42 



L. F. Nilson and Otto Pettersson. [Nov. 18. 



Atom, lie at 

 of oxygen. 



In Ga 3 3 c= 184 . 0-1062=19-54 

 Ga 2 c= 136 . 0-0802=10-91 



8- 63 = 3. 2-88 



„ ln 2 3 c=274'8 - 0-0807=22-17 

 In 3 c=226'8 . 0'0570=12-92 



9- 25 = 3. 3-08 



According to the determinations made by ns under identical con- 

 ditions, and therefore strictly comparable, the atomic beat of oxygen 

 in beryllia=Be 2 3 is the same as that in alumina ; this, however, can 

 by no means be considered as exceptional. Alumina and beryllia are 

 the leading members of a group of sesquioxides, wherein the atomic 

 heat and (as will be seen from the following paper) the atomic volume 

 of oxygen increases with increasing values of the atomic weights of 

 the metals. 



We will now refer to another objection raised to our former re- 

 searches. Mr. Brauner* admits the specific heat of beryllium to be 

 0'4084 between — 100°, but supposes that it may rapidly increase 

 with the temperature in the same way as does the specific heat of 

 C and Bo. If this were the case, he thinks beryllia could be 

 BeO, the atomic heat of beryllium= 3' 76 between — 100°, and 

 normal=6 , 4 first at a much higher temperature, In our detailed 

 paper, we have tried to meet such an objection, by pointing out that no 

 metallic element is as yet known, the atomic heat of which does not 

 agree with the law of Dulong and Petit. However, in order to remove 

 any doubt in this respect, we have determined the specific heat of 

 beryllium at different temperatures lying between — 300° C. We 

 sifted the metal used in our former determinations through a gauze 

 of platinum, the holes of which were - 25 sq. miliim. For the 

 following experiments we used only that part which did not pass 

 through the gauze, on the supposition that this, consisting of globules 

 and larger crystals, was the purer metal. The analysis confirmed 

 this opinion, for the composition was found to be — 



Beryllium 94*41 



Beryllia 4'89 



Iron 0-70 



100-00 



The following table gives only the results, not the details of our 

 * " Ber. d. Deutscli. Chem. Gesellsch.," xi, 1872. 



