This amounts to a daily deficiency extending over the whole period of 11.8 degrees. It will be seen that Greenwich has a rather lower maximum than Herefordshire on account of the sun being more obscured by fog at the former place. It was in fact the coldest December since 1788, as the following list will show :

The average maximum at Ross in December, 1890, 335, was 3.8 degs. less than the previous coldest in 1870, and 10'8 degs. below average of 30 years, and no less than 176 degs. colder than December, 1868. At Greenwich the maximum was 33.3 degs., also 10'8 degs. below the average of 40 years, and 185 degs. below that of 1852; the next coldest being 37 4 degs. in 1878 and 1879.

The following table, furnished by Mr. Ellis, Superintendent of the Meteorological Department, Greenwich, shows periods of early winter since systematic attention has been paid to meteorology at that observatory, i.e., 50 years ago:

There

The 1890-91 period is considerably longer and colder than the others. were two severe frosts not included in the above list which occurred later in the year. That of 1844-45, when the Wye is reported to have been frozen over at Ross no less than three distinct times in the same winter, in December, February, and March; it set in at the close of November, and lasted till Christmas, the temperature of the four weeks being about 9 degrees below average. After a month's interval of warm weather the frost set in again at the end of January and continued till the 21st March, these eight weeks being eight degrees below usual average, extraordinarily low temperatures having occurred on February 11th and

March 13th. In the latter case the temperature at noon was only 23 degrees, and a thermometer in long gras descended at night to -0.2 or slightly below zero This is well remembered by those who attended Ross Spring Fair, when long icicles hung from the animals' mouths! What is called the "Crimean " winter was also a remarkable one. The frost continued unbroken for six weeks and began on January 14th, the weather having been mild at Christmas and for three weeks afterwards. The frost was so intense that a waggon crossed the Wye over the ice at Hoarwithy, and the defect of temperature averaged more than eleven degrees for the six weeks, the intensity of the frost being somewhat greater, but the duration two weeks less than that of last winter. We now come to the winters more than fifty years since. Although tables have been published giving statistics of the weather since 1771, still the figures are not sufficiently reliable to found any very accurate comparative calculation upon them, the error being caused not so much by any defect in the instruments used or of accurate observation, as from want of uniformity in the position of the instruments and the differences of situation and exposure not being sufficiently allowed for. At the close of the last century there were several notable frosts. January, 1776, was so on account of the great fall of snow which lay on the roofs of London houses for 26 days. 1784 was a very cold year, with severe frosts in January and again in December. On the last day of January the frost returned and continued to the middle of March, accompanied and followed by a severe and destructive drought. I have previously referred to the intense cold of January, 1795, when the ice on the Wye, when the frost broke up in the beginning of February, was said by an old man who remembered it well to have measured 18 inches thick. The flood which followed the thaw, which broke so many bridges, and whose height is recorded at so many places on the Wye and Severn, has not since been equalled on the Wye. The twelve weeks' frost of 1813-14 notable for the great depth of snow, as well as that of 1829-30, very severe but not quite so protracted, have been also previously described in our Transactions. In the winter 1890-91 the snow continued on the ground for five weeks, but never exceeded seven inches in depth in this part of Herefordshire. On its breaking up, under the influence of a strong south-west wind, the thaw was very rapid. The ice on the river Wye broke up at Hereford at four a.m. on the morning of the 21st, and six hours later at Ross, some twentyeight miles lower down, the river rose five feet in one hour, the height of the flood being nine feet above summer level at one p.m., and ten feet at four p.m. the same day. For ten hours great blocks of ice from ten to fourteen inches thick rushed down with the stream, breaking off in many cases, and barking and stripping, the trees and other plants on the river banks, so as to leave for months afterwards marks of its course, the sight being a very grand one while it lasted. The last twelve days of January were warm and wet-1'59 of rain falling during that period. February was absolutely rainless, and the driest month in Herefordshire probably since June, 1826, or July, 1800. This was the sixth dry February in succession, curiously enough following seven consecutive wet ones; the means being 0.77 inches and 3.79 inches respectively. It was warmer than usual, although frosts occurred on 15 out of the 28 days in the month, and in the latter

