APRIL 14, 1923] 




















ROF,. RAYMOND PEARL and his students con- 
tinue to make important contributions to the 
biology of man and other organisms on a statistical 
basis. In a recent paper (Pearl and Bacon, Johns 
_ Hopkins Hospital Reports, vol. xxi. Fasc. iii.) an 
analysis is made of the relation of the relative size of 
_ heart, liver, spleen, and kidneys to tuberculosis. The 
data were derived from 1341 autopsies in which there 
_ were tubercular lesions. Six indices for the relative 
weights of the above organs were used as the basis of 
statistical investigation in relation to age, sex, race, 
‘and cause of death. It is shown that the relative 
_weights of liver and heart, and heart and spleen change 
progressively during life; also that in cases of fatal 
tuberculosis the absolute weight of the heart is less 
and of the spleen greater than normal, probably 
because these changes are brought about by the 
disease. Curves of age show that when tuberculosis 
alone is fatal it kills at comparatively early ages. 
Many other interesting facts are brought out in this 
_ statistical study. 
In “‘ dry ” America, experiments with alcohol have 
a particular interest. Stockard has shown with 
‘guinea-pigs and Pearl with fowls that individuals 
which throughout life received daily doses of alcohol 
by inhalation are much longer-lived than their un- 
treated sibs. Ina recent note (Amer. Journ. Hygiene, 
vol. ii. No. 4) Prof. Pearl points out that the actuarial 
data of insurance companies, which are generally 
supposed to show that the consumption of alcohol in 
‘man in any quantity shortens life, are practically 
worthless. From 1569 family history records care- 
fully collected in the vicinity of Baltimore, he con- 
cludes that while heavy or steady drinking lowers the 
expectation of life, the moderate or occasional con- 
sumption of alcohol has no such effect on either sex. 
_ In experimental studies on the duration of life in 
Drosophila (Pearl and Parker, Amer. Naturalist, vols. 
55, 56) the authors compare the percentage of sur- 
_vivals at successive ages with the corresponding curve 
for man. A day in the life of a fruit-fly corresponds 
roughly with a year in the life of a man. Large 
numbers of flies of different stocks were bred under 
_ carefully standardised conditions. The length of the 
imaginal life was noted and the results compared with 
the statistics for man, beginning at the age of fifteen 
; Fundamentally similar curves are obtained in 
the two cases. In Drosophila it is shown that long- 
winged flies have two or three times as great an 
expectation of life at any age as short-winged flies, 
and that other hereditary differences in duration of 
life also occur. The death-rates generally increase 
steadily with advancing age. The mortality curve 
for Drosophila is then compared with that for modern 
man and for the population of the Roman provinces 
in Africa about the beginning of the Christian era 
from data of MacDonnell, Biometrika, 1913). The 
rosophila curve generally runs intermediate between 
these two. The modern curve of human mortality is 
' diverted from the normal by the prolongation of life 
of many of the less rugged by measures of public health 
and sanitation. 
By selection and inbreeding from Drosophila stocks 
it was possible to isolate strains showing large differ- 
ences both in mean duration of life and in the form of 
the mortality distributions, while in inbred lines the 
enetic differencesremained constant for ten to twenty- 
ive generations. It was shown that occasional 
etherisation of the flies has no Sppresee effect in 
lessening their duration of life. There is some evi- 
dence that in crosses, duration of life may segregate 
like a Mendelian character. A pedigree indicating 
something of the same kind in man is presented by 
No. 2789 VOL. 111] 
NATURE 
513 

Biometry and Genetics. 
Pearl (Amer. Journ. Hygiene, vol. ii. No. 3). In the 
father’s family only ro per cent. survived to the age of 
fifty years, in the mother’s family 75 per cent. reached 
that age, and in the offspring 87-5 per cent. 
Using too births/deaths as a “ vital index,’ Pearl 
and Burger (Proc. Nat. Acad. Sci. vol. 8, No. 4) plot 
the curve for this index for England and Wales in the 
years 1838-1920, from statistics in the Annual Reports 
of the Registrar-General. This ratio shows a slow 
but extremely steady increase until 1914, with only 
two slight fluctuations caused by influenza epidemics 
in 1847 and 1890. The birth-rate in the meantime 
showed a slow increase until about 1878, then a more 
rapid decrease until 1914, and a marked recovery since 
the war. Thus while in the year 1838-39 the number 
of births for each death was 1-4, in 1920 it was more 
than 2. The whole curve for the vital index shows a 
remarkably steady increment in the rate of population 
growth, with a high degree of regulation of death-rates 
to variations in the birth-rate. Measured by the 
criterion of the vital index, it is concluded that the 
population of England and Wales is “ biologically 
more vigorous ” than in 1838. But this merely means 
that its net rate of increase is greater, and takes no 
account of the differential character of the birth-rate. 
In another note in the same issue, Pearl considers the 
seasonal fluctuations in the vital index of the popula- 
tion, based on the same data, and finds that in each 
year it has its lowest value in the winter quarter 
(ending March 31), and its highest value in the summer 
quarter. In other words, in the winter months the 
birth incidence is relatively low and the death inci- 
dence relatively high, as might be expected. 
That density of population influences fecundity was 
formerly shown for fowls, and similar results have now 
been obtained for Drosophila (Pearl and Parker, Proc. 
Nat. Acad. Sci. vol. 8, No. 7). The rate of repro- 
duction of this fly is shown experimentally to vary 
inversely with the density of population. This is 
the converse of Farr’s law that the death-rate varies 
directly with density of population. It is suggested 
that the world-wide increase in density of popula- 
tion may account for the general decline in birth- 
rates which has taken place in the last forty years. 
The subject is one which deserves further investi- 
gation. 
A hexadactylous Norwegian family in which the 
postaxial digits (little fingers and toes) are double, is 
described by Aslaug Sverdrup (Journ. Genetics, vol. 
xii. No. 3). The condition is traced through six 
generations, and two types of polydactylism are 
recognised. In type A one finger, usually the fifth, 
is duplicated, while in type B the sixth finger is re- 
presented by asmallattached appendage. Both these 
types are already wellknown. The condition behaves 
in general as a dominant character, but in one line of 
the family, showing chiefly the A-type, there is an 
excess, and in another, showing only the B-type, a 
deficiency of polydactyls. Moreover, an A-type 
individual may have either A- or B-type offspring, 
whereas B-types cannot produce A-types. It is 
concluded that the B-type is probably determined by 
a single Mendelian factor with sometimes a failure of 
dominance, while the A-type is probably due to 
cumulative factors. The A-type of polydactyly is 
sometimes accompanied in this family by a form of 
brachydactyly due to shortening of certain meta- 
carpal bones, but also in some cases to short phalanges. 
Such papers on the inheritance of human abnormalities 
are important in their recognition of the necessity for 
accurate and detailed observations. 
In a study of the inheritance of patching in the 
flower of the sweet pea Prof. Punnett (Journ. Genetics, 
