April ii, 1918] 



NATURE 



113 



The object of the observations was to obtain data for 

 August 21, the date of a solar eclipse, which was total 

 at Theodosia, and for comparison data from some 

 adjacent days. The records were derived with the aid 

 of a magne'tograph of the Mascart type, which is fully 

 described and illustrated by photographs. The curves 

 were read at five-minute intervals for some hours dur- 

 ing the time of the eclipse, and the tabulated results 

 for declination, horizontal force, and vertical force are 

 compared with the corresponding mean results from 

 the adjacent days. The data are exhibited graphically 

 in curves, with corresponding data from De Bilt 

 (Netherlands), Rude Skov (Denmark), Seddin (Ger- 

 many), and Ekaterinburg (Russia). In the case of 

 horizontal and vertical force, it is shown that some 

 movements, which might not unnaturallv be 

 associated with the eclipse if the Central European 

 records only had been available, must be assigned to 

 some other cause, but Prof. Palazzo is disposed to 

 associate some of the declination phenomena with the 

 eclipse. There are a good many references to earlier 

 work on the subject. 



Prof. A. Righi has published a second memoir (i?. 

 Accademki dclle Scienze dell' Instituto di Bologna, 

 November 25, 1917) dealing with the ionisation pro- 

 duced by X-rays in a magnetic field. In the first part 

 of the paper Prof. Righi discusses the question raised in 

 these columns (Nature, vol. c, p. 32, p. 224, 1917) of the 

 possibility of explaining the experimental results as to 

 the increase of current by taking into account- the 

 oblique, and therefore longer, paths of the ions under 

 the joint actions of the two fields. He points out that 

 the kinetic energy of an electron (or of an ion) depends 

 only on the electric field and on the projection of the 

 path on the direction of the said field, and is not 

 affected by the existence of the magnetic field. Prof. 

 Righi 's own view of magneto-ionisation is that an 

 electron in motion can ionise a gaseous atom by colli- 

 sion, when this is in a magnetic field, even if the 

 kinetic energy of the electron does not reach that 

 minimum which is necessary when the field does not 

 exist. On this theory it is possible to explain not only 

 the increase in current due to the magnetic field, but 

 also the fact that when the field is made sufficiently 

 strong there is an inversion of the observed effect, the 

 current diminishing instead of increasing. There are 

 two causes at work, producing opposite effects : mag- 

 neto-ionisation and the magnetic deviation or change 

 in the paths of the particles. The former increases 

 with the magnetic field, but reaches a limiting value; 

 the latter increases indefinitely, and finally gets the 

 upper hand. The paper contains an analytical dis- 

 cussion of the motion of an electron in a uniform 

 electric field on which is superposed a f>erpendicular 

 magnetic field, a problem previously considered by 

 Sir J. J. Thomson (•'Conduction of Electricity through 

 Gases ") and treated elegantly by a purely geometrical 

 method by W. B. Morton (Phys, Soc. Proc, vol. xxi., 

 p. 300, 1909). The last part of the paper gives an in- 

 teresting account of new experiments carried out with 

 an apparatus specially designed to test the existence of 

 magneto-ionisation. Curves are given showing the 

 relation between the current and the applied potential 

 difference for various magnetic fields. These indicate 

 an increase in the current when a magnetic field is 

 applied, the increase being most marked when the 

 potential difference exceeds a certain value depending 

 on the strength of the magnetic field. 



ThB so-called " iminohydrins," or isoam'ides, were 

 first prepared bv Eschweiler in 1897, who gave them 

 the general formula R.C(OH) : NH. They were after- 

 wards (1901) investigated by Hantzsch, and given the 

 Hmolecular formula NH : CR.O.NH, : CR.OH. Dr. 



NO. 2528, VOL. lOl] 



H. G. Rule has studied these compounds afresh, and 

 gives an account of his results in the January issue 

 of the Journal of the Chemical Society. He shows 

 that they are amidine salts of the' general type 

 R.C(NH,):NH,R.CO,H, and that " gIycollimin(v 

 hydrin," the first of Eschweiler's preparations, is really 

 glycollamidine glycollate, 



OH.CH,.C(NH,) : NH,OH.CH,.CO,H. 



The constitution of this and similar compKJunds is 

 proved by its synthesis by the interaction of sodium 

 glycollate and glycollamidine hydrochloride, this 

 method of preparation giving a far better yield than 

 Eschweiler's method of treating the imino-ether hydro- 

 chlorides with moist silver oxiide. Besides the glycol 

 compound methoxyacetamidine methoxyacetate, a'cet- 

 amidine acetate and phenylacetamidine phenylacetate 

 were prepared, whilst mandelamidine mandelate was 

 obtained by Dr. J. E. Mackenzie. Molecular weight 

 determinations, by the cryoscopic method, of these 

 compounds support the new theory of their constitu- 

 tion, on the assumption that they are almost completely 

 ionised in solution. To explain the formation of these 

 amidine salts by the action of water on the imino- 

 ethers, Dr. Rule suggests that the latter first undergo 

 autohydrolysis, forming ammonium salts of the corre- 

 sponding acids, and that ithese then interact with the 

 imino-ethers. 



In a paper on the possibilities of the ferro-concrete 

 ship read by Major Maurice Denny at the Institution 

 of Naval Architects on March 22, the author raises 

 the interesting point of the permissible stress on the 

 steel reinforcement under tension, without the risk of 

 rupture occurring in the adjacent concrete. A usual 

 figure taken in land structures is 16,000 lb. per sq. in. 

 for the working tensile stress in the steel ; with a 

 modular ratio of 12-5 this would produce a tensile 

 stress of about 1300 lb. per sq. in. in the neighbour- 

 ing concrete — i.e. a stress sufficient to produce rupture 

 of some sort. The matter is of serious importance in 

 ship construction, owing to the necessity for maintain- 

 ing watertightness. In the discussion on this paper — 

 reported in Engineering for April 5 — Mr. J. Foster 

 King provided a long and valuable contribution, in 

 the course of which reference was made to the same 

 matter. Taking the elastic modulus of reinforced con- 

 crete to be the same as that of plain concrete 

 — 8 per cent, of that of steel — the permissible 

 stress on the steel must not exceed 5400 lb. per sq. in.' 

 if the concrete is to remain unbroken. As reinforced 

 concrete lost homogeneity under tensile stresses which 

 exceed the breaking stress of the concrete by 45 per 

 cent., the designed working stress on the concrete 

 should be less than its own tensile strength, so as to 

 le^e such a margin between ordinary and extra- 

 ordinary stresses as experience had forced upon ship- 

 builders. Experience of reinforced concrete had been 

 derived from ratios of steel to concrete of about i per 

 cent., and it seemed unreasonable to expect effective 

 bond of steel and concrete when the ratio exceeds 

 8 per cent. Mr. King suggests experiments upon 

 material exposed concurrently to tension and water* 

 pressure, in order to ascertain the point where steel 

 and concrete cease to lend their properties to one 

 another. 



Erratum. — A correspondent points out that it was 

 Pope Innocent VIII. who, in 1484, gave the sanction 

 of the Church to the popular beliefs concerning witches 

 referred to in Nature of April 4 (p. 82), and not Pope 

 Innocent VII., as there stated. The reference in Dt. 

 Withinp:ton's article was correct, but was wrongly 

 given by the reviewer. 