part of it. March, 1891, was principally remarkable for the blizzard which occurred on the evening of the 9th, bringing with it a snowfall of some ten inches deep, blocking many of the roads, and resembling, though not quite equalling in severity, that of January 18th, 1881. In Somerset drifts of snow were reported ten feet deep. The weather continued very cold and dry in April, wind being north or east most of the month. May was cold till the 10th, very hot from 11th to 14th, afterwards cold, stormy, and showery. The contrast in temperature between the 13th, when the thermometer reached 79'8, and that of the 19th, when about five degrees of frost were registered, was very unusual even for May, and particularly destructive to fruit and early vegetables, some accounts describing it as the worst May on record. The rainfall which succeeded the breaking-up of the long-continued drought on May 15th, was, as usual in such cases, much in excess of average. Thus the seven months ending December 14th, 1891, had 27·18 inches of rain, the largest quantity in any similar period since 1852, October being the wettest since 1818, and the only wetter Augusts being those of 1878 and 1879, at the close of the month scarcely any corn being gathered, and much hay being still out. On the contrary, only three drier Septembers occurred in thirty-three years, the 10th, 11th, and 12th had each a maximum temperature exceeding 80°.

I have only now to notice three violent gales, which occurred in the autumn months. That of October 13th and 14th was preceded by a rapid fall of the barometer to 28.724 at 3.5 p.m. This gale was accompanied by torrents of rain, about 1 inches falling in two days-the river rose ten feet and trees were blown down in all directions; large elms being torn up by the roots, and blocking the roads in many parts. In consequence of the warm and windy weather this month dahlias and other tender plants were not injured till October 31st, or thirteen days later than in 1890. Extremely violent gales occurred on October 14th in 1877, and in 1881. The first was described at the time as a tremendous gale doing immense damage both to buildings and trees," and as a great hurricane or cyclone" causing "remarkable blasting of foliage," while that of October 14th, 1881, as "a most terrific hurricane causing immense loss at sea, and also great destruction of property on land." In this neighbourhood all the three storms were about equally violent and destructive. It would appear that the middle of October is specially subject to such visitations. The idea that gales are particularly prevalent at the equinoxes has however been proved to be an erroneous one.

66

The neighbourhood of Ross which more particularly came under the author's observation presented a very singular appearance after the passage of the storm. Specially so in exposed parts like Ross churchyard where large limbs were torn off the elms, and also in the valley where the conformation of the ground caused an outlet to the pent up fury of the gale.

On November 11th, 1891, there was another violent south-west gale, accompanied by an unusually low reading of the barometer, 28.176 (uncorrected) at 9.45 a.m., the greatest violence occurring about one to two p.m., during the exceedingly rapid rise of 743 inch in 7 hours. Many trees, which had stood the pressure of the previous gale, succumbed to this, rooks' nests being blown clean

out of the trees, and fir trees browned by sea spray at a distance of twelve miles from the coast. But we have not yet completed the tale of disasters in this remarkable year.

On December 7th, there was a tornado at Peterstow, near Ross, causing great destruction in a limited track of some three or four hundred yards wide, and, perhaps, a mile long, the roar of which was distinctly heard as it was approaching. In this somewhat limited district the orchard and forest trees fell before the gale like ninepins.

