of Edinburgh, Session 1885 - 86 . 
853 
in one 1 9 days, and in one 1 4 days ; average life duration being 
thus 21 days. With defibrinated blood the time varied from 14 to 
21 days, or an average of 17 J days. 
The longest duration of life of transfused blood corpuscles in 
rabbits may, therefore, be taken as from 2 to 4 weeks. 
This applies naturally to only a few of the corpuscles injected, 
doubtless the youngest and, therefore, the most resistant at the 
time of injection. The great majority become destroyed at a much 
earlier period, oftentimes with great rapidity. 
The quantities of blood injected were very large, varying from 
about 40 to 90 per cent., but after the injection of smaller quan- 
tities the return to the normal was complete in a few days (5-7). 
It is thus probable that the above period does not represent the 
average life duration of transplanted corpuscles. On the contrary, 
all the observations go to show that it is more in virtue of the large 
numbers injected than of any special vitality of the corpuscles, that 
some of them succeed in remaining so long in the circulation. 
Small quantities of blood are, therefore, probably removed from 
the body in a few days, judging at least from the rapidity with 
which large numbers of corpuscles, e.g 40 percent,, are destroyed 
in from 14 to 21 days, i.e ., a destruction of blood at a rate of 
some 2 to 3 per cent, daily, in addition to the normal amount of 
blood destruction going on in the body. 
This consideration is of importance, in view of the value to 
be attached to the operation of transfusion in man, in whom the 
quantities of blood transfusible are relatively so very small. 
The question then arises, how far the results obtainable in rabbits 
are applicable to the case of other animals. Since the metabolism 
of the body probably varies somewhat in each different species, may 
it not be that the duration of life of the blood corpuscles likewise 
varies 1 
As has been seen, the results obtained in dogs and rabbits 
agree in a somewhat striking manner, and certainly do not lend 
much support to the above view. They both alike point to a 
period of some two to four weeks as being the longest duration 
of life of transfused blood corpuscles. 
On the other hand, from two experiments which I made on dogs, 
it would almost appear as if in them the process of blood destruc- 
