
a a 
May 12, 1923] 
skin which covers the rudimentary eye is filled with 
a dark pigment. This considerable but by no means 
complete absorption of light due to the pigment is 
sufficient to arrest the development of the eye, so 
that the normal degeneration occurs. Red light, 
however, causes no pigmentation in the skin, and only 
under the influence of this chemically inactive light 
is the regression overcome. 
The misinterpretation of these data allows me to 
make a further general observation. In order to 
prove completely that acquired characters are in- 
herited we must produce at least one alteration of an 
inborn property. But if we only recognise those 
properties as hereditary which are unchangeable, 
then we have from the very outset excluded all herit- 
able transformations, and at the same time rendered 
useless any investigations of the matter. If that 
which changes cannot be hereditary, and if that which 
is hereditary cannot change, we can only predict the 
immutability of species and therewith dogmatically 
leave on one side, not only the inheritance of acquired 
characters, but also the whole theory of evolution. 
All existing objections, which rendered insoluble 
the inheritance of acquired characters, apply also 
to my breeding experiments on Alytes, and I myself 
would not have attached any special significance to this 
were it not that it is a result of just these experiments 
which has aroused the keenest interest in England— 
the development of a nuptial pad in the male Alytes. 
In male frogs, which pass their mating-time in water, 
there appears before mating, usually on the inner 
fingers, a rough, horny, glandular, dark-coloured pad. 
On the other hand, in Alytes, which mate on land, 
no trace of such a pad is to be seen. Yet it can be 
made io appear after several generations by compelling 
the Alytes to mate in water, like other European frogs 
and toads. This compulsion is brought about by 
raising the temperature, under which condition the 
mating animals stay longer in the water than usual, 
for if they did not do so they would run the risk of 
being dried up. Later in life compulsion becomes 
unnecessary. The stimulus of warmth produces an 
association through which henceforward the Alytes 
take to the water of their own accord when they wish 
to mate. 
Of the many changes which gradually appear in 
_ this water breed during the various stages of develop- 
ment—egg, larva, and the metamorphosed animal, 
young and old—I will describe only one, the above- 
Mentioned nuptial pad of the male. At first it is 
confined to the innermost fingers, but in subsequent 
breeding seasons it extends to the other fingers, to 
the balls of the thumb, even to the underside of the 
lower arm. After spreading, it exhibits an unexpected 
variability, both in the same individual and between 
one individual and another. The variability in the 
same individual is shown by the characters altering 
_ from year to year and in the absence of symmetry 
between the right hand and the left. In one specimen 
the dark pad extended to all the other fingers and 
almost over the whole of the left hand. On the 
_ right hand it was never so marked, and it was even less 
_ developed later, because the skin was stripped from 
_ this hand in the living animal for the purpose of 
histological investigation. The present skin and pad 
: NO. 2793, VOL. 111] 
NATURE 
639 
formation next to the inner finger is to be ascribed 
to regeneration in the mating season which followed. 
Microscopical preparations show the difference between 
the thumb skins of the mating male Alytes in the 
control breed and the padded skins of the water 
breed. The skin on the thumb of the normal male 
is subject likewise to an annual change in thickness. 
Alytes has already in its natural state a tendency 
to pad formation, and therefore does not display such 
a striking novelty as microscopic observation would 
lead one to think. 
The great variability and extent of the pad, which 
can be produced by cultivation, and its independence 
of the testes, as castration experiments show, render 
the hypothesis possible, that what we are dealing with 
is an artificial creation of a new function. On the 
other hand, the Alytes pad can be interpreted as an 
atavism; or again, since the tendency was already 
there, one can quite well deny that the character has 
been acquired. Further, the influence of the heat 
responsible for the change penetrates the whole body 
of the cold-blooded animal and may therefore pene- 
trate to the germ plasm in a purely physical manner. 
It is true that when four generations have altered 
in a similar manner, even after the stimulus has been 
removed, it is not very plausible that parallel induc- 
tion should be the cause, and the subsequent appear- 
ances a mere after-effect. But as the atavism objection 
can always be raised, it is not very clear to me why 
just this experiment (with Alytes) is so often looked 
upon as an experimentum crucis. In my opinion it 
is by no means a conclusive proof of the inheritance 
of acquired characters. 
Not content with any of the previous experiments, 
I carried out, before 1914, what may really be an experi- 
mentum crucis. I have written a few words on it 
in my “Allgemeine Biologie.” There has been no 
detailed publication as yet. The subject is the 
Ascidian, Ciona intestinalis. If one cuts off the two 
siphons (inhalent and exhalent tubes), they grow again 
and become somewhat larger than they were previously. 
Repeated amputations on each individual specimen 
give finally very long tubes in which the successive 
new growths produce a jointed appearance of the 
siphons. The offspring of these individuals have also 
siphons longer than usual, but the jointed appearance 
has now been smoothed out. When nodes are to be 
observed, they are due not to the operation but to 
interruptions in the period of growth, just as in the 
winter formation of rings in trees. That is to say, 
the particular character of the regeneration is not 
transferred to the progeny, but a locally increased 
intensity of growth is transferred. In unretouched 
photographs of two young Ciona attached by their 
stolons to the scratched glass of an aquarium, the upper 
specimen is clearly seen to be contracted ; the lower 
is at rest and shows its monstrously long siphons in 
full extension. They were already there at birth, 
for it was bred from parents the siphons of which 
had become elongated by repeated amputation and 
growth. 
In those animals with artificially lengthened siphons 
we can, furthermore, combine with the amputation 
at the front end another amputation at the hinder 
end. At the hinder end—in the coils of the intestine 

