ZOOLOGY AND BOTANY, MICROSCOPY, ETC. 
273 
most important from the latter point of view are those on the chances 
of death, reproductive selection, and variation in man and woman. 
In the first the mortality frequency for age curve is given. It is 
analysed into five components of chance distributions of mortality 
centering round five distinct ages of life — old age, middle age, youth, 
childhood, and infancy — the latter extending backwards to some nine 
months before birth. The skewness of distribution varies for each of 
these ages. The several mortalities are — 481, 173, 51, 41, and 24 per 
1000 respectively. 
Keproductive selection is the increase in fertility that must be the 
necessary consequence of the inheritability of fertility, so long as the 
increasing fertility is not correlated with any detrimental character that 
would be held in check by natural selection. On the other hand, if there 
be any neutral characters correlated with fertility, then the inevitable 
cumulation of fertility results in an appearance of a definitely progres- 
sive variation of such neutral characters. Thus a possible explanation 
is afforded of the definite variations of many authors. For instance, 
stature seems to vary with fertility in woman ; this would tend to raise 
the height of the race about 3 inches in 1000 years. But, as a matter of 
fact, the death-rate of large families is higher than that of small families. 
Thus natural selection checks reproductive selection in man. There is 
a differential fertility in the classes of human society, but natural selec- 
tion, or at any rate a differential death-rate for the classes, checks here 
also the effects of a reproductive selection. Indeed, it tends to reverse 
those effects. On the other hand, there is a differential marriage rate for 
the classes. The total effect of these three interacting causes is that, on 
the whole, society is recruited rather from the artisan class than from 
the commercial or professional classes. 
The current doctrine that the male is more variable than the female 
is a pseudo-scientific superstition. The statistics by which the hypo- 
thesis has been supported have been mainly drawn from data relative to 
pathological conditions, some even from the investigation of characters 
that are in reality secondary sex characters. The data are always 
inadequate, if not inadmissible. Skull measurements give reliable data. 
They may be collected at random, so as to yield fair samples of the 
population. They are related to the brain, which is so significant in 
human evolution. Absolute relative variability has often been taken as 
the criterion, rather than the comparison of the coefficients of variation. 
The examination of 17 groups of measurements of different parts of the 
body shows that in 11 groups the female is more variable than the male, 
and in six the male more than the female. The differences of variability 
are slight ; less than that between members of the samo race living in 
different conditions. Such as it is, it is probably due to a difference in 
the severity of the struggle for existence. 
Ontogeny and Phylogeny.* — Prof. A. Hyatt entitles his paper 
4 Cycle in the life of the individual (Ontogeny), and in the evolution 
of its own group (Phylogeny).’ In the introductory portion he has an 
interesting historical note on Oken’s (1805) prevision of the cell-theory, 
and on Meckel’s (1811) early statement of recapitulation doctrine. He 
* Proc. Amer. Acad., xxxii. (1897) pp. 209-24. 
