PHILOSOPHICAL TRANSACTIONS.

XII. A description of a vertical floating collimator; and an account of its application to astronomical observations with a circle and with a zenith telescope. By Captain HENRY KATER, V.P.R.S.

Read April 24, and May 1, 1828.

IN the Philosophical Transactions for 1825, I gave an account of a floating collimator, and added a suggestion for the construction of a vertical floating collimator which had not then been carried into effect. I have since had an instrument of that description made, with such improvements as occasion required, and the results which it has afforded have been so satisfactory that I am induced to lay them before the Society.

The collimator which formed the subject of the paper I have mentioned was a horizontal floating collimator. This, in the manner in which I then used it, was the worst form in which the instrument could have been employed; as it was necessary to take the float out of the mercury and replace it in order to complete each observation. The result was therefore liable to be vitiated by any particle of dust or minute bubble of air which might have found a place between the float and the mercury. It cannot therefore but be considered as surprising, that out of one hundred and fifty-one results, only twenty-eight were found in error to an amount exceeding one second, the greatest error being 2".58 and the next 2".

The horizontal floating collimator was tried by the Rev. Dr. BRINKLEY, the present Bishop of Cloyne at the Dublin Observatory, and by the Rev. Dr. RoBINSON at the Observatory at Armagh. An account of Dr. BRINKLEY'S observations is given in the Philosophical Transactions for 1826, where it may be

seen that the mean difference between the results of Dr. BRINKLEY's catalogue of 1823, and those obtained by means of the horizontal floating collimator by observations upon twenty stars, is only 0".03.

Dr. ROBINSON, by ten observations of five stars made with the horizontal floating collimator and his equatorial, obtained a latitude differing only 0".02 from the latitude resulting from his observations of the preceding two years, when the level was employed; and by thirteen observations with the collimator at the winter solstice of 1825, the deduced obliquity of the ecliptic for the beginning of the year, differed only 0".33 from that given in the Nautical Almanac. These results should seem to leave little to be desired in point of accuracy; but the method of using the horizontal floating collimator is so inconvenient, as to constitute no small objection to the general employment of the instrument in this form. To which may be added, the possibility of error arising, as before stated, from the necessity of taking the float out of the mercury and replacing it. From both these objections the vertical floating collimator is wholly free.

The vertical floating collimator has also this further advantage, that it may not only be used with a circle, but may be applied to a telescope, either of the refracting or reflecting kind; such a telescope furnished with a wire micrometer and directed to the zenith, becomes a zenith telescope, free from all the objections to which the zenith sector and the zenith telescope with a plumb-line are liable.

In Plate XIII. I have given plans and sections of the different parts of which the vertical floating collimator is composed. Fig. 1. represents a board of well seasoned mahogany fourteen inches square, and an inch and a half thick. Into this board four legs are screwed, at the distance of an inch and a half from the edge of the board to the centre of each leg. In the middle of the board a circular hole is made, four inches in diameter, into which a tube of sheet iron is firmly driven, of such a length as to project about an inch or an inch and a half above the upper surface of the board. At the distance of five inches and a half from the centre, three brass rollers are let into the board. These are equidistant from each other, and are intended to support the iron pan hereafter to be described, and to facilitate its being moved round about the sheet iron tube as a centre, with but little friction.

Figs. 2. and 3. are the plan and section of a cast iron circular pan; this is one foot in diameter, and has a circular opening of four inches, which passes easily upon the tube of sheet iron just described, and about which the pan is intended to be turned. The sides of the pan (fig. 3.) are 2.4 inches high on the outside, the inside depth being two inches, which leaves four-tenths of an inch for the thickness of the bottom of the pan. The sides of the model for the pan must be sloped as in the section, for the convenience of casting. Into the side of the pan at the bottom a stout iron wire is screwed, intended, as will be seen, to serve as a stop to prevent the pan from being turned round more than 180 degrees.

Figs. 4. and 5. represent (in half its real dimensions) one of two guides made of sheet iron, destined to receive pins, which are intended to prevent the float, presently to be described, from moving horizontally. A piece of plate iron is cut into the form represented at fig. 4. This is afterwards turned up as at fig. 5. leaving the distance of a quarter of an inch between the front and back. Two of these guides are screwed to the bottom of the pan in the inside, (their backs touching its sides,) in that diameter which is at right angles to the iron pin projecting from the outside of the pan.

Fig. 6. represents a spring of about two inches and a half long. The flat part has a longitudinal slit in it, through which screws pass which attach it to the mahogany board, fig. 1. A piece of plate brass is interposed between the screws and the spring. The longitudinal slit is for the purpose of adjusting the spring to its required position. At its extremity the spring is formed into a Y, terminating in a hook, as represented in the figure. Two of these springs are fixed to the mahogany board; the Y of each being distant a little more than six inches from the centre of the board, in opposite directions, and the direction of the spring is at right angles to this diameter. The springs both point the same way.

If now we suppose the pan to be placed upon the sheet iron tube, fig. 1. and it be turned round, the iron pin will come in contact with one of the springs, which will yield to the weight of the pan until the pin is lodged in the Y, beyond which it will be prevented from going by the projecting hook. On turning the pan the contrary way, the inclined plane of the Y will yield to the pressure of the pin and permit it to escape; and when the pin has gone through