An Essay on the ifiedical Effects of Climates. 257 



such wesierly winds predominate, is more than twice as 

 great as in London : or, if we convert the measure into 

 days, that the predominance amounted, in 1794, to 68 days 

 for London, of a wind nearly W.S.W. and to 170 days 

 for Dawlish, of a wind a litile to the south of west. 



The variations of tlie chmate of the same place, with re- 

 spect to mean temperature, are easily collected from the 

 usual meteorological computations. Dr. Heberden has very 

 successfully combated the common opinion rcspectino- the 

 superior salubrity of cold winters ; it appears however that 

 the winter which he particularly observed was more varia- 

 ble, as well as colder, than usual, Mr. Kirwan has at- 

 tempted to account for the greater frequency of colds, which 

 he supposes to occur in spring and in autumn, by the greater 

 variability of the temperature at those seasons : but both 

 the fact and the explanation are very questionable; for in 

 reality the variations of temperature, if estimated by the 

 total range of the thermometer within the 24 hours, are 

 alniost uniformly greatest in the hottest weather. In Lon- 

 don, the greatest variations of successive days at the same 

 hours in the morning are greatest in winter; in the after- 

 noon, in summer; and although the latter are a little greater 

 in April than in some of the succeeding months, the dif- 

 ference is by no means considerable. 



Of the empirical evidence, which may be collected, re- 

 specting the medical effects of different climates, the most 

 authentic is perhaps that which is derived from well regu- 

 lated bills of mortality; since these documents ought to 

 afford us a tolerable criterion of the general healthiness or 

 unheallhiness of a place, from the proportion between the 

 annual deaths and the population, and at the same time 

 a pretty correct determination of the degrees in which dif- 

 ferent diseases are fatal. Thus, when we find that in Stock- 

 holm the annual deaths amount to -j^g- of the population, 

 in London to .jU, in the Pays de Vaud to ■,'•3-, and in some 

 villages in different parts of Great Britain to ^ only, we 

 cannot hesitate to consider a residence in the country as 

 generally nioie healthy, than in a metropolis similar to 

 either of those cities; although it cannot fairly be concluded 

 that the healihiness is precisely in the proportion which 

 might be inferred from this comparison^ until we have con- 

 sidered how far the effect of emigration 10 a great town may 

 influence the appaient mortality. After the age of eight 

 or ten, the j)robable duration of life may be estimated 

 with sufficient accuracy, as Demoivre has very ingenioi.sly 

 •hown, by assumintj that, of a certain number of persons 



Vol. 41. No. ISO. Jpril 1813. R bora 



