680 TRANSACTIONS OF SECTION K. 
of the characteristic traits of any given individual. ‘The question is, What con- 
ception can we form as to their nature, and how are they able to produce the 
observed results? It is not necessary to discuss the evidence that the chromosomes, 
or the materials of which they are composed, play a most important part in con- 
nection with development. All the work of the last decades has tended tu 
emphasise their importance in the transmission of hereditary qualities, and this is 
equivalent to admitting that they contain factors that determine the path of 
development, and are responsible for the production, from the egg, of the form 
and structure of the adult. 
Now it is certain that it is not the chromosome-substance acting as a whole 
which is effective in those processes summed up in the term Ontogeny. It might 
be, and till recently was, thought that in those plants in which there is a marked 
alternation of generations a definite relation existed between the number of the 
chromosomes and the particular stage of the life-history. The double number was 
supposed to be essential for the sporophyte, whilst the halved number was similarly 
regarded as causally related with the appearance of the gametophyte or prothallial 
generation. 
But Loeb and others had already shown that the eggs of echinoderms might 
be stimulated to parthenogenetic development by means other than fertilisation, 
and Wilson found that such larvee only contained the half number of nuclear 
chromosomes, as, indeed, was only to be expected. But the idea of a close 
parallelism between chromosome number and the alternative phases of the life 
history was so deeply rooted that the full significance of Wilson’s discovery was 
not at cence grasped. The comparative neglect was, perhaps, partly justified, 
inasmuch as the larve could not be reared. It may, however, be incidentally 
remarked that no one, so far as I am aware, has yet succeeded in raising the 
normal echinoderm larva beyond the pluteus stage. 
The investigation of cases of apospory that occur in the pteridophytes have 
proved that no causal relation can exist between the number of the chromo- 
somes and the characters that distinguish the gametophyte and the sporophyte 
respectively. For the sporophyte may give rise to the gametophyte aposporously 
without any reduction, whilst the various types of apogamy with which we are 
now acquainted exhibit all gradations between a coalescence of more or less 
differentiated nuclei and the complete absence of all semblance of nuclear fusion. 
In the latter case, when the sporophyte springs from a gametophyte that has 
itself arisen after nuclear reduction, the sporophyte continues to retain the 
smaller number of chromosomes normally associated with the other generation 
only. 
We thus have a complete proof that a single sexual cell which has undergone 
reduction in the number of its chromosomes retains, in so far as its architectural 
contiguration is concerned, the capacity of giving rise to a plant possessed of the 
full complement of characters belonging to the species. But this, after all, is 
only what the facts of heredity might have led us to anticipate. For, whilst we 
are ignorant of the fundamental significance of the sexual fusion of the gametes, 
one of its most obvious veswts consists in the duplication of the primordia of the 
specific characters in the cells of the individual thus produced. This statement is 
not only in accord with results of experiments in breeding, but it is also in 
harmony with the essential features of the heterotype mitosis; and no other 
satisfactory interpretation of the latter series of phenomena has yet been found. 
Furthermore, the facts of Mendelian dominance clearly show that each parent, 
through the gametes to which it gives rise, contributes an independent organisation 
responsible for at least some of its own distinctive characters, as well as for those 
which distinguish the species. Consequently, when two gametes fuse, the embryo 
will be provided with a duplicate stock of agents or primordia which determine the 
appearance of its own specific and individual characters, These will not always 
be similar in the two parents, and when this is the case it often happens that the 
offspring resembles one parent only in respect of a particular feature. Neverthe- 
less the results of further breeding shows that the corresponding, but apparently 
lost, character only is latent, for it reappears in a proportion—and often a fixed 
ae 
