The 1966 Florence flood: books, art and work still unfinished

Exceptional rain on 3 and 4 November 1966 swelled the Arno and rivers in neighbouring Tuscan basins. V01 V13

The Comune di Firenze reports a basin average of 160 mm over 24 hours. Local peaks reached 250–300 mm. V01

The De Marchi commission gives a little over 200 mm basin-wide over about 28 hours. Its window and method differ; both sources describe the same storm. V13

The 751 mm maximum over 48 hours belongs to Barcis, in the northeastern Alps — not Casentino. For the Arno basin, the study gives more than 300 mm over two days. V01

Peak flow changes with the station. The Comune di Firenze reports 4,100 m³/s at Nave di Rovezzano; the official hydrological yearbook records 3,540 m³/s at Nave di Rosano. These are different stations, and the published records do not support reading them as one settled flow for the city. V08

Water left marks at different heights. Civil Protection records 4.92 m on Via dei Neri; at Santa Croce’s refectory, the level is 5.02 m. The Uffizi reports about 6 m at the Cimabue Crucifix’s location. Each height stays tied to its place. V03 V16

The prefectural list commemorated by the municipality counts 35 deaths in Florence and its province: 17 in the city, 18 in the province. V11

CNR-IRPI counts 47 in Tuscany, including 38 in Florence province. Venice stays outside both totals. V11

The Biblioteca Nazionale Centrale di Firenze (BNCF), Florence’s central national library, describes almost one million affected bibliographic units. V11

Its 2026 breakdown: about 300,000 books, 460,000 volumes of newspapers and periodicals, and 400,000 pamphlet units. The library does not call all of them “books”; the categories stay separate, without a more precise total than the one it publishes. V11

The first washing and drying began at Florence railway station’s central heating plant. V03 V06

The books were then dried and stored at Forte Belvedere. Fortezza da Basso hosted the separate art-restoration operation. V03 V06

Rescue took the shape of a work sequence: sorting and collation, disbinding, washing and other wet treatments, paper repair, sewing, rebinding. This production-line effort helped make high-volume book conservation an institutional practice. V05

Water, fuel oil and mud left different marks on the artworks. Santa Croce and the Uffizi estimate that the Crucifix by Cimabue, a Florentine painter, lost about 60% of its painted surface. The figure describes this one work. V02

For movable works as a whole, CNR uses a broad estimate: about 1,500 affected objects. V06 V11

The inventory has limits. It does not automatically include frescoes, furnishings, liturgical objects or every architectural surface. V06 V11

The Uffizi’s photographs and recovery lists preserve uneven damage and specific interventions. They show the scale without combining unlike collections and categories under one denominator. V05 V06

Thousands of volunteers came from Italy and abroad. They cleared mud and recovered books and artworks. Institutional accounts support the scale of the mobilisation, not a complete census. V04

Angeli del fango — “Mud Angels” — became their name later. Giovanni Grazzini is often credited with the phrase, but its exact first use remains disputed. V04

The name’s origin remains disputed; institutional accounts preserve the documented work: mud cleared, books and artworks recovered. V04

Ugo Procacci’s modern restoration cabinet had existed at the Opificio delle Pietre Dure, Italy’s state restoration institute, since 1932, more than three decades before the flood. V07

The works’ immersion sharply increased the volume of work and collaboration. Pressure also grew to formalise scientific and high-volume conservation methods. V07

The modern Opificio took shape in 1975 by merging the historic institution with Procacci’s cabinet. V07

Vasari’s Last Supper at Santa Croce spent more than 48 hours underwater. Its restoration ended with the work’s return in 2016. V07

The BNCF says about 30,000 miscellany units remain inaccessible: their shelfmarks or bibliographic identities were lost. V14

A project begun in 2026 is working to identify them and restore their place in the catalogue. V14

The figure tracks identification and access; it does not count material left physically unrestored. Physical treatment and bibliographic recovery are different kinds of work. V14

On 30 September 2026, the Comune di Firenze declared its urban Arno protection system complete. V15

The package covers about 4.5 km of riverbank interventions, parapets, permanent containment and removable panels. The panels depend on forecasts and timely installation. V15

Upstream, works around Figline — Restone, Pizziconi, Prulli and Leccio — and the raising of Levane remain in design or implementation. Protection has improved and risk has been reduced; risk has not been eliminated, and the whole basin has not been declared safe against every event. V15 V09

In Venice, the tide separately reached 194 cm at 18:00 on the same day. Another city, another water system: that figure stays outside Florence’s flood tally. V10