5 7 2 - C U R R E N T TOP I CS . [J. F, I.

silk. In this respect it even excels na tura l silk. Th is feature per- mits the silk to be woven on looms running at the highest speed.

P lan t has also been installed for the manufac tu re of non- inf lammable celluloid, t e rmed " celastoid." This product has all the advan tages of celluloid wi thou t the g rea t d i sadvan tage of high inf lammabi l i ty . Such a mater ia l is of g rea t value to the celluloid trade, where the high price of c amphor has ser iously res t r ic ted the manufac tu re of celluloid. The day is not far dis- tan t when the use of dangerous highly inflammat~le celluloid will be prohibi ted.

The marve lous insula t ing p rope r ty of cellulose aceta te silk or sheet ing is a factor that has a roused grea t interest in the elec- trical and allied trades, and the long-sought - for non- inf lammable accumula to r boxes, wind screens, and a mul t i tude of other articles of necess i ty will shor t ly be placed on the market .

T h e Crys ta l S t ruc ture of Ice. D. hi. DENNISON. (Phys. Rev., January, I 9 2 I . ) - - D i s t i l l e d wa te r was frozen in a capi l lary tube. The tube with the ice was ro ta ted dur ing a ten-hour exposure to X - r a y s as a resul t of which a series of twelve lines were obtained. The method was tha t of A. W. Hul l ' s c rys ta l powder photo- graphs . T h e in te rpre ta t ion of the resul ts gives an axial rat io of 1.62, while the value furnished by the crystallographer is 1.617. The n u m b e r of molecules of H,_,O per cu. cm. is calculated to be 3.154 )~ lO 22. The volume of an elementary prism can likewise be derived f rom the exper imenta l resul ts and is found to equal o.6478 X lO-22 ,cu. cm. From these two it follows that in an ele- mentary prism there are 2.o4 H 2 0 molecules. " Ice belongs, then, to one of the two a l te rna te a r r a n g e m e n t s which inelastic un i form spheres m a y assume if packed as closely as possible. Th is is the same form tha t magnes ium a toms take. . \Vith increased knowledge of crysta l s t ruc ture these data may furnish informa- tion regard ing the shape of the molecule of water . W i t h the hexagonal close packed type of lattice each pr ism contains on the average one molecule. The resul t calculated above indicates tha t in ice crys ta ls the molecules of wa te r are of the fo rm (H20)2 or H~O~." G . F . S .

An E lec t romagne t i c T h e o r y of Gravi ta t ion. H. A. WILSON. (Phys. Rev., January, I 9 2 1 . ) - - W h e n an electric sys tem is placed in a m e d i u m whose specific induct ive capaci ty var ies f rom point to point it experiences a force tending to move it toward par t s of the field where that quantity is greater. A like force acts in a field wi th var iable magne t i c permeabi l i ty . T h e au thor bases a theory of g rav i t a t ion on these facts wi th the assumpt ion tha t both the specific induct ive capaci ty and the permeabi l i ty of the e ther are modified by the presence of ma t t e r and, further , tha t the d imensions of the body acted on change when it passes

April , I92I. ] CURRENT TOPICS. 573

f rom one par t of the variable field to another. He is able to show that his theory is consis tent with the inverse square law, with the deflection of l ight by the sun and with the gravi ta t ional effect upon the f requency of light emit ted by an atom. " I t may be said that since ma t t e r is cer tainly par t ly electrical and since the refract ive index of the ether certainly varies near large masses, it seems certain tha t part , at least, of the observed gravi ta t ional forces mus t be due to an action of the kind considered in this paper, so that it is sa t is factory to find tha t it is possible to explain the whole a t t ract ion by means of this kind of actio'n."

G. F. S.

The Charge on the Atomic Nucleus and the L a w of Force. J. CI~ADWlCK. (Phil. Mag., December, I92O.)--" The theory of the nuclear const i tut ion of the a tom, put forward by Sir Ernes t Ruth- erford in 191I , has been confirmed by evidence ga thered f rom such var ious sources tha t it now forms, the foundat ion on which the deve lopment of a tomic physics is based. On this theory the posit ive charge associated with an a tom is concent ra ted on a massive nucleus of small dimensions, sur rounded by a distr ibu- tion of electrons extending over a distance comparab le with the d iameter of the atom, as usual ly understood.

" The physical and chemical proper t ies of an e lement are de- t e rmined by the charge on the nucleus, for this fixes the number and a r r angemen t of the external electrons, on which these prop- erties mainly depend. The mass of the nucleus influences the a r r a n g e m e n t of the electrons only to a very small degree. The nuclear charge is thus the fundamenta l constant of the a tom and the question of its actual magni tude of grea t impor tance for the deve lopment of a tomic theory ."

As the consequences and implicat ions of Ruther ford ' s theory became unders tood a t t emp t s were made to connect the nuclear charge with other a tomic constants . " van den Broek suggested tha t the nuclear charge might be equal to the a tomic n u m b e r of the element, i.e., the num ber of the e lement when all the e lements are a r ranged in order of increasing a tomic weight ." The pahna ry significance of the a tomic num ber in this connection was b rought forward by the bril l iant young Brit ish physicist , Moseley, whose un t imely death at Gallipoli was the grea tes t single loss to science inflicted by German aggression. H e showed that the X- ray spec- tra of the e lements was dependent on the square of a number which became grea ter by one when passage was made f rom one e lement to the e lement of next higher a tomic weight . " On the nuclear theory of a tomic s t ruc ture this character is t ic n u m b e r mus t be closely connected with the charge on the nucleus, and Moseley concluded that the number gave in fundamenta l units the actual value of this charge. This is one of the most impor tan t genera l iza t ions in modern physics, and gives a s tar t ing-point for