Mis. No. 48. 63 



two others of being in harmony with the decimal system, and offacilitatin"' 

 calculations. It has tiie advantage over the Fahrenheit of poi:s(\ssint,'a imn- 

 ral zero, a point of departure fixed by nature herseh", easy todeternHne,and 

 which corresponds to a point of temperature so critical, that from' the 

 moment of reaching it, meteorological phenomena are snddeidy and pro- 

 foundly modified, vegetation ceases, and all nature assumes another as- 

 pect. It is then highly proper that this temperature, which is a point of 

 change marked by nature herself, be also the principal one on the ther- 

 mometric scale. 



The English adduce, as one of the reasons Avhich causes diem to retain 

 the Fahrenheit scale; the absence of negative quantities. This, it is true, is 

 an advantage; for 1 admit that the mixing of negaiive and positive quan- 

 tities is a source of error, and embarrasses the calculation. Hut this advau- 

 tage, which is real, disappears the moment we wish to niake a uni- 

 versal one of the Fahrenlieit scale, for in nearly half the countries of the 

 earth, and particularly in North America, the Fahrenheit thermometer fur- 

 nishes every year many negative quantities. It has been said further, that 

 the Fahrenheit degree, measuring a smaller diflerence of temperature, allows 

 of a more exact reading than the centigrade. i3ut it is only requisite 

 to mark half degrees upon the centigrade scale in order to obtain, the 

 tubes being equal, a still greater approximation, that is to say, accord- 

 ing to the ordinary method of estimation of twentieths of degrees; which 

 is much more than it is righttodemandof instruments destined for meteoro- 

 logical observations, since various causes prevent, even with the best, that 

 one could guarantee the temperature of the air within two or three tenths 

 of a degree. 



But the strongest reasons for the adoption of the centigrade scale are, 

 without contradiction, the predominance already secured to it in the greater 

 part of the scientific world, the considerable amount of meterological 

 labors in which it is employed, and the decided tendency contimially to ex- 

 tend itself. America has not here to open a aew path, she has only to as- 

 sociate herself with the movement which is carrying scientific Europe 

 along with it. But I repeat, her influence wid be decisive, she holds in 

 her hands the future, and if she well understands her part, here as in other 

 domains, she should break with the past in its narrowness, and render 

 herself cosmopolitan in the interest of general utility. 



What I have just said applies equally to the barometric scale. The met- 

 rical scale is the most convenient; it is rational, easy to verify, invariable; 

 it has also the future in its favor. 



There remain to be considered material difficulties which attend every 

 change of system. I will take the liberty of calling your attention to the 

 fact that they only concern other institutions. The Smithsonian Institu- 

 tion is perfecdy free to take the course which may seem to it most in accord- 

 ance With the interests of meteorology. It has, moreover, a right here to 

 an influence proportioned to its expenditures in this department. If, lor ex- 

 ample, it takes upon itself the gratuitous publication of the observations 

 made 'in stations already existing, may it not in time claim to regulate also 

 the form of these observations. 



Besides, the questionis not about changing at once the habits ot the 

 people, put simply those of scientific observers, which is a very difierent 

 afl'air The detailed publication of the observations will principally cnxulate 

 in the scientific world tor which it is designed. But nodnng would pre- 



