Description of the District.

The district in which the volcanic eruptions took place is hilly and open, hut with a band of forest running through the lake country to the Patetere plateau. North of Lake Tarawera this band of forest is about ten miles broad, but west of Rotorua it attains a much greater breadth; it is, however, broken by a strip of open land round Rotorua and the west end of Rotoiti. South of Lake Tarawera the country is covered with fern and stunted tea-tree (Leptospermum).

The highest points are—Mount Tarawera (3,609ft.), on the south-east side of Lake Tarawera; Ngongotaha (2,554ft.), west of Ohinemutu; Haroharo (2,529ft.), south of Rotoehu; Whakapongakau (2,524ft.), between Rotoiti and Lake Otakaina; and Moerangi (2,440ft.), north of Rotokakahi. The lakes form two groups, the first of which contains Rotorua, Rotoiti, Rotoehu, and Rotoma. These lakes lie in the same valley, and were formerly united, but have been separated by the lowering of the drainage-channel. They perhaps owe their origin to eruptions from Haroharo blocking up the valley. The second group includes Otakaina, Okereka, Tikitapu, Rotokakahi, Tarawera, and Rotomahana; they may possibly owe their origin to former eruptions of Mount Tarawera. The heights of the lakes above the sea are thus given by Dr. von Hochstetter : Rotorua, 1043ft.; Tarawera, 1,075ft.; Rotomahana, 1,088ft.; but these figures are probably too low, as he made Ngongotahi to be 2,282ft., or 272ft. less than the height given by the Survey Department.

Hot Springs.—Nearly all the mud-volcanoes, fumaroles, siliceous springs, and solfataras of New Zealand are included in a zone, about twenty miles broad, running north-north-east from Lake Taupo; bounded on the north-west by the Patetere page 4 plateau, and on the north-east by the Kaingaroa Plains. To this zone Dr. von Hochstetter applied the name "Taupo zone:" in it he recognized three parallel lines. The first is the main line which extends from Tongariro to White Island : it includes Rotomahana as well as the hot springs of Kakaramea and the valley of the Waiotapu. The second line is about four miles to the north-west of the first, and includes the springs on the north-west side of the Paeroa Range and at Orakei-Korako. The third line is fourteen miles from the first, and passes through Ateamuri, Ohinemutu, Tikitere, and Rotoiti.^* He considered that the second and third of these lines were due to faults by which the low land between the Paeroa Range and the Patetere plateau had been thrown clown; the steam escaping through the fissures produced by the dislocations,^†

The recent volcanic eruptions took place on the first line. The new craters are divided into two groups : (1) on Mount Tarawera, (2) on the plains from the foot of the mountain through Rotomahana and the valley of the Haumi nearly to Lake Okaro.

Observers.—On the night of the eruption the only Europeans on the south and cast of Tarawera were at Galatea, on the Rangitaiki River, sixteen miles from the mountain; but the place lies low, and the mountain cannot be seen. On the north-west side European observers were at Wairoa (seven miles), Rotorua (fifteen miles), and Lake Rotoiti (eighteen miles). From near Wairoa the whole of Mount Tarawera can be seen, but from Rotorua the south-west end of the mountain (Tarawera proper) is hidden. Prom Rotoiti the view is entirely cut off by Whakapongakau.

Geology.

The only published account of the geology of the district is the survey of Dr. P. von Hochstetter, made in 1859 for the Provincial Government of Auckland. We have found this account to be very accurate, and have merely added to it a few additional observations of our own.

Volcanic Rocks.—The central portion of the district is composed of rhyolitic rocks, which are bounded on the south-east by the Whakatane Mountains, a high range lying on the east bank of the Rangitaiki River, composed of Palaeozoic sandstones belonging to the Maitai system. To the west and north they are page break

Plate 1.

Summit of Mt Tarawera from Paeroa. 2^ND July 1886.

page 5 bounded by the trachytic tuffs and breccias of the Patetere plateau and of the hills between the north side of Rotoiti and Matata. Dr. Hochstetter considered that the trachytic rocks of the Patetere plateau were an older formation than the rhyolites of the Taupo zone.

