IV. The Eruption.

The story of the eruption has been so frequently told from the popular side that it is unnecessary to repeat here the thrilling experiences of eye-witnesses. The present account will therefore be confined to those points which are of geological interest, and enable us to understand the character of the eruption.

It should perhaps be stated that the district is very thinly inhabited, and that the nearest spectators of the beginning of the eruption—the Natives who lived at the Moura and Ariki settlements, on Lake Tarawera—did not survive to relate their experiences. The nearest observers were at the Wairoa (eight miles west of Mount Tarawera), Rotorua (fourteen miles to the north-west), and Galatea sixteen, miles to the south-east); and the hilly character of the country did not allow them a complete and unimpeded view of the eruption.

The evening of the 9th of June was fine and clear. There had been a little rain at the Wairoa, though none was registered at the meteorological station at Rotorua. The previous summer had been very dry, and the rainfall in May was 2.02in. A little rain fell on page 30 the 1st and 2nd of June; on the 4th and 5th 1.84in. fell. From that date until after the eruption no rain was registered by the rain-gauge at Rotorua. The variations of the atmospheric pressure as recorded by the self-registering barometer at Rotorua show no apparent connection between any atmospheric disturbance and the outburst of the volcanic forces. The barometer at 10 a.m. on the 8th stood at 29.44. From that time till 4 p.m. on the 9th it gradually fell, reading then 29.02. It then rose very slowly up to the time of the eruption early next morning, when it stood at 29.1 (The Rotorua station is 990ft. above sea-level.)

The moon was just past its first quarter, and there was to occultation of Mars by it at 10.20 p .m.; but no unusual meteorological or other phenomena marked the night until after midnight. About half-past 12 o'clock slight earthquake-shocks were noticed by some of the inhabitants of the Wairoa, and also by the Postmaster at Rotorua. They must have been of but little intensity, for most of those who were asleep at the Wairoa were not aroused till some time later.

The earthquake-shocks followed one another rapidly, and increased in violence. About half-past 1 a.m. the eruption seems to have commenced at Wahanga, the portion of the mountain to the north, for a column of dark vapour, described as black smoke, and probably charged with ashes, was seen rising from it by some of the inhabitants of the Wairoa. About half an hour later the dark column of vapour was also seen ascending from Ruawahia. Only a few people seem fan have seen these dark clouds; but all are agreed that at about ten minutes past 2 a.m. there was a violent earthquake, followed by a loud and prolonged roar. A black cloud could now be seen lying over the whole mountain, ascending in a broad column, which spread outwards at the top. This familiar feature in the behaviour of the steam-column rising from an active volcano is compared in Italy with the stone-pine tree so common there. At Rotorua it was compared with objects more familiar—viz., a mushroom or umbrella.

Red bodies, which were doubtless red-hot fragments of ejected rock, were now seen darting from the black cloud, whilst lightning! began to shoot out from it, accompanied by the roll of thunder, A red glow lit up the column; and from time to time, as a fresh outburst took place on the mountain, the clouds were lit up with stronger glow, and red-hot stones, described by observers as fire-balls were seen falling around the summit. By this time the mountain seems to have been in full eruption, and presented a magnificent spectacle. The electrical phenomena must have been of a very remarkable character.

It is not clear at what time the eruption commenced at Rotomahana, as lofty hills obscure the view in that direction, both from Rotorua and the Wairoa. It is probable, however, that the craters at Rotomahana were in eruption soon after 3 a.m., for about that : time the earthquakes were especially severe, the heaviest one noticed happening at 3.20 a.m. There is a similar uncertainty as to the hour at which the craters to the south-west of Rotomahana were opened; but the point is not one of much importance. It seems clear, however, that by half-past 3 the whole country was in violent eruption along a line nine miles in length, extending from the north-east end of Wahanga to a point 600 yards to the north-northwest of Lake Okaro. The observers at Rotorua noticed a distinct column of lurid yellowish vapour rising over the hills in the direction of Lake Okaro, so that it was thought that the lake had been blown up. The column, however, must have been from one of the last craters at the south-west end of the line.

It is important to note that, whereas the clouds rising from the Tarawera Range were illuminated with an intense red glow, doubtless the reflection of the red-hot lava in the craters beneath, the clouds from Rotomahana and the south-west were not lit up in the same way, nor were "fire-balls" observed there.

