ON 
SYMMETRY AND HOMOLOGY 
IN LIMBS. 
U Y 
.1 E FFRIES WYMAN, M. D. 
From the Proceedings of the Boston Society of Natural History, Vol. XI 
June 5, 1867. 
BOSTON: 
A. A. KINGMAN, 50, B BOM FIELD STREET. 
1 867. ON 
SYMMETRY AND HOMOLOGY 
IN LIMBS. 
BY 
JEFFRIES WYMAN, M. D. 
From the Proceedings of the Boston Society of Natural History, Voi. XI., 
June 5,1867. 
BOSTON: 
A. A. KINGMAN, 50, BROMFIELD STREET. 
1 86 7. SYMMETRY AND HOMOLOGY IN LIMBS. 
SYMMETRY. 
Anatomists who have compared the fore and hind limbs of man 
and animals, have mostly described them as if they were parallel rep- 
etitions of each other, just as are any two ribs on the same side of the 
body. By a few they have been studied as symmetrical parts, repeat- 
ing each other in a reversed manner from before backwards, as right 
and left parts do from side to side.* We have adopted this last mode 
of viewing them, because, though open to grave objections, as will be 
seen further on, the difficulties met with are, on the whole, fewer than 
in the other, and because too, it is supported by the indications of 
fore and hind symmetry in other parts of the body. 
*The following are among the more recent articles in which the homologies of 
the limbs and their symmetry are treated of at length. 
S. R. Pittard. Article Symmetry—Cyclopaedia of Anatomy and Physiology, Vol. 
iv, p. 845. 
Observations on the Limbs of Vertebrate Animals. By George Humphrey, M. D., 
F. R. S., etc., Cambridge, 1860. 
Charles Martins. Nouvelle Comparaison desMembres. Memoires de I’Acad. des 
Sciences de Montpellier T. n, p. 461. 1857. 
Also by the same author, Memoire sur I’Osteologie Comparde des Articulations 
du Coude et du Genou. Ann. des Sciences Naturelles. T. win. 4me s6rie. 1862. 
Homologies des Membres Pelviens et Thoraciques de I’Homme, par le Docteur 
Foltz. Journal de Physiologie, T. vi, 1863, p. 49. 
On Morphology and Teleology, especially in the limbs of Mammalia. By Burt 
G. Wilder, S. 8., Cambridge, 1865. From the Memoirs of the Boston Society of 
Natural History. 
Also by the same author, Morphological Value and Relations of the Human Hand. 
Am. Journal of Science, Vol. xx.iv, July, 1867, p. 44. 4 
Among animals, two organs or parts, generally speaking, are said to 
be symmetrical when they are situated on opposite sides of an axis, 
and are alike in form and size, but one is the reverse of the other, as 
is everywhere obvious in those which are right and left. It is not to 
be understood, however, that this likeness is absolute; for while it is 
very generally true that such right and left parts are alike in size and 
form, or very nearly so, it occasionally happens that they are very un- 
like in these respects, still retaining, however, a certain amount of 
symmetry. We have striking illustrations of this in the claws of many 
Crustaceans, as in Astacus and Gelasimus, and in the right and left 
halves of the body of Bopyrus. Among Acephalous Molluscs, this 
difference is in some cases very remarkable, as in Radiolites, Mono- 
pleura, etc., where one of the valves is conical and spirally twisted, 
while the other is quite flat, and in its relation to the other valve, re- 
sembles the operculum of certain Gasteropods. 
Among Vertebrates such differences are much less frequently met 
with; they however exist, and symmetry of right and left parts, even 
in the human skeleton, is not constant; it may even be doubted 
whether absolute symmetry exists anywhere. Attention has recently 
been called to asymmetry in the base of the skull, and by a compari- 
son of the bones of the fore arm in ten skeletons, we have found the 
right ulna longest in eight, the left in one, and in one the right and 
left bones were equal. Differences in length and weight were also 
found In the clavicles, the humerus, and other parts of the skeleton, 
but in different degrees. The close approach to absolute symmetry in 
some of the minuter details of structure is sometimes quite remark- 
able, as in the arrangement of the papillas on the tips of the fingers 
and toes of most individuals \ in some, however, the asymmetry of 
these parts is quite marked, and in others even the pattern of the 
figure on the two sides is changed. More marked instances of asym- 
metry exist in the unequal nosti’ils, and more or less bent vomer of 
some Cetaceans, in the rudimentary right, but immensely developed 
left tusk of the Narwal, in the lower jaw of the adult male Mastodon 
which has a tusk on the left side, but none on the right, in the unequal 
development of the ovaries and oviducts of most birds, as also in a 5 
similar condition of the carotid arteries. In the human brain the 
hinder lobe of one of the hemispheres, more commonly the left, is 
longer than the other, as may easily be seen in the cranium by the 
depth of the corresponding fossm. In the illustrious Bichat, the in- 
equality of the two sides of the head amounted to deformity. The 
most striking instances of all are to be found in the halibut (Hippo- 
glossus), flounders (Pleuronectes), and other flat fishes, in which the 
bones of the face, the brain and organs of sense are all more or less 
distorted, and the eyes especially are unsymmetrically placed. 
All organs which are thus unlike or unsymmetrical in the adult, 
have been shown by embryologists, as by Steenstrup in the case of 
the flat fishes, and also in the crabs, molluscs, etc., by other observers, 
to be alike in the embryo, the deviations from true symmetry taking 
place as development advances. We have seen a lobster nearly three 
inches in length, in which the right and left anterior claws were still 
symmetrical. 
The organs of the great cavities in 
adult vertebrates are almost uniformly 
unsymmetrical; nevertheless in the em- 
bryo the symmetry of these parts, even 
of the liver, is complete. In some 
fishes this is true of the liver of the 
adult, and in a few instances this is 
divided into a right and left organ. 
The fundamental idea of the organs of 
organic life involves the condition of 
symmetry. The other kind of sym- 
metry, viz., that which is believed to 
exist between the fore and hind parts 
of the body is much less obvious, and 
would in fact be generally overlooked, 
the deviations from true symmetry be- 
ing so great. In many Articulates, however, this deviation is com- 
paratively slight, especially in the genera Jaera (Fig- 1), Oniscus, 
Fig. 1. 6 
Porcellio, Asellus, Cymothoa, and other Isopods, also among Myria- 
pods, as in the genera Scutigera, Scolopendra, etc., in -which the limbs 
are repeated oppositely, though with different degrees of inequality, 
from the centre of the body backwards and forwards. If to the 
general symmetry of such fore and hind parts we add certain details 
of structure occasionally seen, especially the fact that some of the 
worms and Crustaceans have organs of special sense developed in 
the last as well as in the first segments, the evidence that the fore 
parts are repeated in the hinder becomes much stronger.* Among 
Vertebrates, as will be seen further on, the resemblance between the 
fore and hind limbs is quite obvious, and the symmetry of plan easily 
recognized ; but in the majority of animals, whether vertebrate or 
invertebrate, fore and hind parts, though symmetrical in plan, actually 
differ largely from each other, both in size and form, and thus present 
a distorted symmetry like that which has already been noticed be- 
tween right and left parts. The analogy holds still further, since the 
fore and hind limbs, however widely they may differ in the adult, are 
as nearly alike in the early embryo as are corresponding limbs on the 
right and left sides. In right and left parts, however, distorted sym- 
metry is the exception, while in the fore and hind parts of adults it 
is the rule. 
A sufficient explanation of this deviation from complete symmetry 
is found in the circumstance that in right and left parts the functions 
are generally similar and equal, while this is seldom the case in fore 
and hind ones. Identical and homologous parts having similar and 
equal functions will have equal growth and development, while the 
reverse will be the case in those having the opposite conditions. A 
close approach to fore and hind symmetry of limbs'is found in certain 
swimming animals, as the Ornithorhynchus and Ichthyosaurus, while 
extreme asymmetry is found in birds, in which aerial locomotion be- 
longs to the arms, and land or aquatic locomotion to the legs, or as 
* Leuckart and Van Beneden have shown that My sis has an ear in the last seg- 
ment, and Schmidt has described an eye in the same part in Amphicora, a worm 
Nat. Hist. Rev., April, 1862, p. 133. See also Quatrefages, Memoire sur la Famille 
des Polyophthalmiens, Ann. des. Sc. Nat. 3me serie, T. xm, p. 6. in the kangaroos and other jumping animals, where the hind limbs 
predominate so largely over the fore ones, and in the apes and three- 
toed sloths, where the reverse is the case. Similar extreme differences 
are still better indicated by the animals whose tracks are left in the 
Connecticut River sandstones, as for example, in Anomoepus and 
analogous forms. 
The facts brought to light by the study of embryos, offer additional 
evidence in support of the view that the fore and hind portions of the 
body are in idea symmetrical. As already stated, this is the more 
noticeable the nearer the embryo is to the earliest stage of its develop- 
ment ; and it is to this that attention should be carefully turned, for 
as the embryo becomes more specialized, and its organs take on those 
forms which adapt the individual to its future conditions of life, the 
differences rapidly increase. 
