Information Content and Biotopology of the Cell in Terms of Cell Organelles 221 



The cytological and epigenetic criteria of this definition must uUimately 

 be supplemented by cytogenetic criteria, since the organelles are certain to 

 include such elements as some plasmagenes and all kinetosomes (20, 29), 

 and homeostats (30). The use of the term 'organelle' in this concept has 

 several distinct advantages. In the first place it is particularly nebulous and 

 difficult to define at its limit values, an asset because "the boundaries of all 

 natural units are hazy" (52); and it therefore does not lack this attribute of 

 realism. In this way it is less mystical and arbitrary than many of the 'vital 

 granules' of the past, and there is much room left for the contribution of the 

 configurations at the 'hazy lower-boundary' to cell form and function, by 

 the mechanisms discussed by Wassermann (31), Gross (32) and others. 



On the other hand it has sufficient traditional meaning (with or without 

 our four-part definition) for some cytologists to agree that all of the units 

 listed outside of brackets in Fig. 2 are 'organelles.' Our concept goes only 

 a step further in claiming epigenetic kinships. 



III. ORGANELLE DECISION-TREES 



Such an unorthodox but allowable interpretation of the organelle complexes 

 and systems of the unicellular animal Paramecium (shown in Fig. 1) leads to 

 the following postulates that should be useful in describing the information 

 content of other organisms: (a) The units of structural and functional integra- 

 tion in the cytoplasm are organelles ranging in size from less than 0.2 j.i to 1 //, or 

 are composed of such organelles as fusion structures, as complexes, and as systems 

 of complexes, (b) The organelles arise from primary organelles that in some 

 or all cases come from the nucleus as extruded nucleoli. 



According to these postulates, the primary organelle or young nucleolus 

 starts out with a finite set of possibilities and finishes with one having been 

 realized. We can call the point of choice a 'decision point', and discuss the 

 events involved in terms of binary decisions. Decision points are encountered 

 at which specialization is gained and differentiation-potential (uncertainty) is 

 lost. The critical decision points in Paramecium are to remain intranuclear oi 

 become extranuclear; to remain within the fluid phase (karyoplasm or endo- 

 plasm) or to occupy an interface; to remain solitary or to pack; to form a 

 coacervate or not. A provisional decision-tree for Paramecium, as well as for 

 other animal and plant cells, is presented in Fig. 2. The differentiation of 

 organelles in Paramecium insofar as it concerns the present classification can 

 be accomplished in a sequence of 2 to 5 binary decisions; the informational 

 performance accomplished in the sequence of decisions is not more than a few 

 bits. Presumably, only a small number of such decisions will be needed to 

 account for all existing differentiations in any organism. (While it is premature 

 to assign probabilities to each decision point*, the practical methodology 

 is nearly at hand: an animal has approximately 13,000 cilia, and about 

 as many trichocysts; perhaps twice this number of mitochondria; a macro- 

 nucleus at division is capable of numerically twice generating this entire 



* Pure chance, or a probability bias imposed by the micro-environment, may not be re- 

 garded as a sufficient cause for the choice of a particular decision in every case. We do not 

 suppose that each primary organelle contains all messages of the genome, and its specific 

 quality may therefore predetermine some choices and exclude others. 



