of a large country and publish results; while, after the general survey is made, private individuals can easily under-take the local examinations, and may often be paid for doing so. Consequently it is the duty of the Government to take in hand an important work which cannot be done by private-enterprise, and this is what most Governments of civilised countries have actually done.

Now, no systematic geological survey has yet been made of New Zealand—that is, there has been no close and continuous examination of the rocks, starting from a few localities and gradually spreading over the whole country. Nevertheless, in the intervals between the examination of mines and mining districts Sir James Hector has managed to get a sketch-map made of the greater part of the country, while some important districts have been examined more in detail. We thus know a good deal about the general geological structure of New-Zealand although we do not know it accurately; and, as a consequence, in several cases the evidence appears to be conflicting, so that different opinions may be maintained according to the observer's interpretation of the evidence.

But while something has been done towards unravelling the geological structure of New Zealand, the palæontology has been woefully neglected. Large collections of fossils have been made by the survey, but they remain undescribed—most of them, indeed, as yet unpacked—in the Museum at Wellington, and there appears to be no chance of getting them described. According to the last annual report, there are more than thirty thousand specimens in the exhibition-cases, by far the larger part of which are unnamed and undescribed; while, in addition, there are about five hundred boxes still unpacked, many of which have been stored away for years. A slight attempt was made in 1873 to describe the Tertiary shells and Echinoderms; but the plates which were prepared to accompany the text have never been published. Also, in 1880 a report appeared on the Neozoic corals and Bryozoa, but nothing further has been done by the New Zealand Government. Nearly all we know of the palæontology of New Zealand is either due to the publication by the Government of Austria of the fossils collected by Dr. von Hochstetter in 1859, or is the result of private enterprise. The collections of the survey—made with great labour and at considerable expense—are wasted by the apathy of the Government, which appears not to know how great would be their scientific value if described and figured, and how useless they are as they now exist in the Museum at Wellington. It is a great pity that this should be so, for the geographical position of New Zealand gives to its geology a world-wide interest. It is in New Zealand alone that we

which in north-west Nelson and near Reefton lie quite unconformably on the edges of the folded rocks of the Takaka system. It has been called the “Maitai system” from the Maitai River at Nelson, although, it is doubtful whether the rocks of the lower part of the Maitai Valley belong to it.^* This Maitai system is very largely developed in New Zealand. In Otago, Canterbury, and Marlborough it lies directly on the Wanaka system; and it forms the greater part of the mountain-ranges of New Zealand in both Islands, from the Takitimos in Southland to the eastern side of the Bay of Plenty. It is again found in isolated patches on the north-west of Lake Taupo, and in many other places as far as the North Cape. In the North Island these are the oldest known rocks. The thickness of the system in the South Island is estimated by Sir James Hector at from 7,000ft. to 10,000ft.; but it is very difficult to form an opinion, as the rocks are everywhere highly folded and the stratification is often obscure.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Only three species of Brachiopoda have been found, all of which—if they have been correctly named—are also found in the Permo-carboniferous of Tasmania, and one of them—Productus brachythærus—seems to be characteristic of that formation in eastern Australia. Straight but slightly tapering tubes up to 2 in. or 3 in. in length and from 1/10;in. to ¼in. in diameter have also been found in several places in Canterbury as well as near Wellington. They are generally called “annelid tubes,” but more probably they are the shells of pelagic Cephalopoda. The so-called annelid-beds, which contain these fossils, appear to belong to the upper part of the Maitai system, and on the western side of Lake Ohau they have been found together with the remains of plants.^†

In several localities in both Islands red-jasperoid slates occur, sometimes associated with manganese-oxide, and this, together with the paucity of fossils and the general absence of plant-remains, points perhaps to a deep-sea origin.^‡ It seems probable that these beds accumulated in the deeper portions of a sea, the more western and shallower portions of which were at the same time receiving the débris from a Permocarboniferous Australian Continent.

In New Zealand the period closed with an eruption of granite, which is now found at intervals from Stewart Island, through Westland, to near Separation Point in Blind Bay. This granite has penetrated the rocks of the Maitai system, and is found as rolled fragments in the conglomerates of the

Zealand no land plants or animals of the period by which we could test the truth of this hypothesis.

The land then sank and the Maitai system was deposited. Towards the end of the Carboniferous period the New Zealand area appears to have been under a deep ocean, but with shallower water to the west and north-west, for in Tasmania and Australia we find that the rocks, which were probably contemporaneous with our Maitai system, contain shallow-water beds with plant remains; while in New Zealand there is an absence of conglomerates and of fossils generally. Perhaps, also, we have direct evidence of deep-water conditions in our deposits of manganese-ore and red-jasperoid slates.

Probably it was early in the Permian period when elevation, with folding of the rocks, again took place; and this was aecompanied by the intrusion of granite in the South Island along the axis of the present New Zealand Alps. After a long interval the granite was followed by a series of basic volcanic eruptions, and the land began slowly to sink, and this was continued during the Triassic and the first half of the Jurassic periods. During the whole of this long-continued subsidence—while the rocks of the Hokanui system were being laid down—land must have been in the close neighbourhood, for in the rocks we find bands of conglomerates and abundant vegetable remains.

The sedimentary rocks of the Hokanui system have much the same lithological characters throughout New Zealand, and appear to have been chiefly the products of large rivers which drained a continent, and not the products of small island streams; so that probably New Zealand was then placed near the coast-line of a large continent stretching away to the north and west. Whether this land reached to Tasmania and Australia we cannot say until the New Zealand fossils have been compared with those of Australia; but if it be true that a Labyrinthodont lived on the banks of these rivers a land-connection between Australia and New Zealand must have existed. Certainly there is no evidence that the Permo-triassic land was a mountainous region near New Zealand, for there are no signs of any deep local excavations having taken place, such as would have been produced by mountain streams, and no evidence has as yet been found of the presence of Permo-triassic glaciers.

About the middle of the Jurassic period folding of the rocks again occurred along the same north-east and south-west axis; the Alps were formed, and the present land of New Zealand may be said to have been born, for since then it has never been submerged. Up to the present we know of no igneous outbursts which accompanied this third folding. But some of the granites along the West Coast—such as that

covered by marine rocks, which culminated in a limestone, well known as a building-stone, in many parts of New Zealand. This stone, under various names, is found in patches all round the coasts of New Zealand, from Winton in the south to the Bay of Islands in the north, as well as in many of the inland valleys. It belongs to that variety called polyzoal limestone, because it is made up principally of small fragments of calcareous Polyzoa, &c.; and it is, no doubt, the remains of a reef which in the Oligocene period encircled New Zealand. These Tertiary rocks lie unconformably on those of the Waipara system at the Weka Pass, Mount Somers, and a few other places; but elsewhere they lie upon older rocks, belonging to the Hokanui and other, systems. At the time of the formation of the Oamaru limestone there were living in our seas a Zeuglodont whale (Kekenodon onamata), as well as true cetaceans (Squalodon serratus), a penguin (Palæeudyptes antarcticus), the huge shark (Carcharodon angustidens), rays (Trygon and Myliobates), and a sparoid fish (Sargus laticonus), as well as the nautilus called Aturia australis. But with them are found some Cretaceous-looking Echinoderms belonging to the genus Holaster. These last were no doubt survivors from Mesozoic times, and I agree with Dr. Stache and Sir F. McCoy in thinking that the rocks in which they occur belong to the Oligocene period.^* A species of Unio has been described from the coal-beds in Otago.

In the neighbourhood of Oamaru basic volcanic rocks underlie the marine beds, and in other places of Otago and Canterbury volcanic rocks are interbedded with the sedimentaries which belong to the earlier part of the Oamaru series. The most remarkable of these is a hydrated tachylyte, which is found in several places between Lookout Bluff, near Hampden, and Castle Hill, in the valley of the Waimakariri, a distance of a hundred and fifty miles.^† The volcanic system of Dunedin probably belongs to the close of this period, as also do the volcanic rocks in the neighbourhood of Palmerston South, and those on the northern side of the Hurunui Plains at Culverden and Pahau, as well also as a large part of the volcanic rocks of Banks Peninsula. The older rocks in the Dunedin Peninsula are andesites, followed by olivine

sixty miles. The glacier which filled Lake Wakatipu did not drain into the Clutha, but went due south by Athol to the Oreti River. The united glaciers of Lakes Te Anau and Manapouri extended to Blackmount, on the River Waiau, a distance of sixty-five miles, thus being the largest of the New Zealand glaciers. There is also in Otago the remarkable isolated moraine in the Lower Taieri, which forms low hills—some 400 ft. or 500 ft. in height—between Lake Waihola and the sea. This moraine may perhaps be older than any of the others. A few marine fossils have been found in the sandy clays underlying it, which seem to indicate a Miocene age for those beds, so that the moraine itself may belong to the older Pliocene.^*

In Nelson the terminal moraines of the largest of the ancient glaciers are about 2,000 ft. above the present sea-level. In South Canterbury they go to 1,000 ft., and in South Otago to 600 ft.; but in Westland and in the West Coast Sounds the glaciers advanced to below the present sea-level. There are, however, no stratified till-deposits, and nowhere do we find the moraines enclosing marine shells, so that there is no evidence that these glaciers descended into the sea. Another remarkable feature is that no boulder-clay has as yet been detected in New Zealand—nothing but the ordinary moraines of valley glaciers. Neither are there any true erratics—that is, large blocks of rock which have been transferred from one drainage system into another. All our erratics have come down the valley in which we now find them.

Now comes the question, What was the cause of this great accumulation of ice in our mountains? We cannot account for it by a colder climate, for there is not the least palæontological or biological evidence to show that our climate has ever been colder than it is now. On the contrary, all the evidence goes to show that it was formerly warmer. Thus in the south we find local outliers of warmth-loving plants and animals which flourish much better in the North Island. Such are the nikau palm (Areca sapida) on Banks Peninsula and at Westport, where also Lomaria frazeri occurs, a fern which is not found elsewhere south of Auckland. These are survivals of a more genial age in the South Island. Indeed, the greater part of our present flora is of subtropical origin, as also was that of Europe before the cold of the glacial epoch killed it off and largely replaced it by a northern flora. Nothing of that kind has happened in New Zealand. Again, several northern marine shells still live in Foveaux Strait, such as Triton spengleri, Scalaria zelebori, and Cookia sulcata. If New Zealand had lately passed through a cold phase all these plants and

animals would have been killed off in their southern localities, for there is no place near New Zealand to which, in Pleistocene times, the subtropical flora could have temporarily retreated and then returned after the cold was over.

