ON THE ’ 
CHANGE OF THE MEXICAN AXOLOTL 
TO 
AN AMBLYSTOMA. 
BY 
DR. AUGUST WEISMANN. 
TKANSLATED BY HENRY M. DOUGLASS. 
REPRINTED FROM THE SMITHSONIAN REPORT FOR 1877. 
WASHINGTON: 
GOVERNMENT PRINTING OFFICE. 
18 7 8. ON THE 
CHANGE OF THE MEXICAN AXOLOTL 
TO 
AN AMBLTSTOMA 
BY 
DR. AUGUST WEISMANN 
TRANSLATED BY HENRY M. DOUGLASS 
REPRINTED FROM THE SMITHSONIAN REPORT FOR 1877\ 
WASHINGTON: 
GOVERNMENT PRINTING OFFICE. 
1 S 7 8 . ON THE CHANGE OF THE MEXICAN AXOLOTL TO AN AMBLY- 
STOMA. 
By Dr. August Weismann, Professor at Freiburg, Breisgau. 
[Translated by Henry H. Douglass, Champaign, 111.] 
Since first announced the change of a number of Axolotls 
into the so-called Ambly stoma form, this Mexican fish-salamander has 
been bred in aquariums at many European places, principally with a 
view to determine the conditions under which that change occurs, and 
then from that to draw further conclusions upon the real causes of this 
exceptional and therefore so mysterious metamorphosis. 
The creatures were easily propagated everywhere, and that in great 
numbers. Yet not only did the cases in which the transformation 
occurred remain exceedingly rare, but no one succeeded in answering 
the first and most important question, whether this was called forth by 
external circumstances or was dependent upon purely intrinsic causes. 
Much less were auy definite external influences found through whose 
presence one could bring on the metamorphosis with certainty. 
But until this point is decided, all attempts at theoretical interpre- 
tation and use of the phenomenon must remain without any firm found- 
ation. 
To me this history of the Axolotl’s transformation always seemed to 
be of quite a particular interest. I believed, indeed, that possibly this 
one special case might be sufficient to determine the correctness of the 
fundamental principles, according to which the origin of species is repre- 
sented in the two hostile camps of development and heterogeneous or dis- 
tinct creation. So I resolved to set about experiments with the Axolotl 
myself, in the hope that perhaps it would be my good fortune to throw 
some light upon this subject. 
In the year 1872, Herr von Kolliker had the kindness to present to 
me five of the Axolotls he had bred in Wlirzburg. In the following 
year these produced an abundant brood. With them I tried to settle 
the question, whether it was not possible to force all the larvae to the 
change, or at least the greater part, if they were brought into circum- 
stances which rendered the use of the gills difficult, that of the lungs 
easy. This, in other words, would be to compel them from a certain age 
to live half on the land. It will be seen further on, what theoretical 
grounds led me to this idea. 
Meanwhile I reached no results that year. The most of the larvae 
died before the time for such experiments seemed to have come. The 350 
ON CHANGE OF MEXICAN AXOLOTL TO AN AMBLYSTOMA. 
few survivors did not make the change, but lived until the early part of 
the next year, only to die one after another. 
Evidently I had given them too little care and attention, being pre- 
vented by a somewhat lengthened absence from Freiburg, as well as by 
other labors. 
I reached the conviction at that time that without the greatest care- 
fulness and attention in the rearing no result can be obtained. This 
conviction further experience only confirmed. One must just concen- 
trate all his interest upon this one aim, and must not begrudge having 
to devote considerable time every day for months to this work. That I 
could not carry this on myself, without giving up other work for it, 
seemed clear to me, and so I hailed it with joy when the opportunity 
offered of seeing the experiments made by another person. 
Miss von Chauvin, a lady well known to several correspondents in 
this department by her beautiful observations on the Phryganidae (alas, 
not yet published), proposed the following year to take a number of the 
lame just out of the egg, rear them, and make the attempt to bring 
them, in a certain measure, by force into the Amblystoma condition. It 
will be perceived by the following records of the lady herself how com- 
pletely this attempt succeeded, and no less will it be seen that this suc- 
cess was possible only with such carefulness in the treatment and deli- 
cacy in the observation as were here employed. 
EXPERIMENTS. 
“I began the experiments June 12, 1874, with five larvae about eight 
ilays old, all that remained alive of the twelve delivered to me. On 
account of the extraordinary tenderness of these larvae, the quality and 
temperature of the water, the kind and quantity of the food supplied, 
especially in the earliest period, exert a great influence upon them, so 
that one can hardly be cautious enough in their management. 
“ The little creatures were kept in a glass globe of about 30 centi- 
metres in diameter; the temperature of the water was regulated, and 
as nourishment Daphnida were furnished at first, afterward larger 
aquatic animals in greater quantity. In this way the five larvae throve 
exceedingly well. By the end of June the strongest larvae showed the 
beginning of the fore legs, and the 9th of July the hind legs made 
their appearance. At the end of November I noticed that one Axolotl— 
tor brevity I distinguish it "with I., and shall note the others with the 
successive Eoman numerals—kept constantly on the surface of the water, 
which brought mo to the supposition that the right time had come to 
prepare for the change to a land salamander. 
“For this purpose, on December 1,1874,1. was put into a considerably 
larger glass vessel with a flat bottom, which was so placed and filled with 
water that only in one spot could he dive quite under water, while every- 
where else in his repeated creepiugs about the bottom of the vessel he 
came more or less in contact with the air. On the following days the ON CHANGE OF MEXICAN AXOLOTL TO AN AMBLYSTOMA. 
351 
water was gradually diminished, aud during this time the first changes 
in the animal appeared: The gills began to shrivel. At the same time 
the creature showed a striving to reach the shallow places. The 4th of 
December it betook itself wholly to the land and crept away into the 
damp moss which I had placed on a bed of sand in the highest part of 
the bottom. This was followed by the first molt. Within the four days 
from December 1 to 4 a striking change took place in the exterior of I.; 
the gill-tufts shriveled almost entirely together, the crest on the back 
vanished completely, and the hitherto broad tail assumed a rounded 
form similar to a land salamander’s tail. The grayish-brown color of 
the body gradually changed to blackish ; here and there white spots 
came out, faintly marked at first, and growing more distinct. 
‘‘When, on the 4th of December, the Axolotl crept out of the water, the 
clefts of the gills were still open, but they gradually closed, and in about 
eight days they were no longer to be seen and were grown over with a 
skin. 
“Of the other larvoe, at the end of November (that, is at the same 
time when I. came to the surface of the water), three more were just as 
strongly developed as I.—an indication that for them also the right 
time had come for the hastening of the process. They were therefore 
subjected to the same treatment. II. changed in fact at the same time 
as I., and, precisely like it, had perfect gill-tufts when it was placed in the 
shoal water, and after four days had almost completely divested itself of 
these. He went upon the land; and then followed, in the course of about 
ten days, the groicing over of the gill-clefts and the complete assumption of the 
salamander's form. During this last time the animal took food, but only 
when he was urged. 
“With III. and IV. the change went on more slowly. They did not so 
frequently seek for the drier places, and generally did not expose them- 
selves so long to the air, so that the greater part of January passed 
before they took wholly to the land. Yet the drying of the gill-tufts 
required no longer time than I. and II. The first molt also followed as 
soon as they crept upon the land. 
“V. showed much more remarkable deviations in the transformation 
than III. and IV. As this individual from the beginning appeared much 
more feeble than the others, and also staid behind remarkably in its 
growth, this was by no means surprising. It required fourteen days 
instead of four to carry the change so far that it could leave the water. 
It was of quite peculiar interest to watch the condition of this individual 
during this time. As may be inferred from its delicate nature, it was 
more sensitive than the others to all outward influences. If it was ex- 
posed to the air too long it took on a lighter color. Besides it gave 
off a peculiar odor, similar to what salamanders diffuse when they are 
alarmed or endangered. As soon as these appearances were noticed, it 
was immediately put into deep water, where it dived under at once and 
gradually recovered again. Then the gills always unfolded again. The 352 
ON CHANGE OF MEXICAN AXOLOTL TO AN AMBLYSTOMA. 
same experiment was made repeatedly and was every time attended by 
the same result; from which it may well be concluded that through the 
exercise of too energetic constraint with a view to hasten the process of 
transformation a stoppage may be brought about, and death, indeed, by 
continued constraint. 
“Of Axolotl Y. it remains to be stated that he crept out of the water, not 
like all the others at th e first molt, bat at the time of the fourth. 
“All these are to-day alive, healthily and strongly developed, so that, 
as regards their nourishment, nothing would stand in the way of their 
propagation. The largest of the first four has a length of 15 centime- 
tres. Axolotl Y. measures 12 centimetres. 
“From what I have said must be established the correctness of the 
view suggested in the beginning: most Axolotl larvae, if not all, com- 
plete their metamorphose, if, in the first place, they come out of the 
egg healthy and are properly fed, and, in the second place, meet with 
arrangements which force them to change from breathing under water 
to breathing above the water. It will be understood that this compul- 
sion may only be exerted quite gradually, and in a way which does not 
draw too much upon the animal’s vital force. 
