Sessional_Paper_1901 — Page 132

128

5.--I am of opinion that, unless the stop wall is preferable to a puddle wall. below ground with concrete, using puddle

Puddle Trench.

most excellent puddle clay is readily obtainable, a concrete At any rate it is good practice to fill in the puddle trench only in the bank, thus:-

Naturn!

Clay Puddic

Croun

-Concrote

Valve Well and outlet arrangements.

6. The proposed arrangement of the outlet culvert and valve well does not entirely commend itself to my judgment. It would be more in accordance with modern practice to remove it from beneath the bank, constructing it as a tunnel. In determining its section, provision should be made for carrying off the perennial flow of the streams, and even of considerable floods, during construction. It will scarcely be possible to complete the dam during one dry season. If it be not carried up to the full height before the first monsoon so that the permanent bywash can come into play, then during the monsoon the incomplete dam may be overtopped, and if earth be the constructive material, con- siderable damage will result. One of the advantages of concrete is that if an incomplete dam is over- topped, no serious damage ensues. If a masonry dam is adopted, the arrangement at Tytam may be followed.

Size of By-wash.

7.--The size of overflow should be proportioned to the area of gathering ground in the ratio which obtains at Tytam. This sufficed to carry off the extraordinary flood of 1889, one that is not likely to be exceeded. According to the figures given in the report, the area of overflow will probably suffice.

Filtration, Service Reservoirs, and Distributary-system.

8.--The proposed arrangement of filter beds and service reservoirs is, as far as can be seen from the plans, thoroughly sound; also the proposed distributary-system.

Abandonment of old works.

9.--The existing works will doubtless be abandoned ultimately, if for no other reason, because the gathering ground will be wanted for occupation. Mr. GIBBS in his report lays stress upon the saving that will be effected by the substitution of gravitation for pumping. I think he has scarcely made enough provision for the cost of filtration which is required in the case of the gravitation sup- ply, but not in the case of the existing works. I do not think that it would be prudent to dismantle the old works until the new works have been in operation for some years, so that their actual yield will have been ascertained experimentally. It is to be remembered that all calculations of supply are based on the assumption that the rainfall and flow-off is identical with that which obtains in Hongkong-an assumption that is probable, but not certain.

By-pass Channel.

10. In the description of the Reservoir, I see no mention of a by-pass channel or pipe for carry- ing the water of the tributary streams direct to the filter beds, without passing through the body of the reservoir. This arrangement was found to be necessary in the case of the Pokfulam reservoir, and afforded the means of maintaining a supply of filtered water during heavy rain. It was found that after a flood the reservoir remained for a long time so turbid as to render the water unfilterable, whilst the tributary streams cleared rapidly after the cessation of rain.