It is also stated by Captain Flinders, that three of the English Company's Islands which were examined, slope down nearly to the water on their west sides; but on the east, and more especially the south-east, they present steep cliffs; and the same conformation, he adds, seemed to prevail in the other islands.* If this structure occurred only in one or two instances, it might be considered as accidental; but as it obtains in so many cases, and is in harmony with the direction of the ranges, it is not improbably of still more extensive occurrence, and would intimate a general elevation of the strata towards the south-east.

(*Footnote. Flinders Volume 2 page 235.)

Now on examining the general map, it will be seen, that the lines of the coast on the mainland, west of the Gulf of Carpentaria, between Limmen's Bight and Cape Arnhem--from the bottom of Castlereagh Bay to Point Dale--less distinctly from Point Pearce, latitude 14 degrees 23 minutes, longitude 129 degrees 18 minutes, to the western extremity of Cobourg Peninsula, and from Point Coulomb, latitude 17 degrees 20 minutes, longitude 123 degrees 11 minutes, to Cape Londonderry, have nearly the same direction; the first line being about one hundred and eighty geographical miles, the second more than three hundred, and the last more than four hundred miles, in length.* And these lines, though broken by numerous irregularities, especially on the north-west coast, are yet sufficiently distinct to indicate a probable connexion with the geological structure of the country; since the coincidence of similar ranges of coast with the direction of the strata, is a fact of very frequent occurrence in other parts of the globe.** And it is observable that considerable uniformity exists in the specimens, from the different places in this quarter of New Holland which have been hitherto examined; sandstone, like that of the older formations of Europe occurring generally on the north and north-west coasts, and appearing to be extensively diffused on the north-west of the Gulf of Carpentaria, where it reposes upon primitive rocks.***

(*Footnote. It is deserving of notice, that the coast of Timor, the nearest land on the north-west, at the distance of about 300 miles, is also nearly straight, and parallel to the Coast of New Holland in this quarter: part of the mountainous range, of which that island consists, being probably more than 9000 feet high; and its length, from the north-eastern extremity to the South-West of the adjoining island of Rottee, about 300 miles. But, unfortunately for the hypothesis, a chain of islands immediately on the north of Timor, is continued nearly in a right line for more than 1200 miles (from Sermatta Island to the south-eastern extremity of Java) in a direction FROM EAST TO WEST. This chain, however, contains several volcanoes, including those of Sumbawa, the eruption of which, in 1815, was of extraordinary violence. See Royal Inst. Journal volume 1 1816 page 248 etc.

At Lacrosse Island, in the mouth of Cambridge Gulf, on the north-west coast of New Holland, the beds rise to the North-West: their direction consequently is from South-West to North-East; and the rise towards the high land of Timor. The intervening sea is very shallow.)

(**Footnote. A remarkable case of this kind, which has not, I believe, been noticed, occurs in the Mediterranean; and is conspicuous in the new chart of that sea, by Captain W.H. Smyth. The eastern coast of Corsica and Sardinia, for a space of more than two hundred geographical miles being nearly rectilinear, in a direction from north to south; and, Captain Smyth has informed me, consisting almost entirely of granite, or, at least, of primitive rocks. The coast of Norway affords another instance of the same description; and the details of the ranges in the interior of England furnish several examples of the same kind, on a smaller scale.)

(***Footnote. The coastlines nearly at rightangles to those above-mentioned--from the South-East of the Gulf of Carpentaria to Limmen's Bight, from Cape Arnhem to Cape Croker, and from Cape Domett to Cape Londonderry--have also a certain degree of linearity; but much less remarkable, than those which run from South-West to North-East.)

The horn-like projection of the land, on the east of the Gulf of Carpentaria, is a very prominent feature in the general map of Australia, and may possibly have some connexion with the structure just pointed out. The western shore of this horn, from the bottom of the gulf to Endeavour Straits, being very low; while the land on the east coast rises in proceeding towards the south, and after passing Cape Weymouth, latitude 12 degrees 30 minutes, is in general mountainous and abrupt; and Captain King's specimens from the north-east coast show that granite is found in so many places along this line as to make it probable that primitive rocks may form the general basis of the country in that quarter; since a lofty chain of mountains is continued on the south of Cape Tribulation, not far from the shore, throughout a space of more than five hundred miles. It would carry this hypothesis too far to infer that these primitive ranges are connected with the mountains on the west of the English settlements near Port Jackson, etc., where Mr. Scott has described the coal-measures as occupying the coast from Port Stevens, about latitude 33 degrees to Cape Howe, latitude 37 degrees, and as succeeded, on the eastern ascent of the Blue Mountains, by sandstone, and this again by primitive strata:* But it may be noticed that Wilson's Promontory, the most southern point of New South Wales, and the principal islands in Bass Strait, contain granite; and that primitive rocks occur extensively in Van Diemen's Land.

(*Footnote. Annals of Philosophy June 1824.)