“ Freiburg, Breisgau, July, 1875. 
“MABIE v. CHAUVIN.” 
I remark upon the above records, that in all five cases the change 
was a complete one, not to be confounded with that which all the Axolotls 
confined in small glass vessels undergo more or less; namely, there 
are frequently variations which seem to advance toward the Amblystoma 
form without reaching it. In the five full-grown Axolotls which I possess 
at this very time, and of which two at least are four years old, the gills 
are all very much shriveled up, but tail and crest are unchanged. But 
the crest may also disappear and the tail grow smaller without changing 
to the Amblystoma, as shall be shown further on. As regards the time 
used for the transformation, in Axolotls I. to Y. it took 12 to 14 days. 
Four days of that bring us to the first change, during which the animal 
still remains in the water; the rest of the time is given to the comple- 
tion of the change. gives the time of the transformation at 10 
days. 
From the experiments communicated the following seems to me espe- 
cially noteworthy : The five Axolotl larvae, which alone can come into the 
account, since the others died early, all without exception completed the 
change and became Amblystomas. Only one of them, No. I., showed by con- 
stant swimming on the surface, which was noticed at the end of the sixth 
mouth, a distinct inclination to the change, a preference for breathing 
with the lungs. Of this individual, therefore, it may well be assumed, 
that also without artificial aid it would have come to the land and have 
undergone the transformationalist as was the case in some thirty spec- 
imens which observed. ON CHANGE OF MEXICAN AXOLOTL TO AN AMBLYSTOMA. 
353 
For Nos. II., III., and IV., on the other hand, such a supposition has but 
little probability. All three larvae sought to keep themselves in deeper 
water, avoided as long as possible the shallow places which compelled 
them to mere lung-breathing, and only completed the change more than 
a month later. 
With No. V., moreover, it can scarcely appear doubtful that it would 
not have made the change without the forcible habituation to abode in 
the air. 
From these results may well be drawn the conclusion, most Axolotl 
larvae change to the Amblystoma form if at the age of six to nine 
months they are brought into water so shallow that they must chiefly 
breathe with the lungs. The experiments in question are small in num- 
ber, to be sure, but such a conclusion cannot be called hasty, when it is 
considered that among many hundreds (the number is not 
given precisely) of Axolotls, obtained only some thirty Amblystomas; 
that likewise, amoug several hundred Axolotls, von Kolliker bred only a 
single Amblystoma. 
It only remains doubtful whether every larva can be compelled to make 
the change, and this question can only be determined by new experiments. 
It had been my intention to defer the publication of those above given 
until they were repeated in greater fullness by Miss von Chauvin. But 
as my Axolotls have given me no young this year (1875), I had to give 
that up for the time, and could do this the more willingly, since it is 
rather irrelevant to the theoretical interest of the facts whether all or 
only almost all of the Axolotls may be forced to make the transforma- 
tion. On the other hand, I will not forbear to mention, that the curator 
of the zoological museum here, Herr Gehrig, reared a considerable num- 
ber of larvae from the same brood, with which Miss von Chauvin experi- 
mented, and that of these larvae six passed through the winter without 
undergoing the transformation. They were always kept in deep water, 
and therefore constituted the counter-experiment to those communicated 
above, showing that this whole brood by no meaus possessed a predis- 
position to undergo the transformation. 
If, now, these new facts are to be made use of to clear up our conception 
of the nature of this unusual process of transformation, we must, before 
all else, bring to our aid the data already known. 
In the first place, it is to be laid down that Siredon Mexicanus, in its home, 
so far as ire hnow, never undergoes the transformation. From that locality 
it is Tcnoicn only in the Siredon form. The testimony which I find on this 
point comes from De Saussure, * who himself observed the Axolotl in 
the Mexican waters. This naturalist has never seen even a single Am- 
blystoma in the vicinity of the lakes, and “yet the larvae of the Axolotl 
are so common there, that they are brought into market by thousands.” 
De Saussure believes that the Axolotl does not make the change in 
Mexico. 
* Yerhandl. d. Schwoiz. Naturforscli. Gesellschaft. Einsiedeln, 1868. 
28 s 354 
ON CHANGE OF MEXICAN AXOLOTL TO AN AMBLYSTOMA. 
Cope* asserts tlie same thing quite decidedly; and individuals bred 
by him in Americafrom the Siredon Mexicanus, also in captivity, “showed 
no inclination to the transformation.” On the other hand, Tegetmeiert 
saw the change begun in one of five individuals which traced their 
descent to the Lake of Mexico and so the second fact is established, 
that also the genuine and real Axolotl under certain circumstances changes 
in confinement to an Amblystoma. 
This remark would be superfluous, if it were, as was long believed, 
true that the Axolotl of the Paris Garden of Plants, whose metamor- 
phosis was first observed and excited so much attention at the time, 
actually belonged to the species Siredon Mexicanus, the only Siredon 
which in its home bears the name Axolotl. 
In his first communication, Dumdril himself was of this opinion ; he 
called the creature “Siredon Mexicanus ” syn. Huniboldii; J but later, in 
his detailed work § on the change of the Axolotl contained in the Gar- 
d.eu of Plants, he recalled this view, and, after a critical examination of 
the five described Siredons, came to the conclusion that the Axolotls in 
the possession of the Paris Museum were probably Siredon lichenoides 
Baird. 
So all the transformations of Axolotls observed in Europe were to be 
referred to this species; for—at least so far as is known—they are all 
descendants of the Paris colony. Thence also, indirectly, came the sub- 
jects of my experiments. 
To be sure, now it does not agree with this, that the Amblystoma 
form, which Dum6ril obtained from his Axolotls, corresponded with 
Amblystoma tigrinum Cope, while we learn by Marsh || that Siredon 
lichenoides Baird changes to Amblystoma mavortium Baird when it under- 
goes the complete metamorphosis. Marsh found the Siredon lichenoides 
in Alpine lakes (7,000 feet above the sea), in the southwestern portion of 
the United States (Wyoming Territory), and by breeding in aquariums 
obtained from it the Amblystoma mavortium Baird. Nevertheless, he holds 
it doubtful whether the animal ever goes through the change in its 
home; to be sure, without any sure foundation and on purely theoretical 
grounds, namely, because in his judgment “the colder temperature 
there is less favorable to it.” 
If I doubt the correctness of this last opinion, it is only because the 
Amblystoma mavortium in a state of nature has been found in many 
parts of the United States, namely, in California, New Mexico, Texas, 
* Dana and Silliman, Amer. Journ., 3d series, i, p. 89. Aunals Nat. Hist., vii, p. 246. 
t Proceed. ZjoI. Soc. 1870, p. 160. 
t Compt. Rend., tome 60, p. 765 (1865). 
$ Nouvelles Archives du Mus6um d’Hist. Nat. Paris, 1866. Tome ii, p. 268. 
(1 Proceed. Boston Soc., vol. xii, p. 97 ; Silliman, Amer. Journ., vol. xlvi, p. 364; a 
reference to this in Troschel’s Jahreshericht for 1868, p. 37. 
51 Proceed. Boston Soc., vol. xii, p. 97; Silliman, Amer. Journ., xlvi, 364. I could not 
examine these works myself, and quote from the reference in Troschel’s Jahreshericht, 
1868, p. 3-7. ON CHANGE OF MEXICAN AXOLOTL TO AN AMBLYSTOMA. 
355 
Nebraska, and Minnesota. Nevertheless, it is by no means incredible 
that just in the Alpine lakes, from which Marsh obtained it, the species 
occupies a different position in regard to the transformation from what 
it does in other habitats. This will be seen from the observations on 
the Triton adduced below. So, until further observations are made, I 
believe we must suppose that the Paris Axolotls are not Siredon liche- 
noides, but probably a new and very closely related species. 
Meanwhile, there is not much depending upon this in the review of 
the transformation, if it is only established that this Axolotl in its home 
does not undergo the metamorphosis, or only does it as exceptionally 
as in Europe. Unfortunately, there is in communications no 
precise statement of the locality where this “Mexican” animal was 
found; probably the locality was unknown to himself, and so I can 
only assume on Cope’s authority that no Amblystoma has ever been 
brought from south of the provinces of Tamaulipas and Chihuahua; 
that is, from south of the tropic.* 
Of that, however, there is no certainty. Much more important is the 
fact adduced above, that the genuine Axolotl of the lakes around the 
city of Mexico never makes the change to an Amblystoma, but that this 
species also in isolated cases undergoes the transformation in confine- 
ment. Now, from this and from the fact that the Paris Axolotls in con- 
finement are only transformed to the extent of a very small percentage, 
it may be concluded that also in its home it is either not transformed at 
all, or only as an exception. 