In the meantime the columns of vapour, as they rose from the mountain and the country to the south, spread out overhead in a dense pall of clouds. From this the stones and ashes commenced to fall. At the Wairoa the first stones fell about 3 a.m. The cloud spread slowly outwards, and as viewed from Rotorua appeared to be advancing slowly towards it, but also spreading out to the north and south. About 4 a.m. ashes began to fall lightly at Rotorua; but nearly at the same time a fierce south-west gale began to blow, and this drove the ash-laden cloud away from Rotorua and in the direction of the Bay of Plenty. As the cloud travelled before the wind it gradually dropped its burden of ashes over all the country between Rotorua and the sea.

The chief violence of the eruption appears to have been over before 6 a.m., the paroxysm having lasted only three or four hours. The ejection of matter was continued for some time longer, but with declining energy. Owing to the presence overhead of the dark clouds of vapour charged with ash, the district remained in total darkness till some time after the hour of sunrise. Even at Rotorua the darkness lasted till after 9 a.m. At the Wairoa mud or ashes continued to fall until 9 a.m., whilst to the east of Tarawera and in the direction the wind was blowing the fall of ashes and consequent darkness con- page 32 tinned still longer. At Opotiki (47 miles away) it was pitch-dark till 10.20 a.m., at which hour the fall of dust became lighter, and daylight gradually appeared.^*

The earthquakes attending the eruption have been described as severe or heavy. They were not, however, severe enough to upset bottles or other light objects in the houses at Rotorua, fourteen miles away. Nevertheless the shocks, and still more the accompanying sounds, were perceived at a much greater distance. At Auckland, distant 133 miles in a direct line from Tarawera, a large proportion of the population were aroused by loud detonations and the rattling of doors and windows, thought by most of those who heard them to be due to the firing of cannon. It was supposed by some that the sounds came from a ship in distress in the Manukau Harbour. The discharges, however, were not regular like those of minute-guns, but rather resembled the sounds produced by a train of siege guns, including large and small cannon. The column of vapour rising from Tarawera, with its accompanying flashes of lightning, was also visible from Auckland.

Judge Puckey informs me that the sounds were heard even as far as Hokianga, 253 miles north-east from Tarawera, and that next day the Natives who lived near the North Cape came into Hokianga stating that they had heard the strange noises during the night, and feared some disaster had happened to their people.

The sounds of the eruption were heard in an easterly direction from Tarawera, all along the Bay of Plenty. At Gisborne, ninety miles to the cast-south-east, most of the inhabitants were aroused about 2.30 a.m. "by the rumble of distant explosions" and the continuous rattling of doors and windows. The cloud over Tarawera was distinctly seen in the clear moonlit sky; its height was calculated by Archdeacon Williams to be a little more than six miles above the top of the mountain.^†

To the west of Tarawera the sounds reached Hamilton and Cambridge; to the south they were heard all around Lake Taupo, and as far as Wellington (228 miles to the south-south-west), and even as far as Christchurch (420 miles).

It should be observed that the detonations resembling the firing of cannon were not very noticeable at Rotorua, being lost in the general prolonged roar of the craters and rumble of thunder. sounds heard at a distance were probably due to the explosions of page break

Tarawera.

June 14th 1886.

From a sketch by S. Percy Smith.

page break

General View of Craters.

[unclear: by G. S]

page 33 steam deeper in the fissure, the vibrations of which were conducted through the earth.

Decline of the Eruption.—We have already seen that the period of intense paroxysmal eruption was over before 6 a.m. on the 10th June. The eruption continued for some time longer, though with rapidly-diminishing energy. It is impossible to state exactly the condition of the vents during the two or three following days, for the cloud of vapour hung around the mountain, and it was not till the morning of the 13th that its top was clearly seen. Until noon on the day of the outbreak the lightning continued to play around the clouds, and during the whole of the day an almost constant rumble was heard from Tarawera.

When the top of Tarawera Mountain was first seen on the morning of the 13th June, it was found that it had increased in height, whilst its outline was changed (see Pl. p. 32). Columns of steam rose at intervals from its whole length : these were of small size, but Rotomahana and some of the craters to the south-west were emitting vast volumes of steam (see Pl. p. 33). The column from Rotomahana, or rather from the huge crater which included the former site of Rotomahana, rose to a great height, so that it was clearly visible at Auckland 133 miles away. The elevation above the horizon, as observed by myself, corresponded to the height of 15,000ft. above Rotomahana. Mud and numerous stones were thrown out from Rotomahana crater and some others to the south-west for two or three weeks after the eruption; but these gradually diminished in quantity, until at the time of my visit in the following July the ejection of material was for the most part confined to a few stones thrown out now and again when a sharper explosion took place in the craters. Much steam was still emitted from Rotomahana, and usually filled the crater so as to completely obstruct the view of the bottom. At times, however, a favourable breeze would drive away the clouds, and afford a glimpse of the hideous mud-craters and mud-fountains which covered the bottom some 600ft. below the margin.

