The Cruise of the Betsey (3)

Home Up Autobiography First Impressions Scenes and Legends Tales and Sketches Leading Articles Miscellanea Main Index Site Search


 

[Previous Page]


CHAPTER VII.


I BREAKFASTED in the travellers' room with three gentlemen from Edinburgh; and then, accompanied by a boy, whom I had engaged to carry my bag, set out to explore.  The morning was ominously hot and breathless; and while the sea lay moveless in the calm, as a floor of polished marble, mountain, and rock, and distant island, seemed tremulous all over, through a wavy medium of thick rising vapour.  I judged from the first that my course of exploration for the day was destined to terminate abruptly; and as my arrangements with Mr Swanson left me, for this part of the country, no second day to calculate upon, I hurried over deposits which in other circumstances I would have examined more carefully,—content with a glance.  Accustomed in most instances to take long aims, as Cuddy Headrig did, when he steadied his musket on a rest behind the hedge, and sent his ball through Laird Oliphant's forehead, I had on this occasion to shoot flying; and so, selecting a large object for a mark, that I might run the less risk of missing, I strove to acquaint myself rather with the general structure of the district than with the organisms of its various fossiliferous beds.

    The long narrow island of Rasay lies parallel to the coast of Skye, like a vessel laid along a wharf, but drawn out from it, as if to suffer another vessel of the same size to take her berth between; and on the eastern shores of both Skye and Rasay we find the same Oolitic deposits tilted up at nearly the same angle.  The section presented on the eastern coast of the one is nearly a duplicate of the section presented on the eastern coast of the other.  During one of the severer frosts of last winter I passed along a shallow pond, studded along the sides with boulder stones.  It had been frozen over; and then, from the evaporation so common in protracted frosts, the water had shrunk, and the sheet of ice which had sunk down over the central portion of the pond exhibited what a geologist would term very considerable marks of disturbance among the boulders at the edges.  Over one sharp-backed boulder there lay a sheet tilted up like the lid of a chest half-raised; and over another boulder immediately behind it there lay another up-tilted sheet, like the lid of a second half-open chest; and in both sheets, the edges, lying in nearly parallel lines, presented a range of miniature cliff's to the shore.  Now, in the two up-tilted ice-sheets of this pond I recognised a model of the fundamental Oolitic deposits Rasay and Skye.  The mainland of Scotland had its representative in the crisp snow-covered shore of the pond, with its belt of faded sedges; the place of Rasay was indicated by the inner, that of Skye by the outer boulder; while the ice-sheets, with their shoreward-turned line of cliffs, represented the Oolitic beds, that turn to the mainland their dizzy range of precipices, varying from six to eight hundred feet in height, and then, sloping outwards and downwards, disappear under mountain wildernesses of overlying trap.  And it was along a portion of the range of cliff that forms the outermost of the two up-tilted lines, and which presents in this district of Skye a frontage of nearly twenty continuous miles to the long Sound of Rasay, that my to-day's course of exploration lay.  From the top of the cliff the surface slopes downwards for about two miles into the interior, like the half-raised chest-lid of my illustration sloping towards the hinges, or the up-tilted icetable of the boulder sloping towards the centre of the pond; and the depression behind forms a flat moory valley, full fifteen miles in length, occupied by a chain of dark bogs and treeless lochans.  A long line of trap-hills rises over it, in one of which, considerably in advance of the others, I recognised the Storr of Skye, famous among lovers of the picturesque for its strange group of mingled pinnacles and towers; while directly crossing into the valley from the Sound, and then running southwards for about two miles along its bottom, is the noble sea-arm, Loch Portree, in which, as indicated by the name (the King's Port) a Scottish king of the olden time, in his voyage round his dominions, cast anchor.  The opening of the loch is singularly majestic; the cliffs tower high on either side in graceful magnificence: but from the peculiar inward slope of the land, all within, as the loch reaches the line of the valley, becomes tame and low, and a black dreary moor stretches from the flat terminal basin into the interior.  The opening of Loch Portree is a palace gateway, erected in front of some homely suburb, that occupies the place which the palace itself should have occupied.

The Storr of Skye.


    There was, however, no such mixture of the homely and the magnificent in the route I had selected to explore.  It lay under the escarpment of the cliff; and I purposed pursuing it from Portree to Holm, a distance of about six miles, and then returning by the flat interior valley.  On the one hand rose a sloping rampart, full seven hundred feet in height, striped longitudinally with alternating bands of white sandstone and dark shale, and capped atop by a continuous coping of trap, that lacked not massy tower, and overhanging turret, and projecting sentry-box; while, on the other hand, spreading outwards in the calm from the line of dark trap-rocks below, like a mirror from its carved frame of black oak, lay the Sound of Rasay, with its noble background of island and main rising bold on the east, and its long mountain vista opening to the south.  The first fossiliferous deposit which gave me occasion this morning to use my hammer occurs near the opening of the loch, beside an old Celtic burying-ground, in the form of a thick bed of hard sandstone, charged with Belemnites,—a bed that must at one time have existed as a widely-spread accumulation of sand,—the bottom, mayhap, of some extensive bay of the Oolite, resembling the Loch Portree of the present day, in which eddy tides deposited the sand swept along by the tidal currents of some neighbouring sound, and which swarmed as thickly with Cephalopoda as the loch swarmed this day with minute purple-tinged Medusæ.  I found detached on the shore, immediately below this bed, a piece of calcareous fissile sandstone, abounding in small sulcated Terebratulæ, identical, apparently, with the Terebratula of a specimen in my collection from the inferior Oolite of Yorkshire.  A colony of this delicate Brachiopod must have once lain moored near this spot, like a fleet of long- prowed galleys at anchor, each one with its cable of many strands extended earthwards from the single dead-eye in its umbone. For a full mile after rounding the northern boundary of the loch, we find the immense escarpment composed from top to bottom exclusively of trap; but then the Oolite again begins to appear, and about two miles further on the section becomes truly magnificent,—one of the finest sections of this formation exhibited anywhere in Britain, perhaps in the world. In a ravine furrowed in the face of the declivity by the headlong descent of a small stream, we may trace all the beds of the system in succession, from the Cornbrash, an upper deposit of the Lower Oolite, down to the Lias, the for mation on which the Oolite rests. The only modifying circumstance to the geologist is, that though the sandstone beds run continuously along the cliff for miles together, distinct as the white bands in a piece of onyx, the intervening beds of shale are swarded over, save where we here and there see them laid bare in some abrupter acclivity or deeper watercourse. In the shale we find numerous minute Ammonites, sorely weathered; in the sandstone, Belemnites, some of them of great size; and dark carbonaceous markings, passing not unfrequently into a glossy cubical coal. At the foot of the cliff I picked up an ammonite of considerable size and well-marked character,—the Ammonites Murchisonœ, first discovered on this coast by Sir R. Murchison about fifteen years ago.  It measures, when full grown, from six to seven inches in diameter: the inner whorls, which are broadly visible, are ribbed; whereas the two, and sometimes the three outer ones, are smooth,—a marked characteristic of the species.  My specimen merely enabled me to examine the peculiarities of the shell just a little more minutely than I could have done in the pages of Sowerby; for such was its state of decay, that it fell to pieces in my hands.  I had now come full in view of the rocky island of Holm, when the altered appearance of the heavens led me to deliberate, just as I was warming in the work of exploration, whether, after all, it might not be well to scale the cliffs, and strike directly on the inn.  It was nearly three o'clock; the sky had been gradually darkening since noon, as if one thin covering of gauze after another had been drawn over it; hill and island had first dimmed and then disappeared in the landscape; and now the sun stood up right over the fast-contracting vista of the Sound, round and lightless as the moon in a haze; and the downward cataract-like streaming of the gray vapour on the horizon showed that there the rain had already broken, and was descending in torrents.  We had been thirsty in the hot sun, and had found the springs few and scanty; but the boy now assured me, in very broken English, that we were to get a great deal more water than would be good for us, and that it might be advisable to get out of its way.  And so, climbing to the top of the cliffs, along a water-course, we reached the ridge, just as the fog came rolling downwards from the peaked brow of the Storr into the flat moory pea valley, and the melancholy lochans roughened and darkened in the rain.  We were both particularly wet ere we reached Portree.

Sir Roderick Murchison, Scottish geologist.
(1792-1871)

    In exploring our Scotch formations, I have had frequent occasion, in Ross, Sutherland, Caithness, and now once more in Skye, to pass over ground described by Sir R. Murchison; and in every instance have I found myself immensely his debtor.  His descriptions possess the merit of being true: they are simple outlines often, that leave much to be filled up by after discovery; but, like those outlines of the skilful geographer that fix the place of some island or strait, though they may not entirely define it, they always indicate the exact position in the scale of the formations to which they refer.  They leave a good deal to be done in the way of mapping out the interior of a deposit, if I may so speak; but they leave nothing to be done in the way of ascertaining its place.  The work accomplished is bona fide work,—actual, solid, not to be done over again,—work such as could be achieved in only the school of Dr William Smith, the father of English Geology.  I have found much to admire, too, in the sections of Sir R. Murchison.  His section of this part of the coast, for example, strikes from the extreme northern part of Skye to the island of Holm, thence to Scrapidale in Rasay, thence along part of the coast of Scalpa, thence direct through the middle of Pabba, and thence to the shore of the Bay of Laing.  The line thus taken includes, in regular sequence in the descending order, the whole Oolitic deposits of the Hebrides, from the Cornbrash, with its overlying freshwater outliers of mayhap the Weald, down to where the Lower Lias rests on the primary red sandstones of Sleat.  It would have cost M'Culloch less exploration to have written a volume than it must have cost Sir R. Murchison to draw this single line; but the line once drawn, is work done to the hands of all after explorers.  I have followed repeatedly in the track of another geologist, of, however, a very different school, who explored, at a comparatively recent period, the deposits of not a few of our Scotch counties.  But his labours, in at least the fossiliferous formations, seem to have accomplished nothing for Geology,—I am afraid, even less than nothing.  So far as they had influence at all, it must have been to throw back the science.  A geologist who could have asserted only three years ago ("Geognostical Account of Banffshire," 1842), that the Old Red Sandstone of Scotland forms merely "a part of the great coal deposit," could have known marvellously little of the fossils of the one system, and nothing whatever of those of the other.  Had he examined ere he decided, instead of deciding without any intention of examining, he would have found that, while both systems abound in organic remains, they do not possess, in Scotland at least, a single species in common, and that even their types of being, viewed in the group, are essentially distinct.

