March i8, 1909 J 



NA TURE 



87 



fibre in an alternating field of force very accurately formed 

 ellipsoids or cylinders of the substance to be tested. The 

 field had an intensity of about 300 volts per cm., and 

 made sinuous alternations at a frequency of 80 a second. 

 The polarisation couple upon the specimen was found by 

 measuring the periods of small swings with and without 

 the field. From these, and the dimensions of the body, 

 the dielectric constant was calculated. The values so 

 found for quartz and Hint-glass ellipsoids, carefully made 

 by Messrs. Hilger, agree to one part in a thousand with 

 those calculated by the Sellmeior-Ketteler formula from 

 optical data, and with Hopkinson's values for glass. Sub- 

 stances which could be moulded were formed in a split 

 lead mould, using a steel ellipsoid, also by Hilger, as a 

 matri.\. In order that cylindrical specimens could be used, 

 the longitudinal reaction coetlicient N was found for a 

 series of cylinders of known dielectric constants. Liquids 

 were measured by enclosing them in thin paper cylinders 

 suspended in a saddle of silk thread. The air in the test- 

 ing vessel was thoroughly dried over phosphoric anhydride, 

 ar.d the drying was continued in each case until the period 

 reached a steady value. Quite invisible traces of moisture 

 on the surface of the specimens caused them to behave as 

 conductors, and in the case of water the conductivity of 

 the surface masked the polarisation effect completely. The 

 following values were obtained : — Quartz, parallel to optic 

 axis, 4-606; perpendicular to axis, 4548. Flint-glass, 

 A = 4-65, 1064; A = 4i2, 8-52; A = 3-3o, 6q8. Paraffin 

 wax, 2-32. Beeswax, 475. Shellac, 249. Sealing wax, 

 4-56. Gutta-percha, 443. Chatterton's compound, 308. 

 Ebonite, 2-7q. Amber, 2-80. Ivory, 690. Canada balsam, 

 2-72. Resin, 3-oq. India-rubber, 308. Sulphur, 403. 

 Olive oil, 316. Heavy paraffin oil, A = o-885, 2-55. 



Linnean Society, February 18.— Dr. D. H. Scott, F.R.S., 

 president, in the chair; afterwards Lieut. -Colonel Prain, 

 CLE., F.R.S., vice-president. — Alternation of generations 

 in plants : discussion opened by Dr. W. H. Langr. After 

 some introductory remarks and reference to some examples 

 of well-marked alternation of generations, and the nuclear 

 difference between the two generations, the author adduced 

 the ontogeny of organisms without alternation of genera- 

 tions, the concept of a specific cell corresponding to each 

 specific form. The concept of the specific cell must be 

 applied to organisms with alternation ; the bodies of the 

 two alternating individuals in the life-history may be 

 similar or dissimilar. Two alternative explanations of the 

 differences between the two generations in the complete 

 life-history were stated : — (a) that the differences are due 

 to the different state of the specific cell in the spore and 

 zygote respectively ; (b) that they are due to different 

 environmental conditions acting on equivalent germ-cells. 

 The mode of reproduction — sexuality or spore-production — 

 appears to be necessarily associated with the state — haploid 

 or diploid — of the specific cell. While the possibility of 

 the different states of the specific cell in the spore and 

 zygote having some causal influence on the difference of 

 the resulting individuals must be borne in mind, it is sug- 

 gested that this ontogenetic theory of the nature of the 

 alternation seen in Bryophyta and Pteridophyta may prove 

 a useful working hypothesis, that it will lead to work on 

 new lines, and that it is to some extent open to experi- 

 mental test. 



March 4.— Dr. D. H. Scott, F.R.S., president, in the 

 chair. — .A contribution to the montane flora of Fiji, in- 

 cluding cryptogams, with ecological notes : Miss L. S. 

 Gibbs. Tiic Fiji group consists of 200 islands, only eighty 

 of which are inhabited; Viti Levu is about 4100 square miles 

 in area, with forest-clad mountain ranges, the highest point 

 being Mt. Victoria, 4000 feet in height. The botanical 

 history of the group begins with the visit of H.M.S. 