Occasionally short spasmodic outbursts occurred both at Rotomahana and Tarawera; but they were of brief duration and of little intensity, and not due to a revival of the deeper-seated forces, but (amply to minor explosions of steam accumulated in the rocks lying at the bottom of some of the vents. They were commonly accompanied by earthquake-shocks; on one occasion a shock was of sufficient intensity to overthrow light objects at Rotorua. The earth-quakes or earth-tremors were especially frequent after rainy weather, when half a dozen or more would be noticed in the course of a day. page 34 On two or three occasions I observed small clouds of steam rising from the summit of Tarawera immediately after an earthquake-shock

About the middle of July heavy rains fell in the Hot-lakes district, and the earthquakes were especially numerous. On the 21st July, having ascended Tarawera as far as the uppermost crater of the chasm on its south-west slope, I found the side of the mountain covered with a bright-red volcanic ash, reaching the thickness of an inch, which had evidently fallen since the rainfall, for below it the older grey ask was found with a smooth, caked surface where wet by the rain. The rain-water had apparently found its way down to the lava, which lay still hot in the bottom of the crater. Here it soaked into the lava and accumulated until the growing tension of its vapour produced an explosion.

The earthquakes felt in the district after the eruption were thus due to explosions of steam either on Tarawera or at Rotomahana. This is still further confirmed by the records of a simple form of seismograph set up by Professor Brown at the Rotorua Sanatorium, under the charge of Dr. Ginders. The tracings given by this instrument almost invariably showed that the earthquake-shocks originated at Rotomahana or at Tarawera. It should be mentioned, however, that two sharp shocks, which occurred on the 12th July, were shown by the seismograph to have originated not at Tarawera, but at some place a few degrees to the west of south from Rotorua. These earth-quakes were not felt at Tauranga to the north, but were felt strongly at Taupo, reaching the latter place one minute earlier than Rotorua, the time being compared by telegraphic communication between the post-offices. The origin of these earthquakes seems therefore to have been more to the south of the Taupo zone than Rotomahana, and the writer is inclined to attribute them to some place near Orakei-korako, Perhaps these earthquakes indicate sympathetic disturbances at another part of the zone.

Black Terrace Crater.—Mr. Lundius and the writer, whilst travelling over the mud around the Rotomahana Crater on the 21st July, noticed at one place, about a quarter of a mile from the edge of Rotomahana, steam issuing with hissing noise from cracks in the mud. A fortnight later a survey party observed large clouds of steam rising from the spot, and, on approaching, it was found that a new crater had been formed, which was throwing out stones and mud with great vigour. This crater was about 200 yards or so in diameter and occupied the position of the Atetuhi Geyser, known also as the Black Terrace, from the colour of some of the sinter there. This crater, the last which was formed during the eruption, was therefore page break page 35 called the Black Terrace Crater. It continued in active eruption for two days, and, with a smaller degree of energy, for some weeks longer.

The formation of the Black Terrace Crater was the last considerable event in the eruption so far as the subterranean forces were concerned; the changes which have since occurred in the craters are due rather to agencies acting at the surface, and will be considered : further on.

Alterations in the Rotorua Springs.—We have still to note some of the minor events accompanying the eruption. At the Native settlement of Ohinemutu, near Rotorua, a large number of new hot-springs broke out at various points, some of them on the public roads, and others even in the floor of the large Maori meeting-house—Tamatekapua. In front of the Government Agent's house a large spring of boiling water burst out; some months later this spring was still active, and its overflow formed a considerable rivulet of hot water. Most of these fresh springs were formed on the night of the eruption; a few, however, were formed during the following weeks.

The hot-springs at the Government Sanatorium gradually increased in temperature and volume of outflow in the days following the eruption. The Rachel Bath, the normal temperature of which is 170° Fahr., had the temperature of 176° Fahr. on the 13th June, three days after the eruption, and 190° Fahr. on the 16th. On the 1st July it reached 196° Fahr.; and on the 12th and 13th of the same month, after heavy rain, it reached 198° Fahr. At the beginning of July the spring which supplies the Blue Bath (a hot swimming-bath 62ft. in length) was 15° Fahr. above its normal temperature, with unusually copious discharge. At the same date the Priest's Bath, which is situated on the margin of the lake, was a few degrees below its usual temperature, but had a much greater outflow. Afterwards its temperature rose, and on the 4th September was at 104.6° Fahr., some degrees above the normal.

