316 Dr. T. Margé on the Classification 
We think that we must speak in the same way of the 
system made known by M. A. Giard*. This zealous and 
otherwise distinguished zoologist, as is well known, took as 
the basis of his new classification of the animal kingdom the 
amnion, that is the embryonal envelope originating from the 
ectoderm, and so, according as the embryos possess or are 
destitute of such an envelope, divided the Metazoa into two 
great groups (Hymenotoca and Gymnotoca). But that the 
presence or absence of an embryonal envelope can by no means 
serve as a proof for or against genealogical relationship, and 
therefore cannot suffice for the establishment of a grouping of 
the animal kingdom resting upon natural affinity, will be at 
once seen by any one who will compare with each other all 
the different groups of animals, such as the Vertebrata, ‘Tuni- 
cata, Arthropoda, Echinodermata, Acanthocephala, ‘Trema- 
toda, Cestoda, Turbellaria, and Nemertina, which Guard 
united in the group of the Hymenotoca on account of the 
embryonal envelope observed upon their embryos, although in 
every other respect many of them stand very far apart. Obser- 
vation shows us, moreover, that even groups of animals, e. g. 
fishes, amphibia, reptiles, birds, and mammals, which demon- 
strably belong to one and the same natural stem, differ from 
each other in this respect, that while some of them (mammals, 
birds, and reptiles) during their embryonal existence are in 
possession of an amnion, others, on the contrary (Amphibia 
and fishes), are quite destitute of any thing of the kind. 
From this, however, we may evidently conclude that the 
amnion, or the embryonal envelope, where it is actually 
present, is to be regarded as originally not an inherited cha- 
racter, but one acquired by adaptation. As such it might 
originate, under the action of exactly similar conditions in 
quite divergent animals, even belonging to different stems, 
and quite independently of each other. It would therefore 
be very precipitate to conclude at once as to the true relation- 
ships of animals, in all cases, from the presence or non-presence 
of such a character. 
We must therefore regard all the different endeavours of 
systematists to group animals exclusively in accordance with 
one character, whether morphological, embryological, or bio- 
logical, external or internal, as mere experiments,—such a 
grouping or classification of animals can never be the true 
expression of their natural affinities. 
It was an error on the part of L. Agassiz to attempt to 
group the fishes according to the form and structure of their 
scales into cycloid, ctenoid, ganoid, and placoid fishes, or 
* ‘ Revue scientifique de la France et de l’étranger ’ (1876), no. 38, 
