282 Raphael Falk
too simply mechanicaland themetaphysical conceptionrepresent the
Scylla and Charibdis, to steer one’s course between which is indeed
a difficult task, a task which few have hitherto accomplished. [Roux,
1904/1986, 130]
Facing such an ordeal, it may be no wonder that, to the extent that geneticists
were concerned with problems of embryology and development, they preferred the
search for phenomenologicalmodi operandi of unit traits over that of those of
the physico-chemical causes of organismic systems. This shift in perspective is
best seen in the “metamorphosis” of Thomas Hunt Morgan, an embryologist who
consistently attempted to maintain the center of interest of genetic analysis on
problems of development [Allen, 1978]. Morgan overcame his reservations from de
Vries’ preformationism, once he accepted Johannsen’s differentiation between the
“transmission conception” of personal qualities and the “genotype conception” of
discrete particles that are “‘bearers’ of special organizing functions in the mecha-
nism of ontogenesis” [Johannsen, 1911, 132]. Yet, he tried desperately to maintain
a developmental-system approach of the role of genes and their mutations in devel-
opment of the Drosophila fly, as evident from the nomenclature of mutants that he
introduced: a gene was called by its (presumed) function. But this nomenclature
soon became so complicated, and had to be changed with every new mutant, that
he reluctantly gave in and adopted a nomenclature that pragmatically expressed
instead merely the phenotypic deviation of the mutant from the “wild type”, as
a designator or marker of that gene [Falk and Schwartz, 1993]. This change of
nomenclature inevitably led to a change of emphasisaway from the conceptual,
functional integrative aspects of the development of traits, and, methodologically,
turned attention to the specific phenomenological function of genes.
Johanssen too was quite alarmed by the reductionist implications of the intro-
duction of the notion of the gene into the theory of heredity:
My term “gene” was introduced and generally accepted as a short and
unprejudiced word for unit-factors... but originally I was somewhat
possessed with the antiquated morphological spirit in Galton’s, Weis-
mann’s and Mendel’s viewpoints. From a physiological or chemico-
biological standpoint... there are no unit characters at all!...
We may in some way “dissect” the organism descriptively, using all the
tricks of terminology as we please. But that is not allowed in Genetical
explanation. Here, in the present state of research, we have especially
to do with such genotypical units as are separable,...Certainly by far
the most comprehensive and most decisive part of the whole genotype
does not seem to be able to segregate in units; and as yet we are mostly
operating with “characters,” which are rather superficial in compari-
son with the fundamental Specific or Generic nature of the organism.
[Johannsen, 1923, 136–137]
This methodology of viewing the effects of individual genes, relegating interac-
tions to a secondary role, became to an important extent, the conceptual stereotype