The rhyolites form all the hills from the south side of Rotoiti and Ngongotahi to Mount Tarawera. They surround Lakes Rotokakahi, Tarawera, and Rotomahana, and stretch across the Kaingaroa Plains to Galatea and south-west to Ateamuri. No other volcanic rocks were previously known in the neighbourhood of the recent eruptions; but one of the Okaro craters shows in its walls a dark augite-andesite, and has thus got the name of the Black Crater. The rhyolites are partly the stony variety with grains and crystals of quartz known as liparite or quartz-trachyte, and partly the vitreous variety known as pitchstone—the vitrophyre of Vogelsang. This pitchstone varies much in character, and is often spherulitic; but we have seen no true obsidian in the district.

Mount Edgecombe on the north-east, and Tauhara on the south-west, near Lake Taupo, are characteristic volcanic cones, both marked by Von Hochstetter as trachytic; but in the rhyolitic district there is no true cone unless it may be Haroharo, near Rotoehu.

Mount Tarawera.—Mount Tarawera is a flat-topped ridge about three miles long and half a mile to a quarter of a mile broad, running in a north-east and south-west direction. It rises precipitously from an elevated plateau deeply cut into hills and valleys by the rain (see Frontispiece). On the eastern side a long spur, covered with forest, runs out towards Mount Edgecombe, and on the south-west side another shorter spur stretches to the south. The ridge itself is divided by a saddle, estimated by Mr. Percy Smith at 500ft. deep, which divides a smaller north-west portion, called Wahanga, from the chief part of the mountain. The highest peak is called Ruawahia, and is situated just south of the saddle, while the southern peak of the ridge is called Tarawera by the Maoris. By Europeans, however, the name Tarawera is generally applied to the whole mountain, including Wahanga. It bears no resemblance to a volcanic cone, whether composed of tuft or of viscid lava; and Mr. Percy Smith, who ascended it three times previous to the eruption, states that there was no appearance of a crater on the top. On a point of so much importance we may be permitted to quote Mr. Smith's own words, as he is the only competent observer page 6 who has visited the mountain. He says, "Prior to the eruption the two mountains of Wahanga and Ruawahia (for Tarawera is only a local name on the south end of Ruawahia) formed two high table-lands of about three miles in total length by about half a mile in width, divided by the saddle before referred to, the top of which was covered with large angular fragments of trachyte, which had the appearance of having been shivered into pieces by frost; and the top was further divided into hillocks by deep crevasses running irregularly in all directions. The edge of this table-land has steep, precipitous, rocky sides, falling off into gentle slopes all round, on which were several forests of considerable size—now, alas, all destroyed."^*

Lake Rotomahana.—Before the eruption Rotomahana was a shallow lake, with patches of raupo (Typha latifolia) of irregular form, less than a mile in length by a quarter of a mile in breadth, lying in a north-and-south direction. The northern and southern ends were low and swampy; the sides, both east and west, were higher and rocky, but the rocks were decomposed into soft fumarole-clays, red, white, and grey in colour. On the eastern side the hills rose rather abruptly to a height of about 200ft. above the lake, while on the west they sloped up, more gently, to what is called by Hochstetter the Papawera plateau, about 750ft. or 800ft. above the lake. South-west of the lake and about a mile distant is the hill Hape o Toroa, 2,300ft. (Hector) above the sea, and between 900ft. and 1,000ft. above the former level of the lake. Between this hill and the lake is another smaller one called Oruakorako. All these hills were covered with fern and tea-tree (Leptospermum).

Old Lake-basins and Sinter Deposits.—Extensive old lacustrine beds are found on the south side of Rotorua, forming the flat between the lake and "Whakarewarewa, and they extend along each side to Awahou and Te Ngae. These beds are horizontally stratified, but often current-bedded, and consist of sand and pumice-gravel with occasional beds of fine pumice-dust, evidently an old volcanic ash. At the base of the cliff's near Te Ngae we found, in one locality, a bed of angular rhyolitic gravel, the stones being of all sizes up to 3in. or 4in. in diameter. These beds are covered unconformably by a younger set of rhyolite grits, coloured yellow-brown by hydrous ferric oxide, which we had no time to examine sufficiently. Rhyolite grit with quartz occurs also on the Island of Mokoia to an altitude of about 100ft. above the lake.

