1814.} Royal Irish Academy. 451 



may be expected from experiments on the refractive force and varia-' 

 tion of density of air than from observations of circumpolar stars, 

 even with the best instruments. It is still a question, as far as 

 observations of circumpolar stars are concerned, whether at the 

 above mean height of the barometer and thermometer, the refrac- 

 tion be 57", according to Bradley, or 57'5", according to the 

 French tables ; or whether, according to other astronomers, it may 

 not exceed 58". But the French tables agreeing so nearly with the 

 above conclusions, the author conceives that as far as 7^° they 

 ought to be generally adopted. It is of considerable importance that 

 all astronomers should use the same refractions, and these, by their 

 close coincidence with what may be derived as above from the 

 direct experiments on air, seem entitled to the preference. At 7-±° 

 zenith distance (bar. 29*60, and Fahr. therm. 50) the French tables 

 give refraction = \dS"2". 



Dr. Brinkley gives two tables of refraction derived from the for- 

 mula by which the same refraction as far as 74° is deduced as by 

 the French tables. These tables are very convenient for use, 

 equally so as the French tables ; but they are principally given to 

 afford a convenient check in the use of the French tables. 



Below 80° a knowledge of the law of variation of density in the 

 atmosphere is absolutely necessary for forming a table of refraction ; 

 but if even this were known little advantage would thereby be 

 gained in the nicer researches of astronomy, as within 10° or 12° 

 of the horizon the quantity of refraction varies so much, the height 

 of the barometer and thermometer being the same, as to render 

 observations made with the best instruments of little value. 



This paper concludes with the results of observations of circum- 

 polar stars, and remarks thereon. 



IMPERIAL INSTITUTE OP FRANCE. 



Account of the Labours of the Class of Mathematical and Physical 

 i oj' the Imperial '-institute of I'rcuue during theYair 1813. 



(Concluded from p. :J91.) 



VEGETABLE PHYSIOLOCV AND BOTANY. 



M. Henri Cassia H sop of one of our a iderabk 



of observations by whom we announced last year pn the gi 

 family of plant* known by the name of compound, syn 

 synaihtrue, struck with the principle explained by natural philoso- 

 phers, that a classification to give a just idea of bodies ought to be 

 founded on the whole of their character*, has thi< year examined 

 the stamina of the same family, and b overed several facts 



unknown to botanist*, Nobody, fpr example, had observed the 

 artipuUtUro which divides the filament id the neighbourhood of the 

 anthera, a character which M dqCassini has found much more 



•J w 2 



