174* &V* Everard Home on the changes the blood 
smeared as thin as possible upon the glass, in order that all 
moisture may instantly evaporate ; they then remain of their 
full size and colour, perfectly spherical, as in the representa- 
tion, fig. i, PI viii. But if a greater quantity of blood be laid 
upon a glass which retains moisture only half a minute, the co- 
louring matter begins, in a few seconds, to separate and form a 
circle round the globule, and if the blood is diluted with water, 
the separation of the colouring matter is instantaneous, and 
the globule puts on the appearance represented in fig. 2, PL 
viii. Another reason is, the great quantity of colouring mat- 
ter, it being as three to one in proportion to the globules. 
The globules of the blood have neither the same size nor 
the same shape in all animals. Dr. Young, in his introduction 
to Medical Literature, has described them to be of an oval form 
in the skate. Upon examining them in that fish, Mr. Bauer 
found them, while the fish was alive, of the form of an egg, 
but almost immediately after death, flattened. I shall, how- 
ever, reserve the materials I have collected upon this subject 
for another communication. When the globules in the human 
blood lose the colouring matter, they continue floating in the 
serum, and are seen to have an attraction towards one another 
so as to coalesce, /uniting themselves together. In the an- 
nexed drawing is represented their mode of union under differ- 
ent circumstances, surrounded by serum deeply tinged with 
the colouring matter. In one instance three globules are so 
united as to form one line ; in another there is a line compo- 
sed of four globules, with lateral indentations, where the 
union between the globules had taken place. This appear- 
ance, joined to other circumstances, renders it probable at least 
that the globules may be the part of the blood, out of which 