    The three Edinburgh gentlemen whom I had met at breakfast were still in the inn.  One of them I had seen before, as one of the guests at a Wesleyan soiree, though I saw he failed to remember that I had been there as a guest too.  The two other gentlemen were altogether strangers to me.  One of them,—a man on the right side of forty, and a superb specimen of the powerful, six-feet-two-inch Norman Celt,—I set down as a scion of some old Highland family, who, as the broadsword had gone out, carried on the internal wars of the country with the formidable artillery of Statute and Decision.  The other, a gentleman more advanced in life, I predicated to be a Highland proprietor, the uncle of the younger of the two,—a man whose name, as he had an air of business about him, occurred, in all probability, in the Almanac, in the list of Scotch advocates.  Both were of course high Tories,—I was quite sure of that,—zealous in behalf of the Establishment, though previous to the Disruption they had not cared for it a pin's point,—and prepared to justify the virtual suppression of the toleration laws in the case of the Free Church.  I was thus decidedly guilty of what old Dr More calls a pro-sopolepsia,—i.e. of the crime of judging men by their looks.  At dinner, however, we gradually ate ourselves into conversation: we differed, and disputed, and agreed, and then differed, disputed, and agreed again.  I found first, that my chance companions were really not very high Tories; and then, that they were not Tories at all; and then, that the younger of the two was very much a Whig, and the more advanced in life,—strange as the fact might seem,—very considerably a Presbyterian Whig; and finally, that this latter gentleman, whom I had set down as an intolerant Highland proprietor, was a respected writer to the signet, a Free Church elder in Edinburgh; and that the other, his equally intolerant nephew, was an Edinburgh advocate, of vigorous talent, much an enemy of all oppression, and a brother contributor of my own to one of the Quarterlies.  Of all my surmisings regarding the stranger gentlemen, only two points held true,—they were both gentlemen of the law, and both had Celtic blood in their veins.  The evening passed pleasantly; and I can now recommend from experience, to the hapless traveller who gets thoroughly wet thirty miles from a change of dress, that some of the best things he can resort to in the circumstances are, a warm room, a warm glass, and agreeable companions.

    On the morrow I behoved to return to isle Ornsay, to set out on the following day, with my friend the minister, for Rum, where he purposed preaching on the Sabbath.  To have lost a day would have been to lose the opportunity of exploring the island, perhaps for ever; and, to make all sure, I had taken a seat in the mail gig, from the postman who drives it, ere going to bed, on the morning of my arrival; and now, when it drove up, I went to take my place in it.  The post-master of the village, a lean, hungry-looking man, interfered to prevent me.  I had secured my seat, I said, two days previous.  Ah, but I had not secured it from him.  "I know nothing of you, I replied; but I secured it from one who deemed himself authorized to receive the fare; was he so?"  "Yes."  "Could you have received it?" "No."  "Show me a copy of your regulations."  "I have no copy of regulations; but I have given the place in the gig to another."  "Just so; and what say you, postman?"  "That you took the place from me, and that he has no right to give a place to any one: I carry the Portree letters to him, but he has nothing to do with the passengers."  A person present, the proprietor or stabler of the horse, I believe, also interfered on the same side; but what Carlyle terms the "gigmanity" of the postmaster was all at stake,—his whole influence in the mail-gig of Portree; and so he argued, and threatened withal, and, what was the more serious part of the business, the person he had given the seat to had taken possession of the gig; and so we had to compound the matter by carrying a passenger additional.  The incident is scarce worth relating; but the postmaster was so vehement and terrible, so defiant of us all,—post, stabler, and simple passenger,—and so justly impressed with the importance of being postmaster of Portree, that, as I am in the way of describing rare specimens at any rate, I must refer to him among the rest, as if he had been one of the minor carnivore of a Skye deposit,—a cuttlefish, that preyed on the weaker molluscs, or a hungry polypus, terrible among the animalculæ.

    We drove heavily, and had to dismount and walk afoot over every steeper acclivity; but I carried my hammer, and only grieved that in some one or two localities the road should have been so level.  I regretted it in especial on the southern and eastern side of Loch Sligachan, where I could see from my seat, as we drove past, the dark blue rocks in the water-courses on each side the road, studded over with that characteristic shell of the Lias, the Gryphœa incurva, and that the dry-stone fences in the moor above exhibit fossils that might figure in a museum.  But we rattled by.  At Broadford, twenty-five miles from Portree, and nine miles from isle Ornsay, I partook of a hospitable meal in the house of an acquaintance; and in little more than two hours after was with my friend the minister at isle Ornsay.  The night wore pleasantly by.  Mrs Swanson, a niece of the late Dr Smith of Campbelton, so well known for his Celtic researches and his exquisite translations of ancient Celtic poetry, I found deeply versed in the legendary lore of the Highlands.  The minister showed me a fine specimen of Pterichthys which I had disinterred for him, out of my first discovered fossiliferous deposit of the Old Red Sandstone, exactly thirteen years before, and full seven years ere I had introduced the creature to the notice of Agassiz.  And the minister's daughter, a little chubby girl of three summers, taking part in the general entertainment, strove to make her Gaelic sound as like English as she could, in my especial behalf.  I remembered, as I listened to the unintelligible prattle of the little thing, unprovided with a word of English, that just eighteen years before, her father had had no Gaelic, and wondered what he would have thought, could he have been told, when he first sat down to study it, the story of his island charge in Eigg, and his Free Church yacht the Betsey.  Nineteen years before, we had been engaged in beating over the Eathie Lias together, collecting Belemnites, Ammonites, and fossil wood, and striving in friendly emulation the one to surpass the other in the variety and excellence of our specimens.  Our leisure hours were snatched, at the time, from college studies by the one, from the mallet by the other: there were few of them that we did not spend together, and that we were not mutually the better for so spending.  I at least owe much to these hours,—among other things, views of theologic truth, that determined the side I have taken in our ecclesiastical controversy.  Our courses at an after period lay diverse; the young minister had greatly more important business to pursue than any which the geologic field furnishes; and so our amicable rivalry ceased early.  In the words in which an English poet addresses his brother,—the clergyman who sat for the picture in the "Deserted Village,"—my friend "entered on a sacred office, where the harvest is great and the labourers are few, and left to me a field in which the labourers are many, and the harvest scarce worth carrying away."

    Next day at noon we weighed anchor, and stood out for Rum, a run of about twenty-five miles.  A kind friend had, we found, sent aboard in our behalf two pieces of rare antiquity,—rare anywhere, but especially rare in the lockers of the Betsey,—in the agreeable form of two bottles of semifossil Madeira,—Madeira that had actually existed in the grape exactly half a century before, at the time when Robespierre was startling Paris from its propriety, by mutilating at the neck the busts of other people, and multiplying casts and medals of his own; and we found it, explored in moderation, no bad study for geologists, especially in coarse weather, when they had got wet and somewhat fatigued.  It was like Landlord Boniface's ale, mild as milk, had exchanged its distinctive flavour as Madeira for a better one, and filled the cabin with fragrance every time the cork was drawn.  Old observant Homer must have smelt some such liquor somewhere, or he could never have described so well the still more ancient and venerable wine with which wily Ulysses beguiled one-eyed Polypheme:—


                                        Unmingled wine,
Mellifluous, undecaying, and divine,
Which now, some ages from his race concealed,
The hoary sire in gratitude revealed. * * *
Scarce twenty measures from the living stream
To cool one cup sufficed: the goblet crowned,
Breathed aromatic fragrances around.


    Winds were light and variable.  As we reached the middle of the sound opposite Armadale, there fell a dead calm; and the Betsey, more actively idle than the ship manned by the Ancient Mariner, dropped sternwards along the tide, to the dull music of the flapping sail.  The minister spent the day in the cabin, engaged with his discourse for the morrow; and I, that he might suffer as little from interruption as possible, mis-spent it upon the deck.  I tried fishing with the yacht's set of lines, but there were no fish to bite,—got into the boat, but there were no neighbouring islands to visit,—and sent half a dozen pistol-bullets after a shoal of porpoises, which, coming from the Free Church yacht, must have astonished the fat sleek fellows pretty considerably, but did them, I am afraid, no serious damage.  As the evening began to close gloomy and gray, a tumbling swell came heaving in right ahead from the west; and a bank of cloud, which had been gradually rising higher and darker over the horizon in the same direction, first changed its abrupt edge atop for a diffused and broken line, and then spread itself over the central heavens.  The calm was evidently not to be a calm long; and the minister issued orders that the gaff-topsail should be taken down, and the storm jib bent; and that we should lower our topmast, and have all tight and ready for a smart gale a-head.  At half-past ten, however, the Betsey was still pitching to the swell, with not a breath of wind to act on the diminished canvass, and with but the solitary circumstance in her favour, that the tide ran no longer against her, as before.  The cabin was full of all manner of creakings; the close lamp swung to and fro over the head of my friend; and a refractory Concordance, after having twice travelled from him along the entire length of the table, flung itself pettishly upon the floor.  I got into my snug bed about eleven; and at twelve, the minister, after poring sufficiently over his notes, and drawing the final score, turned into his.  In a brief hour after, on came the gale, in a style worthy of its previous hours of preparation; and my friend,—his Saturday's work in his ministerial capacity well over when he had completed his two discourses,—had to begin the Sabbath morning early as the morning itself began, by taking his stand at the helm, in his capacity of skipper of the Betsey.  With the prospect of the services of the Sabbath before him, and after working all Saturday to boot, it was rather hard to set him down to a midnight spell at the helm, but he could not be wanted at such a time, as we had no other such helmsman aboard.  The gale, thickened with rain, came down, shrieking like a maniac, from off the peaked hills of Rum, striking away the tops of the long ridgy billows that had risen in the calm to indicate its approach, and then carrying them in sheets of spray aslant the furrowed surface, like snow-drift hurried across a frozen field.  But the Betsey, with her storm-jib set, and her mainsail reefed to the cross, kept her weather bow bravely to the blast, and gained on it with every tack.  She had been the pleasure yacht, in her day, of a man of fortune, who had used, in running south with her at times as far as Lisbon, to encounter, on not worse terms than the stateliest of her neighbours in the voyage, the swell of the Bay of Biscay; and she still kept true to her old character, with but this drawback, that she had now got somewhat crazy in her fastenings, and made rather more water in a heavy sea than her one little pump could conveniently keep under.  As the fitful gust struck her headlong, as if it had been some invisible missile hurled at us from off the hill-tops, she stooped her head lower and lower, like old stately Hardyknute under the blow of the "King of Norse," till at length the lee chain-plate rustled sharp through the foam; but, like a staunch Free Churchwoman, the lowlier she bent, the more steadfastly did she hold her head to the storm.  The strength of the opposition served but to speed her on all the more surely to the desired haven.  At five o'clock in the morning we cast anchor in Loch Scresort,—the only harbour of Rum in which a vessel can moor,—within two hundred yards of the shore, having, with the exception of the minister, gained no loss in the gale.  He, luckless man, had parted from his excellent sou-wester; a sudden gust had seized it by the flap, and hurried it away far to the lee.  He had yielded it to the winds, as he had done the temporalities, but much more unwillingly, and less as a free agent.  Should any conscientious mariner pick up anywhere in the Atlantic a serviceable ochre-coloured sou-wester, not at all the worse for the wear, I give him to wit that he holds Free Church property, and that he is heartily welcome to hold it, leaving it to himself to consider whether a benefaction to its full value, deducting salvage, is not owing, in honour, to the Sustentation Fund.

