light-waves begin to leap in it, and long, solitary pillars shoot towards the zenith. At this moment there comes life into the sky. From every quarter of the firmament streamers come rushing with the speed of lightning towards the zenith. The little, fiery tongues whirl round, or sway to and fro, appearing as though they were Cupids in golden mantles with borders of purple. They dart and leap in vain to reach the zenith; they begin to move wave-like, slower and slower; they seem to get tired, still they whirl on towards the north, when suddenly they lose in intensity, and, in a fraction of a second, vanish!

It is again dark and cold; a thin veil of light again begins to form over the star-covered sky. This is as the aurora appears in its grandest form, and any description of it would fail to give even an idea approaching its real majesty and even grandeur.

In addition to the meteorological and magnetic observations, those of the aurora borealis were also made during the Polar night by means of the well known theodolite, and from October the electricity of the air was also examined. On the two agreed dates, the 1st and 15th of every month, the magnets and the aurora were examined and registered every fifth minute, and during one hour, every twentieth second. Besides these observations, meteors and shooting stars were watched and carefully noted, attempts made to measure the quantity of the snow, measurements of the aurora borealis effected, along with astronomical determinations of hour and place, absolute magnetic measurements, simultaneous observations every twentieth second of the magnets, the aurora, and the electrometer, and researches on the moisture of the air, and the nightly radiation, while the temperature of the snow was examined at various depths.

Already in October the remarkable depressing influence which darkness exercises on the human mind, with which every one who has wintered in the Arctic regions is familiar, began to be manifest. In that month it was, however, felt only slightly, but with November it rapidly increased, and at the end of December it had reached "the first stage of insanity." This influence caused a remarkable dislike to conversation, accompanied by great lassitude. When lying down, phantoms of the scurvy crept over one's mind, and the thought uppermost was that here, next to us, the bodies of fifteen brave men were found in a horrible condition ten years ago. The best cure for this was, we found, an exhausting walk, a good dinner, and a few glasses of lime-juice accompanied with the cheering thought that our expedition formed one of the moments in the great work of the human race.

The moonlight during midwinter was very remarkable, and imparted in the day a transparency to the air which we had never seen before. The greatest mountains did not oppress the eye, but seemed to assume a lightness which made them appear as if they were floating on the dark background.

On February 19 the sun was to reappear, but already on January 23 it was so light that we could read fine print out of doors, and on February 8 we could, at II a.m., read the thermometers in the cage without a lantern. On February 19 the sun came at last. During these days the scenery was magnificent. On the light sky clouds of every shape floated, coloured in the loveliest tints by the sun's rays, while over the whole was cast a hue of purple and gold.

In the beginning after the sun's return, aurora were still seen in the night, but on March 25 we saw the last of this phenomenon. Eventually on April 19 the sun became circumpolar, and from that date we had perfect daylight.

We often noticed during the spring a thick, cold haze lying over the landscape, in which mock suns and some other optical phenomena were frequently seen, caused by the reflection of the sun's rays in the ice-crystals.

The fjord was in the light period entirely covered with

ice, and, as the sun reappeared, even the open leads which could be seen between the ice-floes became covered with thin ice. Only far out on the horizon above the fjord a "water cloud," bespeaking open water, could be seen, and the increase or decrease of this we watched with great interest.

The migratory birds now began to arrive, and the Procellaria glacialis was already seen on February 7. On April 13 the first snow-sparrow came, soon after followed by the auks, the rodges, and the seagulls. The ptarmigans, which had lived in flocks during the winter, now began to separate, and preferred the mountains to the plains.

