496 
L. DOXCASTEK. 
the hydrogen atom for chlorine and the substance CH3CI has 
different properties^ while those of CH3OH are different 
again. An exactly corresponding series, in composition and 
properties, is provided by the substances C2H5H, CgH^Cl, 
CgHjOH, and it would not be regarded as exaggeration to 
say that the change from a hydrocarbon to a chloride or an 
alcohol is produced by replacing the hydrogen atom by 
another atom or radicle. But the various atoms which may 
replace the hydrogen do not have exactly the same effects 
when attached to CoH^- as when attached to CH3- ; their 
effects are similar, but not identical. So when a chromosome 
of species A is replaced by one of species B, the effects of B 
in A cytoplasm cannot be expected to be identical with those 
produced in its natural environment. And when, as in 
Godlewski^s famous experiment, it was found that Ante don 
chromosomes appeared to have no effect in Echinus eggs, it 
may well be that the environment was too strange for any 
interaction to be possible, just as the replacement of hydrogen 
by chlorine in some compounds is easy, in others difficult or 
impossible. 
Further evidence with regard to the functions of the 
chromosomes in heredity will be given when dealing with sex- 
limited inheritance, after the general question of sex-deter- 
mination has been considered. 
(2) Sex, 
The evidence of connection between chromosomes and 
sex-determination is on the whole more complete than that 
Avhich connects chromosomes with Mendelian factors. Here 
again the bulk of the evidence is indirect rather than direct, 
but since there are only two sexes it is much easier to 
associate a particular chromosome with sex-determination 
than to show that any one is connected with the transmission 
of a particular Mendelian unit. The evidence that a particular 
chromosome in many cases invariably accompanies one sex is 
so clear that it may almost be regarded as direct evidence for 
sex-determination by that chromosome, and some of the facts 