Plate II.

Mt Tarawera from Near Galatea, 4^TH July, 1886.

South of the Hemo Gorge, by which the Taupo Road leaves the Rotorua basin, there is another old lake-basin, now entirely filled up, from the middle of which rises Haparangi, like Mokoia in Rotorua. This basin is ten or twelve, miles long by about four or five in breadth; but we had no time to examine it.

Old siliceous sinter deposits are found in many places on the south side of Rotorua; and the hill called Pukeroa, behind Ohinemutu, as well as Tamate Heria on the lake, are composed entirely of it. The sinter, both here and in the Rotorua Township, is overlain by the lacustrine pumice-sands already mentioned. We also saw some siliceous sinter by the side of the Taupo Road between Haparangi and Horohoro.

Records of Previous Eruptions and Earthquakes.

Volcanic Ash.—Near Ateamuri, on the River Waikato, pumice-sands are seen capping high flat-topped hills, and sections on the Taupo Road show two ash-deposits divided by a chocolate-coloured band representing an old surface-soil. The older of the two is pumice-ash with quartz, and contains masses of pumice as well as angular pieces of rhyolite and, very rarely, black scoria. It is of considerable thickness, but the bottom is not exposed. The upper ash, above the old surface-soil, is 10ft. page 8 thick in places, although usually less, and it can he traced north to Lichfield, where it thins out. It is composed of pumice-sand containing angular fragments of pumice and of compact rocks up to 1in. or 1½in. in diameter. These rocks are chiefly variously-coloured rhyolites; more rarely porphyritic basalt with olivine. At the base, just above the surface-soil, there is often a thin band of grey, very fine pumice-ash with small quantities of quartz. It is from½n. to¼in. thick, and is very like the fine ash lately thrown out of Mount Tarawera. The coarse pumice-ash contains fragments of carbonized wood, and shows no appearance of getting finer upward.

At Galatea, on the banks of the Rangitaiki River, overlying the liparites, there is a pumice-ash with quartz, very like the newer ash at Ateamuri and also containing carbonized wood. It is, however, covered by 2ft. to 4ft. of loose, coarse rhyolitic sand.

At Lake Rotorua the fine pumice-ash, already mentioned as occurring in the cliffs at Te Ngae, contains no quartz, while the beds below it are composed largely of quartz grains. This is probably owing to the ash having fallen into the lake and the quartz grains having sunk rapidly, while the lighter pumice remained in suspension for some time.

All the beds at Earthquake Flat are pyroclastic, but in the upper part, above the fumarole-clays, we have a fine pumice-dust (No. 3) overlying coarser beds much as at Te Ngae, and we cannot doubt but that this dust is the finer portion of the ash which fell into still water, and consequently settled after the coarser part.

Earthquake Fissures and Crateriform Hollows.—A well-marked earthquake fissure, which was noticed by Dr. von Hochstetter in 1859, crosses Earthquake Flat south of the point where the Galatea and Paeroa roads meet, and runs for about half a mile in a north-east and south-west direction—that is, parallel to the fissure which has lately been opened in Mount Tarawera. Crateriform hollows, like the one already mentioned in Earthquake Flat, occur in several places in the district, and are probably due to former earthquakes, like the circular hollows formed in Calabria by the earthquake of 1873,^* although ours are often much larger. In Calabria they were found to have a pipe at the bottom, filled with loose sand, up which the water spouted. This no doubt was the same with ours, for they closely resemble the so-called Okaro craters, lately opened, the

^* Lyell: "Principles of Geology," vol. 2, p. 127.

page break

South end of Rotomahana, 27^TH June. 1886.

From a Photograph by Wheeler & Son Christchurch.

page 9 most southerly of which shows a ring of red fumarole-clay round its mouth very like that seen in the hollow in Earthquake Flat.