In the general development of the ver- 
tebrate embryo, the first fact which strikes 
us, as it increases in size, is, that this in" 
crease is not from a growth from before 
backward, but from a central, and, as it 
were, a neutral point, both backwards and 
forwards, so that the two ends are made 
to recede from the centre in opposite di- 
rections, just as do the radicle and plum- 
ule of a plant from the point where these 
are continuous. By this process the head 
and tail both become free, while the cen- 
tral part of the body remains attached to the yelk. Secondly, the 
primitive groove of the nervous axis in its earliest stage (Fig. 2) is 
nearly symmetrically enlarged at either end, so as to form two op- 
posite dilatations :* one the precursor of the future cerebral vesicles, 
Fig. 2. 
* In some adult fishes the spinal marrow ends in a ganglionic enlargement, form- 
ing a kind of caudal brain. We have found such a ganglion quite conspicuous in 
the American Lophius, and Quatrefages describes one in Amphioxus, which is all 
the more striking, since it is only a diminutive repetition of the fore end of the 
axis. The distribution of the last pair of nerves, as seen in his admirable figures, 
is also symmetrical with that of the first. 8 
and the other of the rhomboidal sinus, which last has only a temporary 
existence in the mammals, but is permanent in the birds. Thirdly, 
when the spinal groove closes up by the union of the dorsal laminse, it 
does so, as Reichert has shown, by the union of its lips, first in the 
middle portion, and then gradually in a symmetrical manner towards 
either end. Fourthly, in the four classes of vertebrates the first traces 
Fig. 3. 
Fig. 4. 
of the vertebral segments are to be found in three or four pairs of 
plates which appear on either side of the primitive axis midway be- 
tween the two ends, from which region they are multiplied and ex- 
tended forwards to the head, and backwards to the tail, by the devel- 9 
opment of new plates, thus lengthening the column in a symmetrical 
manner (Figs. 3, 4). The ossification of the bodies of the vertebras 
takes place in the same order, beginning in the middle and extending 
in either direction. During the first half of foetal life, and even be- 
yond it, in the human body, the bodies of the vertebrae first ossified, 
viz., those in the middle, are the largest, and from these the column of 
bodies gradually tapers to the head and coccyx. It is only towards 
the end of foetal life that the lumbar vertebrae assume larger propor- 
tions. Fifthly, a resemblance analogous to that which exists between 
the opposite ends of the nervous system and of the vertebral column, 
can also be traced in the intestinal canal. Oken first maintained the 
idea that the oral and anal portions of this canal repeat each other. 
Notwithstanding the ridicule which has been directed to this view of 
his, fairly examined, it will be found to have, underlying it, at least 
the semblance of a truth. The two opposite ends agree in this, that 
in the embryos of all air-breathing animals there is developed from the 
abdominal side of each end of the straight symmetrical intestine a sac ; 
that in front forming the lung, and that behind the allantois, and each 
opening into the intestine by a narrow neck. The allantois as well as 
the lung is a respiratory organ, but it is not justifiable to cite a function 
as an indication of homology. The correspondence between the two 
ends of the canal is still further, but less clearly, indicated in the sub- 
sequent division of them, the fore end into mouth and nostrils, or 
respiratory and digestive portions, and the hind into anal and genito- 
urinary portions, and still further by the development of the tongue 
on the floor of the mouth, and of the male organ on that of the geni- 
tal portion of the intestine. In most mammals the genito-urinary por- 
tion becomes wholly separated from the digestive, while the respiratory 
does not. Nevertheless in the embryo the first retains its connection 
for a certain time, as is permanently the case in the birds and reptiles. 
The most striking facts bearing upon the idea of fore and hind sym- 
metry are to be found in the development of the limbs. Yon Baer 
has described the phases which the limbs assume, as also their attitudes 
as they advance towards their permanent condition. Ye have traced 10 
these successive positions in several embryos, and from them the fol- 
lowing description is drawn. The limb-buds, when first formed, are 
simply tegumentary outgrowths, and project at right angles from the 
sides of the body in the form of half oval discs. As they increase in 
length they become divided into a somewhat flattened, disc-like end, 
which becomes the future hand or foot, and a pedicle which connects 
this with the trunk. This pedicle is transformed into the fore arm and 
leg, and partly into the arm and thigh; the remaining portions of 
these last, when developed, result from a still further outgrowth from 
the body which takes place at a later period. In the second stage the 
limbs are each bent to an angle at a point near the trunk, so that 
their ends are directed downwards, and what were previously the under 
sides of the disc-like hands and feet (the body being supposed to be 
horizontal), have now become vertical and face inwards ; they are the 
soles and palms. The angle formed in the limbs corresponds with the 
elbows and the knees, and thus marks off fore arms and legs from arms 
and thighs, these last being very short. Thirdly, during the stage just 
mentioned, both upper arms and thighs projected at right angles from 
the sides of the body. They now begin to change their direction with 
reference to the trunk, but in a perfectly symmetrical manner. The 
elbow which has thus far projected outwards, now swings backwards 
to the side of the thorax, and the knee forwards towards the side of 
the abdomen. This condition of things is readily seen in a human 
embryo in the writer’s collection, measuring 0.60 inch in length, and 
represented in Fig. 5. The effect of this change of position would be 
to make the palms face forwards, and it 
makes the soles face partly backwards. 
This tendency, however, as regards the 
hand, is counteracted by a remarkable 
change of position in the fore arm. This 
is slowly rotated inwards, so that the palm 
is made to face almost backwards. In 
those animals which walk on the palms 
of their hands, the hand is bent on the fore arm, so that its back is 
raised, the palm directed more or less downwards, and thus the limb is 
Fig. 5. 11 
adapted to walking or crawling. At tlie same time a similar change 
takes place in the feet, the backs of which are raised towards the fore 
part of the leg, and the toes directed forwards, and the sole down- 
wards. The rotation of the fore arm just mentioned is the chief cause 
of the interference with the symmetry of the limbs ; were it not for 
this the hands and feet would have assumed exactly symmetrical atti- 
tudes, the toes projecting forwards and the fingers backwards, the 
back of the foot looking towards the fore part of the leg, and the 
back of the hand towards the hind part of the fore arm. 
If we admit the idea of symmetry in structure between arms and 
legs, and would compare the movements of the two In man and ani- 
mals, we must change in some respects the terms flexion and extension, 
from those ordinarily used in the description of the human body. We 
will suppose the human skeleton suspended with the vertebral column 
horizontal, the limbs slightly flexed, the toes and fingers pointing 
downwards, the palms facing forwards and the soles backwards. Flex- 
ion of the humerus would be backwards, of the femur forwards ; of the 
fore arm forwards, of the leg backwards; of the hand backwards, that 
is by carrying the back of it towards the back of the fore arm, and 
the foot forwards. Thus the movements would be symmetrical through- 
out in the two limbs. Supposing the limbs to be of equal strength, 
somewhat flexed, the soles and palms resting on the ground, they 
would antagonize each other in their action; the fore limbs, if ex- 
tended, would, in consequence of their obliquity, tend to push the 
body upwards and backwards, and the hind limbs under the same 
conditions upwards and forwards. The two acting together would 
give rise to a resultant motion upwards. By the rotation of the fore 
arm in the embryo the action of the fore limb is reversed, and thus in 
the forward movement of the body cooperates with, instead of antag- 
onizing, the hind limb. 
When fully developed, therefore, the fore and hind limbs of ani- 
mals have a general symmetry except in the following respects ; the 
bones of the fore arm cross each other, while those of the legs do not, 
and the toes and fingers are both directed forwards. If, however, the 
fore arm be rotated outwards through half a circle, so as to make 12 
the bones parallel, as in the leg, thus counteracting the change which 
took place during development, the general symmetry would be 
restored, and be complete. There would, however, exist a certain 
amount of special asymmetry in the position of the thumb and great 
toe, for these would be on opposite sides of the two limbs. Of this 
difficulty we shall speak again further on. 
The symmetry of disease to which attention has of late years been 
called,* also helps to sustain the idea of fore and hind symmetry. Cer- 
tain maladies, as psoriasis, leprosy, syphilis, etc., not only attack cor- 
responding or symmetrical portions of right and left parts, but also of 
fore and hind parts. Certain skin diseases attack the backs of the 
hands and feet, or the palms and soles, or the elbows and knees. The 
earthy deposits in the arteries show a similar tendency to symmetrical 
distribution. Such instances, however, are quite rare in comparison 
with the vast proportion of diseases in which no such tendency is ap- 
parent. They nevertheless tend to show that homologous parts, either 
on the right and left, or fore and hind parts of the body, have such a 
constitution that they are more amenable to the influence of a given 
disease than other parts. 
We pass by only with a mention a third kind of symmetry, which 
has been much insisted on by some anatomists, viz., that between the 
dorsal and abdominal parts. Under certain circumstances this kind 
of symmetry becomes almost exact, as may be seen, for example, in a 
vertical section of the tail of a fish, where the arrangement of the 
bones, etc., below a horizontal line, passing through the vertebral col- 
umn, is only a reversed copy of the parts above. 
The statements which have been made in the previous pages are 
intended to show that even in right and left parts, symmetry is of vari- 
ous degrees, and is rarely, if ever, absolute. Asymmetry, however, 
is in most cases slight, but may in certain others become as great as 
that between fore and hind limbs, in accordance with the degree of 
* William Budd. Medico-Chirurgical Transactions, .London. Yol. xxv, 1852. 
James Paget. Surgical Pathology. Philadelphia. 1860. p. 27. 
