November 4, 1897] 



NATURE 



17 



utilise twenty races for obtaining the common racial mean. His 

 method of discussion is based on the safe ground that, if racial 

 variability be associated with individual varial)ility, whenever 

 any specified character is more variable than another specified 

 character in the one, it will (on the whole) be more variable 

 in the other also. Therefore if the reduced quartiles 

 of the two characters, in the series of means of races, be called 

 A and B. and those in that of the individuals of any given race 

 he called a and b, then if A be greater than a, B would (on the 

 whole) be greater than b, and conversely. More briefly and 

 fully, if the sipts of the differences (A- B) and {a-b) are alike, 

 the evidence, so far as it goes, favours the suggested idea of a 

 connection between racial and individual variability ; if the 

 ■■iirn>- are different it discountenances it. 



The test is rough, but is of value when applied on a con- 

 siderable scale with concurrent results, as in the present in- 

 stance. The eighteen characters admit of \ (i8 x 17) = 153 

 different pairs of combinations of the form (a-<J), in each of the 

 eight races ; that is 1224 pairs in all. Each of these has been 

 compared with its associated pair in the series of means of races, 

 of which there are altogether \ (8x7)= 28. The result is that 

 in seven out of the eight races the cases testifying to the exist- 

 ence of the suggested association are from twice to thrice as 

 numerous as the others, and in the eighth race they are i^ times 

 as numerous. Nay more, there is some evidence that the most 

 variable characters in the one, are the most variable in the other. 

 This conclusion is corroborated by three other inquiries of the 

 same kind, two into rodents and one into carnivorae. The error 

 introduced by the strained assumption, that the ordinary law of 

 frequency holds good for the series of means of allied races, 

 does not seem likely to invalidate the general conclusion to 

 a serious extent. It therefore appears that Mr. Brewster has 

 provisionally established his thesis that whenever any specified 

 character varies much in individuals of the same race, it is 

 probable that it will be found to vary much in " allied " races, 

 and conversely. 



ON THE CONSTITUTION OF THE ELECTRIC 

 SPARK} 



T F a Leyden jar is discharged through metal electrodes, and 

 the spectrum of the spark is examined, it is found that the 

 metallic lines are not confined to the immediate neighbourhood 

 of the poles, but are seen sometimes in the centre of the spark, 

 several millimetres away from the electrodes, from which they 

 must have been projected with considerable velocity. 



It has always seemed to me to be a problem of interest, to 

 measure the velocity of projection. A knowledge of it may teach 

 us something concerning the mechanism of electric sparks and, 

 in addition, we may hope to obtain information on some im- 

 portant points in spectrum analysis, which are at present under 

 discussion. Thus, for instance, if the speed with which a 

 molecule is pushed forward into the centre of the spark depends 

 on molecular weight, we may separate from each other those 

 lines of the spectrum which belong to different molecular com- 

 binations. For many years past I had made various unsuccessful 

 attempts to deal with this problem, when I became acquainted 

 with the elegant method, used by Prof. Dixon in some of his 

 recent experiments, in which a photograph is taken on a film 

 fixed to the rim of a rapidly revolving wheel, of which the speed 

 may easily be made sufficiently large to measure velocities of 



moving luminous particles up to 2000 -• This number 



might be doubled or trebled with improved appliances. 



The experiments were conducted by Mr. Gustav Hemsalech, 

 to whose care and skill their success is largely due. Without 

 entering into a detailed description of the apparatus, it will be 

 sufficient to say that the photographs, which I now submit to 

 the Section, were taken on a film m.oving with a linear speed of 



about 80 '"^^'^ -^ in a direction at right angles to the slit of the 



second 

 spectroscope. The lines of the metal appear inclined instead of 

 straight, in consequence of the finite velocity of the luminous 

 molecules. The air lines, on the other hand, though slightly 

 broadened, remain straight. The sparks were takeii from five 

 large Leyden jars, charged by means of a Voss machine. Each 



1 By Prof. Arthur Schuster, F.R.S. (Read before Section A of the British 

 Association at the 'loronto meeting ) 



single spark produces a good spectrum, reaching approximately 

 from A = 50CO to A = 4000. 



