MICROMETER E Y E - P I E C E. 121 



shows how many complete turns of the screw must have been 

 made in thus separating the threads ; while the number to which 

 the index points on the milled head, shows what fraction of a 

 turn may have been made in addition. It is usual, by employ- 

 ing a screw of 100 threads to the inch, to give to each division 

 of the scale the value of l-100th of an inch, and to divide the 

 milled head into 100 parts ; but the absolute value of the divi- 

 sions is of little consequence, since their micrometric value de- 

 pends upon the objective with which the instrument maybe em- 

 ployed. This must be determined by means of a ruled slip of 

 glass laid upon the stage ; and as the distance of the divisions 

 even in the best ruled slip is by no means uniform, 1 it is advisa- 

 ble to take an average of several measurements, both upon dif- 

 ferent slips, and upon different parts of the same slip. Here the 

 draw-tube will be of essential use, in enabling the microscopist 

 to bring the value of the divisions of his Micrometer to even num- 

 bers. Thus, suppose that with a l-4th-inch object-glass, the tube 

 being pushed in, a separation of the lines by one entire turn 

 and 37-100ths of another were needed to take in the space be- 

 tween two lines on the ruled slip, whose actual distance is 

 l-1000th of an inch ; then it is obvious that 137 divisions on the 

 milled head are equivalent with that powder to a dimension of 

 1- 1000th of an inch, or the value of each division is 1-137, 000th 

 of an inch. But as this is an awkward number for calculation, 

 the magnifying power may be readily increased by means of the 

 draw-tube, until the space of l-1000th of an inch shall be repre- 

 sented by a separation of the cobweb threads to the extent of 

 150 divisions ; thus giving to each division the much more con- 

 venient value of l-150,000th of an inch. The Microscopist who 

 applies himself to researches requiring micrometric measure- 

 ment, should determine the value of his Micrometer with each 

 of the objectives he is likely to use for the purpose ; and should 

 keep a table of these determinations, recording in each case the 

 extent to which the tube has been drawn out, as marked by the 

 graduated scale of inches which it should possess. The accuracy 

 with which measurements may be made with this instrument, is 

 not really quite so minute as it appears to be ; for it is found 

 practically that when the milled head is so graduated, that, by 

 moving it through a single division, the cobweb threads are se- 

 parated or approximated by no more than l-10,000th of an inch, 

 it needs to be moved through four divisions, for any change in 

 the position of the threads to be made sensible to the eye. Con- 

 sequently, if three entire turns, or 300 divisions, were found to 

 separate the threads so far as to coincide with a distance of 

 l-1000th of an inch on the ruled glass under a l-8th of an inch 



1 Of the degree of this inequality, some idea may be formed from the statement of 

 Hannover, that the value of the different divisions of a glass ruled by Chevalier to 

 1-lGOth of a millimetre, varied between the extreme .ratios of 3T36, the mean of all 

 being 34. 



