271 
indicate a very strong, almost convincing, correlation. Twenty-one 
of them suffered each from one pyrexial period, lasting from one to 
seven days and preceded or followed by apyrexia; and in all except 
Case 32, the asexual parasites were very much more numerous 
during the pyrexial period. Nine of the cases (7, 17, 18, 19, 20, 23, 
24, 26, 32) suffered from one or two relapses each (thirteen relapses 
altogether); and every relapse, as well as every original attack, was 
associated with a comparatively large number of parasites (forty-three 
pyrexial periods altogether). Similarly, there were forty-six apyrexial 
periods (including the four cases without fever), all associated with 
comparatively small numbers of asexual parasites ; and conversely 
not a single marked rise in the number of asexual parasites took 
place without corresponding fever. 
For further details we compare the numbers of parasites per 
c.mm. counted on days of fever with those counted on days without 
fever. A febrile day is taken as any one on which the patients 
temperature exceeds 98'6° F. ( = 36° Haematothermic Scale), or any 
one which comes between two tertian paroxysms (Cases 9> *4» 2 4)- 
The non-febrile days on which no asexual forms could be found in 
1 c.mm. of blood are omitted. The results are: 
Cases 
Days 
Total 
parasites 
Average 
per day 
Ratio 
P. max— 
Fever . 
8 
3 ° 
136,062 
4,535 
3^3 
No fever. 
8 
68 
8,487 
125 
— 
P- falciparum— 
Fever . 
21 
97 
1,319,880 
13,607 
29-6 
No fever. 
x 9 
83 
38,277 
461 
If we had not omitted the non-febrile days on which no parasites 
were found, the non-febrile daily averages would have been much 
smaller than the figures given above, and the ratio of the febrile 
averages to them much larger; so that the febrile excess is very 
marked. An examination of the details will further convince the 
reader of the great fall in temperature which accompanies the fall 
