes were lined throughout by infiltration, with a coating of quartz or calcite a line or *Vol. xxxv, p 166, 1863.

See article by the writer "On the Intrusive Nature of the Triassic Trap Sheets of New Jersey," in this Journal, xv, 277, 1878.

The occurrence of "bitumen" in amygdaloid trap is briefly mentioned by E. S. Dana in an article on the Trap Rocks of the Connecticut Valley. Am. Assoc. Sci., 1874, B 47 [also, with an explanation of its origin, by . W. Hawes, in this Journal, ix, 456, 1875.]

two in thickness, before the carbonaceous material was introduced. Above the amygdaloid is found a metamorphosed shale which still retains its bedded structure, and in places presents something of the usual reddish color of the unaltered shales. This altered rock is traversed in various directions by seams and fissures, which are frequently filled with the same albertite-like mineral. Resting upon these metamorphosed beds occur slates, shales and sandstones, which contain fossil fishes and a considerable abundance of obscure vegetable remains. It seems evident that these organic bodies furnished by their decomposition the carbonaceous material in the associated rocks. The heat derived from the slowly cooling injected rocks may have played an important part in this process.

The mineral whose geological occurrence we have thus described, gives, when subjected to chemical tests, almost precisely the same reactions as albertite. It is insoluble in heated acids and alkalies, and is but sparingly if at all soluble in alcohol, ether, or oil of turpentine. Like albertite, also, it is infusible, but softens by heat and burns with a yellow flame, emitting an agreeable odor. It gives when incinerated less than 0.10 per cent of ash.

The occurrence of petroleum in the cavities of the igneous rocks of Gaspé and Sicily, and of a solid hydrocarbon in the trap rocks of New Jersey would seem to be but different stages in the same process. If the cavities in a rock were filled with petroleum by infiltration, and evaporation slowly removed the more volatile portions, and oxidation took place to some extent, the result would be the formation of a deposit of solid hydrocarbon in the cavities. A similar process sometimes occurs with bottled samples of petroleum, by which the interior of the bottle is left coated with a solid carbonaceous layer. In the rocks, if a fresh supply of oil was furnished from time to time by infiltration, the cavities would eventually become completely filled with the solid carbonaceous residue. A vesicular lava might in this manner be changed to an amygdaloid, the cavities of which would be filled with solid hydrocarbons instead of quartz, zeolites, etc.

Such, it appears to us, must have been the history of the Triassic amygdaloid we have described, the cavities of which must at one time have been filled with mineral oil. This is but an epitome of what took place on a grand scale at the great fissure over 1,400 feet deep, in New Brunswick, which was filled with albertite, and in the case of the Grahamite in West Virginia, which also occupies an immense fissure.

Since writing the above, our attention has been called, through the kindness of Prof. J. D. Dana, to the fact that Percival in his report on the geology of Connecticut, published in

1842, records the occurrence of "bitumen" in connection with the trap rock of Connecticut. Those who are fortunate enough to possess a copy of this report will find that Percival with his usual accuracy of observation, mentions several times the occurrence of this substance while describing the Triassic formation of the Connecticut Valley. Mr. Percival speaks particularly of "indurated bitumen" as occurring in the cavities of amygdaloid trap, and in small veins in the indurated shale adjoining. Associated with these rocks occur, also, bituminous limestones and shales containing fossil fishes. Similar bituminous rocks he describes as occurring in the small isolated Triassic area of Southbury and Woodbury in Western Connecticut. In reference to the reported discovery of coal in Connecticut the statement is made that "This substance, however, is a more or less indurated bitumen, similar to that occasionally occupying the pores of amygdaloid, or accompanying metallic veins in the trap and the adjoining indurated sandstones, and is perhaps derived from the same volcanic source as the trap it accompanies."

It will be noticed from the above that the bitumen described by Percival has the same geological associations as the mineral occurring at Plainfield, N. J. A specimen of this mineral from Connecticut which we have just received from Dr. H. C. Bolton of Trinity College, and obtained by him from seams in trap rock near the new college buildings at Hartford, seems identical in its physical and chemical properties with the solid hydrocarbon we have described from New Jersey. The rocks with which this mineral is associated in Connecticut correspond in lithological characters and geological position with the eruptive rock of New Jersey, and are a portion of the great system of trap ridges which traverse the Triassic formation in Connecticut and Massachusetts.

