395 
ontogenic cycles. One cannot accurately speak of the ‘‘ growth”’ 
of a phylum, nor ought the word ‘‘ development’’ to be used for 
the phylum. Development should be restricted to the zoén or 
individual or its morphic equivalent among Protozoa, since it 
expresses more clearly the differences that exist between ontogeny 
and phylogeny than their similarities, and for the same reason it is 
advantageous to use evolution for the phylum alone in the sense in 
which it is commonly employed. ‘The necessity of subdividing the 
embryonic stage is admitted, and in all probability this really 
includes several stages with their own substages, but the discussion 
of this problem must be left to the future. 
The paragerontic stage is in no sense ‘atavistic’’ or ‘‘ rever- 
sionary,’’ as it is defined by Buckman and Bather. Reversions are 
the returns or recurrence of ancestral characteristics in genetically 
connected organisms which have been for a time latent in inter- 
mediate forms. Ido not think that we can include in this category 
purely morphic characteristics which habitually recur in the same 
individual as the result of paraplasis, or which occur in the paracme 
of a type more or less invariably. In the individual the resem- 
blance of the smooth round shell of the whorl of the paragerontic 
ammonoid after it has lost the progressive characteristic of the 
ephebic stage cannot be considered as a reversion. It is simply an- 
alogy of form, not structural similarity of characteristics. A better 
known and more easily understood case is the resemblance of the 
lower jaw of the infant before it has acquired teeth and that 
of the extremely old human subject in which these parts have been 
lost and the alveoli and upper parts of the bony mandible have dis- 
appeared through resorption. ‘The forms are alike, but no one 
would venture to consider the infant’s cartilaginous jaw and that of 
the old man as similar in structure. 
The best example of similar phenomena in the phylum known to 
me is the close resemblance of form between the straight Baculites 
of the Cretaceous or Jura and Orthoceras of the Paleozoic, which 
has been described above, and is figured further on. One occurs 
in the paracme and the other in the early epacme of the group of 
chambered shells. They are widely distinct in their structural 
characteristics, and these differences are greater in the young than 
at any subsequent stage of their ontogeny, Baculites having a close- 
coiled shell in the nepionic stage, and Orthoceras is straight from 
the earliest stage. The return of a similar form in Baculites in the\, 