But still another series of facts comes very essentially into the account 
in reviewing this history of transformation : I mean the existence of a 
considerable number of species of Amblystoma in nature. In the “ Revision 
der Salamandriden-Gattungen” (Review of the genera of Salamanders) 
which Strauch f gave a few years ago, there are represented, accord- 
ing to the examples of Cope,§ twenty species of the Amblystoma Tschudi 
living in North America. Now, although a few of these species are 
based only upon one specimen, and, therefore, as Strauss correctly 
says, “must be suppressed in time,” yet there is left a whole range of 
species which certainly live and propagate as Amblystoma, and which 
have their locality from tbe latitude of New York to that of New Mexico. 
lienee there are certainly species of Siredon which, under their natural con- 
ditions of life, regularly take the Amblystoma form and propagate in it; 
ichile on the other hand there are at least two species which, under their 
present natural conditions of life, only propagate as Siredon. It is only 
another form of expression for this fact when one says the Mexican 
Axolotl, and the Paris Siredon as wTelI, be this latter named lichenoides 
or something else, stands upon a lower phyletic stage of development than 
the other species of Amblystoma which propagate in the salamander 
* Daua and Silliman, Amer. Journ., 3d ser. i, p. 89; Annals of Nat. Hist, vii, p. 246. 
t Proceedings Acad. Philad., xix, 1867, pp. 166-209. 
$ Mem. Ajad. Petersb., tome xvi. 356 
ON CHANGE OF MEXICAN AXOLOTL TO AN AMBLYSTOMA. 
form. No one can object to this statement; while the other proposi- 
tion, either expressed or tacitly held by all authors, includes a theory 
which I regard as incorrect. That proposition is : The Mexican Axolotl 
has remained upon a lower phyletic stage of development. 
All zoologists who have expressed themselves on the transformation 
of the Mexican Axolotl, and who are not, like its first observer, still 
attached to Cuvier’s views on the immutability of species, took up the 
matter as if here was a species which through some sort of special cir- 
cumstances had remained behind on a lower stage of development, and 
was now stimulated by some sort of influences to its advance to a 
higher stage. 
For a long time also I did not believe that the thing could be other- 
wise understood, little as I was able to bring all the phenomena into 
harmony with this view. Thus in 1872 I used the following expres- 
sions “Why should not a sudden change in all the relations of life 
(the removal from Mexico to Paris) have had a direct influence upon 
the organism of the Axolotl, so that he suddenly attained a higher stage 
of development, which many of his kindred species have long since 
attained, which evidently lies in the nature of his organization, and 
which perchance he also would have reached in his native haunts, though 
at a later period *? Or would it be too much to suppose that by the 
sudden transportation from 8,000 feet above the sea-level (the Mexican 
highland) to the altitude of Paris, the organs of respiration had re- 
ceived a shock which just brought them to the change already close at 
hand ? So in all probability we have to do with a direct effect from 
altered conditions of life.” 
That the purport of the last sentence must also be held true to-day 
follows as a matter of course from the experiments communicated above, 
which surely prove that by the application of definite external influences 
we have to a certain extent it in our power to call forth the transforma- 
tion. Just in that fact lies the new light which these experiments have 
brought. 
But must we also understand the phenomenon in the manner above 
indicated—that is to say, as a phyletic onward development of the species, 
appearing suddenly and in a measure resulting from a shoclc f I believe 
not. What first perplexed me in regard to this view was the sight of 
my living Amblystomas reared from the larvae of the Axolotl. These 
creatures by no means show a variation from the Axolotl iu single traits 
merely, but they are distinguished from it iu their entire habit; they 
differ somewhat in all parts; although slightly in many, in others quite 
decidedly; in short they have become entirely different animals. Corre- 
sponding to this they also live quite differently; no longer go into the 
water, but prefer iu the daytime to keep hidden in the damp moss of 
their prison, and at night come forth to seek their food on the dry land. 
*“Ueber (leu Eiulluss der Isolirung a,uf die Artbilduug” (Ou the Influeuce ot Isolation 
upon the Formation of Species), Leipsic, 1872, p. 33. ON CHANGE CF MEXICAN AXOLOTL TO AN AMBLYSTOMA. 
357 
To be sure I could have perceived the great difference between the 
two stages of development from the anatomical data long familiar to 
me which Dumdril had given on the structure of his Amblystomas; 
but the reading over of many statements in detail gives no living pict- 
ure. At any rate the sight of the living animal first brought me to the 
consciousness how comprehensive is the change we have to do with 
here; that by no means does it merely concern those parts which are 
directly affected by the alteration in the mode of life, but that most, 
if not all, parts of the animal undergo a transformation, which can be 
very well explained in part as the morphological adaptation to new con- 
ditions of life, in part as consequences of these adaptations (correlative 
changes), but cannot by any possibility be explained in a word as the sud- 
denly appearing effect of these changed conditions of life. Such at least is 
my view, according to which a sudden development of the species, such 
as must have taken place here, is quite inconceivable. I willingly ac- 
knowledge that a few years ago the question whether such sudden de- 
velopment occurs was still an open one for me, but since then my invest- 
igations have kept urging me to the conviction that it does not occur at 
all, as I shall show in another place. Here I have to confine myself to 
the examination of this individual case; a case which appears to me— 
as was above intimated—quite peculiarly suited to shed light decisively 
upon the great alternative, about which at the present moment the war 
of opinions is centered, in regard to the doctrine of descent. 
I may well assume that hitherto it has been with most students in 
regard to the metamorphosis of the Axolotl very much as with myself; 
it did not come to their consciousness how far the transformation goes ; 
and so it may need to be made plain that the theoretical importance of 
the case and its value as a basis of inference was not properly emphasized 
by either side. But it is evidently a case of quite unusual significance 
for the principles involved. I believe it may be easily shown that the 
hitherto pretty generally admitted explanation of the history of the 
Paris Axolotl’s transformation includes at the same time the recogni- 
tion of a very far-reaching principle, namely: if this explanation were 
the correct one, then in my judgment would be at the same time demon- 
strated as correct the opinion of those who, like Kolliker, Askenasy, 
Nageli, and among the philosophers Hartmann and Huber, would 
refer the transmutation of the species in the first instance to a motive 
power dwelling within the organisms, to an active, i. e. spontaneous, 
“law of development,” a “principle of perfection,” or, as I should pre- 
fer to name it, a phyletic vital energy, in contrast to the exactly corre- 
sponding vital energy of the so-called “philosopher domain” in the 
nature of ontogenesis. 
If, namely, the Axolotls that have become Amblystomas are to be taken 
as individuals which, stimulated by external influences, have hastened 
on in advance of the remaining individuals in their phyletic develop- 
ment, then this advance can only be placed to the account of a phy- 358 
ON CHANGE OF MEXICAN AXOLOTL TO AN AMBLYSTOMA. 
letic vital energy; for the transformation is a sudden one; it leaves no* 
time for gradual adaptation in the course of generations. Indirect influ- 
ence of the external relations of life, i. e. natural selection, is accordingly 
excluded from the first; but direct influence of the altered circumstances 
of life does not suffice by a great way to explain the total transforma- 
tion of the whole structure, as I have already intimated and will uow 
show more minutely. 
The distinctions between the Paris Axolotl and its resulting Ambly- 
stoma are, according to Dumisril, Kolliker, and my own observations, 
the following: 
1. The gills disappear, the clefts of the gills close up, and only the 
foremost of the arches of the gill remains in existence; the other arches 
vanish. At the same time the Os hyoideum is changed (Dumi-iril). 
2. The crest on the back disappears entirely (Dum^ril). 
3. The rudder-like tail changes to a tail like a salamander’s (Dum£ril), 
which nevertheless is not quite circular in section as in the salamander, 
but is somewhat compressed laterally (Weismann). 
4. The skiu gets yellowish-white patches, irregularly distributed on 
the sides and th.e back (DumJsril), while at the same time its former 
ground color of grayish-black changes to a shining greenish-black (Weis- 
mann), and, besides, the shiny secretion from the skiu is lost and the 
glands of the skin become indistinct (Kolliker). 
5. The eyes become promiuent and the pupils small (Kolliker), and 
eyelids are formed which can close the eye completely, while in the 
Axolotl only a narrow circular fold surrounds the eye, and it cannot be 
closed (Weismann). 
6. The toes diminish in size and lose their skin-like appendages (Kol- 
liker), or, more precisely, the half web-membranes which unite the 
iuuer ends of the toes on all the feet (Weismann). 
7. The palatal teeth in this, as in all Amblystomas, stand in a diag- 
onal row, while in the Axolotl, like the Triton larvae, they stand at the 
side of the palatal vault forming an arched band, with several rows of 
teeth.* (Dumeril. See his figure. Zeitsclirift f. wissensch. Zool., xxv 
Bd., Sup., p. 279.) 
8. In the Axolotl the under jaw has, besides the teeth on the upper edge 
of the bone, some very small teeth disposed in several rows; these latter 
disappear after the metamorphosis (Dumeril). I add thatthepermanent 
teeth belong to the os dentate, the temporary ones to the os operculareA 
* Dum£ril has the teeth of the vomer separated fiom those of the os palatmum by a 
gap. Probably this was an artificial one, as Gagenbaur (Friedrich & Gagenbaur, 
“Der Scbiidel des Axolotl”—The Skull of the Axolotl—Wurzburg, 1849) figures the 
rows of teeth without interruption, passing over from one bone to the other. The 
same was true in three Axolotls which 1 examined in regard to this point; moreover, 
this little difference is quite unimportant in the question here treated. 
t See O. Hertwig, “ Ueber das Zahnsystem der Amphibien und seine Bedeutung fur 
die Genesis des Skelets der Mundhohle.” (On the Dental System of the Amphibia, and 
its Significance for the Genesis of the Skeleton of the Mouth.) Arch. f. Mikroskop. 
