INSTRUMENTS FOR MEASURE OF PHOTOGRAPHIC REGISTERS. (15) 



This combination evidently possesses the power of measuring the distance of any part of the 

 photograph from some part of the millimetre scale, subject to a constant correction ; and therefore 

 possesses the power of measuring the distance between any two parts of the photograph. 



We now proceed to the verification of the sub-divisions of the millimetre scale. A second 

 millimetre scale is provided, similar to the first, but containing only 64 millimetre spaces. By 

 placing this secondary scale in different positions under the photograph-microscope, we can use 

 different portions of the principal scale to measure the whole or any part of the secondary scale. 

 Thus, adopting as fundamental divisions of the principal scale the Nos. 20 and 148, we could use 

 both the spaces 20 84 and 84 148 for comparison with the whole length of the secondary scale, 

 and thus could find the relative error of division 84. This being established, we can adopt half 

 the length of the secondary scale, and can use with it the spaces 2052 and 52 84 for the error 

 of 52, and the spaces 84 116 and 116 148 for the error of 116. This process of bisection was 

 carried on, with an extension for the first and last numbers of the scale, till the error of every 

 graduation was determined. 



The photographic plates were mounted in the measuring apparatus by the following arrange- 

 ment : The glass plates for the British Expedition were each 6 inches square ; those for the 

 Australian photographers were 6| inches square. Circular brass plates of sufficient thickness 

 were prepared ; that for the British photographs was lOf inches in diameter, and that for the 

 Australian photographs 12^ inches in diameter. These plates were pierced with square holes, 

 fitted to receive the photographic glass plates, and having ledges and 'springs for supporting and 

 steadily holding them. The brass-plate was laid upon the base-plate of the micrometric instru- 

 ment and pushed in laterally between two guides as far as was necessary, and was held steady 

 by a spring. As the brass plate could be turned to any position in azimuth and could be pushed 

 in to any extent, there was no difficulty in adjusting the direction of the line to be measured so 

 as to make it coincide with the line of the beam-compass. For so adjusting the line joining the 

 centers of the Sun and Venus a moveable wire-frame was provided : carrying a central wire ; a 

 wire on each side parallel to it, at a distance very approximately equal to the radius of Venus ; and 

 a wire on each side at a distance very approximately equal to the Sun's radius. The wires were 

 made tangentially to touch the two images ; marks for the center wire were made on the brass 

 plate at the place of the central wire ; and the adjustment on the base-plate was made with great 

 facility. 



I now proceed to explain the use of the instrument for determining the distortion of images 

 formed by the photoheliograph. 



Some years ago, Mr. Warren De la Rue constructed for a special purpose a scale of equal parts, 

 15 feet in length. The frame of this scale consists of three iron tube-columns, each 15 feet long, 

 and % inch in diameter, braced together with diagonal bracings, so as to form a triangular prism 

 15 feet long with sides (center to center) about 12| inches apart. To one side of the prism are 

 fixed eight iron plates, each 2 feet broad in the direction transverse to the length of the frame and 

 1 foot broad in the direction of the frame's length, with intermediate spaces of 1 foot each : it is 

 this succession of plates and spaces which constitutes the scale of equal parts. The whole is 

 beautifully worked. This instrument is (by Mr. De la Rue's kind permission) now preserved in 

 the Royal Observatory. 



Our first requirement was, to satisfy ourselves, by our own examination, that the spaces are 

 equal, with insignificant error. For this purpose, a curtate triangle (1) was prepared, whose sides 

 approach at the ratio of 1 approach to 12 length of side ; and another frame (2) adapted to embrace 



