919 
of Edinburgh, Session 1885 - 86 . 
katabolic, Alternation of generations is indeed but a rhythm 
between a relatively anabolic and katabolic preponderance. The 
conception must be applied, however, not only to the general facts, 
but to specially difficult cases. Thus the peculiarity in the life- 
history of mosses, where the naturally vegetative or asexual genera- 
tion is not independent, but grows, as it were, parasitically upon 
the sexual, explains the comparative failure of that line of evolu- 
tion, and the more successful development of those other cryptogams 
in which the asexual generation, developing its predominantly 
anabolic tendency, finds sufficient foothold in the struggle for exist- 
ence. So phenomena, like apogamy and apospory in ferns, the 
shortening of the sexual generation in phanerogams, and the like, 
must be analysed. In short, all the modifications of form, floral and 
other, in one or in two generations, must be explained — not in terms 
of “spontaneous variation,” i.e., by unaccountable variations in 
each special case, with natural selection varying also with circum- 
stances for each special case, as has been usually believed and 
maintained — but in terms of the fundamental protoplasmic pro- 
cesses, by reference to the rhythm of anabolism and katabolism. 
§ 3. Nature of Sex. A. Inductive . — In what has been already 
noted in regard to incipient sex, it was the object to indicate how 
sexual reproduction was related to the asexual process, and how 
both were associated with a preponderance of katabolism. There 
remains, however, the further problem of the real nature of sex, or 
the rationale of sexual dimorphism. In attempting to define the 
distinctive characteristics of male and female, it is necessary to begin 
with the sexual elements themselves. The difference between male 
and female is there exhibited in its fundamental and most concen- 
trated expression. It is in the sexual elements, indeed, that the 
continuity of organic life is secured, the vegetative organisms being 
but appendages to the direct immortal chain of sex-cells. In com- 
paring ovum and spermatozoon it is necessary, in the first place, to 
refer briefly to the possible phases of cellular life. 
Starting from an undifferentiated and amoeboid cell, it is a 
simple fact of observation that the continuance of life in varying 
environment opens up the possibility of great variation in the 
protoplasmic metabolism, so that the algebraic sum of anabolism 
and katabolism must vary within the widest limits. Suppose, 
