parts approaching a grey syenite and in others becoming greenish by an admixture of chlorite. No very distinct stratification is visible here, but further south, towards the Swallow Tail Light, this is more apparent, the beds becoming at the same time less crystalline and associated with considerable beds of fine grained indurated shales. These beds near the centre of the peninsula exhibit a series of low undulations, but as the last named headland is approached their inclination becomes greater and their dip (to the northward) more uniform. They are here associated with altered gray sandstones and some thin beds of impure limestone, and are traversed by veins of heavy spar, holding small quantities of galena and copper pyrites. Considerable masses of diorite are occasionally met with along this shore, and at one point broad lenticular sheets of fine-grained flesh-red felsite.

In the small indentation known as Spragg's or Pette's Cove, a somewhat abrupt transition in the character of the rocks may be seen, for while the Eastern side of this Cove, forming the promontory of the Swallow-Tail Light, has the uniform grey colour and other features alluded to in the preceding remarks, the Western exhibits a most marked contrast, being conspicuous, even for a considerable distance, from the almost chalky whiteness of its low cliffs. This appearance is due to the peculiar weathering of a thick mass of pale liver-grey micaceous slates, which here form the shore, dipping northward (N. 50° E.) at an angle of 50°. Between these slates and the grey rocks first alluded to, thinner beds of grey micaceous shales and impure pyritiferous dolomites are poorly exposed along the beach, and near its northern side fine-grained fissile black shales. The contact between these two sets of rocks is obscure, but so far as I could judge, they appear to be conformable and to be connected together by intermediate. gradations. That those last enumerated form a single series is evident from their frequent alternation, as may be well seen on either shore of the promontory separating Spragg's or Pette's from Flag's Cove. The pale grey unctuous or nacreous slates and black slates are here associated with hard grey somewhat slaty sandstones (including thin layers of black slate) and coarse grey and purplish sandy shales, many of the beds being more or less filled with veins of brown spar or dolomite, and the whole several times repeated by faulting. On that side of the peninsula looking toward's Flag's Cove, some of the finer purplish beds are

VOL. VI.

D

No. 1.

ribbanded, and exhibit numerous and abrupt corrugations. Their general dip, however, is northward (N. 20° to 30° E.).

The section afforded by the peninsula above described between Whale Cove and Flag's Cove may be taken as affording a fair representation of the whole metamorphic belt of the Eastern shore of Grand Manan, strata similar in their general aspect to those alluded to being met with at various points along the latter as well as in the adjacent islands. With these, however, are some beds but imperfectly represented or altogether wanting in the area first al

luded to.

Along the Western side of Flag's Cove the metamorphic belt is greatly reduced in breadth, being confined to a narrow strip along the shore and to a series of ledges mostly covered by the tide. The rocks exposed here are coarse greenish-grey somewhat chloritic sandstones, with strong slaty cleavage, having numerous imbedded nodules of mixed quartz and spar from a quarter of an inch to two inches in length, besides numerous little crystalline spots resembling spathic iron. Beds of precisely similar character may be seen on Big Duck Island several miles to the southward, being here associated with blueish-grey somewhat unctuous feld. spathic schists, and pale grey dolomites. There are also upon this island (beneath the first named beds) white weathering nacreous slates with green and purple shales, the whole porphyritic as above with numerous little rhombohedral crystals, and more or less filled with sparry nodules. The white nacreous slate has been examined by Dr. Sterry Hunt, who finds it to consist of an admixture of silicious matter with a hydrous potash-mica, containing only traces of magnesia and iron. The imbedded crystalline spar is a triple carbonate, consisting of carbonate of iron 39.20, carbonate of magnesia 40.40, carbonate of lime 20.40 = 100.00. These beds have here a breadth of over 100 rods, and rest upon coarse purplish-grey quartzose grits, the general dip being westerly at an angle of about 60°. The whole series is evidently the same as that of Flag's Cove, with which these rocks are connected through those of Long Island.

Along the road connecting Flag's Cove with Woodward's Cove and Grand Harbour the rocks met with are chiefly grey lightweathering felsites and coarse grey feldspathic sandstones and slates, much broken and seamed with quartz, and sometimes becoming true quartzites. Similar rocks form Nantucket Island, near the entrance of the last named Cove, but here the quartzite of

a nearly pure white colour, rises into a conspicuous ridge, having a low westerly dip (W. 10° N. > 20°), and a breadth of over one-eighth of a mile.* It rests upon soft dark green shales, and with these extends through the length of the island, reappearing in Gull Rock and in Chalk Cove towards the northern end of Ross Island. This large island, as well as the shore from Woodward's Cove to Grand Harbour, I had not leisure to examine, but in passing around the shore of the last named haven, and thence along the beach to Red Head, was enabled to obtain a fair idea of the structure of the remaining portion of the metamorphic belt.

