ISM.] Mr. Ekelergi o 43 



suavity of disposition which threw a glow of kindness over his coun- 

 tenance. But during the latter part of his life his face was marked 

 by a languor, or melancholy, sufficiently conspicuous. He was sus- 

 ceptible of friendship, and sincerely devoted to his friends. But 

 his learning and science constituted the only solace of his life ; for 

 lie was burdened with a large family, and had but very slender 

 means of providing for them. The following is a catalogue of all 

 his writings : — 



Pullished separately. 



1. Dissertatio Mus. Nat. Acad. Ups. Pars III. 1787. 



2. de Oleis Seminum expressis. Ups. 1/88. ' 



3. de Cake Phosphorata. Ups. 1793. 



4. de Topazio. Ups. 179fi. 



5. de Materiis Oleosis e Regno Animali. Ups. 1796* 



6. de nova Analysi Aquarum Mediviensium (in con* 



cert with Dr. Berzelius). Ups. 1800. 



In the Memoirs of the Royal Academy of Sciences. 



7. Experiments on the black Stone from Ytterby, and on the 

 new Earth which it contains. 17^7- 



8. Elucidation of the peculiar Properties of Yttria, and particu- 

 larly a Comparison between it and Glucina: with an Account of 

 the Minerals in which this new Earth is found, and a Description 

 of a new Body of a metallic Nature (Tantalum). 1802. 



9. Chemical Analysis of a Swedish Titanite. 1803. 



In the Journal of Swedish Literature. 1795. 



10. Account of the present State of Chemical Science; with 

 many Observations. 



1 1 . Poem on the Peace of 1 790. Besides this, he published 

 various other Poems. 



Article II. 



On the Composition of Arragonite. 



Tiik striking difference between arragonite and calcareous spat 

 has been long known to mineralogists. The two minerals are dis- 

 tinguished from each other by their hardness, their specific gravity, 

 the forms of their crystals, which cannot he considered as de- 

 rived from the same primitive form. Bernhardt indeed has endea- 

 roored to reconcile the crystalline forms of the two minerals with 

 each other. But we have the tuthority of Haiiy for savin-- that the 

 attempt has not hecn successful. To what is ibis <! • ,,1' pro- 



perties to be ascribed? Many analyses of anagonile have bcea 



