An animal injected with the antidiphtheritic serum becomes refractory to diphtheria almost immediately; but the immunity does not last, as it diminishes from time to time, disappearing in some days or weeks according to the strength and the quantity of the serum administered.
To give an idea of the strength of the serum used in his experiments on children and animals, Roux stated that one tenth of a cubic centimetre, about a drop and a half, of the toxine he employed would kill a guinea pig weighing five hundred grammes, or one pound, in forty-eight hours; but a mixture of one tenth of a cubic centimetre of this serum with nine tenths of a cubic centimetre of those toxines did not even cause a local effect in the guinea pig that received it. In other words, a dose of toxine that would suffice to kill nine average guinea pigs was rendered inert by one ninth of its quantity of serum. Experiments on animals have shown that where the toxine is introduced first it is necessary to give more of the serum, and after a certain delay the serum exercises no antagonistic effect on the toxine that was administered.
The preceding facts have served to bring us to the consideration of the use of the serum in the treatment of diphtheria. The prevalence of the latter disease may be judged from the reports of the United States census of 1880 and of 1890, the statistics of the former year showing that diphtheria and croup caused 77·96 deaths per thousand deaths, and those of the latter year showing that they caused 49·54 deaths per thousand deaths. This experience of a lessening of the mortality from diphtheria has not been confirmed by the statistics of New York city, in which there was a rise and fall from 1883 to 1892 inclusive:
Mortality per Thousand of Deaths.
| 1883 | 48·60 | 1888 | 63·54 |
| 1884 | 52·46 | 1889 | 59·55 |
| 1885 | 61·09 | 1890 | 44·46 |
| 1886 | 72·15 | 1891 | 45·12 |
| 1887 | 78·49 | 1892 | 47·55 |
In fact, it may be noticed that the year the census statistics were obtained the mortality in the city of New York was less than in any other year of the series. This oscillation has occurred in other countries. Thus, in England, during the decade 1861 to 1870 the diphtheria mortality was one hundred and eighty-seven per million living; during the decade 1871 to 1880 it fell to one hundred and twenty-one per million living; while from 1881 to 1890 it increased to one hundred and fifty-nine per million; and this increase was the more conspicuous because both the general death-rate and the death-rate from infectious diseases had been
