View report page

91
this ? If the 97 weeks period may vary from 95 to 97.5, and the 87 weeks period may be represented by
a period of 86 weeks, and the 114 weeks period by one of 112 or 113 and so on, are we not, on a broad
view of the whole series of ampl'tudes ranging from 87 to 114 weeks, compelled to conclude that we are
dealing in London with an approximate periodicity of from something rather less than 90 to rather less
than 120 weeks, i.e., just about 15 weeks on either side of a biennial period, with a tendency to lean to
less rather than to more than two years ? If so much be admitted, and if it be agreed that special causes,
forexample, exhaustion of susceptibles, season, preventive measures, altered birth-rate, immigration
and emigration, etc., etc., are likely to operate in the direction possibly of lengthening, but as a rule,
under London conditions, rather of shortening the period to a greater or less extent, at one and another
time, then the hypothesis that there is some special biological significance attaching to a 97 weeks period
in North London and an 87 weeks period in South London falls to the ground.
A word must be said here regarding the " most ingenious paradox "seemingly presented by
the behaviour of measles north and south of the Thames ; on the one hand, the "97 week" measles
crops up, so it is said, all over the town; on the other, the " 87 week " measles radiates outwards
from St. Saviour's, Southwark. Consideration of the special distribution of population in London
north and south of the river, might, perhaps, be held to account for such a state of things, for central
London on the north side stands in remarkable contrast with central London on the south; the
former area is mainly commercial with very few susceptible child inhabitants, the latter area contains
the closest and largest aggregation of susceptible children in the whole town. Hence, if we may be
permitted to glance at environment at all, the paradoxical nut may be crushed without recourse to a
steam-hammer of hypothesis concerning differing types of germs. Moreover, the tendency to
manifestation of a shorter periodicity on the south side of the river may here find its explanation.
The exposition of the evidence upon which Dr. Brownlee relies for his account of the spread
outward from St. Saviour's, Southwark, was given in a communication to the Royal Society, read
January 31st, 1918. Here, however, no account whatever is taken of the question of distribution of
population qua crowding and qua poverty. Dr. Brownlee, moreover, finds, in the argument which
he developed in this Royal Society paper, support for his objection to the theory that " an epidemic
ceases because the number of susceptible persons is exhausted "; for on this theory, he says, " the course
and the severity of the secondary epidemics, in the districts surrounding the central focus, should not
be essentially different from that in the central area itself." But he finds that the mortality is less in
Wandsworth and in Lewisham than in Southwark and Bermondsey. As a matter of fact, the order
of the districts of South London, as regards mortality, precisely follows, as it might safely have been
predicted it would do, their order in relation to poverty and overcrowding. Dr. Brownlee, of course,
will not allow us to think of environment, but gives two alternatives, loss of virulence and loss of infectivity
on the part of the organism. He adds: "I prefer to interpret it as due to loss of infectivity, the smaller
number of deaths being the result of a smaller number of persons being infected." Upon this evidence,
the hypothesis that measles displays loss of infectivity on proceeding south-eastwards down the Old
Kent-road, or south-westwards down the Wandsworth-road, is, he suggests, established; the fallacy,
here, surely lies in the assumption that death-rates from measles are unaffected by the question as to
whether the populations concerned dwell in the poorer central courts and alleys, or the outlying suburban
villa residences.
The paradox above presented is deserving of notice, if only for the reason that it illustrates the
need for considering the influence of environment. C. S. Calverley's warning may again be referred to
in this connection.
It seems clear, therefore, on the one hand, that there is nothing to indicate that the intervals
between prevalences are the outcome of some predetermined and invariable length of life cycle of special
strains of organisms. There is, on the other hand, abundant proof that variations in periodicity are
directly and closely associated with alterations in environmental conditions. Two of the latter are
specially indicated for study. Increased aggregation of susceptibles (for instance by increase in the
birth-rate as in the example already given above), and activity or passivity as regards the putting into
operation of preventive measures.
The influence of these two environmental conditions goes far, it is submitted, towards explaining
the behaviour of measles in the last fifty years. Taking first the birth-rate, this reached a maximum,
in London, in the late eighties and it has since steadily declined until within quite recent years. Then,
secondly, preventive measures (exclusion of particular scholars and, in some instances, school and class
closures) were first seriously undertaken some twenty years ago, and appreciation of the importance of
prompt use, more especially of exclusion of particular scholars, has fairly uniformly and increasingly
grown since that time. The operation of the combined influence of the two environmental conditions
referred to in large measure serves to explain the behaviour of measles in London, e.g., first the shortening
of the interval between epidemics up to the late eighties, with a gradual tendency to lengthen the
interval subsequent to that date; second, the falling away in the intensity of the individual prevalences
as preventive measures were developed in the early years of the century; and, thirdly, following upon
this, with the consequent increasing accumulation of susceptible persons (the reduction in the number
of those attacked more than compensating for the continued decline in the birth-rate), the rise observed
of late vears culminating first in the more extended epidemic of 1911, and then in that of 1920.
A recently developed phenomenon is, however, deserving of further consideration; it is the
altered form of the mean seasonal distribution of measles as exhibited over say, the last ten years;
thus the contrast presented on comparing years prior to 1874 with the years subsequent thereto was
exhibited in diagram D in my annual report for 1912, and the suggestion was made that the deepening of
55580 M 2
The altered
seasonal
incidence
of the last
ten years.