At Whakarewarewa, two miles south of the Rotorua township, the geysers have been distinctly more active since the eruption. This increased activity did not appear at the time of the eruption, but afterwards. For instance, on the 25th June a large and very deep pool of clear and almost boiling water, which had long been used by the Natives as a cooking-place, partly because it was always at the same high temperature and free from spasmodic action, suddenly burst into eruption, throwing out boiling water and stones. The Maoris, thinking of Tarawera, were terrified by so unusual a circumstance, and left the settlement for the time. Waikite, a large geyser page 36 which had been altogether silent for more than ten years, suddenly became active, and threw up a column of water 40ft. and, at times, 60ft. in height. Other geysers have also become more than usually active, and have remained so up to the present time.

Natives from Orakei-korako reported that the hot-springs then were much disturbed, both at the time of the eruption and afterwards. The earthquakes also were severely felt there.

A question of much interest at once suggests itself here: Do the outbreak of the new springs at Ohinemutu, the increased temperature of the baths near the Sanatorium, and the greater activity of the Whakarewarewa geysers indicate increased subterranean activity near Rotorua, sympathetic with the eruption of Tarawera? Tempting as such an hypothesis may seem, a simpler explanation of the phenomena is to attribute them to the opening by the earthquake-waves of the fissures by which the heated water or steam finds its way to the surface. It may seem strange that the increased action should have shown itself at Ohinemutu immediately, whereas at Whakarewarewa it only appeared some days after the eruption. It must, however, be remembered that the earthquakes continued for some weeks after the eruption, and that the formation of new springs at Ohinemutu was not confined to the night of the eruption, but took place to a certain extent afterwards.

The rise in the temperature of the springs at the Sanatorium may be explained as clue in part to the rise in the level of Lake Rotorua along its southern shores, which occurred on the morning of the eruption, as will be shown immediately. It has always been noticed at Rotorua that the hot springs are more active when the water in the lake is high.

Level of Lake Rotorua.—During the morning of the eruption Tarawera the waters of Lake Rotorua rose along the shores near the Sanatorium. At the Government baths a register of the height of the lake was kept, and it was found that the lake rose 18in. up to 6.30 a.m., then sank 6in., only however to rise again. At Ohineamutu a similar rise of the lake was noticed, and the water in the Utuhina Stream was observed to be flowing backwards by Mr. Alloam and others. The Natives who lived around the lake, and the boat men, stated that the lake had fallen at the northern end. As the matter was one of much importance, we sailed round the lake, and calling at various points, we were shown at the Ngae, Awabou, and Mokoia stumps which were said to have been recently uncovered by the fall of the waters. Such statements, if to be relied upon, would point to a tilting of the basin of Lake Rotorua. They were however page 37 called in question, and even the rise of the lake-level at Ohinemutu and the Sanatorium denied. Mr. James Stewart, C.E., had taken the levels along the projected line of railway to Rotorua only a few months previously, and, when these levels were taken again after the eruption, it was found that no change of level had occurred along the railway-line. It should, however, he stated that the railway first approaches Lake Rotorua at a point 1½ miles south of Awahou, and that it keeps at some distance from the shores, ending in the Government township a quarter of a mile from the lake. The observation may therefore he perfectly consistent with a local subsidence of the ground close to the southern shores of the lake. That the lake-level here did rise at the time of the eruption must be accepted as an established fact from the careful observations of the Government officials at the Sanatorium. This rise of level could not be due to any influx of rain-water, for there had been comparatively little rain lance the summer. The lake is always higher in winter than in summer: and, when some time later the usual winter rains came, there was an additional rise of the lake, its level being higher than had ever been noticed before, so that the Native cemetery on the point near the old submerged pa was washed out. Further, the rise at the time of the eruption was a permanent one. I observed the lake-levels with some care during the summer of 1886—87, and found that the waters did not fall so low as during previous summers. I had been well acquainted with the shores of the lake near Ohinemutu (having frequently bathed in the lake), and noticed that familiar landmarks remained submerged even at the end of summer. The tea-tree which grew just at the former winter-level of the water had been killed, and a few inches of water still covered the place where it formerly grew, the dead sticks projecting from the water.

There is therefore ample testimony that a rise of level of the lake along the southern shores from Ohinemutu to the Sanatorium took place at the time of the eruption, and that this rise was permanent. Unfortunately such clear evidence as to the levels of the other parts of the lake was not obtainable, and, accepting only such evidence as is perfectly clear, it follows that a local subsidence took place along the southern shores of the lake. These shores are largely composed pumice sands and gravels; they are undermined by hot-springs which bring much mineral matter in solution to the surface, to be deposited as sinter around the boiling springs, or to be carried in solution into the lake. The rocks must therefore be more or porous or cavernous from the removal of the mineral matter, and the earthquakes at the time of the eruption may have caused page 38 these porous strata to settle down, thus causing a subsidence of the surface. The settling-down of the strata would doubtless assist in the formation of new outlets for the heated water, which would now find a free passage upwards by the fissures opened by the passage of the earthquake-waves.

Earthquake-fissures.—About five miles to the west of Rotomahana is a valley-fiat about three-quarters of a square mile in area, known as Waikorua or Earthquake Flat (see Map II.). The valley is surrounded by hills, and its bottom is quite flat, and has the appearance of a lake-basin which has been filled up with material washed off the surrounding hills. Crossing the flat in a N.E. by E. direction are a number of old earthquake-cracks, as well as some holes said to have been produced by earthquakes. I counted as many as eleven of these parallel cracks : some extend completely across the flat; others cannot be traced so continuously. During the recent eruption several of the old fissures were reopened, and it will be noticed that their direction is nearly parallel to the direction of the great fissure formed in the eruption.

Other earthquake-fissures were also formed at the time of the eruption, as, for instance, on the hills near Pareheru, about a mile south-east of Earthquake Flat. Here a crack was formed which could be traced for the distance of over a mile along the sides of the hills, All these cracks are comparatively narrow, usually only a few inches wide and rarely reaching as much as a foot in width. Still farther to the south-east old earthquake-cracks may be noticed amongst the hills, and some of them were reopened during the eruption. In some cases these earthquake-fissures are also lines of fault, the ground to the south-east being thrown down a few feet. Most of them have a direction nearly parallel to those on Earthquake Flat and to the line of the Great Fissure. One crack, however, had a north-and-south direction.

It can scarcely be an accidental circumstance that the old lines of earthquake-fissure are parallel to the new and much more extensive fissure formed during the recent eruption. It must point rather to former earthquakes originating in disturbances at the same volcanic focus where the pent-up subterranean forces found vent in the eruption of 1886.

During the earthquakes which happened about the 20th July a fissure was opened on the road from Rotorua to the Wairoa. About I half-a-mile on the Rotorua side of the Tikitapu bush was a dry basin among the hills. The ground on every side sloped down to the bottom of the basin, so that there was no escape for the rain-water page 39 except underground. After the fall of the ash, water collected at the bottom of this basin, forming a pool across the road. The fissure which was subsequently opened could be traced for the distance of some 250yds., its direction coinciding with the longer diameter of the basin. Towards the central part it was of considerable breadth, and subsidences had occurred along it. The largest of the depressions was a hole having the length of 70yds., and extreme width in the middle of 34yds., though at the ends it was only 10yds. and 4yds. wide respectively. The greatest depth was 98ft., and the materials shown in the sides of the chasm were pumice sands and gravels in horizontal beds. The fissure was opened along an old earthquake-crack, and it seems probable that the existence of this crack had determined the course taken by the drainage from the basin. The subsidences were probably due to the fall of loosely-compacted pumice-strata into hollows produced by the removal or settling of material along this subterranean watercourse. Deep holes of various sizes are of common occurrence in the pumice-gravels of some parts of the Waikato Valley and the Patetere country, and these seem to be due to similar subsidences along underground watercourses in the porous strata.

Sympathetic Action at other Centres.—From what has already been stated it will have been inferred that the eruption of Tarawera had deep-seated origin. It docs not follow that because the eruption bad such an origin it would be accompanied by disturbances at other volcanic centres in the same district. Nevertheless, any such sympathetic activity would furnish interesting evidence that the eruption had its source in changes acting along the line of fissure of the Taupo zone at a depth far below the surface. Some evidence of this kind may be cited, as, for instance, the rise of steam from Ruapehu ft phenomenon' which has frequently been seen since the eruption. A few slight subterranean rumblings from Tongariro or Ngauruhoe were also heard by the shepherds and road-makers who were near the mountains on the night of the eruption.

The earthquakes originating to the south of Rotomahana (probably at Orakei-korako) may also be mentioned here (see p. 34).

The increased activity of the springs in the Rotorua district should be ascribed to the effect of the earthquake-waves, and of the rise in the level of Rotorua, rather than to any sympathetic action.

White Island is reported to have shown greater activity after the eruption; and even after allowing for the much closer attention which the island received then, and the natural tendency to exaggeration at such a time, the reports appear to have been well founded. The increase of activity was not continuous, but from time to time page 40 unusually large clouds of vapour were observed rising from the island.