The uniformity of the coastlines is remarkable also in some other quarters of Australia; and their direction, as well as that of the principal openings, has a general tendency to a course from the west of south to the east of north. This, for example, is the general range of the south-east coast, from Cape Howe, about latitude 37 degrees, to Cape Byron, latitude 29 degrees, or even to Sandy Cape, latitude 25 degrees; and of the western coast, from the south of the islands which enclose Shark's Bay, latitude 26 degrees, to North-west Cape, about latitude 22 degrees. From Cape Hamelin, latitude 34 degrees 12 minutes, to Cape Naturaliste, latitude 33 degrees 26 minutes, the coast runs nearly on the meridian. The two great fissures of the south coast, Spencer's, and St. Vincent's Gulfs, as well as the great northern chasm of the Gulf of Carpentaria, have a corresponding direction; and Captain Flinders (Chart 4) represents a high ridge of rocky and barren mountains, on the east of Spencer's Gulf, as continued, nearly from north to south, through a space of more than one hundred geographical miles, between latitude 32 degrees 7 minutes and 34 degrees. Mount Brown, one of the summits of this ridge, about latitude 32 degrees 30 minutes, being visible at the distance of twenty leagues.

The tendency of all this evidence is somewhat in favour of a general parallelism in the range of the strata, and perhaps of the existence of primary ranges of mountains on the east of Australia in general, from the coast about Cape Weymouth* to the shore between Spencer's Gulf and Cape Howe. But it must not be forgotten, that the distance between these shores is more than a thousand miles in a direct line; about as far as from the west coast of Ireland to the Adriatic, or double the distance between the Baltic and the Mediterranean. If, however, future researches should confirm the indications above mentioned, a new case will be supplied in support of the principle long since advanced by Mr. Michell,** which appears (whatever theory be formed to explain it) to be established by geological observation in so many other parts of the world, that the outcrop of the inclined beds, throughout the stratified portion of the globe, is everywhere parallel to the longer ridges of mountains, towards which, also, the elevation of the strata is directed. But in the present state of our information respecting Australia, all such general views are so very little more than mere conjecture, that the desire to furnish ground for new inquiry, is, perhaps, the best excuse that can be offered for having proposed them.

(*Footnote. The possible correspondence of the great Australian Bight, the coast of which in general is of no great elevation, with the deeply-indented Gulf of Carpentaria, tending, as it were, to a division of this great island into two, accords with this hypothesis of mountain ranges: but the distance between these recesses, over the land at the nearest points, is not less than a thousand English miles. The granite, on the south coast, at Investigator's Islands, and westward, at Middle Island, Cape Le Grand, King George's Sound, and Cape Naturaliste, is very wide of the line above-mentioned, and nothing is yet known of its relations.)

(**Footnote. On the Cause of Earthquakes. Philosophical Transactions 1760 volume 51 page 566 to 585, 586.)

DETAILED LIST OF SPECIMENS.

The specimens mentioned in the following list have been compared with some of those of England and other countries, principally in the cabinets of the Geological Society, and of Mr. Greenough; and with a collection from part of the confines of the primitive tracts of England and North Wales, formed by Mr. Arthur Aikin, and now in his own possession. Captain King's collection has been presented to the Geological Society; and duplicates of Mr. Brown's specimens are deposited in the British Museum.

RODD'S BAY, on the East Coast, discovered by Captain King, about sixty miles south of Cape Capricorn.* Reddish sandstone, of moderately-fine grain, resembling that which in England occurs in the coal formation, and beneath it (mill-stone grit). A sienitic compound, consisting of a large proportion of reddish felspar, with specks of a green substance, probably mica; resembling a rock from Shap in Cumberland.

(*Footnote. In Captain King's collection are also specimens found on the beach at Port Macquarie, and in the bed of the Hastings River, of common serpentine, and of botryoidal magnesite, from veins in serpentine. The magnesite agrees nearly with that of Baudissero, in Piedmont. (See Cleaveland's Mineralogy 1st edition page 345.)

CAPE CLINTON, between Rodd's Bay and the Percy Islands. Porphyritic conglomerate, with a base of decomposed felspar, enclosing grains of quartz and common felspar, and some fragments of what appears to be compact epidote; very nearly resembling specimens from the trap rocks* of the Wrekin and Breeden Hills in Shropshire. Reddish and yellowish sandy clay, coloured by oxide of iron, and used as pigments by the natives.

(*Footnote. By the terms Trap, and Trap-formation, which I am aware are extremely vague, I intend merely to signify a class of rocks, including several members, which differ from each other considerably in mineralogical character, but agree in some of their principal geological relations; and the origin of which very numerous phenomena concur in referring to some modification of volcanic agency. The term Greenstone also is of very loose application, and includes rocks that exhibit a wide range of characters; the predominant colour being some shade of green, the structure more or less crystalline, and the chief ingredients supposed to be hornblende and felspar, but the components, if they could be accurately determined, probably more numerous and varied, than systematic lists imply.)