Burt G. Wilder. Pathological Polarities, or What has heen called Symmetry 
of Disease. Boston Medical and Surgical Journal, April sth, 1866. difference in the function of the two sides. These differences do not, 
nevertheless, prevent our recognizing the idea of symmetry of plan 
underlying the structure of such parts. In comparing fore and hind 
limbs, it has been shown that they are sometimes nearly symmetrical, 
but generally the symmetry is largely distorted. But if we bear in 
mind the fact that the limbs which in the adult are the most unsym- 
metrical, are quite symmetrical in the embryo, the hypothesis that the 
idea of symmetry underlies their structure is rendered highly prob- 
able. ANALOGY BETWEEN SYMMETRY AND POLARITY. 
From what has been stated it is obvious that in the early stages of 
development there is at work a force which regulates the distribution 
of the particles of matter out of which the embryo is formed in a 
symmetrical manner, and that up to a certain stage there is sym- 
metry, not only of right and left, but of fore and hind parts. The 
essential characteristic of this force is that it gives rise to similar but 
reversed forms on the two sides as well as on the two ends of the axis 
of the body. 
If we look for any thing among known forces analogous to this 
force, it is to be found, if anywhere, in those known as polar forces. 
The essential features of polarity, as of symmetry, are antagonism, 
oppositeness or inversion, either of qualities or forms. Studying the 
subject from the most general point of view, there are striking resem- 
blances between the distribution of matter capable of assuming a polar 
condition, and free to move around a magnet, and the distribution of 
matter around the nervous axis of an embryo. 
In every complete series of magnetic curves formed by particles in a 
polar condition, there are two neutral lines (Fig. 6, A), one extending 
lengthwise of the magnet, so that the curves formed may be divided 
into right and left; secondly a transverse one, the particles on each 
side of which form the north and south curves, or which for purposes 
of comparison might be called fore and hind curves. In the right and 
left series those which are on one side of the long axis are symmetrical 
with those on the other and not in themselves, and in the north and 
south series those on either side of the transverse neutral line are sym- 
metrical with each other, and not in themselves. 15 
If these curves are projected on paper, and this be folded on the 
line of the longitudinal, or north and south axis, the cuxwes of opposite 
sides or opposite ends will correspond as right and left hands or other 
double organs do when applied to, or placed opposite each other. 
The same is true of the north and south curves when the paper is 
folded on its transverse axis. 
Fig. 6. 
The distribution of particles just described, corresponds, first, to all 
that we designate as right and left in normal development; second, to 
all that we designate as fore and hind with reference to the long axis 
of the body, and which is characterized by symmetry in structure. 
Not only is there this analogy between the distribution of matter 
around a magnet, and that around the nervous axis of the normal 
embryo, but the analogy is still more striking in the curves formed by 
the combined action of two adjoining magnets and the appearances 
found in more or less double monsters. 
If two magnets are placed parallel to each other (Fig- 6, B), and at a 
distance, two sets of curves are formed as in the usual way; but if they 
are brought so as to be within each other’s influence, the two magnetic 
figures are combined, and now form a single compound one, the mid- 
dle portion of which consists of the curves from the two adjoining 
sides of the magnets; and the particles from either series of curves do 16 
not pass beyond the line where the forces of the two magnets are in 
equilibrium, but are deflected upwards or downwards, north or south. 
In this manner that portion of the figure formed by the particles ar- 
ranged between the two magnets becomes symmetrical, one half 
consisting of particles belonging to the right magnet, and the other 
half of those belonging to the left. The symmetry of the whole 
compound figure thus formed is in all respects as perfect as that of 
the ordinary figure from a single magnet. If such a compound figure 
is projected on paper, and this be folded on either the longitudinal or 
transverse axis of the whole figure, the opposing halves will corres- 
pond. The right and left curves belonging to the same magnet will 
not now be symmetrical with each other, but all the curves formed by 
one magnet will be symmetrical with those formed by the other. 
In abnormal development, if two nervous axes are formed on oppo- 
site sides of one and the same yelk, each axis, or rather the symmet- 
rical force, the axis of action of which corresponds with the nervous 
axis, will distribute the organic matter under its influence without 
coming In collision with that of the other, except near the umbilicus, 
so that two nearly perfect embryos will be formed, as In the case of 
the Siamese twins. 
If the two axes are formed side by side, and so near to each other 
that the particles under their respective influence come in contact, 
then at the line of contact a series of intermediate organs or limbs, 
will be formed in connection with the two axes, as is commonly seen 
in those double monsters which present what Is called lateral doubling. 
Here, too, as in the case of the double magnetic curves, the right and 
left parts connected with one and the same nervous axis will no longer 
be symmetrical with each other, but those connected with the right 
half of one axis will be symmetrical with those connected with the 
left half of the other, or in other words, the two bodies thus united 
will be symmetrical with each other, but will not be bilaterally sym- 
metrical in themselves. 
If the magnets are now inclined towards each other (Fig. 6, C), so 
as to touch at one end, forming a Y, then we shall have the particles 
arranged so as to produce a figure double at one end, but single at the 17 
other. The ends of the magnets which are separated will arrange 
their particles so as to form a double series of curves, which are com- 
pletely independent of each other; but as the magnets come nearer 
together the intervening curves are gradually modified or suppressed, 
and at last a single symmetrical terminal figure is formed. One half 
of this terminal portion is however formed under the influence of the 
right, and the other of the left magnet. 
Likewise in certain double embryos if the axes are inclined so as 
to form a Y-shaped figure, the two separated ends will have a head 
more or less complete, but as the two axes converge below, the or- 
gans become more or less fused and suppressed, and at length the 
hindmost are reduced to the normal type, all intermediate ones having 
become obsolete. The body will be provided with two legs, but one 
will be connected with, and under the control of the right axis, and 
the other under that of the left. Precisely such a case is found in 
the well known instance of Ritta-Christlna. The completeness of the 
intermediate organs in a given instance, as in the case of the inter- 
mediate magnetic curves, will depend upon the degree of divergence 
or separation of the axes. 
In comparing the results of the symmetrically acting force in ani- 
mals with a polar force like that of magnetism, it is not intended to 
imply that the two forces are the same, but only that they have like 
modes of acting, and that when left to themselves undisturbed by 
other influences, each tends to produce symmetrical figures. The 
type or general idea of any of the double monsters may be imitated 
by the combined action of two magnets. 
In the preceding paragraphs we have spoken of the symmetrical 
distribution of the particles of matter around the nervous axis as if 
the distributing force emanated from this axis. It is not to be inferred 
that such is actually the case, since the force is already in action be- 
fore the nervous axis itself is formed, and may be said to be manifested 
in the first stage of the segmentation of the yelk ; tor when the whole 
yelk, or as in some cases, a limited portion of it, divides into two dis- 
tinct segments, such division shows that a symmetrically acting force is 
already present; in fact, we have now right and left parts. Such force 18 
is again manifest when a new division takes place at right angles to 
the first, separating each right and left portion into two others, which 
may be compared to the fore and hind divisions of the body; even the 
nervous axis itself is symmetrically developed under the influence of 
this same force. The nervous axis, however, when formed, coincides 
with the axis of the symmetrically-distributing force. 
The use of the term polarity in connection with organic structures 
has long been familiar to physiologists, but apparently with very varied 
signification. Oken in his celebrated “Programm” uses the following 
words with regard to the skeleton : “This skeleton repeats itself at the 
two poles ; each pole repeats itself in the other, and they are head 
and pelvis.” It does not appear from this, nor from any other state- 
ment of his that we have seen, precisely what he understood by the 
word pole. At the present day few will agree with him that the head 
and pelvis repeat each other. Still, although these parts are not com- 
parable, the idea underlying his statement, viz., that the two ends of 
the body do repeat each other, may be, and we believe is, correct. 
In order, however, that they may be repeated as if under the influ- 
ence of a polar force, or of a force acting in a manner analogous to 
one, there should be a more or less symmetrical repetition of homolo- 
gous parts. 
The term pole is often used in the description of eggs and of cells; 
in the former to distinguish the portion of the egg where the oily mat- 
ters are collected from the opposite side where there is only albumen, 
and in the latter, simply to designate the two ends without intending 
thereby to imply any difference of quality or force. In the nervous 
system cells are described as “unipolar,” “bipolar,” or “multipolar,” 
which only means that they are prolonged into one or more points, 
“caudate appendages,” or processes which connect them with the nerve 
tubes. 
Owen appears to make use of the term polarity in the sense in 
which it is made to stand as representing the quality of a force acting 
in animal bodies, and producing symmetrical results. After compar- 
ing the dorsal and abdominal portions of the vertebral arches, and 
showing that they repeat each other, he says “symmetry and polarity, 19 
or serial homology of the parts of the same vertebral segment is usually 
still more strictly observed in the transverse direction, and is so obvi- 
ous as to have immediately led to the detection of the homologous 
parts, which are accordingly distinguished as right and left.”* He 
does not, however, recognize the symmetry of fore and hind parts. 
Prof. Dana uses the term in the same manner. “An animal is em- 
bodied or concentrated force, which force manifests polarity in the 
results of its action in development, that is in the oppositeness of the 
anterior and posterior extremities of the structures evolved, and also in 
the dorso-ventral relation of these structures.” f 
Polarity, according to Mr. Faraday, may be considered as “an axi 
of power, having contrary forces exactly equal in opposite directions.” 