Anat., Bd. xi, Supplementheft, 1874. ON CHANGE OF MEXICAN AXOLOTL TO AN AMBLYSTOMA. 
359 
9. The posterior face of each vertebra is slightly hollowed out before 
as after the transformation, but the anterior face is less concave in the 
Amblystoma than in the Siredon (Dumeril). 
The details cited from Dumeril under 7 and 9 I have not so far been 
able to confirm by personal examination, as I was not willing to kill any 
of my living Amblystomas, only for the purpose of verifying the details 
of a naturalist in whom perfect confidence may surely be reposed. In 
like manner, I have not yet seen the change in the arches of the gills. 
All the other data given by Kolliker or Dumeril I can corroborate 
completely. 
, The differences in structure which exist between Axolotl and Am- 
blystoma are considerably greater and more important than those 
between neighboring genera, yes, greater than are found between 
the families of the Urodela. The genus Siredcn without any doubt 
belongs to a different suborder from the genus Amblystoma, into which 
it is occasionally transformed. Strauch, who has made the latest ar- 
rangement of this group, separates the suborder of the Salamandrida 
from that of the Ichthyodea by the possession of eyelids' and by the 
position of the palatal teeth in a simple row on the posterior edge of the 
palatal bone ; while in the Ichthyodea the lids are wanting and the pal- 
atal teeth either “ stand on the front edge of the palatal bone or cover 
the whole surface of the palate as brush-like clusters.” How would it 
be possible now to regard anatomical characters standing so far asunder 
as transformations which had been suddenly called forth by a single 
operation of varying conditions of life ? 
Hand in hand with the falling out of the old palatal teeth and the ap- 
pearance of new goes a change in the anatomical structure of the verte- 
bral column, and, as we may conclude from Kolliker’s entirely ac- 
curate observation on the stoppage of the slimy secretion from the skin, 
in the histological structure of the skin itself. 
Who would undertake to explain all these deep-lying changes as the 
direct and sudden effect of any external influences whatever operating 
but once f And even if any one had an inclination to explain them as 
results of the loss of the gills, and therefore as correlative changes, what 
would such a correlation be but the newly christened vital energy above 
spoken of? Tor if from one variation caused by direct influence of 
external agencies the whole body can through correlation transform itself 
in a couple of days just so in all its parts as it appears best adapted for 
the new conditions of life in which it is henceforth to be, then the word 
correlation is only a term, by which nothing is explained, but the 
search for a better explanation is hindered. Then it is preferable that 
we simply acknowledge our belief in a phyletic vital energy. 
Moreover it certainly is not allowable even to wish to seek an explana- 
tion of that sort (by correlation), for ice know some Urodela in the adult 
state have no gills, and yet possess all other marks of the Ichthyodea : lack  ON CHANGE OF MEXICAN AXOLOTL TO AN AMBLYSTOMA. 
of the eyelids, characteristic type of the palatal teeth, and of the arrange- 
ment of the lingual bones. Such are the genera Amphiuma L., Meno- 
poma Hark, and Cryptobranchus v. d. Hoev. The first two geuera, as is 
well known, still have the clefts of the gills. Cryptobranchus, on the 
contrary, has lost these clefts, which are grown over with skin as in the 
Amblystoma, and yet, by the unanimous testimony of all systematic 
zoologists, it is a genuine fish-lizard in habit, arrangement of the lingual 
bone, palatal teeth,* &c. It must be added that even the Axololl itself 
may lose the gills without, on that account, changing to an Amblystoma. I 
have mentioned elsew'here that in Axolotls which are kept in water that 
is shallow and still, the gills frequently grow small; but it also happens •• 
that they shrivel up entirely. I have an Axolotl preserved in spirits, 
in which the gills are shriveled to little irregular bunches; at the same 
time the crest on the back is so completely wanting that a longitudinal 
furrow has appeared in its place, and on the tail the border of skin has 
entirely vanished from the lower margin, and about half from the upper. 
Nevertheless, the animal is widely separated in structure from the 
Amblystoma : it has the arches of the gills, the palatal teeth, the skin, 
&c., of the Axolotl. 
This demonstrates, therefore, that the loss of the gills by no means must 
always bring after it all the other variations which we see take place in the 
metamorphosis of the Axolotl; that these, therefore, are by no means the 
necessarily and immediately appearing result of that loss. Whether they 
must necessarily appear after long successions of generations, whether 
also the descendants of Cryptobranchus will some time in the future take 
the structure of the Salamandrida, that is another question which I 
should not like to answer just in the negative, but which does not come 
into the account here, as we are considering only a possible sudden result 
from the loss of the gills. 
The question, therefore, seems to stand thus : Either our apprehension 
up to this time of the transformation history of the Axolotl as a further 
development of the species is incorrect, or the existence of a phyletic vital 
energy is demonstrated by the case of the Axolotl beyond the possibility of a 
refutation. 
Now the question comes up, whether the facts of this transformation 
history do not also admit of another explanation. I believe that this is 
at any rate possible and that another interpretation may be shown as 
the correct one with a good degree of probability. 
I esteem those Amblystomas which in individual cases have developed in 
confinement from Siredon Mexicanus (syn. pisciformis), as well as from the 
Paris Axolotl, not as forms of advancement, but forms of retrogression. 
I believe that the Axolotls which to-day live in the lahes of Mexico were 
already Amblystomas a geological (or better a zoological) epoch earlier, but 
* See Strauch, Zeitschrift. f. Wissenscli. Zool., xxv Bd., Suppl., p. 10. ON CHANGE OF MEXICAN AXOLOTL TO AN AMBLYSTOMA. 
361 
that through alterations in their conditions of life they have sunk hack to 
their earlier stage of the Perennibranchiates. 
Without doubt I was first led to this intepretation by the results I 
had reached in my studies ou the season-dimorphism of butterflies.* 
There also we have the question of two different forms under which ono 
and the same species appears, and of which it may be proved as prob- 
able that one is phyletically the older, the other the younger. The 
younger summer form has in my view proceeded through the gradual 
warming of the climate from the winter form that in an earlier zoologi- 
cal epoch was the only one; but this latter, the primary form, has not 
ou that account ceased to exist, but to-day still alternates every year 
with the secondary, the summer form. 
Now it is easy with the season-dimorphous butterflies to make the 
summer brood assume the winter form, and that by exposing their 
pupse rather a long time to a lower temperature; and it may be made in 
the highest degree probable that this suddenly appearing variation or 
transformation, often very far reaching, is sudden in appearance only, 
and is but apparently the result of cold acting upon this generation. 
Much rather in truth the variation depends upon a reversion to the 
primary form of the species, and therefore the cold that appears but 
once is only the impulse to the reversion, but not the true cause of the trans- 
formation. This cause must be sought in the long-continued operation 
of the cold, to which thousands of generations of the ancestors of our 
living butterflies were subjected, and whose final result was just the 
winter form. 
If we now assume for a moment that my interpretation of the Axolotl’s 
transformation-history as just given is correct, we have relations here 
which are in many respects analogous to those of the season dimorphism. 
To be sure the two forms in this case no longer alternate regularly with 
each other, but the primary form may occasionally appear in place of the 
secondary, and this result from the shock of external circumstances. 
As there we succeed in stimulating the summer brood by the action of 
cold to lay aside the summer form and adopt the winter dress, so here 
we are able to lead the Axolotls into the Amblystoma state by com- 
pelling them at a certain age to breathe air. 
Further, as in the season-dimorphism, it may be shown that this 
transformation called forth by artificial means is only in appearance a 
sudden new formation, but in truth a reversion to the much older win- 
ter form, so here we also had to do, not with an actual new formation of 
the species, but only with an apparent one, a reversion to the phyletically 
older form of the species. To be sure that sounds very paradoxical if 
here is a form that must have come by reversion, and yet it must 
undoubtedly be accounted as the more highly developed. But I believe 
* Studien zur Descendenztheorie, I. Ueber den Saison-Dimorphism der Schmetter- 
linge (Studies on the Doctrine of Descent, I. On the Season-Dimorphism of Butter- 
flies), Leipsic, 1675.  ON CHANGE OF MEXICAN AXOLOTL TO AN AMBLYSTOMA. 
that on more careful consideration much of the paradox which seems to 
lie in this view will disappear. 
Before all else, it is to be considered that the phyletic development 
of the species need by no means have always gone directly forward. 