    It was ten o'clock ere the more fatigued aboard could muster resolution enough to quit their beds a second time; and then it behoved the minister to prepare for his Sabbath labours ashore.  The gale still blew in fierce gusts from the hills, and the rain pattered like small shot on the deck.  Loch Scresort, by no means one of our finer island lochs, viewed under any circumstances, looked particularly dismal this morning.  It forms the opening of a dreary moorland valley, bounded on one of its sides, to the mouth of the loch, by a homely ridge of Old Red Sandstone, and on the other by a line of dark augitic hills, that attain, at the distance of about a mile from the sea; an elevation of two thousand feet.  Along the slopes of the sandstone ridge I could discern, through the haze, numerous green patches, that had once supported a dense population, long since "cleared off" to the backwoods of America, but not one inhabited dwelling; while along a black moory acclivity under the hills on the other side I could see several groups of turf cottages, with here and there a minute speck of raw-looking corn beside them, that, judging from its colour, seemed to have but a slight chance of ripening.  The hill-tops were lost in cloud and storm; and ever and anon as a heavier shower came sweeping down on the wind, the intervening hollows closed up their gloomy vistas, and all was fog and rhime to the water's edge.  Bad as the morning was, however, we could see the people wending their way, in threes and fours, through the dark moor, to the place of worship,—a black turf hovel, like the meeting-house in Eigg.  The appearance of the Betsey in the loch had been the gathering signal; and the Free Church islanders—three-fourths of the entire population—had all come out to meet their minister.

    On going ashore, we found the place nearly filled.  My friend preached two long energetic discourses, and then returned to the yacht, a "worn and weary man."  The studies of the previous day, and the fatigues of the previous night, added to his pulpit duties, had so fairly prostrated his strength, that the sternest teetotaller in the kingdom would scarce have forbidden him a glass of our fifty-year-old Madeira.  But even the fifty-year-old Madeira proved no specific in the case.  He was suffering under excruciating headache, and had to stretch himself in his bed, with eyes shut but sleepless, waiting till the fit should pass,—every pulse that beat in his temples a throb of pain.


 
CHAPTER VIII.


THE geology of the island of Rum is simple, but curious.  Let the reader take, if he can, from twelve to fifteen trap-hills, varying from one thousand to two thousand three hundred feet in height; let him pack them closely and squarely together, like rum-bottles in a case-basket; let him surround them with a frame of Old Red Sandstone, measuring rather more than seven miles on the side, in the way the basket surrounds the bottles; then let him set them down in the sea a dozen miles off the land,—and he shall have produced a second island of Rum, similar in structure to the existing one.  In the actual island, however, there is a defect in the inclosing basket of sandstone: the basket, complete on three of its sides, wants the fourth; and the side opposite to the gap which the fourth should have occupied is thicker than the two other sides put together.  Where I now write there is an old dark-coloured picture on the wall before me.  I take off one of the four bars of which the frame is composed,—the end-bar,—and stick it on to the end-bar opposite, and then the picture is fully framed on two of its sides, and doubly framed on a third, but the fourth side lacks framing altogether.  And such is the geology of the island of Rum.  We find the one loch of the island,—that in which the Betsey lies at anchor,—and the long withdrawing valley, of which the loch is merely a prolongation, occurring in the double sandstone bar: it seems to mark—to return to my illustration—the line in which the superadded piece of frame has been stuck on to the frame proper.  The origin of the island is illustrated by its structure: it has left its story legibly written, and we have but to run our eye over the characters and read.  An extended sea-bottom, composed of Old Red Sandstone, already tilted up by previous convulsions, so that the strata presented their edges, tier beyond tier, like roofing slate laid aslant on a floor, became a centre of Plutonic activity.  The molten trap broke through at various times, and presenting various appearances, but in nearly the same centre; here existing as an augitic rock, there as a syenite, yonder as a basalt or amygdaloid.  At one place it uptilted the sandstone; at another it overflowed it: the dark central masses raised their heads above the surface, higher and higher with every earthquake throe from beneath; till at length the gigantic Ben More attained to its present altitude of two thousand three hundred feet over the sea-level, and the sandstone, borne up from beneath like floating sea-wrack on the back of a porpoise, reached in long outside bands its elevation of from six to eight hundred.  And such is the piece of history, composed in silent but expressive language, and inscribed in the old geologic character, on the rocks of Rum.

    The wind lowered and the rain ceased during the night, and the morning of Monday was clear, bracing, and breezy.  The island of Rum is chiefly famous among mineralogists for its heliotropes or bloodstones; and we proposed devoting the greater part of the day to an examination of the hill of Scuir More, in which they occur, and which lies on the opposite side of the island, about eight miles from the mooring ground of the Betsey.  Ere setting out, however, I found time enough, by rising some two or three hours before breakfast, to explore the Red Sandstones on the southern side of the loch.  They lie in this bar of the frame, to return once more to my old illustration,—as if it had been cut out of a piece of cross-grained deal, in which the annular bands, instead of ranging lengthwise, ran diagonally from side to side; stratum leans over stratum, dipping towards the west at an angle of about thirty degrees; and as in a continuous line of more than seven miles there seem no breaks or repetitions in the strata, the thickness of the deposit must be enormous,—not less, I should suppose, than from six to eight thousand feet.  Like the Lower Old Red Sandstones of Cromarty and Moray, the red arenaceous strata occur in thick beds, separated from each other by bands of a grayish-coloured stratified clay, on the planes of which I could trace with great distinctness ripple markings; but in vain did I explore their numerous folds for the plates, scales, and fucoid impressions which abound in the gray argillaceous beds of the shores of the Moray and Cromarty Friths.  It would, however, be rash to pronounce them non-fossiliferous, after the hasty search of a single morning,—unpardonably so in one who had spent very many mornings in putting to the question the gray stratified beds of Ross and Cromarty, ere he succeeded in extorting from them the secret of their organic riches.

Louis Agassiz, Swiss-born geologist
(1807-73).

    We set out about half-past ten for Scuir More, through the Red Sandstone valley in which Loch Scresort terminates, with one of Mr Swanson's s people, a young active lad of twenty, for our guide.  In passing upwards for nearly a mile along the stream that falls into the upper part of the loch, and lays bare the strata, we saw no change in the character of the sand stone.  Red arenaceous beds of great thickness alternate with grayish-coloured bands, composed of a ripple-marked micaceous slate and a stratified clay.  For a depth of full three thousand feet, and I know not how much more,—for I lacked time to trace it further,—the deposit presents no other variety: the thick red bed of at least a hundred yards succeeds the thin gray band of from three to six feet, and is succeeded by a similar gray band in turn.  The ripple-marks I found as sharply relieved in some of the folds as if the wavy undulations to which they owed their origin had passed over them within the hour.  The comparatively small size of their alternating ridges and furrows give evidence that the waters beneath which they had formed had been of no very profound depth.  In the upper part of the valley, which is bare, trackless, and solitary, with a high monotonous sandstone ridge bounding it on the one side, and a line of gloomy trap-hills rising over it on the other, the edges of the strata, where they protrude through the mingled heath and moss, exhibit the mysterious scratchings and polishings now so generally connected with the glacial theory of Agassiz.  The scratchings run in nearly the line of the valley, which exhibits no trace of moraines; and they seem to have been produced rather by the operation of those extensively developed causes, whatever their nature, that have at once left their mark on the sides and summits of some of our highest hills, and the rocks and boulders of some of our most extended plains, than by the agency of forces limited to the locality.  They testify, Agassiz would perhaps say, not regarding the existence of some local glacier that descended from the higher grounds into the valley, but respecting the existence of the great polar glacier.  I felt, however, in this bleak and solitary hollow, with the grooved and polished platforms at my feet, stretching away amid the heath, like flat tombstones in a graveyard, that I had arrived at one geologic inscription to which I still wanted the key.  The vesicular structure of the traps on the one hand, identical with that of so many of our modern lavas,—the ripple-markings of the arenaceous beds on the other, indistinguishable from those of the sea-banks on our coasts,—the upturned strata and the overlying trap,—told all their several stories of fire, or wave, or terrible convulsion, and told them simply and clearly; but here was a story not clearly told.  It summoned up doubtful, ever-shifting visions,—now of a vast ice continent, abutting on this far isle of the Hebrides from the role, and trampling heavily over it, now of the wild rush of a turbid, mountain-high flood breaking in from the west, and hurling athwart the torn surface, rocks, and stones, and clay,—now of a dreary ocean rising high along the hills, and bearing onward with its winds and currents, huge icebergs, that now brushed the mountain-sides, and now grated along the bottom of the submerged valleys.  The inscription on the polished surfaces, with its careless mixture of groove and scratch, is an inscription of very various readings.

