but large parts of it can be read without going deeply into the 

 mathematical procedures used. The author very usefully suggests 

 several shortened "reading routes" for various categories of 

 readers . 



The theory is based in large part on a generalization of 

 Sperry's "chemical neurospecif icity" hypothesis, in conjunction 

 with Weiss' "molecular ecology" concept. The underlying tenet 

 is that each individual somatic cell in a metazoan organism 

 (with certain exceptions) is chemically uniquely specified, and 

 that this specification is under genetic control from the begin- 

 ning of development in a completely specific, non-random manner. 

 The specification is postulated to be due to "cell-unique pro- 

 teins", and is called "microdif ferentiation" (this concept ex- 

 cludes all types of morphological and chemical differentiation 

 hitherto distinguished, which are all brought under the heading 

 of "gross differentiation"). On the basis of a sophisticated 

 model for the structure of eukaryotic chromosomes and for the 

 transcription process the author argues that the organism con- 

 tains more than enough DNA to specify 10 chemically distinct 

 cells (on the auxiliary hypothesis that the one-cistron - one- 

 polypeptide rule does not hold for cell-unique proteins). 



On this foundation a huge theoretical superstructure is then 

 erected, which makes use of a multitude of newly coined terms 

 and concepts, and encompasses no less than 36 "fundamental hy- 

 potheses". A central part of the theory is a basically simple, 

 but very elaborate "process algorithm" for the explanation of 

 microdifferentiation. This is based on the notion of a "devel- 

 opmental tree" (with numerous "sub-trees"), a cell genealogy 

 defined by the use of enumeration techniques. 



Among other things, the theory envisages asymmetrical mitoses 

 which ensure that the derepressed state of a particular gene is 

 passed on to one of two daughter cells and its descendents. At 

 certain times hypothetical "inductive stimulus substances" of 

 relatively low specificity, produced by cells other than the 

 generator cell of a sub-tree, trigger off further development 

 of the sub-tree in question. The cytoplasm is assumed to play 

 no role whatsoever in the selective derepression of the genes 

 concerned in microdifferentiation (in contrast to those in- 

 volved in gross differentiation, but here the controlling cy- 

 toplasmic constituents would again be cell-unique proteins). 



The cell-unique proteins are postulated to serve as compo- 

 nents of two basic types of structure: "protein chains" and 

 "molecular maps". The former are predominantly intracellular 

 and, among other things, can specify intracellular axes (e.g. 

 the polar axis of the egg). The latter are assumed to consti- 

 tute the bounding layers of triple membranes, located both in- 

 ternally and at the cell periphery; in the latter position they 

 are responsible (in conjunction with "protuding protein chains" 

 attached to them) for cell recognition, cell matching, cell 

 contacts, and even morphogenetic movements (for which a number 

 of auxiliary hypotheses are introduced). All these processes 

 are thus ultimately under strict genetic control. 



Apart from this rigid, genome-centered control system, which 

 is basically a switching system, the theory also embodies con- 

 tinuous variables which are required to explain the flexibili- 

 ty of development (e.g. regulation, regeneration). Among these 

 are the concentrations of cell-unique proteins, and the concen- 

 trations of "stabilizer substances"; the latter are less spe- 

 cific diffusible factors which, if present in amounts exceeding 

 critical thresholds, can stabilize the derepressed states of 

 genes. The author disagrees with those who invoke gradients 



186 