This power will produce perfectly symmetrical figures, however, only 
when wholly free from the influence of a superior force. In the pres- 
ence of such it may be interfered with, or have the results of its ac- 
tion so changed as to give rise to forms whose symmetry is more or 
less distorted, as is so frequently the case in crystals; these, we are 
told, being exactly symmetrical only in idea. 
In the vertebrate animal, the plan of construction appears to be in 
accordance with the idea of general symmetry, which in the early 
stages of development is maintained at the two ends as well as the 
two sides of the axis, but subsequently is more or less interfered with to 
adapt the animal to its special conditions of life. This force, produc- 
ing symmetry, acts in a manner analogous to a polar force. To desig- 
nate the disturbing, or rather adapting force, by which the force 
tending to act in a symmetrical manner is interfered with, we must 
still retain the term vital or life-force, although this may in the end 
prove only a physical force acting under special conditions. 
* Archetype and Homologies of the Vertebrate Skeleton. London, 1848. p. 166. 
t Am. Journal of Science. Yol. xxvxi, p. 157,1864. See also Vol. lxi, p. 164, 
1866. See also Wilder op. cit. p. 9. HOMOLOGY. 
Owen defines a homologue to be “the same organ in different 
animals under every variety of form and function.” * When parts 
are repeated in the same animal, not from right to left, but from 
before backwards, either on the middle line of the body, as the ver- 
tebra; or sternal pieces, or on the same side, as the ribs, such parts 
are homologous, but not in the same sense as when they are re- 
peated in different animals; In the first case he calls them homotypes 
and in the second homologues. 
As a general rule, homologous parts resemble each other in form 
and uses, and by these are easily recognized. But there are instances 
in which form and use are insufficient for the determination of the 
homology of a given part. The “prickly pear” (Opuntia) which has 
the appearance of being all leaves, is in reality all stem, and this has 
not only the form but the function of leaves. Two homologous bones 
from different animals may be so unlike that an anatomist might be 
excused for taking them for different parts; and on the other hand, 
two different parts may be so nearly alike that were one not on his 
guard they would be considered homologous. Even Cuvier, in his 
earlier days, mistook the coracoid of a turtle for the scapula, on ac- 
count of its shape.f Among teeth molars and premolars interchange 
forms. Even here Cuvier was misled by this circumstance and de- 
scribed the la.-ge back teeth of the feline carnivora as true molar. 
Owen has shown that these supposed molars were the successors of de- 
ciduous teeth and therefore premolars. Human anatomists have gen- 
* Archetype and Homologies of Vertebrate Skeletons, p. 7. 
t Lepons sur I’Anat. Comp. lr edit. T. i, p. 252. 21 
erally misunderstood the homology of the articulating surfaces of the 
atlas and the upper part of the axis; and from their use have de- 
scribed them as if they were “ articulating processes,” while in man 
and most mammals, these processes do not exist in the vertebras men- 
tioned. True, articulating processes are found in the Cetacea, and 
some birds and reptiles, co-existing with the articulating surfaces 
above mentioned, and occupying the true position of articulating 
processes which the “surfaces” do not. We might extend the list of 
such instances, but it is unnecessary. 
As the form and use of a given part under certain circumstances 
may leave us in doubt as to its homology, we need some other 
guide. This may be found, as Geoffrey St. Hilaire long since pointed 
out, in the relative position. The tusks of the elephant and mastodon 
are only known to be incisors by their position in the intermaxillary 
bones, and the radius and ulna of Icthyosaurus and Plesiosaurus are 
identified by their relation to the humerus and carpus, and not by 
their forms or uses. Had these been the only parts of the above 
mentioned animals which had been discovered, naturalists would have 
hardly suspected them to be ulna and radius. So in the determination 
of homotypes we are not to expect precise similarity of forms, as these 
are liable to differences analogous to those of homologues, as is 
especially the case in the carpus and tarsus. 
If we are justified in accepting the conclusions set forth in this pa- 
per, then by an application of them the homotypes in the two limbs 
may be readily determined, for those parts will he homotypes which have 
the same relative position, and are symmetrically placed with regard to 
each other. SCAPULAR AND PELVIC ARCHES. 
The general homology of these has attracted less attention than 
the determination of the corresponding parts of the two with each 
other. Oken, Spix and Cams, considered them as ribs, and in this 
respect have been followed by Owen, who has presented his views 
with much more precision than his predecessors. Admitting them 
to be specially modified ribs, to what vertebrae do they belong ? The 
pelvic arch has been assigned with much unanimity to the sacral 
vertebrae in the immediate neighborhood of which it always is; but 
the scapular arch offers a much more difficult problem, since in the 
three higher classes of vertebrates, the vertebrae near which it is found 
are all provided with ribs, and in many animals all the cervical verte- 
brae are rib-bearing, independently of the arch in question. Spix, 
who has been followed by Owen, regards this as being made up of 
the ribs of the occipital vertebra, and Owen urges in confirmation 
of this view, that in fishes the scapular arch is an appendage to the 
occiput. 
The objections to this view, though we are not sure but that they 
are more apparent than real, are, first, that in only one of the four 
classes of vertebrates, viz., fishes, would the arch be found, so to speak, 
in its normal place; second, each such rib in the larger portion of the 
vertebrate series higher than fishes, would be provided with two “car- 
tilages” or “haemapophyses,” viz., the coracoid and clavicle. This same 
difficulty presents itself in connection with the pelvic ribs, since these 
would also have two cartilages, the ischium and pubes. Third, the 
objection urged by Agassiz appears to have much weight. He objects 
that this arch and the limbs supported by it, derive their nerves in all 
classes from the spinal, not from the cranial series, while the reverse 23 
should be the case if they were truly cranial ribs. A branch of the 
vagus, or of the lateral nerve formed by the union of the vagus and 
trigeminus is, it is true, distributed to the scapular arch in addition to 
its spinal pairs. This, however, would not make the scapular arch an 
appendage to the head, any more than it does the trunk or even the 
tail to which this same nerve sends branches in fishes and Urodel 
batrachian reptiles. 
In view of these objections one cannot but feel that additional evi- 
dence is needed, especially that to be derived from embryology, be- 
fore definite conclusions can be reached. Extensive observations on 
the development of the scapular and pelvic arches maybe expected to 
throw much light upon this problem. We have studied the develop- 
ment of the pelvis in frogs, and find that this dofes not take place after 
the manner of ribs. In the tadpole of the Bull-frog (Rana pipiens, L.) 
the first trace of the pelvis consists of a plate of cartilage situated 
on the median line in front, and which corresponds in position with 
the future ischium and pubes. This cartilage subsequently becomes 
extended on either side, and at last forms a connection with the trans- 
verse processes of the sacral vertebra;. Its first stage, it will be seen, 
corresponds very nearly with the permanent condition of the pelvis in 
most fishes. In Cetaceans, the pelvis consists of ischia, or ischia and 
the bones of the pubes alone ; the ilia do not exist. This would seem 
to be an indication that in mammals the development of this part 
followed the same course as in frogs and fishes; that is, it begins its 
development on the median line in front. 
In most vertebrates the special homology of these arches is more 
obvious, though somewhat masked, since the physiological require- 
ments of the two differ so widely ; strength and solidity being essen- 
tial conditions in the pelvis and mobility in the shoulder. If, how- 
ever, they are compared in those animals whose habits are more 
strictly aquatic, and whose locomotion is effected more nearly equally 
by the fore and hind limbs, as, for example, in the Ornithorhynchus 
among mammals, and the marine saurians among reptiles, the scapular 
arch, in consequence of the greater development of the coracoids, and 24 
of the union of these as well as of clavicles with the sternum, or with 
each other, will be found to have almost the firmness of a pelvis. On the 
other hand the attachment of the pelvis to the vertebral column may 
become comparatively slight. In the chameleon the pelvis is nearly 
as moveable as the scapular arch, and in the Icthyosaurus it appears 
to have been entirely free, and to have embraced the ribs behind as 
the scapular arch does in front. (See Cuvier’s Oss. Foss., PI. 260.) 
Although the two arches never repeat each other exactly in one and 
the same animal, they do sufficiently to show that they are constructed 
upon one and the same plan. The scapular arch of one animal, how- 
ever, often very nearly resembles the pelvic arch of another, as, for 
example, the first taken from a frog, Fig. 7, when compared with the 
second taken from a chameleon, Fig. 8, but in the Enaliosaurians the 
resemblance of these parts in the same individuals is still more 
striking. 
Fig. 7. 
Fig. 8. 
Fig. 7. Scapular arch of a frog. A, scapula; b, clavicle; c, coracoid. 
Fig. 8. Pelvis of a chameleon, a, ilium; b, ischium; c, pubes. 
As each arch consists of three pieces, the first step is to determine 
which are the corresponding ones. Yicq d’Azyr comparing bones 
from opposite sides, Gerdy, Bourgery, Blandin, Flourens, Cruveilhier, and Owen* comparing those of the same side, agree in considering the 
ihnm as the serial homologue or homotype of the scapula, the pubes 
of the clavicle and the ischium of the coracoid. 