The same may be said of almost the entire faunas and floras of the islands lying to the south of New Zealand. Yet we find in the Auckland Islands a peculiar genus of duck (Nesonetta) which cannot fly, and in Antipodes Island a peculiar parroquet (Cyanorhamphus unicolor) which has almost lost its powers of flight. These birds must have been developed on the islands where we now find them, and the process must have been a slow one; yet during the whole of that time the islands could not have been covered with ice. We may extend this argument to other islands in the Antarctic Ocean, such as Kerguelen Land and the Crozets. These possess several peculiar plants and animals, and it is certain that the islands could not have been covered with icesince the first appearance of their present floras. A general reduction of the temperature of the whole Southern Hemisphere being therefore out of the question, we must look for other and local causes for the extension of the glaciers. Two theories have been advanced—one is that our mountains during the great glacier epoch were flat-topped, forming plateaux on which large masses of snow collected; the other is that the mountains stood at a greater altitude than at present, due to a general elevation of the whole Island.

Now, passing over the question whether large snow-covered plateaux necessarily imply large glaciers—they do not do so in Norway—we are met with the fact that most of our river-valleys had been cut down to their present level before the Oligocene period, for rocks of that age fill several of them nearly or quite to the bottom. For example, it is certain that in the Eocene period the Rakaia River ran at a lower level than it does at present. As this is an important point I will give the proofs of my statement.

In the valley of the Rakaia, opposite the east end of Lake Coleridge, there is an outlier of Oligocene limestone called Redcliff. It is lying almost horizontal in its original plane of deposition in a lateral valley on the south side of the river, and is, no doubt, a fragment of a set of beds which once filled all that part of the valley. Now, as this limestone passes under the gravels and descends below the present level of the river, it is evident that when the Rakaia scoured out the valley in the Eocene period it must have been running at a lower level than at present, for it now runs on the top of alluvial gravels which partly fill up the whole valley. Also the junction of the limestone with the Palæozoic rocks must mark the limit of the valley of the Rakaia at that place when

the limestone was formed. If, therefore, any great lateral denudation had taken place since that time, the line of junction between the two rocks ought to stand out as part of a prominence. But it does not do so; consequently, the lower portion of the Rakaia Valley cannot have been greatly enlarged since the Eocene period. This is confirmed by the fact that on the southern slopes of Mount Algidus, in the Upper Rakaia, there is another outlier of Tertiary marine rocks, showing that there also the valley was very deep long before the Pliocene period.

It is therefore very unlikely that a great plateau in the upper part of the Rakaia Valley has been lately removed; and we may say, generally, that as the rivers of the South Island had cut such deep valleys before the Oligocene period it seems impossible, from what we know about river erosion, to believe that any large plateaux could at that time have been in existence—that is, none to which we could attach any great importance.^* If also, as we have seen, the New Zealand Alps were formed in the middle of the Jurassic period, and have been exposed to the action of the weather ever since, plateaux of any size could not have existed from the Jurassic to the Pliocene and then have rapidly disappeared, especially at a time when, by the hypothesis, they were protected by a covering of perpetual snow.

This plateau hypothesis failing, we are left with that of elevation to account for the phenomena; and it has been calculated that an elevation of between 3,000 ft. and 4,000 ft. would, at the present day, be sufficient to expand our glaciers to their former dimensions.^† That the New Zealand Alps did formerly stand higher than they do now we have direct evidence in the deep fiords of south-west Otago and Marlborough, which must have been excavated when the land was considerably elevated. The greatest depth recorded in the West Coast Sounds is 1,728 ft., in Breaksea Sound; but in many places no bottom was reached with the line used, and we may safely assume that when the valleys were scooped out they stood more than 2,000 ft. higher than they do now. And this agrees fairly well with the quite independent estimate that an elevation of 3,000 ft. or 4,000 ft. would be sufficient to reproduce all the phenomena. In Canterbury also we find evidence of a former elevation, for in sinking a well in Christchurch a quantity of solid timber was found at a depth of more than 400 ft., which must either have grown on the spot or have been brought there by a river. How deep the shingle-beds of the Canterbury Plains go we do not know.

Lastly comes the question, When did this elevation take place? I will take the biological evidence first. The great similarity between the faunas and floras of the two Islands of New Zealand shows undoubtedly that they were once united; and an elevation of 500 ft. at Cook Strait would connect them again. Nevertheless, we find six different kinds of birds represented by different species in each Island; and this is not due to differences in climate or in the physical features of the two Islands, but to changes in the animals which have taken place since the Islands were separated. It is the same with the extinct moas. Nearly all—perhaps all—of the known species were confined to one or other of the Islands, and certainly none were abundant in both. But this implies that the Islands have been separated by Cook Strait for a long time, during which, of course, there could have been no general elevation.

Still, as I said before, it is evident that the two Islands were once united. Indeed, we may go further and say that in all probability the Chatham Islands, the Auckland Islands, and perhaps Campbell and Macquarie Islands were at one time united to New Zealand by land, for their faunas and floras are closely allied to those of New Zealand, and are quite unlike those of Tasmania.

As no lizards nor land-shells have passed between Tasmania and New Zealand, and as very few birds, insects, and plants are common to both countries, although the distance between them is not much more than twice that between New Zealand and the Chatham Islands, it is evident that our connection with these and with the Auckland Islands must formerly have been much closer than it is now. And the presence of a slug common to all three suggests that there was a land-communication between them.

Now, these outlying islands contain many endemic species of plants and animals, and, in the case of the Chatham Islands, we cannot explain the existence of these distinct species by differences of climate. Out of twenty-one land-birds on the Chathams, seven—that is, one-third of the whole—are endemic. And of the plants about 15 per cent are endemic. This implies that the Chatham Islands have been isolated for a very long time, and we can say with some confidence that this isolation must have lasted ever since the close of the Pliocene period. But when New Zealand extended so far as to include the Chatham Islands it probably stood at a much higher elevation than at present; and the Pliocene period, therefore, is the time we should expect that the greatest extension of our glaciers took place.

We will now take the geological evidence. In the first place, it is significant that there are no marine Pliocene beds

north-west there was the Continent of Australia, with vast mountain-ranges covered with snow, and with glaciers glittering in every valley.

This state of things lasted into the Permian period, by which time the bed of the ocean had been gradually raised, so that the sea became shallow, and the New Zealand area lay near the shore-line of a continent stretching away towards Tasmania and Australia, to which, perhaps, it was joined. This land was covered with ferns and Cycads, and probably there were a number of active volcanoes ejecting rhyolitic lavas. But what animals lived on the land we do not, as yet, know.

In the middle of the Jurassic period came a violent upheaval. The rocks were crumpled up, the coast-line was changed into a mountain-range, and the land between it and Australia sank, forming the Tasman Sea. The new land, which we may now call New Zealand, for it has never since been entirely covered by the sea, extended in a westerly direction to at least twice its present breadth, and to the north it joined New Caledonia and New Guinea, which at that time probably formed part of a South Pacific continent. Plants and animals—including snails, worms, and insects, but no birds—came trooping down from the north to form the basis of our flora and fauna.

A long period followed, in which the western side of the mountains of the South Island were constantly being worn away by the heavy rains brought by cyclones sweeping over the Tasman Sea; but this did not take place to so great an extent in the north, for in those latitudes westerly winds are not so prevalent.

In the Upper Cretaceous the land subsided, and New Zealand was reduced to comparatively small limits. This land, however, supported many angiospermous trees, as well as gymnosperms, whose descendants are still living; while in our seas were marine reptiles and shells which have long since become extinct.

A little before the commencement of the Tertiary era the rocks were folded once more, the land rose again, and again it stretched far away to the north, but was not again united to New Guinea nor to northern Australia. A second invasion from the north followed, and quantities of plants of all descriptions, accompanied by animals—among which were many land-birds—migrated into New Zealand, and it is the descendants of this Eocene invasion which form the greater part of our present flora and fauna.

This was the last folding of rocks in New Zealand on an extensive scale, for all the younger rocks usually lie in the same position in which they were originally deposited, and

circle round the bases of hills formed by older rocks. Not only was the last touch given in the Eocene period to the internal structure of the mountains, but the chief valleys were also deeply scoured out, so that when the land sank again in the Oligocene period these valleys were filled up with marine limestones and other rocks.

The Oligocene and Miocene were periods of depression, separated by a slight upheaval which lasted only for a short time. During most of the Middle Tertiary era New Zealand must have formed a narrow ridge of land, very irregular in shape, running north-east and south-west, with some detached islands on each side, three or four on the south-east side, and a dozen or more to the north-west, none of them being very high above the sea.

In the older Pliocene came the last great upheaval. All the islands were joined together, and the land stretched away to the east and south so as to include the Chatham and Auckland Islands, as well, perhaps, as Campbell and Macquarie Islands; while to the north it certainly extended to the Kermadecs, and perhaps much further. On the mountains of the South Island large glaciers were formed; and the torrential rivers running from them tore into disconnected fragments the Miocene marine rocks which obstructed their valleys. Probably at this time more land than at present existed in the Antarctic Ocean, for New Zealand added to its flora and fauna many antarctic plants and marine animals. But this land could not have connected New Zealand with either Patagonia or South Africa, for if it had done so we should certainly have had many more immigrants, including land birds, and, probably, mammals.

It is possible that this large extension of land to the eastward may have produced desert or steppe-like conditions in a portion of New Zealand, evidence of which some botanists think they find in our flora; also, in the old lake at Kapua, near Waimate, there is some slight evidence of a dry epoch having, at that place, succeeded the diluvial epoch during which the moas were buried.^* But this may have been due to quite local causes.

Subsidence seems to have commenced first in the southern portion of the North Island, for in the newer Pliocene large portions of what are now dry land were under the sea, and Cook Strait had been formed. But at a later date sinking began in the South Island also, so that in the Pleistocene period the sea at Amuri Bluffstood at least 500 ft. higher than it does now. This sinking has again been followed by an elevation of all parts of New Zealand, the centre of the

good sections can be studied by the way, more especi-ally along the watercourses, which generally proceed, in the direction of the dip of the beds. The country from the mountains to Puketitiri may be included as forming a part of the range, although the rocks on the range of hills fronting Puketitiri are, speaking geologically, different. The mountains themselves in some places show rocks with a slaty cleavage. These are mixed with a pale-red sandstone of a fairly fine texture, and corresponding to the rocks at the top of the Ruahine Range at the back of the Whakararas. They are highly denuded, and since the destruction of the scrub and native grass the high winds have bared them, and now thousands of acres consist of bare rock, which sun and frost and rain break up at a rapid rate—so rapid, in fact, that no growth is at present possible.