No scoria cones were known in the district, but we were informed by Mr. Frazer, of Taheke, that on the south side of Mount Tarawera, in a line with Rotomahana, there used to be a hill about 400 yards long, 200 yards wide, and 120 yards high, rising up from a flat, with a crater-like hole on the top and hundreds of boulders of obsidian lying about it. The inside of the crater sloped and was covered with fern.

Dr. von Hochstetter also mentions that Rotomakariri, near Rotomahana, had curious circular coves which reminded him of the tuff-craters near Auckland;^* and he said that the solfatara of Ruahine, near Tikitere, had the appearance of being an active crater.^† Viewed with our present knowledge, we may well believe that the crateriform springs of the White and Pink Terraces, of Rotopunamu and other places on Rotomahana, commenced with explosions like those which happened on the 10th of last June.

Premonitory Symptoms.

These appear to have been very few and slight. Ruapehu, of which no previous record or even tradition of activity exists, had been observed to be steaming for several months previous to the eruption. On the 16th April, 1886, Mr. L. Cussen; ascended the mountain and found a steaming crater on the top; and on the next day he saw a large column of steam, 100ft. high, ascending from it. No extra activity has been observed in Tongariro, but Mr. A. B. Wright informs me that his men, who were camped at the north-east base of Tongariro on the night of the 9th June, heard a series of explosions come from the mountain, so loud and in such rapid succession that they expected an eruption would follow. They felt no earthquakes, although a week or so after the eruption Mr. Wright felt several severe ones between Tongariro and Lake Taupo, followed by a rumbling noise and distinct explosions in Tongariro. No unusual activity has been observed at White Island, and all through the eruption it appears to have retained its ordinary condition.

Earthquakes of a local character have occasionally been felt in the Tarawera district ever since Europeans inhabited the country, but during the last few months they had become much page 10 more common—not enough, however, to excite alarm, or even to arouse a suspicion that anything unusual was going to happen. The hot springs at Rotorua had been gradually declining, but this, as well as the low level of the lake, was, no doubt, due to the exceptionally dry season that had passed. On Monday, the 7th June, a party of excursionists from Wairoa visited Rotomahana, accompanied by the well-known guide Sophia, and reported nothing unusual there.

Heavy rain fell at Rotorua on the 4th and 5th June. From then to the time of the eruption it was fine at Rotorua, but at Wairoa it was showery on Wednesday, the 9th.

The self-registering barometer at the Government Sanatorium shows that at Rotorua on the forenoon of the 9th the glass gradually fell until 4 p.m., when it reached 29.00in. It then began to rise, and at 1 a.m. on the 10th was at 29.10in., which it maintained all through the eruption.

General Account of the Eruption.

When it is remembered that the eruption of Mount Tarawera was quite unexpected, and that it occurred in the middle of the night in a hilly country sparsely inhabited, it will be allowed that a strictly correct narrative of the eruption is impossible. Nevertheless we are of opinion that the following gives a fairly accurate account of the sequence of events that took place.

Commencement of the Eruption.—The night of Wednesday, the 9th June, 1886, was fine and starlight, the moon in her first quarter, and it was nearly calm. At 12.30 a.m. on the 10th earthquakes commenced, slight at first, but increasing in intensity. At 1.15 or 1.20 a.m. Wahanga broke out with a flash followed by an explosion, after which everything remained quiet until 1.45 a.m., when the main eruption commenced, probably from Ruawahia. At 2.10 a.m. there was a violent earthquake, and immediately afterwards Tarawera broke out with a loud roar; and at 2.30 a.m. the whole mountain from Wahanga to Tarawera was in violent eruption. The column of steam was at first continuous, but afterwards broke up into seven or more distinct columns. These spread slowly out at the top into abroad black cloud. Forked lightning darted out from the column, and brilliant electrical scintillations illuminated the margin of the cloud. Showers of red-hot stones were shot upwards like rockets, and fell outwards in all directions; the column was also lighted up by the reflection from the red-hot rocks below. The red-hot stones were seen distinctly at Gis- page break

Plate IV.

Te Hape O Toroa. 16^TH June 1886.

