The vesica piscis (a figure formed by the developing arcs of two equilateral triangles having a common side) which in so many cases seems to have determined the main proportion of a cathedral plan—­the interior length and width across the transepts—­appears as an aureole around the figure of Christ in early representations, a fact which certainly points to a relation between the two (Illustrations 42, 43).  A curious little book, The Rosicrucians, by Hargrave Jennings, contains an interesting diagram which well illustrates this conception of the symbolism of a cathedral.  A copy of it is here given.  The apse is seen to correspond to the head of Christ, the north transept to his right hand, the south transept to the left hand, the nave to the body, and the north and south towers to the right and left feet respectively (Illustration 44).

[Illustration 47]

The cathedral builders excelled all others in the artfulness with which they established and maintained a relation between their architecture and the stature of a man.  This is perhaps one reason why the French and English cathedrals, even those of moderate dimensions are more truly impressive than even the largest of the great Renaissance structures, such as St. Peter’s in Rome.  A gigantic order furnishes no true measure for the eye:  its vastness is revealed only by the accident of some human presence which forms a basis of comparison.  That architecture is not necessarily the most awe-inspiring which gives the impression of having been built by giants for the abode of pigmies; like the other arts, architecture is highest when it is most human.  The mediaeval builders, true to this dictum, employed stones of a size proportionate to the strength of a man working without unusual mechanical aids; the great piers and columns, built up of many such stones, were commonly subdivided into clusters, and the circumference of each shaft of such a cluster approximated the girth of a man; by this device the moulding of the base and the foliation of the caps were easily kept in scale.  Wherever a balustrade occurred it was proportioned not with relation to the height of the wall or the column below, as in classic architecture, but with relation to a man’s stature.

[Illustration 48:  FIGURE DIVIDED ACCORDING TO THE EGYPTIAN CANON]

It may be stated as a general rule that every work of architecture, of whatever style, should have somewhere about it something fixed and enduring to relate it to the human figure, if it be only a flight of steps in which each one is the measure of a stride.  In the Farnese, the Riccardi, the Strozzi, and many another Italian palace, the stone seat about the base gives scale to the building because the beholder knows instinctively that the height of such a seat must have some relation to the length of a man’s leg.  In the Pitti palace the balustrade which crowns each story answers a similar purpose:  it stands in no intimate relation to the gigantic arches below, but is of a height convenient for lounging elbows.  The door to Giotto’s campanile reveals the true size of the tower as nothing else could, because it is so evidently related to the human figure and not to the great windows higher up in the shaft.