    We passed along a transverse hollow, and then began to ascend a hill-side, from the ridge of which the water sheds to the opposite shore of the island, and on which we catch our first glimpse of Scuir More, standing up over the sea, like a pyramid shorn of its top.  A brown lizard, nearly five inches in length, startled by our approach, ran hurriedly across the path; and our guide, possessed by the general Highland belief that the creature is poisonous, and injures cattle, struck at it with a switch, and cut it in two immediately behind the hinder legs.  The upper half, containing all that anatomists regard as the vitals, heart, brain, and viscera, all the main nerves, and all the larger arteries, lay stunned by the blow, as if dead; nor did it manifest any signs of vitality so long as we remained beside it; whereas the lower half, as if the whole life of the animal had retired into it, continued dancing upon the moss for a full minute after, like a young eel scooped out of some stream, and thrown upon the bank; and then lay wriggling and palpitating for about half a minute more.  There are few things more inexplicable in the province of the naturalist than the phenomenon of what may be termed divided life,—vitality broken into two, and yet continuing to exist as vitality in both the dissevered pieces.  We see in the nobler animals mere glimpses of the phenomenon,—mere indications of it, doubtfully apparent for at most a few minutes.  The blood drawn from the human arm by the lancet continues to live in the cup until it has cooled and begun to coagulate; and when head and body have parted company under the guillotine, both exhibit for a brief space such unequivocal signs of life, that the question arose in France during the horrors of the Revolution, whether there might not be some glimmering of consciousness attendant at the same time on the fearfully opening and shutting eyes and mouth of the one, and the beating heart and jerking neck of the other.  The lower we descend in the scale of being, the more striking the instances which we receive of this divisibility of the vital principle.  I have seen the two halves of the heart of a ray pulsating for a full quarter of an hour after they had been separated from the body and from each other.  The blood circulates in the hind leg of a frog for many minutes after the removal of the heart, which meanwhile keeps up an independent motion of its own.  Vitality can be so divided in the earthworm, that, as demonstrated by the experiments of Spalanzani, each of the severed parts carries life enough away to set it up as an independent animal; while the polypus, a creature of still more imperfect organization, and with the vivacious principle more equally diffused over it, may be multiplied by its pieces nearly as readily as a gooseberry bush by its slips.  It was sufficiently curious, however, to see, in the case of this brown lizard, the least vital half of the creature so much more vivacious, apparently, than the half which contained the heart and brain.  It is not improbable, however, that the presence of these organs had only the effect of rendering the upper portion which contained them more capable of being thrown into a state of insensibility.  A blow dealt one of the vertebrata of the head at once renders it insensible.  It is after this mode the fisherman kills the salmon captured in his wear, and a single blow, when well directed, is always sufficient; but no single blow has the same effect on the earthworm; and here it was vitality in the inferior portion of the reptile,—the earthworm portion of it, if I may so speak,—that refused to participate in the state of syncope into which the vitality of the superior portion had been thrown.  The nice and delicate vitality of the brain seems to impart to the whole system in connection with it an aptitude for dying suddenly,—a susceptibility of instant death, which would be wanting without it.  The heart of the rabbit continues to beat regularly long after the brain has been removed by careful excision, if respiration be artificially kept up; but if, instead of amputating the head, the brain be crushed in its place by a sudden blow of a hammer, the heart ceases its motion at once.  And such seemed to be the principle illustrated here.  But why the agonized dancing on the sward of the inferior part of the reptile?—why its after painful writhing and wriggling?  The young eel scooped from the stream, whose motions it resembled, is impressed by terror, and can feel pain; was it also impressed by terror, or susceptible of suffering?  We see in the case of both exactly the same signs,—the dancing, the writhing, the wriggling; but are we to interpret them after the same manner?  In the small red-headed earthworm divided by Spalanzani, that in three months got upper extremities to its lower part, and lower extremities, in as many weeks, to its upper part, the dividing blow must have dealt duplicate feelings,—pain and terror to the portion below, and pain and terror to the portion above, so far, at least, as a creature so low in the scale was susceptible of these feelings; but are we to hold that the leaping, wriggling tail of the reptile possessed in any degree a similar susceptibility?  I can propound the riddle, but who shall resolve it?  It may be added, that this brown lizard was the only recent saurian I chanced to see in the Hebrides, and that, though large for its kind, its whole bulk did not nearly equal that of a single vertebral joint of the fossil saurians of Eigg.  The reptile, since his deposition from the first place in the scale of creation, has sunk sadly in those parts: the ex-monarch has become a low plebeian.

    We came down upon the coast through a swampy valley, terminating in the interior in a frowning wall of basalt, and bounded on the south, where it opens to the sea, by the Scuir More.  The Scuir is a precipitous mountain, that rises from twelve to fifteen hundred feet direct over the beach.  M'Culloch describes it as inaccessible, and states that it is only among the debris at its base that its heliotropes can be procured; but the distinguished mineralogist must have had considerably less skill in climbing rocks than in describing them, as, indeed, some of his descriptions, though generally very admirable, abundantly testify.  I am inclined to infer from his book, after having passed over much of the ground which he describes, that he must have been a man of the type so well hit off by Burns in his portrait of Captain Grose,—round, rosy, short-legged, quick of eye but slow of foot, quite as indifferent a climber as Bailie Nicol Jarvie, and disposed at times, like the elderly gentleman drawn by Crabbe, to prefer the view at the hill-foot to the prospect from its summit.  I found little difficulty in scaling the sides of Scuir More for a thousand feet upwards,—in one part by a route rarely attempted before,—and in ensconcing myself among the blood stones.  They occur in the amygdaloidal trap of which the upper part of the hill is mainly composed, in great numbers, and occasionally in bulky masses; but it is rare to find other than small specimens that would be recognised as of value by the lapidary.  The inclosing rock must have been as thickly vesicular in its original state as the scoria of a glass-house; and all the vesicles, large and small, like the retorts and receivers of a laboratory, have been vessels in which some curious chemical process has been carried on.  Many of them we find filled with a white semi-translucent or opaque chalcedony; many more with a pure green earth, which, where exposed to the bleaching influences of the weather, exhibits a fine verdigris hue, but which in the fresh fracture is generally of an olive green, or of a brownish or reddish colour.  I have never yet seen a rock in which this earth was so abundant as in the amygdaloid of Scuir More.  For yards together in some places we see it projecting from the surface in round globules, that very much resemble green pease, and that occur as thickly in the inclosing mass as pebbles in an Old Red Sandstone conglomerate.  The heliotrope has formed among it in centres, to which the chalcedony seems to have been drawn, as if by molecular attraction.  We find a mass, varying from the size of a walnut to that of a man's head, occupying some larger vesicle or crevice of the amygdaloid, and all the smaller vesicles around it, for an inch or two, filled with what we may venture to term satellite heliotropes, some of them as minute as grains of wild mustard, and all of them more or less earthy, generally in proportion to their distance from the first formed heliotrope in the middle.  No one can see them in their place in the rock, with the abundant green earth all around, and the chalcedony, in its uncoloured state, filling up so many of the larger cavities, without acquiescing in the conclusion respecting the origin of the gem first suggested by Werner, and afterwards adopted and illustrated by M'Culloch.  The heliotrope is merely a chalcedony, stained in the forming with an infusion of green earth, as the coloured waters in the apothecary's window are stained by the infusions, vegetable and mineral, from which they derive their ornamental character.  The red mottlings which so heighten the beauty of the stone occur in comparatively few of the specimens of Scuir More.  They are minute jasperous formations, independent of the inclosing mass; and, from their resemblance to streaks and spots of blood, suggest the name by which the heliotrope is popularly known.  I succeeded in making up, among the crags, a set of specimens curiously illustrative of the origin of the gem.  One specimen consists of white, uncoloured chalcedony; a second, of a rich verdigrishued green earth; a third, of chalcedony barely tinged with green; a fourth, of chalcedony tinged just a shade more deeply; a fifth, tinged more deeply still; a sixth, of a deep green on one side, and scarce at all coloured on the other; and a seventh, dark and richly toned,—a true bloodstone,—thickly streaked and mottled with red jasper.  In the chemical process that rendered the Scuir More a mountain of gems there were two deteriorating circumstances, which operated to the disadvantage of its larger heliotropes: the green earth, as if insufficiently stirred in the mixing, has gathered, in many of them, into minute soft globules, like air-bubbles in glass, that render them valueless for the purposes of the lapidary, by filling them all over with little cavities; and in not a few of the others, an infiltration of lime, that refused to incorporate with the chalcedonic mass, exists in thin glassy films and veins, that, from their comparative softness, have a nearly similar effect with the impalpable green earth in roughing the surface under the burnisher.

    We find figured by M'Culloch, in his "Western Islands," the internal cavity of a pebble of Scuir More, which he picked up on the beach below, and which had been formed evidently within one of the larger vesicles of the amygdaloid.  He describes it as curiously illustrative of a various chemistry: the outer crust is composed of a pale-zoned agate, inclosing a cavity, from the upper side of which there depends a group of chalcedonic stalactites, some of them, as in ancient spar caves, reaching to the floor; and bearing on its under side a large crystal of carbonate of lime, that the longer stalactites pass through.  In the vesicle in which this hollow pebble was formed three consecutive processes must have gone on.  First, a process of infiltration coated the interior all around with layer after layer, now of one mineral substance, now of another, as a plasterer coats over the sides and ceiling of a room with successive layers of lime, putty, and stucco; and had this process gone on, the whole cell would have been filled with a pale-zoned agate. But it ceased, and a new process began. A chalcedonic infiltration gradually entered from above; and, instead of coating over the walls, roof, and floor, it hardened into a group of spear-like stalactites, that lengthened by slow degrees, till some of them had traversed the entire cavity from top to bottom.  And then this second process ceased like the first, and a third commenced.  An infiltration of lime took place; and the minute calcareous molecules, under the influence of the law of crystallization, built themselves up on the floor into a large smooth-sided rhomb, resembling a closed sarcophagus resting in the middle of some Egyptian cemetery.  And then, the limestone crystal completed, there ensued no after change.  As shown by some other specimens, however, there was a yet farther process: a pure quartzose deposition took place, that coated not a few of the calcareous rhombs with sprigs of rock-crystal.  I found in the Scuir More several cellular agates in which similar processes had gone on,—none of them quite so fine, however, as the one figured by M'Culloch; but there seemed no lack of evidence regarding the strange and multifarious chemistry that had been carried on in the vesicular cavities of this mountain, as in the retorts of some vast laboratory.  Here was a vesicle filled with green earth,—there a vesicle filled with calcareous spar,—yonder a vesicle crusted round on a thin chalcedonic shell with rock-crystal,—in one cavity an agate had been elaborated, in another a heliotrope, in a third a milk-white chalcedony, in a fourth a jasper.  On what principle, and under what direction, have results so various taken place in vesicles of the same rock, that in many instances occur scarce half an inch apart?  Why, for instance, should that vesicle have elaborated only green earth, and the vesicle separated from it by a partition barely a line in thickness, have elaborated only chalcedony?  Why should this chamber contain only a quartzose compound of oxygen and silica, and that second chamber beside it contain only a calcareous compound of lime and carbonic acid?  What law directed infiltrations so diverse to seek out for themselves vesicles in such close neighbourhood, and to keep, in so many instances, each to its own vesicle?  I can but state the problem,—not solve it.  The groups of heliotropes clustered each around its bulky centrical mass seem to show that the principle of molecular attraction may be operative in very dense mediæ,—in a hard amygdaloidal trap even; and it seems not improbable, that to this law, which draws atom to its kindred atom, as clansmen of old used to speed at the mustering signal to their gathering place, the various chemistry of the vesicles may owe its variety.

    I shall attempt stating the chemical problem furnished by the vesicles here in a mechanical form.  Let us suppose that every vesicle was a chamber furnished with a door, and that beside every door there watched, as in the draught doors of our coal-pits, some one to open and shut it, as circumstances might require.  Let us suppose further, that for a certain time an infusion of green earth pervaded the surrounding mass, and percolated through it, and that every door was opened to receive a portion of the infusion.  We find that no vesicle wants its coating of this earthy mineral.  The coating received, however, one-half the doors shut, while the other half remained agap, and filled with green earth entirely.  Next followed a series of alternate infusions of chalcedony, jasper, and quartz; many doors opened and received some two or three coatings, that form around the vesicles skull-like shells of agate, and then shut; a few remained open, and became as entirely occupied with agate as many of the previous ones had become filled with green earth.  Then an ample in fusion of chalcedony pervaded the mass.  Numerous doors again opened; some took in a portion of the chalcedony, and then shut; some remained open, and became filled with it; and many more that had been previously filled by the green earth opened their doors again, and the chalcedony pervading the green porous mass, converted it into heliotrope.  Then an infusion of lime took place.  Doors opened, many of which had been hitherto shut, save for a short time, when the green earth infusion obtained, and became filled with lime; other doors opened for a brief space, and received lime enough to form a few crystals.  Last of all, there was a pure quartzose infusion, and doors opened, some for a longer time, some for a shorter, just as on previous occasions.  Now, by mechanical means of this character,—by such an arrangement of successive infusions, and such a device of shutting and opening of doors,—the phenomena exhibited by the vesicles could be produced.  There is no difficulty in working the problem mechanically, if we be allowed to assume in our data successive infusions, well-fitted doors, and watchful door-keepers; and if any one can work it chemically,—certainly without door-keepers, but with such doors and such infusions as he can show to have existed,—he shall have cleared up the mystery of the Scuir More.  I have given their various cargoes to all its many vesicles by mechanical means, at no expense of ingenuity whatever.  Are there any of my readers prepared to give it to them by means purely chemical?

    There is a solitary house in the opening of the valley, over which the Scuir More stands sentinel,—a house so solitary, that the entire breadth of the island intervenes between it and the nearest human dwelling.  It is inhabited by a shepherd and his wife,—the sole representatives in the valley of a numerous population, long since expatriated to make way for a few flocks of sheep, but whose ranges of little fields may still be seen green amid the heath on both sides, for nearly a mile upwards from the opening.  After descending along the precipices of the Scuir, we struck across the valley, and, on scaling the opposite slope, sat down on the summit to rest us, about a hundred yards over the house of the shepherd.  He had seen us from below, when engaged among the bloodstones, and had seen, withal, that we were not coming his way; and, "on hospitable thoughts intent," he climbed to where we sat, accompanied by his wife, she bearing a vast bowl of milk, and he a basket of bread and cheese.  And we found the refreshment most seasonable, after our long hours of toil, and with a rough journey still before us.  It is an excellent circumstance, that hospitality grows best where it is most needed.  In the thick of men it dwindles and disappears, like fruits in the thick of a wood; but where man is planted sparsely, it blossoms and matures, like apples on a standard or espalier.  It flourishes where the inn and the lodging-house cannot exist, and dies out where they thrive and multiply.

    We reached the cross valley in the interior of the island about half an hour before sunset.  The evening was clear, calm, golden-tinted; even wild heaths and rude rocks had assumed a flush of transient beauty; and the emerald-green patches on the hill-sides, barred by the plough lengthwise, diagonally, and transverse, had borrowed an aspect of soft and velvety richness, from the mellowed light and the broadening shadows.  All was solitary.  We could see among the deserted fields the grass-grown foundations of cottages razed to the ground; but the valley, more desolate than that which we had left, had not even its single inhabited dwelling: it seemed as if man had done with it for ever.  The island, eighteen years before, had been divested of its inhabitants, amounting at the time to rather more than four hundred souls, to make way for one sheep-farmer and eight thousand sheep.  All the aborigines of Rum crossed the Atlantic; and at the close of 1828, the entire population consisted of but the sheep-farmer, and a few shepherds, his servants: the island of Rum reckoned up scarce a single family at this period for every five square miles of area which it contained.  But depopulation on so extreme a scale was found inconvenient; the place had been rendered too thoroughly a desert for the comfort of the occupant; and on the occasion of a clearing which took place shortly after in Skye, he accommodated some ten or twelve of the ejected families with sites for cottages, and pasturage for a few cows, on the bit of morass beside Loch Scresort, on which I had seen their humble dwellings.  But the whole of the once peopled interior remains a wilderness, without inhabitant,—all the more lonely in its aspect from the circumstance that the solitary valleys, with their plough-furrowed patches, and their ruined heaps of stone, open upon shores every whit as solitary as themselves, and that the wide untrodden sea stretches drearily around.  The armies of the insect world were sporting in the light this evening by millions; a brown stream that runs through the valley yielded an incessant poppling sound, from the myriads of fish that were ceaselessly leaping in the pools, beguiled by the quick glancing wings of green and gold that fluttered over them; along a distant hill-side there ran what seemed the ruins of a gray-stone fence, erected, says tradition, in a remote age, to facilitate the hunting of the deer; there were fields on which the heath and moss of the surrounding moorlands were fast encroaching, that had borne many a successive harvest; and prostrate cottages, that had been the scenes of christenings, and bridals, and blythe new-year's days,—all seemed to bespeak the place a fitting habitation for man, in which not only the necessaries, but also a few of the luxuries of life, might be procured; but in the entire prospect not a man nor a man's dwelling could the eye command.  The landscape was one without figures.  I do not much like extermination carried out so thoroughly and on system;—it seems bad policy; and I have not succeeded in thinking any the better of it though assured by the economists that there are more than people enough in Scotland still.  There are, I believe, more than enough in our workhouses,—more than enough on our pauper-rolls,—more than enough huddled up, disreputable, useless, and unhappy, in the miasmatic alleys and typhoid courts of our large towns; but I have yet to learn how arguments for local depopulation are to be drawn from facts such as these.  A brave and hardy people, favourably placed for the development of all that is excellent in human nature, form the glory and strength of a country; a people sunk into an abyss of degradation and misery, and in which it is the whole tendency of external circumstances to sink them yet deeper, constitute its weakness and its shame; and I cannot quite see on what principle the ominous increase which is taking place among us in the worse class, is to form our solace or apology for the wholesale expatriation of the better.  It did not seem as if the depopulation of Rum had tended much to any one's advantage.  The single sheep-farmer who had occupied the holdings of so many had been unfortunate in his speculations, and had left the island: the proprietor, his landlord, seemed to have been as little fortunate as the tenant, for the island itself was in the market; and a report went current at the time that it was on the eve of being purchased by some wealthy Englishman, who purposed converting it into a deer-forest.  How strange a cycle!  Uninhabited originally save by wild animals, it became at an early period a home of men, who, as the gray wall on the hill-side testified, derived, in part at least, their sustenance from the chase.  They broke in from the waste the furrowed patches on the slopes of tile valleys,—they reared. herds of cattle and flocks of sheep,—their number increased to nearly five hundred souls,—they enjoyed the average happiness of human creatures in the present imperfect state of being,—they contributed their portion of hardy and vigorous manhood to the armies of the country,—and a few of their more adventurous spirits, impatient of the narrow bounds which confined them, and a course of life little varied by incident, emigrated to America.  Then came the change of system so general in the Highlands; and the island lost all its original inhabitants, on a wool and mutton speculation, inhabitants, the descendants of men who had chased the deer on its hills five hundred years before, and who, though they recognised some wild island lord as their superior, and did him service, had regarded the place as indisputably their own.  And now yet another change was on the eve of ensuing, and the island was to return to its original state, as a home of wild animals, where a few hunters from the mainland might enjoy the chase for a month or two every twelvemonth, but which could form no permanent place of human abode.  Once more, a strange and surely most melancholy cycle!

    There was light enough left, as we reached the upper part of Loch Scresort, to show us a shoal of small silver-coated trout, leaping by scores at the effluence of the little stream along which we had set out in the morning on our expedition.  There was a net stretched across where the play was thickest; and we learned that the haul of the previous tide had amounted to several hundreds.  On reaching the Betsey, we found a pail and basket laid against the companion-head,—the basket containing about two dozen small trout,—the minister's unsolicited teind of the morning draught; the pail filled with razor-fish of great size.  The people of my friend are far from wealthy; there is scarce any circulating medium in Rum and the cottars in Eigg contrive barely enough to earn at the harvest in the Lowlands, money sufficient to clear with their landlord at rent-day.  Their contributions for ecclesiastical purposes make no great figure, therefore, in the lists of the Sustentation Fund.  But of what they have they give willingly and in a kindly spirit; and if baskets of small trout, or pailfuls of spout-fish, went current in the Free Church, there would, I am certain, be a percentage of both the fish and the mollusc, derived from the Small Isles, in the half-yearly sustentation dividends.  We found the supply of both,—especially as provisions were beginning to run short in the lockers of the Betsey,—quite deserving of our gratitude.  The razor-fish had been brought us by the worthy catechist of the island.  He had gone to the ebb in our special behalf, and had spent a tide in laboriously filling the pail with these "treasures hid in the sand;" thoroughly aware, like the old exiled Puritan, who eked out his meals in a time of scarcity with the oysters of New England, that even the razor-fish, under this head, is included in the promises.  There is a peculiarity in the razor-fish of Rum that I have not marked in the razor-fish of our eastern coasts. The gills of the animal, instead of bearing the general colour of its other parts, like those of the oyster, are of a deep green colour, resembling, when examined by the microscope, the fringe of a green curtain.

    We were told by John Stewart, that the expatriated inhabitants of Rum used to catch trout by a simple device of ancient standing, which preceded the introduction of nets into the island, and which, it is possible, may in other localities have not only preceded the use of the net, but may have also suggested it: it had at least the appearance of being a first beginning of invention in this direction.  The islanders gathered large quantities of heath, and then tying it loosely into bundles, and stripping it of its softer leafage, they laid the bundles across the stream on a little mound held down by stones, with the tops of the heath turned upwards to the current.  The water rose against the mound for a foot or eighteen inches, and then murmured over and through, occasioning an expansion among the hard elastic sprays.  Next a party of the islanders came down the stream, beating the banks and pools, and sending a still thickening shoal of trout before them, that, on reaching the miniature dam formed by the bundles, darted forward for shelter, as if to a hollow bank, and stuck among the slim hard branches, as they would in the meshes of a net.  The stones were then hastily thrown off,—the bundles pitched ashore,—the better fish, to the amount not unfrequently of several scores, secured,—and the young fry returned to the stream, to take care of themselves, and grow bigger.  We fared richly this evening, after our hard day's labour, on tea and trout; and as the minister had to attend a meeting of the Presbytery of Skye on the following Wednesday, we sailed next morning for Glenelg, whence he purposed taking the steamer for Portree.  Winds were light and baffling, and the currents, like capricious friends, neutralized at one time the assistance which they lent us at another.  It was dark night ere we had passed Isle Ornsay, and morning broke as we cast anchor in the Bay of Glenelg.  At ten o'clock the steamer heaved-to in the bay to land a few passengers, and the minister went on board, leaving me in charge of the Betsey, to follow him, when the tide set in, through the Kyles of Skye.


 
CHAPTER IX.


NO sailing vessel attempts threading the Kyles of Skye from the south in the face of an adverse tide.  The currents of Kyle Rhea care little for the wind-filled sail, and battle at times, on scarce unequal terms, with the steam-propelled paddle.  The Toward Castle this morning had such a struggle to force her way inwards as may be seen maintained at the door of some place of public meeting during the heat of some agitating controversy, when seat and passage within can hold no more, and a disappointed crowd press eagerly for admission from without.  Viewed from the anchoring place at Glenelg, the opening of the Kyle presents the appearance of the bottom of a landlocked bay;—the hills of Skye seem leaning against those of the mainland: and the tide-buffeted steamer looked this morning as if boring her way into the earth like a disinterred mole, only at a rate vastly slower.  First, however, with a progress resembling that of the minute-hand of a clock, the bows disappeared amid the heath, then the midships, then the quarter-deck and stern, and then, last of all, the red tip of the sun-brightened union jack that streamed gaudily behind.  I had at least two hours before me ere the Betsey might attempt weighing anchor; and, that they might leave some mark, I went and spent them ashore in the opening of Glenelg,—a gneiss district, nearly identical in structure with the district of Knock and Isle Ornsay.  The upper part of the valley is bare and treeless, but not such its character where it opens to the sea; the hills are richly wooded; and cottages and corn-fields, with here and there a reach of the lively little river, peep out from among the trees.  A group of tall roofless buildings, with a strong wall in front, form the central point in the landscape: these are the dismantled Bernera Barracks, built, like the line of forts in the great Caledonian Valley,--Fort George, Fort Augustus, and Fort William,--to overawe the Highlands at a time when the loyalty of the Highlander pointed to a king beyond the water; but all use for them has long gone by, and they now lie in dreary ruin,--mere sheltering places for the toad and the bat.  I found in a loose silt on the banks of the river, at some little distance below tide-mark, a bed of shells and coral, which might belong, I at first supposed, to some secondary formation, but which I ascertained, on examination, to be a mere recent deposit, not so old by many centuries as our last raised sea-beaches.  There occurs in various localities on these western coasts, especially on the shores of the island of Pabba, a sprig coral, considerably larger in size than any I have elsewhere seen in Scotland; and it was from its great abundance in this bed of silt that I was at first led to deem the deposit an ancient one.

Bernera Barracks, Glenelg, strategically located to control the crossing at Kylerhea, it was built to house a garrison of 200 soldiers.


    We weighed anchor about noon, and entered the opening of Kyle Rhea.  Vessel after vessel, to the number of eight or ten in all, had been arriving in the course of the morning, and dropping anchor, nearer the opening or farther away, each according to its sailing ability, to await the turn of the tide; and we now found ourselves one of the components of a little fleet, with some five or six vessels sweeping up the Kyle before us, and some three or four driving on behind.  Never, except perhaps in a Highland river big in flood, have I seen such a tide.  It danced and wheeled, and came boiling in huge masses from the bottom; and now our bows heaved abruptly round in one direction, and now they jerked as suddenly round in another; and, though there blew a moderate breeze at the time, the helm failed to keep the sails steadily full.  But whether our sheets bellied out, or flapped right in the wind's eye, on we swept in the tideway, like a cork caught during a thunder shower in one of the rapids of the High Street.  At one point the Kyle is little more than a quarter of a mile in breadth; and here, in the powerful eddie which ran along the shore, we saw a group of small fishing-boats pursuing a shoal of sillocks in a style that blent all the liveliness of the chase with the specific interest of the angle.  The shoal, restless as the tides among which it disported, now rose in the boilings of one eddie, now beat the water into foam amid the stiller dimplings of another.  The boats hurried from spot to spot wherever the quick glittering scales appeared.  For a few seconds rods would be cast thick and fast, as if employed in beating the water, and captured fish glanced bright to the sun; and then the take would cease, and the play rise elsewhere, and oars would flash out amain, as the little fleet again dashed into the heart of the shoal.  As the Kyle widened, the force of the current diminished, and sail and helm again became things of positive importance.  The wind blew a-head, steady though not strong; and the Betsey, with companions in the voyage against which to measure herself, began to show her paces.  First she passed one bulky vessel, then another: she lay closer to the wind than any of her fellows, glided more quickly through the water, turned in her stays like Lady Betty in a minuet; and, ere we had reached Kyle Akin, the fleet in the middle of which we had started were toiling far behind us, all save one vessel, a stately brig; and just as we were going to pass her too, she cast anchor, to await the change of the tide, which runs from the west during flood at Kyle Akin, as it runs from the east through Kyle Rhea.  The wind had freshened; and as it was now within two hours of full sea, the force of the current had somewhat abated; and so we kept on our course, tacking in scant room, however, and making but little way.  A few vessels attempted following us, but, after an inefficient tack or two, they fell back on the anchoring ground, leaving the Betsey to buffet the currents alone.  Tack followed tack sharp and quick in the narrows, with an iron-bound coast on either hand.  We had frequent and delicate turning: now we lost fifty yards, now we gained a hundred.  John Stewart held the helm; and as none of us had ever sailed the way before, I had the vessel's chart spread out on the companion-head before me, and told him when to wear and when to hold on his way,--at what places we might run up almost to the rock edge, and at what places it was safest to give the land a good offing.  Hurrah for the Free Church yacht Betsey! and hurrah once more!  We cleared the Kyle, leaving a whole fleet tide-bound behind us; and, stretching out at one long tack into the open sea, bore, at the next, right into the bay at Broadford, where we cast anchor for the night, within two hundred yards of the shore.  Provisions were running short; and so I had to make a late dinner this evening on some of the razor-fish of Rum, topped by a dish of tea.  But there is always rather more appetite than food in the country;--such, at least, is the common result under the present mode of distribution: the hunger overlaps and outstretches the provision; and there was comfort in the reflection, that with the razor-fish on which to fall back, it overlapped it but by a very little on this occasion in the cabin of the Betsey.  The steam-boat passed southwards next morning, and I was joined by my friend the minister a little before breakfast.

    The day was miserably bad: the rain continued pattering on the skylight, now lighter, now heavier, till within an hour of sunset, when it ceased, and a light breeze began to unroll the thick fogs from off the landscape, volume after volume, like coverings from off a mummy,--leaving exposed in the valley of the Lias a brown and cheerless prospect of dark bogs and of debris-covered hills, streaked this evening with downward lines of foam.  The seaward view is more pleasing.  The deep russet of the interior we find bordered for miles along the edge of the bay with a many-shaded fringe of green; and the smooth grassy island of Pabba lies in the midst, a polished gem, all the more advantageously displayed from the roughness of the surrounding setting.  We took boat, and explored the Lias in our immediate neighbourhood till dusk.  I had spent several hours among its deposits when on my way to Portree, and several hours more when on my journey across the country to the east coast; but it may be well, for the sake of maintaining some continuity of description, to throw together my various observations on the formation, as if made at one time, and to connect them with my exploration of Pabba, which took place on the following morning.  The rocks of Pabba belong to the upper part of the Lias; while the lower part may be found leaning to the south, towards the Red Sandstones of the Bay of Lucy.  Taking what seems to be the natural order, I shall begin with the base of the formation first.

    In the general indentation of the coast, in the opening of which the island of Pabba lies somewhat like a long green steam-boat at anchor, there is included a smaller indentation, known as the Bay or Cove of Lucy.  The central space in the cove is soft and gravelly; but on both its sides it is flanked by low rocks, that stretch out into the sea in long rectilinear tines, like the foundations of dry-stone fences.  On the south side the rocks are red; on the north they are of a bluish-gray colour; their hues are as distinct as those of the coloured patches in a map; and they represent geological periods that lie widely apart.  The red rocks we find laid down in most of our maps as Old Red, though I am disposed to regard them as of a much higher antiquity than even that ancient system; while the bluish-gray rocks are decidedly Liasic. [2]  The cove between represents a deep ditch-like hollow, which occurs in Skye, both in the interior and on the sea-shore, in the line of boundary betwixt the Red Sandstone and the Lias; and it "seems to have originated," says M'Culloch, "in the decomposition of the exposed parts of the formations at their junction."  "Hence," he adds, "from the wearing of the materials at the surface, a cavity has been produced, which becoming subsequently filled with rubbish, and generally covered over with a vegetable soil of unusual depth, effectually prevents a view of the contiguous parts."  The first strata exposed on the northern side are the oldest Liasic rocks any where seen in Scotland.  They are composed chiefly of greenish-coloured fissile sandstones and calciferous grits, in which we meet a few fossils, very imperfectly preserved.  But the organisms increase as we go on.  We see in passing, near a picturesque little cottage,--the only one on the shores of the bay,--a crag of a singularly rough appearance, that projects mole-like from the sward upon the beach, and then descending abruptly to the level of the other strata, runs out in a long ragged line into the sea.  The stratum, from two to three feet in thickness, of which it is formed, seems wholly built up of irregularly-formed rubbly concretions, just as some of the garden-walls in the neighbourhood of Edinburgh are built of the rough scoria of our glass-houses; and we find, on examination, that every seeming concretion in the bed is a perfectly formed coral of the genus Astrea.  We have arrived at an entire bed of corals, all of one species.  Their surfaces, wherever they have been washed by the sea, are of great beauty: nothing can be more irregular than the outline of each mass, and yet scarce anything more regular than the sculpturings on every part of it.  We find them fretted over with polygons, like those of a honeycomb, only somewhat less mathematically exact, and the centre of every polygon contains its many-rayed star.  It is difficult to distinguish between species in some of the divisions of corals: one Astrea, recent or extinct, is sometimes found so exceedingly like another of some very different formation or period, that the more modern might almost be deemed a lineal descendant of the more ancient species.  With an eye to the fact, I brought with me some characteristic specimens of this Astrea [3] of the Lower Lias, which I have ranged side by side with the Astreæ of the Oolite I had found so abundant a twelvemonth before in the neigbourhood of Helmsdale.  In some of the hand specimens, that present merely a piece of polygonal surface, bounded by fractured sides, the difference is not easily distinguishable: the polygonal depressions are generally smaller in the Oolitic species, and shallower in the Liasic one; but not unfrequently these differences disappear, and it is only when compared in the entire unbroken coral that their specific peculiarities acquire the necessary prominence.  The Oolitic Astrea is of much greater size than the Liasic one: it occurs not unfrequently in masses of from two to three feet in diameter; and as its polygons are tubes that converge to the footstalk on which it originally formed, it presents in the average outline a fungous-like appearance; whereas in the smaller Liasic coral, which rarely exceeds a foot in diameter, there is no such general convergency of the tubes; and the form in one piece, save that there is a certain degree of flatness common to all, bears no resemblance to the form in another.  Some of the recent Astreæ are of great beauty when inhabited by the living zoophites whose skeleton framework they compose.  Every polygonal star in the mass is the house of a separate animal, that, when withdrawn into its cell, presents the appearance of a minute flower, somewhat like a daisy stuck flat to the surface, and that, when stretched out, resembles a small round tower, with a garland of leaves bound round it atop for a cornice.  The Astrea viridis, a coral of the tropics, presents on a ground of velvety brown myriads of deep green florets, that ever and anon start up from the level in their tower-like shape, contract and expand their petals, and then, shrinking back into their cells, straightway become florets again.  The Lower Lias presented in one of its opening scenes, in this part of the world, appearances of similar beauty widely spread.  For miles together,--we know not how many,--the bottom of a clear shallow sea was paved with living Astreæ: every irregular rock-like coral formed a separate colony of polypora, that, when in motion, presented the appearance of continuous masses of many-coloured life, and when at rest, the places they occupied were more thickly studded with the living florets than the richest and most flowery piece of pasture the reader ever saw, with its violets or its daisies.  And mile beyond mile this scene of beauty stretched on through the shallow depths of the Liasic sea.  The calcareous framework of most of the recent Astreæ are white; but in the species referred to,--the Astrea viridis,--it is of a dark-brown colour.  It is not unworthy of remark, in connection with these facts, that the Oolitic Astrea of Helmsdale occurs as a white, or, when darkest, as a cream-coloured petrifaction; whereas the Liasic Astrea of Skye is invariably of a deep earthy hue.  The one was probably a white, the other a dingy-coloured coral.

    The Liasic bed of Astreæ existed long enough here to attain a thickness of from two to three feet.  Mass rose over mass,--the living upon the dead,--till at length, by a deposit of mingled mud and sand.--the effect, mayhap, of some change of currents, induced we know not how,--the innumerable polypedes of the living surface were buried up and killed, and then, for many yards, layer after layer of a calciferous grit was piled over them.  The fossils of the grit are few and ill preserved; but we occasionaly find in it a coral similar to the Astrea of the bed below, and, a little higher up, in an impure limestone, specimens, in rather indifferent keeping, of a genus of polypifer which somewhat resembles the Turbinolia of the Mountain Limestone.  It presents in the cross section the same radiated structure as the Turbinolia fungites, and nearly the same furrowed appearance in the longitudinal one; but, seen in the larger specimens, we find that it was a branched coral, with obtuse forky boughs, in each of which, it is probable, from their general structure, there lived a single polype.  It may have been the resemblance which these bear, when seen in detached branches, to the older Caryophyllia, taken in connection with the fact that the deposit in which they occur rests on the ancient Red Sandstone of the district, that led M'Culloch to question whether this fossiliferous formation had not nearly as clear a claim to be regarded as an analogue of the Carboniferous Limestone of England as of its Lias; and hence he contented himself with terming it simply the Gryphite Limestone.  Sir R. Murchison, whose much more close and extensive acquaintance with fossils enabled him to assign to the deposit its true place, was struck, however, with the general resemblance of its polypifers to "those of the Madreporite Limestone of the Carboniferous series."  These polypifers occur in only the Lower Lias of Skye. [4]  I found no corals in its higher beds, though these are charged with other fossils, more characteristic of the formation, in vast abundance.  In not a few of the middle strata, composed of a mud-coloured fissile sandstone, the gryphites lie as thickly as currants in a Christmas cake; and as they weather white, while the stone in which they are embedded retains its dingy hue, they somewhat remind one of the whitelead tears of the undertaker mottling a hatchment of sable.  In a fragment of the dark sandstone, six inches by seven, which I brought with me, I reckon no fewer than twenty-two gryphites; and it forms but an average specimen of the bed from which I detached it.  By far the most abundant species is that not inelegant shell so characteristic of the formation, the Gryphœa incurva.  We find detached specimens scattered over the beach by hundreds, mixed up with the remains of recent shells, as if the Gryphœa incurva were a recent shell too.  They lie, bleached white by the weather, among the valves of defunct oysters and dead buccinidæ; and, from their resemblance to lamps cast in the classic model, remind one, in the corners where they have accumulated most thickly, of the old magician's stock in trade, who wiled away the lamp of Aladdin from Aladdin's simple wife.  The Gryphœa obliquita and Gryphœa M'Cullochii also occur among these middle strata of the Lias, though much less frequently than the other.  We, besides, found in them at least two species of Pecten, with two species of Terebratula,--the one smooth, the other sulcated; a bivalve resembling a Donax; another bivalve, evidently a Gervillia, though apparently of a species not yet described; and the ill-preserved rings of large Ammonites, from ten inches to a foot in diameter.  Towards the bottom of the bay the fossils again become more rare, though they re-appear once more in considerable abundance as we pass along its northern side; but in order to acquaint ourselves with the upper organisms of the formation, we have to take boat and explore the northern shores of Pabba.  The Lias of Skye has its three distinct groups of fossils: its lower coraline group, in which the Astrea described is most abundant; its middle group, in which the Gryphœa incurva occurs by millions; and its upper group, abounding in Ammonites, Nautili, Pinnæ, and Serpulæ.

Nautilus

Gryphœa incurva

Turbinolia


     Friday made amends for the rains and fogs of its disagreeable predecessor: the morning rose bright and beautiful, with just wind enough to fill, and barely fill, the sail, hoisted high, with miser economy, that not a breath might be lost; and, weighing anchor, and shaking out all our canvass, we bore down on Pabba to explore.  This island, so soft in outline and colour, is formidably fenced round by dangerous reefs; and, leaving the Betsey in charge of John Stewart and his companion, to dodge on in the offing, I set out with the minister in our little boat, and landed on the north-eastern side of the island, beside a trap-dyke that served us as a pier.  He would be a happy geologist who, with a few thousands to spare, could call Pabba his own.  It contains less than a square mile of surface; and a walk of little more than three miles and a half along the line where the waves break at high water brings the traveller back to his starting point; and yet, though thus limited in area, the petrifactions of its shores might of themselves fill a museum.  They rise by thousands and tens of thousands on the exposed planes of its sea-washed strata, standing out in bold relief, like sculpturings on ancient tombstones, at once mummies and monuments,--the dead, and the carved memorials of the dead.  Every rock is a tablet of hieroglyphics, with an ascertained alphabet; every rolled pebble a casket, with old pictorial records locked up within.  Trap-dykes, beyond comparison finer than those of the Water of Leith, which first suggested to Hutton his theory, stand up like fences over the sedimentary strata, or run out like moles far into the sea.  The entire island, too, so green, rich, and level, is itself a specimen illustrative of the effect of geologic formation on scenery.  We find its nearest neighbour,--the steep, brown, barren island of Longa, which is composed of the ancient Red Sandstone of the district,--differing as thoroughly from it in aspect as a bit off granite differs from a bit of clay-slate; and the whole prospect around, save the green Liasic strip that lies along the bottom of the Bay of Broadford, exhibits, true to its various components, Plutonic or sedimentary, a character of picturesque roughness or bold sublimity.  The only piece of smooth, level England, contained in the entire landscape, is the fossil-mottled island of Pabba.  We were first struck, on landing this morning, by the great number of Pinnæ embedded in the strata,--shells varying from five to ten inches in length,--one species of the common flat type, exemplified in the existing Pinna sulcata, and another nearly quadrangular, in the cross section, like the Pinna lanceolata of the Scarborough limestone.  The quadrangular species is more deeply crisped outside than the flat one.  Both species bear the longitudinal groove in the centre, and, when broken across, are found to contain numerous smaller shells,--Terebratulæ of both the smooth and sulcated kinds, and a species of minute smooth Pecten resembling the Pecten demissus, but smaller.  The Pinnæ, ere they became embedded in the original sea-bottom, long since hardened into rock around them, were, we find, dead shells, into which, as into the dead open shells of our existing beacnes, smaller shells were washed by the waves.  Our recent Pinnæ are all sedentary shells, some of them full two feet in length, fastened to their places on their deep-sea floors by flowing silky byssi,--cables of many strands,--of which beautiful pieces of dress, such as gloves and hose, have been manufactured.  An old French naturalist, the Abbe Le Pluche, tells us that "the Pinna with its fleshy tongue" (foot),--a rude inefficient-looking implement for work so nice,--"spins such threads as are more valuable than silk itself, and with which the most beautiful stuffs that ever were seen have been made by the Sicilian weavers."  Gloves made of the oyssus of recent Pinnæ may be seen in the British Museum. Associated with the numerous Pinnæ of Pabba, we found a delicately-formed Modiola, a small Ostrya, Plagiostoma, Terebratula, several species of Pectens, a triangular univalve resembling a Trochus, innumerable groups of Serpulæ, and the star-like joints of Pentacrinites.  The Gryphæ are also abundant, occurring in extensive beds; and Belemnites of various species lie as thickly scattered over the rock as if they had been the spindles of a whole kingdom thrown aside in consequence of some such edict framed to put them down as that passed by the father of the Sleeping Beauty.  We find, among the detached masses of the beach, specimens of Nautilus, which, though rarely perfect, are sufficiently so to show the peculiarities of the shell; and numerous Ammonites project in relief from almost every weathered plane of the strata.  These last shells, in the tract of shore which we examined, are chiefly of one species,--the Ammonites spinatus,--one of which, considerably broken, the reader may find figured in Sowerby's "Mineral Conchology," from a specimen brought from Pabba sixteen years ago by Sir R. Murchison.  It is difficult to procure specimens tolerably complete.  We find bits of outer rings existing as limestone, with every rib sharply preserved, but the rest of the fossil lost in the shale.  I succeeded in finding but two specimens that show the inner whorls.  They are thickly ribbed; and the chief peculiarity which they exhibit, not so directly indicated by Mr Sowerby's figure, is, that while the ribs of the outer whorl are broad and deep, as in the Ammonites obtusus, they suddenly change their character, and become numerous and narrow in the inner whorls, as in the Ammonites communis.

