W. B. Crow 
98 
to a large extent, relied upon as systematic distinctions, as was recog¬ 
nised by the earlier workers. The presence of pigments, pseudo¬ 
vacuoles and certain types of stratified membranes may be also of 
importance, when the life-history is fairly well known. It also appears 
that cytological characters such as the degree of differentiation of 
the protoplasm and particularly the distribution of the pigment are 
very significant, although such characters have hiterto received little 
recognition in schemes of classification. On the other hand, the 
character of the stratum or colony has only slight morphological 
significance, although this has been largely relied upon by svs- 
tematists. The distinctions between many species and even genera 
must be regarded as differences in life-history. So although some 
specific and generic forms are convertible one into the other, yet for 
any given organism, growing under normal conditions, there may 
be one predominant morphological form. Another organism may be 
capable of presenting precisely the same phases of life-history but 
the predominant phase may be different from that of the first organ¬ 
ism. I11 general, among the Chroococcacese we are dealing with 
quantitative rather than qualitative differences. The number of 
genetic factors producing qualitatively different effects in the 
organisms investigated need be relatively small. It is not possible, 
of course, to enumerate these factors, but it can be seen from the 
foregoing analyses of the morphology that they are such as affect 
rate of growth, secretion of various substances, and those regulating 
the surface of the cell or affecting the polarity of the protoplasm. 
These characters, especially the first two, are also likely to be in¬ 
fluenced by external factors. But that there are internal factors 
playing an important part in determining the structure of the organ¬ 
ism is obvious from the fact that two or more forms, growing under the 
same conditions, sometimes even in the same stratum, may remain 
quite distinct, even after growing side by side for many months. 
The group of the Chroococcacese shows such a number of features 
that may be regarded as primitive that it is unlikely that as a whole 
the family consists of reduced forms. It seems fairly clear that the 
family is practically homogeneous from the phylogenetic point of 
view. The differences rather than the resemblances of the various 
forms have been dealt with in the previous pages, but in most cases 
true transitions have been found between the extreme types. Among 
the most prominent indications of the primitive character of the 
group are the production of several pigments in the plasma, the 
variability of these pigments in amount and distribution, the un- 