One of the photographs, in which zinc poles were used, shows 

 that the velocity of the molecules is gradually diminishing as 

 they move away from the pole. Close to it the speed seems 

 very great, the average velocity up to a distance of about one 

 millimetre being about 2000 metres per second. At a distance 

 of four millimetres the speed is reduced to something like 400. 



In another experiment one pole was zinc, while the other was 

 bismuth. Some bismuth lines are found to be decidedly more 

 curved than those of zinc, indicating a smaller velocity. But 

 the line of bismuth, which lies at 4560, seems almost straight. 



When the poles are moistened with a solution of calcium 

 chloride interesting results are obtained, the calcium line at 4226 

 being more inclined than H and K. 



The experiments were made with comparatively rough ap- 

 pliances, but a more perfect apparatus is in course of construction ; 

 and the author hopes to continue the research in conjunction 

 with Mr. Hemsalech. 



PHYSIOLOGY AT THE BRITISH 

 ASSOCIATION. 



'T'HE section of Physiology at the British Association meeting 

 at Toronto was a large and active one. Under the presi- 

 dency of Prof. Michael Foster a large body of physiologists 

 attended in about equal numbers from the east and west of the 

 Atlantic. Meetings were held on the Thursday and Friday, 

 August 19 and 20, and on the Monday, Tuesday and Wednesday, 

 August 23, 24 and 25. 



Among those present were Messrs. Bowditch, Boyce, Cushny, 

 Crookshank, Gaskell, Halliburton, Huber, Huerthle, Lee, Loeb, 

 Lombard, Billings, Lister, Macallum, Osier, Wesley Mills, Noel 

 Paton, Porter, Waymouth Reid, Sherrington, G. N. Stewart, 

 Anderson Stuart, Kellog, W. H. Thompson, Charles Richet, 

 Waller, Welby, Shore, MacAlister, O. Grunbaum, Baldwin, 

 Braun, Reynolds Green, Meldola, and A. S. Grunbaum. 



Prominent among proceedings in the Section were the follow- 

 ing : — 



Prof. Bowditch read a paper on the physiology of unstriped 

 muscular tissue as exemplified in the wall of the stomach of the 

 frog. The rhythmic contractility of the tissue is well seen. 

 When the organ or a strip of it is fitted to a recording apparatus, 

 in the majority of instances after a lapse of not more than three 

 hours, contractions of rhythmic recurrence are registered. 

 Often two or more sets of rhythmic contractions are superposed. 

 This may be best accounted for by supposing the muscle- 

 cells to contract not all together but in two or more groups. 

 In the discussion Prof. Sherrington referred to experiments 

 published by him in which had been recorded contractions of 

 the urinary bladder removed from freshly-killed monkeys and 

 placed in warm normal saline solution, while connected with a 

 volume recorder. In these the isolated bladder began to " beat " 

 almost at once, and continued " beating" for an hour or less, at 

 rate somewhat quicker than once a minute. The contractions 

 thus obtained from the viscus only exceptionally displayed the 

 compound character .shown in Prof. Bowditch's curves. 



Prof. Carl Huber brought forward observations on the cells of 

 the sympathetic system of vertebrates. In Amphibia nearly all 

 sympathetic cells are unipolar. In other vertebrates the pre- 

 vailing type is multipolar. All sympathetic neurons have one 

 axon only. The dendrites form a network between the cell- 

 bodies of the neurons constituting a ganglion. The axon of 

 each sympathetic nerve-cell becomes either a non-myelinated 

 nerve-fibre (grey fibre) or a fine myelinated nerve-fibre. The 

 fibres of the white rami are axons of cells lying within the spinal 

 cord, and these axons reach the sympathetic ganglia through the 

 white rami, and in the ganglia undergo branching, ending in 

 baskets which enclose the perikarya of the sympathetic neurons. 

 In mammalia and birds the circumcellular baskets are compara- 

 tively simple networks of varico.se fibrillse. In Reptilia and 

 Amphibia instead of simple end baskets the fibre is spirally wound 

 and completely contorted. This is the explanation of Beale's 

 spiral fibre in the sympathetic cells of the frog figured by him 

 forty years ago. In all vertebrates the pericellular baskets are 

 intracapsular. Langley and his pupils have shown that an 

 impulse travelling along a spino-sympathetic efferent chain 

 may be blocked in a sympathetic ganglion by the injection of 

 nicotin. This has been applied to practically all regions of the 



NO. 1462. VOL. ^']'\ 