The observations were steadily continued, and the particular object of the researches of the meteorologist at this period was the radiation from the snow's surface. We thus believe we have discovered that the thermometers in the cage did not give the true temperature of the air, which was to be tested by means of a "swing" thermometer, i.e. a thermometer fastened to a cord, and then swung rapidly round, as such a thermometer will give the air's exact temperature as near as possible. Under these observations, which were made every hour, it, however, often happened that the cord broke, and the instrument suffered injury. In order to avoid this a mechanism was constructed, driven by hand, which kept the thermometer in a constant rotary motion, and from May 4 until the end of the month, when the thaw set in, this thermometer was read every hour. Another subject also investigated, from February 15, was the temperature of the snow on the surface and at three different depths.

During the light period three hydrographic-magnetic excursions of research were made on the ice in the Ice Fjord, viz. on April 19, April 24, and May 24. The longest of these, the one on May 24, extended six miles from the shore, and it was very difficult work to drag the sleigh over the rough ice. The results of the same were several absolute magnetic measurements, observations of the temperature of the sea at various depths, and testings of the saltness of the water. The greatest depth found was 250 metres.

At the same time, while the snow still remained on the ground, several topographical works were effected. A base some 600 metres long was measured between the universal instrument and a pole south of the same, while two signal posts were erected on two crests south-west and north-east of the station, and three miles apart. Afterwards the greater base was determined by means of triangular measurements from the smaller, in order to serve as a basis for further work. In addition to this there was built, on the sun's return, an astronomical observatory for the universal instrument, which was finished on February 14, and finally a magnetic hut was built for the Wrede's variation instrument, finished on May 19.

There was, during the dark period, one question which was much discussed, and which we were anxious to test, viz. whether the Polar night has the effect of turning the complexion white. On January 23, therefore, when it was light enough to see out of doors, we assembled in the open to examine our faces, and the concensus of opinion was that the darkness had not affected the skin in the least.

In the end of May the thaw set in in earnest, and soon mosses and shrubs came forth. In the beginning of June the fjord was still covered with ice, but by the 11th it commenced to open towards the sea, and by the 21st it began to break up and drift. On July 4 the fjord was free

from ice.

The fauna now began to appear: thus already on June 2 the red blossoms of Saxifraga oppositifolia came out from the snow; on June 11 Salix polaris was in bloom, as well as Draba wahlenbergii, and soon the plains were covered with flowers.

At that time some exceedingly interesting experiments in horticulture were commenced. A small garden was first formed by breaking up the layer of turf on the surface, to enable the sun to thaw the frozen earth underneath, and in this manner sufficient mould was obtained to lay out proper beds. In these were then planted seeds, among others radishes brought from Sweden, while several species of the Spitzbergen fauna were planted here. Both flourished remarkably, as did also the rye and oats which we planted here. The latter grew well, although slowly, and were, at the end of July, six to eight centimetres long. Their growth was measured every fifth day, while studies of the sun's chemical influence on the same were simultaneously prosecuted.

The migratory birds continued to arrive: thus on June 2 the brent geese put in their appearance, and in great flocks took possession of the innumerable lagoons. They were, however, very shy, and comparatively few were shot. Of wild reindeer several were shot, and one Polar bear was seen, but escaped.

At last on June 26, at 4 p.m., the first reminder of the outside world appeared in the shape of a fishing smack, but, although every effort was made to attract attention, she passed northwards. On July 8 an expedition was despatched to Cape Staratschin, the "general post-office" of Spitzbergen, which brought back news, letters, and the literature of the civilised world for a whole twelvemonth, the period of our isolation.

Shortly afterwards we had several calls of Norwegian hunters, among whom may be mentioned the well known Capt. Kjeldsen, of the Isbjörnen, who participated in the Payer-Weyprecht expedition of 1872, and in the Austrian to Jan-Mayen, 1882-83. He made the remarkable report that he had found the sea at the Norse Islands early in July this summer entirely free from ice, not even seeing the "ice-blink," i.e. the light reflected from new ice formed out of sight. This was in the exact spot where the Swedish expedition was compelled to return on account of enormous pack-ice, at the same period in 1882. He was of the opinion that a steamer would have been able to penetrate very far north of the Seven Islands this summer.

