wanted. A telescope of sufficient power to separate the two stars, and a 

 wire micrometer, are the necessary apparatus : of the principle of the latter 

 we shall give a general description, not entering into any of the niceties 

 of its construction, and supposing throughout that the instrument is perfect.* 



The wire micrometer is an apparatus to be annexed to a telescope, such 

 that when inserted in the tube the field presents the usual appearance of a 

 luminous circle cut by four very fine wires parallel two and two, the first pair 

 beinj 



at right angles to the second. 



Fig. 12. 



It is found that the apparatus can be turned round so as to give any de- 

 sired direction to the wires. One pair of wires is placed at a fixed distance 

 from each other. Of the other pair, one is moveable, so as alternately to ap- 

 proach to or recede from that to which it is parallel, preserving, however, its 

 parallelism, during the motion. In fact, the interval between one of the pairs 

 of parallel wires can be increased and diminished at pleasure. This motion 

 is given by a screw which has a small circular head, the edge of which is 

 divided into a certain number of divisions, say 100. The threads of this 

 micrometer-screw are so small that a whole revolution of the graduated 

 carries the moveable wire toward or from that to which it is parallel, through 

 a very small space, and if there be one hundred divisions on the circumfer- 

 ence of the head which are sufficiently distinct to be read to a quarter of a 

 division, we can ascertain a motion of the wire which corresponds to the 

 four hundredth part of the effect of one entire revolution. If we desire to 

 measure the interval between two stars which are near each other, as is 

 always the case with the individuals of a double star, we have now only to 

 adjust the instrument until one of the two stars moves (by the diurnal motion 

 of the heavens along the fixed wire), and then by turning tjie screw adjust 

 the moveable wire, so that the other star shall move along it. It is then only 

 necessary to ascertain how many revolutions and parts of a revolution of the 

 screw are necessary to bring the moveable wire to coincidence with the 

 fixed wire. The distance between the stars will then be Joiown, provided we 

 have previously ascertained what space of the field of view corresponds to one 

 revolution of the screw. 



It might perhaps be imagined, that in the original construction of the 

 micrometric apparatus, the screw would be cut so that each revolution might 

 correspond to a certain space, such as one second. Mechanical art, however, 

 has not, and probably never will attain to the degree of perfection necessary 

 to accomplish this. It is very possible to cut a fine screw with threads which 

 in a practical sense may be said to be equal to each other, but he can not 

 do this and also insure a result which will make a certain number of these 



* De Morgan, pp. 81-'2, pp. 84-'5, pp. 94-'5. 



