October 22, 1903] 



NATURE 



607 



of a complete record, but the set of fourteen drawings of 

 the polar cap which accompany the paper show very clearly 

 the diminution of the cap from October 15, 1902, to March 

 15, 1903, and its augmentation from then untij July i, the 

 minimum apparently taking place at an earlier date than 

 usual. 



In addition to detailed descriptions of the most interest- 

 ing observations, the paper contains reproductions of ten 

 excellent drawings showing various features on the planet's 

 surface. 



Natal Government Observatory. — The report of the 

 Government Astronomer for Natal, Mr. E. Nevill, for 1902 

 is chiefly devoted to the various meteorological records of 

 the colony, and forms a valuable addition to the meteorology 

 of last year. 



After giving brief descriptions of the staff, the instru- 

 ments, the management of the time signals, the magnetic 

 observations, and the tide records, the report gives a number 

 of tables containing very complete records of the meteor- 

 ological results obtained at the Durban Observatory and 

 twenty-two inland stations, and the less complete records 

 of twenty-six subsidiary stations which are scattered 

 throughout the colony. 



In dealing with this section of the report Mr. Nevill 

 directs special attention to the importance of obtaining the 

 fullest possible records of the meteorological conditions in 

 Natal, because, in addition to their local importance, it has 

 been shown that there is a very close connection between 

 them and the conditions obtaining in Australia and India. 

 In the latter case there are trustworthy indications that the 

 meteorological conditions of Natal are those which are 

 likely to prevail in India during the following season ; this 

 is especially marked in the case of the rainfall. 



IXHERITANCE OF PSYCHICAL AND 

 PHYSICAL CHARACTERS IN MAN.' 

 'T'HERE are probably few persons who would now deny 

 ■*• the immense importance of ancestry in the case of 

 any domestic animal. A majority of the community would 

 probably admit also that the physical characters in man are 

 inherited with practically the same intensity as the like 

 characters in cattle and horses. 



But the preeminence of man in the animal kingdom 

 is justly attributed, not to his physical, but to his psychical 

 character. The latter is seen developing apparently under 

 the influences of home and of school, and we conclude, 

 perhaps too rashly, that home and school are the chief 

 sources of the psychical qualities. We are too apt to over- 

 look the possibility that the home standard is itself a pro- 

 duct of stock, and that the relative gain from education 

 depends in a surprising degree on the raw material pre- 

 sented to the educator. 



It is possible to hold this view and yet believe that 

 moral and mental characters are inherited in either a 

 qualitatively or a quantitatively different manner from the 

 physical characters. Both may be influenced by environ- 

 ment, but one in a far more marked way than the other. 



.Some six or seven years ago, then, I set myself the 

 following problem : What is the quantitative measure of 

 the inheritance of the moral and mental characters in man, 

 and how is it related to the corresponding measure of the 

 inheritance of the physical characters? 



The problem really resolved itself into three separate 

 investigations : — 



(a) .\ sufficiently wide inquiry into the actual values of 

 inheritance of the physical characters in man. 



For this investigation upwards of 1000 families were 

 measured, giving ample means of determining the quanti- 

 tative measure of resemblance for both parental and 

 fraternal relationships. 



(b) A comparison of the inheritance of the physical 

 characters in man with those in other forms of life. 



No substantial difference in this inheritance has been dis- 

 covered. 



(c) An inquiry into the inheritance of moral and mental 

 characters in man. 



Owing to the great difficulty of comparing the moral 



1 Abstract of the Huxley Ateniorial Lecture for 1903. Delivered before 

 the Anthropological Institute on October 16, byProK Karl Pearson, F.R.S. 



NO. 1773, VOL. 68] 



characters of a child with those of its adult parents, I con- 

 fined my attention to fraternal resemblance, for if fraternal 

 resemblance for moral and mental characters is less than, 

 equal to, or greater than its value for physical characters, 

 the same must be true for parental inheritance. 



In the next place it seemed impossible to obtain moder- 

 ately impartial estimates of the psychical characters of 

 adults. The inquiry, therefore, was limited to children, so 

 that the partial parent or relative could be replaced by the 

 fairly impartial school teacher. 



After much consideration and some experimenting, 

 schedules were prepared in which teachers could briefly 

 note the chief characteristics of the children under their 

 charge. These schedules were white for a pair of brothers, 

 pink for a pair of sisters, and blue for a brother and sister. 

 With the schedules specially devised headspanners were 

 distributed, directions for the use of the headspanner, and 

 general directions as to the estimation of the physical and 

 mental characters. 



The material took upwards of five years to collect. 

 Appeal was inade through the columns of the educational 

 journals to teachers of all kinds, and the observations were 

 made not only in the great boys' public schools and the 

 grammar schools of the country, but in modern mixed 

 schools, in national and elementary schools of all kinds, in 

 board schools, and private schools throughout the kingdom. 

 Some 6000 schedules were distributed, and between 3000 

 and 4000 returned with more or less ample data. I have 

 most heartily to thank the masters and mistresses of some 

 200 schools in which observations have been made for me. 

 In the midst of arduous professional claims on their time and 

 energy, they have, in many cases at considerable personal 

 inconvenience, recorded and measured the children in their 

 charge for a purpose only dimly foreshadowed for them. 



Much of what I have to say upon the nature of the theory 

 applied will not be new to those who have examined recent 

 biometric work, and some of it will not be intelligible 

 except to the trained mathematician. Still we must strive 

 i.i broad lines to see how the work has been done, and, 

 above all, to justify our treatment of the psychical character. 



[To illustrate the method the lecturer examined the degree 

 of resemblance between the cephalic indices of brothers, 

 the cephalic index of a person being 100 X the ratio of 

 breadth to length of head. This scarcely changes with 

 growth after the first two years of life. A table was ex- 

 hibited showing the cephalic index for 1982 pairs of 

 brothers.] 



Taking the boys, for example, with cephalic indices 

 between 74 and' 75, these boys had seventy-eight 

 brothers who were distributed according to the column 

 headed 74 to 75. Brothers are not alike in cephalic index, 

 but distributed with a considerable range of variation. 

 The arithmetic mean of the cephalic indices of this array 

 of brothers is 77.45. Thus the average brother of a boy 

 with cephalic index' 74.5 has a cephalic index =7745. This 

 is the phenomenon of regression towards the general popu- 

 lation mean (78-9) discovered by Francis Galton. 



We now find by taking all the arrays that whatever the 

 cephalic index of first brother be, cephalic index of mean 

 second brother 



= (l -a){mean cephalic index of whole population} 

 + {cephalic index of first brother} 

 and that in the case of cephalic indices for two brothers 

 the quantity o, defined as the " resemblance," has the 

 value 05. 



Now from this result we have learnt two great features 

 about inheritance in man. Firstly, that part of the cephalic 

 index of the second brother depends in the above linear 

 manner on that of the mean of the whole population and 

 part on that of the first brother ; and, secondly, that these 

 parts are about equal. Are these true for other characters 

 than the cephalic index? Undoubtedly, for all physical 

 characters. And further, the fraction a, which we have 

 called the resemblance, is, for brethren, in all cases 

 about 05. 



This surprising uniformity in the inheritance of the 

 measurable physical characters can be extended to physical 

 characters not capable of accurate measurement, and to 

 psychical characters provided we assume a certain distri- 

 bution of frequency for such characters in human popu- 



