1895-96.] Mr R. C. Mossman on Death-rate in Edinburgh . 105 
On the Seasonal Death-rate from Certain Diseases in 
Edinburgh during the Period 1878-94, with Re- 
marks on the Relation between Weather and Mor- 
tality. By R. C. Mossman, F.R.S.E., F.R. Met. Soc. 
(With a Plate.) 
(Read March 2, 1896.) 
This inquiry embraces the seventeen years beginning with 1878 
and ending with 1894, the material utilised being the weekly 
mortality returns published by the Registrar-General for Scotland. 
Particulars are given in these returns of the number of deaths 
from small-pox, measles, scarlet fever, diphtheria, whooping-cough, 
diarrhoea, fever, croup, laryngitis, bronchitis, pneumonia, and 
pleurisy. Some of the diseases are classed together, viz., typhus 
and typhoid fever, croup and laryngitis, and bronchitis, pneu- 
monia, and pleurisy. The data discussed also embraces returns of 
deaths from all causes, deaths of infants under one year of age, 
and of persons aged sixty years and upwards. The deaths have 
all been entered as having occurred during the calendar weeks in 
which they were registered, but as notification usually follows 
decease by several days, a slight element of error is introduced. 
Since the returns do not give any information as to the duration of 
the illness, we are unable to ascertain whether specific weather 
types are in casual relationship to disease. There are every week 
a number of deaths from uncertified and unspecified causes. These 
omissions are subsequently rectified, but their presence in the 
returns does not affect the general results of the investigation. 
It is our intention merely to discuss some of the broader results of 
the inquiry, the chief object of which is to give a first approxima- 
tion to the seasonal death-rate of the diseases under review. In 
discussing the relation between weather and disease, it is well to 
keep in mind the fact that different diseases exhibit varying 
degrees of sensitiveness to weather influences, and that climatic 
changes take some little time to affect the human subject. 
The meteorological data for the period under review were sup- 