PERCY ISLANDS, about one hundred and forty miles north of Cape Capricorn. Compact felspar of a flesh-red hue, enclosing a few small crystals of reddish felspar and of quartz. This specimen is marked "general character of the rocks at Percy Island," and very much resembles the compact felspar of the Pentland Hills near Edinburgh, and of Saxony. Coarse porphyritic conglomerate, of a reddish hue. Serpentine. A trap-like compound, with somewhat the aspect of serpentine, but yielding with difficulty to the knife. This specimen has, at first sight, the appearance of a conglomerate, made up of portions of different hues, purplish, brown, and green; but the coloured parts are not otherwise distinguishable in the fracture: It very strongly resembles a rock which occurs in the trap-formation, near Lyd-Hole, at Pont-y-Pool, in Shropshire. Slaty clay, with particles of mica, like that which frequently occurs immediately beneath beds of coal.

REPULSE ISLAND, in Repulse Bay, about one hundred and twenty miles north-west of the Percy Islands. Indistinct specimens, apparently consisting of decomposed compact felspar. A compound of quartz, mica, and felspar, having the appearance of re-composed granite.

CAPE CLEVELAND, about one hundred and twenty miles north of Repulse Island. Yellowish-grey granite, with brown mica; "from the summit of the hill." Reddish granite, of very fine grain; with the aspect of sandstone. Dark grey porphyritic hornstone, approaching to compact felspar, with imbedded crystals of felspar.

CAPE GRAFTON, about one hundred and eighty miles west of north from Cape Cleveland. Close-grained grey and yellowish-grey granite, with brown mica. A reddish granitic stone, composed of quartz, felspar, and tourmaline.

ENDEAVOUR RIVER, about one hundred miles west of north from Cape Grafton. Grey granite of several varieties; from a peaked hill under Mount Cook and its vicinity. Granular quartz-rock of several varieties: and indistinct specimens of a rock approaching to talc-slate.

LIZARD ISLAND, about fifty miles east of north from Endeavour River. Grey granite, consisting of brown and white mica, quartz, and a large proportion of felspar somewhat decomposed.

CLACK ISLAND, near Cape Flinders, on the north-west of Cape Melville, about ninety miles north-west of Lizard Island. Smoke-grey micaceous slaty-clay, much like certain beds of the old red sandstone, where it graduates into grey wacke. This specimen was taken from a horizontal bed about ten feet in thickness, reposing upon a mass of pudding-stone, which included large pebbles of quartz and jasper; and above it was a mass of sandstone, more than sixty feet thick. (Narrative volume 2.)

SUNDAY ISLAND, near Cape Grenville, about one hundred and seventy miles west of north from Cape Melville. Compact felspar, of a flesh-red colour; very nearly resembling that of the Percy Islands, above-mentioned.

GOOD'S ISLAND, one of the Prince of Wales group, about latitude 10 degrees, thirty-four miles north-west of Cape York. The specimens, in Mr. Brown's collection from this place, consist of coarse-slaty porphyritic conglomerate, with a base of greenish-grey compact felspar, containing crystals of reddish felspar and quartz. This rock has some resemblance to that of Clack Island above-mentioned.

SWEER'S ISLAND, south of Wellesley's group, at the bottom of the Gulf of Carpentaria. A stalactitic concretion of quartzose sand, and fine gravel, cemented by reddish carbonate of lime; apparently of the same nature with the stem-like concretions of King George's Sound: (See hereafter.) In this specimen the tubular cavity of the stalactite is still open.

The shore, in various parts of this island, was found to consist of red ferruginous matter (Bog-iron-ore ?) sometimes unmixed, but not unfrequently mingled with a sandy calcareous stone; and in some places rounded portions of the ferruginous matter were enveloped in a calcareous cement.

BENTINCK ISLAND, near Sweer's Island. A granular compound, like sandstone recomposed from the debris of granite. Brown hematite, enclosing quartzose sand.

PISONIA ISLAND, on the east of Mornington's Island, is composed of calcareous breccia and pudding-stone, which consist of a sandy calcareous cement, including water-worn portions of reddish ferruginous matter, with fragments of shells.

NORTH ISLAND, one of Sir Edward Pellew's group. Coarse siliceous sand, concreted by ferruginous matter; which, in some places, is in the state of brown hematite. Calcareous incrustations, including fragments of madrepores, and of shells, cemented by splintery carbonate of lime.

CAPE-MARIA ISLAND, in Limmen's Bight, was found by Mr. Brown to be composed principally of sandstone. The specimens from this place, however, consist of grey splintery hornstone, with traces of a slaty structure; and of yellowish-grey flint, approaching to chalcedony; with a coarse variety of cacholong, containing small nests of quartz crystals.

GROOTE EYLANDT is composed of sandstone, of which two different varieties occur among the specimens. A quartzose reddish sandstone, of moderately fine grain; and a coarse reddish compound, consisting almost exclusively of worn pebbles of quartz, some of which are more than half an inch in diameter, with a few rounded pebbles of chalcedony. The latter rock is nearly identical with that of Simms' Island, near Goulburn's Island on the north coast.

