170 ROBERTS, ON BLOOD-CORPUSCLES. 
Hab. In the streams running through the Chinamen’s 
Garden, near the lake at Mahabuleshwar, about 5000 feet 
above the sea-level. 
In this form each angle of the end lobe appears to be 
bifid; but as the longer ends come into focus before the 
shorter ones, I am of opinion that the shorter ones are the 
ends of the opposite side, which only slightly come into view, 
and give the bifid appearance ; but with a Wenham’s binocular 
this would easily be proved. The ends are narrow, and 
minutely toothed. 
May 12th, 1863. 
On Pecuuiar AppEarances exhibited by BLoop-coRPUSCLES 
under the influence of So.uttons of Macrnta and TANNIN. 
By Wiiu1aM Roserts, M.D., Physician to the Manchester 
Royal Infirmary. Communicated to the Royal Society 
February 18, 1868. 
(‘ Proc, Roy. Soc.,’ vol. xii, p. 481.) 
Tue object of the following paper is to give an account of 
certain observations which seem to indicate that the cell-wall 
of the vertebrate blood-disc does not possess the simplicity 
of structure usually attributed to it. 
It is well known that the blood-corpuscles, when floating 
in their own serum, or after having been treated with acetic 
acid or water, appear to be furnished with perfectly plain 
envelopes, composed of a simple homogeneous membrane, 
without distinction of parts. But, as will appear from the 
observations here to be related, when the blood is treated 
with a solution of magenta (nitrate of rosanilin) or with a 
dilute solution of tannin, the corpuscles present changes 
which seem irreconcileable with such a supposition. 
Attention is first asked to the effects of magenta. When 
a speck of human blood was placed on a glass slide and 
mixed with a drop of a watery solution of magenta,* the 
following changes were observed. The blood-dises speedily 
lost their natural opacity and yellow colour; they became 
perfectly transparent, and assumed a faint-rose colour; they 
also expanded sensibly, and lost their biconcave figure. In 
* The solution I found to answer best in these experiments was a nearly 
saturated solution of nitrate of rosanilin, made by boiling the salt in water, 
and filtering after it had stood twenty-four hours, then diluting slightly with 
water to prevent precipitation. 
