Photography to Micrometry. By Dr. Woodward. 149 
and the corpuscles measured in less than a quarter of the time 
necessary to measure the same number of corpuscles in the micro- 
scope by means of a glass eye-piece micrometer, and in less than 
a tenth of the time necessary if a cobweb micrometer be used. 
As to accuracy, I have already mentioned the advantages of the 
plan proposed. 1 have next to refer to certain sources of inaccu- 
racy in the method ordinarily employed. This method consists in 
the use of an eye-piece micrometer to v^hich values are first given 
by focussing on a stage-micrometer and comparing the two sets of 
lines ; the stage-micrometer is then removed and replaced by the 
slide to be measured. But the values thus obtained are only true 
so long as all the optical conditions under which they were pro- 
cured are rigorously maintained ; and unintentional errors may be 
introduced in various ways. Especially must it be noted that the 
least alteration of the cover-correction of the objective, whether by 
accident or for the purpose of improving definition, will be found 
to modify the magnifying power of the objective, and of course to 
alter the value of the eye-piece micrometer. If, bearing this fact 
in mind, the observer first finds the best position of the screw-collar 
for the slide of blood to be measured, and then inserts the stage- 
micrometer, to give values to the eye-piece micrometer, he will very 
often discover, when he replaces the blood-slide and begins to 
measure it, that the cover is of different thicknesses in difierent 
parts, and that he must either change the correction or be content 
to measure the corpuscles as seen somewhat out of focus. 
Besides this source of error, to which too little attention has 
been paid, there is another, which appears to have been altogether 
neglected. Most of the published measurements of blood-corpuscles 
have been made on blood dried in thin films on glass. But how- 
ever daintily this operation is performed, a large proportion of the 
corpuscles become more or less elliptical in drying. Yet in the 
micrometry of the blood-corpuscles, as practised hitherto, they are 
measured in only one direction. Under these circumstances inac- 
curate results are of course inevitable. 
With the comparatively low powers used by most of those who 
have published measurements of the blood-corpuscles, this source of 
error was naturally overlooked : but with a thousand diameters and 
upwards it becomes evident enough. In making the measurements 
reported in my former paper, I endeavoured to escape this error by 
measuring only those corpuscles which appeared to be perfectly 
round. " Large and small forms were not searched for, but all the 
perfectly formed corpuscles brought into view by the movement of 
the stage were measured as they passed under the micrometer, 
without selection, until the required number was recorded." I am 
now satisfied that the larger corpuscles are more frequently deformed 
than the smaller ones, and that, by pursuing the plan I did, I 
measured an undue proportion of the smaller corpuscles, and thus 
