He soon devised a simple method of expelling the air and restoring the requisite identity of level, by fixing a stop-cock for the exit of the air at the vertical bend, where the iron piping, after rising from the bottom of the well to within a few inches of the surface of the ground, begins to slope downwards towards the sea. This bend has necessarily to be made at a point a little below the level of the lowest high-water tide, and, consequently, on opening the stop-cock at high-water, all the air inside the pipe is of course immediately expelled, and then the water inside the well at once assumes the same level as that of the sea. But for this expedient it would have been impossible to carry on the operations continuously for any length of time, as there was found to be a decided tendency for air to collect in the pipes. It was most fortunate that this was discovered during the experimental observations at Bombay, for there stop-cocks could be readily constructed and attached to the piping, which could not possibly have been done at either of the stations in the Gulf.

At Okha Captain Baird found some difficulty in keeping the stop-cock

dry and having access to it. He therefore had a water-tight box 3 feet long and 1 foot square made in halves and fitted over and under the stop-cock, holes having been cut to admit the pipes, and carefully caulked up after the box had been fitted over the pipe (see figure); in this way no water could get at the stop-cock except over the top of the box. Underneath the first 7 or 8 feet of the pipe leading to the sea, a layer of mud and stones of considerable thickness was made, and a wall of similar material built all round the stop-cock, leaving a space about 3 feet square for standing in and steps for getting down to it; also mud and sand were thrown down

between the iron cylinder and the masonry wall right up to the level of the stop-cock bend. It was found that by this means the stop-cock was quite dry and access could be had to it at any time however high the tide was.

Captain Baird spent the recess of 1873 at Bombay in preparing for the operations of the field season of 1873-74. Cylinders as above described

were constructed in suitable lengths to be easily transported to the tidal stations and there put together. The several self-registering instruments. which were to be employed-the tide-gauges, the aneroid barometers and the anemometers or anemographs-were overhauled and put into good working order. The tide-gauges were tested by being employed to register the tides in the harbour of Bombay for several weeks continuously, and were set up over wells connected with deep water by piping, in order that the experimental observations should be taken under precisely similar circumstances to the actual observations. Sundry alterations and improvements were made in them, and in fact everything was done which could be thought of to ensure the instruments being found in a satisfactory condition when they were set up for work at the tidal stations. It was a matter of great importance to have all this done at Bombay, because the advantages of excellent workshops and skilled artificers were not to be met with in the places where the instruments had to be set up for observation.

While in Bombay, Capt. Baird also constructed three portable observatories for erection at the tidal stations. These observatories were made in such a manner as to be readily put together, or taken to pieces and re-erected at any other place where they might afterwards be required. They were about 12' x 9' x 9' and about 12 feet high in the centre, the roof sloping from the ridge to the sides which were about 8 feet high. They were clinker-built, but it was found necessary to cover them with a tarpaulin to keep out the heavy rain during the monsoons.

Operations at Okha Station.-Okha station being near the mouth of the Gulf and the nearest to Bombay, where all the preliminary arrangements were made, was selected as the first to be taken in hand. There all the instruments and stores, and the European assistants, including Mr. Peters, a skilled artificer of the Bombay Harbour Works, whose services had been obligingly placed at Capt. Baird's disposal by Major Merewether, R. E., 6 sub-surveyors and 24 men were sent, on the 13th October, 1873, direct from Bombay in a large pattimar (or native sailing vessel). Meeting with contrary winds, the "Kotia Romani" took such a long time to perform the voyage that Capt. Baird began to fear that she was lost with all hands on board; at last, however, she arrived with her passengers and crew nearly starved, for they had only taken a week's provisions for a voyage which lasted a fortnight. On the 5th November the stores were landed and the first thing to be done was to make an excavation for the iron cylinder. At Okha, as well as at Nawanár, the soil being sandy, it was necessary to take measures to prevent the sides of the well from falling in during the excavation, and therefore a masonry well of sufficient diameter to receive the iron cylinder and vertical shaft of the piping had to be sunk in much the same manner as the wells so frequently used in this country for the foundations of bridges and aque

