250 PHILOSOPHICAL TRANSACTIONS. [aNNO 1777. 



nary glass of about 4 feet, where its effect, for the sun's diameter, must be 

 much greater than an inch of motion in a minute ; for the other planets one 

 may have 10 or 15 lines in a second, or even more. Generally the scale is the 

 whole length of the glass for the total refraction of the prism, which is also the 

 case in the Abbe Rochon's prism, for the difference of the 2 refractions. But 

 one may vary the angle by applying the prism without the glass near the object- 

 glass, and turning one of the parts on its axis. In that case, the scale of the 

 excess of the sum of the refractions of the 2 parts of the prism above the dif- 

 ference, will be in length no more than half the circumference of a circle, 

 though the circle may be made as large as one pleases ; but the difference of the 

 distance of the images, will not be proportional to the difference of the arcs 

 run through by the index. In order to determine the relation which the mo- 

 tion of the index bears to the variation of the distance between the 2 images, 

 one must have the solution of a geometrical problem, which is easily gained by 

 spherical trigonometry ; but it will always be better to determine this relation by 

 an actual terrestrial observation of a divided ruler, observed at a given distance. 



When the angle happens to be a large one, the colours would naturally be 

 such as would greatly deform one of the 2 images of the object, namely, that 

 given by the rays that pass through the prism ; but this is easily remedied, at 

 least in a great measure, by making each prism of 2 pieces, one of common 

 and the other of flint-glass. We may multiply the composed achromatic prisms 

 with variable angles, by making the one give degrees from 5 to 5, or from 1 to 

 2, and the other the minutes. We may put 2 on the outside near the object- 

 glass, which will change the distance of the images by the circular mqtion, and 

 give the angle required a little larger than the real one : and another within, 

 which will exactly give the seconds. I have already thought of the instruments 

 requisite for obtaining all these objects with accuracy, as well as for the applica- 

 tion of a variable prism to the common sailor's octant, and have by me the 

 solution of the necessary problems : this will be the object of a work I am pre- 

 paring. In the mean time I publish this, to give others an opportunity of 

 hitting on something better concerning the mechanical construction of these 

 instruments. 



XXXVl. Of a New Instrument for Measuring Small Angles, called the Pris- 

 matic Micrometer. By the Rev. Nevil Maskelyne, D. D., F. R. S., i^c. 



P- 799- 



Practical astronomy was much benefited by the invention of the wire micro- 

 meter, for measuring differences of right ascension and declination : and it re- 

 ceived at least equal advantage from Mr. Savery's ingenious invention of the 

 divided object-glass micrometer, which has been rendered more commodious by 