ART. XII.-On a new and remarkable mineral locality in Fairfield County, Connecticut; with a description of several new species occurring there; by GEO. J. BRUSH and EDWARD S. DANA. First Paper.

[Continued from page 46.]

3. DICKINSONITE.

Physical characters.-Dickinsonite occurs most commonly in crystalline masses, which have a distinctly foliated, almost micaceous, structure. It is also lamellar-radiated and sometimes stellated, the laminæ being usually more or less curved. This massive variety forms the gangue in which crystals of eosphorite are often imbedded, and also sometimes triploidite. It more

* Percival's Geol. Rep. of Conn., p. 452.

over occurs in minute scales distributed through the massive eosphorite and giving it a green color, and is sometimes imbedded in the rhodochrosite. Minute tabular crystals are rare; they are observed implanted upon the gangue, and also scattered through the reddingite. In general aspect the mineral resembles some varieties of chlorite though very unlike in its brittleness. It has perfect basal cleavage. The hardness is 3-5-4, and the specific gravity 3.338-3-343. Luster vitreous, on the cleav age face somewhat pearly. The color of the purest crystal is oil- to olive-green, in the massive varieties generally grass-green though sometimes quite dark; the streak is nearly white. Transparent to translucent, the crystals being perfectly clear. The laminæ are very brittle; fracture uneven.

5.

Crystalline form.-Distinct crystals of dickinsonite are not often found, and owing to the extremely brittle character of the mineral, it is only in very rare cases that they can be obtained showing more than the basal plane. The crystallographic data which are given here were all obtained from two crystals, which, though extremely small and yielding only approximate angles, yet served to decide all the essential points. Other less perfect crystals gave confirmatory results.

a

с

Dickinsonite crystallizes in the MONOCLINIC SYSTEM. The axial ratio and obliquity were obtained from the following angles :

Plane angle of the base
сла, 001 100,
001301,

120° 0'

= =

61° 30'

42° 30'

For the unit prism (not observed),

1.I = 66° 36', and 113° 24'

The observed planes are as follows:

B=61° 30′

The adjoining figure shows all of these planes except the clinopinacoid, which was only once observed.

The following are the most important angles, measured and calculated:

*

It will be seen from the above table that the angle between the base and one of the two pyramids (cp=61° 8') differs but little from the angle between the base and the orthopinacoid (ca=61°30'); there are thus three planes which have nearly equal inclinations to the base. This fact, which is analogous to that true of the Vesuvian biotite (meroxen) as pointed out by Tschermak, gives to the crystals a marked rhombohedral aspect especially as the planes x (301) and s (221) have usually a minor development. As exact measurements were not possible the true relations could hardly be established beyond doubt until recourse was had to an optical examination. This showed that the cleavage planes are not isotrope as they must be if rhombohedral; on the contrary one plane of vibration is exactly parallel to the edge c/a, and the other normal to it.

The rhombohedral pseudo-symmetry is also shown in the fact that the plane angle of the base differs very little if at all from 120°. The most careful measurements practicable failed to establish any variation. That the angle really is 120° seems, moreover, to be indicated by the fact that on many cleavage lamina triangular markings are visible, which are apparently equilateral the angles measuring 60°; other analogous markings have four or five sides but always with angles of 60° or 120° as near as the measurements can be made.

The above facts show that crystallographically dickinsonite is related to the micas and chlorites, although most unlike chemically.

The plates of dickinsonite are sometimes striated parallel to the edges c/p, c/p', and also c/a, corresponding to the triangular markings mentioned and still more increasing the rhombohedral aspect of the crystals. No twins have been observed, although some very imperfect crystals early suggested their possible occurrence.

The cleavage plates show a marked dichroism, parallel to the edge ca, the rays being grass-green and much absorbed

* Groth, Zeitschrift für Krystallographie, ii, p. 19, 1877.