a 
1908. | Hem-agglutinins, etc., in the Blood. 539 
examined in the last stages of the disease. This, however, is not an example 
of true agglutination, but it is clumping of short rouleaux. Shattock noted, as 
already stated, that if we diluted human serum with saline, the red blood 
corpuscles did not run into rouleaux when mixed with this diluted serum, as 
in the case of the pure serum. A similar effect is the rule in the case of 
agglutination. In this respect, rouleaux formation and hem-agglutination 
have points in common. If the blood is examined, eg., from a case of 
pulmonary tuberculosis in which the rouleaux formation observed in the 
fresh film is well-nigh perfect and free from agglutination, when this serum is 
mixed with normal red cells it gives a striking appearance to the blood. No 
rouleaux formation can be observed, but the red cells are collected into large 
and tight clumps, while the shape of the cells is considerably altered from 
the large red circular disc to small retracted bodies apparently less than half 
the size of the original cell, highly refractile and resembling an oil droplet. 
This is probably a physical action and may be dependent upon the salt. 
contents of the serum and cells. 
From what has been stated in this communication, and from very numerous. 
_ unrelated observations, rouleaux formation and haem-ageglutination must be 
regarded as distinct phenomena. 
Heem-opsonins. 
Technique.—The white cells for these experiments were obtained by the 
method which is usually adopted for such investigations on phagocytosis. In 
every instance normal leucocytes were employed, but in some instances 
washed immune cells were obtained from various cases, and the results com- 
pared with those obtained with the normal cells. It was found, however, that 
practically the same result occurred whether the immune cell or the normal 
cell was employed for this special purpose. 
A 5-per-cent. suspension of washed red blood corpuscles in normal saline 
(0°85 per cent.) was used in all these experiments. One volume of normal 
or immune leucocytes was drawn up into a capillary tube with an equal 
volume of washed red cells and blood serum. These were carefully mixed, 
sealed in a glass tube, and incubated in the usual manner. Various times 
were used for the incubation, varying from 15 minutes to several hours, but: 
there was no advantage, while in many cases a disadvantage occurred in 
incubating for periods of longer time than half an hour. 
In many instances the serum, before it was drawn up into the capillary 
tube, was diluted to a definite strength either with normal salt solution or 
with a standard normal serum. At the end of the incubation period the 
contents of the capillary tube were blown out on to a glass slide, and thin film 
