420 HORARY VARIATIONS OF THE BAROMETER. 



tlae cliimr.ey-screen if not well secured in its position. The odor of soot per- 

 vades the apartment, distinctiy indicating the presence of a descending current 

 of air in the flue of the chimney. Xow, it is this descending movement of the 

 air in the chimney which represents, in the diurnal period of the barometer, the 

 phase commencing at three or four o'clock in the afternoon and continuing until 

 after midnight. 



Is it ijot a curious thing to find, in what occurs at the fireside of the most 

 simple dwelling, so complete an analogy with all the circumstances of the plie- 

 nomenon, apparently so complicated, which is called the diurnal period of the 

 barometer 1 Need we hesitate once more to repeat that it is the presence or 

 absence of the sun, the warming or cooling of the earth which thence results, 

 which is the sole cause of all meteorological phenomena, from the ascent and 

 passage of a soap-bubble to the most fearful tempests ? Let us carefully observe, 

 then, these effects of heat and cold, not by studying the phases of the moon and 

 the tides which it may produce at the limits of our atmosphere, not by consider- 

 ing the stars, but by attending to what passes around us on the surface of the 

 earth. 



Biot has said, and M. Poey has recently repeated the remark, that it is neces- 

 sary to commence our study of meteorology from above ; we are not at all of 

 that opinion. Almost all the phenomena of meteorology — rains, snow, hail- 

 are accidents solely due to differences of temperature ; they are produced at the 

 surface of the earth, and it is they which occasion the disorderly movements of 

 the lower strata of the atmosphere ; a little above these there is only a blue_ or 

 black sky, absolutely a stranger, let us believe, to all those phenomena which 

 occupy so large a space in our own experience. 



LETTER OF MARSHAL VAILLANT, ON THE FORKGOING THEORY, TO M. CHARLES 

 SAINTE-CLAIRE DEVILLE, MEMBER OF THE INSTITUTE.* 



I have attentively read your memoir, or, as you term it, monograph, On the 

 Barometric Phenomena of the Antilles and Neighboring Countries. Thefollowing 

 are some of the reflections Avhich that examination has suggested to me. 



The observations which you report confirm the truth of the theoiy which I 

 have developed in my note on the horary variations, and I have no doubt that 

 observations which will be made in the future, and for which, I trust, that theory 

 Avill be accepted as a guide, will lead to the most satisfactory confirmation. 



You are right in saying that the different elements of the total oscillation un- 

 dergo the constant influence of the solar heat. Yes, it is to the influence of that 

 heat I hat the ascensional movement of the mercury in the barometer is due, in 

 consequence of the inertia which the air and Avatery vapor mingled together 

 oppose to the ascension which the heat communicated tends to impress upon 

 them ; and it is also to the diminution of the calorific effect on the air and watery 

 vapor, and to the inertia v/hich retards their being put in motion, that is due the 

 subsidence of the mercury in the barometric tube. Inertia, that force which is 

 ever and everywhere encountered, has been, perhaps, rather too much neglected 

 by the meteorologists. 



■ The cause being known, or at least suspected, we can calculate its effects and 

 proceed in advance of direct observation ; this will come afterwards to the sup- 

 port of the theory. Thus, as you very properly say, the pihenomenon of the 

 diurnal variation, at the level of the sea, considered in the establishment of the 

 tropical hours or in the extent of the oscillation, is manifestly connected with the 



* This letter forms a sequel to the previous article, On the Horary Variations of the Barolnt- 

 ter, and was designed to fumLsh to the explorers and correspondents of the scientific commis- 

 sion of Mexico indications respecting the direction to be given to their barometrical observa- 

 tions. 



