^»^. 5. 1875] 



NATURE 



281 



THE MORTALITY OF THE LARGE TOWNS 



OF THE BRITISH ISLANDS IN RELATION 



TO WEATHER* 



'TTHE materials for this inquiry have been obtained from the 



-*■ Weekly Reports of the Registrars'-General for England 



and Scotland for the ten years, 1865-74. The data discussed 



embrace returns of deaths from all causes and at all ages, deaths 



of persons under one year of age, of persons above sixty years of 



age, and deaths from diarrhoea. The weekly averages have 



been calculated on the annual rate of mortality per i,ocX) of the 



population. 



The results for'every one of the large iown?, show during the 



vnxvitx months an excess above the average mortality. A 

 regards the English towns, that excess is greatest at Norwich, 

 Wolverhampton, and Nottingham, and least at Bradford, Leeds, 

 Salford, and most other towns in the north. In Scotland the 

 winter excess is greatest at Aberdeen, and least at Leith and 

 Greenock. At Dublin, the largest monthly mortality, 22 per 

 cent, above the weekly average, occurs during February and 

 March, being from a month to six weeks later than the time of 

 the maximum of the English and Scottish towns. 



In all the English towns, the minimum mortality of the year 

 is in the spring months, the amounts below the averages of each 

 town being greatest at Norwich, Wolverhampton, Birmingham, 

 Leicester, and Nottingham. In Scotland, on the other hand, 



§^ 



Fig. 



-Showing the Weekly Deaths among Infants under one year of age on the Annual Mortality po 

 Leicester, Curve i ; Liverpool, 2 ; I^ondon, 3 ; and Bristol, 4. 



of the^whole population. For 



autumn is the healthiest season. In Glasgow and Edinburgh 

 the deaths fall about 20 per cent, below the average in the month 

 of September. 



It is, however, to the summer death-rate that the greatest \ 

 nterest attaches, since it is during the hottest weeks of the year I 

 that the differences in the rates of mortality of the different I 

 British towns stand most prominently out. During the period 

 of high temperature in summer, every one of [the large towns of 

 England shows an excess of deaths above the average, with the single | 

 exception of Bristol, at which place, while there occurs an increased j 

 mortality at this season, it only comes near to, but never quite j 

 reaches, the average. As regards the time of absolute maximum, 

 it occurs in London in the end of July, but at other places more 



generally about the beginning and middle of August. Taking 

 any two consecutive weeks which indicate the highest mortality, 

 the excess per cent, above the average is for Wolverhampton, 6 ; 

 Manchester, 8 ; Portsmouth, 12 ; London, 14 ; Hull, 20 ; and 

 Leicester, 47. The excess above the average at Leicester being 

 thus eight times greater than that of Wolverhampton. 



In Scotland no town exceeds its average during the hottest 

 weeks of the year, but on the contrary the death-rate everywhere 

 is under the average, and in most cases very considerably so. 

 At Aberdeen the rate below the average is 18 per cent, during 

 each of the months, July, August, and September ; and at Dublin 

 the annual minimum occurs in July, when the death-rate falls 

 25 per cent, below the average during the second and third 



Jan. Feb. March. April. May. June. July. Aug. Sept. Oct 



Fig. 2.— Showing the Weekly Deaths from Diarrhoea on the Annual Mortality per t.ooo of the whole population. 

 Liverpool, 2 ; London, 3 ; Bristol, 4 ; Portsmouth, 5 ; and Edinburgh, 6.. 



For Leicester,. Curve i; 



weeks of that month. Though none of the Scottish towns 

 exceed the average at this season, yet Glasgow and Dundee show 

 a decidedly increased mortahty, their curves though rising towards 

 never quite reach the average. 



In a paper on the mortality of London by Dr. Arthur Mitchell 

 and myself, it has been shown that if the deaths of children 

 under one year of age be deducted from the total mortality, the 

 summer excess disappears from the curve ; and it is further showni 

 that, if deaths from diarrhcea be deducted from the whole mor- 

 tality, the summer excess disappears equally as in the former 

 case. Now, these results hold good for every one of the large 

 towms for which the required data have been published. It 

 follows, therefore, that curves of the death-rate for infants and 



• The substance of this paper was read at th« jeneral raetting of the 

 Scottish Meteorological Soaety, held at Edinburgh on July 13, 1875. 



diarrhoea have a peculiar interest in discussions of this nature. 

 Fig. I gives for Bristol, London, Liverpool, and Leicester 

 curves representing the average weekly mortality among children 

 under one year of age, calculated on the annual mortality of 

 I, coo of the whole population, the .'averages being dealt with 

 after Mr. Bloxam's method, according to which each average is 

 calculated so as to include that of the week immediately preced- 

 ng and that of the week immediately following. 



Of all the large towns of England, Bristol has the least summer 

 excess of infant mortality, the highest average of any week being 

 only at the rate of eight calculated on an annual mortality of 

 I, coo of the whole population. In London, the rate rises to ten 

 in the end of July and beginning of August ; and in Liverpool it 

 rises to sixteen, a rate which is also reached by the deaths in 

 Leeds, Hull, and Sheffield, and closely approached_by a number 



