Recent Advances in the Study of Heredity. 327 
of attempts to deal with the problem on these lines. The work 
based on this conception achieved apparently conclusive and 
immediate results, because certain elementary conditions necessary 
to experiments of this kind were disregarded; and it gave distinct 
colour to the generalization, advocated in Geddes and Thomson’s 
“ Evolution of Sex,” that the production of males is determined by 
general katabolic conditions and that of females by anabolic ones. 
Yung, for instance, found that favourable conditions of nutrition 
resulted in the production of a high percentage of males in a batch 
of tadpoles; and that opposite conditions resulted in a corres¬ 
pondingly high proportion of females. In experiments of this kind it is 
essential that a record of the number of individuals at the beginning 
and at the end of the experiment is kept, because, if this is not done, 
there is no means of knowingWhether the high final percentage of 
one sex is due to a particular effect produced on the sexless embryo 
in all cases , or whether it may not be due to a differential death- 
rate resulting from the abnormal experimental conditions. No 
experiment of this kind is of the slightest value unless a record of 
the total number of individuals at the beginning and at the end of 
the experiment is kept. 
The cause of the existence of this theory of the determination 
of sex is to be sought in a too cautious inference from the facts of 
development. The sex of an animal or plant is not discernible in 
the fertilized egg-cell which gives rise to it; nor are the structural 
features which distinguish the sexes manifested until fairly late in 
development. And any inference, from these facts, that the sex of 
the adult was already determined in the fertilized egg-cell which 
produced it, ran an obvious risk of being objected to as being too 
far in advance, or ahead, of the facts. This objection is in our 
opinion a legitimate one ; and the way out of the difficulty is that 
the facts of the development of primary and secondary sexual 
characters do not constitute the basis from which we should start 
in elaborating a theory of the determination of sex. I say this is a 
way out of the difficulty. But I am by no means convinced that 
these facts ought to be neglected in the consideration of this 
question. 
The theory that the sex of an organism is already determined 
in the fertilized egg-cell from which it develops has become widely 
adopted of recent years because the suggestion that sex is inherited 
in Mendelian fashion has received very general credence. 
This suggestion was bound to be made sooner or later. The 