On the top of the highest part of the mountains I was much interested at observing a series of parallel lines of loose rocks arranged irregularly in line as is done by little children when playing with stones. I could only account for their presence by supposing ice-movement from a higher elevation, when the stones are brought down and deposited in irregular lines as morainic débris. Snow does not rest for more than three months in the year upon the mountains, but the falls appear to be heavy at times, and possibly a slight movement of the snow takes place as soon as the sun is sufficiently powerful to act upon it, at a time when the atmospheric changes are most rapid. The ridge that separates the sources of the rivers is quite narrow, and one can imagine the time when the country further westward was open to the east, and when only a single large river flowed from the lands in the direction of Ruapehu and Taupo, bringing down great volumes of pumice, shingle, and timber, the latter being the result of the destructive outbursts from the volcanoes in the district.

The range of hills immediately joining the mountains on the east side is known as the Birch Range. The Makahu Stream flows between the two, and runs to the north-east to join the Mohaka, that comes further from the westward. These hills may be set down as forming a part of the main range, as they are geologically the same. Denudation, however, has played havoc with them, and they seem as if they had been shattered and shaken and broken at the time when the mountains were in process of elevation. This range is again separated from another line of hills which forms the northern end of a series of rocks—limestones, sandstones, and other—which are met with along the foot of the entire Ruahine Ranges. In some places the rocks are fairly compact limestone; sometimes they are hungry-looking sands, such as are seen topping the slates in different places.

At the top of the hills on the Napier—Kuripapango Road, known as “Blowhard,” the sandstones run into a peculiar fluted limestone, as described by me in a former paper. Further northward the limestone disappears, except here and there, and a brownish-grey hungry sandstone, mixed in places with a grit conglomerate, takes its place. This sandstone country has been subject to great denudation, and the whole of the Puketitiri district presents remnants of this sandstone and nodular limestone, into which the former passes as it dips to the south-east.

In order, however, to comprehend the full sequence of rocks in the district under notice it is necessary to imagine what the country was at the time when the drainage was to the south-west. The slope was towards Hawke's Bay, but a deep valley lay between the rising mountains and the range of limestones, whose scarp showed a fracture running north-east and south-west, and facing the north-west. This deep valley can be traced for many miles, for the scarp is as definite to-day as when the upward pressure fractured the limestones which at that time covered the entire area in the direction of Ruapehu. I took a photograph of one section of the scarp, on the ridge between Puketitiri and Hawkeston, the residence of Mr. J. Hallett, which shows a face as if cut with a knife. The elevation of the Kawekas and the fracturing of the rocks to the eastward caused a break sufficiently large for the entrance of the sea so as to form deep bays and inlets, and in various parts of the district fossiliferous sands and impure limestones are met with topping limestones which belong to the Upper Miocene beds. I do not think these younger beds of fossiliferous sands and limestones are to be met with to the north of the 39th parallel of latitude, which may be said to be the northern boundary of Hawke's Bay; but it is also a curious fact that shingle conglomerates and attendant sands do not appear to the northward of this parallel, whilst they are very highly developed to the southward. The limestones which present in their scarps such a characteristic feature in the landscape belong to the Upper Miocene beds. They abound in fossils, and in some places the large oyster Ostrea ingens, which is characteristic of what are known as the Te Aute limestones, forms immense banks, presenting the appearance of artificial banks of oyster-shells. On the hills a mile or so to the north of Mr. Hallett's homestead there are scores of acres of these shells, and they are arranged so regularly atop of each other that it is difficult to imagine how they lived.^* Certainly they represent a long period of deposition;

and yet there is no trace of a break, an intrusion, or a change in the direction of currents. In some places hardly another variety of shell—except a Patella or a Balanus—is to be found, and yet within a score or two yards are other banks containing a whole museum of specimens. I have taken photographs of several of the banks, and have made lantern-slides to show what life there must have been thereabouts in the days when the oysters were at their prime. Such deposits in these days would satisfy even the demands of a city like London for years, although I doubt whether epicures of this bivalve would have wished for an oyster-supper where one or two oysters at the most was sufficient for a meal!

The Miocene beds continue from Puketitiri in the direction of Patoka, changing somewhat their rock characters as they proceed. The limestones are seen to rest upon a pale-blue sandy limestone and marl, containing plenty of fossils; such as Struthiolaria, Cytherea, Pecten, Natica, Murex, and Ostrea. At the Patoka Hill laminated limestone is inter-bedded with fossiliferous sands, and these are seen to be above the marly limestone containing the fossils just named. Few or no fossils are to be seen in this laminated limestone, although traces of broken shell, such as Pectèen and Balanus, can be distinguished.

Proceeding towards Napier from Patoka the country appears rugged and broken. We are now on the eastern slope of the Titio-kura limestone range, of which Te Waka is such a prominent point from the Napier bluff. Immediately following the Patoka Hill is a smaller one at its foot, and this is quite different in structure from the rocks that have been hitherto met with. We are now in the line of the conglomerates, which, commencing north of Pohui in sands and grit, strike to the south-west and intrude themselves everywhere, sometimes resting beside the limestones, sometimes replacing them, and sometimes being apparently mixed with shelly limestone, and so thrusting themselves every where till they partly lose themselves in the Ruataniwha Plain.

Whenever I come among these conglomerates there always arises in my mind a doubt as to their age, and yet they can be traced regularly over a large area of this district. Sometimes I have been inclined to the opinion that they are contemporary with the Pliocene limestones which appear overtopping the Miocene deposits as far back towards the mountains as the Birch Hills; at other times they seem to me as belonging to the Miocene beds; and yet there can hardly be a doubt that they were deposited subsequently to the Pliocene beds, and during their deposition the Miocene and Pliocene deposits were greatly denuded. In fact, with

the exception of the higher lands, all the Miocene and Pliocene limestones were subjected to severe erosion, and were replaced by enormous accumulations of sand, shingle, conglomerate, and lignite lands such as now cover such a large extent of country.

Between Patoka and Rissington the whole area is covered with a conglomerate deposit which varies in structure, sometimes presenting walls almost like a face of limestone, sometimes being of a deep-brown grit, and at others passing into sands and shelly conglomerate. The shells where seen are mainly the cockle and the oyster. It would appear that the shingle-conglomerates were deposited within the vicinity of salt-water, for bones are not uncommon, as the workmen who quarry the conglomerates for roading often find large bones, which appear to belong to a cetacean of some kind. I have several of the bones so found, and specimens were sent to Sir James Hector by a Napier gentleman, who received intimation that they belonged to a cetacean.

At Rissington the blue-clay marls, which form the lowest beds of the Pliocene deposits, are well exposed, and on the top of them are seen resting shingle and conglomerate which have evidently planed down the clays and carried away the limestones.

As you rise the hills in the direction of Mr. Bennett's homestead at Wharerangi the limestones again make their appearance, and with them here and there are traces of the shingle-conglomerates, which evidently at one time swept across the tops of what are now the highest hills hereabouts. Traces of the shingle may be noticed descending the hills into the valley which opens at Puketapu into the Tutaekuri River, but their appearance is such as to bring doubts into the mind regarding their true stratigraphical position. From the valley the road passes over the hills in the direction of Napier, and here the well-known upper limestones of the Napier series are met with, whilst in certain places of the inner harbour the shingle-conglomerates make their appearance on the top of the blue marls, which represent the middle beds of the Napier series.

It may be that the shingle deposits that are met with here and there from the top of the hills beyond Wharerangi to Napier were deposited from a different stream from that which swept over the whole country between Patoka and Rissington, or perhaps the stream with its burden of shingle was diverted somewhat further to the south ward. In any case, the limestones were left in the district between Wharerangi and Napier, whilst they were replaced further to the south-west, where remnants remain mixed with shingle, as if solidification had taken place after the shingle had passed over the district.

Island in a general north-west and south-east direction, including as far north as the Philippines, and so as to embrace the greatest number of the Pacific islands. The actual trend of the islands will be found lying generally in a west-northwest and east-south-east direction, although many vary from those points.

Professor Milne places this area in his map as one of subsidence, wherein it will be seen how greatly we differ. The trinolith at Tongatabu happens to be a very good landmark as to sea-levels, and it shows that that island, at any rate, has not subsided 6 in. a century for the past three or four thousand years. Instances are within our own knowledge of the rapid growth of volcanic islands. “In 1796 a volume of smoke was seen to rise from the Pacific Ocean about thirty miles to the north of Unalaska. The ejected materials having raised the crater above the level of the water, the usual volcanic phenomena occurred. Repeated eruptions have increased the dimensions of the island until now it is several thousand feet in height and between two and three miles in circumference.” So that there is really no geological objection to the upheaval and formation of any of the groups of islands in the Pacific.

I am at once met with the objection that there is no line of present volcanic activity along the 20th parallel of south latitude. That is so; but nevertheless a glance at the map shows that there has been a line of upheaval, and my duty is to record in such a paper as this all the evidence I can collect bearing upon volcanic action in the Pacific, a region which hitherto has not met from vulcanologists that attention which it merits—the grandest volcanic region, I take it, upon the face of the globe. Great earth-oscillations doubtless occur in this immense water-area—seventy million square miles. There could not, of course, be so much volcanic activity without this vast water-area. The pressure of 30,920 ft. of sea-water near Tonga upon each foot of the ocean-bed at 62 ½lb. to the foot can be readily calculated. Such a column of water would readily find out any weak crack or crevice to reach the central heat or the imprisoned lava within the earth's crust. But, of course, the more surface-water that pours in the more quickly steam is generated, which by upheaval closes up the fault or crevice.

Thus, as I have already said, the bottom of the ocean is blistered by upheavals and volcanic growths—viz., the various groups of islands—and although we find no active volcanoes along this third line, yet I am not prepared to admit that volcanic action is quiescent, for one of the greatest tidal waves on record within the past twenty-five years arose from a submarine explosion in 21° 22′ S. latitude and 71° 5′ W. longitude—the 1877 wave described by Milne and others, which,

like the wave of 1868 and all the other great waves, travelled almost the length and breadth of the Pacific.

Accepting as a maxim the statement that an earthquake is an incomplete volcano, the islands along this line experience pretty sharp earth-movements; at present correct observations of them are not taken at all. Thus as I write news comes (dated 8th August, 1899) of a tidal wave which burst into Valparaiso Bay, damaging Government property to the extent of millions of dollars. Great parts of the embankment were carried away, and railway-cars and locomotives were dashed off the rails and embedded in the débris, the rails being torn and cranes smashed. Many thousand tons of merchandise were destroyed. The State railway between Bellavista and Baron was completely wrecked. This may only have been a local phenomenon, but a few of the islands in the Paumotus are sometimes washed clean by tidal waves.

