SEPTEMBER 25, 1902 | 
interpreted. Taken in conjunction with other characters of 
little less importance, conspicuously those of the plastron and 
pelvis, this decides the question of affinity, and proves the 
Chelonia to have had a lowly ancestry, as has generally been 
maintained, : 
Recent research has fully recorded the facts of development 
of the rare New Zealand reptile Sphenodon, and it has more 
than justified the conclusion that it is the sole survivor of an 
originally extensive and primitive group, the Rhynchocephalia, 
as now understood. To confine our attention to its skeleton, as 
that portion of its body which can alone be compared with 
both the living and extinct, it may be said that positive proof 
has been for the first time obtained that the developing verte- 
bral body of the terrestrial vertebrata passes through a paired 
cartilaginous stage, and that in its details the later development 
of this body is most nearly identical with that of the lower 
Batrachia. There has long been a consensns of opinion that 
the forward extension of the pterygoids to meet the vomers in 
the middle line, known hitherto in this animal and the croco- 
diles alone, is for the terrestrial Vertebrata a primitive char- 
acter ; and proof of this has been obtained by its presence in 
all the Rhynchocephalia known. The same condition has also 
been found to exist in the Plesiosaurs, the Ichihyosaurs, the 
Pterodactyles, the Dicynodontia, the Dinosaurs, and with modi- 
fication in some Chelonians. It has, moreover, been found in 
living birds ; a most welcome fact, since Archzeopteryx, in the 
possession of a plastron, carries the avian type a stage lower 
than the Dinosaurs. It is pertinent here to remark that, inas- 
much as in those Dinosaurs (e.g. Compsognathus) in which the 
characters of the hind limbs are most nearly avian, the pelvis, 
in respect to its pubis, is at the antipodes of that of all known 
birds, and the fore-limb is shortened in excess of that of Archz- 
opteryx itself, the long supposed dinosaurian ancestry for birds 
must be held in abeyance. 
Passing through the Rhynchocephalia to the Batrachia, we 
have to countenance progress most definite in its results. * The 
skull, the limbs and their girdles are chiefly concerned, and this 
in a very remarkable way. 
In the year 1881 there was made known by Prof. Froriep, of 
Tiibingen, the discovery that the hypoglossus nerve of the embryo 
mammal is possessed of dorsal ganglionated roots. Again and 
again have I heard Huxley insist on the fact that the ventral 
roots of this nerve are serial with the spinal set, but never did 
he suspect the rest. It is, however, a most intensely interest- 
ing fact that, whereas by a Huxleian triumph the vertebral 
theory of the skull was overthrown, in these later Huxleian 
. days the proof of the incorporation.of a portion of the vertebral 
region of the trunk into the mammalian occiput should have 
marked the succeeding epoch in advance. The existence of 
twelve pairs of cranial nerves which all the Amniota possess 
involves them in this change ; and the fact that in all Batrachia 
there are but ten, enables us to draw a hard-and-fast line between 
batrachian and amniote series. 
It may be urged, as an objection, that since we have long 
been familiar with a fusion of vertebrae and skull in various 
piscine forms, the force of this distinction is weakened. But 
this cannot be; since, in respect to the investing sheaths and 
processes of development which lie at the root of the genesis of 
the vertebral skeleton, the fishes stand distinct from the 
Batrachia and Amniota, which are agreed. So forcible is this 
consideration that it behoves us to express it in words, and I have 
elsewhere proposed to discriminate between the series of terres- 
trial Vertebrata as arvchae- and syn-craniate. 
Similarly there is no proof that any batrachian, living or 
extinct (and in this I include the Stegocephala as a whole) 
possesses a costal sternum. So far as their development is 
known, the cartilages in these animals called ‘* sternal ” are either 
coracoidal or sa generis. The costal sternum, like the syn- 
craniate skull, is distinctive of the Amniota alone. Had the 
Stegocephala possessed it even in cartilage, there is reason to 
think it might have been preserved, as it has been in the colossal 
Mososaur Tylosaurus of the American Cretaceous. When to 
this it is added that whereas, in the presence of a costal sternum, 
the mechanism of inflation of the lung involves the body-wall, 
in its absence it mainly involves the mouth (as in all fishes and 
batrachians), the hard and sharp-line between the Batrachia and 
Amniota may be expressed by the formula that the former are 
archaecraniate and stomatophysous, the latter syncraniate and 
somatophysous. 