In the middle of August the relief boat Urd arrived, and, after having cleared the houses, and nailed up the windows and doors, we went on board, and steamed out of the Ice Fjord on August 25, having for a period of exactly 400 days, contributed our quota to International Polar research.

THE WEIGHTS OF BRITISH NOBLEMEN DURING THE LAST THREE GENERATIONS T is of considerable interest to know in an exact way the amount of change that may have occurred in our race during recent generations. I therefore send the following results concerning the changes in weight, which I have calculated from data obligingly furnished to me by Messrs. Berry, of 3, St. James's Street, London. Messrs. Berry are the heads of an old-established firm of wine and coffee merchants, who keep two huge beam scales in their shop, one for their goods, and the other for the use and amusement of their customers. Upwards of 20,000 persons have been weighed in them since the middle of last century down to the present day, and the results are recorded in well-indexed ledgers. Some of those who had town houses have been weighed year after year during the Parliamentary season for the whole period of their adult lives. I examined two of the ledgers at my own house, and was satisfied of their genuineness and accuracy; also that they could be accepted as weighings in "ordinary indoor clothing" unless otherwise stated. Much personal interest attaches itself to these unique registers, for they contain a large proportion of the historical names in our upper classes.

I have ventured to discuss only a small and definite

part of this mass of material, and I selected the nobility for the purpose, because the dates of their births could be easily learnt, which had to be done in order to connect the years in which they were weighed with their ages at the time. They formed a more homogeneous group than one that included younger brothers and men about town, who marry late and lead less regular lives. I therefore begged Messrs. Berry to find a clerk for me who should make the required extracts under their direction in an anonymous form for statistical purposes. I also asked to be furnished with an alphabetical list of the persons weighed, that I might know generally with whom I was dealing, and that each schedule should bear a reference to the folio whence it was extracted, so that, whenever verification was needed, the original might be referred to. All this was done, and I am in possession of 139 schedules referring to as many different persons, namely, 109 peers, 29 baronets (who were added as makeweights), and I eldest son of a peer. They were born at various times between 1740 and 1830, or thereabouts. Each schedule gives the age and year of the several weighings, the highest and lowest weights recorded in that year, and a copy of such remarks as were entered at the time about the dress. An age-weight trace similar to those in Figs. 1 and 2 was plotted on a Specimens of the Age-Weight Curves of Individuals

FIG. 2.-One-fourth of the Series are less irregular than this Specimen. (The Lower Quartile.)

large scale on each schedule. My best thanks are due to Messrs. Berry for their careful oversight of the tedious clerical work and for the intelligent assistance they gave in having it satisfactorily accomplished.

The age-weight traces differ widely and in many ways: (1) in the annual range of weight, (2) in its fluctuations from year to year, (3) in the age at which the weight reaches its maximum, (4) in the bluntness of the culminating point.

The annual range is shown in Figs. 1 and 2 by the short, v rtical lines that connect the upper and lower contours. The top of each line corresponds to the highest weight recorded in the year to which it refers, and the bottom of the line to the lowest. I find the average annual range in my whole series of cases to be 6 lbs., and that, in the successive decades extending over ninety years, it has decreased prettily steadily from 7 lbs. to 5 lbs. This points to an irregularity in the mode of life that was greater two or three generations back than now, and we shall shortly see that it is by no means a solitary indication of this well known fact. It would be interesting to learn how much annual irregularity in the weight of an adult is consistent with perfect health.

1

The only evidence I know that could throw much light upon it is summarised in a Parliamentary paper on prison discipline, whence it appears (p. 54) that a certain amount of irregularity is normal among prisoners, that they are heavier in summer than in winter, and that the changes are abrupt; also, that fluctuations in weight, bearing no sort of proportion to previous changes of diet, are of constant occurrence.

