180 Reflexions sur itn Memoire de M. Nicollet 



The result of this analysis gave 



Alumine ^S'TS 



Silex 07-90 



Water 48-55 



Peroxide of iron 00-80 



100-00 

 If this be taken as the more correct determination, the sub- 

 stance may be considered as a silicate of alumina, in which 

 the alumina is rather more than five times the weight of the 

 silica. Klaproth and Berthier have analysed two minerals, 

 the one called Kollyrite, the other, siliciferous hydrate of alu- 

 mine, in both of which the alumina is about three times the 

 weight of the silica ; and there is a mineral called AUophane, 

 and another called Lenzinite, in which it appears, from the 

 analyses of Stromeyer and Dr. John, that the proportions of 

 these elements are nearly equal, the weight of the alumina in 

 allophane being somewhat greater than that of the silica. The 

 atomic weights of alumina and silica are not perfectly settled ; 

 but it is evident that they do not differ widely from each other, 

 and that the atom of alumina weighs rather more than the 

 atom of silica. It is probable, therefore, that in the two last 

 of these minerals, one atom of silica is combined with one 

 of alumina ; that in the minerals analysed by Klaproth and 

 Berthiei", one atom of silica is combined with three atoms of 

 alumina, and that, in that now described, one atom of silica 

 is combined with five atoms of alumina. Should this view 

 of its composition prove to be correct, it will require a sepa- 

 i-ate mineralogical name, and may be distinguished by the 

 appellation of Scarbroite. 



XXVIII. Beflexioris sur un Memoire de M. T. N. Nicollet, 

 inserc dans la Connaissance des Tems pour VAn 1831, "&//• 

 un Nouveau Calcid des Latitudes de Mont-Jouy et de Barce- 

 lojie, 2^our servir de Supplement au Traite de la Base du Sy- 

 stcme Metrique*." 



TL y a long tems que Ton connait le phenom^ne assez etrange, 

 -^ que les observations des latitudes faites en 1792, avec un 

 soin extreme avec des ceixles-repetiteurs de Lenoir, par Me- 

 chain, a Mont-Jouy, et I'annee suivante a Barcelone, ont donne 

 une amplitude de Tare du m6ridien plus forte de 3",24, qu'elle 

 ne devrait eti'e suivant la distance geodesique tres-bien con- 



* Comniiinicated by a correspondent. 