There are allied topics which might bz considered did our 
NO. 1717, VOL. 66] 
NATURE 
925 
time permit; but one certain outcome of this is that there is 
an end to the notion of a batrachian ancestry for the Mammalia. 
And when, on this basis, we sum up the characters demanded 
of the stock from which the Mammalia have been derived, we 
find them to be precisely those occurring outside the Mammalia 
in the Anomodont Reptiles alone. Beyond the sternum and 
skull, the chief characters are the possession of short and equal 
pentadactyle limbs, with never more than three phalanges to a 
digit, a complete fibula and clavicle, a doubly ossified coracoid, 
a heterodont dentition—a combination which, wholly or in part, 
we now associate with the Permian genera Procolophon, Paria- 
saurus, and others which might be named, the discovery of 
which constitutes one of the morphological triumphs of our 
time. . 
Beyond this, it may be added, concerning the Batrachii, that 
among living pedate forms the Anura have alone retained the 
pentadactyle state and the complete maxillo-jugal arch, an | that 
the Eastern Tylototriton, in the possession of the latter, 
becomes the least modified urodele extant. These facts lead to 
the extraordinary conclusion that the living Urodela, while of 
general lowly organisation, are one and all aberrant; and it is 
not the least important sequel to this that, despite their total 
loss of limbs, the Apoda, in the retention of the dermal armour 
and other features which might be stated, are the most primitive 
Batrachia that exist. 
The batrachian phalangeal formula 22343 was until quite 
recently a difficulty in the determination of the precise zoological 
position of the class; but it has now been overcome, by the 
discovery of a Keraterpeton in the Irish Carboniferous having 
three phalanges on the second digit of both fore- and hind-limbs, 
and by that in the Permian of Saxony of a most remarkable 
creature, Sclerocephalus, which, if rightly referred to the 
Stegocephala, had a head encased, as its name implies, in an 
armature like that of a fish, and the phalangeal formula of a 
reptile, 23454. 
Passing trom the Batrachia to the Fishes, we have still to 
admit a gap, since an interminable discussion on fingers and fins 
has not narrowed it in the least. In compensation for this, 
however, we have to record within the fish series itself 
progress greater, perhaps, than with the higher groups. Cer- 
lainly is this the case if, as to bulk, the literature in systematics 
and palzeontology be alone taken into account. 
Ot the Dipnoi our knowledge is fast becoming complete. We 
know that Lepidosiren forms a burrow ; and, in consideration 
of a former monstrous proposal to regard this animal, with its 
fifty-six pairs of ribs, and Protopterus, with its thirty to thirty- 
five, as varieties of a species, it is the more interesting to find 
that the Congo has lately yielded a Protopterus (?. Do/Joz) with 
the lepidosiren rib formula, viz., fifty-four pairs. 
As a foremost result of American palzeontological research we 
have to record the occurrence, in the Devonian of Ohio, of a 
series of colossal fishes known as the Arthrodira, the supposed 
dipnoan affinities of which are still a matter of doubt. 
We have evidence that the osseous skeleton in a plate-like 
form first appeared as a protection for the eye of a primitive 
shark. And coming to recent forms having special bearings on 
the teachings of the rocks, we have to acknowledge the capture 
in the Japanese seas of a couple of ancient sharks, of which 
one (Cladoselachus), since observed to have a distribution ex- 
tending to the far North, isa survivor from Devonian times ; 
the other (Mitsukurina) a genus whose grotesqueness leaves no 
doubt of its identity with the Cretaceous lamnoid Scapano- 
rhynchus. In the elucidation of the Sturiones and the deter- 
mination of their affinities with the ancient Palzeoniscide a 
master stroke has been achieved. In the Old Red genus 
Palzeospondylus we have become familiar with an unmistakable 
marsipobranch, possessing, as do certain living fishes, a noto- 
chord, annulated, but not vertebrated in the strict sense of the 
term. The climax in Ichthyopalzeontology, however, has been 
reached in the discovery of Silurian forms, which, there is every 
reason to believe, explain in an unexpected way the hitherto 
anomalous Pteras- and Cephalaspidians, by involving them ina 
community of ancestry with the primitive Elasmobranchs. The 
genera Thelodus, Drepanaspis, Ateleaspis and Lanarkia chief 
among these annectant and ancestral forms, are among the most 
remarkable vertebrate fossils known. 
Passing to the Recent Fishes alone, the discovery which must 
take precedence is that of the mode of origin of the skeletogenous 
tissue of their vertebral column. The fishes, unlike all the 
higher Vertebrata, have, when young, a notochord invested in 
