Free Brunelleschis Dome by Ross King Page A

diminishing from seven feet at its base to just under five feet at the top. And the exterior shell — added to protect the inner one from the elements as well as to make the entire structure appear più magnifica e gonfiante, “larger and more inflated” — was to narrow from a width of slightly over two feet at its base to just over a foot at the oculus. Likewise the eight vertical ribs at each corner of the octagon were to taper as they rose skyward. And while the dead load of the Pantheon had been lessened by the use of pumice stone and empty bottles, in Santa Maria del Fiore the shells were to be built out of stone for the first 46 feet, then from either brick or tufa, the latter being a light, porous stone formed from volcanic ash. The building program also outlined, albeit vaguely, the incorporation of a number of rings of sandstone beams held together with cramps of leaded iron — the chains that Neri di Fioravanti had envisioned encircling the dome’s circumference. These would be embedded in the masonry and therefore hidden from view.
    It was the twelfth point that raised the most doubt. The wardens agreed that for the first 30 braccia of their height — that is, for a distance of about 57 feet above the drum — both shells were to be built without any scaffold-supported centering. Thereafter, from 30 braccia upward, the dome was to be built “according to what shall then be deemed advisable, because in building only practical experience will teach that which is to be followed.”
    This vital condition reveals the reservations of the wardens in the face of Filippo’s daunting plan. Acceptance of it represented a concession on Filippo’s part, a means of appeasing the nervous wardens by committing himself to building only the first fifth of the dome without centering. If he succeeded, he would then have to make his case to raise the rest of the cupola in similar fashion. He must have felt frustrated by the wardens’ continuing lack of faith, but he may also have felt relieved that he was given some time to consider his plans. It is conceivable that even he was unsure of himself at this early stage. Uncertainty about how to execute his audacious plan, and not simply fear of someone stealing his ideas, may have been one of the reasons why he refused to divulge to the incredulous wardens the secret behind the procedure of vaulting without centering. As late as the summer of 1420, for example, he had still to work out the design of the circumferential stone chains. He would not in fact devise a plan for the first one until June of the following year, barely a month before its construction was due to begin. And plans for the second were not completed until 1425, when yet another model had to be made.
    Neither Filippo’s brick model of the dome nor the cupolas he built for the two chapels could quite have prepared him for the task ahead. It had long been known that architectural models were poor guides to statics, because what worked structurally in a model could not necessarily be achieved when the proportions were magnified. In the Middle Ages and Renaissance, proportionally identical models behaved differently depending on their respective sizes, and scale models were generally misleadingly strong. *
    Given the experimental nature of Filippo’s plan, the 30- braccia limit seems to have been a wise precaution, especially since a sound logic governs the restriction. At a height of 30 braccia the bed joints of the masonry would have risen to form an angle of 30 degrees to the horizontal, or just inside the critical angle of sliding. 3 Friction alone would keep the stones in place up to an angle of 30 degrees, even when the mortar was green; therefore, no centering would have been required until that point. Above that level, however, each course of masonry would incline more sharply, reaching a maximum angle, near the top, of 60 degrees to the horizontal. No doubt it was impossible for the wardens to imagine how