Miscellanea 
419 
original figures seem to have been the crude rate for males and females separately from 1835 
onwards." Why the writer of these words should have assumed them without any inquiry 
of me, or any examination of the values of the crude death-rates (which are accessible to every- 
body) to be "crude death-rates," I do not know, but they illustrate his readiness to form a 
biased judgment when his feelings are stirred by unfavoui'able criticism. As a matter of fact 
the rates were standardised rates reduced to the population of 1901, and most kindly 
provided at my special request by the General Register Office. It is of interest to oljserve that 
Dr Weinberg of Stuttgart — recently made precisely the same charge as Mr Major Greenwood 
with tlie same over-hasty assumption that the reality must be the desired, if undemonstrated, 
•error*. With the German as with other foes, it is well to leave ample 02)portunity for their 
assuming you to be foolish ; their assumption may lead them to riui against hard reality. 
K. P. 
IV. Note on Reproductive Selection. 
By DAVID HERON, D.Sc. 
The fact that in the case of mant fifty per cent, of one generation comes from twenty-five 
per cent, of the preceding one was first noted by Professor Karl Pearson in the Chances of 
Death (Vol. I. p. 80) and in dealing more fully with this important generalisation in the Ground- 
vjork of Eugenics, p. 27, he said : " It is very difficult from any English statistics to determine 
how many adults never marry. No information on this point is asked in the death schedule for 
males ; it is asked but imperfectly answered in the case of the schedule for females." In 
a footnote he adds : " The Registrar-General informs me that the record of civil condition 
in the case of female deaths is worthless and that no useful return can be made from it." 
He found that in the Argentine and in Scotland 60 per cent, died unmarried, in the United 
States 51 per cent., and from the last two English Censuses and the Annual Reports 48 per 
cent., and added " This indirect method of reaching the result is, however, not very satisfactory. 
We may, I think, conclude in round numbers that 40 per cent, of the population dies before it 
reaches the age of 21 and that probably another 20 per cent, are never married." On this 
assumption Professor Pearson proceeds to show that " about 12 per cent, of all the indi\'iduals 
born in the last generation provide half the next generation." 
Some data published in Bulletin of Population and Vital Statistics No. 30 for the Cominon- 
wealth of Australia (Tables 48 and 84 a and h) prove that the assumptions made lie very close to 
the facts. The data are shown in the following table which gives the conjugal condition and 
issue of the males and females who died in Australia in 1912. From this we find that half the 
total number of children came from 3337 of the parents (all those who had at least 9 children 
and part of those who had each 8 children). It thus appears that of the males 17,404 out 
of 30,285 = 57'5 died unmarried while half the total offspring came from 25-9 of those who 
married and ll'0°/„ of the whole number of males, so that approximately three-fifths of the 
males born die unmarried and one-half of one generation comes from one-quarter of the married 
population or from one-ninth of all the males born in the preceding generation. The diagram 
gives a graphical illustration of the argument. 
In exactly the same way we find that nearly one-half of the females born in Australia die 
unmarried and that one-half of one generation comes from one-quarter of the married and from 
one-seventh of all the females born in the preceding generation. 
* Archiv fiir Rassen- und Gesellschufts-Biologie, ix. Jahrgang, S. 87. Leipzig, 1912. 
t It has also been dealt with in various mammals. See the Groundwork of Kuyciiic^, Eugenics 
Lecture Series ii (Dulau and Co.), p. 29. 