CHASM ISLAND, WINCHELSEA ISLAND, and BURNEY'S ISLAND, are of the same materials as Groote Eylandt: and sandstone was found also on the western shore of BLUE-MUD BAY.

On the shore of the mainland, opposite to Groote Eylandt, a little north of latitude 14 degrees, Mr. Brown observed the common sandy calcareous stone, projecting here and there in ragged fragments.

MORGAN'S ISLAND, in Blue-Mud Bay, north-west of Groote Eylandt, is composed principally of clink-stone, sometimes indistinctly columnar. But among the specimens are also a coarse conglomerate of a dull purplish colour, including pebbles of granular quartz and a fragment of a slaty rock like potstone: the hue and aspect of the compound being precisely those of the oldest sandstones. Reddish quartzose sandstone, of uniform and fine grain. A concretion of rounded quartz pebbles, cemented by ferruginous matter, apparently of recent formation.

ROUND HILL, near Cape Grindall, a prominence east of north from Blue-Mud Bay, was found by Captain Flinders to consist, at the upper part, of sandstone. The specimens of the rocks in its vicinity are, dark grey granite, somewhat approaching to gneiss, with a few specks of garnet; and a calcareous, probably concretional stone, enclosing the remains of shells, with cavities lined with crystals of calcareous spar.

MOUNT CALEDON, on the mainland, west of Caledon Bay, consists of grey granite, with dark brown mica in small quantity; and on the sides and top of the hill large loose blocks of that rock were observed, resting upon other blocks.

A small island, near Cape Arnhem, is also composed of granite, in which the felspar has a bluish hue.

Smaller of the MELVILLE ISLANDS, north-east of Melville Bay.* A botryoidal mass of ferruginous oxide of manganese, approaching to hematite; the fissures in some places occupied by carbonate of lime.

(*Footnote. The relative position of the islands and bays on this part of the coast is represented in the enlarged Map.)

MELVILLE BAY. Granite, composed of grey and somewhat bluish felspar, dark brown mica, and a little quartz; containing minute disseminated specks of molybdena, and indistinct crystals of pale red garnet.

RED CLIFFS, south-west of Arnhem Bay; on the line of the first chain of islands mentioned by Captain Flinders. (See the Map, figure 3.) Friable conglomerate, of a full brick-red colour, consisting of minute grains of quartz, with a large proportion of ochreous matter.

MALLISON'S ISLAND. (Map, figure 4.) The cliffs of this island are composed of a fissile primitive rock, on which sandstone reposes in regular beds. The specimen of the former resembles gneiss, or mica slate, near the contact with granite: the sandstone is thick-slaty, quartzose, of a reddish hue, with mica disseminated on the surfaces of the joints; and one face of the specimen is incrusted with quartz crystals, thinly coated with botryoidal hematite. Light grey quartzose sandstone of a fine grain, with a thin coating of brown hematite, was also found in this island: And a breccia, consisting of angular fragments of sandstone, cemented by thin, vein-like, coatings of dark brown hematite, was found there, in loose blocks at the bottom of perpendicular cliffs. The specimen of this breccia is attached to a plate of granular quartz, and may possibly have been part of a vein.

The shore of INGLIS' ISLAND, the largest of the ENGLISH COMPANY'S RANGE (2. 2. 2. in the Map) is formed of flat beds, of a slaty argillaceous rock, which breaks into rhomboidal fragments; but the specimen is indistinct. Ferruginous masses, probably consisting of brown hematite, come also from this island.

ASTELL'S ISLAND, north-east of Inglis' Isle. Very fine-grained greyish-white quartzose sandstone; identical with that of Mallison's Island, and very closely resembling some of the specimens from Prince Regent's and Hunter's Rivers.

Among the remaining islands of this range, BOSANQUET'S, COTTON'S, and POBASSOO's Isles, were found by Mr. Brown to consist, in a great measure, of sandstone, of the same character with the specimens above-mentioned.

POBASSOO'S ISLAND, a small islet south-east of Astell's Isle. Fine-grained, somewhat reddish, sandstone. Another specimen of sandstone is friable, of a light flesh-red colour, and apparently composed of the debris of granite. A crystalline rock, consisting of greenish-grey hornblende, with a very small proportion of felspar (Hornblende rock ?). Fragment, apparently from a columnar mass, of a stone intermediate between clink-stone and compact felspar.

Such of the English Company's Islands as were examined by Captain Flinders, are stated by him to consist, in the upper part, of a grit, or sandstone, of a close texture; the lower part being argillaceous, and stratified, and separating into pieces of a reddish colour, resembling flat tiles. The strata-dip to the west, at an angle of about 15 degrees.

