THE STRUCTURE OF PROTOPLASM 
285 
which began with the Mendelian revival nothing has been discovered which 
in any essential way invalidates the evidence from cytology that at least 
certain hereditary characters are in some way transmitted through the 
chromosomes. Indeed, the parallelism between the processes involved in 
chromosome reduction as described by cytologists and the theoretically 
postulated behavior of factors in segregation is regarded as one of the strong 
points in favor of the whole Mendelian theory. 
On the other hand, when we touch the question as to just how the chro- 
mosomes function in heredity, and more definitely how the qualities of the 
tissues and organs and of the organism as a whole which are borne by them 
come to expression in the morphogenesis both of the embryo and the adult, 
we find ourselves again at an impasse. It is easy to say that the serration 
of a leaf edge is an inherited character transmitted by the chromosomes of 
any parent plant which possesses it, and the statement harmonizes with 
practically all known data of cytology and experimental genetics, but to 
attempt to give this generalized formula of concepts concrete reality by 
telling how the serration is represented in the chromosome and how it 
comes to expression in the many-celled leaf is quite beyond us as yet. It is 
quite possible that we are astray in our conception of representation on the 
one hand and the characters of many-celled organs on the other. At any 
rate, the method of functioning of the chromosomes in heredity is a problem 
of the future, though we can have no doubt that they are just as definitely 
related to hereditary transmission, whatever that implies, as are the plastids 
to carbohydrate metabolism. 
From the standpoint of our question as to the structure of protoplasm, 
I think we may say that the chromosomes are each regions or portions of 
the protoplasm which by reason of the localization and specialization of 
certain functions and processes in them have come in some degree as has 
the cell itself to have a permanent unity and identity, and to arise only by 
division of parent chromosomes. The protoplasm is not an aggregate of 
such bodies, but its activities have been specialized and localized till such 
bodies as chromosomes have resulted. As a polyphase colloidal system it 
has furnished the internal condition for the development of the greater 
and greater difTerentiation, specificity, and fixity of its phases. 
The centrosome or central body with its obvious relation to cell dy- 
namics in the division both of nucleus and cytoplasm, and with its more 
recently discovered relation to cell movements as the blepharoplast, with 
its frequently, if not always, minute size and resemblance to other granules 
or groups of granules in the protoplasm, has been a main support for all 
theories of cell structure involving the idea of living granules of ultra- 
microscopic or at least sub-cellular size. Heidenhain makes it the type 
of one of his grades of life units. The evidence that the granules (centrioles) 
or groups of granules (microcentra) regularly arise by division of parent 
granules or groups in nuclear division in animals and lower plants is con- 
