May 20, 1909J 



NA TURE 



339 



These results cannot be obtained on Planck's theory. 



I hope shortly to publish a full account of the assump- 

 tions involved in this calculation, together with additional 

 results. R. A. Houstou.n. 



Physical Laboratory, University, Glasgow, May 14. 



Dimensional Changes produced in Iron and Steel 

 Bars by Magnetism. 



While engaged on research work of an engineering 

 nature, I came upon some facts with regard to the 

 ■behaviour of magnetism on iron and steel bars in the 

 semi-plastic state beyond the yield point that I am un- 

 aware have been noted before. I propose, therefore, to 

 ^ive a brief account of the experiments carried out and 

 the results obtained, on the chance that they may prove 

 of interest to others who have time to pursue the matter 

 further. 



S. specimen of mild-steel about 18 inches long, ^-inch 

 •diameter, and 8 inches between gauge points, having been 

 fixed in the jaws of the testing machine, was surrounded 

 by a solenoid, and a current supplied sufficient to cause 

 magnetic saturation. The specimen had then a tensile 

 load applied to it in the usual way until it ruptured, the 

 magnetism being kept at the saturation point all the 

 time. Other bars were then tested, with and without 

 magnetism, and in the result it was found that the 

 magnetised bars were distinctly less in length between 

 gauge points than the unmagnetised — in other words, that 

 the elongation was less in the first case than in the second. 



In order to make the comparison as fair as possible, and 

 •to eliminate the effects of difference of composition and of 

 manufacture, the specimens for each experiment (consist- 

 ing of the rupturing of one unmagnetised and one 

 magnetised specimen) were each cut from the same bar. 

 A few of the results are given in the table below. These 

 were taken at random from a large number of examples, 

 and will serve to give some idea of the nature of the 

 changes. The material in each case (with the exception 

 of experiment No. 10, in which it was wrought iron) was 

 ordinary mild-steel taken from bars about 12 feet long, 

 just as they were delivered to the laboratory. 



Units = inches and pounds. 



The results may be summarised as follows : — 

 (a) The amount of the decrease of elongation caused by 

 'the magnetism varies from about 3 per cent, to 16 per cent. 

 NO. 2064, VOL. 80] 



(b) The composition of the steel, its hardness, &c., affect 

 the amount of the decrease of elongation. 



(c) The average maximum load without magnetism seems 

 higher than the average maximum load with magnetism. 



(J) The average breaking load without magnetism seems 

 lower than the average breaking load with magnetism. 



(With regard to (c) and id), nothing definite can be put 

 forward, as it is an extremely difficult matter to gauge 

 the maximum and breaking points to a hundred pounds 

 or so on a 70-ton testing machine.) 



(e) Careful measurement shows that, after rupture, the 

 magnetised specimen is thicker all over its length than 

 the unmagnetised, but that the greatest difference is at 

 the place of local extension. This points to the likelihood 

 that the magnetism hinders the flow of the metal, and 

 that this hindering action begins just after the yield point 

 is reached, and attains its maximum value at local 

 extension. This is also brought out in the case of the 

 experiment with wrought iron (No. 10), which shows on 

 fracture numerous planes of cleavage that no doubt 

 hindered the formation of " waist," and caused the 

 relatively small decrease of elongation. 



The following are readings, taken inch by inch, between 

 S-inch gauge points on J-inch mild-steel specimens cut 

 from the same bar : — 



Ajter Rupture. 



Unmagnetised 

 Magnetised ... 



.3rd 



i-38"i7o' 

 I 'So" I -50' 



1-36" 1-24" 

 I '26" I '24'' 



I '20 

 I '20" 



The unmagnetised specimen broke almost exactly 

 between the fourth and fifth inches, and the magnetised 

 at the end of the fourth inch. 



It was thought possible that if the diminution in elonga- 

 tion were due to the magnetism hindering the flow of the 

 metal, tests on a Brincll hardness testing machine 

 might give some results, but though many were carried 

 out, nothing decisive was obtained. A few compression 

 tests were also made, but insuflicient to give trustworthy 

 data. W. J. Crawford. 



Municipal Technical Institute, Belfast. 



" Blowing " Wells. 



In a village about three miles from Norwich, and 

 situated about 140 feet above sea-level, there are three 

 of these " blowing " wells. They are, roughly, about 

 100 yards apart, and each is 3 feet in diameter and from 

 70 feet to So feet in depth. When last opened, some 

 years back, they were found to be empty of water. One 

 of them was then domed over with an iron dome, which 

 after a time blew off owing to the pressure of air (or 

 other gases) within the well. The other two wells have 

 since been domed over in a similar manner, but it was 

 necessary to insert a 3-inch ventilation pipe into the dome 

 in each case because of the great pressure of air that 

 sometimes accumulates within. 



Observation shows that this pressure is sometimes 

 positive for several consecutive days, and that the air then 

 comes out of the ventilation pipe with considerable force, 

 so much so that, in the case of one well which has a 

 grating over the end of thi< pipe, the well " roars " so 

 loudly that it can be heard for a distance of several yards. 

 At other times the pressure in the well is negative, and 

 then leaves and other debris get sucked into the grating. 

 There is a strong belief, locally, that an accurate fore- 

 cast of the weather can be gauged by the intensity of the 

 " blowing." 1 have never heard or read of similar 

 " blowing " wells, and it is difficult to assign an adequate 

 explanation for this alternating positive and negative 

 pressure in the well. Can there be any connection between 

 the blowing and changes in atmospheric pressure, as is 

 locally supposed, or do the rise and fall of the level of 

 the water in the river Yare (which is about two miles 

 distant, and is at this point only about 4 feel above sea- 

 level) have anv possible effect on this curious phenomenon? 



Norwich, May 3. Sydney H. Long. 



