earth sooner than a heavy shower of rain. The moisture makes the earth subside. Any one can satisfy himself of this fact by filling a flower-pot with fresh earth and then pouring a little water over it, he will find the earth to subside fully one-tenth. But the Bradfield embankment was wetted much more than it could have been by the heaviest rain. It was deluged during the whole day with spray, driven by a stormy wind, blowing right down the valley, from a full reservoir reposing on a sloping bank. Every accessory was favourable to the production of spray, which must have been enormous. We have it in evidence from Mr Gunson, that during the day he could not cross over the embankment, from the spray that was blowing over. This was quite sufficient to saturate the upper portion of the outer embankment, next the puddle-wall, to the depth of forty or fifty feet, whose settlement accounts amply for the depression of the few feet that caused the catastrophe. It thus most wonderfully appears, that notwithstanding all the care that was taken, from first to last, to separate the waters of the reservoir from the outer slope, by means of the puddle-wall, their junction was ultimately effected over its top, in a manner never contemplated before, and which might have been obviated by the simple expedient of a wall or screen along the top of the embankment. Since writing the foregoing, I have seen an account of a sudden settlement of newly raised earthwork by the agency of water, which fully corroborates all I have just said as to the immediate cause of the Bradfield calamity. It is given in a paper on the Erith explosion and the repair of the Thames embankment, by Lewis Moore, read before the Society of Engineers, 7th November curt., and reported in the "Mechanic's Magazine" of 18th November. After giving an account of the manner in which they had successfully raised an embankment, and kept ahead of the tide, which was following very closely, Mr Moore says:-" An alarm was now raised that the water was coming in, and it became apparent that there was considerable percolation under the dam-probably through the broken ground upon which so much labour had been bestowed in puddling; and making its appearance many yards at the back of the work, it was difficult to arrive at the treacherous point. This was most disheartening, by the knowledge that a small stream would soon increase in newly made earthwork, and undermine the whole. The exertions were redoubled, but it is feared that these would have proved of no avail but for the timely settlement of the structure, and consequent compression of the substrata. By this providential circumstance the leak was staunched, so that the settlement, apparently a source of great apprehension, proved really a matter of congratulation." On the next morning (Sunday) the bank was found to have settled some four feet. This was a very great settlement to take place in a bank of moderate height. It is plainly attributable, like that of the Bradfield embankment, to the action of water on the newly formed bank. It happily assisted in the success of one of the greatest engineering feats of the day, the shutting out of the Thames from the Erith marshes in the course of a single tide. In conclusion, I would beg to offer my opinion:-That, but for this one fault, that of leaving too little margin betwixt the levels of the waste-weir and embankment, the works of the Bradfield reservoir were unexceptionable. That this fault is by no means limited to the Bradfield water-works or their engineers. That the remedy is to have a larger margin betwixt the levels of waste-weir and embankment, which may be reduced, after the settlements are made up, by additions to the height of the waste-weir, which, however, should be easily removable on occasion. Observations on the Rain-fall on an area of 4504 acres in the Glencorse District in the year 1863, and in the separate months of April, May, June, July, August, September, and October (being the period of the drought) of 1864. By ALEXANDER RAMSAY, V.P.R.S.S.A., Manager, Edinburgh Water Company.* My special object in the preparation of these observations was to show the quantity of water delivered in Edinburgh— * Read before the Society 23d January 1865, and awarded the Society's Silver Medal. the quantity sent down to the mills-the quantity run to waste during floods-and especially the quantity absorbed by the land, or lost by evaporation,-the whole particulars bearing reference to the rain-fall. I may mention that the observations from which the various results are derived, originated in a desire which I had long entertained to ascertain the quantity of water that could be collected, or otherwise accounted for, out of an ascertained rain-fall on a drainage area of sufficient extent to afford somewhat of an approximation to a general principle. Before, however, proceeding to the more specific objects of this paper, I must ask your permission to say a few words regarding the valley to which the observations apply. It is familiar to many persons in this city, and the crowds who flock to it from early summer till approaching winter testify to its beauties and attractions. Whether for a day's quiet recreation in a delicious atmosphere amid the pastoral scenery of the district, or for a social pic-nic, scarcely a day passes without a crowd of visitors. The good taste of the different classes who frequent the valley is evinced by their