I calculated a rough numerical measure of the irregularity of each trace for the purpose of classifying them. I did so on the same principle that one might adopt to measure the discursiveness of a rambling path, in comparison with that of a straight turnpike road between the same points, namely, by finding the proportion that the length of the one bore to the other. I measured the trace and also the general sweep of the trace with a mapmaker's "perambulator," divided one by the other, and corrected each result on the principle that a fluctuation of 12 lbs. in a man of 16 stone should not count more than one of 9 lbs. in a man of 12 stone. I also exercised some judgment in my measurements, to avoid the error of dealing with ups and downs in the trace that were apparently due to the fragmentary character of the observations (sometimes only one record in a year, and sometimes two), as if they were real fluctuations. Each available trace was marked on this principle, and the traces were classified according to their marks. Figs. 1 and 2 are the “quartiles" of this class.

much as Elaine did on the dints in the shield of Launcelot, and on looking at some huge notch in the trace, may hazard the guess, "Ah, what a stroke of gout was there!"

The age at which the weight reaches its maximum is earlier in the earlier generations. I attempted eye estimates, and found it comparatively easy to form them in respect to the traces of the earlier period, where the culmination was usually distinct, and found that it frequently occurred at an early age; the number of times in which it took place in the successive decades of life in those days being as follows: under the age of 29, 2 cases; 30-9, 5 cases; 40-9, 6; 50-9, 7; 60-9, 12; 70 and upwards, 2. In the latter generations the culminating point was frequently too indistinct to be localised, so that I am unable to offer a corresponding statement for comparison that would be trustworthy. In short, the development of the latter generations was more regular.

The clearest evidence of the different age-weights in the three generations, A, B, and C, is obtained by comparing their Means. The following is a brief numerical abstract of them to which the number of cases upon which each mean is based is added in a different type below it. The figures in parentheses are doubly meaned results, those to the left being derived from observations made at the ages of 26 and 28, and those to the right from observations at 68 and 72. For purposes of comparison I subjoin the weights of the professional classes, extracted by interpolation from the table, published by the Anthropometric Committee of the British Association in their Report, 1883, p. 40. The number of observations on which these are based, are given in a form that does not admit of strict comparison with those of my series. They are 24, for observations at the ages 30-35; 24, for 35-40; 44, for 40-50; 13, for 50-60; 5, for 60-70.

160,

FIG. 3.-Mean Age-Weight of British Noblemen in three successive generations

WEIGHT

One quarter of all the traces are more irregular than Fig. 1, one quarter are less irregular than Fig. 2, and the remaining two quarters lie between them. The "median" trace occupies the half-way position; it is unnecessary to reproduce it here, as an imaginary interpolation between Figs. 1 and 2 will suffice.

I next divided the traces into three divisions, A, B, and C, according to the dates of birth of the persons they referred to. It happened that each division covered a period of thirty years, so A, B, and C may be taken to represent three successive generations, born respectively between 1740 and 1769, 1770 and 1799, 1800 and 1829. The numbers of traces available for the present purpose were 21, 22, and 26 respectively. It appeared that the most irregular trace in Group C would rank only as the seventh in Group B, and as the fifth in Group A, and yet C contains the greatest number of cases.

There can be no doubt that the dissolute life led by the upper classes about the beginning of this century, which is so graphically described by Mr. Trevelyan in his "Life of Fox," has left its mark on their age-weight traces. It would be most interesting to collate these violent fluctuations with events in their medical life-histories; but, failing such information, we can only speculate on them,

Copies of correspondence between the Secretary of State for the Home Department and the Inspector of Prisons, &c., and the Report of a Committee, &c. Ordered to be printed May 20, 1864.

