XXXVI NOTES. 



words peifav or fhaffo-wv (Cod. Par. N 2389). Tycho Brahe expressed this 

 increasing or diminishing by points. 



( 155 ) p. 81. Sir John Herschel, Outl. of Astr. p. 520527. 



( lo6 ) p. 82. This was the application of mirror sextants to the determina- 

 tion of the light of stars which I employed in the tropics still more than 

 diaphragms, which had been recommended to me by Borda. T began the 

 work under the fine sky of Cumana, and continued it subsequently up to 1803, 

 undei less favourable circumstances, on the high plains of the Andes, and on 

 the coast of the Pacific at Guayaquil. I had formed for myself an arbitrary 

 scale, in which I made Sirius, as the brightest of all the fixed stars, = 100 ; 

 stars of the 1st magnitude-between 100 and 80 ; 2d magnitude between 80 

 and 60 ; 3d magnitude between 60 and 45 ; 4th between 45 and 30 ; and 

 5th between 30 and 20. I passed in review more particularly the constella- 

 tions of Argo and Grus, in which I believed I should find alterations 

 since Lacaille's time. It appeared to me, after careful combinations of esti- 

 mation, and employing other stars as intermediate gradations, that Sirius is 

 as much superior in the strength of its light to Canopus, as a Centauri is to 

 Achernar. On account of the above-mentioned mode of classification, my 

 numbers do not admit of direct comparison with those given since 1838 by 

 Sir John Herschel. (See my Recueil d'Observ. astr. Vol. i. p. Ixxi. ; and 

 Relat. hist, du Voy. aux Regions equin. T. i. p. 518 and 624 ; also Lettre 

 de M. de Humboldt a M. Schumacher en Fevr. 1839, in the Astr. Nachr. 

 N 374.) In this letter I say : "M. Arago, qui possede desmoyens phoco- 

 metriques entierement differents de ceux qui ont ete publics jusqu'ici, m'avait 

 rassure sur la partie des erreurs qui pouvaient provenir dn changement d'in- 

 clinaison d'un miroir entame sur la face interieure. II blame d'ailleurs le 

 principe de ma methode, et le regarde comme peu susceptible de perfectionne- 

 ment, non-seulement a cause de la difference des angles entre Tetoile vue 

 directement et celle qui est amenee par reflexion, mais surtout parce que le 

 resultat de la mesure d'intensite depend de la partie de 1'ceil qui se trouve en 

 face de 1'oculaire. II y a erreur lorsque la pnpille n'est pas tres-exactement 

 a la hauteur de la limite inferieure de la portion non entamee du petit 

 niiroir." 



( 157 ) p. 82. Compare Steinheil, Elemente der Helligkeit's-Messungen am 

 Sternenhimmel Munchen, 1836 (Schum. Astr. Nachr. No. 609) and John 

 Herschel, Results of Astronomical Observations made during the years 1834 

 1838 at tbe Cape of Good Hope (Lond. 1847), p. 353357. In 1846, 