We have examples enough of development backward, although in 
rather a different sense, as in parasites and such forms, which have sunk 
from free locomotion to the sedentary mode of life. 1 do not ignore the 
difference which exists between this sort of development backward by 
the pining away of certaiu organs or systems of organs, and proper re- 
version. The latter is the return to an animal form already once existent, 
but in the former case, in spite of all simplification of the organization, 
something wholly new is always formed. But I am able to see nothing 
absurd as to principle in the supposition that a proper reversion, also, 
be it of a whole species or of the individuals of a species withiu a cer- 
taiu habitat, is thought of as possible, and I do not for the present demand 
a further eoncessiou. Why, for example, should it be so entirely in- 
credible that the Axolotl, in a time long past, was adapted to living on 
the land, that it gradually, through direct and indirect operation of al- 
tered conditions of life, had acquired the salamander form, but that 
later, by a new change of the circumstances of life unfavorable to its pres- 
ent organization, it has again fallen back into the old form or one lying 
near it ? At any rate, such a supposition contains nothing which would 
stand in contradiction with well-knowu facts, but can be sustained in 
several ways; and finally, it recommends itself by furnishing, at least 
in my opinion, the only possible explanation of the facts before us. 
The above-mentioned existence of a whole group of species of Ambly- 
stoma shows at once that species of Siredon can rise to the salamander 
form, and can regularly propagate in that form ; and further, that this 
phyletic advance has already actually taken place in very many species. 
But that a sinking back from this higher stage of development to the 
lower can also take place, several observations on our water salaman- 
ders show. 
It is well known that Tritons under some circumstances become, as it 
is usually expressed, “ sexually mature in the larval state.” 
In the year 1861 De Filippi* found in a swamp by Lake Maggiore 
fifty Tritons, of which only two showed the structure of the full grown 
water salamander; but all the others still had their gills, yet corre- 
sponded to mature animals in bodily size and in the development of the 
sexual organs, and that iu both sexes. 
Filippi determined the fact that these “sexually mature larvae” did 
not merely resemble larvae externally by the possession of gills, but that 
they also presented all the other anatomical marks of the larvae; that 
is, the characteristic clusters of palatal teeth standing on both sides, 
instead of the later simple row, and a vertebral column, which still has 
the Chorda dorsalis running through its whole length. 
* Sulla larva del Triton alpestris, Arcbivio per la Zoologia, 1861. ON CHANGE OF MEXICAN AXOLOTL TO AN AMCLYSTOMA. 
363 
According to my view this would be a case of reversion of the Triton 
to the phyletic stage lying next behind it, that is, the Pcrennibranchiate stage; 
and in this case most zoologists who staud at all upon the footing 
of the doctrine of descent will agree with my view. I at least would 
count it a profitless playing with words if any one should speak here of 
larval reproduction and believe he had thereby explained anything. In 
any event the animal becomes sexually mature in the condition in which 
it first appears as a larva. But we first get an insight into the nature of 
this phenomenon by the reflection that this so called “sexually mature 
larva” has precisely the structure which the preceding phyletic stage 
of the species must have possessd; that therefore we have a reversion 
of the individual to the older phyletic stage of the species. I es- 
teem it an error when Dumeril puts this case of the Triton in par- 
allel with the genuine larval reproduction of Wagner’s Cecidomyia 
larvae. There it is certainly not reversion to an older phyletic stage 
which makes the larvae capable of reproduction ; for these larvae present 
no older phyletic stage of the species, but must have arisen at the same time 
icith the species in its present form. The vast difference in the structure 
of the larva and the fly is not explained by assuming that the latter 
has arisen from the former as from a finished given quantity, but by 
this, that both at the same time have adapted themselves to conditions 
of life further from each other.* Begarded phyletically, these larvae are 
not at all a necessary point of transit for the origin of the fly. They 
could also be built quite otherwise, without the form of the fly needing 
to be likewise changed ; for the stages of insect transformation vary 
independently of each other, corresponding to the conditions of life to 
which they are subjected, and exert no influence upon each other, or 
only a very slight one, in the determination of form, as I shall attempt 
to demonstrate in auother place. At any rate, “the ability of these 
larvae (the Cecidomyia) to multiply asexually has first been acquired as 
a secondary matter, as already follows from the fact that there are 
numerous species of the same genus of flies which do not feed their 
young.” “ Iu the form which they possess to-day they can never have 
played the part of the final stage of ontogenesis, and therefore also 
cannot have possessed at a former time the power of sexual repro- 
duction.! In short, we have to do here with genuine larval reproduction, 
but iu the Tritons with reversion to an older phyletic stage. 
Nor can I agree with my friend Haeckel, when he designates the 
reversion of the Tritons as “adaptation to continued life iu the water.”£ 
One would only be able to speak of “ adaptation ” in this case by tak- 
ing the word in a sense quite different from that with which Darwin 
* Compare, also, Lubbock, On the Origin and Metamorphoses of Insects. London,. 
1874. 
t Seo my paper “ Ueber den Saison-Dimorphismus der Schmetterlinge,” Leipsic*. 
1875, p. CO. 
t See Haeckel’s Anthropogenie, p. 449. 364 
ON CHANGE OF MEXICAN AXOLOTL TO AN AMBLYSTOMA. 
and Wallace have introduced it into science. Those naturalists dis- 
tinguished by the word a gradual remodeling of the body, taking place 
in the course of generations, corresponding to the new demands of new 
conditions of life ; in other words, the operation of natural selection, but 
not a result from causes of variation suddenly and directly acting once 
and in one generation. 
Just because the word “adaptation” may, according to customary 
use of language, be used in many different senses, it would be desirable 
to take it only in one acceptation and to have that fixed exactly ; before 
all, not to speak of adaptation when there is no morphological change, but 
only a sort of exchange of functions in Dohrn’s sense.* Thus, for exam- 
ple, when FoiiELf shows that fresh-water air-breathing snails, whose 
organization is calculated for the direct breathing of the air, could 
nevertheless pass to the greatest depths of the Alpine lakes, while they 
used their lungs as gills. That with this not the slightest change has 
taken place in the lungs, is shown by the observations of v. Siebold4 
He saw the Pulmonates of shallow water use their lungs alternately 
for direct air-breathing and for water breathing, according as the atmos- 
pheric contents of the water was less or greater. If with v. Siebold one 
should apply the word “ adaptation ” simply to such cases, it would 
lose the special sense which was originally attached to it; as a technical 
term the word would have to be given up. 
At any rate, there is as little a case of genuine adaptation in the Triton 
“larvae” that were capable of reproduction as in the Axolotl exception- 
ally changing to the Amblystoma. In both cases, also, the transforma- 
tion in consideration is not at all indispensable for the life of the individual. 
Mature Tritons (without gills) live, as I have witnessed, many months, 
probably for years, in deep water, although they are fitted for simple 
air-breathing; and Axolotls, as I have already stated above, can live quite 
well for years in shallow and quiet water. If their gills shrivel, yes, 
vanish entirely, this also yet is no adaptation in the Darwinian sense, but 
a direct result of external influences, principally indeed of diminished use. 
A case quite analogous to Filippi’s was observed in 18G9 by Jullieu. 
Four female larvae of Lissotriton punctatus Bell, (synonym for Triton 
tceniatus Schnd.) were fished out of a swamp and showed themselves 
sexually mature. In their ovaries they had matured eggs, ready for 
laying, and two of them also actually laid the eggs. Four male larvae, 
which were taken in the same swamp, showed themselves likewise 
developed in regard to bodily size, though they had no spermatozoa in 
the testicles, but only sperm-producing cells.§ 
* “ Der Ursprung der Wirbelthiere und das Princip des Functionswechsels ” (The 
Origin of the Vertebrates and the Principle of the Exchange of Functions). Loipsic, 
1875. 
t Faune profonde du lac Ldman; Verhandl. d. Schweiz. Naturforsch. Gesellschaft 
iDeep Fauna of Lake Geneva). Schafihausen, 1873. 
t Zeitschrift f. Wissenschaftl. Zool. (Journal of Scientific Zoology), bd. xxiii, 1873. 
$ Compt. Rend., t. lxviii, p. 938, 939 (Proceedings of the French Academy of Sciences). ON CHANGE OF MEXICAN AXOLOTL TO AN AMBLYSTOMA. 
365 
A third case of the same kind I find cited* by Seydig, in his essay, 
“Ueber die Molche der Wiirtembergischen Fauna” (On the Salamanders 
in the Fauna of Wiirtemberg), so rich in interesting detail. Schreibers, 
the former director of the Vienna Cabinet of Natural History, likewise 
found “larvae” of Triton with gills well developed, but of the size of “ adult 
males,” and, as anatomical examination showed, with well developed 
sexual organs, as well as ovaries tilled with eggs. 
So, therefore, it is established that species which long ago attained 
the stage of Salamandrida in the phyletic development may occasionally 
sink back to the stage of Perennibranehiates. Evidently, this fact makes 
my view of the Axolotl as a form of reversion appear much less para- 
doxical. Indeed, the cases of reversion in the Triton are directly anal- 
ogous to the transaction I suppose for the Axolotl. 