    The tide began to flow, and we had to quit our explorations, and return to the Betsey.  The little wind had become less, and all the canvass we could hang out enabled us to draw but a sluggish furrow.  The stern of the Betsey "wrought no buttons" on this occasion; but she had a good tide under her keel; and ere the dinner-hour we had passed, through the narrows of Kyle Akin.  The village of this name was designed by the late Lord M'Donald for a great sea-port town; but it refused to grow; and it has since become a gentleman in a small way, and does nothing.  It forms, however, a handsome group of houses, pleasantly situated on a flat green tongue of land, on the Skye side, just within the opening of the Kyle; and there rises on an eminence beyond it a fine old tower, rent open, as if by an earthquake, from top to bottom, which forms one of the most picturesque objects I have almost ever seen in a landscape.  There are bold hills all around, and rocky islands, with the ceaseless rush of tides in front; while the cloven tower, rising high over the shore, is seen, in threading the Kyles, whether from the south or north, relieved dark against the sky, as the central object in the vista.  We find it thus described by the Messrs Anderson of Inverness, in their excellent "Guide Book,"--by far the best companion of the kind with which the traveller who sets himself to explore our Scottish Highlands can be provided.  "Close to the village of Kyle Akin are the ruins of an old square keep, called Castle Muel or Maoil, the walls of which are of a remarkable thickness.  It is said to have been built by the daughter of a Norwegian king, married to a Mackinnon or Macdonald, for the purpose of levying an impost on all vessels passing the Kyles, excepting, says the tradition, those of her own country.  For the more certain exaction of this duty, she is reported to have caused a strong chain to be stretched across from shore to shore; and the shot in the rocks to which the terminal links were attached is still pointed out."  It was high time for us to be home.  The dinner hour came; but, in meet illustration of the profound remark of Trotty-Veck, not the dinner.  We had been in a cold Moderate district, whence there came no half-dozens of eggs, or whole dozens of trout, or pailfuls of razor-fish and in which hard cabin biscuit cost us sixpence per pound.  And now our stores were exhausted, and we had to dine as we best could, on our last half-ounce of tea, sweetened by our last quarter of a pound of sugar.  I had marked, however, a dried thornback hanging among the rigging.  It had been there nearly three weeks before, when I came first aboard, and no one seemed to know for how many weeks previous; for, as it had come to be a sort of fixture in the vessel, it could be looked at without being seen.  But necessity sharpens the discerning faculty, and on this pressing occasion I was fortunate enough to see it.  It was straightway taken down, skinned, roasted, and eaten; and, though rather rich in ammonia,--a substance better suited to form the food of the organisms that do not unite sensation to vitality, than organisms so high in the scale as the minister and his friend,--we came deliberately to the opinion, that, on the whole, we could scarce have dined so well on one of Major Bellenden's jack-boots,--"so thick in the soles," according to Jenny Dennison, "for by being tough in the upper leather."  The tide failed us opposite the opening of Loch Alsh; the wind, long dying, at length died out into a dead calm; and we cast anchor in ten fathoms water, to wait the ebbing current that was to carry us through Kyle Rhea.

Castle Maoil


    The ebb-tide set in about half an hour after sunset; and in weighing anchor to float down the Kyle,--for we still lacked wind to sail down it,--we brought up from below, on one of the anchor-flukes, an immense bunch of deep-sea tangle, with huge soft fronds and long slender stems, that had lain flat on the rocky bottom, and had here and there thrown out roots along its length of stalk, to attach itself to the rock, in the way the ivy attaches itself to the wall.  Among the intricacies of the true roots of the bunch, if one may speak of the true roots of an alga, I reckoned up from eighteen to twenty different forms of animal life,--Flustræ, Sertulariæ, Serpulæ, Anomiæ, Modiolæ, Astarte, Annelida, Crustacea and Radiata.  Among the Crustaceans I found a female crab of a reddish-brown colour, considerably smaller than the nail of my small finger, but fully grown apparently, for the abdominal flap was loaded with spawn; and among the Echinoderms, a brownish-yellow sea-urchin about the size of a pistol bullet, furnished with comparatively large but thinly-set spines.  There is a dangerous rock in the Kyle Rhea, the Caileach stone, on which the Commissioners for the Northern Lighthouses have stuck a bit of board, about the size of a pot-lid, which, as it is known to be there, and as no one ever sees it after sunset, is really very effective, considering how little it must have cost the country, in wrecking vessels.  I saw one of its victims, the sloop of an honest Methodist, in whose bottom the Caileach had knocked out a hole, repairing at Isle Ornsay; and I was told, that if I wished to see more, I had only just to wait a little.  The honest Methodist, after looking out in vain for the bit of board, was just stepping into the shrouds, to try whether he could not see the rock on which the bit of board is placed, when all at once his vessel found out both board and rock for herself.  We also had anxious looking out this evening for the bit of board: one of us thought he saw it right a-head; and when some of the others were trying to see it too, John Stewart succeeded in discovering it half a pistol-shot astern.  The evening was one of the loveliest.  The moon rose in cloudy majesty over the mountains of Glenelg, brightening as it rose, till the boiling eddies around us curled on the darker surface in pale circlets of light, and the shadow of the Betsey lay as sharply defined on the brown patch of calm to the larboard as if it were her portrait taken in black.  Immediately at the water-edge, under a tall dark hill, there were two smouldering fires, that now shot up a sudden tongue of bright flame, and now dimmed into blood-red-pecks, and sent thick strongly-scented trails of smoke athwart the surface of the Kyle.  We could hear, in the calm, voices from beside them, apparently those of children; and learned that they indicated the places of two kelp-furnaces,--things which have now become comparatively rare along the coasts of the Hebrides.  There was the low rush of tides all around, and the distant voices from the shore, but no other sounds; and, dim in the moonshine, we could see behind us several spectral-looking sails threading their silent way through the narrows, like twilight ghosts traversing some haunted corridor.

    It was late ere we reached the opening of Isle Ornsay; and as it was still a dead calm, we had to tug in the Betsey to the anchoring ground with a pair of long sweeps.  The minister pointed to a low-lying rock on the left-hand side of the opening,--a favourite haunt of the seal.  "I took farewell of the Betsey there last winter," he said. "The night had worn late, and was pitch dark; we could see before us scarce the length of our bowsprit; not a single light twinkled from the shore; and, in taking the bay, we ran bump on the skerry, and stuck fast.  The water came rushing in, and covered over the cabin-floor.  I had Mrs Swanson and my little daughter aboard with me, with one of our servant-maids who had become attached to the family, and insisted on following us from Eigg; and, of course, our first care was to get them ashore.  We had to land them on the bare uninhabited island yonder, and a dreary enough place it was at midnight, in winter, with its rocks, bogs, and heath, and with a rude sea tumbling over the skerries in front; but it had at least the recommendation of being safe, and the sky, though black and wild, was not stormy.  I had brought two lanthorns ashore: the servant girl, with the child in her lap, sat beside one of them, in the shelter of a rock; while my wife, with the other, went walking up and down along a piece of level sward yonder, waving the light, to attract notice from the opposite side of the bay.  But though it was seen from the windows of my own house by an attached relative, it was deemed merely a singularly distinct apparition of Will o' the Wisp, and so brought us no assistance.  Meanwhile we had carried out a kedge astern of the Betsey, as the sea was flowing at the time, to keep her from beating in over the rocks; and then, taking our few moveables ashore, we hung on till the tide rose, and, with our boat alongside ready for escape, succeeded in warping her into deep water, with the intention of letting her sink somewhere beyond the influence of the surf, which, without fail, would have broken her up on the skerry in a few hours, had we suffered her to remain there.  But though, when on the rock, the tide had risen as freely over the cabin sole inside as over the crags without, in the deep water the Betsey gave no sign of sinking.  I went down to the cabin; the water was knee-high on the floor, dashing against bed and locker, but it rose no higher; the enormous leak had stopped, we knew not how; and, setting ourselves to the pump, we had in an hour or two a clear ship.  The Betsey is clinker-built below.  The elastic oak planks had yielded inwards to the pressure of the rock, tearing out the fastenings, and admitted the tide at wide yawning seams; but no sooner was the pressure removed, than out they sprung again into their places, like bows when the strings are slackened; and when the carpenter came to overhaul, he found he had little else to do than to remove a split plank, and to supply a few dozens of drawn nails."



[Next Page]
 

NOTES.

 
2.    Sir R. Murchison considers these rocks Silurian.  See "Quarterly Journal" of the Geological Society, Anniversary Address.
 
3.    Probably one of the Ieastrea of Edwards.
 
4.    See a paper by the Rev. P. B. Brodie, on Lias Corals, "Edinburgh New Philosophic Journal," April 1857.

 


[Home] [Up] [Autobiography] [First Impressions] [Scenes and Legends] [Tales and Sketches] [Leading Articles] [Miscellanea] [Main Index] [Site Search]

Correspondence should be sent to Webmaster@Gerald-Massey.org.uk