2S2 On dcftn'de Proportions, 



subsulphate oF the oxide. The sokition was evaporated to 

 dryness, and the rcnjainder was heated in a platina cruci- 

 ble, in order to expel all the water. It weighed 5 1 gr. 

 When dissolved in water, it deposited some subsalt, of 

 which tlie acid had been driven away with the water ot' 

 crystallizalion, and vvhich, after ignition, \Neighed "24 gr. 

 The sohiiion in water was decomposed by adding to it first 

 caustic ammonia, and then muriate of baryta. The red 

 oxide after ignition weighed 2" ] 6", answering lo 1'498 gr. 

 of pure iron : the salt of baryta 9'7 gr. which contained 

 3-3 of sulphuric acid, or 1-335 of sulphur. Consequently 

 100 parts of iron had been united with 8y of sulphur, that 

 is, with half as much more, within '3, as is contained in 

 the sulphuret at a minimum. This gives for 100 parts of 

 sulphuric acid 65-46 of oxide of iron, containing 20' I of 

 oxygen, or again, within -IQ, the same quantity as we have 

 already found to be appropriated to the base uniting with 

 100 parts of eulphuric acid. According to this analysis, the 

 lieutral sulphate of the oxide of iron consists of 

 Sulphuric acid 60-44 JOO'O 



Oxide of iron 5956 63*5 



The yellow powder, which vi^as deposited bv the neutral 

 sulphate of the protoxide during the digestion in an open 

 vessel, which is considered as a neutralised oxide by The- 

 nard, is a subsulphate. 1 placed it on a filter, washed it 

 very carefully, and dissolved it, while still ujoist, in pure 

 muriatic acid, the solution being efltcted without difficulty, 

 although the san)c salt when dry could only be dissolved 

 by long tioiling. The caustic ammonia threw down oxide 

 which wi igheil -855 gr. alter ignition ; the muriate of 

 baryta -045 of sulphate of barvta, weighed with the same 

 precaution, implying 'd^l of sulphuric acid. Conse- 

 quently 100 parts of sulphuric acid were here conibined 

 with 2G6 of oxide of iron, and the subsali consists of 

 Sulphuric acid 27-33 100 



Oxide of iron 7'2-G7 266 



Since 100 parts of sulphuric acid neutralise 65-5 of 

 oxide of iron, and in the subsalt take up ^66 parts, that is 

 four times as much, it is evident that the same lav.' prevails 

 here as with regard to the sub-^alts of copper. I his gives 

 22 parts of sulphur for 100 of iron,, which indeed is one 

 quarter of the Cjuaiitily of sulphur coniained in the neutral 

 salt of the oxide, but stands ni no sin)|)le relation to the 

 quantity of sul|)hur with which iron can he combined with- 

 out the presence of a third body ; a proof that, under cer- 

 tain circumstances, nature departs, in the case of compli- 

 cated 