We only need suppose Amblystomas in the place of the Tritons, and 
consider the swamp in which de Filirpi found his “sexually mature 
Triton larvae” expanded to the Lake of Mexico, regarding, also, the un- 
known, and perhaps in this case temporary, causes of the reversion as 
permanent, and we have everything which is necessary to the restoration 
of the Axolotl as we know it to-day, ice obtain a Perennibranchiate popu- 
lation of the lake. 
It is not even yet determined whether, in that swamp of de Filippi 
the Pereunibranchiate form of the Triton does not actually prevail per- 
manently, for it has not, to my knowledge, been examined since in regard 
to this subject. 
But if we assume for a moment that it actually were so, that a colony 
of Pereunibranchiate Tritons lived there, carrying on reproduction sexu- 
ally, should we wonder if occasionally, also, a genuine Triton came from 
our brood, if we could succeed in stimulating most individuals of this 
brood by removal into shallow water to the metamorphosis into Tritons ? 
But just so it is according to my view with the Mexican Axolotl. 
But I need not limit myself to support my hypothesis, but must also 
directly assail the tenability of the one hitherto held, for it stands in op- 
position to facts. 
If we really had in the Axolotl a suddenly occurring phyletic further 
development, then one fact would remain wholly incomprehensible, 
namely, the sterility of the Amblystomas. 
Of about thirty Amblystomas which Dum6ril had obtained up to the 
year 1870, full sexual maturity had not appeared in a single one; neither 
copulation nor the simple laying of eggs had taken place, and the indi- 
viduals that were examined anatomically showed the eggs immature and 
the spermatozoa present indeed, but without the undulating membrane 
belonging to all the Salamandrida. They were not destitute of all power 
of motion, but as Quatrefages stated, only moving imperfectly.! 
# Archiv f. Naturgeschichte (Archives for Natural History), 1867. 
t Compt. Read., t. lxx, 1^70. 366 
ON CHANGE OF MEXICAN AXOLOTL TO AN AMBLYSTOMA. 
The five Amblystomas, on which I have reported here, have shown, up 
to the present time, no appearances of reproduction. 
It is evidently an objection that will very poorly stand the test when 
Saco* attributes the sterility of the Amblystomas reared from Axolotls 
to “poor food.” Why, then, do the Axolotls reproduce so easily when 
they are fed exactly the same ? I can also state expressly that my Am- 
blystomas areqnite excellently fed. To be sure, these have hardly reached 
the age of one and a half years; but the Axolotls propagate the second 
year, and some of DumIsril’s Amblystomas were five years old in 1870. 
The fact of sterility stands in clear opposition to the view that these 
Amblystomas are the regular advanced guards of the genus Siredon as 
it moves forward in phyletic development. To be sure I will by no 
means maintain that my theory of reversion could actually explain the 
sterility, but at least it does not stand directly in contradiction with it. 
Mere forms of reversion may perish without reproducing themselves. 
A new form called forth by the working of an unknown phyletic vital 
energy cannot be a sterile one, because this directly destroys again the 
“purpose” which the vital energy is pursuing. Tbe idea of the vital 
energy includes that of teleology. 
Moreover, from our stand-point the sterility of the Amblystomas may 
be, if not completely understood, yet shown as a phenomenon that does 
not stand quite alone. In the case of the Lissotritonpunciatus, as quoted 
above, the female “larvae” were at any rate sexually mature and laid 
eggs, but at the same time the males had no perfected spermatozoa in 
the testicles. 
Other cases of the kind are not known to me. At the lime when I made 
my experiments with butterflies, as mentioned above, this point of view 
was still unseen, and so I neglected to examine the artificially produced 
forms of reversion in regard to their organs of generation. But general 
principles also lead to the conclusion that atavistic forms may easily 
remain sterile. 
Darwin* finds the most immediate causes of sterility, first, in the 
operation of widely varying circumstances of life; and, second, in the 
crossing of individuals with widely differing constitutions. Varying 
circumstances of life are at any rate what induce the transformation of 
the Axolotl, and from this point of view it could not be surprising if we 
find those individuals sterile, which have precisely shown themselves as 
especially affected by these altered conditions of life, since they have 
reverted to the salamander form. In this reasoning it is by no means as- 
serted that reversion always and tcitliout exception is accompanied by ster- 
ility. Nor can it be objected to my interpretation of the Axolotl’s trans- 
formation that through reversion a colony of Axolotls capable of reproduc- 
tion never could have arisen. On thecontrary, Jitllien’s female Triton 
*Bull. Soc. Neuck&tel, Bd. viii, p. 192, a refereuce to the place in Troschel’s Jalires- 
bericht (Annual Report), for 1869. 
* Origin of Species, 5th edition, p. 325. ON CHANGE OF MEXICAN AXOLOTL TO AN AMBLYSTOMA. 
367 
larvae that laid eggs exactly prove that even in reversion the power of 
reproduction may remain complete. From the universal causes of ster- 
ility mentioned above, it may be inferred that through those causes 
fruitfulness may be lost in different degree; and, further, it may in a 
measure be understood why the fruitfulness is more completely lost in 
the reversion to the Amblystoma form than in the reversion of the Tri- 
ton to the Perennibranchiate form. 
If in these cases the reversion is called forth by altering the con- 
ditions of life, one may perhaps imagine that the extent of this altera- 
tion will also fix the degree of fertility which the atavistic form can pre- 
serve; but still more will the fertility be influenced by the extent of the 
morphological skip which is made in the reversion. 
We know that the blending of widely varying constitutions (e. g., in 
the crossing of different species) produces sterility. Something similar 
takes place also in the sudden reversion to a stage of development widely 
varying in its whole structure. Here also takes place to a certain extent 
the union of two very different constitutions in one individual, a sort of 
crossing. 
From this point of view it may be, to some extent, understood why 
sterility may be a result of the reversion, but, on the other hand, we ob- 
tain no explanation why with the same breadth of morphological varia- 
tion there occurs in one case complete sterility, in the other relative 
fruitfulness. The extent of the morphological contrast is exactly the 
same between Axolotl and Amblystoma as between Triton and its 
‘‘sexually mature larvm.” The difference in the two cases of reversion 
lies entirely in the direction of the skip, which in the first case is made 
exactly in the contrary direction to that in the second. 
Just in that I would seek the reason for the varying force with 
which the power of reproduction is affected; not in the direction of the 
leap in and for itself but in the differences of the ontogenesis, which 
are conditioned upon the direction of the leap. 
The reversion of the Triton to an older pliyletic stage coincides with 
remaining upon a younger ontogenetic stage; or, in other words, the older 
stage of phylogenesis, to which the reversion takes place, is still entirely in- 
cluded in the ontogenesis of every individual. Every Triton is a Perenni- 
branchiate for a considerable portion of its life; the individual that 
makes the reversion simply reverts to an older pliyletic stage by re- 
maining in the larva stage of its individual development. 
Quite otherwise with the reversion of the Axolotl to the Amblystoma 
form, which had been once reached at some former time, but long since 
given up. This is not included in the ontogenesis of the Axolotl, hut has 
entirely fallen out. For a long succession of generations—so we must 
suppose—the ontogenesis has always proceeded only to the Perenni- 
branchiate form. If now single individuals are made to revert to the 
Amblystoma form, no greater leap in respect to morphology is made 
than in the reversion of the Triton to the Perennibranchiate form. At 368 
ON CHANGE OF MEXICAN AXOLOTL TO AN AMBLYSTOMA. 
the same time it involves a leap in an opposite direction, namely, a leap 
away over a long succession of generations, back to an animal form, 
which for a long time the species had not produced, which to a certain 
extent had become foreign to it. 
We should therefore have here also the coincident effect upon the 
Axolotl’s constitution of a widely different constitution, or—if one pre- 
fers—the blending of two widely varying constitutions. 
Of course I am very far from wishing to give this explanation for an 
exact one. It is nothing more than an attempt to point out the 
force in which we are to seek for the cause of the varying extent to 
which the power of reproduction is affected. To investigate more 
deeply and to demonstrate specially in what manner this force comes 
into operation must be left to a later time. For the present it must 
suffice to have shown in this connection that in general an essential 
difference exists between the two kinds of reversion, as well as to have 
made it in some measure comprehensible that this difference may be 
the motive force in relation to the question of sterility. Perhaps the 
law involved here may some day be formulated thus : Atavistic individ- 
uals lose the power of reproduction the more completely as the series of 
generations of their ancestors is longer whose ontogenesis no longer contains 
the phylctically older stage to which the reversion takes place. 
Consequently our hypothesis, which regards the transformation of the 
Axolotl as a reversion, offers us at once the possibility of learning to 
understand the sterility of the Amblystomas produced from Axolotls. 
So on the contrary the observed sterility of these Amblystomas is for 
those who think that a phyletic vital energy was here exerted—let it be 
named whatever it may—not only “ a veritable scientific enigma,” as 
DuarERiL expresses himself, but a complete parodox. Of such a design- 
ing, impelling principle it ought to be expected that it would bring forth 
new forms capable of life and not decayed to the point of dying out; and 
this so much the more since there is concerned a combination of struc- 
tural peculiarities, which, when it lias originated in another way (namely, 
from other species of Siredon), has already long ago shown itself capable of 
life arid reproduction. We kuow species of Amblystoma which reproduce 
as such, and every one of which comes from a larva like the Axolotl. One 
cannot, therefore, explain the sterile Amblystomas which the Paris Axo- 
lotl produces as an unsuccessful effort of the vital energy—an explana- 
tion, which to be sure, would be in and for itself sufficiently presump- 
tuous. 