From a Photograph by C. Spencer, Tauranca.

page 11 borne—seventy-five miles distant—and the height to which they ascended was calculated by Archdeacon Williams to be 32,000ft., which, however, seems to be excessive.

At 2.30 a.m. another black (one observer says yellow) steam column arose straight up from near Lake Okaro, and attained an elevation considerably greater than that from Mount Tarawera. All observers agree that this column came from a point much to the west of Rotomahana. It was thought to proceed from Kakaramea, but it is now known that the southernmost crater is considerably to the north of Kakaramea, but on the direct line between Kakaramea and Wairoa. At 3.30 a.m. very violent earthquakes and eruptions commenced, which were probably caused by the outbreak of Rotomahana. It is impossible to ascertain this with certainty, as no one was in a position to observe this portion of the line; but the evidence obtained by the boating party that visited Te Ariki on the 14th June is strongly confirmatory of the opinion that some of the Okaro craters were active before those of Rotomahana. The boating party found the Kaiwaka Creek, which formerly drained Rotomahana, to be dry, and with Avails 20ft. to 25ft. high; the lower; portion formed of scoria, stones, and white ash, covered by a layer of mud from 2ft. to 5ft. thick. The bed of the creek was dry and firm near Lake Tarawera, but higher up the bottom got soft. The layer of mud had been cut through as well as the scoria, so that the stream must have flowed after the mud had been deposited. This, therefore, must have taken place before the eruption of Rotomahana, which dried up the steam.

The steam columns from Rotomahana and Okaro are not reported to have been accompanied by red-hot stones, neither were they lit up by reflections from red-hot rocks. They spread slowly to the north, and obscured the cloud from the mountain.

Decline of the Eruption.—At 5.30 a.m. on the 10th the crisis of the eruption was over, although heavy earthquakes continued until 6 a.m. The pillar of steam from Mount Tarawera was seen at 9 a.m., and thunder and lightning-flashes continued to 10 or 11 a.m. Mrs. Blyth, who at this time was on the Kaingaroa Plains trying to get from Galatea to Rotorua, states that there was then no eruption from Wahanga or from Ruawahia, only at Tarawera proper, as well as at Rotomahana and Okaro. At 4 or 5 p.m. the eruptions from Rotomahana and Okaro were much less, although Tarawera was still very active and making a great noise. It continued in that state until about midnight.

On the morning of the 11th only small quantities of steam were issuing from the mountain, but the craters of Rotomahana and Okaro continued to throw out immense quantities of steam together with sand and stones. On Sunday, the 13th, the steam column was observed by Mr. Humphries at New Plymouth—120 miles distant—and calculated to reach 25,000ft. above the sea Stones were ejected from these craters for ten or twelve days but with diminished quantities of steam; and from that time to the end of our visit the amount of steam remained practically the same. On the 8th July, and again on the night of the 12th slight spasmodic increases in violence of the Rotomahana craters took place.

It thus appears that the eruption of the mountain began at 1.15 a.m., continued for about twenty-three hours, and then rapidly declined. The eruptions from the flat commenced at 2.30 a.m., were most violent from 3.30 to 5.30 a.m., them slowly decreased, but continued for ten or twelve days to eject solid materials, and, with the exception of the southern crater, they were during our visit gigantic fumaroles.

Results of the Eruption.^*

Openings formed in the Earth.

A comparison of photographs of Mount Tarawera, taken from Wairoa and from near Te Ariki before and after the eruption shows that very little change has taken place in the shape of the mountain. Wahanga, Tarawera (proper), and the saddle between Wahanga and Ruawahia appear almost unaltered, but the ridge between Ruawahia and Tarawera has undergone changes in out-line, although no true cone has been formed.

On the eastern side of the mountain, near the top, a long fissure extends from Wahanga to the southern slope of Tarawera (Plates I. and II.). This is seen very plainly from the Kaingaroa Plains, near Galatea (Plate II.).^† From Paeroa, which is on the direct line from Mount Tarawera through Kakaramea, this easterly fissure can also be seen, as well as a crater on the top of Tarawera, which is probably connected with the fissure. Plate I. represents this as seen through a telescope; the pointed hill on

^* Detailed accounts of the results of the eruption will be given by Professor F. D. Brown and Professor A. P. Thomas in separate reports. The outline here presented is only intended to explain the conclusions arrived at as to the cause of the eruptions.

^† In this plate the fissure is represented too far down the mountain; it ought to be close under the top from Tarawera to Ruawahia. The line of steam running down from Ruawahia towards the north may perhaps come from a lava-stream.

page break

Plate V.

The Okaro Craters from Pareheru, 27^TH June 1886.

From a Photograph by Wheeler & Son, Christchurch.

page 13 the right foreground being the summit of Kakaramea, which is just about half-way between Tarawera and Paeroa.

Half-way up the south-west side of Tarawera there is another fissure, which runs up to Ruawahia, and probably joins the easterly fissure. This westerly fissure was noticed soon after the eruption, and is the one mentioned by Mr. Percy Smith and Dr. Hector (D) in their reports to the Government. The easterly fissure is that marked A—C on Dr. Hector's diagram. These two fissures may, for convenience, be called the craters on the mountain.

On the plains, near the foot of the mountain, Rotomahana has been much enlarged, including now Rotomakariri to the north-east and beyond the Pink Terraces to Oruakorako on the south-west. The bottom does not appear to be deeper than it was before, but it is filled with mud in which are several crater-Rings, either steaming or full of water. Plate III. is taken from Oruakorako, looking east. Between Rotomahana and the foot of Mount Tarawera a new lake, with precipitous walls, has been formed; it is about a mile long and of irregular outline.

South-west of Rotomahana, in the valley of the Haumi, there is a row of six or eight crateriform hollows, known as the Okaro craters (Plates IV. and V.). The third of these craters from the south, or the Black Crater, lies at the foot of a low hill;^* which circumstance has led to the opinion that the hill itself has been formed by ejections from the crater, and it was called Mount Haszard. The hill, however, is part of the old surface, and was in existence long before the eruption. These openings from Okaro to the foot of Mount Tarawera may be called the craters on the plains.

Our knowledge of the size and positions of all these openings at present very imperfect, and must remain so until an accurate topographical survey of them has been made.

Materials ejected.

Two different rocks have been ejected—viz., rhyolite and augite-andesite—and each is either compact or vesicular. The augite-andesite, when compact, is greyish-black, vesicular in places, and with opaque-white angular fragments of decomposed rhyolite. In the vesicular state it is black scoria.

The rhyolite is of several varieties, but all are pale in colour, and more or less decomposed by the felspar being kaolinized. In page 14 the vesicular state it is white glassy pumice, with quartz grains more or less abundantly developed.

From the mountain came vesicular scoria and pumice as well as fine ash. It always fell dry, and was hot eighteen miles from the mountain. Scoria is far more abundant than pumice.

From the plains came compact blocks, grit, and dust, chiefly rhyolite, without any pumice or scoria. These fell warm close round the openings, but farther off as intensely cold mud, and still farther as fine dry dust.

The boundary between these two deposits is tolerably well marked, and passes along the western base of Mount Tarawera through the eastern end of Rotoiti to Maketu. To the east of this line the deposit is all from the mountain, while to the west the deposit from the mountain is overlain by the deposit from the craters on the plains.

Cause of the Eruption.

Eruption of the Mountain.—The eruption of Mount Tarawera was a true volcanic explosion of a paroxysmal character, throwing up large quantities of red-hot scoria and pumice, although no lava is known to have been emitted. It may also be looked upon as the opening of a new volcanic vent; for, although it is probable that further investigation may prove Mount Tarawera to be an old volcano, still it is evident that the new fissures are not in the same position as the old crater. The great interest, therefore, from a scientific point of view, is that we see in it the method of establishment of a new volcanic vent, while the eruption was so short and so violent that the openings have not been buried, but can be examined at leisure, which is not the case with either Jorullo or Monte Nuovo.

An examination of the ejectamenta from the craters shows that—(1) A complete series can be made, from compact andesite with decomposed rhyolite to the same rock highly vesicular, but still showing fragments of the rhyolite and quartz; (2) a similar series can be made from decomposed rhyolite to quartzose pumice; (3) the Black Crater has thrown out blocks of an old andesitic lava, which has overflowed a still older rhyolitic lava, the two being intimately connected at the line of junction. The same intimate connection between scoria and pumice is seen in fragments thrown out by Mount Tarawera, and it is very improbable that this similar intimate connection between two similar rocks should have been brought about in two different ways.

General View from Lake Rotorua

From this it follows that the scoria and pumice emitted by Mount Tarawera are old surface-rocks that have been lately heated up to the point of fusion, for if this were not so they could not have become vesicular. Evidently, therefore, we have in this reheating of surface-rocks, which had been previously saturated with water, the immediate cause of the eruption of the mountain.

If now we further inquire, How came these rocks to be reheated? we see at once that the cause was local, in the mountain, for the reheated rocks have all previously undergone atmospheric decomposition. Consequently the heat could not have been caused by upward conduction, through the solid crust, of the internal heat of the earth. If such had been the case the heat would have been widespread, and Rotomahana would have been heated as well as the mountain. No difference in the conducting power of rocks can explain the fact that Mount Tarawera was much more strongly heated than Rotomahana. No chemical changes, at all competent to do the work, suggest themselves as the cause of reheating.

The crushing of rock, due to strains in the mountain, and the conversion of these mechanical movements into heat, seems at first sight very plausible, as mechanical movements might be expected to occur along old lines of weakness, such as the line from Tongariro to White Island may be conceived to be. But when closely examined this theory is found beset with difficulties. It has been calculated that ten volumes of rock must be rapidly crushed in order to fuse one volume, and even then the heat of the ten crushed volumes must, in some way, be focussed into the one fused volume. Leaving, however, the localization of the heat out of the question, we find that in the case of Mount Tarawera several millions of tons have been fused and ejected, and therefore several tens of millions of tons must have been crushed but not fused. But very nearly all the material ejected by the mountain has been fused : where, therefore, is the immense amount of crushed but unfused material? and why was none of it ejected? Instead of the mountain having been crushed, there are now two deep gaping fissures in it; and photographs of one of them, taken by Mr. C. Spencer, show that it has solid rock on each side. Again, several tens of millions of tons of rock could hardly have been rapidly crushed without producing violent earthquakes; but the earthquakes which preceded the eruption were not violent, but so slight that many people, even in Wairoa, were not awakened page 16 until after the eruption had begun, and none of the earthquakes were so violent as those due to subsequent explosions of steam from Rotomahana. The heaviest earthquakes were after the eruption had commenced, not before it, and probably therefore all were the effects, not the causes, of explosions. Also many greater earthquakes have occurred in the district without causing an eruption; consequently the local production of heat by crushing appears to be very improbable.

The only other explanation left to us is that the local increase of temperature was due to molten rock having been forced upwards from below into the mountain. We know that molten rocks have, at some former time, been erupted all along the Taupo zone; and the fact that heat has also been lately developed in Ruapehu, without any accompanying earthquakes, points also to this movement of molten rock as the true explanation of the case. When we remember the slow conducting-power of rocks and the great thickness of surface that has been heated, it seems evident that this intrusion of melted rock into Mount Tarawera must have taken place a long time before the eruption; and this is in accordance with the facts that the heat had been developed in Ruapehu in April, and that for several months earthquakes had occurred occasionally in the Tarawera district.

Two hypotheses have been put forward to explain the upward rise of molten rock in the earth. The first is that incandescent rock lying deep down in the earth gradually absorbs condensed water-vapour by capillary action under great pressure, and the quantity gradually increases until the pressure of the overlying rock is overcome, and the water-vapour flashes into steam. The two principal objections to this hypothesis are—(a) that it is very doubtful if the necessary pressure exists in the interior of deeply-seated rocks to force the condensed water-vapour into the incandescent rock; and (6) this condensed water-vapour must have had the same temperature and pressure as the rock when it was absorbed; it is difficult therefore to see why after absorption it should have more expansive force than before it was absorbed. The second hypothesis is that the upward rise of molten rock is clue to pressure arising from movements of the earth's crust over a liquid or viscid interior. This hypothesis is, of course, open to all the objections that have been urged against a viscid interior of the earth. It will not, however, be necessary for me to discuss these rival hypotheses, for I do not sec that the present eruption throws any new light on the subject.