appreciation of its attractions; and at such seasons the glen is frequently crowded with persons of every grade of social life. The foot-sore and dusty pedestrian who has sacrificed his day's wages in the workshop or manufactory to enjoy the simple beauty of the glen, may be seen toiling his solitary way, or pointing out to other wayfarers the scenes of its manifold traditions, and in the spirit of true patriotism and pure taste, telling "Of all those hills have seen, Of all the glories of that green, Omnibuses, clustered over with merry crowds of the youth and beauty of middle rank, waken the echoes of the glen with shouts of joyous laughter. Picturesque groups from still humbler vehicles wander by the side of the burn, or stroll over the meadow, or climb the grassy slopes of the hills, and fill the valley with the sounds of mirth and of pleasant songs sung by sweet and tuneful voices. Dashing carriages, filled with visitors of higher grade and less demon strative in their enjoyment, are there. Beautiful horses, and servants in gay liveries, add a pleasant variety to the scene, and all parties are bent on enjoyment, and-with, I am happy to say, but rare exceptions-all are respectful to each other. The glen, indeed, is not less rich in pastoral beauty than interesting in its historical and traditionary associations. The scene of the battle between the poor Covenanters and Hamilton's dragoons, on its eastern slope, is commemorated by a monument erected on the spot where those who fell were buried; and the route of the dragoons, as they crossed the hill from the north and skirted the slope on the south of the Glencorse reservoir, is still distinguishable. Here, too, is the scene of that memorable chase commemorated in Roslyn Chapel by the sculptured representation of a knight in armour with his foot on the neck of a dog. A little farther on is Logan Cottage, the property and occasional residence of our esteemed and respected President, who is proprietor of nearly the whole of this beautiful region. Higher up the glen are the remains of Logan House, a building in the castellated form, and bearing over one of its doors the date of 1230. It is said that its occupant or proprietor about the middle of last century furnished the original of Allan Ramsay's Sir William Worthy, in his play of "The Gentle Shepherd;" and this notion derives a sort of confirmation from the well-known scene at the upper end of the valley bearing the name of " Habbie's Howe." Midway between the ruins of the castle and Habbie's Howe, standing on a picturesque precipice, and bearing the picturesque name of the "Hoolet House," are the remains of an arched building, which, with an enclosure on its north front, suggest the idea of a chapel and graveyard. It is evidently a building of great antiquity, and in times long anterior to the Reformation was most probably the resort of the neighbouring inhabitants for the exercise of their religious rites. As a scene of quiet and peaceful rural beauty, the glen is probably not excelled by any other in Scotland; and here, indeed, one cannot help adopting, with a slight alteration, the words of the poet, "Sweet Logan glen, Thy very name is redolent of beauty." Such, then, is the district to which the matter-of-fact observations which it is my object to bring before you refer. And now applying myself to the more immediate purpose of this paper, I must premise that in any attempt to collect water in reservoirs from surface drainage, the first step in the process is to ascertain the quantity of rain which usually falls in the district in which the reservoirs are situated. The next is to ascertain the extent and character of the drainage area; and as the quantity of water that area may be expected to afford can only be determined by these primary conditions being satisfactorily ascertained, their importance is sufficiently obvious. The laws which govern the rain-fall, however, although no doubt as immutable as those which regulate the tides of the ocean or the rising and setting of the sun, have hitherto eluded scientific inquiry; and hence, although it is known that a certain quantity of rain will fall in a given district within a given cycle of years, no one can predicate the quantity the clouds will discharge on any given day, or week, or month, and only approximately in any one year, or series of years. The collection of water in reservoirs under such conditions, therefore, must necessarily be a matter attended with some degree of uncertainty, and this uncertainty is increased by a variety of circumstances, to some of which I shall take leave to advert. It is Here and everywhere, then, first in the order of claims for water comes the earth, gaping for its natural and preferential share, and to the extent of its necessity as inexorable in its demands, and only less insatiable than the grave. what a water company would call a preferential consumer. Its claims of priority are inalienable, and till it has obtained. its full share, it will concede none of its demands. The quantity it requires, however, is a matter of circumstance. A rocky, mountainous district obviously requires but little water for supporting its scanty and stunted vegetation, while the rich and luxuriant crops of an alluvial plain absorb a large portion of the whole rain-fall. The bare and desolate mountains which enclose the valley of Glencoe afford a striking example of the former, and the fertile VOL. VII. C |