These figures are rendered much more expressive by translating them into smoothed curves; those from which A was drawn are shown by crosses; those from which C was drawn are shown by small circles; but those from which B was drawn are omitted for clearness' sake.

values, which is by no means so clear as may at first Whatever may be the exact significance of these mean sight be imagined, and whatever may be their absolute worth, which I do not rate very highly, there can be no doubt as to their differential importance. They show with great distinctness that the noblemen of the generation which flourished about the beginning of this century attained their meridian and declined much earlier than nearly a stone heavier at the age of 40. those of the generation 60 years their juniors. They were

The weights of these two generations were identical at the age of 62 or 63, but at that period of life the earlier generation was declining in weight with almost the exact

speed at which the latter was continually rising. The steadiness of the rise of the latter from early manhood to late years is very striking; it is almost in a straight line. I have not sufficient data to justify me to say when its curve culminates; I have closed it at 70 with a dotted line.

It is only necessary to add that the ledgers of Messrs. Berry are a quarry from which, with some labour, much further information of the kind just given might be drawn. Perhaps the publication of this paper will suggest methods of treating them that have not occurred to myself. FRANCIS GALTON

THE ERUPTION OF KRAKATOA1

"SIXTEEN volcanoes now working between the spot

where Krakatoa was before and Sebesie." Such was one of the first reports which was sent by cable to Singapore, and which we heard at Pontianak. Never before had we been so longing for news from Java, for when H.M. ship Hydrograaf steamed into the PadangTikar River, we heard heavy detonations and explosions like far-off shots, so that we were alarmed about Java. As we expected, our ship was soon ordered to survey the Sunda Straits. This survey was finished at the end of October, and the reader will probably feel interested to know what really has happened there.

Krakatoa has not entirely disappeared, while, till now, new volcanoes are visible in the neighbourhood. But the report that new islands were said to have

no

after half of the mountain had crumbled away, had flowed over the wall, which is still there. What remains of the slopes is covered with a grayish-yellow stuff (which, as plainly appears, had been in a melted or fluid state), full of cracks or splits from which steam is continually coming out.

In the same way steam is also coming forth from the deeper cracks of the steep wall, which is still remaining. Sometimes this is accompanied by slight explosions; at that time clouds of brown dust fly up from the cracks, and stones roll down which are often so big as to disturb the sea around the entire base of the mountain. Our

FIG. 2.-Krakatoa after the eruption in May, after a drawing of the Military Survey Bureau, Batavia.

entire survey of the north of Krakatoa suggested the idea that we were above a crater which had been filled with water and quenched by it, and this idea was still strengthened on observing that the decrease of depth, south of Sebesie, bad principally been caused by matters which were cast out and flung away.

Almost in every place here the lead came up from the bottom, filled with black sand or carbonised dust, sometimes mixed with pulverised pumice-stone and little black stones, which apparently had been in a red-hot or melted state. Moreover, the soundings were very different, and the new rocks resemble clods of substances which, when

FIG. 1.-Krakatoa during the eruption of May, after a drawing of the Military Survey Bureau, Batavia.

arisen between Sebesie and Krakatoa is easily to be explained, for the new islands are like a mass of smoking and steaming rocks, and if seen from afar they may easily suggest the idea of a great number of working volcanoes. But, when looked at closely, it appeared that the masses of rock were composed of hot pumice stone, mixed with eruptive masses. In them there were a great many cracks and splits, in which, by the heavy breakers, steam of water was continually generated.

The northern part of the island has entirely disappeared. At what is now the northern edge the peak rises nearly perpendicularly from the sea, and forms a crumbled and rugged wall, and shows a vertical cutting (which is more than 800 metres high) of Krakatoa.

Where was land before, there is now no bottom to be found; at least we could not fathom it with lines of 200 fathoms (360 metres) long. When we had quite calm weather, and steamed slowly and cautiously to and fro along the base of the peak, or had turned off steam and let the ship drift, and were busy in measuring the depth, we could distinctly see the different strata and rocks of the bare, opened mountain. Only here and there a slight trace of melted volcanic matter was to be seen, which, By M. C. van Doorn, officer in command of H.M ship Hydrograaf Translated (an partially abridged) by E. Metzger from Eigen Haard, 1883, No. 51.

