Conduit from Tai Tam Reservoir to Albany [Filter Beds] [1887- ] | Gwulo: Old Hong Kong

Conduit from Tai Tam Reservoir to Albany [Filter Beds] [1887- ]

Current condition: 
Date Place completed: 
c.1887-08-01 (Month, Day are approximate)

Shortage of water was a problem in Hong Kong since the early days. In 1860 the new governor Sir Hercules Robinson reported that "the population of Victoria had suffered very keenly during the previous winter from the scarcity of water". To overcome future shortages, a small reservoir was built at Pok Fu Lam. The winter of 1865 was dry again and plans were made to built a bigger reservoir at Pok Fu Lam. The governor Sir Richard MacDonnell decided to built this reservoir against the concerns of his advisors. When the reservoir was finished, it was already too small regarding the fast growing colony. During the next years no major break-through occurred, actually all governors tried to avoid decisions and left the problem of water shortage to their successors. 

In 1873 finally, valleys to build dams were examined, unfortunately all were on the "wrong side of the mountains" regarding the location of Victoria. One of the important requirements was the water should flow freely to the city and no pumping being necessary. The only valley that met this criterion was Tai Tam. Plans were were made in 1873 to build a dam, a tunnel, and a conduit. Unfortunately, in 1875, the colony was hit by an "unprecedented recession", and nothing happened. It took as long as 1882 until the original layout was accepted and the work started.



There have been other departures from the former abridged project which explain other minor additions to the aggregate estimate of cost. The most important of these changes is the reversion to the original scheme of a surface conduit running along the contours of the hills overlooking the Happy Valley and Wantsai, in lieu of the cheaper, but less desirable expedient of an underground pipe down the Wong-nei-Chung Valley and under the Race Course meadows, to which recourse was bad in 1875 only in order to avoid initial expense.

The superiority of a surface conduit over an iron pipe is obvious. An iron pipe would rust and perish in the course of a few years while there is no reason why the conduit should not endure for many generations. An iron pipe buried in the earth is incapable of augmenting the town supply in any way, while a conduit winding along the surface of the ground is able to convert into tributaries all the mountain streams that cross its path, and thus to augment the supply by wayside feeders. There is an additional advantage special to the conduit, which is that in case of a breakdown at the outlet works in Tytam, or in the event of any temporary stoppage for repairs—a most serious thing for a town—the supply can be continued for a day or two from the conduit. The conduit is not being built in an even or gradual incline or fall from the tunnel mouth to the town, like the bed of a flowing stream, but in a series of flats or level steps resembling a concatenation of narrow elongated tanks, each tank about a quarter of a mile long, and each lower than its neighbour by about one foot. These water-compartments may therefore in case of emergency, be converted into temporary sources of supply and their contents retained or eked out at will by means of stop-planks or water-gates, until the main supply is re-opened from the fountain head.


Tunnel—The tunnel in course of construction from Tytam to Wong-nei-Chung for the passage of the waters from the new reservoir to the northern side of the island will be when completed, about 7,300 feet long or roughly speaking, a little under a mile and a third. It is six feet wide and seven feet high from the floor to the crown of the arch, these being the least admissible dimensions consistent with convenience of working. The excavation has been driven so far entirely through a hard, solid and fissured granite formation, that has proved most adverse to progress and there is now no reasonable hope left that the character of the rock will change to anything more favourable

The mountain is being pierced from both sides in the same straight line and on the same level, and the two headings are now each advancing towards one another at the rate of twenty feet per week and therefore lessening the distance between them by forty feet every week. The Tytam heading has been carried in to a distance of 2,132 feet, and the Wong-nei-Chung heading 1,868 feet, making a total of tunnel finished of 4,000 feet, or over one half of the whole work. There remain therefore about 3,300 feet of tunnel to drive, and at the rate of actual progress which I have mentioned, i.e.: forty feet per week,—a progress which there is no reason to doubt will be maintained—it will take eighty-two weeks to perforate the rest of the mountain and to connect Tytam reservoir-water with the conduit on this side. Eighty-two weeks are equivalent to one year and eight months, so that the completion of the tunnel may be looked for on or about July 1887.

Conduit—The masonry surface-conduit which winds along the mountain contours from Wong-nei-Chung to the Albany is intended to convey the water by gravitation from the tunnel mouth to a proposed terminal tank and filter-bed to be built on the hill-side in the Albany Valley. I have already in a preceding page, described the manner in which this water-channel is being built, with a view in case of emergency, to the utilization of its successive compartments as tanks whence to derive relief pending any temporary stoppage of the water at the dam. The conduit is a square red brick culvert 3 feet wide, rendered inside with cement to prevent leakage, and covered with granite slabs to protect the purity of the supply and to keep down the temperature of the water. With the exception of the Wantsai Valley which it is intended to cross with a syphon of iron pipes, the water will be carried over all streams by aquaeducts of arched masonry, and these aquaeducts will be so paved and concreted over as to enable them and the entire length of conduit to be used as a continuous level road for pedestrians. With reference to the progress of this particular branch of the work, it may be briefly mentioned that it has been divided into two sections, the first extending from the tunnel month at Wong-nei-Chung as far as the Wantsai Valley, and the second from the Wantsai Valley to the Albany. The first section now rapidly approaching completion will be finished or very nearly so, by the end of the present year, whereupon the Wantsai syphon works and the second section will be commenced and terminated by the end of 1886.


Further details of tunnel and conduit can be found here: Tai Tam Tunnel [1887- ] and Tai Tam Conduit [1887- ]





Presented to the Legislative Council by Command of His Excellency the Governor,

PUBLIC WORKS DEPARTMENT, Hongkong, 1st November, 1885.

Seen at:

The full report can be read there.

Further details at:



Hi David, could you please add Tai Tam Tunnel [1887- ] and Tai Tam Conduit [1887- ] as "children to this parents place". Regards, Klaus


Hi Klaus & David,

There seem to be three different pages for the Tai Tam conduit - this one, Bown Road Viaduct and Tai Tam Conduit. I wonder if it's possible to merge them or link them more clearly perhaps? 




Hi Liz,

the idea for the different places/descriptions is as follows:

Conduit from Tai Tam Reservoir to Albany [Filter Beds] [1887- ] is the "parent page"

Tai Tam Tunnel [1887- ] and Tai Tam Conduit [1887- ]  are "children pages" (like cross harbour tunnel with two "children" north entrance and south entrance).

Only David can do that, and I will point that out more clearly in the parent page. 

The two places added by tngan are special spots on the conduit, I will create a link to these from Tai Tam Conduit.

Thank you for pointing out this unclear descriptions/relations. 



Thanks, Klaus, I understand now.