MR. HATCHETT ON MAGNETICAL PYRITES. g[Y 



ployed to form artificial magnets) with that of fulphuret and. 

 phofphuret of iron; each being firft examined in the form of a 

 iingle mafs or bar of equal weight, and afterwards in the ftate 

 of compound magnets, formed like the large horfe-ftioe mag- 

 nets, by the feparate arrangement of an equal number of bars 

 of the fame fubftance in a box of brafs. 



The efFedsof the above compound magnets fliould then be An extenfive 

 tried againft others, compofed of bars of the three different ^"^'^ ^*^^,^^H"'- 



/•in • ■ I 1 • *ii ^ f ments of intereft 



labftances, various m number, and mihemodeor arrange- and importance. 



tnent; and, laftly, it would be interefting to make a feries of 

 experiments on chemical compounds, formed by uniting differ- 

 ent proportions of carbon, fulphur, and phofphorus, with one 

 and the fame mafs of iron. Thefe quadruple compounds, 

 which, according to the modern chemical nomenclature, may- 

 be called carburo-fulphuro phofphurets, or phofphuro-fulphuro- 

 carburets, &c, of iron, are as yet unknown as to their chemi- 

 cal properties, and mayalfo, by the inveftigation of their mag- 

 netical properties, alTord fome curious refults. At any rate, 

 an unexplored field of extenfive refearch appears to be opened, 

 which poffibly may furnifti important additions to thehiftoryof 

 magnetifm, a branch of fcience which of late years has been 

 but little augmented, and which, amidft the prefent rapid pro- 

 grefs of human knowledge, remains immerfed in confiderable 

 obfcnrity. 



III. 



Extradtofa M^noir of Mr, Erman, entitled Obfervations and 



Doubt.'i concerning Atmofpheric Eledricily, 



(Concluded from p. 300, Vol. X.) 



N electrometer furniftied with a rod three feet in length. Two ele^irome- 

 and placed in the open air, does not fliewany divergence; but^^^^J."^^ ^. ' 

 when the bent point of another ele6trometer, which alfo ex- vergence alone, 

 hibits no divergence, is moved above the firft, and even when ^*^^ ^^ ^^'^^" 

 the motion is parallel to the horizon, the leaves of the latter above the other, 

 win be feen to diverge negatively, without the fecond giving 

 any fign of electricity. 



ft is very probable that the effeCl of thcfe vapourous and This effea pro- 

 8quer>u=; meteorological raafle^ is raanifeftcd at the points of the ^/^^'v arises horix 



\t xrx ,, .^ the divifion or 



Vol. XI.— May, 1805. C eleclrometer, theeieilridty. 