South-west bay of GOULBURN'S SOUTH ISLAND, two hundred and fifty miles west of the Gulf of Carpentaria (Narrative 1). Coarse-grained reddish quartzose conglomerate and sandstone; resembling the older sandstones of England and Wales, and especially the mill-stone grit beneath the coal formation. Fine greyish-white pipe-clay; of which about thirty feet in thickness were visible, apparently above the sandstone last mentioned. Coarse-grained, ferruginous sandstone, containing fragments of quartz, from above the pipe-clay. The appearance of the cliff from which these specimens were taken, is represented in the view of the bay on the south of Goulburn Island (volume 1); and a distant head in the view consists of the same materials.

SIMMS ISLAND, on the west of Goulburn's south Island (Narrative 1) is composed of a reddish conglomerate, nearly identical with some of the specimens above-mentioned.

The western side of LETHBRIDGE BAY, on the north of MELVILLE ISLAND, consists of a range of cliffs like those at Goulburn's Island; the upper part being red, the lower white and composed of pipe-clay. The western extremity of BATHURST ISLAND, between CAPE HELVETIUS and CAPE FOURCROY, is also formed of cliffs of a very dark red colour.

LACROSSE ISLAND, at the mouth of CAMBRIDGE GULF, about one hundred miles from Port Keats. Reddish, very quartzose sandstone; from a stratum which dips to the south-east, at an angle of about ten or fifteen degrees. Micaceous and argillaceous fissile sandstone, of purplish and greenish hues, in patches, or occasionally intermixed; precisely resembling the rock of Brecon, in South Wales, and, generally, the old red sandstone of the vicinity of Bristol and the confines of England and Wales. Fine-grained thin-slaty sandstone, resembling certain beds of the coal formation, or of the millstone grit, is found in large masses, under an argillaceous cliff, on the north side of Lacrosse Island.

The specimens from the interior of Cambridge Gulf are from ADOLPHUS ISLAND, and consist of reddish and grey sandstone, more or less decomposed.

VANSITTART BAY, about one hundred and forty miles north-west of Cambridge Gulf. Reddish quartzose sandstone, or quartz-rock. Indistinct specimens of greenstone, with adhering quartz; apparently a primitive rock.

PORT WARRENDER, at the bottom of Admiralty Gulf, about forty miles south-west of Vansittart Bay (Narrative volume 1). Epidote and quartz, in small crystals confusedly interlaced; apparently from veins, or nests, but unaccompanied by any portion of the adjacent rock. The structure in one of these specimens approaches to the amygdaloidal. A compact greenish stone, with disseminated crystalline spots of epidote, and of quartz, and apparently consisting of an intimate mixture of those minerals, is also among the specimens from Port Warrender.

All these specimens are from detached water-worn masses at the foot of Crystal Head, on the south-west of the port. The summit of the head is flat and tabular, and the rocks in the vicinity are described by Captain King as consisting of siliceous sandstone. Chalcedony, apparently from amygdaloid of the trap formation, was also found at Port Warrender.

The epidote of this place is in general of a pale-greenish colour, but is mixed with, and sometimes appears to pass into, spots of a rich purplish-brown. The specimens resemble generally the epidote of Dauphiny and Siberia; but Mr. Levy, who has been so good as to examine them, informs me that the crystals exhibit some modifications not described either by Hauy, or by Mr. Haidinger in his paper on this mineral, and which are probably peculiar to this locality.

WATER ISLAND, on the west side of CAPE VOLTAIRE, at the south-west entrance of Port Warrender, is described (volume 1) as consisting of quartzose sandstone; as is also KATER ISLAND, in Montagu Sound. And the same rock appears to occur throughout the islands on this part of the coast. (Narrative 1.)

MONTAGU SOUND, about five-and-twenty miles south-west of ADMIRALTY GULF (Narrative 1). Greyish granular quartz; like that of the Lickey Hill, in Worcestershire. Fine-grained quartzose sandstone, of a purplish hue, resembling a rock on the banks of the Severn, near Bridgenorth. Grey and reddish sandstone; apparently composed of the debris of granite, and very nearly resembling that of Simms Island above-mentioned.

HUNTER'S RIVER, falling into YORK SOUND, on the north-east side. Somewhat coarse reddish-white sandstone; like that of the coal formation, and some varieties of millstone grit. Fine-grained, reddish-grey quartzose sandstone, having the appearance of stratification, and resembling the rocks of Cambridge Gulf.

ROE'S RIVER, at the eastern termination of York Sound (Narrative 1) runs between precipitous banks of sandstone, in nearly horizontal strata, which rise to the height of three hundred feet.

CAREENING BAY, between York Sound and Prince Regent's River (Narrative volume 1. See the plate volume 1). Crystalline epidote, and whitish quartz, apparently from a vein. Purplish-brown epidote, with small nests or concretions of green epidote and quartz; forming a sort of amygdaloid. Conglomerate, containing angular fragments of yellowish-grey quartz-rock, in a base of compact epidote. A nearly uniform greenish compound of epidote intimately mixed with quartz, also occurs at this place. Flat lamellar chalcedony. Very fine-grained reddish-grey quartzose sandstone, with traces of a slaty structure, resembling that of York Sound, and Cambridge Gulf, was found in the north-east end of this bay; and fine-grained greenstone, on the summit of the adjacent hills.

