13G Mr. J. H. Bickct on the 



produced by a single bar-magnet and fixed permanently to a 

 slip of card by gum was suspended lightly upon a needle- 

 point by means of a glass cap. It set itself in the magnetic 

 meridian, and was found capable of being deflected on the ap- 

 proach of a steel magnet. The fixed magnetic curves are 

 therefore themselves magnets. 



Y. Magnetic Figures of three dimensions. 



The writer has several times essayed to produce magnetic 

 figures of three dimensions. The difficulty in producing them 

 arises from the weight of the iron filings when unsupported, 

 as they must be when the whole figure does not lie in one ho- 

 rizontal plane. With even the most powerful electromagnets 

 the forms of the curves cannot be actually constructed in iron 

 filings for more than a few millimetres length. 



Attempts to float iron filings are also difficult, as there is 

 no transparent liquid nearly approaching the density of iron. 

 The writer has tried heavy paraffins and strong solutions of 

 mercuric iodide in potassic iodide. Better effects were ob- 

 tained, however, when iron filings were employed which had 

 previously been coated with shellac varnish, and which there- 

 fore possessed greater buoyancy. The experiment is curious ; 

 but the difficulty of seeing across the forests of lines of filings 

 reduces the observation to one of curiosity only. 



Another process attempted consisted in plunging a small 

 magnet into a soft paste of plaster of Paris and iron filings. The 

 plaster shortly hardened ; and then sections were cut in various 

 directions. The figures observed, however, were poor ; and 

 no observations were made of any additional interest as the 

 result of the attempt. 



XVII. On the Dissociation of Aniline Colours. By J. H. 

 Bicket, Assistant in the "Young" Laboratory of Tech- 

 nical Chemistry, Glasgoiv*. 



IN the course of their researches on dyeing, an account of 

 which has recently been communicated to the Chemical 

 Societyt, it was found by Professor Mills and Mr. Thomson 

 that a dilute aqueous solution of rosaniline acetate or hydro- 

 chloride is entirely decolorized by boiling, nevertheless im- 

 parting the rosaniline salt in its normal red state to a piece of 

 silk immersed in the heated liqued On Professor Mills's 

 suggestion, I have made some further experiments in con- 

 nexion with this subject. 



* Communicated by Dr. Mills, F.R.S. 



t ' Journal ' (1879), i. p. 20. 



