¹ We were very fortunate to receive considerable help in Split and Trogir from Cvito Fisković and Ivo Babić, who ensured the success of our visits, We were also very greatly helped by the cathedral authorities of Trogir, by Sheila McNally and Janet Buerger of the Split Excavation Team, and by Tomislav Marasović, Architectural Historian of the Town-Planning Institute, Split. The project was the result of John Ward-Perkins's interest in architects' methods of work, and he provided invaluable help and encouragement throughout, In 1975 we were most grateful to be financed by a Michael Ventris Memorial grant from the Architectural Association. Our attention was drawn to Trogir by the late Xenia Radulić, whose death was a great loss to Dalmatian studies and to whose memory we would like to dedicate this article.

² The parapets and tiled roofs must be later than the drawings, since the parapets cover some of the incised lines, and since over the north aisle the drawings are now very badly lit and can only be seen at a stoop; here at least they must have been incised when the whole area was open. That the flag-stoned aisle roofs were originally open to the sky is proved by the outward fall for drainage and by the gutters, rendered useless by the addition of the tiled roofs. The tower-drawing in the south aisle (drawing O), the last datable one in the series, could perhaps be later than the addition of the new roof.

³ Iveković, i, p. 6.

⁴ Iveković, i, p. 24. See also Lucio, pp. 448–9 and Farlati, p. 398.

⁵ This inscription is now lost. It is, however, reproduced in Jackson, ii, p. 138, fig. 50.

⁶ The first storey above the porch is probably late fourteenth-century, with additions by Goyković, and the second storey built (or rebuilt) entirely in this period 1420–1 (Folnesics, pp. 29–33). It is very unclear if the cathedral was originally intended to have a tower in this position. The south end of the thirteenth-century porch, which supports the tower, does not seem to have been built particularly strong; in 1909 it was necessary to reinforce it (Iveković, i, p. 26).

⁷ Ed. Fisković, pp. 35–6, records contracts in 1427, 1431, and 1437. Lucio, p. 488, followed by Iveković, i, p. 25, records only the first two contracts.

⁸ Ed. Fisković, p. 36. Dated to 1440 by Lucio, p. 488; followed by Iveković, i, p. 25.

⁹ Ed. Fisković, pp. 17 and 47–8. Also Lucio, p. 488. Iveković, i, p. 25, quotes part of the contract. Wrongly dated by Jackson, ii, p. 127, to 1458.

¹⁰ Ed. Fisković, pp. 44–5 which is probably reliable. Dated by Lucio, p. 488, to 1446–50, and by Iveković, i, p. 25, to 1446–52. Jackson, ii, p. 125, says the building cost 4020 zecchini. In 1460 Andrea Alessi, later to work on the baptistry and Chapel of St. John, was paid for some work here: probably minor repairs (ed. Fisković, p. 45).

¹¹ Ed. Fisković, pp. 44–5. Dated to 1458 by Lucio, p. 488, and Iveković, i, p. 25.

¹² Jackson, ii, pp. 133–4, says the building cost 4980 zecchini.

¹³ Ed. Fisković, p. 41; Iveković, i, p. 26; Jackson, ii, pp. 128–30; Lucio, p. 488. The contract is published by Kolendié.

¹⁴ Ed. Fisković, pp. 43–4: these statues were commissioned between 1482 and 1559. The building was probably far advanced by 1472, when both Alessi and Fiorentino accepted a contract for work in Split (Iveković, i, p. 26).

¹⁵ Ed. Fisković, pp. 42–3. The only original lighting was through the oculus windows above the niches and through the entrance into the chapel. This lighting has been restored by the placing of curtains over the four later windows.

¹⁶ Fisković, C., ‘Tri Angela Nikole Firentinac’, Nova Tiskara, 1940, pp. 3–8Google Scholar.

¹⁷ Ed. Fisković, p. 44.

¹⁸ Published by Fisković, C., Trifun Bokanii graditdj zvonika trogirske katedrale, Split, 1940, pp. 10–11Google Scholar .

¹⁹ Lucio, p. 488; Iveković, i, p. 26.

²⁰ This is to be found at about eye-level of the human figure in fig. 3.

²¹ The trefoils of the blind arcades on the first storey are about 70 cm. wide and those in the windows (also in the first storey) are 36 cm. wide.

²² These consist of two very small pointed arches and a larger round one. These could not be immediately identified with structures inside the church.

²³ Unfortunately it was not possible to take measurements at the level of the vaulting, which would prove conclusively (or disprove) this identification.

²⁴ On our drawing of the elevation as built the height is approximated; in reality this height might well be much closer to that which appears on the incised drawing.

²⁵ Ed. Kolendié, p. 74 sq.

²⁶ From the contract of 4th January 1468 as published by Kolendić: ‘… che Ii dene romper el muro, dov' e l'altar de Sancta Orsola ne la giesia cathedral da uno pilastro al'altro, et 11 de esser un scalin, sopra el qual de esser doi pillastri de grosseza de pie un et mezo da la faza davanti dentro da la giesiae per l'altra faza grosso pie doy, altii diti pillastri pie XVII, et sopra quelli uno archo in mezo tondo, el diametro suo pie X ve l circa, grosso per la faza davanti un piee mezoe per l'altra pie doy; e ne la straia del dicto archo a chace una figura de piedi 3; el dicto archo de esser intaiado da tute due faze, come apar nel desegno;…’ We are very grateful to Professor Fisković for obtaining for us a copy of this contract.

²⁷ Iveković, i, p. 25, thought the drawing was a project never put into effect by Matej Goyković, the early fifteenth-century architect of much of the lower part of the tower; but this idea has rightly been dismissed by successive scholars: Jelić, p. 30 and Folnesics, p. 33, note 1.

²⁸ See Branner, R., ‘Villard de Honnecourt, Reims and the Origin of Gothic Architectural Drawing’, Gazette des Beaux Arts, ser. 6, lxi (1963), 129–46Google Scholar, with a very useful bibliography. Recently Professor Lucos Cozza has identified a similar incised drawing of Roman date outside the Mausoleum of Augustus in Rome; this awaits publication by Sheila Gibson and John Ward-Perkins. One of the finest groups of drawings (that on the ambulatory-terrace of Clermont-Ferrand Cathedral) is as yet unpublished.

²⁹ For instance, in England, at Wells and York: Harvey, J. H., ‘Wells Cathedral’, Arch. Journ. cxxxi (1974), 214 and fig. 2Google Scholar; Colchester, L. S. and Harvey, J. H., ‘The Tracing Floor in York Minster’, 40th Annual Report of the Friends of York Minster for 1968, pp. 9–13Google Scholar .