Along the western side of Grand Harbour the strata exposed to view, near its head, are greenish-grey chloritic and grey feldspathic schists, and grey feldspathic sandstones, with a strong slaty cleavage and variable dip; while nearer its entrance there are with these fine-grained greenish and purplish rocks, containing epidote, and more or less amygdaloidal. A few beds of finegrained grey felsite, or felsite with an admixture of quartz and chlorite, or talcoid mica, are intercalated with these. The dip here is N. 60 to 70 E. > 30° to 50°. Similar beds, but with a larger proportion of shales, sometimes purple and sometimes dark green with films of chlorite, skirt the shore westward of the entrance of the harbour, forming the promontories of Mike's and Oxnard's Points. A long curving beach, broken at intervals by beds of yellowish-grey slaty felsite, separates this point from a line of low bluffs running out and terminating in the promontory of Red Head. The beds exposed in these bluffs bear much resemblance to some of those described above, as seen upon the shores of Flag's Cove, towards the head of the island. They are grey and bluish-grey (sometimes purple or black) fine-grained beds, conspicuously ribbon-banded and thrown into innumerable sharp corrugations. With these are grey feldspathic sandstones, coated with specular iron, and coarse green chloritic beds, similarly plicated, but having a general northerly dip at an angle of about 30°. Towards the head the finer beds predominate, becoming soft and rubly and conspicuously stained with red oxide.

• The quartz rock is here associated with dark grey fissile shales and green chloritic schists, dipping S. 40 W. > 30. It has almost the aspect of a white quartz vein. Similar rocks form conspicuous cliffs on the western side of Whitehead Island but have not been visited by me.

of iron, having evidently suggested the name by which the promontory is known. The latter, as stated in a preceding paragraph, marks the southern limit of the metamorphic belt, the contact of this with the Mesozoic traps being well exposed in a small cove upon its western side. The red slates last described, dipping northward, here meet and are covered by a coarse conglomerate made of dark trap pebbles, which in turn underlies and passes into coarsely columnar trap, these being the first of a succession of such beds forming the northern shore of Benson's Cove.

Several small groups of islands lie to the south and east of the promontory last described. These I have only partially examined, but as they exhibit some features not met with upon the mainland, they may be briefly alluded to here. The first of these groups is that known as the Wood Islands, distinguished as the Inner and Outer Wood Islands. Upon the former the rocks bear much resemblance to those seen along the western side of Grand Harbour, described above. They are rather fine grained rocks, of bright green, red, and purple colours, often diversified with paler bands and blotches, and more or less filled with amygdules of calcite and epidote. These beds are associated with sandstones (and some conglomerates) of deep red and purplish red colours, sometimes finely banded and alternating with thinner beds of pale grey feldspathic schist and impure dolomite. These rocks, with occasional masses of trap, form nearly the whole of the wes tern side of the island, as well as its northern extremity, their dip being somewhat variable, but where most regular, about N. 20 to 60° E. > 40°. The sandstones are at some points very curiously and conspicuously marked by narrow veins (one-fourth of an inch wide) of fibrous calcite or satin spar, which fill short lenticular cavities arranged in parallel and overlapping lines, at right angles to the bedding of the rock.

Outer Wood Island, at the only point seen by me (on its eastern side), is composed of hard greenish-grey silico feldspathic rocks, with very obscure stratification.

The group of the Three Islands lies to the south and east of that last described, and with the exception of Gannet Rock, on which a light-house is built, is the most southerly of the chain of islands about the entrance of the Bay of Fundy. On the larger island of this group, known as Kent's Island, are beds of crystalline limestone. They are mostly light coloured but mottled with shades of green, grey, or pink, and are rendered impure by a

considerable admixture of quartz. The associated rocks are pale grey light weathering feldspathic grits, somewhat granitoid in aspect, grey feldspathic quartzites, and greenish and purplish altered schists, all much broken and disturbed. The other islands in this group I have not examined.

With reference to the age of the metamorphic rocks described above, I can only add to the various conjectures already made by other authors. In doing so, however, I may say that I have had the advantage of being able to compare them directly with the formations of the mainland, and thus of arriving at a more probable estimate of their true position than is likely to be obtained from the mere study of the rocks themselves. Of the recognized formations in New Brunswick, they bear no resemblance to either the Laurentian, Primordial, Upper Silurian, or Carboniferous. They are equally unlike the Devonian rocks, so far as these have been clearly determined on palæontological evidence. They do, however, bear much resemblance to an assemblage of strata met with at various points along the southern coast of the Province as well as in the interior, and to a portion of which a Devonian age has been assigned in earlier publications. The rocks in question, embracing like those of Grand Manan a series of coarse red sediments, grey clay slates, chloritic slates and grits, with some limestones and dolomites, were at some points found to rest upon undoubted Devonian beds, and were for this reason referred to that horizon. It is not yet certain that such is not their age, but a careful study of the district having shewn the existence therein of several great faults and overlaps, it is possible that the beds in question, notwithstanding the superposition referred to, are really much more ancient. If this is the case, there can be no doubt that they are to be looked upon as a subordinate division of the great Huronian series, to the other members of which, as recognized in southern New Brunswick, they bear much resemblance. The metamorphic rocks of Grand Manan have been compared by Dr. Dawson (from Prof. Verrill's description) with what has been termed the Kingston series on the mainland of the Province. They differ from these latter in some respects, but as these Kingston rocks are now also believed to be a subdivision of the Huronian system, (and not Upper Silurian, as at one time supposed) this comparison may be taken as an additional argument in support of the view here advocated. Prof. Verrill has suggested that possibly more than one group