and those of certain other Mammals. 
71 
were made with a glass eye-piece micrometer ruled in two hundred 
and fiftieths of an inch, and with such a magnifying power that 
each division corresponded to the fifty thousandth part of an inch 
(•0005079 mm.). The objectives used were an immersion X V of 
Powell and Lealand, and an immersion No. 13 of Hartnack, either 
of which permitted the above value to be given to the divisions of 
the eye-piece micrometer by properly adjusting the draw-tube. 
The stage-micrometer used in effecting this adjustment is an excel- 
lent one in -rg-oths and yoVoths of an English inch, in which the 
several hundredths and thousandths, as nearly as I can measure, 
are equal to each other, and the ten divisions of the latter value to 
any one division of the former, a quality in which the stage 
micrometers in the market are generally deficient. I have com- 
pared this micrometer with a standard scale ruled on silver — a 
centimeter in millimeters and tenths — the property of the United 
States Coast Survey, kindly loaned for this purpose by Mr. J. E. 
Hilgard, who assures me that it is “very accurate.” I made 
several comparisons both by means of an eye-piece micrometer, and 
by the contact method described by Welcker. These comparisons 
showed that the divisions of my stage-micrometer were nearly 
two per cent, (exactly 1 • 945 per cent.) larger than they ought to 
be, and this correction was accordingly applied in adjusting the 
value of the eye-piece micrometer. The value assigned to the 
divisions of the eye-piece micrometer for these measurements cannot, 
therefore, I think, differ from their absolute value by a quantity 
large enough to modify the results appreciably. 
As the divisions represent a value twelve and a half times 
less than that of the divisions of Mr. Gulliver’s eye-piece micro- 
meter, and more than three times less than those of Welcker’s eye- 
piece micrometer, I did not find it necessary to estimate fractions 
of a division, as they did, but read the nearest number of whole 
divisions corresponding to each corpuscle. Fifty corpuscles, or 
about that number, were measured in each sample of blood. An 
assistant noted the number of eye-piece divisions corresponding to 
each corpuscle as the measurements were made, and the mean was 
obtained in each case by adding together all the values and 
dividing by the number of corpuscles measured. Of course, the 
number of eye-piece divisions found only required to be multiplied 
by two to convert it into decimals of an inch. I endeavoured at 
first to make these measurements with a dry Powell and Lealand’s 
sVth of an inch, with the draw-tube so adjusted that each division of 
the eye-piece micrometer should equal one hundred thousandth of 
an inch, but I found the outline of the corpuscles, with this power, 
was not sharp enough to permit me to measure them as exactly as 
I wished, and I therefore gave the preference to the immersion 
objectives above mentioned. 
