44 CELLULAR STRUCTURE OF THE RED BLOOD-CORPUSCLES. 
tive, giving a power of almost 1200 diameters, by adjusting 
its component lenses, for a covering glass slightly thinner 
than that actually employed, and then cautiously screwing 
down the objective, so as to compress a blood disk beneath 
it; under this finely graduated pressure and high magnifying 
power, the apparent expanding of fold after fold, in the pli¬ 
cated wall of a previously wrinkled corpuscle, became strik¬ 
ingly evident. After tinting the external portion of the red 
disk with aniline solution, and then applying considerable 
force to the covering glass, either by means of a mounted 
needle under a low power, or, with the extremity of a high 
objective itself, so as to empty out all the haemato-crystallin, 
the shrivelled envelope could be traced after the removal of 
the pressure closely applied to the surface of the nucleus, 
and under such circumstances occasionally presented an 
obscurely granular appearance. 
Sometimes a few of the corpuscles situated near the edge 
of the thin glass, and therefore most exposed to the action of 
the air, appeared, after three or four hours, to become cracked 
in various places from the circumference to their centre; 
those fissures seem to involve not only the cell contents, but 
also the supposed cell wall; although at first sight this phe¬ 
nomenon may be deemed inconsistent with the older theory, 
in regard to the structure of the red disk, yet I think that it 
can be explained by supposmg that the haemato-crystallin 
had in these cases undergone a sort of troubled crystalliza¬ 
tion, causing it to form a mass of tolerable firmness, which 
split into fragments as it became dry, and at the same time 
cracked its membranous envelope, just as a piece of muslin 
frozen fast to a lump of ice is sometimes broken with the 
fracture of the surface to which it is attached. In some in¬ 
stances, the delicate and transparent cell wall could be de¬ 
tected in the flaw of the haemato-crystallin, its outer edge 
showing a concave line across the peripheral extremity of the 
fissure. 
The addition of water to the fresh blood gave very inter¬ 
esting results, and occasionally afforded an admirable proof 
of the existence of a membranous envelope. The first effect 
of diluting the liquor sanguinis was to increase the thickness 
of the corpuscle, and under its further action the disk gra¬ 
dually became less elongated, until it assumed a spheroidal 
form, the coloured portion being rapidly dissolved out, and 
leaving the nucleus and cell wall more distinctly visible. In 
one instance, a corpuscle which had become quite decolorized 
attached itself to some little mass of granular matter, so tha^ 
it could be retained under observation while I set up cur. 
