108 
PROF. T. CARNELLEY, MR. J. S. HALDANE, AND DR. A. M. ANDERSON 
rence, it is true (found in 90 per cent, of the fatal cases of diphtheria, according to 
Thaon, loc. cit.) But here the chances of broncho-pneumonia are about equal, 
whatever the number of micro-organisms in the ah’, because in these diseases the 
broncho-pneumonia is due to the specific poison of the disease, and caused by the 
spread downwards of infection from the throat. In diphtheria the patches of broncho¬ 
pneumonia are, according to Thaon, full of the same organisms as are present in the 
false membrane, and which produce artificial diphtheria in an im als (Loeffler, 
‘ Mittheilungen a. d. K. Gesundheitsamte, 7 vol. 2, p. 421). The death-rates both in 
Dundee and Buda-Pesth are in consistence with this hypothesis. 
The curious relation in the death-rate from phthisis appears to depend on the 
mutual interaction of constitutional predisposition and bad hygienic conditions. The 
mere fact that, the worse the hygienic conditions, the more do children who are 
weakly die off in childhood, does not altogether account for the differences in the 
death-rate from phthisis. For, in spite of this survival of the fittest, the general 
death-rate above 20 increases steadily from the better to the worse houses 
(including Infirmary and Poor-houses), although the increase is nothing like so great 
as it is for children. Were phthisis due merely to the influence of bad hygienic 
conditions, we should, therefore, still expect the death-rate to increase up to the two- 
and one-roomed houses. As a matter of fact, although those under good hygienic 
conditions suffer by far the least from phthisis, yet it tells most heavily on those 
under only moderately bad hygienic conditions, and this is the case even when all the 
deaths in the Infirmary and Poor-houses are set down against those under the worst 
hygienic conditions, i.e., those in the one- and tw T o-roomed houses. This would seem 
to indicate a predisposition to tubercular disease in a certain proportion of the 
individuals, quite apart from general constitutional weakness. Under good hygienic 
conditions the great majority of these will escape altogether. Under medium con¬ 
ditions a large proportion will survive the tubercular diseases of childhood, but only 
to fall victims to phthisis later on in life. Under the worst conditions most will die 
of tubercular disease during childhood. So few will survive that, although a larger 
proportion of them will die than of the survivors under medium conditions, yet the 
death-rate from phthisis on the whole population in the worst houses will be less. 
The fact that the mortality from croupous pneumonia increases so markedly from 
four- to two-roomed houses, and out of all proportion to the death-rate above childhood, 
is an important confirmation of the theory of its being caused by micro-organisms. 
The fact that it is rather less fatal in one- than in two-roomed houses is due probably 
to the fact that it is to a very large extent treated in the Infirmary, where very many 
recover who would certainly die at home. 
