TRANSACTIONS OF SECTION G. 767 



earthen dam has been destroyed in that manner, and so far as 1 am able to form 

 an opinion the accidents due to other causes have been as frequent in the case of 

 masonry dams as in that of earthen dams. The destruction of masonry dams has 

 in some instances been the result of too great reliance having been placed on 

 theoretical calculations, without sufficient allowance having been made for the 

 many defects in material and workmanship which might occur in a work of that 

 kind. It was the opinion of the late Mr. Thomas Hawksley that in some cases 

 the destruction of masonry dams had been occasioned by the neglect of the effects 

 of uplift due to the pressure exerted by water finding^ its way beneath the bottom 

 of the dam, a possible condition which he was very careful to take into account 

 when designing the masonry dam of the Vyruwy reservoir of the Liverpool 

 Waterworks. 



Examples of large masonry dams in the United Kingdom may be seen in that 

 constructed by Mr. G. H. Hill at Thirlmere Lake, from which the city of Man- 

 chester is partly supplied with water. Also at the Vyrnwy reservoir of the 

 Liverpool Corporation AVaterworks, designed by and partially carried out under 

 the direction of the late Mr. Thomas Hawksley, after whose retirement it was 

 completed by Mr. G. F. Deacon, who presided over Section G on the occasion of 

 the visit of the British Association to Toronto in 1897 ; and again at the reservoirs 

 near Rhayader, in Wales, now approaching completion, from the designs and 

 under the direction of Mr. James Mansergh, F.R.S., Past President Inst.C.E., for 

 the supply of water to the city of Birmingham. 



From the impounding reservoir the water has to be conveyed to the point of 

 distribution by an aqueduct. This aqueduct, which is sometimes of great length, 

 may consist either wholly of metal pipes, usually of cast iron, or partly of a 

 conduit constructed of masonry, brickwork or concrete following the contour of 

 the ground, with occasional tunnels where high ground has to be passed through, 

 and metal (inverted syphon) pipes where valleys have to be crossed. These con- 

 duits may be either open or covered, the latter method being generally adopted, 

 when they become what is technically known as ' cut and cover ' conduits. In 

 the case of a continuous pipe-line of considerable length it is divided into sections 

 by means of break-pressure tanks interposed at suitable elevations, each tank 

 being say 100 feet or thereabouts below the preceding tank, by which means the 

 pipes are relieved from the excessive pressure to which they would be subjected 

 if the head due to the elevation of the impounding reservoir was carried forward 

 to the service reservoir, from which the water is distributed to the consumer. 

 Steel pipes are frequently used abroad where the cost of carriage is great, but 

 they have not yet been much employed in this country, sufficient experience not 

 having yet been gained in reference to the deterioration of steel pipes due to the 

 action of the water from within and of the ground in which they are laid from 

 without. 



The lines of pipe are provided at intervals with suitable stopcocks, sluice- 

 valves, and air-valves, and also in some cases with self-acting valves which close 

 automatically in the event of the velocity of the water in the pipe becoming 

 abnormally increased owing to the bursting of a pipe beyond. 



I have already stated that most waters obtained from gathering grounds are 

 much improved by filtration. The process of filtration may be carried on where 

 the water leaves the impounding reservoir or at any convenient point on the line 

 of conduit thence to the place of distribution, provided the filter-beds are situate 

 at such an elevation as to place them on the line of hydraulic gradient. Various 

 considerations will influence the determination of their position, but it is desirable 

 that the water should not be subjected to long exposure to light after filtration. 

 Filtration by the slow passage of the water through a bed of sand from two to 

 three feet in thickness, supported by small gravel or other suitable material, is 

 the method usually adopted in Europe, though what is known as mechanical 

 filtration has been used to a considerable extent in the United States, and may 

 under certain conditions be usefully employed. However I do not think it is likely 

 to take the place to any considerable extent in this country of the efficient svstem 

 of sand-filtration introduced so long ago as the year 1828 by the late Mr. James 



