662 



NATURE 



[October 31, 1907 



primary beam, and a maximum in, or opposite to, that 

 direction. 



On the neutral pair hypothesis, if we only assume that 

 the chance of ejection from an atom is equal in all direc- 

 tions in the plane of rotation, it may as simplv be shown 

 that the directions of minimum and maximum intensity 

 are the same as on the previous hypothesis ; but whereas 

 on the ether pulse theory the intensity of the secondary 

 rays m the direction of propagation 'of the primary is 

 double that in a direction at right angles, on the " neutral 

 pair ' hypothesis it varies as the cosecant of the angle 

 which the direction of propagation of the secondary makes 

 with that of the primary, becoming infinite along the 

 direction of propagation of the primary. In other words, 

 the intensity of secondary radiation varies as the density 

 of the lines of longitude (or as the secant of the latitude) 

 on a sphere with the secondary radiating mass at its 

 centre, the direction of primary propagation being along 

 the axis. 



I have made experiments to test the two hypotheses, 

 using an electroscope to compare the intensities of 

 secondary radiation as nearly as possible in these two 

 directions. Taking into account the finite section of the 

 beams and consequent obliquity of the rays, the ratios on 

 the two hypotheses would be roughly i-q : i and 8 • i 

 assuming perfect scattering and neglecting the effect of 

 tertiary rays in the first case and assuming the pl.ane of 

 rotation to contain accurately the direction 'of propagation 

 in the second. If the assumption is only approximately 

 correct m either case, the ratio will be somewhat reduced. 

 It IS evident that great accuracy in the experiments was 

 not essential. They, however, leave no doubt as to the 

 conclusion, for the ratio of intensities was roughly i-b : i— 

 one that might be expected on the ether pulse theory and 

 appears impossible on the other. It is possible that 'with 

 suitable primary rays and thickness of secondary radiator 

 results showingmore perfect scattering will be obtained. ' 



Ihese preliminary experiments, however, to my mind 

 furnish quite conclusive evidence in favour of the ether 

 puse theory. Charles G. Barkla. 



University of Liverpool, October 26. 



On Correlation and the Methods of Modern Statistics. 



I DO not know that much profit is likely to arise from 

 continuing this discussion further; it appedrs to me to be 

 merely unwrapping considerable convolutions in Mr. 

 Hinks's mental attitude towards Miss Gibson and myself 



he chief charge made at the British .Association was that 

 we h,ad overlooked a curved regression line between magni- 

 tude and parallax— that now appears to have disappeared 

 into hmbo. In his first letter to Nature Mr. Hinks 

 apparently objected to our finding " a quite significant 

 and_ important " relation between parallax and proper 

 motion, but one not more than half-way up the correlation 

 scale. He has now discovered that "' the point of most 

 general interest " is that of colour. He charged us with 

 stating a far-re.^ching suggestion on the basis of the Cape 

 stars. It turns out now that the element in our far- 

 reaching suggestion is not the sug!;cstion at all but what 

 I am prepared to assert as a fact, namely, that the magni- 

 tude of the stars " is not mainly determined by parallax 

 or distance, but is more closely associated with colour, 

 and thus prob.ably with chemical or physical condition." 

 The colour and magnitude correlation is essenti.ally th.at 

 determined by Miss Gibson, 0-3 ; the values for the snec- 

 tral class and magnitude correlations run up according 

 to the classification used to double this value, and even 

 to 0-7. The colour and snectral class correlations reach, 

 as we might expect, a still higher value. Meanwhile, thn 

 magnitude and parallax relation in its best determination 

 IS 0.28. I agree with Mr. Hinks that this is a point of 



general interest." and I am glad that his last letter 

 enables me to assert it, not as " the vaguest of sugges- 

 tions, which words had reference to the discontinuity of 

 frrfa^iency in star counts, but as a fact which mav be 

 slightly modified when more data are reduced, but is 

 substantially correct as I have given it. 



Karl Pearson. 



NO. 1983, VOt. 76] 



The Interpretation of Mendelian Phenomena. 