Again, in December, there was a great storm of wind from the 10th to the 13th, scarcely inferior in effects to that of October. The Hereford Times of December 19th, 1891, thus describes the gale of Sunday, December 13th :-" The gale on Sunday caused much damage in Hereford and district. As in the case of the October storm, trees were the principal sufferers, a large number having been uprooted. A small tree growing on the Hereford Castle Green bank by the side of the Castle Pool was blown down, and the branches from another tree close by were broken off. In order to remove the fallen timber it has been found necessary to drain the pool. A portion of a large elin tree standing at the rear of a cottage occupied by Mr. George Minton, gardener, Whitecross Road, Hereford, was blown down. The limb fell on the roof and very much damaged it. Luckily no one was injured. At the Green Crize, ten of the large elm trees were on Sunday morning blown down within twenty minutes. They fell across the road, and it was not until Tuesday evening that all of them were removed. The trees in this avenue are about 85 years of age, and the average height of those which were levelled to the ground would be about 100 feet. A large tree was also blown down on the railway near Redhill junction. Fortunately it was quickly observed, and was removed before any further accident could result. During Sunday, the high wind was accompanied by heavy rains, and on Monday morning large tracts of land were covered. The Lugg was very much swollen, and the Wye registered 15 feet, which is about a foot more than in October."

A similar succession of fierce gales occurred in the autumn and winter of 1859-60, commencing with what has been called the " Royal Charter" gale of October 25th, 1859, and ending with that of February 28th, 1860, which was the most violent one of that stormy season, and probably, the most so since the memorable one of January, 1839.

I have not yet alluded to the effects of these exceptional climatic conditions on vegetation. There is not time to enter fully into the subject. Suffice it to say that while many plants were completely overcome by the great frosts of 1890, it is by no means clear that quite as much permanent injury has not been done by the cold springs which have been the rule now for several past years. It has, however, afforded proof of the great recuperative powers of nature. It is hardly to be expected that New Zealand, Californian, or Mediterranean plants, to say nothing of some tropical species, should not resent the severe changes we have experienced. Gardeners will now perhaps know better how to protect or preserve the more delicate subjects, so that our gardens may still be beautified by the most striking plants which travellers can collect from distant parts of the world.

Woolhope Naturalists' Field Club.

AUGUST 25TH, 1892.

THE Brecon Beacons were visited on Thursday, August 25th, under the most favourable conditions. Owing to the peculiarly rarefied state of the atmosphere, objects on the distant horizon could be discerned with remarkable distinctness, insomuch that those who ascended the two rival summits had cause to congratulate themselves on being favoured for once in a lifetime with such an unusual atmospheric phenomenon. These conditions of atmosphere are rare, and would surpass belief if not faithfully recorded. We have notes, for instance, of a similar rarefaction of atmosphere which prevailed on Monday, July 8th, 1888, to such an extraordinary degree in the English Channel that every prominent object along the French coast could be picked out from Dover and Folkestone; the lighthouse at Cape Grisnez, Calais, and the dome of the Cathedral and Napoleon's column at Boulogne, could be distinctly seen with the naked eye. The distance from Dover to Boulogne, as the crow flies, is 28 miles, and the column is about two miles further inland.

For the benefit of posterity it is considered advisable to leave on record how best the ascent of the Beacons can be managed in one day from Hereford. Having guaranteed a party of say fifty members, you must request the Midland Railway Company to make arrangements with the Brecon and Merthyr Railway Company to give you a special train along their line from Talyllyn Junction to Torpantau. Proceeding along this latter line, the first station from Talyllyn is Talybont, a pretty village near the junction of the River Cafanog with the Usk. From Talybont the line ascends along the shoulder of one of the hills for a distance of seven miles to the next station, Torpantau, overlooking upon the right hand the valley, seven miles in length, of Glyn Collwng, presenting only for a short interval occasionally a glimpse of one or other of the three angular summits of the Beacons in the background, as they peep out between the intermediate heights. The scenery upon the left of this line having been already described on page 82, on the occasion of the visit of the Club on August 26th, 1890, it is unnecessary here to repeat it. On reaching the summit of the seven miles ascent, the train enters a tunnel, immediately on emergence from which the traveller finds himself landed on the platform of the railway station at Torpantau, at an elevation of 1,314 feet above the sea. There is no refreshment room at the station, nor are any to be found on the Beacons, you will therefore act wisely in soliciting the assistance of the gude-wife of the station master to prepare a good fire, hot water, and other creature comforts, in event of your return to Torpantau in preference to making the descent over the Beacons to Brecon.

For the ascent of the Beacons you cannot do better than follow the latest