Several of these specimens are almost identical with those of Port Warrender; from which place Careening Bay is distant about sixty miles.

BAT ISLAND (Narrative volume 1) western entrance of Careening Bay. Quartz from thin veins, with particles of an adhering rock, probably chlorite-slate. Quartz, containing disseminated hematitic iron-ore and copper pyrites. Quartz crystals, with chalcedony, from nodules in amygdaloid. Quartz with specular iron ore. Greenstone, with chalcedony and copper pyrites. A decomposed stone, probably consisting of wacke. The specimens of trap-rocks from this place are from a cavern.

GREVILLE ISLAND, near the entrance of Prince Regent's River. Reddish, coarsely granular, siliceous sandstone; in horizontal strata, intersected by veins of crystallized quartz.*

(*Footnote. Narrative volume 2.)

HALF-WAY BAY, within Prince Regent's River on the west of the entrance, near Greville Island. Hornblende rock ? nearly agreeing with that of Pobassoo's Island, on the north-west of the Gulf of Carpentaria (see above). Calcedony, apparently from nodules in amygdaloid. Greenish quartz, approaching to heliotrope. Red, somewhat slaty jasper, mixed with quartz and chalcedony, and containing specular iron ore.

The specimens from this place much resemble some of those from Sotto i Sassi, in the Val di Fassa in the Tyrol, which I have seen in the collection of Mr. Herschel; and which consist of reddish jasper with chalcedony, and a greenish flinty stone, like heliotrope, the whole belonging to the trap-formation.

POINT CUNNINGHAM, east of south from Cape Leveque, and about one hundred and fifty miles south-west of Prince Regent's River. Very compact and fine-grained reddish granular quartz, with a glistening lustre, and flat conchoidal fracture. This stone, though so compact in the recent fracture, has distinct traces of stratification on the decomposed surface, which is of a dull reddish hue. Bright red ferruginous granular quartz (Eisen-kiesel ?) with a glistening lustre, and a somewhat porous texture. A specimen of the soil of the hills at Cygnet Bay, consists of very fine reddish-yellow quartzose sand. A large rounded pebble, consisting of ferruginous granular quartz, of a dark purplish-brown colour, and considerable density, was found here; near a fireplace of the natives, by whom it is used for making their hatchets; with a fragment of a calcareous incrustation, like that of the west coast hereafter mentioned.

The next specimens in Captain King's collection--a space of more than three hundred miles on this coast not having been examined by him--are from MALUS ISLAND, in Dampier's Archipelago (see Narrative volume 1) they consist of fine-grained greenstone, and what appears to be a basaltic rock, of amygdaloidal structure.

DIRK HARTOG'S ISLAND, west of Shark's Bay. A compound of rather fine-grained translucent quartzose sand, cemented by carbonate of lime, of various shades of reddish and yellowish grey. This stone has in some places the structure of a breccia; the angles of the imbedded fragments, which are from half an inch to two inches in diameter, being very distinct--but in other parts, the fracture exhibits the appearance of roundish nodules, composed of concentric shells--or bags as it were, of calcareous matter, which vary in colour, and are filled with a mixture of the same substance and quartzose sand: and the spaces between these nodules are likewise occupied by a similar compound.*

(*Footnote. The following description given by the French naturalists of the rocks at Bernier's Islands, was probably taken from a large suite of specimens; and M. Peron states (1 page 204) that it is strictly applicable to all the adjacent parts of the continent, and of the islands that were examined by the French voyagers:

Le sable du rivage (de l'ile Bernier) est quartzeux, mele d'une grande proportion de debris calcaires fortement attenues. La substance de l'ile meme se compose, dans ses couches inferieures, d'un gres calcaire coquillier, tantot blanchatre, tantot rougeatre, depose par couches horizontales, dont l'epaisseur varie de 2 a 8 decimetres (7 a 11 pouces) et qui toutes etant tres uniformes dans leur prolongement, pourroient offrir a la maconnerie des pierres de construction naturellement taillees.

Les coquilles incrustees dans ces massifs des roches sont presque toutes univalves; elles apartiennent plus particulierement au genre Natice de M. de Lamarck, et ont les plus grands rapports avec l'espece de Natice qui se trouve vivante au pied de ces rochers. Elles sont sans doute petrifiees depuis bien des siecles, car, outre qu'il est tres difficile de les retirer intactes du milieu de ces gres, tant leur adhesion avec eux est intime, on les observe encore a plus de 50 metres (150 pieds) au dessus du niveau actuel de la mer.