Charles Martins compares the bones of the two arches in the fol- 
lowing manner: “The reader may place the shoulder of one side re- 
versed over the ilium of the opposite ; or what amounts to the same 
thing, he may place on his left an ilium of the right side, and on his 
right the reversed scapula of the same side. The outer surfaces of 
the two bones will then be opposite to him, and maybe studied in this 
position, the spectator being between the two.”f Any one who will 
follow his directions with the bones in his hands, will find that the two 
ways above mentioned do not amount to the same thing. In the first 
method, the glenoid and cotyloid cavities will face in the same direc- 
tion, and in the second in opposite directions. In the last these posi- 
tions will be symmetrical, but not in the first. After comparing the 
axillary border of the scapula with the inguinal border of the ilium, 
the coracoid with the ischium, and the clavicle with the pubes, he con- 
cludes as follows: “En resume I’homologie de I’epaule et du bassin me 
parait complete. Un ceinture massive, ala colonne ver- 
tebrale, se’st renversee et transformee en un appareil leger et mobile 
suspendu dans les chairs.” J 
If it be true, as he says, that the shoulder is the pelvis reversed, then 
it follows that the part of the latter which is farthest backwards, viz. 
the ischium, would be repeated in the shoulder by the clavicle, which 
is farthest forwards. This arrangement of the parts would be in ac- 
cordance with the idea of symmetry, and with which M. Martins’ 
description certainly does not agree. 
* “I commence with ilium as being the homotype or correlative of the scapula in 
the fore limb. The ischium, which is the homotype of the coracoid, is confluent 
with the ilium, as the coracoid is with the scapula; the pubes, which is the homo- 
type of the clavicle, is confluent with both ilium and ischium.” Philos. Trans., 
1859, p. 809. 
t Nouvelle Comparaison des Membres Pelviens et Thoracique chez I’horame et 
chez les Mammiferes. Par Charles Martins. Extrait des Mem. de TAcad. des Sci- 
ence 3de Montpellier. T. m. p. 515. 
X Ibid. p. 518. 26 
Mr. Humphrey’s determination of the parts, though based upon 
use and not upon the idea of symmetry, is strictly in accordance with 
it; the hinder edge of the scapula, according to him, being repeated 
in the fore edge of the ilium, the coracoid in the pubes, and the clavi- 
cle in the ischium. 
Foltz adopts a method of viewing these parts quite different to 
that of either of the authorities alluded to above, but not unlike that 
made use of by Cruveilhier, and also by Martins, as will be seen fur- 
ther on, who compare a portion of one and the same bone in one limb, 
to two different bones in the other. The body of the pubes, according 
to Foltz, corresponds with the coracoid process of the scapula, while 
the descending branch corresponds with the clavicle.* He seems to 
overlook the fact that the body and descending branch are all of one 
piece, are ossified from a single centre, and never show themselves in 
any other way. Under these circumstances they cannot be consid- 
ered other than as parts of one and the same bone, unless all ideas of 
the individuality of bones are abandoned. Following the principles 
of symmetrical development already laid down, the homologous parts 
will stand as follows : 
Scapula, Fig. 6, a Ilium, Fig. 7, a. 
Clavicle, “ n Ischium, “ b. 
Coracoid, “ c. . . . . . Pubes, “ c. 
• Homologie des Membres Pelviens et Thoracique. Par le Docteur Foltz. Jour- 
nal de Physiologic. Paris, 1863. T. vi, p. 53. Y. 
LIMBS. 
The general homology of Ijmbs, as has also been the case with 
the scapular and pelvic arches, has attracted much less attention 
than the special. Oken regarded them as “ liberated ribs.” In this 
view he has had but few followers, though Maclise in the article “ Skel- 
eton” in the Cyclopaedia of Anatomy and Physiology, advocates some- 
what at length a similar interpretation of these parts. Limbs are 
more commonly spoken of as “ appendages ” to their respective arches, 
and as “diverging appendages” by Owen, who at the same time con- 
siders them as serially homologous with the “ oblique processes ” on 
the ribs of birds, crocodiles and most fishes. With regard to the ob- 
lique processes just referred to, he suggests that while such rudiment- 
ary limbs in these animals never come to be more than spines attached 
to the edges of the ribs in the actual vertebrates, they might possibly 
under other circumstances, or in other worlds, rise to the dignity of 
perfect limbs. The possibilities of the vertebrate archetype may not 
as yet have, been exhausted.* It does not appear, however, that the 
processes in question ever take on in the actual vertebrates an ap- 
proach to a form which might be considered a limb, and it might be 
urged as an objection to Prof. Owen’s view, that in fishes they seem to 
be an integral part of the muscular syrstem, and that in neither fishes, 
reptiles or birds, do they occupy a position homologous with that of 
limbs, viz., at the junction of the rib with its cartilage, or of the 
pleurapophysis with its hasmapophysis. 
*On the Archetype and Homologies of the Vertebrate Skeleton. London, 1848. 
p. 102. 28 
It is unfortunate that in the attempts to determine the general hom- 
ology of limbs so little attention has been given to their development, 
which is indispensable to the complete solution of the question. On 
studying their transitional phases in the embryo, we are, first of all, 
struck with the fact that in their primary conditions limbs have so 
strong a resemblance to the median fins of fishes and the flukes of ceta- 
ceans. The fins on the median line in fishes all agree in this, that they 
appear as an outgrowth from the integuments, in the form of a ridge 
extending continuously along the back, around and under the tail. 
This ridge is a mass of embryonic cells, all alike, but which subsequently 
become differentiated into fin-rays and other structures. The fin-rays 
are secondary structures, and cannot therefore be said, as has some- 
times been asserted, to push out from beneath and carry the integu- 
ments with them. The adipose fin of the Salmonidse permanently 
retains an early embryonic condition, and no fin-rays are formed at 
any period. The fore and hind limbs, in like manner, are outgrowths 
of the tegumentary cells, and for a time the cells undergo no differentia- 
tion into bones or other tissues. These are, at length, developed in the 
limbs, and subsequently grow pari passu with them. The bones do not, 
therefore, force the Integuments out by their protrusion, but the integu- 
ments themselves have already grown out, and the limb is formed before 
the bones are developed. Limbs in their primary condition do not ap- 
pear to be dependencies of the scapular and pelvic arches, any more than 
the median fins of fishes, or the flukes of cetaceans, which last have some- 
times been compared to limbs, are dependencies of the vertebral column, 
or teeth are dependencies of the jaws, with which, notwithstanding 
their totally different origin, they become so intimately united at last 
In view of the above considerations, and in view of the fact, also, that 
in fishes the ventral fins never pass beyond the condition of appendages 
to the integuments, we believe there is ground for the hypothesis that 
limbs belong to the category of tegumentary organs, and that their 
connection with the vertebral column through the scapular and pelvic 
ai'ches is only secondary, as is that of the teeth with the jaws. 
All agree that these two are homotypes, the only question is whether 
they are to be compared as parallel or as symmetrical bones; this an- 29 
HUMERUS AND FEMUR. 
Fig. 9. Femur of an alligator. 
Fig. 9. 
Fig. 10. Humerus of an alligator. 
Fig. 10. 
swered, the parts which correspond are easily determined. As has al- 
ready been stated, the majority of anatomists describe them as parallel 
repetitions. One of the difficulties which is encountered in this mode, 
is the fact that the knees and elbows in all animals are bent so as to 
form angles pointing in opposite directions. To meet this, Yicq 
d’Azyr, in comparing the limbs, turned the elbow forwards, but in do- 
ing so the head of the humerus was found to face in one direction, and 
that of the femur in the opposite one. He then compared the limbs 
of opposite sides ; in this case, not only was the limb placed in an un- 
natural position, but although the heads of the bones faced in the same 
direction, the thumb was on one side of the hand, and the great 
toe on the other side of the foot. He simply exchanged asymmetry 
of the heads of the humerus and femur for asymmetry of hands and 
feet, and so the difficulty was not obviated. Maclise, in his article 
“ Skeleton,” in the Cyclopedia of Anatomy and Physiology, and Martins 
in his Memoir, meet this difficulty by supposing the humerus twisted 
through 180°. “In primitive construction both members are identical, 
but this secondary modification, viz., the torsion of the humerus, is 
that circumstance which distinguishes them one from the other. While 
in idea I untwist the humerus, by bringing its back to the front, I 
at the same time unravel the gordian knot of that problem which has 
so long existed as a mystery for the homologist.” * According to 
* Cyclop. Anat. and Physiol. Article Skeleton, p. 666. Martins, “ the femur is straight and has no torsion. The humerus be- 
ing a twisted femur, if we wish to compare the two, it is first of all 
necessary to untwist it (detordre I’humerus) the effect of which is to 
place the epitrochlea (inner condyle) outwards, and the epicondyle 
(outer condyle) inwards. This done, the comparison of the pelvic and 
thoracic extremities offers no further difficulty.”* That this does not 
quite clear up the matter even to M. Martins, appears a little further 
on, where he admits what he "calls a “ metaphysical difficulty ” (diffi- 
culte metaphysique'), viz., that the humerus never was literally twisted ; 
“itis a virtual torsion, which was never mechanically effected “ but 
this virtual torsion has produced all the consequences of a real one.” f 
In another place he informs us that “L’humerus n’est point un os 
d’abord droit, qui se torde ensuite. II y a mieux I’humerus est tordu 
avant d’exister.”! In order to put the bone in the untwisted condition he 
supposes it plastic, as when the lime has been removed by acid, or, to 
make a mechanical illustration practicable, he cuts the bone in two, 
puts a peg in the medullary cavity and rotates the lower half 180°. 
