656 
conveyed more effectively by speaking of 
geme differences than by adding extra syl- 
lables. The “type” part of Johannsen’s 
words has served only to confuse the issues, 
as in the passage where Shull says that pheno- 
types and genotypes are abstractions relating 
to types but not to groups. How can there be 
typical differences, in any biological sense, 
unless groups are compared? The fact seems 
to be that Johannsen was not using the word 
type in accord with biological traditions, but 
in a loose metaphysical way that renders the 
terms more abstract instead of more concrete. 
There should have been no difficulty in 
finding suitables names for the two classes of 
Mendelian hybrids that Shull has pointed out, 
instead of allowing them to become confused 
with Johannsen’s genotypes and phenotypes. 
As the so-called genotypes are supposed to 
have the same gens, they could be described as 
isogenic. hybrids or isogens. Any group 
treated as having biological unity may be 
called an isogen. Johannsen approached the 
idea of biological unity in the passage ex- 
plaining the use of phenotype, but did not 
provide a name for such groups except indi- 
rectly through the genotype concept. 
The hybrids that have different germinal 
constitutions, and yet look alike, could be de- 
scribed as isophanic hybrids or isophans. 
They have the same dominant characters, but 
this does not involve any complete statistical 
or phenotypic unity. The groups are formed 
with reference to alternative, Mendelian 
characters, instead of on the basis of statis- 
tical measurements of continuous variations. 
As Johannsen pointed out, even genotypical 
unity does not preclude phenotypical differ- 
ences. 
Pluralizing the word gen is another difficulty 
encountered by geneticists. Johannsen used 
the term mostly in its German plural form, 
Gene. Our writers have added another letter 
making a double plural, “genes,” something 
like “ memorandas.” 
Johannsen proposed gen as a simplification 
of Darwin’s term pangen, to avoid the impli- 
cations of Darwin’s theory of pangenesis: 
Instead therefore of pangen (das Pangen) and 
SCIENCE 
[N.S. Vou. XXXV. No. 904 
pangens (die Pangene), we shall simply say gen 
(das Gen) and gens (die Gene). 
Along with this word gen, to represent an 
invisible rudiment or transmitted germ of a 
character, it will be useful to have a corre- 
sponding term, phan, to represent an external 
manifestation or expression of a character. 
To be able to refer to the external expression 
or phanic relations of characters is quite as 
important as to discuss them from the stand- 
point of theories of transmission. From 
these two roots it will be easy to develop a 
simple and appropriate terminology for many 
of the facts of heredity. 
O. F. Cook 
WASHINGTON, D. C., 
February 24, 1912 
CROSS CUTTING AND RETROGRADING OF STREAM- 
BEDS 
In the October (1911) number of the Amer- 
ican Journal of Science, I read with interest 
an article by Mr. John Lyon Rich on “ Re- 
cent Stream Trenching in the Semi-arid 
Portion of Southwestern New Mexico, a Re- 
sult of Removal of Vegetation Cover,” on 
which I have ever since intended making 
brief comment, because it seemed to me Mr. 
Rich presented only one phase of the subject. 
While the stated factor, “removal of vegeta- 
tion cover,” may in some localities, accelerate 
the retrograding (trenching) of stream-beds, 
it is not, in my opinion, the cause of retro- 
grading. I noted the same characteristics 
(and others probably also noted) years ago in 
places where there were no cattle and never 
had been any. 
The “trenching,” Mr. Rich says, “is still 
in progress,” which is true, for it has always 
been and always will be, in progress, cattle or 
no cattle, vegetation or no vegetation, not 
only in semi-arid regions but everywhere. 
There are differences in degree and rate— 
that is all—and in arid regions the rate is 
conspicuous. 
There are two forces at work wherever 
water runs or ice flows, which, so far as I 
know, have not been sufficiently defined up to 
the present. They are cross-cutting and retro- 