I AM sorry Mr. Lock should mistake what I devoutly 

 hope is a sense of proportion for a desire to belittle 

 -Mendelian work. In science clear ideas are of importance, 

 and I wished to elicit something more definite than the 

 vague notion that Mendelism will someday and somehow 

 furnish a master key to the problems of heredity. I made 

 no complaint that Mendelism " does not immediately lead 

 to the solution of all the most difficult problems which 

 biology affords," as Mr. Lock rather extravagantly 

 asserts, but merely asked what conceivable bearing it can 

 have on any problem save that of sex. By the problem of 

 sex I mean the problem of the function of sex — or of con- 

 jugation if Mr. Lock prefers. I confess I cannot imagine 

 what light Mendelism has shed on the question of the 

 alleged transmission of acquirements, and as for the 

 " problems of the actual transmission of characters," these, 

 as dealt with by Mendelians, are nothing other than 

 problems of sex. That is, Mendelian experiments demon- 

 strate nothing more than the degree in w'hich certain 

 characters (mutations) are transmitted or distributed under, 

 or affected by. conditions of conjugation. Doubtless it is 

 true that the majority of Mendelian cases have been 

 observed in self-fertilised types, but I am not aware that 

 they have ever been observed unless cross-fertilisation had 

 previously occurred. In parthenogenesis the individual 

 arises from an unfertilised ovum ; how, then, is segre- 

 gation possible? What segregates? 



The evidence on which I base my assertion that there 

 is no segregation in the mulatto is that of my own eyes. 

 Mulattoes vary amongst themselves, but the blend is 

 usually very obvious, and is reproduced in subsequent 

 generations when breeding is inter se. With every infusion 

 of European blood the negro type — skin colour, hair 

 texture, shape of features, and the like — grows fainter, 

 until at length the " touch of the tar-brush " is hardly if 

 at all perceptible ; and this blending, so far as I am aware, 

 occurs, not only in all crossed human varieties, but in 

 other natural varieties as well. There may be exceptions ; 

 in fact, I believe there are ; but blending appears to be 

 the rule in the vast majority of instances. 



How can the fact that human races have crossed more 

 often than any other animal complicate the problem? My 

 statement implied, not that every human race is a chaotic 

 mixture of types, nor even that there are no pure types, 

 but only that we have here a very large and varied inass 

 of material on which to found our judgments. Nor did 

 I imply that mutations are especially frequent under con- 

 ditions of cultivation. I believe they are quite as common 

 in nature. Our hospitals and asylums are full of them — 

 hare-lips, cleft-palates, club-feet, haemophilia, colour-blind- 

 ness, deaf-mutism, feeble-mindedness, and so forth. Their 

 inheritance is usually Mendelian, but I never heard of a 

 human mutation that was useful. I implied merely that 

 artificial selection is founded on mutations, and that the 

 striking difference between artificial and natural varieties 

 indicates that natural selection is not founded on them. 

 We know the past and present of man better than that of 

 any other type, certainly of any natural type. Men are 

 fond of noting wonders, and we have a written history of 

 thousands of years ; but never yet has the differentiation 

 of a human variety by mutation been recorded. On the 

 other hand, so surely as a human race separates into 

 sections, between which there is little or no intercourse, 

 gradual differentiation sets in, which, under conditions of 

 savage warfare and very restricted intercourse, may be 

 seen in the inhabitants of quite small tracts of country, as 

 in New Guinea. Amongst plants and lower animals 

 parthenogenetic types are particularly rich in varieties. 

 '* Thousands of forms may be cultivated side by side in 

 the Botanical gardens and exhibit slight but undoubted 

 differentiating features, and reproduce themselves truly by 

 seed " (de Vries, " Species and Varieties," pp. 5Q-60). 

 When reproduction is bi-parental, varieties are few if in- 

 dividuals from distant parts of a wide area are able to 

 mate, and proportionately more numerous if intercourse 

 is more restricted. Thus in every valley of Samoa is found 

 a distinct variety of snails ; but species of birds, mammals, 

 and fishes which possess consider.able powers of locomotion 

 have few varieties. Is Mr. Lock able to conceive any 