The continuous pressure upon the ocean-bed at the deepest soundings between Tonga and New Zealand at 800 tons to the foot (34,848,000 tons to the acre, and 22,400,000,000 tons to the square mile) is so enormous that it is no wonder we find great volcanic activity within a certain radius from it. Where are we to look for the balance ? At our hotlake district in New Zealand, or at Tanna or Ambrym, in the New Hebrides ? Supposing we found thermal action going on regularly at eight- to ten-minute intervals in the geysers or fumaroles at our hot lakes, or similar discharges at Tanna or Ambrym in the New Hebrides, might we not conclude that these are the safety-valve escapes from the enormous pressure referred to, and that the regularity of the escaping steam proves a certain connection within the whole circle of which the points referred to are radii? The pulsations fairly average twelve minutes at the three points named, yet some sixteen hundred miles apart; so I think Sir James Hector ought to grant a greater area of unity in volcanic phenomena than he does.

It will be seen, too, that I differ very considerably from Milne's chart in my three lines or areas of activity and upheaval, as he gives them all as subsiding. It may be he is right with regard to those islands near to and north of the equator and with Easter Island and with some of the Paumotus; but the really subsiding area in the Pacific—viz., an ocean band of about a thousand miles in width following the south - east and north - west trend of the western coast of the American Continent—he does not give at all. But even this subsidence is so slight as to be almost unnoticeable, for we can even not be guided by the Easter Island images, whose gradual subsidence maybe only a local phenomenon, as I fancy much of the subsidence amounts to in the Pacific.

The one real and prominent fact in regard to the island groups in the Pacific is that they have been upheaved, since which a certain amount of subsidence has followed, but so slight that, as I have said, the Langiis at Tonga still remain much as they were built some three or four thousand years ago. There may be now a period of rest, but even that supposition I cannot agree with, so many slight changes do I know of taking place. I must say that Milne concludes his book in quite as uncertain a frame of mind as I am myself. The field is so vast and our knowledge so slight. The coral borings at Funafuti may tell us something, but even the sinking of a volcanic hill or blister near the equator some 1,500 ft. below the sea after upheaval, somewhat like Falcon Island, but to a greater depth, may only be a local phenomenon, telling us nothing of the general law of lines of activity or subsidence to which I am referring in this paper.

The opinion of leading geologists belonging to this Institute is that there are no “lines of activity,” each volcanic rent being local to itself. What, then, of the great line running north to south, bordering the Pacific, in the two American continents? And another line might be drawn fringing the Pacific on the other side—namely, through Japan and the Kurile Islands.

Beginning with the phenomena upon the islands where the first two lines of activity intersect each other—viz., the Loyalty Islands: Attention has already been called to the Rev. Mr. Turner's remarks “that Lifu is an uplifted coral formation, the highest land on the islands being about 300ft. above sea-level; and Mare a mass of uplifted coral, also bearing marks of two distinct upheavals.”

The peculiar formation of the islands lying off the mouth of Nei-afo Harbour in Vavau (Tonga) now require reference. The whole of them appear to be at an exact uniform level of 100 ft. to 200 ft. above the sea, evidently showing the same upheaval. They look just as if they were bits of reef upheaved. In no other way could they have acquired their flat tops and straight sides. “The power which exhausts itself in the eruption of a volcano often shows itself by the changes it produces in the level of the surrounding country.” I do not think that any land in the three great islands of Tonga is much more than 300 ft. above sea-level; Tongatabu much lower. The highest land I know is in Vavau: that island contains a great open grassy plain about 200 ft. high, unstocked, owing to the absence of water. The soil of Vavau and the surrounding islands is an excellent rich brown one, evidently volcanic. No doubt this soil came from the original volcanic rock, which has weathered, decomposed, and speedily clothed itself with vegetation.

It will be seen at a glance what an excellent soil the brown gravel in the little specimen-bottle upon the table from Falcon Island (now sunk below sea-level) would decompose into if allowed sufficient time; so that from this specimen it can be seen how much of the volcanic soil in the Pacific has been formed. I obtained the gravel before the island disappeared.

There is no water, as I have said, in Tonga except that which is caught in tanks, nor is rain very abundant there, but the night-dews are heavy. The natives bathe in the sea, and use very little water for cooking purposes; their chief drink being from the cocoanut. Their cooking is done by pouring about a pint of water into a huge earthen pot, closing the neck, and converting the water into steam. Clothes they do without, so that the thirty thousand people there rub along very well without much water. The Europeans, of course, use tanks, but the water these tanks contain must be a living mass of microbic germs.

Pylstaart, Kao, Letté, and Tofua are separate small islands to Tongatabu, Hapaai, and Vavau, and tower up, as I have said, from 700 ft. to 3,000 ft. They may have attained these heights by sudden growth, and yet not altered the levels of the three large islands.

Writing of Savage Island, Mr. Turner says, “It is an uplifted coral island 300 ft. above the level of the sea, about forty miles in circumference, in 19° S. latitude and 170° W. longitude.” It will be noticed that Savage Island, the Tongan Group, and the Loyalties, stretching some twelve hundred miles west to east, show an upheaval of 50 ft. to 300 ft. above sea-level.

The Cook or Hervey Group, over five hundred miles to the eastward, may also be embraced in this line, for with the exception of Rarotonga, which is volcanic and mountainous, the other islands consist of ancient coral formation raised 20 ft. to 200 ft. above the sea, some of them lower, and all surrounded by living coral reefs.

Mangaia, the southernmost island of the group, is of coral formation, but otherwise differs from most of the South Sea islands outside this group. It is about 650 ft. high, and at a distance appears quite flat. There is a fringing reef all round, about 2 cable lengths from the shore, and about 2 ft. above high-water mark, but with no passage for boats. Boats anchor outside the reef on a ledge, and canoes come off for passengers, &c. The natives then look out for the rise of the swell, land the canoes on the reef, jump out quickly, and drag the canoe to land before the receding sea can sweep it back into deep water. These facts go to prove a late upheaval at Mangaia.

Yet Rarotonga, a hundred miles west by north of Mangia,

is 2,920 ft. in height, seven miles long, and four miles wide, with the usual barrier reef. Of course, there is nothing to have prevented Rarotonga having been upheaved 200 ft. to 600 ft. at the same time as Mangaia was uplifted. So with Atiu or Vatiu Island in this group (latitude 19° 59′ S., longitude 158° 6′ W.), whose formation much resembles Mangaia, with a reef closely fringing the shore. Its highest point is 394 ft. above sea-level.

I cannot say whether the sides of these two islands are perpendicular or show two or more upheavals. My friend Mr. Moss, the late British Resident there, may be able to tell us, but I should think they would show either one or two upheavals, as nature generally acts slowly in such changes. The islands at the entrance to Vavau Harbour and the heads of that harbour itself are straight up and down—so much so that the men-of-war use them for target-practice. I should therefore suppose that this area of upheaval, when it occurred, rose about 200 ft. On the other hand, it may only have been tilted up by slow degrees, like the western foot-coast of the South American Andes, in the supposed crumpling or buckling of the earth's crust. But the islands present so up-heaved an appearance that one is led to that conclusion. A careful study of the rock-formation of each islet will settle the question. If the rock is old coral-formation upheaval cannot be questioned.

The 200 ft. to 300 ft. upheaval is also shown in the Austral Islands, lying further to the east and southward. These islands are high and fertile, Rurutu having a high central peak with lower eminences sloping to the shore. Around the foot of the mountains is a plain about a quarter of a mile wide, which consists of coral-formation, well covered with earth washed from the sides of the adjacent eminences, which has gradually constituted a soil teeming with luxuriant vegetation. Large coral masses rise here and there, abruptly in some instances, to the height of more than 200 ft. above the beach.

The Society Islands, rising 7,000 ft. above sea-level, show great ancient volcanic disturbance, but space will not allow me fully to describe them now. In Tahiti, the most important island of the group, volcanic substances, stratified, broken, and thrown up in the wildest disorder, are every where to be met with.

In the Paumotu, or Low, Archipelago there are many evidences of upheaval, and, of course, subsidence. The soil of Pitcairn Island, where the mutineers of the “Bounty” took refuge, is very rich but porous, a great portion being decomposed lava, the remainder a rich black earth. I should consider Henderson or Elizabeth Island (latitude 21° 21′ S., longitude 128° 19′ W.) within the range of the line of up-heval

I am now referring to. According to Captain Beechey the island is five miles long and one mile wide, and has a flat surface, nearly 80 ft. above the sea. All sides, excepting the north, is bounded by perpendicular cliffs, about 50 ft. high, composed entirely of dead coral, which are considerably undermined by the action of the waves. (This is exactly what one sees at Vavau, Tonga, and I recommend visitors not to miss pulling into one of these caves at the entrance to Vavau Harbour). Byron's Cave there, I expect, was similarly formed.

The external form of the Gambier Islands, in the Paumotus, conveys at once an impression of their volcanic origin, but the seventy-eight islands or groups of islands comprising the Paumotus are generally low islands almost a-wash. The hurricane of 1878, indeed, swept over some of them, carrying every living thing away. It is not often this group is visited by such cyclones. Many of the inhabitants saved themselves by tying themselves to the trunks of trees. It may be that the whole of the Paumotu Islands, under the 200 ft. level, appeared above the sea at the same upward movement to which I am at present referring. The uprising tapered off, as it were, at a little beyond Easter Island, and affected the whole equatorial area right through, perhaps to the western coast-line of the Australian Continent. It will be noticed that the 20th parallel of south latitude very nearly cuts all these islands.

The distance from the Loyalties to Henderson Island is about 3,580 miles, and the evidence which I have been able to produce shows one area of upheaval. It will be noticed, too, that this line is almost at right angles to the first line of present volcanic activity referred to in this paper, bounding the 180th parallel of longitude from Tarawera to Nei-afo, a distance of fifteen hundred miles, or in its fullest extent from Mounts Erebus and Terror to the equator, a distance of about four thousand five hundred miles; the second line of activity, from Hunter Island and through the New Hebrides, the Solomons, and New Britain, &c., to the longitude of Vulcan Island in New Guinea or Uap in the Carolines, being fully 2,250 miles; but this latter line would be better extended another two thousand miles of eastern longitude, so as to include the whole line of present activity through the Malay Archipelago.

Reference will be made directly to the islands immediately bordering on and north of the equator, which are all, with few exceptions, low and small. All these show subsidence. There are a great number of crescent- and low-shaped atolls in these northern, central, and eastern portions of the Pacific; also small, circular, sunken patches of coral, showing that during the

subsidence which accompanied and succeeded the upheavals the coral-polyp started its labours on the top of an extinct volcano or from a rugged ridge or peak. The evidences are common in the Pacific for a volcanic hill to subside, the crater to become a lake or a lagoon, the island to sink still further, and end in being an atoll or a crescent- or low-shaped reef, or finally a sunken coral-patch. At the same time, with these evidences of subsidence, there is this third area of upheaval. Nor can there be subsidence in the earth's crust without upheaval somewhere. But my task is only to record the facts I have seen or collected regarding these islands.

