APRIL 26, 1912] 
Therefore it would be proper to designate the 
statistically prominent type as an apparent type 
(Erschemungstypus), or, more briefly and di- 
rectly, a phenotype (Phaenotypus). Such pheno- 
types are in themselves measurable realities; some- 
thing that can be observed as typical; that is, the 
centers among series of variations, around which 
the variants are grouped. The word phenotype 
serves only to make the necessary mental reserva- 
tion that from the appearance alone no further 
conclusion can be drawn. A given phenotype may 
be an expression of biological unity (Ausdruck 
einer biologischen Einheit), but it does not at all 
need to be. Indeed, this is not true, in a great 
majority of cases, of the phenotypes found in 
nature by statistical investigations of variations.t 
Of course it would be presumptuous to as- 
sume that any translation would convey the 
exact meaning of such a passage, but at least 
it can be seen that phenotype was being used 
by Johannsen as a concrete collective term, 
and not merely as an abstract conception, as 
Shull has supposed: 
‘*Phenotype’’ and ‘‘genotype,’’ when both are 
tightly used, are contrasted terms, both being 
abstractions referring to the type to which an 
individual or group of individuals belongs, and 
not to the group of individuals belonging to that 
type. To illustrate the use of ‘‘phenotype’’ in 
its correct sense, reference may be made to the 
¥, of a Mendelian hybrid.? 
When the phenotype idea was brought later 
on into direct contrast with the genotype idea, 
the two were compared as abstractions, but 
this conceptual refinement was for purposes of 
explanation and did not necessarily supplant 
the more concrete application of phenotype 
previously made. Shull need not apologize 
for himself or for Jennings on account of 
having used phenotype in a concrete sense. 
It may be that the first use of the term, as re- 
stricted to the statistically prominent center 
of the group, was too narrow for convenience, 
but any group that has been found to show a 
statistical unity could be described at least as 
phenotypic. 
1 Johannsen, W., ‘‘Hlemente der Exacten Er- 
blickertslehre,’’ p. 123. 
?Shull, G. H., ‘‘ ‘Phenotype’ and ‘Clone,’ ’’ 
Science, N. S., Vol. XXXV., February 2, 1912, 
p. 182. 
SCIENCE 
655 
That Johannsen did not contemplate the 
employment of “genotype” in any such. con- 
crete sense as phenotype seems plain from the 
statement that accompanies his definition: 
Very obvious phenotypical differences may be 
shown where no genotypical difference is present; 
and there are also cases where with genotypical 
diversity the phenotypes are equal. Just for this 
reason it is of the greatest importance to separate 
clearly the conception phenotype or apparent type 
(Erscheinungstypus) from the conception geno- 
type or germ-type (Anlagetypus), as one might 
say. With this latter conception, to be sure, we 
shall not be able to work (nicht operiren kénnen) 
—a genotype does not make its appearance in pure 
form (tritt eben nich rein in die Erscheinung) ; 
but the derived concept of genotypical difference 
will be of use in manifold ways. 
Phenotypes, as we learned from the previous 
quotation, are found in nature, but genotypes 
are not. To the unregenerate reader Johann- 
sen’s genotype appears to be nothing more 
than an unframed conception of a germinal 
or genetic constitution, considered as some- 
thing apart from the external manifestation 
of the characters. It is an indirect and com- 
plicated substitute for the old distinction be- 
tween latent and patent characters, between 
transmission and expression. 
To replace the word genotype because it was 
preoccupied in taxonomic biology may not 
seem so necessary if it be considered merely 
as the name of an abstract. conception with no 
real existence that needs to be discussed in 
biological literature. But that geneticists 
should wish to keep the word in active use as 
a major term after its previous history has 
been pointed out is only one more way of 
showing disregard for the taxonomic frame- 
work of biology. 
The adjective use of genotype is hardly 
more fortunate than the substantive applica- 
tion. Why we should say genotypical differ- 
ences instead of genetic differences or ger- 
minal differences is not obvious, but perhaps 
the longer word means more to geneticists. 
Tf the object was to keep closer to the idea of 
a germinal constitution made up of separate 
units or gens, the meaning could have been 
