638 

and has therefore unwittingly confirmed my breeding 
experiments. The development of typica into taeniata 
is reversible, for it also happens that forma taeniata will 
change back to forma typica. Mr. E. G. Boulenger has 
confirmed this in experimental animals which he kept 
in the larval condition on colour backgrounds. He 
obtained in this way results far more beautiful and 
significant than my own. 
At the end of the experiments, then, I have two 
types of striped Salamandra: first, the Salamandra 
which are found in Nature ; and, secondly, those which 
have been bred in the laboratory from spotted parents. 
The former is an anciently established natural race, the 
latter a “new” laboratory race, and both of these 
are externally identical. I used both types for inter- 
crossing and inter-transplantation and also to complete 
my transmutation experiments. 
If spotted Salamandra be crossed with naturally 
striped Salamandra, the offspring are of either one race 
character or the other in the Mendelian fashion. 
Spottedness is dominant over stripedness. If one 
crosses naturally spotted Salamandra with experi- 
mentally striped Salamandra the hybrids are of an 
intermediate character (stripe-spottedness) and Men- 
delian segregation does not occur. The hybrid indicates 
therefore a difference between “old” and “new” 
characters, even though it happens that externally both 
are identical. Doubiless both are heritable, but only 
the long-established race characteristic obeys the 
Mendelian laws. The new characteristic does not 
exhibit any atavistic tendency toward the grandparent 
race. These facts acquire a special interest when we 
recall that the vast majority of Mendelian experiments 
has been done on long-established race characters. 
These old and new characteristics can be distinguished, 
not only by means of crossing-experiments, but also 
by means of experiments on ovarian transplantation. 
If ovaries of spotted females are transplanted into the 
naturally striped ones, then the appearance of the 
young is determined by the origin of the ovaries— 
according to the érue mother and not according to 
the foster-mother. They are always irregularly 
spotted. If, on the other hand, ovaries of spotted 
females are transplanted into artificially striped ones, 
then, if the father is spotted, the young are line-spotted ; 
if the father is striped, the young are wholly striped. 
The ovary of the spotted female brings into the body 
of the naturally striped foster-mother only its own 
hereditary properties as effective in fertilisation. In 
the body of an artificially striped foster-mother this 
same ovary behaves as if it had been derived from the 
body of a striped female and as if the eggs of the 
striped female had been used in the crossings. 
The objection cannot be raised that the operation 
was not thorough—that portions of the original | 
ovaries may have been left behind in the foster-mother, 
as in Guthrie’s experiments on fowls, which were after- 
wards tested by Davenport and found to be merely 
cases of regeneration of the original ovaries. Thanks 
to its enclosing membrane, the ovary of the Salamandra 
can be removed from the surrounding tissue as a whole. 
It is impossible that any remnants could have been 
left behind and that the descendants were derived from 
these remnants regenerated. 
These experiments on ovarian transplantation first 
NO. 2793, VOL. IIT] 
NATURE 
[May 12, 1923 


led me to consider the possibility of the true inheritance 
of somatic characters. This conception of mine was 
supported by the experiments of Secerov, who, to 
begin with, had obtained analogous changes in Sala- 
mandra maculosa (forma taeniata) by influencing the 
larve. Secondly, Secerov had nfeasured the amount 
of light which was able to penetrate the interior of the 
body of the Salamandra. Only one-sixth of one per 
cent. of the outer light reached the ovaries, and the 
colours of the surroundings were reduced by absorption 
by the skin. It is improbable, therefore, that there 
could have been any direct colour influence on the 
cells of the ovary and a colour-adaptation by “ parallel 
induction.” After considering this, together with the 
results of transplantation, only one plausible hypothesis 
remains, namely, that the colour changes become 
inherited by a “somatic induction”; by a process 
similar to, if by no means the same as, that which 
Charles Darwin had already imagined in his theory of 
Pangenesis and Cunningham and Hatschek brought 
forward later on to explain the phenomena of heredity. 
The different reactions of old and new, inherited 
and acquired, characters in transplantation and cross- 
ing, I have tried to make intelligible by an analogy, 
I must confess, provisional and crude. A new piece 
of clothing irritates, but this irritation diminishes the 
longer the clothing is worn, and it ultimately disappears. 
Likewise, there is a morphological irritation from each 
part of the body, and this diminishes in the same way. 
When there have been recent changes the irritation 
is stronger. 
and intensity the irritation penetrates to the germ 
plasma. There it renders permanent a potentiality 
for repetition of the actual change which brought it 
into play. In a new character, as time goes on, the 
morphological irritation diminishes. Its germ-plasmic 
induction is no longer effective. It now belongs to 
the past. For the present it is no longer necessary, 
because without it the corresponding tendency is fixed 
in the germ cells. The znductive dependence is a rela- 
tion existing between the germ plasm and only newly 
acquired characters. Between germ plasm and old 
characters, the morphological irritation of which has 
by use long since disappeared, there exists a complete 
independence as demanded by Weismann’s theory 
and proved by Mendel’s experiments. 
I will now touch briefly on further results of my 
experiments, though these now deal not with in- 
heritance but with changes induced on one generation. 
Since it is just these experiments which are cited as 
evidence against the inheritance of acquired characters, 
it will not be out of place to give a brief refutation 
of this, as I think, mistaken interpretation. I have 
succeeded in developing the rudimentary visual organ 
of Proteus unto a full-sized functioning eye by red 
illumination to which the animals were exposed for 
five years from birth. The degeneration of the eyes 
in cave-dwelling animals, according to the other view, 
cannot be made hereditary. It is only a non-hereditary 
modification, a mere environmental change. Other- 
Under suitable conditions of duration — 
hea. 
oe 
wise it would be impossible, by exposure to light for — 
a single generation, to undo what life in darkness for 
so many generations had produced. 
What contradicts this view is that exposure to 
ordinary daylight is not effective. In daylight the 
ee ee ee eee 