We must ask residents in the islands to keep a careful record of observation in land- and sea-levels. At present it may only be that the Pacific Ocean is deepening slightly in consequence of a slight shallowing of the Atlantic or Indian Oceans, which would account for the marvellous energy of the coral-polyp—an animal that must go on building as the waters deepen. Neither in the Atlantic nor Indian Oceans is there anything approaching the coral-growth we find in the Pacific. At Easter Island the carved tuff images are slowly descending into the sea. In the physical geography of the earth I am inclined to the belief that all change is slow and gradual, and not violent. At times, here and there, we experience a great volcanic eruption, but it is confined strictly to a very circumscribed locality.

Soundings amongst the islands are very steep—200 to 600 fathoms (1,200 ft. to 3,600 ft.), and this sometimes close up to the reefs. One can easily understand this, however, looking at the rugged volcanic shape of the islands of Fiji, Samoa, or Tahiti. And the deepest soundings on earth, as already pointed out, lie between New Zealand and Tonga—5,155 fathoms (30,930 ft.), latitude 30° 27′ S. and longitude 176° 39′ W.; so that the oceans would have to overflow the tops of the Himalayas a couple of thousand feet before the supposed sunken Pacific continent could be again exposed.

Let us suppose a piece of land a hundred miles square, like the narrow neck between Auckland and Onehunga, containing a similar number of extinct craters sunk beneath the sea in the tropics. The coral-polyp would begin its labours directly from the top of the different extinct craters, go on building upwards to the sea-surface, and we should have all the evidences of atolls and circular-shaped reefs, but not, of course, on so large a scale as we find in the Pacific.

In the Sandwich Islands the United States steamer “Tuscarora,” at a distance of only forty-three miles from Molokai, found 3,023 fathoms, or over 18,138 ft. Add this to the height of Maunakea, the highest point in Hawaii (13,805 ft.), and we have 31,943 ft., or 3,773 ft. higher than the loftiest peak of

the Himalayas, which is only 28,170 ft. Am I therefore justified in slightly doubting this “sunken continent” theory, and in thinking that our equatorial belt and polar depressions must have been fairly fixed in their present positions at the original cooling of the planet, and, from the volcanic phenomena in the Pacific, that our views of geological science must be modified.

I would point out the great lengths and breadths of the volcanic lines I have been speaking of. The islands forming the Wallis Group show the 200 ft. upward or outward thrust equally with the Tongan islands. These small islands do not show any lines of fracture on the land, but rather a distinct outward thrust. It appears as if the earth's crust uprose in this special geographical area in prehistoric times some 200 ft. to 250 ft. in one gentle movement, since which time the islands have remained about stationary, the ocean waves, however, washing them slowly away. Their flat tops or sharp peaks to me show every evidence of their former submergence. Nevertheless, each insular spot may only be a local upheaval, like the Sandwich Island volcanoes. But I am endeavouring to prove that there has been upheaval, and not all subsidence, in the Southern Pacific (the present accepted belief). It looks as if there has been a tilting, the islands along the 20th parallel of south latitude rising, and those at the equator and to the north of it sinking. Then, a crumpling is evidently taking place between Tahiti and South America, the Andes rising slightly and the ocean-bed sinking slightly, the Paumotus being proof of the upheaval being all old coral. Of course, I do not doubt a local slickenside action at any one spot, such as occurred at the Hanmer Plains, in New Zealand, or at Tacoma quite lately, on the North Pacific coast of America, when 600 ft. of docks belonging to the Northern Pacific Railway Company disappeared into the bay. The local subsidences are common in all volcanic regions, and easily understood. But what I am now pointing out are the evidences of a former bulging-out of the earth's crust some 200 ft. to 250 ft. in this particular ocean-area. Nor do I think that any particular harm at the time was done by such a displacement of the water. The tidal wave formed by the movement would have been very serious certainly, and injurious to any native peoples then inhabiting the shores of the ocean near this area, but I doubt whether the peoples surrounding the Indian or Atlantic Oceans would have been much affected by it.

Mr. J. P. Russell, of Wangaimoana, Palliser Bay, New Zealand, pointed me out the spot where an anvil of his had been carried up by the tidal wave (caused by the earthquake here in 1854) some 30 ft. above high-water mark. I dare say

an earthquake such as that of 1854 would form as big a tidal wave as the great and general former upheaval I refer to in the islands of the Pacific.

In the boat-cove at the entrance to Vavau Harbour I saw fractures in the coral caused by upheaval, but these may only have been local. The terraces at Cape Quiros, in Espiritu Santo (New Hebrides), should be counted, as these terraces may represent successive upheavals.

The following is the account of the Tacoma disaster: “More complete details relative to the Tacoma disaster are given in a Reuter's telegram dated Tacoma, Washington, 29th November. A mysterious accident which resulted in great damage to property occurred here last night. At 11 o'clock a loud roaring was heard, like that preceding the advance of a tidal wave, and 600 ft. of the docks suddenly disappeared into the bay. Two steamers were disabled and sunk. The ground in the vicinity subsided to the extent of 6 in. to 1 ft., causing a panic and stampede among the crowd which had collected in the vicinity. The cattlepens of the Northern Pacific Railway, and the company's offices, besides a freight-house 1,400 ft. in length, collapsed, the last mentioned catching fire. Various theories are advanced as to the cause of the disaster. The steamboat men maintain that it was due to a tidal wave 25 ft. in height, while others assert that, owing to a submarine land-slide, a great fissure or hole was formed beneath the bay, causing the docks to be swallowed up. Two lives were lost.”

I ought, perhaps, now to refer to Easter Island. We all know of the stone images there, 5 ft. to 37 ft. high, but usually 15 ft. to 18 ft. These are all cut out of a grey compact lava found in the crater of Hotuiti, at the east end of the island, where there are still many in an unfinished state. Their shape is the human trunk, terminating at the hips, the arms close to the sides, the hands sculptured in low relief, and clasping the hips. The head is flat, and the top of the forehead cut off level, so as to allow the crown, which is made of red tuff (found in the Te Rano Kao crater), to be put on. The face is square, massive, and sternly disdainful in expression, the aspect always upwards. Easter Island is volcanic, and has numerous extinct craters rising from different parts of the island, none of which have been active for a long time. The red tuff found in the Te Rano crater, from which the crown of the images are made, shows previous submarine volcanic origin; calc. tuff—which I suppose this red tuff to mean—being a mineral nearly identical with limestone and marble. The statement of the present inhabitants of Easter Island that their ancestors cut these images need not be credited. As already mentioned, my opinion is that the images were cut

by a race of people previous to the special local migration of the present inhabitants. I know of no race of islanders in the Pacific now acquainted with sculpture; neither do they possess even the tools with which they could do the work.

There is an exception, perhaps, to these two statements—viz., the arragonite money of Uap (or Yap), in the Carolines, and the Spaniards may have taught some of the Western Pacific islanders, since the sixteenth century, to roughly carve in coral. But the extraordinary money mentioned is composed of large discs of arragonite, often of great size; 6 ft. in diameter, 12 in. thick, and about 3 tons in weight are not uncommon dimensions. It is not used as a medium of exchange, but for purposes of ostentation. The arragonite is brought from a quarry in the Harbour of Malakal, at Korror Island, in the Palao or Pelew Group.

At the Duke of York and New Britain Islands it was the custom for a chief to place all his treasures before a visitor, and after inspection to have them put away. Thus several large coils of cowry money, about the size of lifebuoys, were placed before the Rev. Messrs. Brown and Fletcher at Blanche Bay when opening the first mission to those islands in 1875. These islands are about thirteen hundred miles from Uap, not an excessive distance for a canoe voyage in the Pacific. It would be well if some officer of our warships visiting the Carolines would inform us how this arragonite money is cut and removed from the quarry.

The trinolith and Langiis at Tongatabu (of which I present photographs) were cut from the coral reef, I believe, as the stones are not far from shore. (I am waiting Mr. A. W. Mackay's paper for a full and minute description of these ruins.) Arragonite is a mineral essentially consisting of carbonate of lime, and much like calcareous spar. The two minerals only differ in their form of crystallization. The rhombic prisms of arragonite are easily divided by the hammer, so that there should be little difficulty in quarrying them. Arragonite is a mineral found usually in volcanic districts, and in the neighbourhood of hot springs. Its crystals are sometimes prisms shortened into tables, which this money resembles. It appears to be the product of a crystallization taking place at a higher temperature than that in which calcareous spar is produced, showing great submarine volcanic action in days gone by at Uap. I should consider that the coral-polyp first deposited the lime, great subsidence and followed by volcanic action subsequently breaking up and converting the reef into columns of basalt, calcareous spar, and arragonite.

Mr. F.J. Moss, in his book, “Through Atolls and Islands in the Great South Sea,” writes of Ponape (or Ascension

Island, the largest of the Seniavine Group in the Carolines) and its neighbouring islets “as being a number of volcanic islands varying in size, representing the mountain-tops of an ancient land,” quite forgetting that the basaltic columns of which the ruins at Ponape are composed are really a variety of volcanic lava formed from lime. The ancient mountains must first of all have the lime deposited upon them, then have been subjected to great volcanic action, which melted them, and afterwards threw them up from the sea. Evidences show that these islands are now sinking again, which looks to me as if the bed of the Pacific ever since the globe cooled has been the seat of a constant and steady deposit of lime and subsequent great volcanic action, including upheaval and subsidence.

I might mention here that Ponape itself rises some 2,860 ft., its shores and hillsides strewn with loose blocks of basalt, many of them perfect hexagonal prisms of considerable size. Mr. Moss considers its summit “as probably to have formed the backbone of an ancient great mountain-range of the submerged continent.” I refer members to his interesting work. Yet he tells us “that so thickly is the place strewn, so numerous are the basaltic blocks, and so extensive an area do they cover, that it looks as if the whole island had been at one time terraced and cultivated, and that these rocks and prisms are the ruins of the terraces washed or fallen from the hills to the shore below.” But the basaltic columns forming the walls of the great temple ruin at Ponape show little or no wear from water. Their rhomboids and angles are still intact, which makes me think that the volcanic action was submarine, and that subsequent upheaval tumbled the columns and broke them up as they now are found. Moreover, the particular islet upon which this great temple is found, like many of the other islets near it, is embanked with massive walls of the same style as the building. “These careful embankments, the great walls, and the solemn silence gave to the whole the appearance of a city dead and deserted now, but with canals once crowded with canoes filled with devotees eager to attend the savage rites and sacrifices of which the ruined mass before us may have been the sacred scene.” Ancient Mexican history and rites are recalled to mind by this extract. The embankments forming these canals show that the sea was encroaching when they were built. Also the fact that the canals are still wadable shows either that they could not have been embanked so many centuries ago or the extreme stationary condition of sea- and land-level for the past 3,500 years in that locality.

