88 



L. F. Nil son and Otto Pettersson. [Not. 18, 



cnlar volumes of Be 3 3 with A1 2 3 , and also by Ebelmen,* who ob- 

 tained the oxide of beryllium in crystals isomorphous with A1 2 3 . 

 Subsequently an elaborate work, " De Glucium et de ses Composees," 

 1855, by Debray,f once more caused a change in the prevalent opinion. 



Debray regarded beryllia as an isolated member of the series occupy- 

 ing a position intermediate between the monoxides and the sesqui- 

 oxides, and showing marked analogies with both groups, but not 

 intimately connected with either of them by isomorphism. As the 

 analyses of its compounds in most cases agreed better with the simple 

 formula BeO, this was preferable to Be^Og. 



Klatzo.i in 1868, endeavoured to decide this matter finally by the 

 assumption of a complete isomorphism between the sulphates of Be 

 and Mg, Co, Fe, M, and though Marignac,§ in 1873, proved that 

 this pretended isomorphism did not exist, and was founded on a grave 

 mistake, the opinion that beryllia was a monoxide was at this time 

 universally accepted by chemists. Theoretically this opinion was 

 founded on the " periodic law " of Mendeleen\|| The classification of 

 beryllium at the head of the second group among the diatomic elements 

 is a leading point in the theory of Mendeleeff. 



If the composition of beryllia was Be 2 3 , and the atomic weight of 

 beryllium =13"8, instead of 9*2, the place of Be = 9*2 in the system 

 would be vacant, and the order of the series partially reversed, the 

 " atomic analogies " would be overthrown, and, still worse, beryllium 

 = 13*8 would find no place at all in the system, except in the fifth 

 group among the five-atomic elements, to which it certainly does not 

 belong. 



The final decisive proof still wanting to confirm the ideas of 

 Mendeleeff's was furnished, 1877, by Reynolds, 1 ^ who found the 

 specific heat of metallic beryllium = 0"642, which (Be = 9*2) makes the 

 atomic heat=5*9, in accordance with the law of Dulono- and Petit. 



About a year before the publication of Mr. Reynolds, we bad suc- 

 ceeded, by means of a new method, in isolating metallic beryllium from 

 its chloride. We employed** a massive crucible of wrought iron, 

 hermetically closed by a screw-plug, wherein equivalent quantities of 

 beryllium chloride and metallic sodium were heated to redness. 

 Metallic beryllium was thus obtained, partly fused into globules, partly 

 forming aggregations of little prismatic crystals, which in brightness 



* "Ann. d. Oh. n. Pharm.," lxxx, p. 211. 



f " Ann. de Chimie et de Phys.," [3], xliv, p. 5. 



X " TJeber die Constitution der Beryllerde." Dorpat, 1868. " Journ. f. Prakt. 

 Ch.," cri, p. 227. 



§ " Ann. de Chim. et de Phys.," [4], xxx, p. 45. 



|| "Ann. d. Ch. u. Pharm., Suppl.," viii, p. 151 (1871). 



1 "Phil. Mag.," [5], iii, p. 38. 



** For the details of the experiment see " Darstellung und Talenz des Beryl- 

 liums," "Pogg. Ann.," [2], iv, p. 554 (1878). 