Any one, however, who will take an arm from an articulated skeleton, 
place the palm on the table, and then following his directions in im- 
agination, or by the mechanical process, and rotates the humerus 180° 
from within outwards, (the direction necessary to untwist it) he wil 
find a real, and not a metaphysical difficulty. In the ordinary prone 
position of the hand, the radius is already partly wound around the 
ulna, and will completely encircle it when the humerus is “untwisted,” 
thus producing a much greater distortion than the one attempted to be 
obviated. Mr. Humphrey has already pointed out other difficulties in 
the way of this view of Martins, and we will only add that the lines 
supposed to indicate torsion simply grow out of the mode of attach- 
ment of the muscles to the bone, and in no way indicate a twisting of 
it. The bone primarily was smooth, and the ridges were built up on 
it as the muscles themselves were developed, without the ends of the 
bones having in any way changed their relative position. 
If the two bones are supposed to be symmetrical repetitions, no diffi- 
culties arise. They will have the position which is natural to them in 
* Memoir, p. 482. 
t Idem, p. 490. the animal series; the axis of the humerus will incline backwards, and 
that of the femur forwards. The articulating convex surfaces of the 
lower end of the humerus will face forwards, while those of the femur 
will face backwards. The back of the humerus, which is on the side 
of the extensor muscles of the fore arm, will be opposed to that part 
of the thigh which is on the side of the extensor muscles of the leg. 
FORE-ARM AND REG. 
No portions of the limbs have given rise to more widely differing 
opinions than these, and it is in connection with them that nearly all 
the discussion of the homologies of the parts of the limbs have been 
made. Yicq d’Azyr, comparing opposite sides, considers the tibia as 
the homotype of the ulna, the fibula of the radius, and the patella of 
the olecranon. Meckel and others homologize the same parts, but 
compare limbs of the same side. Gerdy, on the other hand, compares 
the radius and tibia, the ulna and fibula, but asserts at the same time 
that the olecranon and patella are homotypes, in which case it is obvi- 
ous that the patella should be attached to the fibula. He gets over 
this difficulty by assuming that its union with the tibia is an “anomaly.” 
Fig. 11. 
Fig. 12. 
Fig. 11. Bones of the leg of an alligator. A, tibia; b, fibula. 
Fig. 12. Bones of the fore arm of an alligator, a, ulna; b, radius, 
Bourgery, and more recently Cruveilhier, seeking for a solution of 
the question by studying only the resemblances in the form and uses 
of parts, adopted the singular “ hypothese de croisementwhich Cru- 
veilhier states as follows : 32 
“Ist. No bone of the leg singly represents one of the bones of the 
fore arms. 
“2d. In each of the bones of the leg we find characters, some of 
which belong to the ulna, and some to the radius. 
“3d. We admit that the upper end of the tibia is represented by 
the upper end of the ulna, and the lower half of the tibia by the lower 
half of the radius ; while the fibula is represented by the upper half 
of the radius, and the lower half of the ulna.”* 
Martins homologizes the parts in question as follows : “ The femoral 
end of the tibia in Monadelph mammals is formed by the humeral heads 
of the radius and ulna.” “ The upper third of the fibula is repre- 
sented by the anterior or coronoidal part of the ulna.” “In all mam- 
mals the two lower thirds of the tibia represent the corresponding part 
of the radius, and the two lower-thirds of the fibula that of the ulna.” f 
The view of Prof. Owen is as follows: “ The skeleton of the 
Phalangista or Phascolomys plainly demonstrates that the tibia is the 
homotype of the radius, and that the fibula is the homotype of the 
ulna.”! The same conclusions are adopted by Humphrey. 
Thus putting together the views of the anatomists above cited, we 
have: 
Ist. The ulna the tibia of the opposite side. 
2d. The ulna the tibia of the same side. 
3d. The ulna the fibula. 
4th. The head of the tibia = the heads of the radius and ulna. 
The fibula = the coronoidal and lower two-thirds of the ulna. 
The lower two-thirds of the tibia = the corresponding part 
of the radius. 
sth. The tibia = the upper half of the ulna and the lower half 
of the radius. 
The fibula = the upper half of the radius and the lower half 
of the ulna. 
There seems to be no sufficient reason for entering into a discussion 
of the views of those who consider a bone of one limb homologous with 
*Traite d’Anatomic Descriptive. Paris, 1843. T. i., p. 342. 
f Op. cit., p. 534. 
t Archetype, p. 167. 33 
parts of two different bones in another. We do not know of any 
unquestioned analogy in the whole range of comparative anatomy 
which can be brought forward in support of it. 
Prof. Owen sustains his opinion, already cited, by the following 
statement, and in this he is followed by Humphrey and others : ‘‘ln the 
Wombat the part of the fibula representing the olecranon is a de- 
tached sesamoid, as the olecranon Itself is in the penguin and the bat. 
In the Ornithorliynchus the fibula (Fig- 14, B) assumes the proportions 
and developes the process from its proximal end, the want of which 
in man and most mammals, deceived Yicq d’Azyr as it misled, more 
recently, M. Cruveilhier.” 
We are somewhat at a loss to know what is to be understood by the 
expression “ detached sesamoid,” inasmuch as a sesamoid, as commonly 
understood by anatomists, is always detached. The only interpreta- 
tion which suggests itself to us is, that it is used synonomously with 
an epiphysis, and that the bone connected with the upper end of the 
fibula in the Wombat is the detached epiphysis of that bone, and this 
he considers homologous with the large process on the upper end of 
the fibula of the Monotremes. No evidence is brought forward to 
prove that the sesamoid referred to is in any way different from the 
bones found in tendons in other parts of the body, as e.g., in the heads 
of the gastrocnemius of the opossums, and also of many rodents, and in 
the flexor tendons of the fingers and toes of many mammals, in the, 
peroneus longus in man, and as an anomaly in the tendon of the 
human biceps cruris near its insertion into the fibula. Furthermore 
the part in question can hardly be considered as a detached epiphysis, 
since in those animals where a sesamoid is developed in the tendon 
over the head of the fibula, the normal epiphysis exists in the bone 
itself. 
In the Penguin it does not appear that the part corresponding to 
the olecranon is of any less size than in other birds, although the sesa- 
moid bones, which Prof. Owen supposes replace this process, are pres- 
ent. In the arm of a large Pteropus we have found the patella-like 
bone attached to the ulna by a ligament, as the patella is attached to 
the tibia. The ulna, however, has a projection at its upper end of 34 
about the same size as the olecranon in birds. But before this patella- 
like bone can be claimed to be a detached epiphysis, it must be ascer- 
tained whether the ulna has, or has not, an epiphysis in the immature 
bone. In many of the anatomical descriptions of the ulna, the ole- 
cranon process and the epiphysis have been described as if they were 
the same thing, which most certainly they are not. In mammals the 
olecranon is often much longer than in man, the epiphysis forming but a 
very small part of it, and in man is only a very thin scale covering its 
end. Properly speaking, the olecranon is a continuation of the shaft 
of the ulna. The cartilage of ossification interposed between the shaft 
and the epiphysis is remarkably thick, and has given rise to the belief 
that this was itself the epiphysis. The last only forms at a late period. 
As to the argument drawn from the great development of the pro- 
cess on the top of the fibula in the Ornithorhynchus, it seems to us 
that this is an instance in which we are liable to be misled by form, 
and that the resemblance between the process in question and the 
olecranon is no proof of homology. It should be remembered that 
processes, and even bones, in different animals, are liable to every de- 
gree of variation according to the physiological requirements in indi- 
vidual cases. Compare the extraordinary processes of the humerus 
of the Mole, the Mylodon, or of the Ant-eaters, with the diminutive 
ones of the same bone in the Three-toed Sloth, the Cetaceans, or of 
the marine Saurian reptiles. The olecranon itself is a very variable 
process; it does not exist in Cetaceans, is hardly apparent in birds and 
most reptiles, but in mammals may become, as in the great Armadillo, 
almost half as long as the whole ulna. This development is in relation 
to the extensor muscles of the fore arm. The remarkable develop- 
ment of the fibula in the Ornithorhynchus may justly be compared 
with the great development of the olecranon in the Armadillo or the 
Mole, as a means of increasing the power of the particular muscles 
attached to it, but this resemblance is physiological only, and has no 
bearing whatever on its homology. The head of the fibula in the 
Ornithorhynchus is, however, not developed in relation to the extensor 
muscles of the leg, but of some of the unusually large muscles of the 
foot, which take their origin from it; viz., the large head of the gas- 35 
trocnemius, the soleus, the two peroneal muscles, the common flexor of 
the toes, the long extensor of the great toe, the common extensors of 
the toes and posterior tibial muscle, but has not that relation to the mus- 
cles of the leg which the olecranon has to those of the fore arm. The 
mere form of°the process of the fibula is no proof that the fibula and 
ulna are homotypes, although in many respects it resembles an ole- 
cranon. (Fig. 14, B.) The prolonged upper end of the tibia in 
Fig. 13. 
Fig. 14. 
~ I} fibula; c, upper end of tibia prolonged so as 
Fig. 13. From a grebe, a, tibia, b. nuui*, w 
to form a process analogous to the olecranon, d, emur. 