There is little doubt to my mind that colonies from ancient Mexico and Peru voyaged westward into the Pacific and left

traces of their handiwork at Easter Island and on Tonga. The Carolines doubtless were colonised from Asia.^*

Again, Mr. Moss says, “The rude character of the structures is apparent. Not a vestige of art or workmanship of any kind is to be seen.” Now, the Langiis at Tongatabu show sculpture, as do the images at Easter Island.

At Tele, forming one of the Ualau or Strong Island Group (Ualau rises 2,000 ft., and is a basaltic island), are similar interesting ruins built of enormous blocks of basalt, showing that the people who shifted them about knew how to handle great weights, like the pyramid-builders in Mexico, Peru, and Egypt, and those who erected the stone ruins we find at Stonehenge or in northern France.

There are also several artificial canals and a canoe-harbour at Tele. These artificial canals show an acquaintance with the ancient canal system of Mexico and Peru, and that the migration of people who built them may have come direct from America. The ruins in Tele are stated by the natives to have been built by the former inhabitants partly for their defence and partly in honour of the dead, the large blocks of stone being brought from the main island on rafts. This, again, shows an acquaintance with the manner the red-granite blocks of Syene in Upper Egypt were anciently rafted down the river to Lower Egypt in flood-time, also with the way the Assyrians rafted great weights down their rivers. I do not think the present inhabitants of any of the Carolines know how to remove these heavy weights. A great weight is sometimes moved in the islands by rolling or pushing it into deep water, lifting it and fastening it under two strong canoes, and then sailing or paddling it to any required distant point. But this, again, only proves an acquaintance with ancient raft-movement, as it were.

Captain C. A. Bridge, of H.M.S. “Espiègle,” read an excellent paper, I believe, upon these ruins before the Royal Society, but I have not seen it. All that I wish to show is the part played by volcanic action in their origin.

The Hogulu group of islands, in the Carolines, is composed of ten lofty basaltic islands and numerous coral islands enclosed in a vast lagoon like a large lake in the sea. Yap, or Uap, or Guap is also of volcanic origin.

Asking pardon for this digression, I now proceed onwards to the eastern Pacific, so as to finish this portion of my subject.

Some of the Paumotus are rising, some sinking, some in a state of rest. Henderson Island is 80 ft. high. Pitcairn and the Gambiers rise to 1,000 ft. Osnaburg Island appears to have changed since 1790 from a “reef of sunken rocks” to an island fourteen miles long; whilst Archangel Island (20° 29′ S.) appears to have sunk out of sight since 1606. Actual volcanic action is most peculiar. It really appears to delight in confining itself (in the one given line) to one spot or pipe at a time, striking a blow or thump from below to “knock out,” as it were, one identical spot; which, if weak enough, “gives” to the imprisoned giant; and we have an elevation or a crater. At sea this thump from below strikes a vessel as if she had gone crash upon a rock. Quite recently a vessel sailing from San Francisco to Japan met with such an experience in the northern Pacific, and it made the crew dazed and sick. The greater the pressure of water the more the imprisoned giant likes to assert itself; but directly it meets with no resistance it acts quite gently, and expands its force in slight elevations above the sea. But I regard volcanic phenomena as the great ultimate friend of man, notwithstanding any immediate damage they may do.

Many of these Paumotu Islands descend sheer 1,500 ft. to 4,000 ft. within 1,000 yards of the reef. Aurora Island (15° 48′ S.) is an uplifted coral island about 230 ft. high, with the usual perpendicular sides found all along this line, and nowhere else that I know of on the planet except in Possession Island, in the Antarctic, where Sir James C. Ross landed in 1841. The Tongan Islets are positively square, having flat tops and straight sides. Of course, there are many coasts with perpendicular cliffs, but what I wish to say is that these straight-up-and-down islands show upheaval, and not subsidence.

Of course, a vast deal of geological work is required to be done in the Pacific, now the very home of volcanic activity, as it were, of this planet—here and in the antarctic region. In Appendix B I give a brief account of the antarctic volcanoes from a paper by Captain Borchgrevinck, in order that all the information I can collect of volcanic action within the sphere of the Pacific Ocean, as it were, may be collected in one paper for the use of future observers.

The Marquesas lie nine hundred miles to the north-east of

the Society Islands, nearly midway between them and the equator, being situated between the parallels of 8° and 11° S. and the meridians of 138° and 141° W. They have a warmer climate than Tahiti, and are all mountainous and volcanic, rising to upwards of 5,000 ft. above the sea (Magdalene, 3,675 ft.; Santa Christina, 3,280 ft.; Adams Island, 4,042 ft.; and Masse Island, 2,000 ft. high). The mountain-peaks are extremely broken and rugged, and the centres of some of the islands are occupied by piles of rocks of most fantastic shape. The volcanic precipices in many places extend abruptly down to the sea, presenting barren walls of black and naked lava; but the intermediate valleys are singularly fertile and picturesque, and are copiously watered by streams which descend in numerous cascades, one of which (in Nukahiva) is 2,000 ft. high, and is amongst the most beautiful in the world. They have no active volcanoes, and do not appear to be subject to earthquakes.

From Angas's “Polynesia” I also extract certain remarks upon active volcanoes in the Pacific; also an extract from Miss Bird's “Hawaiian Archipelago” (see Appendix A).

I refer briefly to the Sandwich Islands. On the 24th February, 1877, a slight shock of earthquake was felt at Kaavoloa, Hawaii, and steam was observed to be rising from the sea off Cocoanut Point. On visiting the spot it was found that lumps of porous lava, some nearly a cubic foot in size, were rising to the surface, when, on the contained gas escaping, they sank again. At the time of the earthquake a crack opened in the ground from Cocoanut Point in an east-south-east direction, extending for more than a mile, in some places 4 in. broad and 50 ft. deep. (This, again, shows the east-south-east trend. I have sometimes in this paper referred to the trend as south-east and north-west: I believe I should be more correct in saying east-south-east and west-north-west.)

Mauna Haleakala, on the Island of Maui, is somewhat like Mauna Kei, in Hawaii. The craters upon it are inactive, the natives having no tradition of any eruption.

Space does not permit me to refer to the phenomena in the various other islands of the Sandwich Group. Oahu (on which is Honolulu) is the principal island of the group, and the extensive plain on which that city stands is purely volcanic. About three miles north-west of Honolulu there is a remarkable circular salt-water lake, about half a mile in diameter, so impregnated with salt that twice every year the natives take out large quantities of fine, hard, clear, crystallized salt, which furnishes a very valuable article of commerce. At the time of the visit of the United States Exploring Expedition it was believed by the natives to be fathomless, but on examination by Commodore Wilkes it proved to be only

18 ft. deep. I mention this now as I have often heard of other lakes in the Pacific believed by the natives to be similarly fathomless.

It will be noticed that the Sandwich Island volcanoes are quite outside of my three lines of phenomena. But Kilauea may only be a huge safety-valve in this particular portion of the earth's crust, showing a great and permanent fault near it. That it has long been so the immense height of the volcanic lava, cinder, and ash heap forming the mountain (nearly 32,000 ft.), with a base of a hundred miles in diameter, proves; so that this safety-valve must have retained this one escape for many thousands of years; or it may be that the whole bed of the ocean for more than a thousand miles round the group has been slowly subsiding, and that the volcanoes on the Sandwich Group, and Cotopaxi and others in Central America, are safety-valves. Certainly the islands in the Pacific on and immediately to the north of the equator, as I of have pointed out, have been also slowly subsiding. The result of this subsidence has been upheaval along the 20th parallel of south latitude, as the evidences show.

I might also point out that the trend of the Sandwich Group, south-east to north-west, somewhat contradicts my theory of islands north of the equator trending south-west to north-east, like the Japan, Kurile, and Aleutian Islands.

I am particular in giving members all the information I can upon this subject, principally in the southern Pacific (without making detailed reference to the New Zealand craters, which more able observers have described), so that the heights and distances of the various active and extinct volcanoes from each other may be seen almost at a glance by any persons studying the map of the ocean. It would be necessary to examine the records taken by the “Challenger” and other expeditions as to the depths of the Pacific Ocean between the different groups of islands. It is a mistake to suppose that there are vast stretches of ocean-bed between these groups, because that is not generally the case by any means. For the fixity of all the continents and oceans'; the limitation of the supposed great glacial epochs and, ice regions almost to their present position; the fixity of the poles and the equator from the original setting of the planet after cooling to the positions we find them now, would follow as corollaries to my present doubt of the former existence of a great southern continent in the Pacific. But why the shores of the Pacific Ocean and its central southern bed should be subject to so much volcanic action is remarkable. We find few such phenomena around the shores of the Atlantic or Indian Oceans The fact that active volcanoes are confined within a short distance of sea-coasts might show a weakness in the

crust of the earth where the great sea-beds join the continents. But I do not see how this weakness could extend right across the arch in the crust of the earth under the great sea-beds from 40° S. latitude to 50° N. latitude, yet it is known that the great earthquake of Lisbon in 1755 did shake such an arch of the earth's crust over an area of 700,000 miles.

This, however, is a question to be hereafter considered. My present task is to set down the actual facts as I find them in the Pacific. I have thought it right to add to the two lines of volcanic action I first purposed pointing out the line of upheaval along the 20th parallel of south latitude, from the Loyalty Islands to the Cook Group, as well as the other facts collected. I have not seen them specially referred to by any other observer before. Huge active volcanoes exist in Central America, the Sandwich Islands, Vulcan Island, and New Guinea. With Mount Hecla near the North Pole, and Mounts Erebus and Terror (and other active volcanoes in Graham Land) near the South Pole, the Pacific volcanoes, with Vesuvius and Etna in the Mediterranean, are nearly the only great safety-valves the planet now possesses. Evidences of extinct volcanoes are abundant, but these are now far removed from sea-water.

The matter has its practical side, too, seeing that the market need never be short of sulphur whilst there are such great deposits of that mineral in Tanna or Ambrym in the New Hebrides, or at the Mother and Two Daughters in New Ireland, &c. The cultivators of the grape-vine in Australia use hundreds of tons of sulphur. They will find plenty in the different spots I have named.