 Sulphur in 1840, and in the same year the Wilkes expedi- 

 tion touched at the islands. The Herald called in 1856, 

 and Dr. Seemann visited the group in 1860-1, and embodied 

 his results in his " Flora Vitiensis." Mr. Home, director 

 of the Botanic Gardens at Mauritius, spent a year collect- 

 ing in the late 'seventies of last century. Thanks to these 

 investigators, the flora of the lower parts of the chief 

 islands are fairly well known. The author therefore 

 decided to confine her investigations to the region lying 

 at 2qon feet and above, and the three spring months of 



NO. 2055, VOL. 80] 



.August, September, and October were spent at Nadarivatu, 

 the highest inhabited point. From the collections the flora 

 may be described as Indo-Malayan. They contain about 

 forty new species and many new records. Thus, of the 

 eight species of Piper, Mr. C. de Candolle found five to 

 be new, and of Peperomia all seven proved novelties. The 

 introduction concludes with some observations as to the 

 origin of the flora, and is followed by a systematic enumera- 

 tion of the whole collection. 



Physical Society, February 26. — Dr. C. Chree, F.R.S. , 

 president, in the chair. — A laboratory machine for apply- 

 ing bending and twisting moments simultaneously : Prof. 

 E. G. Coker. The paper describes a machine built by 

 students of the City and Guilds Technical College, Fins- 

 bury, in which uniform bending and twisting moments can 

 be applied simultaneously over the whole length of the 

 specimen, and in any desired proportion to each other. 

 The principle of the design is to suspend a rod at two 

 intermediate points by wires depending from a fixed sup- 

 port. The equal overhanging ends of the rod are loaded 

 by weights W, so that the applied couple between the 

 points of support is uniform and of amount Wa, where a 

 is the length of the lever-arm. The rod is also twisted by 

 weights W, attached to equal arms of length b, so that 

 there is a uniform twisting moment of amount W,b 

 between the points of suspension. The two systems of 'oad- 

 ing are independent, and their ratio can be adjusted to 

 any value desired. — The self-demagnetising factor of bar 

 magnets : Prof. S. P. Thompson and E. W. Moss. This 

 paper consists of three parts : — (1) a discussion of the 

 significance and definition of the self-demagnetising factor 

 of magnets in general, and of bar magnets in particular; 

 (2) a re-determination of the values of the self-demagnet- 

 ising factor for bar magnets of circular section ; (3) deter- 

 mination of the values of the self-demagnetising factor for 

 bar magnets of rectangular cross-section of various pro- 

 portions. It is shown that, in general, for every bar 

 magnet there is a self-demagnetising action, the value of 

 which at the middle of the bar depends, for a given 

 intensity of magnetisation, on the length of_ the bar 

 relatively to its cross-section, on the permeability of its 

 parts, and on the distribution of its surface-magnetism. 

 Owing to the circumstance that with every kind of steel 

 the permeabilitv is neither constant nor stands in any 

 simple relation 'to the flux-density, any calculation of the 

 actual polar distribution for rods and bars is impracticable. 

 The only form of magnet that is practicable for calculation 

 is that of the ellipsoid, the properties of which are that for 

 any and every value of the permeability, and in any 

 uniform field, the surface magnetism is so distributed that 

 the magnetic force which this distribution exerts in the 

 interior is uniform at every point within, and therefore the 

 internal demagnetising force everywhere within is constant. 



Zoological Society, March 2.— Mr. G. A. Boulenger, 

 F.R.S., vice-president, in the chair. — The development of 

 the subdivisions of the pleuro-peritoneal cavity in birds : 

 Miss Margaret Poole.— The growth of the shell of Patella 

 i'tils;ata, L., E. S. Russell. The breeding season of this 

 limpet extends from July to January. Sexual maturity is 

 reached at a length of 20-2-; mm. An average size for a 

 limpet of the las't season's brood in January or February 

 is 10 mm. ; at the end of the first year it may be 29 mm. 

 Probable sizes at the end of the second and subsequent 

 years are 38 mm., 44 mm., 48 mm., 53 mm. Shells more 

 "than 50 mm. mav be considcrablv more than five years old. 

 Sexual maturitv "is reached in the first vear, and when the 

 limpet is onlv 'half-grown. The rate of growth decreases 

 with age and maturitv, and is slower during the colder 

 months^ of the vear. 'Consider.able changes take place_ m 

 the ratios of the shells' dimensions during growth, being 

 probably in large part the expression of " laws of growth, 

 and not due to natural selection.— The life-history of the 

 ngrionid drason-flv : Frank Balfour-Browne.— Growth 

 stages in the British species of the toral genus Parasmilia : 

 W. D. Lang. , ,„ T, 



Mathematical Society, March 11.— Prof. W. Burnside, 

 vice-president, in the chair.— The transformation^ of the 

 electrndvnamical equations and the laws of motion : H. 

 Bateman. The paper is occupied with the development of 

 some ideas introduced into the subject by Mmkowski. 



