322 
eliminated about 3 grams of urobilin, the large majority in the faeces; 
this is equivalent to 10 per cent, of his total circulating haemoglobin, 
and corresponds with the elimination in ordinary cases of this 
infection ; (under normal circumstances on full diet his faecal urobilin 
is about O' 18 gram per diem). 
A week later he had another attack of fever, accompanied by 
haemoglobinuria the complex known as blackwater fever—and in 
the first seven days ol this attack he excreted the enormous amount 
33 grams of urobilin. In spite of marked urobilinuria during 
the period of haemoglobinuria only 1-25 grams of urobilin appeared 
in all in the urine, and of the remainder nearly the whole (293 
grams) was present in one stool, of which some particulars were 
given in the preliminary remarks on faecal urobilin (this stool 
contained also a considerable amount of unaltered bile) Owing to 
his collapsed and unconscious state, several stools were lost, but his 
next three recoverable stools contained in all only some 0'8 gram of 
urobilin, (though the presence of diarrhoea may have led to an unduly 
low yield). 
As will be remembered, a man of his weight (60 kilos) contains 
potentially some 30 grams from the haemoglobin of his circulating 
ood and probably about 20 grams more from other sources, so this 
patient seems to have suffered an almost total destruction of his 
circulating blood. The haemoglobinometer reading fell to 25 per 
cent and so apparently some of the destroyed haemoglobin was 
api ) replaced from other sources. The haemoglobinuria itself 
on y accounted for the haemoglobin of 20 c.c. of blood, less than 
0 5 per cent, of the total. 
I do not propose to allude further to this case, but would like to 
indicate one other possible point of interest. Barratt and Yorke* 
state that haemoglobinuria arises in man. if a grade of haemoglob- 
maemm is attained exceeding that obtained by the solution of the 
haemoglobin of o'2 5 per cent, by volume of wet red cells in the 
00 pasma; (roughly speaking, the volumes of plasma and 
corpuscles are equal in human blood, and so on this scale total 
Haemolysis represents 100 per cent). 
We have seen that in Cases 2I and 23 from 20 to 25 per cent, of the 
corp uscles were destr oyed, and in these cases the period of pyrexia 
* Loc. cit., Section III. - 