Quelque regularite que ces bancs puissent affecter dans leur disposition generale, ils ne sont cependant pas tous homogenes dans leur substance; il est sur-tout une variete de ces roches plus remarquable par sa structure. Ce sont des galets calcaires, agreges dans une terre sablonneuse ocracee, qui leur est tellement adherente, qu'on ne sauroit detruire cette espece de gangue sans les briser eux memes. Tous ces galets affectent la forme globlueuse, et se composent d'un grand nombre de zones concentriques, qui se developpent autour d'un noyau central d'un gres scintillant et brunatre. Ces diverses couches ont a peine quelques millimitres d'epaisseur, et affectent des nuances agreables, qui varient depuis le rouge-fonce jusqu'au jaune-clair. La disposition generale de cette breche lui donne donc quelques rapports grossiers avec le granit globuleux de l'ile de Corse; et, par ses couches rubanees, concentriques, elle a quelque chose de l'aspect des Agathes-Onyx...Les bancs de gres divers dont je viens de parler, constituent, a bien dire, la masse entiere du pays qui nous occupe, etc. (Volume 1 page 110. See also Freycinet page 187.)

(*Footnote. See Mr. Koenig's Paper. Philosophical Transactions volume 104 1814 page 107 etc.)

(**Footnote. Captain King informs me that the soundings in this part of the coast bring up a very fine quartzose-sand like that cemented in the breccia.)

ROTTNEST ISLAND, about four hundred and fifty miles south of Dirk Hartog's Island. Indistinct specimens containing numerous fragments of shells, in a calcareous cement; the substance of these shells has at first sight the appearance of chalcedony, and is harder than ordinary carbonate of lime.

The characters of the shells in Captain King's specimens from this place are indistinct; but the specimens at the Jardin du Roi, which, there is reason to suppose, have come from this part of the coast, contain shells of several species, belonging among others to the genera, corbula, chama, cardium, porcellanea, turbo, cerithium. M. Prevost, to whom I am indebted for this account, observes that notwithstanding the recent appearance of the shells, the beds which contain them are stated to occur at a considerable height above the sea: and he remarks that the aspect of the rock is very like that of the shelly deposits of St. Hospice, near Nice.

KING GEORGE'S SOUND, on the south coast, east of south from Cape Leeuwin. Beautifully white and fine quartzose sand, from the sea-beach. Yellowish grey granite, from Bald-head. Two varieties of a calcareous rock, of the same nature with that of Dirk Hartog's Island; consisting of particles of translucent quartzose sand, united by a cement of yellowish or cream-coloured carbonate of lime, which has a flat conchoidal and splintery fracture, and is so hard as to yield with difficulty to the knife. In this compound, there are not any distinct angular fragments, as in the stone of Dirk Hartog's Islands; but the calcareous matter is very unequally diffused.

A third form in which this recent calcareous matter appears, is that of irregular, somewhat tortuous, stem-like bodies, with a rugged sandy surface, and from half an inch to an inch in diameter; the cross fracture of which shows that they are composed of sand, cemented by carbonate of lime, either uniformly mixed throughout, or forming a crust around calcareous matter of a spongy texture; in which latter case they have some resemblance to the trunks or roots of trees. A mass, which seems to have been of this description, is stated to have come from a height of about two hundred and fifty feet above the sea, at Bald-head, on the South Coast of Australia. These specimens, however, do not really exhibit any traces of organic structure; and so nearly resemble the irregular stalactitical concretions produced by the passage of calcareous or ferruginous solutions through sand* that they are probably of the same origin; indeed the central cavity of the stalactite still remains open in some of the specimens of this kind from Sweer's Island in the Gulf of Carpentaria. The specimens from Madeira, presented to the Geological Society by Mr. Bowdich, and described in his notes on that island,** appear upon examination to be of the same character. But there is no reason to suppose that the trunks of trees, as well as other foreign substances, may not be thus incrusted, since various foreign bodies, even of artificial production, have been so found. Professor Buckland has mentioned a specimen of concreted limestone from St. Helena, which contains the recent shell of a bird's egg;*** and M. Peron states that, in the concretional limestone rock of the South Coast of New Holland, the trunks of trees occur, with the vegetable structure so distinct as to leave no doubt as to their nature.****

(*Footnote. Tubular concretions of ferruginous matter, irregularly ramifying through sand, like the roots of trees, are described by Captain Lyon as occurring in Africa. Lyon's Travels Appendix page 65.)

(**Footnote. Excursions in Madeira 1825 page 139, 140; and Bull. des Sciences Naturelles volume 4 page 322.)

(***Footnote. Geological Transactions volume 5 page 479.)

(****Footnote. Peron 2 page 75.)

INSTRUCTIONS FOR COLLECTING GEOLOGICAL SPECIMENS.

It so often happens that specimens sent from distant places, by persons unpractised in geology, fail to give the instruction which is intended, from the want of attention to a few necessary precautions, that the following directions may perhaps be useful to some of those, into whose hands these pages are likely to fall. It will be sufficient to premise, that two of the principal objects of geological inquiry, are, to determine, first, the nature of the MATERIALS of which the earth is composed; and, secondly, the relative ORDER in which these materials are disposed with respect to each other.

1. Specimens of rocks ought not, in general, to be taken from loose pieces, but from large masses in their native place, or which have recently fallen from their natural situation.