Our own White Island, off Poverty Bay, is still in a state of volcanic activity, and must be regarded as the summit of a crater but little elevated above the sea. It emits from time to time volumes of white smoke. It produces, as I have said, a great quantity of sulphur. Several cargoes have been sent to Europe, and realised £8 a ton. It is very pure, containing 90 per cent.

The intermittent action of volcanic energy referred to is noticed in nearly all active volcanoes and geysers. It is very marked at Tanna, and in some of our New Zealand geyser-fountains the discharge is very regular as to time. There is, indeed, one geyser in our hot-lake district so regular in its discharge that I think it is called the “twelve-minute geyser” (eight to ten minutes at Tanna, and ten to fifteen minutes at Ambrym).

The crater at Cotopaxi is situated in latitude 0° 41′ S. and longitude 78° 42′ W., at a height of 19,493 ft. above the present level of the Pacific Ocean, showing that once a safety-valve is formed the internal fires keep as much as possible to

loose slags or hardened streams of lava, and whose summit, would have presented a rugged wall of scoria, forming the rim of a mighty cauldron; but instead of this we found ourselves on the edge of a steep precipice, with a vast plain before us, fifteen or sixteen miles in circumference, and sunk from 200 ft. to 400 ft. below its original level. The surface of this plain was uneven, and strewed over with large stones and volcanic rocks, and in the centre of it was the great crater, at a distance of about a mile and a half from the walls of the precipice on which we were standing. Our guides led us round towards the north end of the ridge, in order to find a place by which we might descend to the plain below. The steep down which we scrambled was formed of volcanic matter, apparently a light-red and grey kind of lava, vesicular, and lying in horizontal strata, varying from 1 ft. to 40 ft. in thickness. In a small number of places the different strata of lava were also rent in perpendicular or oblique directions from the top to the bottom, either by earthquakes or other violent convulsions of the ground connected with the action of the adjacent volcano. After walking some distance over the sunken plain, which in several places sounded hollow under our feet, we at length came to the edge of the great crater itself, where a spectacle sublime and even appalling presented itself before us. Immediately before us yawned an immense gulf, in the form of a crescent, about two miles in length, from north-east to southwest, nearly a mile in width, and apparently 800 ft. deep. The bottom was covered with lava, and the south-western and northern parts of it were one vast flood of burning matter, in a state of terrific ebullition, rolling to and fro its fiery surge and flaming billows. Fifty-one conical islands, so to speak, of varied form and size, containing so many craters, rose either round the edge or the surface of the burning lake. Twenty-two of them constantly emitted columns of grey smoke or pyramids of brilliant flame; and several of these at the same time vomited from their ignited mouths streams of lava, which rolled in blazing torrents down their black indented sides into the boiling mass below. The grey and calcined sides of the huge crater before us; the fissures which intersected the surface of the plain on which we were standing; the long banks of yellow sulphur on the opposite side of the abyss; the vigorous action of the numerous small craters on its borders; the dense columns of vapour and smoke that rose at the north and south ends of the plain; together with the ridge of steep rocks by which it was surrounded, rising probably, in some places, 300 ft. or 400 ft. in perpendicular height, presented an immense volcanic panorama, the effect of which was greatly augmented by the constant roaring of the vast furnaces below.” At night the grandeur of the scene

reached its climax. “The dark clouds and heavy fog that after sunset had settled over the volcano gradually cleared away, and the fires of Kirauea, darting their fierce light athwart the midnight gloom, unfolded a sight terrible and sublime beyond all we had seen. The agitated mass of liquid lava, like a flood of melted metal, raged with tumultuous whirl. The lively name that danced over its undulating surface, tinged with sulphurous blue or glowing with mineral red, cast a broad glare of dazzling light on the indented sides of the insulated craters, whose roaring mouths, amidst rising flames and eddying streams of fire, shot up at intervals, with very loud detonations, spherical masses of fusing lava or bright ignited stones. The dark, bold outline of the perpendicular and jutting rocks around formed a striking contrast with the luminous lake below, whose vivid rays, thrown on the rugged promontories and reflected by the overhanging clouds, combined to complete the awful grandeur of the scene.”

From Miss Bird's “Hawaiian Archipelago” I extract the following description of the great lava-flow of Kilauea of the 2nd April, 1868 (I think it only right to include it here): “I could fill many sheets with what I have heard, but must content myself with telling you very little. In 1855 the fourth recorded eruption of Mauna Loa occurred. The lava flowed directly Hilo-wards, and for several months, spreading through the dense forests which belt the mountain, crept slowl shorewards, threatening this beautiful portion of Hawaii with the fate of the cities of the plain. Mr. C. made several visits to the eruption, and on each return the simple people asked how much longer it would last. For five months they watched the inundation, which came a little nearer every day. Should they fly or not? Would their beautiful homes become a waste of jagged lava and black sand, like the neighbouring district of Puna, once as fair as Hilo? Such questions suggested themselves as they nightly watched the nearing glare, till the fiery waves met with obstacles which piled them up in hillocks, eight miles from Hilo, and the suspense was over. Only gigantic causes can account for the gigantic phenomena of this lava-flow. The eruption travelled forty miles in a straight line, or sixty including sinuosities. It was from one to three miles broad, and from 5 ft. to 200 ft. deep, according to the contours of the mountain-slopes over which it flowed. It lasted for thirteen months, pouring out a torrent of lava which covered nearly three hundred square miles of land, and whose volume was estimated at 38,000,000,000 cubic feet! In 1859 lava-fountains 400 ft. in height, and with nearly equal diameter, played on the summit of Mauna Loa. This eruption ran fifty miles to the sea in eight days,

but the flow lasted much longer, and added a new promontory to Hawaii. These magnificent overflows, however threatening, had done little damage to cultivated regions, and none to human life; and people began to think that the volcano was reformed. But in 1868 terrors occurred which are without precedent in island history. While Mrs. L. was giving me the narrative in her graphic but simple way, and the sweet-wind rustled through the palms, and brought the rich scent of the ginger-plant into the shaded room, she seemed to be telling me some weird tale of another world. On the 27th March (five years ago) a series of earthquakes began, and became more startling from day to day, until their succession became so rapid that ‘the island quivered like the lid of a boiling pot nearly all the time between the heavier shocks. The trembling was like that of a ship struck by a heavy wave.’ Then the terminal crater of Mauna Loa (Mokua-weoweo) sent up columns of smoke, steam, and red light; and it was shortly seen that the southern slope of its dome had been rent, and that four separate rivers of molten stone were pouring out of as many rents, and were flowing down the mountain-sides in diverging lines. Suddenly the rivers were arrested; and the blue mountain-dome appeared against the blue sky without an indication of fire, steam, or smoke. Hilo was much agitated by the sudden lull. No one was deceived into security, for it was certain that the strangely pent-up fires must make themselves felt. The earthquakes became nearly continuous; scarcely an appreciable interval occurred between them; the throbbing, jerking, and, quivering motions grew more positive, intense, and sharp; they were vertical, rotary, lateral, and undulating, producing nausea, vertigo, and vomiting. Late in the afternoon of a lovely day, 2nd April, the climax came. ‘The crust of the earth rose and sank like the sea in a storm.’ Rocks were rent, mountains fell, buildings and their contents were shattered, trees swayed like reeds, animals were scared and ran about demented, men thought the judgment had come. The earth opened in thousands of places, the roads in Hilo cracked open, horses and their riders and people afoot were thrown violently to the ground; ‘it seemed as if the rocky ribs of the mountains and the granite walls and pillars of the earth were breaking up.’ At Kilauea the shocks were as frequent as the ticking of a watch. In Kau, south of Hilo, they counted three hundred shocks on this direful day; and Mrs. L.'s son, who was in that district at the time, says that the earth swayed to and fro, north and south, then east and west, then round and round, up and down in every imaginable direction, everything crashing about them, 'and the trees thrashing as if torn by a strong rushing wind.'

He and others sat on the ground bracing themselves with hands and feet to avoid being rolled over. They saw an avalanche of red earth, which they supposed to be lava, burst from the mountain-side, throwing rocks high into the air swallowing up houses trees, men, and animals, and travelling three miles in as many minutes, burying a hamlet with thirty-one inhabitants and five hundred head of cattle. The people of the valleys fled to the mountains, which themselves were splitting in all directions, and, collecting on an elevated spot, with the earth reeling under them, they spent the night of the 2nd April in prayer and singing. Looking towards the shore they saw it sink, and at the same moment a wave, whose height was estimated at from 40 ft. to 60 ft., hurled itself upon the coast and receded five times, destroying whole villages, and even strong stone houses, with a touch, and engulfing for ever forty - six people who had lingered too near the shore. Still the earthquake continued, and still the volcano gave no sign. The nerves of many people gave way in these fearful days. Some tried to get away to Honolulu; others kept horses saddled on which to fly, they knew not whither. The hourly question was, What of the volcano? People put their ears to the quivering ground and heard, or thought they heard, the surgings of the imprisoned lava-sea rending its way among the ribs of the earth. Five days after the destructive earthquake of the 2nd April the ground south of Hilo burst open with a crash and roar which at once answered all questions concerning the volcano. The molten river, after travelling underground twenty miles, emerged through a fissure two miles in length with a tremendous force and volume. It was in a pleasant pastoral region, supposed to be at rest for ever, at the top of a grass-covered plateau covered with native and foreign houses, and rich in herds of cattle. Four huge fountains boiled up with terrific fury, throwing crimson lava, and rocks weighing many tons, to a height of from 500 ft. to 1,000 ft. Mr. Whitney, of Honolulu, who was near the spot, says, ‘From these great fountains to the sea flowed a rapid stream of red lava, rolling, rushing, and tumbling like a swollen river, bearing along in its current large rocks that made the lava foam as it dashed down the precipice and through the valley into the sea, surging and roaring throughout its length like a cataract, with a power and fury perfectly indescribable. It was nothing else than a river of fire from 200 ft. to 800 ft. wide and 20 ft. deep, with a speed varying from ten to twenty-five miles an hour.’ This same intelligent observer noticed as a peculiarity of the spouting that the lava was ejected by a rotary motion, and in the air both lava and stones always rotated towards the south. At Kilauea I noticed that the lava was ejected in a southerly direction.

Christensen and Sarsee, and through Mount Haddington, an extinct volcano in Trinity Land, to Paulet and Bridgman Islands' active volcanoes. The volcanic zone bends easterly from here on account of the easterly trend in the fold, which appears to make a loop towards South Georgia before it swings back towards Cape Horn. That there is a real easterly trend in the earth-fold at Trinity Land and the South Shetlands is proved by the observations made by the “Astrolabe” and “Zélée” expedition, which record a strike in a north-north-east and south-south-west direction to the greyish-white limestones and phyllite-schists at the South Orkneys. Toward Cape Horn from near South Georgia the fold probably trends west-north-westerly, then follows an approximately meridional direction parallel with the chain of the Andes.