1* , . finia* x? fibula, with a process analo*" 
Fig. 14. From an Ornithorhynchus. a, tibia, b, noma, w 
gous to an olecranon; c, patella, n, femur. 
the Grebes (Fig. 13, C), Loons, Ponguins, Gannots. etc. can be 
brought forward with far more force to show that the tab.. » the hom- 
otype of the ulna, for, in addition to its resemblance to an olecranon 36 
in form, “it affords extensive attachments by way of insertion to the 
extensors of the tibia,” * which the fibula does not. 
If the fibula is the homotype of the ulna, then it follows that the 
flexor muscles of the leg, which are inserted into the upper end of it, 
are the homotypes of the extensor muscles of the fore arm, which are 
inserted into the olecranon ; with this comes another difficulty grow- 
ing out of the fact that the internal angle formed by the leg and thigh, 
and which is on the side of flexion, corresponds with, or is the homo- 
type of, the external angle formed by the arms and fore arm, which is 
on the side of extension. 
Prof. Owen derives additional support to his view as to the homol- 
ogy of the ulna and fibula, from the mode of articulation in some 
animals of the tibia with the fibula, as compared with that of the ulna 
and radius. “ The correspondence of the fibula with the ulna is very 
remarkably maintained in the Petaurus taguanoides, in which the 
proximal articular surface of the fibula is divided into two facets, one 
playing upon the outer condyle of the femur, the other concave verti- 
cal and receiving an adapted convexity on the outer end of the head 
of the tibia, which rotates thereupon exactly like the radius in the 
lesser sigmoid cavity of the ulna.” f 
In connection with this statement it may be remarked that as re- 
gards the articulation of the fibula with the femur, it is the exception 
among mammals to find it coming in contact with that bone, but even 
in Monotremes, Birds and Reptiles, the extent of its articulation with 
the femur is always secondary to that of the tibia, and never becomes 
as extensively articulated with it as the ulna does with the humerus. 
As regards the rotation of one bone on the other, we have found, 
after a careful examination of the parts in an OrnithorJiynchus which, 
through the kindness of Prof. Agassiz, we have recently had an op- 
portunity to dissect, that it Is the fibula which rotates, while the 
tibia is fixed, and in this respect the latter more nearly resembles 
the ulna. The fibula in the Petaurista resembles the ulna in having a 
sigmoid notch. This cannot be considered as a decisive character, 
* Owen Cyclop. Anat. and Physiol. Vol. i, p. 287. 
t Cyclop. Anat. and Physical. Art. Marsupialia. p. 285. 37 
when it is remembered that the articulating surface of the lower jaw 
is convex in mammals, but concave in other vertebrates ; the lower 
jaw does not in consequence change its homology in these last. 
Admitting all the arguments which have been adduced to prove 
the fibula to be the homotype of the ulna, we have still another and 
greater difficulty than any thus far mentioned, growing out of its rel- 
ative position. For of all the means of determining homologies, this 
is the most trustworthy. If the two bones of the fore arm and of 
the leg are placed in planes at right angles to the axis of the body, 
those bones must be considered homotypes, which occupy correspond- 
ing positions. The bone on the outside of the fore arm, viz., the 
radius, can only be the homotype of that on the outside of the leg, 
viz., the fibula. But few anatomists have made any allowance for the 
pronation of the fore arm, and most of them overlook the fact that 
the proper position of the bones of this segment for comparison with 
those of the leg, is supination. If the position of pronation is to be 
retained for the fore arm, the leg should go through a corresponding 
rotation in the opposite direction. Viewed in connection with the 
idea of symmetry, the homotypes are determined-without difficulty, 
and are as follows; 
The Radius is homologous with the I ibula 
The Ulna is homologous with the Tibia. 
THE PATELLA. 
By Yicq d’Azyr, who has been followed by Meckel, Blainville, 
Martins, Humphrey and others, this bone has been regarded as the 
homotype of the olecranon, differing from it,, however, in being at- 
tached to bone by a ligament. Soemmering, who has been followed by 
Bertin Bichat, Flourens, Owen and Cruveilhier, maintains that the 
patella belongs to the class of sesamoid bones, and therefore, properly 
speaking, does not belong to the skeleton at all. 
AVc believe the latter view to be the correct one, for the same rea- 
son that the separate bone in the leg of the Wombat, already referred 
to, is a true sesamoid, and not a detached epiphysis ot the fibula. Mr. 38 
Humphrey, like the others, who regard the radius as the homotype of 
the tibia, and the olecranon as the homotype of the patella, is obliged 
to meet the difficulty which arises from the connection of the patella 
with the tibia, when, if it were the homotype of the olecranon, it 
should be connected with the fibula. Mr. Humphrey admits the pos- 
sibility of a part of a bone, an epiphysis, being detached, and of 
becoming connected with another, and sustains his statement by the 
analogy of the ribs, which, he says, may be transferred from the upper 
to the lower transverse process, or vice versa.* This does not seem to 
be an analogous case; for, in point of fact, the typical rib is attached 
to both, as, for example, in the foremost, ribs of the alligator, and either 
the upper or lower attachment may become obsolete, without really 
shifting the relation of the ribs to the vertebra. There is not a single 
unequivocal Instance in the whole range of comparative anatomy, of 
an epiphysis undergoing such a displacement as is claimed for the 
patella. 
There are some facts, however, which seem to support the view 
under discussion. The researches of Robin have set at rest the ques- 
tion as to the nature of the odontoid process of the axis, and have 
proved that it is the body of the atlas coossified with that of the 
axis. This, however, would not be similar to the transfer of the 
epiphysis of one bone to the shaft of another parallel to it. The 
bodies of the vertebrae are in a linear series, and may be united 
without changing their relative position, as happens, for example, with 
the occiput and sphenoid, or with the epiphyses of the sacral vertebrae 
which may become united with each other before either of them be- 
comes united with their respective bodies. The epiphyses of the ulna 
and radius in some ruminants, as the ox, may become cobssified with 
each other before they are coossified with their shafts. But in all such 
cases the relative position of parts is strictly preserved, and there is 
consequently no transfer of an epiphysis from one bone to another, as 
is required by the hypothesis which is referred to above. We therefore 
conclude that the view of Soemmering and others, which regards the 
patella as a sesamoid, is the more reasonable. 
* Memoir, p. 20. 39 
HANDS AND FEET. 
Admitting the existence of fore and hind symmetry, no difficulty is 
met with in comparing either the pelvic or scapular arches, the hu- 
merus and femur, or the bones of the fore arm and legt provided the 
bones of the fore arm are rotated outwards enough to counteract the 
rotation inwards which took place in foetal life. If we would compare 
hands and feet as symmetrical parts, the first step should be to put 
them in symmetrical positions. For this purpose suspend a human 
skeleton with the vertebral column horizontal, allow the legs and 
arms to hang vertically from it, and rotate the fore arm completely 
outward. The palms will now face forwards, and the soles backwards, 
the bones of the fore arm and leg will be in parallel planes, and these 
at right angles to the axis of the body. If now the foot is raised for- 
wards, so that the sole shall be horizontal and at right angles with 
the leg, the hands should be raised backwards to the same position 
with regard to the fore arm; the fingers will now point backwards, 
and the toes forwards, and thus the general symmetry of all the seg- 
ments of the limbs is secured. It is in this that the limbs of man and 
animals would have ended in the process of development, had not the 
tendency to symmetry been interfered with, for the purpose of adapt- 
ing the skeleton to the different kinds of locomotion. 
While the homology of tarsal and carpal bones, as groups, is obvi- 
ous, that of individual bones is quite difficult to determine. The fact 
that in mammals the bones of these two groups are not conformably 
placed in the two limbs, and in addition to this the constant variations 
in the vertebrate series of the form and number of the pieces, some- 
times reduced to two, as in the tarsus of a frog, and sometimes in- 
creased to eleven, as in the carpus of the armadillo, renders the 
probability of a satisfactory result being reached in the direction of 
special homology well-nigh hopeless. The homologies, in man and 
mammals, of the individual bones in the tarsus and carpus hitherto 
determined, rest largely on their physiological correspondence. The 
pisiform and the calcaneum, for example, are regarded as homotypes, 
because each, in certain animals, as the Carnivora, has a large tuberos- 40 
ity, and each is a lever for increasing the muscularpower applied to the 
motions of the foot. But in these animals they are highly specialized 
parts, and are the farthest removed from the typical conditions which 
are most generally the best represented in the lowest animals of a 
given group. If anatomists had begun their studies with reptiles, such as 
Plesiosaurus and Icthyosaurus, or with lizards and turtles, the homolo- 
gies now generally recognized would not have been so persistently 
brought forward. 
The mammalian foot, which connects with the leg solely by the 
astragalus, gives one quite a different idea of the tarsus and its rela- 
tions, from that derived from the same segment in reptiles, where, for 
the most part, the astragalus articulates chiefly with the tibia, the os 
calcis with the fibula, and a third bone is interposed between them. 
Furthermore, the os calcis, which is so highly specialized in mammals 
as to be distinguished at sight, has in reptiles the appearance of the 
other tarsal pieces, and resembles a cuneiform bone. Even in mam- 
mals the pisiform, as in man, is so reduced that it becomes relatively 
insignificant, which circumstance, and its relation to the tendon of the 
ulnar flexor, led Cruveilhier to class it among the sesamoid bones. 