But now it may be asked, What change in the conditions of life could it 
have been which made the Amblystoma in the Lake of Mexico* revert to 
the Siredon form ? To be sure, I can only answer this question with con- 
jectures, which can claim but a limited value so long as they cannot be 
sustained by a more exact knowledge of the circumstances there and 
the habits both of the Axolotl and the Amblystoma. 
* As we flo not know tlio origin of the Paris Axolotl, I must coniine myself in the 
following to Siredon Mcxicanus Shaw. ON CHANGE OF MEXICAN AXOLOTL TO AN AMBLYSTOMA. 
In general it may be presumed that the same external influences 
demanded the reversion which at an earlier time called forth the formation 
of the Perennibranchiate stage. 
In favor of this conjecture we may adduce the experiments here com- 
municated, for evidently it is the stimulus of the air breathing which 
brings the young Axolotls to the reversion to the Amblystoma form, 
i. e., the same stimulus under whose dominating influence the Ambly- 
stoma form must have arisen. 
But the case is quite similar with the season dimorphous butterflies. 
There a reversion of the summer brood to the winter form is most easily 
called forth by the operation of cold, i. e., by the same influences under 
whose control the winter form was developed. We know, at any rate, 
that reversion may arise also from the crossing of races and species, 
and I sought to show that reversion in butterflies may also be called 
forth by other influences than cold. But the most probable presumption 
is evidently the one that the reversion was induced through a recurrence 
of the same cause which to a certain extent produced the Perenni- 
branchiate form. That this form has been shaped under the influence 
of life in the water admits of no doubt; and so my conjecture is that the 
hypothetical Amblystoma Mexicanum, the supposed ancestral form of 
the present Axolotl of the Lake of Mexico, must have been brought to 
revert to the Perennibranchiate form by the circumstance that the possibility 
of going on land was taken from it and it icas constrained to stay in the 
water. 
I will not reject beforehand every other opinion. We must carefully 
distinguish between the mere provocations which are able to produce 
sudden reversion and actual causes of variation which have for their 
result, directly or indirectly, the molding of a, species. So it would 
not be inconceivable a priori that reversion should occur through the 
working of an impulse which has nothing to do with the origin of the 
phyletically older form. Certainly temperature has had no share, or 
only a very slight one, in the fashioning of the Perennibranchiate form. 
Yet cold, in and for itself, might quite well be one of the provocations 
which one day caused the Amblystoma form to revert to the Siredon 
form; and one could not a priori contradict De Saussure when he 
expresses the opinion that the low temperature of the Mexican winter 
must hinder the transformation (of the Axolotl to the Amblystoma) 
which had taken place “ in the hot reptile room” of the Paris Jardin des 
Plantes. He supports his view by the fact that “ Tschudi has found 
the Amblystoma” (of course another species) “in the hottest part of the 
United States.” “ On the plateau of Mexico, however, it snows every 
winter; and although the lake does not freeze, its temperature must sink 
very low near the surface.” 
But although no theoretical considerations oppose this view, yet I do 
not think it correct. I question very much that it was the temperature 
which induced the change from the Amblystoma back to the Axolotl, 
24 s 370 
on change of Mexican axolotl to an amblystoma. 
or—according to De Saussure’s explanation—which at the present 
hinders the transformation of the Axolotl in the Lake of Mexico. This 
is the reason for my doubt: From all portions of the United States to the 
northward of New York Amblystomas have been collected; a proof that 
even a much greater degree of winter cold than that found on the 
plateau of Mexico is no hinderanee for the transformation of the 
Axolotl; that the genus does not prove more sensitive in this respect 
than our native genera of Salamandrida. 
More consideration seems to me to be deserved by the following 
remarks of De Saussure, in which he points out the character of the 
Mexican lake : “ The bottom of this lake is flat, so that one comes im- 
perceptibly from the lake to wide, swampy regions before he reaches 
firm ground. Perhaps this condition makes the Axolotl unable to get 
to the dry laud, and so hinders the transformation.” 
At any rate the Lake of Mexico offers very peculiar conditions of life 
for an amphibious animal. My esteemed friend, Dr. Y. Frantzius, 
called my attention to it that this lake, like many others also of the 
Mexican lakes, is brackish. At the time of the conquest of Mexico 
by Ferdinand Cortez this circumstance caused the final surrender, as 
the Spaniards cut off the water from the besieged and the lake water 
is not drinkable. The old Mexicans had already built conduits from 
the distant mountains, and at the present day the city is still depend- 
ent on the water brought iu by aqueducts. 
Now, this saltness, in and for itself, could be no cause for the falling 
back to the Pereunibranchiate form, but might be such a cause, in 
connection with other peculiarities of the lake. The shallowest part of 
the lake is the eastern, and only in this part does the Axolotl live. In the 
winter violent storms from the east blow regularly and persistently, which 
come down from the mountains and drive the water before them so power- 
fully that it rises in the western part of the lake and frequently causes 
overflows there, while from the flat eastern shore the bottom is laid com- 
pletely dry for 2,000 feet.* Now if one puts together these two peculi- 
arities, the salt and the periodical drying of a part of the lake-bottom by 
continued winds, he gets at any rate conditions of life for the Axolotl 
such as can be found iu very few other places. To be sure, one might 
attempt to turn them to use just in an opposite sense, unfavorable to my 
theory, for the withdrawing of the water from a great part of the lake 
bottom ought—so might one think—rather to make easier the animal’s 
transition to living on the land; yes, just compel it thereto. But one for- 
gets, however, that the bared lake-bottom is a sterile plain, without food 
and without hiding-place; above all, without vegetation; and further, 
that through the pretty considerable saltness of the water (specific 
gravity, 1.0215) all the surface laid dry must be covered with a crust of 
salt, a condition which will make feeding on the land just impossible. 
# Muhlenpford't, Ycrsucli einer getreuen Schilderung tier Republik Meijco, Hanover, 
1S44,. ii, p. 252 (Attempt at a true Picture of the Republic of Mexico). ON CHANGE OF MEXICAN AXOLOTL TO AN AMBLYSTOMA. 
371 
Chloride and carbonate of soda, chiefly, are dissolved in the water in so 
sensible a quantity that they are regularly precipitated as a crust on the 
shore of the lake, and this crust is gathered there during the dry sea- 
son and comes into the market under the name of tequisquite. 
So, therefore, there is no lack of points of support for the conjecture 
that peculiar circumstances made life on the land more difficult to the 
animal than life in the water, and this alone might have been sufficient 
to bring it back to the habit of water-life only, and with that also to the 
reversion into the Perennibranchiate or Ichthyodea form. 
Yet a truce to conjectures. We cannot lament that from the great dis- 
tance and lapse of time, we are not in the condition to ascertain with 
definiteness the causes which compelled the Axolotl to give up the 
Amblystoma stage, so long as we are not able to solve the case of rever- 
sion that lies much nearer to us in the Tritons of Filippi and Jullien. 
Yet here, also, universal causes affecting the whole colony of Tritons 
must have been at the foundation, since, at least in Filippi’s case, the 
great majority of the individuals remained in the larva state. It must 
be that experiments with Triton larvae would bring greater clearness 
here; they would have to determine, before all else, whether the rever- 
sion can be called forth artificially, and, if this is the case, through what 
influences. 
According to the above-quoted experiences with butterflies, as well 
as according to the results attained with Axolotls, we should have to 
expect with the Tritons that the reversion to the Ichthyode form would 
occur if one would continue the stimulus of the water bathing the gills 
and the whole body, and at the same time would take away the stimu- 
lus under whose operation the Salamandrida form has been fashioned— 
the stimulus of the air bathing the gills, the skin, and the lungs. I 
hope at a later time to be able to report on experiments of this kind.* 
FTo one will wish to object to my hypothesis of reversion, that on one 
side it opposes what it of itself postulates on the other side: a sudden 
change of structure. The characteristic of the reversion is precisely in 
reaching at a bound an older, that is, an earlier existing phyletic stage. 
That this occurs is a fact; while the reaching at a bound, to express 
myself figuratively, of an aim (pardon the word) that lies forward has 
never yet been proved or even made probable. 
But as we succeeded in finding in the Axolotl’s present conditions of 
life-forces which make its life on the land difficult or quite impossible, 
and therefore show a motive for that return to the Ichthyodea form 
which seems to have taken place; so can the other side of my hypothesis 
*At any rate, Schreibers seems, in his essay already cited above, to have communi- 
cated experiments from which it follows, as Leydig recapitulates them in the place 
referred to, that the last change, i. e., the loss of the gills, “ may be delayed by forcible 
means.” To be sure, it does not follow from this that the animals of the experiment 
also became sexually mature at the samo time. Unfortunately I could not myself ex- 
amine the paper, as the volume of the Iris for 1833, as referred to, contains nothing 
of the kind, and I have lived for a long time at a distance from any large library. 372 
ON CHANGE OF MEXICAN AXOLOTL TO AN AMBLYSTOMA 
be supported by facts, namely, the assumption that the ancestors of the 
Axolotl had at an earlier time already been Amblystomas. 