2. The specimens should consist of the stone unchanged by exposure to the elements, which sometimes alter the characters to a considerable distance from the surface. Petrifactions, however, are often best distinguishable in masses somewhat decomposed; and are thus even rendered visible, in many cases, where no trace of any organized body can be discerned in the recent fracture.

3. The specimens ought not to be too small. A convenient size is about three inches square, and about three-quarters of an inch, or less, in thickness.

4. It seldom happens that large masses, even of the same kind of rock, are uniform throughout any considerable space; so that the general character is collected, by geologists who examine rocks in their native places, from the average of an extensive surface: a collection ought therefore to furnish specimens of the most characteristic varieties; and THE MOST SPLENDID SPECIMENS ARE, IN GENERAL, NOT THE MOST INSTRUCTIVE. Where several specimens are taken in the same place, a series of numbers should be added to the note of their locality.

5. One of the most advantageous situations for obtaining specimens, and examining the relations of rocks, is in the sections afforded by cliffs on the seashore; especially after recent falls of large masses. It commonly happens that the beds thus exposed are more or less inclined; and in this case, if any of them be inaccessible at a particular point, the decline of the strata will frequently enable the collector to supply himself with the specimens he wishes for, within a short distance. Thus, in Sketch 4, which may be supposed to represent a cliff of considerable height, the observer being situated at a, the beds b, c, d, though inaccessible at that place, may be examined with ease and security, where they successively come down to the shore, at b prime, c prime, and d prime.

6. To examine the interior of an unknown country, more skill and practice are required: the rocks being generally concealed by the soil, accumulations of sand, gravel, etc., and by the vegetation of the surface. But the strata are commonly disclosed in the sides of ravines, in the beds of rivers and mountain-streams; and these, especially where they cross the direction of the strata, and be made, by careful examination, to afford instructive sections.

7. Among the distinctive circumstances of the strata, the remains of organized bodies, shells, corals, and other zoophytes, the bones and teeth of animals, fossil wood, and the impressions of vegetable stems, roots, or leaves, etc., are of the greatest importance; affording generally the most marked characters of the strata in which they occur. These should, therefore, be particularly sought after, and their relative abundance or rarity in different situations noticed. The petrified bodies should, if possible, be kept united with portions of the rock or matrix in which they are found; and where they are numerous, in sand, clay, or any moist or friable matrix, it is in general better to retain a large portion of the whole mass, to be examined afterwards, than to attempt their separation at the time of collecting.

8. The loose materials which are found above the solid rocks, in the form of gravel, silt, rolled pebbles, etc., should be carefully distinguished from the solid strata upon which they repose. And the more ancient of these loose materials, found on the sides or summits of hills, etc., should be distinguished from the recent mud, sand, and gravel, brought down by land-floods, or rivers. The bones and teeth of animals are not unfrequently found in gravel of the former description; and the collection of these remains from distant quarters of the globe, is an object of the greatest interest to geology.

9. Besides a note of the locality, there ought, if possible, to accompany every specimen, a short notice of its geological circumstances; as:

Whether it be found in large shapeless masses, or in strata?

If in strata, what are the thickness, inclination to the horizon, and direction with respect to the compass, of the beds? [If these cannot be measured, an estimate should always be recorded, while the objects are in view.] Are they uniform in dip and direction? curved, or contorted? continuous, or interrupted by fissures or veins?

Is the whole cliff, or mass of strata in sight, of uniform composition? or does it consist of different kinds of stone?

If the strata be different, what is the order in which they are placed above each other successively?

10. A label, distinctly written, should accompany every specimen, stating its native place, its relative situation, etc., etc. And these labels should be connected with the specimens immediately, on the spot where they are found. This injunction may appear to be superfluous; but so much valuable information has been lost to geology from the neglect of it, that every observer of experience will acknowledge its necessity; and it is, perhaps, in practice one of the most difficult to adhere to.

11. A sketch of a coast or cliff, however slight, frequently conveys more information respecting the disposition and relations of rocks, than the longest memorandum. If numbers, denoting the situation of the specimens collected, be marked upon such sketches, much time may be saved at the moment of collecting. But in all such cases, the memorandum should be looked over soon afterwards, and labels distinctly explaining their situation, etc., be attached to the specimens themselves.

12. The specimens should be so packed, that the surfaces may be defended from exposure to air, moisture, and friction: for which purpose, if strong paper cannot be obtained, dry moss, or straw, or leaves, may be used with advantage. Where paper is used for wrapping the specimens, they are best secured by fastening the envelope with sealing-wax.

Lastly, The collector must not be discouraged, nor be prevented from collecting, by finding that the place which he may chance to visit in a remote situation, has not a striking appearance, or the rocks within his view a very interesting character; since it frequently, and even commonly, happens, that facts and specimens, in themselves of very little importance, become valuable by subsequent comparison; so that scarcely any observation, if recorded with accuracy, will be thrown away.

The Instruments required by the geological traveller will vary, according to the acquirements and specific objects of the individual. The most essential are:

The Hammer (Sketch 5); which, for general purposes, may be of the form here represented: