290 



NATURE 



[March 2, 1922 



nodal diameters, but is important in its effect on the 

 " symmetrical " modes and on modes having one 

 nodal diameter. In the other extreme case, when 

 the flexural rigidity may be neglected, the central 

 constraint has no effect upon the natural frequencies. 

 In the general case, in which both flexural and centri- 

 fugal stresses are considered, the gravest frequencies 

 in modes which have nodal diameters may be calcu- 

 lated by the formula previously given ; a special 

 investigation is made of the gravest frequency in a 

 symmetrical mode. — A. E. Oxley : Magnetism and 

 atomic structure. II. — The constitution of the 

 hydrogen-palladium system and other similar systems. 

 The susceptibility of palladium black charged with 

 hydrogen is less than that of pure palladium black. 

 From this it is concluded that the occluded hydrogen 

 is neither in the atomic nor molecular state. The 

 results agree with the existence of a chemical com- 

 pound, probably Pd-H. In the hydrogen molecule, 

 each atom thrusts its electron into the other atom, 

 the bond being represented by a pair of electrons held 

 in common. The palladium atom has 46 electrons, 

 the hydrogen atom i electron, the latter being thrust 

 into the outer shell of the palladium atom. If these 

 47 electrons take up a configuration like that of the 

 silver atom (atomic number 47), wliich is diamagnetic, 

 the fall of susceptibiUty may be accounted for. 

 Paramagnetic manganese fused in hydrogen becomes 

 ferro-magnetic. The occluded hydrogen atoms prob- 

 ably thrust their electrons into the outer shells of 

 the manganese atoms, producing in them electron 

 configurations analogous to that of the iron atom. — 

 T. Carleman and G. H. Hardy : Fourier's series and 

 analytic functions. If /(X) is integrable in the in- 

 terval (o, 27r), and the associated function ^(w) = 

 \{^{a +w) +/(a -m)}, where o<:fl<27r, is harmonic and 

 bounded in a certain neighbourhood of m = o, then 

 the necessary and sufficient condition for the con- 

 vergence of the Fourier series of /(X), at X = a, is that 

 ^(m) should tend to a limit when u tends to zero 

 through positive values. — A. McAulay : Multenions 

 and differential invariants. Pts. II. and III. The 

 quadratic form is introduced : all multenion formulae 

 may be put into invariant form. Tests, both by finite 

 and by infinitesimal transformation, are given, for 

 ascertaining whether invariance of each one of six 

 types subsists for any given function. Multenion 

 methods are compared with those of the Theory of 

 Tensors and details are furnished for translating from 

 one mode of presentation to the other. Pt. III. is 

 a general survey of the apphcations of a Riemann 

 manifold to relativity. In relation to matter and 

 gravitation no new principle is introduced but the 

 electro-magnetic field is treated in a novel manner. 

 The scalar and vector potentials are wholly ignored. 

 The application to matter in bulk is kept in mind, 

 and it is considered imperatively necessary to adapt 

 relativity methods to a sufficiently general set of 

 relations as at least to leave Maxwell's explanation 

 of crystalline reflection, refraction, and transmission 

 of hght intact. 



Linnean Society, February 16. — Dr. A. Smith Wood- 

 ward, president, in the chair. — R. R. Gates : The 

 inheritance of flower size in plants. Reciprocal 

 crosses were made between CEnothera rubricalyx 

 and CE. biennis, with petals 40 mm. and 20 mm. 

 in length respectively. The size of flowers in Fj 

 was intermediate and relatively uniform. In Fg 

 there was a marked difference in size of flowers 

 (a) on different plants, (6) in different flowers of the 

 same plant, and (c) sometimes in the different petals 

 of a flower. Measurements on Fj and F4 plants 

 show that the hypothesis of several Mendelian factors 

 for length of petal is an insufficient explanation. 

 Variation curves show a tendency to segregation in 



NO. 2731, VOL. 109] 



flower-size between different plants, but also a ten- 

 dency for the occurrence of smaller flowers, some 

 petals being only 7 mm. in length. Segregation is 

 therefore not confined to germ-cell formation, and 

 is not Mendelian. Probably cytoplasmic differences 

 are involved in this type of inheritance and variation. 

 — W. Dallimore : The effect produced by wind at 

 Llandudno. Wind causes remarkable dwarfing of 

 trees and shrubs on the exposed rocks of the Great 

 Orme's Head. — J. L. North: The possible successful 

 growth of Glycine Soja, Sieb. & Zucc, as a profitable 

 crop in Great Britain. The flattening of the branches, 

 the result of close sowing — the Chinese method — is 

 retained even when plants are grown wide apart. 

 Also if a plant starts at a wrong angle it twists itself 

 upon its base to bring it into line with the other 

 plants. By using the earUest ripening seeds of the 

 previous year of a so-called German acclimatised 

 plant earlier maturity has been obtained. Plants 

 in 191 4 ripened on November 28, while last year 

 they reached a corresponding degree of ripeness 

 early in September. 



Royal Microscopical Society, February 15. — Prof. 

 F. C. Cheshire, president, in the chair. — A. L. Booth : 

 The microstructure of coal from an industrial stand- 

 point. Microstructure can be applied to augment 

 chemical analysis in fuel selection. Thin sections of 

 bituminous coals can be classified as humic, spore, 

 and cannel. These are arbitrary standards, thus the 

 use of the microscope in industrial coal problems is 

 purely empirical. For selection purposes an unknown 

 coal is compared with coals of known properties. 

 Little is known of the influence of the various coal 

 constituents on the properties of coal. Chemical 

 analysis does not necessarily give information as to 

 the coal type. Much pubUshed work on coal lacks 

 the essential common factor, the microanalyses, 

 whereby the whole can be correlated. 



Cambridge. 



Philosophical Society, February 6. — Prof. A. C. 

 Seward, president, in the chair. — H. Hamshaw 

 Thomas : On some new and rare Jurassic plants from 

 Yorkshire (V) : Fertile specimens of Dictyophyllum 

 rugosum L. and H. Dictyophyllum rugosum has been 

 known since 1828 and its sporangia have now been 

 found near Scarborough. They are similar to those 

 of the modern ferns Cheiropleuria and Platycerium 

 and confirm the suggested relationship to the Dipteri- 

 dineae.— F. A. Potts: On the food of Teredo, the 

 shipworm. The Teredinidae are invariably found 

 burrowing in wood. The minute fragments excavated 

 by the rotating shells pass through the alimentary 

 canal and are in part digested by it. The stomach 

 has an enormous coecum which retains quantities of 

 wood, but digestion takes place in the so-caUed 

 " liver," some of the cells of which are gorged with 

 particles of wood. In living tissue their amoeboid 

 pseudopodia are active and many of the cells are 

 detached and float freely in the lumen. There are 

 no commensal organisms assisting in wood digestion 

 and it seems unlikely that plankton forms play any 

 significant part in the nutrition of these molluscs. — 

 E. H. Neville : The definition of an envelope. 



Manchester. 

 Literary and Philosophical Society, November 29, 

 1921. — Mr. T. A. Coward, president, in the chair. 

 — F. T. Peirce : Electromagnetic valency and the 

 radiation hypothesis. A magnetic doublet, consist- 

 ing of two electrons in small orbits (ring or vortical 

 electrons), furnishes a probable physical basis for a 



