320 
In five instances, only isolated stools shortly after the paroxysm 
were examined, and usually also a control stool, when the patient was 
again convalescent and on meat diet. 
Case Number 
Day of disease 
Faecal Ub. 
(grams) 
Faecal Ub. 
(per cent, of Hbg.) 
Fall of Haem. 
(Haemoglobin¬ 
ometer) 
17 . 
4 
,,, 3 
4 
20 
13 (relapse) 
0-45 
i*6 
10 
21 (control) 
0-28 
1*0 
— 
2 4 . 
20 (relapse) 
o-8 
2-4 
5 
16 (control) 
0*14 
0-42 
2 5 . 
I 
0-12 
— 
— 
7 
0-51 
— 
— 
28 . 
2 
*•13 
3-5 
5 
In the two remaining cases the faecal excretion was followed 
throughout the attack, save for occasional unavoidable losses. The 
history of these cases (21 and 23) has already been alluded to in 
the section dealing with urinary urobilin. 
In Case 23 the urobilin output during convalescence and on meat 
diet was from 0 0 1 to o - 02 gram per diem, and during his pyrexia 
this rose to as much as i'73 grams. 
In the first seven days after the onset of pyrexia his total urobilin 
output was 6 4 grams, representing no less than 25 per cent, of his 
total haemoglobin (of this amount 4 3 grams appeared in the faeces, 
2 " i grams in the urine). His haemoglobinometer reading fell 20 per 
cent, and recovered 6 per cent, in the last two days of the week ; the 
number of red blood cells destroyed (calculated from urobilin 
excretion) must have been about 1,250,000 per mm. 3 , while the 
maximum number of parasites observed was only 50,000 per mm. 3 . 
In two subsequent relapses his urobilin output was very much 
low^er (about 1 5 grams), but in the first of these especially a much 
more marked fall occurred, as shown by the haemoglobinometer, and 
the recovery was much slower, suggesting possibly that the reserves 
of haemoglobin precursors had become exhausted. 
Case 21 had a rather higher output of urobilin during convales¬ 
cence, amounting to 0-15 gram per diem, but after pyrexia it 
