INFECTION IN INTERMEDIATE HOST: BLACKWATER 
211 
for half an hour after the rigor and as a 
general rule the peak of the fever curve 
occurs near the beginning of the attack and 
the temperature tends to fall as the hemo¬ 
lysis decreases. Exceptions to this are, 
(a) the drop to normal or subnormal level 
in the case of anuria and (b) the continu¬ 
ance of fever beyond the hemoglobinuric 
phase (“post-hemoglobinuric pyrexia”). 
Instances of hyperpyrexia have been de¬ 
scribed but appear to be uncommon. 
Vomiting is not an uncommon symptom 
in hemoglobinuric fever but is more prone 
to occur in the severe toxic and anuric 
types. Commonly it is bilious in appear¬ 
ance although some writers have described 
a black color distinguishable from the 
“coffee-grounds” vomitus of yellow fever. 
This distressing symptom may be limited 
to the onset or may continue for a variable 
period during the attack in an intermittent 
manner. Much bile may be brought up. 
Hiccoughing is not infrequently associated 
with vomiting and like it seems to be most 
common in the same (severe) types of 
blackwater fever. 
Splenomegaly and Hepatomegaly. Deeks 
and James (1911) found the spleen always 
to be enlarged and tender as long as the 
fever persisted. This is of interest because 
in primary attacks of P. falciparum ma¬ 
laria the spleen is rarely palpable. 
Where “post-hemoglobinuric fever” exists, 
the liver is more enlarged and tender than 
in malaria alone. Blackloek and Mac¬ 
donald (1928) considered the splenomegaly 
as a necessary preliminary to hemolysis and 
Manson-Bahr (1931) stated that the spleen 
shrank as the hemolysis cleared up and 
considered this as evidence that hemolysis 
occurred primarily in the spleen. 
Urine. Hemoglobinuria is, of course, the 
cardinal symptom of blackwater fever and 
it usually appears within an hour or so 
after the rigor occurs. Occasionally it is 
delayed by as mueh as 12 hours or more 
and in many of such instances the delay 
is probably due to some degree of suppres¬ 
sion. Spectroscopic examination may show 
the presence of both oxyhemoglobin and 
methemoglobin in the urine. The former 
is more frequently found. In 102 speci¬ 
mens which showed absorption bands, Ross 
(1932) noted that 55 per cent contained 
oxyhemoglobin, 39 per cent both, and 6 per 
cent methemoglobin. The urine, however, 
may be negative spectroscopically and yet 
give a positive benzidine reaction. Pseudo- 
methemoglobin, the new pigment described 
by Fairley and Bromfield (1934), has not 
been found in the urihe although it does 
occur in the plasma. These authors (1937) 
concluded that this pigment in the plasma 
had been “erroneously regarded as methe¬ 
moglobin by all observers.” Foy and 
Kondi (1938) found no methemoglobin in 
the sera of blackwater fever patients in 
Greece. More recently Fairley (1939) 
stated that the “haem” of the hemoglobin 
molecule is oxidized to hematin which 
unites with the serum albumin to form 
methemalbumin (pseud o-methemoglobin). 
The concentration of hemoglobin in the 
urine frequently shows considerable varia¬ 
tion when samples are examined over a 
period of time, and, of course, is influenced 
by the volume excreted. Occasionally the 
first specimen voided after the rigor may 
be clearer than later ones. The variation 
may also be affected by other possible fac¬ 
tors such as additional hemolytic events 
and irregular release of hemoglobin into 
the main blood channels from certain areas 
(spleen?, liver?). The duration of hemo¬ 
globinuria is likewise variable. It is com¬ 
monly less than 24 hours but may vary 
from 12 hours to a week or more. It nat¬ 
urally depends upon the degree of hemoly¬ 
sis and its rate as well as the amount of 
urine excreted. Possibly there are other 
factors. Yorke and Nauss (1911) advanced 
the hypothesis that hemoglobin is excreted 
by the epithelial cells of the convoluted 
tubules which are damaged in the process. 
Deeks and James (1911) noted that “in 
all types at the onset of the attack there 
is some suppression of urine, followed as 
a rule by hypersecretion.” They had but 
one recovery among their patients experi¬ 
encing complete suppression. Barratt and 
Yorke (1909) advanced the hypothesis that 
anuria was due to precipitation of the 
