changing its _habitat_, may be erroneous. Dr Newsholme, by taking a
wider basis of experience, has arrived at the opposite conclusion, and
finds that diphtheria does not, in fact, flourish more in
sparsely-peopled districts. "When a sufficiently long series of years is
taken," he says, "it appears clear that there is more diphtheria in
urban than in rural communities." The rate for London has always been in
excess of that for the whole of England and Wales. Its distribution at
any given time is determined by a number of circumstances, and by their
incidental co-operation, not by any property or predilection for town or
country inherent in the disease. There are the epidemic conditions of
soil and rainfall, previously discussed, which vary widely in different
localities at different times; there is the steady influence of regular
intercourse, and the accidental element of special distribution by
various means. These things may combine to alter the incidence. In
short, accident plays too great a part to permit any general conclusion
to be drawn from distribution, except from a very wide basis of
experience. The variations are very great and sometimes very sudden. For
instance, the county of London for some years headed the list, having a
far higher death-rate than any other. In 1898 it dropped to the fifth
place, and was surpassed by Rutland, a purely rural county, which had
the lowest mortality of all in the previous year and very nearly the
lowest for the previous ten years. Again, South Wales, which had had a
low mortality for some years, suddenly came into prominence as a
diphtheria district, and in 1898 had the highest death-rate in the
country. Staffordshire and Bedfordshire show a similar rise, the one an
urban, the other a rural, county. All the northern counties, both rural
and urban,--namely, Northumberland, Durham, Cumberland, Westmorland,
Lancashire, Yorkshire, Cheshire and Lincolnshire,--had a very high rate
in 1861-1870, and a low one in 1896-1898. It is obviously unsafe to draw
general conclusions from distribution data on a small scale. Diphtheria
appears to creep about very slowly, as a rule, from place to place, and
from one part of a large town to another; it forsakes one district and
appears in another; occasionally it attacks a fresh locality with great
energy, presumably because the local conditions are exceptionally
favourable, which may be due to the soil or, possibly, to the
susceptibility of the in