In order to do this I have put together all the partial results on which I believe that I can rely, have taken into account my own measurements, and have adopted the most justifiable compromise amongst them all. The canon of proportion for the anthropologist, I need scarcely say, is the vertical figure of a man divided into one hundred parts, in which are represented the segments of the body, each with the number of parts which enter into its composition in the vertical as well as in the transverse directions - so far, at least, as possible.

I have not considered it necessary,' he proceeds, .to endeavour to obtain an approximation nearer than that of 0.5, although two-tenths added or subtracted from any part of the body have often a great importance in the differentiation of races from this aspect, and in the determination of the influences of environment and of education on the proportions. The following table gives the elements which have served for the construction of the figure:

Mean canon of the European male. Stature = 100.

Head, vertex to chin

13.3.

100

Neck, chin to supra-sternal notch -

4.2

Trunk, notch to seat ...

35.0

Inferior extremity, seat to ground -

47.5

Superior Extremity

Arm, acromion to olecranon

19.5

45.0

Fore-arm, olecranon to styloid process

14.0*

Hand†......

11.5

Inferior Extremity

Thigh, seat to centre of knee

20.0

47.5

Leg, knee to malleolus -

23.0

Malleolus to ground -

4.5

Foot.......

15.0

Height of umbilicus ... 60.0

" " pubis .... 50.5

Span of arms.....104.4

Maximum width of shoulders - - 23.0 " " pelvis - - 16.9

" " hips - - - 18.8

'It will now be interesting to compare the canon which has just been given with that already stated as the canon of the studios. The chief differences are as follow: The head, higher than that of Cousin and Gerdy, is practically the same as that of Blanc; that is to say, it is contained seven and a half times in the stature. The neck, which all artists find too long in the canon of Blanc, is nearly that of Cousin. The inferior extremity in its entire length, estimated by the height of the pubis to permit of comparison, is notably too long in the system of Blanc. It is, on the contrary, too short in the two canons of Cousin and Gerdy; the divisions are bad in the system of the latter. The span of the arms used by artists is absolutely false, for it is equal to the height only once in every ten cases. The shoulders and the hips are too large, and the umbilicus is too high. The height of the pubis can only† 'This proportion, being the total of the three segments of the limb, is less when the member is measured in a straight line from the acromiou to the extremity of the middle finger.

I have considered, however, that this difference might be neglected.' be measured approximately, but it has an exceptional importance, because it is in its neighbourhood that the centre of the body in the vertical direction is to be found. M. Sappey, whose measurements relate especially to this point, places it 13 mm. below the pubis, at the root of the penis. This agrees very well with my conclusions, but not with the canon of Blanc, which places it lower still.'

* The line of separation between the arm and fore-arm is here taken at the superior part of the olecranon.

Scheme of proportions of Paul Topinard. The human body divided in the vertical direction into one hundred equal parts.

Fig. 4. Scheme of proportions of Paul Topinard. The human body divided in the vertical direction into one hundred equal parts.

Before leaving for the present the observations of Topinard, to which I shall have again to recur, it should be remembered that his conclusions are of the greatest weight, being based upon accurate measurements and comparisons. They will be used as a touchstone by which the various canons may be judged at a later part of this work.

In Spain, Philip Borgogna is the author of a system which estimates the stature of the adult male as being equivalent to nine and one-third times the height of the face. Join de Arphe y Villafane, who, like Borgogna. studied at Toledo, published in Seville, in 1685, a work entitled 'Varia Commensuracion para la Escultura y Arqui-tectura,' in four books, of which the second dealt with human proportions. According to Choulant, this book contains a large number of plates, some of which give figures of the whole body, and others separate portions thereof, with scales of measurement, from which we gather that the author had seen Durer's figures of proportions. The representations are, however, more true to nature, more living and more spirited. Two male and two female figures are represented, in each of which the stature is made equal to the length of the faces. Chrisostomo Martinez (1650-1691 or 4) was the author of a work in which appeared the plate represented in Fig. 5. According to Quetelet, he made the stature contain eight heads.

It may be noted of this last writer, that his figures of skeletons were regarded by the great anatomist AWinslow as models of what such drawings should be.

In Holland, S. van Hoogstraeten, born at Dordrecht in 1627, published at Rotterdam in 1678 his 'Polymnia,' in which he gave three plates illustrative of the proportions of man. In the first plate he represents two men, one being fifteen, the other sixteen, palms in height; and as his head contained two palms, it follows that these two figures were respectively of seven and a half and eight heads in height. In his second plate he represents a female figure divided into fifteen parts: seven of these are from the ground to the genitalia, and seven from this point to the line of the eyes, by which method of division the legs are made shorter than the upper part of the body. The arms and hands of both his male and female figures are made too short.

Plate by Chrisostomo Martinez (Choulant).

Fig. 5. Plate by Chrisostomo Martinez (Choulant).

In Belgium, Johan de Laet, of Antwerp, published an edition of Vitruvius, in the appendix of which he quotes Pomponius' canon, as given in his work on sculpture, as being nine faces. Geerardt de Lairesse (1040-1711) makes the following statements: 'The eyes are at such a distance apart that a third can be placed between the two. The nose is one-third of the length of the face. The mouth is as large as an eye. The ears are at the level of the eyes above and of the nose below, however long or short it may be.' Van Bree of Antwerp published in 1821 his 'Lecons du Dessin,' in which he uses the head as a modulus, dividing it into four parts. This writer, who was the first professor in the Academy at Antwerp, gives in his books a number of measurements from ancient statues, which are, however, according to Quetelet, of doubtful accuracy. The most useful work which has appeared in Belgium, an epoch-making book, is that of the last-named author - 'Anthropometric ou mesure des differentes Facultes de l'Homnie.' In this work Quetelet commenced by giving a sketch of the labours of former writers in the same field, to which I have to express my indebtedness for many of the facts which I have laid before you.