TRANSACTIONS OF SECTION A. 659 



2. The Magnetic Curve. 

 By the Rev. T. J. Smith, M.A. 



By the ' magBetic curve ' is here meant a stream-line in plane space due to a 

 source and an equal sink, or to two equal sources. The author gives a short 

 ■description of a method for drawing it by the use of linkages. Suppose that the 

 source and sink are in a horizontal line at distance a apart. Attach two strings of 

 equal length — one to the point at which the source is fixed, another to the point at 

 which the sink is placed. Attach the other ends of these strings to the ends of a 

 rod, whose length is equal to a, which can move in a horizontal line — the one 

 string being attached to those ends of the two equal lines of length a which are 

 nearest together, while the other is attached to the extreme ends. If the middle 

 points of the cords be now weighted, and the movable line of length a be slid along 

 liorizontally, the point of intersection of the lines proceeding from the source and 

 sink traces the required curve. If the middle point of the cord attached to the 

 near ends of the lines of length a be attached to a string which passes round a 

 pulley placed vertically over the middle point of the cord (so that the cord is 

 stretched upwards instead of downwards) the intersection of the cord, which is 

 attached to one of the fixed points, with the prolongation of the other cord, which 

 is attached to the other fixed point, traces a stream-line due to two equal sources 

 placed at these points. 



3. Certain Volume Effects of Magnetisation. 

 By Professor Cargill G. Knott, B.Sc, F.B.S.E. 



Five tubes of Swedish iron and five tubes of Bessemer steel were made of the 

 same length (45'7 cm.) and of the same external diameter (1'92 cm.). The tubes 

 were distinguished in each case by number, No. 1 having the widest bore, and No. 

 5 the narrowest bore. The diameters of the bores were in order, 3-19, 2'.56, 1-93, 

 1-40, and 0'7 cm. Each tube was clo?ed below, and into the upper end a nut 

 screwed tightly, through a perforation in which issued a fine capillary glass tube. 

 The nut was adjusted under water, so that the whole interior space of the metal 

 tube was filled with liquid, and also part of the glass tube. When the tube was 

 set vertically in the heart of the magnetising coil, the changes of volume were 

 measured by the motions of the liquid meniscus in the capillary tube. These 

 motions were observed by a powerful microscope. 



The changes of volume were observed as each field was applied and removed, 

 first in one direction and then in the other. If this alternation of direction of field 

 were not adopted the metal tube got into a magnetically biassed condition. This 

 magnetic bias was particularly noticeable when the field applied was smaller than 

 the field that had just been previously applied. After exposure to a high field the 

 tube could be neutralised by reversals in the well-known fashion. 



In the following brief abstract of the broad effects, the iron tubes are named by 

 Roman numerals, and the steel ones by Arabic, and the dilatation is regarded as 

 positive when the core space enlarges, and negative when it contracts. jN'os. I, 1, 

 and 2, show negative dilatations in fields lower than from 100 to 200 c.g.s. units, 

 and positive dilatations in higher fields. Nos. II., III., and IV. (all iron) give 

 negative dilatations throughout, up to the highest fields used. No. V. begins with 

 positive dilatation in very low fields, but in moderate and high fields behaves like 

 II., III., and IV. In IV. and V. the negative dilatation increases rapidly in high 

 fields and shows no sign of reaching a limit up to a field of 1,400 luiits. No. 3 

 (steel) begins with positive dilatation in low fields, and continues so throughout, 

 approaching a limit as the ],40U field is reached. Nos. 4 and .5 begin with positive 

 dilatation^, attain a sharp maximum, and aljove fields of 300 and 200 respectively 

 show negative dilatations, which grow rapidly in higher fields as in the case of the 

 iron tubes IV. and V. In the following scheme are given the measured values 

 (multiplied by 10') of the dilatations for all the tubes in fields 100, 1,000, and in 

 fields in which a maximum or minimum is obtained, if such a maximum or mini- 

 mum exist. The values of these last-named fields differ for the different tubes, and 



u n 2 



