The head was by the ancients generally considered to be contained eight times in the body, though this proportion is one which, as we have already had occasion to note, is frequently departed from. I give here a few figures, with the authority for each, and others appear in the more elaborate table of Topinard:

Statue.

Heads.

A uthority.

Pythian Apollo......

Humphry.

Farnese Hercules........

8

Quetelet.

Laocoon ..........

7 2/3

Duval.

Antinous ..........

Ibid.

Gladiator .................

8

Ibid.

Moreover, more modern artists have varied the canon considerably; thus, in some of Michael Angelo's figures the size is equal to nine, or even to twelve heads, in order to communicate more grace to a stooping attitude (Humphry). Roberts errs in making it the seventh part of the whole height, though he also says that the proportion may vary between six and eight, and in the case of giants, nine times; while in dwarfs it may form a fourth part of the height.

Quetelet makes the male head 7.4, or very nearly seven and a half times included in the stature.

Topinard gives the following table showing the proportion of the head to the body, as expressed by various artists. The second column shows what this amounts to in numerical terms of the stature, the latter being taken as 100.

Canon.

Heads.

Stature = 100

Hindoo......................

14

Egyptian (two statues)

13.2

Greek (mean of 11 statues varying from 7 to 8¼ ......

7 2/3

13.0

Roman (Vitruvius) ....

8

19

Italian (Alberti) .....

13.2

Prussian (Schadow) ....

7 1/3

13.6

French (Cousin) ....

8

12.5

" (Gerdy).....

8

12.5

Having thus laid down the figures employed by various artists, and after tabulating a number of figures ascertained by anthropologists, he makes the following remarks upon the two sets: The canon of Vitruvius adopted by Gerdy and Cousin exists only in the imagination of the authors; the Greek canon is that of Europeans with small heads, and more particularly, perhaps, of those of Mediterranean races; the Hindoo canon, which relates to the yellow Dravidian races, is approximately correct; and, finally, the canon of Schadow, which was formed from fair races of tall stature with long and narrow faces, is also approximately correct. The European races have shorter heads, although amongst these are met with higher types, such as the Belgians. The yellow races have very notably higher heads. The negroes of Africa are in this respect nearer the first,' and the negroes of Oceania are nearer the second. Using the language of artists, and speaking of the large average, it may be said that the stature of Europeans is equal to seven and a half heads, that of negroes to seven, and that of typical yellow races to six and a half.

This figure Topinard has expressed in his own canon, which I gave at a former period; he there makes the head 13.3 of the stature, an amount which is contained a trifle more (.25) than seven and a half times in the one hundred parts allotted to the stature. This also coincides with the canon of Marshall, in which the head is nine parts of a stature of sixty-seven, giving seven heads and four units, or very nearly seven and a half heads for the stature.

We may, I think, conclude that in representing the average European male this figure may be accepted as accurate. When the number of heads is decreased an appearance of heaviness and dwarfishness is imparted to the figure. When it is increased slightly it may give an appearance of greater gracefulness to the person, yet still without sacrifice of truth; but when it is carried to eight heads the boundary is passed. A good idea of the effect produced by altering the number of heads in the stature of a figure will be gained by examining the linear scheme of the Germanicus and Apoxyomenos (Fig. 3, p. 27), the former containing rather more than seven and a half heads, and the latter nearly eight and a half. Camper has given an example of the difference produced by adopting these two standards, by comparing the pictures of Watteau with those of Rubens. The figures of the former, having eight heads instead of seven, are more graceful than those of the latter, notwithstanding the wonderful power of execution and colouring exhibited by that great master. It should also be remembered that some of the great artists - and this specially applies to the sculptors - varied the proportions, and even totally falsified them, because of the peculiar circumstances under which their work was to be viewed.

Thus, as I have already mentioned, Michael Angelo made some of his stooping figures as much as twelve heads; and, as Topinard points out, if the head was to be seen from below and in perspective, being placed in an elevated situation, it was increased in size, and the body was made to contain it no more than six times. With regard to differences in proportion between the male and female head, there is some variety of opinion. Quetelet says that the male head is contained 7.4 times in the stature, and the female 7.2, thus the head in woman is somewhat longer proportionately than in man. Topinard also says that in general the head is higher in women than in men, and that this is probably the case in all races. On the contrary, Marshall makes his female figure contain exactly sewn and a half heads, and his male seven and four-ninths, the former thus having proportionately a smaller head. The following relations between the different parts of the face are given by Quetelet, who says: 'We may remark an admirable harmony which exists between the principal parts of the human physiognomy. Each of its essential parts has an extremely simple relation with the neighbouring portion, and this harmony is so striking that it cannot escape the most superficial observation, even without the aid of measurements.

Thus, artists have well recognised that in a regularly proportioned body the size of the eye is equal to the distance between the two eyes; it is also equal to the length of the nose. This proportion is so simple, and at the same time so constant, that it enters into the first notions of design. It has, perhaps, been less remarked that the ear, an organ apparently of little importance and of irregular form, remains at all ages exactly equal to the size of the two eyes. The measurement must be taken in the direction of the greatest size of the ear. This rule is subject to so few variations that in my tables the greatest differences in the averages do not amount to more than a millimetre; this regularity is still more remarkable since the ear is of all the organs of sense that which attracts usually the least attention. The size of the ear is also half the distance froms its opening to the summit of the head. A relationship not less curious is that which exists between the size of the eye and that of the mouth, the values being in the ratio of two to three. This relation is absolute at the period of puberty; the mouth is smaller in infancy on account of the fatness of the cheeks; it becomes a little larger at a more advanced age.

These relationships can be pushed still further, and it will then be found that the eye is contained five times in that diameter of the head which is taken through the temples, and seven times in the antero-posterior diameter.

Neck

When we come to consider the measurements of the remaining portions of the axial part of the body, we are met with the difficulty that different observers have not always taken the same points for their observations, which makes any comparison of them exceedingly difficult. This is especially the case in connection with the measurements of the trunk proper, as we shall shortly have occasion to notice; but it is not less true of the neck. According to Quetelet, this is defined as being the area included between two parallel lines drawn, the one below the chin, the other above the point of junction of the clavicles. This is a trifle higher than the measurements which are taken, as in Topinard's work, to the suprasternal fossa, but so little so as to be negligable in the case of artists. Marshall makes the length of the neck in the male three units, or one-third of a head; and in the female three and a half units, or a little longer, the difference in proportional length being explained by him by the fact that in the female the sternum is placed at a lower level, the clavicles being thus also depressed internally, and the upper ribs have a greater obliquity.

If we compare these measurements with Topinard's standard, which for the neck is 4.2 parts of one hundred, we find that, calculated in the same manner, Marshall's figure would amount to 4.4 for the male, or a little longer than that of the French author. Blanc, on the contrary, makes it one nose or one-fourth head in length, which is too short. The transverse measurement of the neck in the male is four and a half units, or exactly one half head; in the female it is four units. The anteroposterior measurements in the two sexes are live units and four and a half units respectively; thus the female neck is proportionately more slender than that of the male.