We know from Humboldt* that the surface of the Lake of Mexico 
once lay considerably higher than at present, and tliat in a time com- 
paratively modern. We know, further, that the plateau of Mexico was 
covered with forest, while now the forest has vanished where the settle- 
ments of man, especially of the Spaniards, have reached. Now, if one 
may suppose that somewhere about the diluvial epoch the mountain 
forest extended to the edge of the lake, then still deep and with abrupt 
shores and containing considerably less salt, we have indicated condi- 
tions of life not only different essentially from the present ones, but also 
such as were quite specially favorable for the shaping of a species of 
the Salamandrida. 
In the light of all this I believe people will not be able to cast upon 
my attempt to explain the exceptional metamorphosis of the Axolotl 
from the Lake of Mexico the reproach of being too free a flight of fancy. 
At any rate it is the only possible explanation which can be opposed to 
that other one which assumes that the occasional transformation of the 
Axolotl is not reversion, but au effort to advance. And this assumption 
must, in my estimation, be rejected on purely theoretical grounds by 
every one who thinks a sudden transformation of species inconceivable, at 
least when it is joined tcith adaptations to new conditions of life. That as- 
sumption must be rejected by every one who looks upon adaptations not 
as the work of magic arising at a stroke, but as the final result of a long 
succession of natural causes, though separately slight and imperceptible. 
Should my interpretation of the facts be correct, this history of trans- 
formation would not have a significance so far-reaching as if it could 
have been taken in favor of heterogeneous creation, namely: in that 
case, demonstrating the existence of heterogeneous creation, it would 
have settled the question between that and transmutation. Now, on the 
contrary, it brings no definite decision, because, strictly taken, the 
refutation of sudden transformation in one case only proves it as not 
present for this one case. 
But it is, after all, a contribution to the gradual and complete rejection 
of such sudden transformation. If one case after another which seemed 
to speak for heterogeneous creation is proved untenable on that theory, 
the argument by induction must finally acquire sufficient strength to be 
acknowledged as satisfactory. 
If my view of the facts is correct, a few corollaries result from it which 
here at the close I should like to mention briefly. 
First, a thing that is more external: 
If the Siredon Mexicanus Shaw only assumes the Amblystoma form 
by occasional reversion, but never reproduces as such, but only as Sire- 
don, we cannot approve the action of the latest writers on systematic 
zoology, who simply strike the genus Siredon out of the system, and 
*See Miihlenpforclt’s book already quoted, vol. i. ON CHANGE OF MEXICAN AXOLOTL TO AN AMBLYSTOMA. 
373 
bring- in the Siredon Mexicanus as an unwelcome addition under the 
genus Amblystoma. So long as there are not one only, but several species 
of Siredon on the earth, which are regularly reproduced as such and only 
as such, so long the genus exists. And if we would not quite rob the 
systematic writers of the hope that some time these species of Siredon 
would rise to be Amblystomas, yet it corresponds better to the state of 
things now existing on the earth if we still allow the genus Siredon to stand 
among the genera of fish salamanders and reckon in it all those species 
which, like the Paris Axolotl, the Siredon Mexicanus Shaw, and proba 
bly also Siredon lichenoides, only take the Amblystoma form as an excep- 
tion or by artificial influences, but without reproducing themselves in it. 
On the other hand, all those species may probably be added to the 
genus Amblystoma which reproduce in this state, and in which the 
Perennibranchiate stage appears only as a larval condition. 
To make this distinction in the individual case will be chiefly the task 
of tbc American naturalists, from whose ever-increasing activity we may 
hope indeed for fuller details on the reproduction of the numerous spe 
cies of Amblystoma in their native land. I should rejoice if my expla- 
nations here presented should give an impulse to such investigations. 
The second corollary to which I referred is of a purely theoretical 
nature. It concerns an addition to the 11 fundamental law in the genesis 
of lifefi first set forth by Fritz Muller and Haeckel. It is well known 
that this is stated in the following proposition : The Ontogenesis contains 
in itself the Phylogenesis, more or less contracted, more or less modi- 
fied. Although the proposition cannot be rigidly proved, because we 
have no means of seeing the phyletic development directly unfolded be- 
fore our eyes, yet its correctness and general validity can be made so 
highly probable, in an indirect way, that few naturalists of the present 
time doubt it who have occupied themselves with the history of devel- 
opment and comparative morphology. 
Now, according to this proposition, every stage of the phyletic devel- 
opment, when it is displaced by a later one, must remain included in 
the Ontogenesis, and therefore come to light in the form of an ontoge- 
netic stage in the development of every individual. Now, with this my 
explanation of the Axolotl’s transformation appears to stand in contra- 
diction, for the Axolotl which had in former generations been an Am- 
blystoma contains nothing of the Amblystoma in its Ontogenesis. Never- 
theless the contradiction is only apparent. As soon as a further develop- 
ment is actually in question, and therefore the attainment of a new stage 
not yet realized, so soon the older stage is taken up into the Ontogen- 
esis. But it is not so when the new is not actually new, but has at a 
former time presented the final stage of the individual development, or, 
in other words, when there is a reversion, not of the single individual, 
but of the species as such to the preceding phyletic stage, and, therefore, 
a phyletic sinking back of the species. In this case the final stage of the 
Ontogenesis is simply eliminated, it falls out, and ice can only recognize its 374 
ON CHANGE OF MEXICAN AXOLOTL TO AN AMBLYSTOMA. 
presence by the fact that it may occasionally appear as a reversionary form. 
Thus the Triton, under some circumstances, sinks back to the Perenni- 
branchiate stage, but not in such a way that the individual would first 
become a Triton and then be transformed back to a Perennibranchi- 
ate, but, as I have already made prominent above, simply by no longer 
reaching the stage of the Salamandrida and remaining upon the stage 
of the Ichthyodca. Thus, also, according to my hypothesis, the Sala- 
mandrida that formerly lived on the shores of the Lake of Mexico, the 
Amblystoma Mexicanum, has sunk back to the stage of the fish-salaman- 
der, and the only trace which remains to us of bis former height of de- 
velopment is just the inclination, more or less present in every individ- 
ual, to reach under favorable circumstances the salamander stage again. 
But the third and last consequence which my explanation of the facts 
brings with it lies in the altered part which would be assigned by it to 
reversion in organic nature. Hitherto atavistic forms have been re- 
garded only as isolated, exceptional cases, interesting, to be sure, in a 
high degree for our knowledge, but without significance for the course 
of development of organic nature. Now a real importance would have 
to be allowed them in this latter regard. 
I should assume that reversion may in a twofold manner be a con- 
trolling power for the preservation or restoration of a form of life. In 
one case, as in the Axolotl, where the newer form, standing organically 
higher, becomes untenable from external causes, and now, as a further 
development in the other direction does not seem possible, instead of 
simply dying out, a reversion of the species to the older and less highly 
organized step follows. But, second, in this manner, that the older pliy- 
letic form is not altogether given vp, while the younger is developed from 
it, but that it alternates periodically until the younger, as we see in the 
season-dimorphous butterflies. One will hardly urge any objection to 
it if I regard the alternation of summer and winter form in these as a 
periodically occurring reversion to the phyletically older form (the winter 
form). 
Though the total reversion of a species, as I assume it for the Axolotl, 
may be a rarely-occurring case, the periodically or cyclically occurring 
reversion surely is not; it certainly plays a considerable part in the 
origin of various forms of the alternating or cyclical mode of reproduc- 
tion. 
It was intimated in the foregoing discussion that the causes from 
which I derived the reversion of the hypothetical Amblystoma Mexi- 
canuvn to the Axolotl of to day, did not seem to me to suffice completely 
for the explanation of the phenomenon. For one thing, they appeared 
to me of too local a nature, as they could only be applied with certainty 
to the Axolotl from the lake of the Mexican capital, while also the Paris 
Axolotl, coming from another part of Mexico, requires an explanation 
that will apply to him. But, on the other hand, they did not seem to 
POSTSCRIPT. ON CHANGE OF MEXICAN AXOLOTL TO AN AMBLYSTOMA. 
375 
me cogent enough. For should we even learn at a later time that the 
Paris Axolotl also comes from a salt lake, which is exposed to winds 
similar to those of the Lake of Mexico, yet there lies after all in these 
peculiarities of the lakes only a force which renders difficult the meta- 
morphosis of the larva and the gaining of a suitable new dwelling- 
place on the land. The impossibility of attaining such a dwelling, or 
indeed the total lack of it, does not necessarily result from them. 
Evidently it would be a much more substantial support for my 
hypothesis if I succeeded in pointing out forces in the physical relations 
of the country which entirely preclude the existence of Amblystoma there.