586 EEPORT — 1887. 



Indeed, calculation shows that according to Fresnels formulae the reflection 

 ■would be nearly insensible at all parts of the spectrum when the index is adjusted 

 for the yellow. The outstanding refl^ection is not due to a difference of wave- velo- 

 cities, but to some other cause not usually taken into account. 



Such a cause might he found in the presence of a tilm upon the surface of the 

 glass, of index ditl'ering from that of the interior, and not removable by mere clean- 

 irg. The glass plate was accordingly repolished with putty powder, after which 

 the reflection was very decidedly diminished. But neither by this nor by any other 

 treatment (e.ff. with hydrofluoric acid) has it been found possible to render the re- 

 flection of a candle-flame at moderate coincidences even ditficult of observation 

 although the adj ustment of indices was as good as could be. 



It would, however, be hardly safe to conclude that no sufficient tilm was opera- 

 tive ; and I do not see how the question is to be decided unless an experiment can 

 be made upon a sm-face freshly obtained by fracture. 



12. On the Magnetisation of Iron m Strong Fields. 

 By Professor J. A. Ewing, B.Sc, F.B.S., and William Low. 



In March of the present year we communicated to the lioyal Society (' Pro- 

 ceedings,' vol. xlii. p. 200) the results of experiments on this subject, in which 

 the magnetism of a narrow neck or isthmus ot iron placed between the pole-pieces 

 of a large electro-magnet was examined, by suddenly drawing out the piece, or 

 turning it end for end so that the direction of its magnetisation was reversed. The 

 piece examined was in the form of a bobbin with a short central neck turned to a 

 small diameter, and with large spreading conical ends, which were in contact with 

 the pole-pieces and provided an easy path for the lines of induction to converge to 

 the central neck. The metal of the neck was in this way subjected to a much 

 greater magnetic force than it would be practicable to produce by the direct 

 action of a magnetising solenoid. The induction in the iron was measured 

 hallistieallj' by means of a coil of fine wire, in a single layer, round the iron neck. 

 The magnetic force in the air-space clo.sely contiguous to the neck was also 

 measured by means of a second or outer induction coil of a slightly greater 

 diameter than the inner one. This determination of the field allowed a correction 

 to be applied for the air-space enclosed by the inner coil, and it also gave what 

 was probably a close approximation to the value of the magnetic force within the 

 metal itself. 



The object of the present note is to describe shortly the results of further 

 experiments of the same kind, the details of which may be reserved for subsequent 

 publication. 



In the former experiments an electro-magnet with pole-pieces 5^^ cms. square 

 was used. The conical ends of the bobbins tested brought the central neck down 

 to a diameter of 0'65 cm. in one form of sample and 0'923 cm. in another form. 

 With this we succeeded in forcing the induction SS in Lowmoor and Swedish 

 wrought iron up to values lying between 32,000 and 33,000 c.g.s. units, the 

 strength of the magnetic field in the air close to the neck being then about 1 1,000 

 c.g.s. units. 



Large as those values were, they have been greatly exceeded in the present 

 series of experiments. In the former paper we pointed out that the magnetic 

 induction of the iron examined showed no sign of approaching a maximum, and 

 that the value to which it might be forced by the ' isthmus ' method depended ou 

 the scale of the experiments. Through the kindness of Professor Tait in allowing 

 the large electro-magnet of the Edinburgh University Laboratory to be brought 

 to Dundee, we have now been able to subject iron to much higher magnetising 

 forces, and to seciu'e a much greater concentration of the lines of induction. The 

 Edinburgh magnet is one of exceptional power. Its limbs, which are vertical, are 

 about 60 cms. long, and the cores are 10'7 cms. in diameter. Rectangular blocks 

 of soft wrought iron 9'6 cms. square serve for pole-pieces. To allow the old 

 bobbins to be effectively used we added a pair of conical intermediate pieces of soft 