Part of the Black Crater Throwing up Stones

Eruptions on the Plains.—The eruptions from the Rotomahana and Okaro craters were very different in character from those of Mount Tarawera. No great heat was developed. They ejected no red-hot stones, nor scoria nor pumice, but up to the last threw out angular slightly-heated fragments of surface-rock. They were therefore not truly volcanic, but only hydro-thermal in action, although they ejected a large amount of fine ash, a phenomenon which was previously supposed to accompany true volcanic action only. It must be remembered, however, that the rocks through which these eruptions burst were already decomposed and soft, with water probably all through them.

The eruptions on the plains followed those from the mountains, and were, no doubt, caused by them. Their position shows that they are in some way connected with the fissure on the mountain; but they can never have been directly connected with that fissure, because, if they had been, (a) scoria and pumice would have been ejected from them as from the mountain, and (b) the eruption of Rotomahana would probably have preceded that of the Okaro craters instead of having followed them. Although following in a general way the direction of the fissure, the craters are by no means arranged in a straight line; indeed, those in and about Rotomahana are scattered over a belt at least half a mile broad. Also no fissure can be seen at the surface as it can on Mount Tarawera; the cliffs now surrounding Rotomahana being due merely to landslips of the much-decomposed rock. Each of the Okaro craters is isolated, so that it is possible to walk round any of them without crossing a fissure, and no fissure is visible on either side.

Still, the position of these craters evidently connects them in some way with an underground extension of the Mount Tarawera fissure. Probably molten rock was injected, as a dyke, into the fissure below, and the earthquakes, caused by the explosions at Mount Tarawera, deranged the subterranean water-reservoirs at Okaro, and allowed some of this water to find its way towards the heated dyke. Before reaching this, however, it was violently ejected as steam, forming the so-called Okaro craters. Subsequent earthquakes caused surface-fissures, which drained off the water of Rotomahana towards the incandescent mass, but this water also was flashed into steam at a considerable distance from the dyke, and consequently no direct communication was established between the subterranean fissure and the surface. The Rotomahana craters are, in fact, only colossal fumaroles, page 18 which we know from observation often commence by throwing up mud and stones.

Lessons of the Eruption.—(1.) The most important point on which this eruption gives evidence is, as I have already said, the method of establishment of a new volcanic vent; but there are other very interesting points in the eruption. For example:

(2.) It is proved that surface-rocks are reheated in a volcano.

(3.) Evidence is forthcoming that this reheating is not due to crushing. But on this head much more evidence can be collected by a close survey of the fissures.

(4.) Evidence is produced in favour of an opinion held by many geologists, that surface water cannot come into contact with molten rock by means of an open fissure.

(5.) Fine dust can be formed from soft rocks without much heat, and without true volcanic action.

Theory of Declining Volcanic Action.—Dr. von Hochstetter and all subsequent observers have agreed in the opinion that volcanic energy was dying out in New Zealand. This opinion was based on the theory of Waltershausen and Bunsen that our alkaline siliceous springs belonged to a second stage of decreasing temperature; the first, or hotter stage, giving rise to acid springs and solfataras.

There are several facts in the Rotorua district which this theory does not explain. For example, the decaying solfatara of Sodom and Gomorrah, between Rotorua and Whakarewarewa, has covered old sinter deposits with thick layers of sulphur; and Waikite, the principal geyser at Whakarewarewa, has lately formed a thin coating of sulphur over the upper portion of its sinter terraces, although the water from it gives a decided alkaline reaction at present. Nevertheless, the theory is no doubt correct on the whole, but it applies equally to dormant and to extinct volcanic districts. Whether the Taupo zone is merely in a dormant state and the late eruption indicates a renewal of activity, or whether the eruption was merely a spasmodic death-struggle, time alone can show; but the probabilities are much in favour of the latter supposition.

Part of the Blank Crater.