FIG. 3.-Peak of Krakatoa after the eruption in August, by M. C. van Doorn. in a melted or very hot state, had contact with water. Probably such a whimsical shape of the rocks above the sea-level suggests the state of the bottom of the sea in the neighbourhood. The stones were still too hot to allow us to discover whether massive stones are under the pumice-stone also. It was not difficult, it is true, to knock off large pieces of these rocks by a hatchet or a chopper, but when a big block fell unexpectedly down, the sailors had often to flee on account of the gases which suddenly arose. The knocked off pieces which been in the boat for an hour. were brought on board were still warm after they had

FIG. 4.-Peak of Sebesie and the volcanic rocks before it, by M. C. van Doorn.

As is to be seen from the map, a great part of the lost ground of Krakatoa is found again at the bottom of the sea, a few miles to the north at least, if we suppose After that no undulations of the ground took place. having passed the limits to which the matters were thrown out, one finds the same soundings as were found before, and the decrease of depth is so local that the idea elevation had taken place, it certainly would be remarked of an upraised bottom is dissipated at once. If such an over a far greater extent and be more regularly ascending and descending. The firmer and stronger part of the crater wall, the peak of Krakatoa, which is still there,

remained standing when the lower and feebler part dropped down, and the water found its way into the fearful boiling pool. We cannot wonder therefore that then a quantity of steam came forth (of which we are not able to form an idea), which caused a strong explosion. The movements of the sea which followed it caused tidal waves, the destroying force of which was experienced in such a fearful manner at the coast of Bantam and the Lampongs.

It is also worth mentioning that a change took place in the figure of Verlaten Island; the area is now triple what it was before, though it is plainly visible that large pieces of the beach were there knocked off a short time ago.

Lang Island, in size and formation, has remained almost unaltered. The sight of these islands, which were formerly covered by a luxurious vegetation, is now very

melancholy. They are now buried under a mass of pumice-stone, and appear like shapeless clods of burst clay (i.e. covered with cracks). After a torrent of rain, the coming forth of steam is sometimes so dense that these islands, when seen from afar, appear like hilly ground covered here and there with snow. If looking at these spots with the telescopes, one can plainly see that these white specks are formed by a great number of clouds, which issued like steam from the fissures.

Sebesie is also covered with ashes up to the top-859 metres-which appear like a grayish-yellow cloth. But it seems that the cover is already less thick here, for here and there one sees the stumps of dead trees peeping out from the crust.

Sebuku shows a dreadful scene of devastation. Perhaps all that lived here is not so completely destroyed as was the case on the southern islands, but the sight of the bare

fields of ashes, alternating with destroyed woods, the trees of which are all either dead or uprooted, gives one a still better idea of the destructive powers which were here at work. It is not until we come to the small islands northward of Sebuku that our eyes are gladdened by little specks of green.

I do not try to describe the scene of destruction and misery which we saw at Anjer and the villages along the coast. The papers have already reported the full particulars, and therefore I do not care to repeat melancholy facts which are already known.

It was a dreadful narrative which was related to us by a native, a lighthouse-keeper of Fourth Point, one of the few men at the lighthouse, who by a wonder was saved. When the wave approached, all fled to the tower (the light was 46 metres above the sea), which, though shaking, resisted the violent waves for a long time. It

was a terrible moment, when at last an enormous rock, which was swept away by the stream, crushed the base of the tower, which then fell down. The man who was saved saw his wife and his children drowned before his eyes. He related this fact in the very resigned way of a Javanese, and considered it the most natural thing in the world that he was now obliged to light the interim light, which was erected as soon as possible.

It has been almost a month that we have been in the Sunda Straits, and even in this short period we could observe that the coasts of Bantam commence to revive. From many places from the heavy rain the ashes are washed down, and a fresh green appears again. Even on the beach young cocoanut trees and banana trees are shooting out between the chaos of dead trees, blocks of rocks, &c.

Off Batavia, October 23, 1883