It may be noted, however, that, whereas the Erebus chain of Victoria Land is on the east side of the fold, the Christensen-Bridgman group are apparently on the opposite side. This may be due to the fact that at the latter locality the eastern slope of the fold is steeper than the western, as seems probable from the presence of the deep ocean abyss east of Graham Land, as shown on Dr. Murray's map. It is probable, therefore, that the volcanic chain of Victoria Land will continue towards the south pole, probably bending somewhat to the eastward, and will thence change its position to the fold on the other side of the antarctic continent, so as to run through the Christensen-Bridgman lines of volcanoes. In any case it is almost certain that high land, covered, of course, more or less by snow and glaciers, will be found at the south pole.

The honour of being the first man to discover the antarctic continent probably belongs to Captain James Cook, who, in the year 1772, reached latitude 71° 10′ S. in longitude 106° 54′ W., where he sighted the great ice-barrier which formed the seaward boundary of Antarctica. Speaking of this discovery, Sir James Clark Ross says, “I confidently believe that the enormous mass of ice which bounded his view when at his extreme south latitude was a range of mountainous land covered with snow.” In 1819 William Smith, in the brig “William,” discovered the archipelago of the South Shetlands, south of Cape Horn. In 1820–23 Weddell visited the South Shetlands, including the active volcano Bridgman. Powell, the discoverer of the South Orkneys, visited the volcanic island of Bridgman in 1882, and found it to be at that time 200 ft. high. Weddell, who visited it during the following year, estimates its height at 400 ft., and describes the island as being of sugar-loaf shape, whereas at the time of Powell's visit there was a crater on the west side

the south-east culminates in Castle Rock, 1,724 ft. in height. From the top of this peak the southern prolongation may be traced for several miles. Exactly how far south the dyke reaches I cannot say, the only satisfaction derived from an attempt to solve this question being that of spending a miserably wet night in the bush. This much is, however, certain: that the dyke extends in a fairly straight line for nine miles, and may possibly reach to Sugar-loaf Hill, in the Waiwawa Creek, a distance of fourteen miles from Kikowhakarere Bay.

The life-history of Castle Rock, or Motutere, to use the more euphonious native name, may be here briefly sketched. It owes its birth to the same agency that covered the whole of the Hauraki Peninsula area with its immense deposit of igneous rocks—viz., the folding of the Palæozoic slates in a direction parallel to the protaxis of New Zealand. From the concomitant line of weakness welled forth immense flows of lava similar to, but lesser in degree than, those of the Hawaiian Islands at the present day. After the deposition of the Upper Eocene andesites a period of quiescence ensued, to be broken in Miocene times by volcanic outbursts at Coromandel, and on the east coast from Port Charles to Whangapoua. That these outbursts were very different in character from preceding eruptions is clear from the nature of the ejecta, and from the numerous dykes that everywhere radiate from the foci of eruption. For both the breccias and the dykes of this period indicate a superabundance of steam, in the former case by the propulsion of débris from a crater, and in the latter case by the rending and fissuring of the adjacent rock.

In the case of the Castle Rock dyke, the centre of eruption was probably to the west of Kikowhakarere Bay, for at the south end of this bay occur true lava-flows of an identical rock. Moreover, the coarser breccias of Beeson's Island, a little to the south, are petrologically identical with the rock of the dyke and of the lava-flow, thus pointing to a common origin.

In few places along its course does the dyke show greater resistance to the weathering agents than the enclosing rock. At Motutere itself the rock is seen to be horizontally columnar, the columns being about 10 ft. in length and 1 ft. to 2 ft. in diameter. Additional evidence in favour of slow local cooling is furnished by the superior hardness and porphyritic nature of the rock at this place.

In hand specimens the rock is grey, with large porphyritic hornblende and feldspar crystals. The specific gravity is about 2.6. Under the microscope the base is seen to be feldsitic, apparently the result of devitrification. The feldspars are both plagioclase and orthoclase, the former

the geology of a strange district enables one to excuse Mr. Reynolds for not being able to see the facts which support that theory; but when he hastily arrives at conclusions about this wonderful formation, and then bases estimates thereon involving the expenditure of thousands of pounds, I cannot help thinking that a little more caution would have been advisable. An eminent geologist from Australia told me he would require at least three months' residence here before he could express any opinion upon the geology of this district.

The origin of the Boulder Bank is a geological question, and most geologists when dealing with the subject have spoken more or less cautiously, realising the difficulties of the problem. The drift theory from Mackay's Bluff might be the true solution, but there is a good deal of evidence in favour of the other theory, that the Boulder Bank is the upturned edge of a stratum of rock or boulder drift.

I shall now proceed to give the facts upon which this theory is based, and then mention the difficulties which make it almost impossible to accept the opposite theory. In the hill above the Rocks Road may be seen a series of stratified rocks, inclined at very high angles. These rocks incline outwards from the face of the cliff, or, in geological language, they dip easterly. On the beach below the Rocks Road are also rocks standing almost on edge. In some places they have been planed down almost to a dead level by the action of the sea, and are covered by every tide; but in other parts, where the rocks are harder or less exposed, they are standing up in wall-like ridges above the general level of the beach. These rocks underlie, and are therefore older than those seen in the cliff. This series of rocks may be traced seawards till the Arrow Rock is reached, upon which they evidently lie at a very high angle. Now, it is strikingly apparent that the rocks in the cliff, and also those on the beach, must at one time have extended seawards in the form of an arch, forming a rounded hill. The Arrow Rock must therefore be regarded as the core of this hill, and this hill was not a mere cone, but had extension northwards and southwards. The strike of the inclined rocks on the beach and in the cliff is north-north-east and south-south-west, and it is also a remarkable fact that the Boulder Bank lies in a similar line. Keeping in mind the fact that a hill once covered the Arrow Rock, that this hill had a northward extension, as is shown by the parallelism of the stratified rocks in that locality, it follows naturally enough that a ridge of hills once occupied the present site of the Boulder Bank. The core of this ridge is represented by the Arrow Rock, which so far has resisted the denuding action of the sea. This, of course, is deduction, but it is deduction based upon solid facts which cannot be gain-

said. This ridge of hill, being composed of soft sandstones and clays similar to the rocks in the cliff and on the beach, has, with the exception of what remains at the cliffs, been washed away. The hard core of this ridge, composed of rock similar to the Arrow Rock remains, however, as an upstanding reef, forming the basis of the Boulder Bank.

Having shown that the existence of a reef under the Boulder Bank is highly probable, the next point for consideration is the nature of the rock of which this reef is composed. That the reef is similar to the Arrow Rock has already been remarked, but the Arrow Rock is only a fragment of an extensive belt of rock, and probably does not adequately represent the whole. An inspection of the Arrow Rock shows that it is made up partly of solid syenite rock and partly of syenite boulders firmly cemented together into a conglomerate. There is also in the syenite an intrusive sheet of lava. The boulders referred to form a part of the Arrow Rock, but lying around its base there are numbers of loose boulders which at one time doubtless formed part of the solid mass. These boulders are syenite, and quite similar to some of those found on the bank.

When the Torpedo Corps were improving the entrance to the harbour they blasted away rock made up of syenite boulders, and owing to its stubborn resistance found they could make but very little impression upon it with their charges of gun-cotton. Such, then, is the kind of reef that probably under lies the Boulder Bank—a boulder stratum, underlain by solid syenite, turned up on its edge by the upheaving force that raised the Port Hills.

There is no need to go into the origin of this boulder stratum beyond stating that it is probably the result of glacial action. In the cliffs above the Rocks Road several boulders of syenite have been unearthed. These stones prove conclusively that boulders were being carried in that direction while the rocks in which they are imbedded were being laid down as horizontal strata.

Given such a reef as has been described, then the origin of the Boulder Bank and the formation of Nelson Haven become simple matters, unbeset by any difficulties, and easily understood.

This theory of the Boulder Bank does not preclude the possibility of drift having come from Mackay's Bluff. In all probability—one might almost say certainly—boulders and shingle have drifted from there along the bank. This theory of the underlying reef, however, gives to the Boulder Bank its alignment, and removes some of the difficulties which make it hard to accept the purely drift theory. Some of these difficulties will now be mentioned.

It is a geological certainty that on the bottom of the greater part of the inner harbour—perhaps the whole of it—there are rocks standing practically on end. How far this arrangement of rocks passes beyond the Boulder Bank it is difficult to say, but that they do pass beyond must also be considered a certainty. Now, if the purely drift theory is to be accepted, we must suppose that these rocks were planed down by the action of the sea for at least 20 ft. below low water before any boulders were deposited. This, of course, is not an impossible thing to happen. Geological sections often reveal such an arrangement of rocks on their edges covered by horizontal strata. But in this case there is a difficulty. If these rocks were planed down, say, 20 ft. below the level of the sea, why were the Arrow Rock and the reefs near it not also planed down to the same depth? How would this denuding action reach from Mackay's Bluff to the Arrow Rock and then stop short? The thing is well-nigh impossible; when the sea planes it usually planes pretty smoothly. Then, supposing the planing-down process to the depth indicated did actually take place, what became of the boulders that were being formed at Mackay's Bluff at the same time? Did they stay there till the adjacent rocks had been planed down 20 ft. below sea-level and then start drifting south-west? Not a likely thing to happen. Another difficulty against the acceptance of this theory is the largeness of some of the stones at the extreme south of the bank. The advocates of the drift theory realise the difficulty, and suggest that heavier seas must at one time have prevailed in the bay. Now, how could those heavier seas have been produced—seas mighty enough to roll huge boulders ten miles along a level sea-bottom at a depth of 20 ft. from the surface? Only by so altering the coast-line that Tasman Bay would be more exposed to the ocean than it is at present. To effect this change of coast-line in the direction indicated we should have to give up some of our fundamental ideas of New Zealand geology. It would involve first the submergence and then the reappearance, in comparatively recent geological time, of some of the land by which the bay is at present sheltered. Then, too, the absence of gradation in the size of the stones on the bank, when viewed lengthwise, is another difficulty against the acceptance of the drift theory. It is a fact that some of the stones at the south end of the bank are as large as any to be found at the north end of the bank. If all these stones had been drifted from one place, one would expect to find large stones near the source, and a gradual diminution in size down to gravel or even sand as the distance from the source was increased. But such is not the case on the Boulder Bank.