490 
L. DONCASTER. 
leading to constancy of behaviour of the hypothetical units 
which ^^bear^^ the Mendelian factors, the whole hypothesis 
would fall to the ground. Others, while admitting the conju- 
gation of chromosomes in synapsis, maintain that it is not a 
mere coming together in pairs, followed by complete separation, 
but that the two chromosomes which pair fuse so intimately 
as to make separation of the parts almost or quite impossible 
(10, 10a). Others, again, deny the existence of a reduction 
division’^ in Weismann\s sense, and maintain that parts of 
both the paternal and maternal chromosomes go into all the 
gametes (32). If any of these conditions were general it 
would completely destroy the basis of the hypothesis under 
consideration. When, however, the various objections are 
critically examined, they begin to appear less fatal to the 
hypothesis. In the first place, the three classes of objection 
mentioned are mutually exclusive ; if the supposed facts on 
which any one of them is founded are genuine, the others 
must of necessity be mistaken. Secondly, they are largely 
founded on negative evidence, and if an observer, however 
competent, fails to find certain phenomena in the material on 
which he is working, this not only does not prove tliat the 
phenomena do not occur in other cases, but it may not even 
prove that they do not occur in his own material. The ease 
with which the alleged phenomena are observed varies greatly 
in different species, and they may quite possibly occur in 
cases in which the nature of the material makes proof or dis- 
proof of their occurrence impossible. In the opinion of the 
writer the wide-spi*ead agreement among workers on chromO’ 
somes in favour of the existence of some kind of individu- 
ality, and of a genuine reduction division, makes the first and 
third of the objections mentioned of sinall importance. 
With regard to the complete separation of chromosomes 
which have paired in synapsis there is much less unanimity, 
and the question must be regarded as still open.^ 
^ In this connection, the discussion by W. E. Agar, ‘ Quart. Journ. 
Micr. Sci,,’ 57, 1911, p. 1, based on peculiarly favourable material 
(Lepidosiren), is of importance. 