ducts. The operations were considerably facilitated by the employment of Bull's Patent Dredgers for scooping out the soil under the sinking masonry. (At Hanstal, where the soil was firm, the masonry well was unnecessary). The masonry well was completed to the full depth of 25 feet by the 4th December, and by the 20th the cylinder had been set up, the piping connected with the sea had been laid out into deep water, the observatory was erected and the several self-registering instruments-a tide-gauge, an anemometer and an aneroid barometer-were all in position and ready for the preliminary trial of their performances. By the 23rd everything was complete, the instruments were all working well, and Capt. Baird was about to proceed to the next station, when an accident happened through a native boat drifting down past the station about 3 in the morning of the 24th December, and dragging her anchors across the flexible pipe, smashing it and carrying off a large portion of it as well as the buoys, anchors &c. Being on the spot, Capt. Baird was able to rectify the damage and to arrange for protecting the piping by laying out and anchoring hawsers around. it. Guards were also provided to prevent boats from approaching the buoys. Thus this accident, though very annoying at the time, proved of use in showing the necessity of taking special precautions for the protection of the piping from injury. Similar measures were taken at the other stations, and these precautions were essential to the success of the operations, because in case of any similar accident happening to injure the piping, the native subordinate who would ordinarily be left in sole charge of the station to keep the instruments in order and look after their performances, would be unable to repair the damage without the personal help of Capt Baird or the European assistant, to obtain which would probably cause suspension of the tidal registrations for a fortnight or more, and greatly impair the value of the observations.

While engaged in completing the arrangements at Okha, Capt. Baird sent his assistants in advance to Hanstal and Nawanár to sink the wells, erect the observatories and get everything ready for him to set up the instruments.

The observatory at Okha was fixed on three cross-beams fitted on the tops of six large piles embedded 8 feet in the sand. The cylinder was about 2 feet from the eastern end; the tide-guage being of course as nearly as possible in the centre of the building. The aneroid barometer was placed carefully on a shelf at one corner. The anemometer was fixed to a shelf so that the upright pipe passed through the roof close to the ridge at the western end (the rain-gauge being close to it on the outside). A platform was made to get at the anemometer easily, and this served a double purpose, as it was also the framework for a guard to protect the pendulum and clock of the tide-gauge, and cloth having been fixed all round it, kept any wind from getting to the pendulum.

The self-registering tide-gauge was carefully so placed that the band allowed the float to be 3 inches from one side of the cylinder, while it was the same distance itself from the other side. The instrument having been carefully levelled by wedges, the trestle was secured with screws to the floor; a hole was cut in the floor and a small box let down (properly fitted so as to allow no sand to come in), in order that the counterpoise weight might be able to act for the entire range of the instrument.

The float band was made 35 feet 6 inches long and 33 feet of chain was added to this, and fixed at its other end to the hook under the float, forming a continuous band as it were. The scale of wheels adopted here was; the barrel being 5 feet long, that scale was the largest that could be used for a 14.90 feet range of tide. The float had a swivel to which the band was attached, and the band also passed through two guides fixed to an upright scale on one side, and through another guide fixed to the trestle on the other.

The temporary tide-gauge, consisting of a pile firmly imbedded in the sand and standing about 8 feet out of the ground, was placed about the

level of low-water neaps; to this was attached a box containing a copper float, and to the float was attached a deal rod with a pointer at the end of it about 6 feet above the float. In the bottom of the box a pipe, two feet long, of small diameter was fixed so as to permit the sea to have access to the float; the box itself was about 6 inches square and 6 feet high, the side of it attached to the pile was extended upwards for 6 feet and had a groove in it in which the upright rod with the index worked (vide figure), so that by having this upper part numbered from a certain zero, the level of the sea below this zero could be at once read off. It was found by careful trial in a bucket of water that the pointer always recorded 6 feet 2 inches above the level of the water in which the float worked-the scale was made accordingly; levels were taken to connect the temporary tide-gauge with the top of the cylinder, for comparison of level of water inside and outside.

It will be unnecessary to enter into details of the operations at each station as they were similar to those at Okha, of which the principal outlines have been given. Full details will be found in Capt. Baird's report appended to the General Reports of the Great Trigonometrical Survey for 1873-74 and 1874-75.

Commencement and Progress of the Observations.-The regular tidal registrations were commenced at Okha by the end of December 1873; at Hanstal by the end of March 1874, and at Nawanár by the end of April. It was hoped that they might have been carried on continuously for at least a year, or perhaps longer, at each station, in order to furnish the requisite data for investigations of the separate influence of each as well as the combined of all the principal tidal constituents, and the least that is needed for this purpose is a series of observations extending over a year. During this time the crrors of the clocks for driving the barrels of the selfregistering instruments would have to be frequently determined and the clocks corrected; the instruments would also have to be examined and cleaned, and possibly repaired also, and the relations between the curves on the diagrams and their zero lines would have to be carefully re-determined from time to time; and all this would have to be done either by Capt. Baird or by his assistant Mr. Rendell.

Capt. Baird accordingly drew up a programme for the periodic inpection of the stations and arranged that he and Mr. Rendell should make tours of inspection in turn. It was calculated that a tour embracing all three stations, would take about a month from the date of leaving to that of returning to recess quarters; these had been established in Rajkote, the nearest town to the Gulf where house accommodation suitable for Europeans could be obtained. Anticipating that during the monsoon months the weather at the tidal stations might be found too cloudy to permit of astronomical observations for determining the clock errors, Capt. Baird provided