61

         (b) 12 patients were injected both subcutaneously and intravenously,
               the initial injection in these patients being done subcutaneously.
               Of these, 8 died, giving a mortality percentage of 66.66 per cent.
               and, lastly,—

         (c) 13 patients had initial injections into the veins of 40 c.c. followed
               in 12 hours by another injection of the same quantity; some
               few of these cases were also injected subcutaneously, the total
               quantity of serum injected amounting in some cases to 400 and
               500 c.c. Of this group, 8 died, giving a mortality of 61.53 per cent.

       Among the 68 cases treated by the serum, there were 45 deaths, whereas,
among the 68 cases not so treated (control cases), there were 41 deaths.

       The serum cases thus show a death-rate of 66.17 per cent., whereas the
control cases have a death-rate of 60.29 per cent., a difference in favour of the
control cases of 5.86 per cent.

       In studying the attached tables, the chief point appears to be—Were the
larger number of the more serious cases in the group of the serum-treated
patients or not ?

       With regard to the general condition of the patient on admission (vide
Table II), it will be noticed that there were ten patients in such a feeble condi-
tion that these were classed among the moribund in the serum group, whereas
there were only four such cases among the control group. But, on the other
hand, there were 16 cases in the control group admitted in a comatose condition,
all of whom died, whereas there were only six such among the serum cases. The
distinction between "comatose" and "moribund" patients, however, was one
difficult to settle in the hospital, and it seems better to amalgamate these two
headings. If this be done, we find 20 patients comatose and moribund on the
control side, and 16 similarly conditioned on the treated side.

       Taken as a whole, then, the conditions were slightly against the non-treated
cases under this head.

       With regard to the sex of the patient, it will be noted that the highest
mortality occurred on the side of the males; of these, the serum cases had a
preponderance, the figures being 54 and 48, respectively.

       Also with reference to the age of the patient, the highest mortality occurred
in those patients of 46 years and over. In this group the serum cases had
five cases, the control cases one.

       Then, again, with regard to the nationality of the patient, the highest morta-
lity occurred among the Goanese and Native Christians; here the serum cases
had a preponderance of six cases, the numbers being 4 and 10, respectively.

       Similarly with regard to the finding of microbes in the blood on date of
admission, the serum group had 19, while the control group had 8 only.