Tue Microscope. 33 
advantage in point of accuracy to compensate for its additional 
cumbersomeness and expensiveness.” 
Up to that time I had never made any comparisons of these 
two methods of micrometric measurement, other than the 
measurement of the blood-corpuscles above referred to, if that 
can be called a comparison; and, as Mr. Tolman habitually uses 
an eye-piece micrometer of the ordinary pattern ruled into 
spaces of ;j;, inch, and is not accustomed to use the filar mi- 
crometer, while at that time I had had little experience with 
the use of any other than a filar micrometer, I have thought that 
the deductions drawn by him from a comparison of measure- 
ments of different objects were hardly warranted by the facts, 
and having thus had my attention drawn to the subject and not 
having seen anything in microscopic literature which to my 
mind satisfactorily settled the question, I have for some time 
past been working upon the problem and herewith present the 
facts and conclusions arrived at, which will, I hope, clear up a 
question which while not difficult, appears not to have received 
so much attention as it deserves. 
The first step in order to compare the results of the two 
methods, obviously was to secure an accurate standard of com- 
parison and accurately to determine the value of one division 
of each micrometer for each one of the objectives used in this 
work. This has engaged my attention for a very considerable 
period, and the values assigned to one division of the respective 
eye-piece micrometers are the mean of a very large number of 
measurements made at different times from different standards, 
whose relative and absolute corrections are well known, so that 
I feel confident that they are correct within exceedingly narrow 
limits. While I have never yet seen a micrometer whose sub- 
divisions were exactly equal, I may state in passing that the 
micrometers manufactured by Prof. W. A. Rogers are so nearly 
uniform and accurate that values derived from means of series 
made from different micrometers at intervals of from one day 
to several months uniformly agree infractions of an inch to the 
sixth, seventh or eighth decimal place, depending on the power 
used ; and that the results thus obtained are identical with those 
obtained from the direct measurement of “ Centimeter Scale A, 
