261 
appeared advisable to make a more careful study of the arrangement 
of the blood-vessels of the glomerulus by means of the method of 
reconstruction. ' 
The requirements for such a reconstruction are a perfect set of 
serial sections through a well injected glomerulus, the sections being 
thin enough to pass at least twice through any of its vessel which 
may be struck parallel to the plane of cutting as well as a conception 
of the outward form of the glomerulus previous to cutting. 
Preliminary injections of the dog’s kidney with a variety of sub- 
stances brought out the advantages of a supersaturated aqueous so- 
lution of Berlin blue over other injection masses and the advisability 
of selecting and cutting a single glomerulus. Adult human kidneys 
from the autopsy table were usually abnormal and always failed to 
be well injected. In order, therefore, to obtain a good injection of 
a normal glomerulus, the kidney of a child three months old, dead 
but a few hours, was injected in situ through the abdominal aorta 
until the Berlin blue appeared in the renal vein. The difficulty of 
obtaining a faultless series of very thin sections was greater than that 
of selecting and cutting out a well injected glomerulus from cleared bits 
of this kidney, though very many seemingly perfect glomeruli proved 
to be but partially injected. A chosen glomerulus from the child’s 
kidney was imbedded in paraffin in the usual way and cut into serial 
sections 3 « thick. The 34 sections through this glomerulus were 
then stained in Upson’s carmine and mounted in balsam. Drawings 
of each of these sections enlarged 1333 diameters, the greatest con- 
venient enlargement, were made with a camera lucida (Figs. 4, 5, 6) 
and the corrected drawings transferred with carbon paper to wax 
plates 4 mm thick, i. e. 1333 times as thick as the original sections. 
Before beginning the reconstruction that part of each plate 
representing the glomerulus proper was cut out, the line of incision 
following the outer borders of the external vessels, leaving BowMAN’s 
capsule in the outer shell. The remaining wax shells thus obtained 
were carefully piled in order and a plaster of Paris cast made of the 
cavity. The solid cast roughly indicated the external form of the 
enlarged glomerulus. As a further guide to the reconstruction, the 
sections of the blood vessels appearing in each plate were cut out 
with the exception of wax bridges connecting them. The internal 
relation of these sections in wax representing the blood vessels was 
thus preserved, which aided materially in piling and blending the 
individual sections. 
From the model thus made it appears that the afferent vessel 