In the ideal vertebrate skeleton the tarsal and carpal bones having 
no special development, would be represented by two rows of polyg- 
onal or circular discs, all alike, as is actually the case in the marine 
saurians. Reducing all the bones in question to one and the same 
form, as in the lowest groups, form would cease to be a guide to the 
determination of the homology of any particular bone of the tarsal or 
carpal series. Those parts will be homologous which occupy similar 
and symmetrical positions; the inner bone of the wrist articulating 
with the ulna, will be homologous, or the homotype of the inner bone 
of the tarsus articulating with the tibia; or in other words, homolo- 
gous bones are determined by the principle of symmetry and relative 
position, and not by their teleological relations. 
Making use of this principle, we shall have the following bones 
homologous in the two limbs ; beginning on the inside : 41 
HAND, FOOT. 
Pisiform 
Pyramidale Scaphoid. 
Lnnare Os calcis. 
Scaphoid Astragalus. 
Unciform Ist cuneiform. 
Magnum . 2d cuneiform. 
Trapezoid . 3d cuneiform. 
Trapezium. Cuboid. 
When applied to the human hand and foot with their high degree 
of specialization, this mode of comparison seems at first sight inadmis- 
sible, but bearing in mind the fact that the type is much more distinctly 
indicated in the lower, than the higher members of a series, and 
beginning our comparisons in the lowest, the difficulties growing out 
of the special developments are obviated. It will be found that in 
reptiles the tarsus, for the most part, articulates with the leg by two, 
and even three, bones instead of by one, as in ordinary mammalia; 
that the astragalus and os calcis neither of them have the peculiar 
characters which are exhibited in the higher vertebrates, while in the 
seals among mammals the astragalus develops a tuberosity backwards 
equal to that of the os calcis of the same animal. 
We have placed no bone opposite the pisiform as its homologue. 
Some homologists, and among them Owen, regard the os calcis as, in 
itself, repeating the pisiform and pyramidale, and as consisting really 
of two bones combined in one, as the scaphoid and lunare are in the 
carpus of the cat and some other mammals. This view docs not seem 
to be well supported; for while the coalescence of the scaphoid and 
lunare in various animals is a matter of observation, the existence of 
an os calcis in two parts has not been observed in any. In its mode 
of ossification, except in the development of a thin scale on the end 
of its tuberosity, it follows that of the other tarsal bones, viz., from a 
single centre. And in having an epiphysis it agrees with the astragalus 
in some lower animals, as the seal. The relation ot the pisiform to 
the tendon of the “ ulnar flexor ” (extensor) of the wrist seems to 
justify Cruveilhier’s view that it belongs to the category of sesamoids. 
The metacarpal and metatarsal bones offer no difficulty in either of 
the two methods of studying the skeleton, since they agree in their 42 
relative position, and the only differences are purely teleological ones. 
Excepting in the thumb and great toe the same may be said of the 
phalanges. We have already alluded to the great difficulty which the 
exceptions just mentioned offer when the limbs are studied as symmet- 
rical parts, and we know of no way in which it can be fairly and sat- 
isfactorily met. 
The thumb and great toe are assumed by most anatomists to be 
homotypes. First, on account of their relative size. Secondly, because 
they have similar relative positions in the ordinary attitude of the 
fore arm. Thirdly, and chiefly, because they have only two phalanges 
each, while each of the other digits has three or more. 
If the human hand and foot are alone examined, the relative size 
of the parts in question favors the view that they are homotypes. 
But this characteristic of size loses its value when they are studied in 
the lower animals. In the seal the thumb might, as regards its size, 
be considered the homotype of either the first or the fifth toe, which 
are the two largest and of equal size. In the walrus the first digit of 
the hand and the fifth of the foot, or the thumb and little toe, are the 
largest in their respective limbs. In the great ant-eater the third digit 
of the hand is longest, while the fourth is in the foot. If size were 
the criterion of homology, either of the fingers might in turn become 
the homotype of either of the toes, for the size of these parts being 
determined by their physiological adaptations, either may in turn be- 
come the largest or the smallest in the series. 
The second reason, that based on the fact that they are both on the 
inside of their respective limbs, loses its force when it is remembered 
that the parts compared are, as it were, in a false position. That but 
for the rotation of the fore arm in the embryo, the thumb would have 
been on the outside of the hand, and would consequently have con- 
formed to the position of the little finger. 
The third argument, derived from the existence of two phalanges in 
each of the parts, is not so easily disposed of, and forms the greatest diffi- 
culty In our way. Notwithstanding the wide difference in the physiolog- 
ical value of these parts in different species of animals, and the conse- 
quent range of variation in the size of them, the number of phalanges 43 
may be said to be almost constant. It is true that in Icthyosaurus, Plesi- 
osaurus, and other marine saurians, the thumb and great toe, like all 
the other digits, have their phalanges multiplied, and if our comparisons 
were confined to such animals as these the question of homologies 
would be easily answered, as it would also in some of the land turtles, 
where the number of the phalanges in all fingers and toes is reduced 
to two. If, too, we might apply the saying of Goethe, which holds true 
in so many instances, viz., “ that it is in her monstrosities that nature 
reveals to us her secrets,” we might call to mind an occasional mon- 
strosity in which the thumb and great toe are each provided with three 
joints, and thus made to conform with the other digits. Lastly, we 
might call to mind the fact that in their mode of ossification, the 
metatarsal and metacarpal bones of the two parts in question agree 
with the phalanges, that is, in having the proximal epiphysis the last 
to unite with the shaft instead of the distal. Still the preponderance 
of facts is the other way, and, if we adopt the idea of symmetry, we 
must rest content with the assumption that the thumb with its two 
phalanges is the homotype of the outer toe with its three phalanges. 
In the preceding pages the object has been to set forth some rea- 
sons for studying the fore and hind parts of vertebrated animals, but 
more especially their limbs, as if these parts and limbs were con- 
structed not only after one and the same type, but in a symmetrical 
manner. They would repeat each other exactly in an ideal animal, 
just as the right and left parts do in the actual. In the actual animal 
the fore and hind parts are so modified as to adapt them to special 
conditions of existence, and as the conditions fulfilled by the two 
kinds of limbs are generally different, the limbs take on different de- 
grees and phases of development. Eight and left parts repeat each 
other almost exactly, because their conditions are the same, though 
even these, as we have already seen, may sometimes vary, and then 
we have a diversity in their development. In fore and hind limbs 
diversity is the rule, while in right and left it is the exception. Nev- 
ertheless, as we go back to the early stages of embryonic life, the 
symmetry and equality of fore and hind parts becomes nearly exact, 
however much they may vary in the adult. 44 
We have not forgotten that in attempts like the present, compari- 
sons should be made, not only between the bones of the limbs, but also 
between the muscles, nerves and vessels. We have confined our re- 
marks chiefly to the bones, because their homologies are the most 
accurately determined. The attempts hitherto made for the determi- 
nation of the homologies of the other parts, have been far less satisfac- 
tory. If a serious objection can be brought against the mode we have 
adopted of viewing the bones, far more serious objections can be 
brought against such a method of viewing the other structures. If 
the method fails in the skeleton, it will certainly fail elsewhere. On 
the other hand, if antero-posterior symmetry can be shown to exist in 
the bones, then we can feel some confidence, that whatever the diffi- 
culties at present may be with regard to the muscles, nerves and ves- 
sels, they will sooner or later be overcome. We may go still further, 
and assert that if the idea of fore and hind symmetry enters into the 
composition of animal structures at all, it will be traced not only in 
the limbs, but in all the great systems of organs. Unity of plan in 
the structure and composition of animals is much more likely to prove 
true than diversity. 
Attempts have been made to construct an ideal skeleton, an “arche- 
type” which is presumed to contain all the essential elementary parts 
of a vertebrated skeleton; these parts nowhere so developed as to 
be adapted to the wants of any individual, but capable, by a variation 
in the quantity and proportions of each, of being adapted to the con- 
ditions of life of every member of the series. Carus in his “ Urtheilen 
des Knochen und Schalen geriistes ” attempted such an archetype. 
Owen in his “ Homologies and Archetype of the Vertebrated Skele- 
ton,” has constructed another on essentially the same basis, but far 
more complete. Neither Carus nor Owen have, however, admitted 
the idea of fore and hind symmetry. If we admit this, then the 
archetype must be so modified as to conform to it. The typical struc- 
ture which represents or occupies the place of the head at one end, 
if we will carry out the idea, must be represented by a similar re- 
versed structure at the other. If, for example, we would adopt 
Owen’s modification of Carus’s archetype, we must divide it in the 45 
middle and replace the hinder half with a reversed repetition of the 
fore half. The skeleton would not in this way be provided with two 
heads, but only with the rudiments of these capable of being devel- 
oped or arrested in development, in such a manner as the conditions 
of individual existence may require. 
Whether we adopt the doctrine of fore and hind symmetry or not, 
such a conception as an archetype involves is necessary in our 
attempts to study the creative idea which underlies all animal struc- 
tures apart from their adaptation to the modes of existence in each 
species; and just in proportion as such conception is based upon a 
more and more complete knowledge of the plan of structure and of 
development, anatomy will, In the same degree, become philosophical.