Go back to article: Giovanni Canestrini’s models of Leonardo da Vinci’s friction experiments

Canestrini’s models and his sources

The exhibit from the 2016 Leonardo exhibition in London is shown in Figure 1, together with a schematic diagram. The accompanying display board stated ‘Leonardo systematically studied friction, which he considered would be important for the functioning of machines. This bench allowed him to experiment with the contact between different surfaces, by distinguishing between sliding and rolling.’ The same exhibit had also been included in related exhibitions in Paris (Cité des Sciences et de l’Industrie, 23 October 2012–18 August 2013), Munich (Deutsches Museum, 11 October 2013–3 August 2014) and São Paulo (Federação das Indústrias do Estado de São Paulo, 11 November 2014–10 May 2015). It belongs, as did many of the other models on display in this loan exhibition, to the Museo Nazionale della Scienza e della Tecnologia ‘Leonardo da Vinci’ (MUST) in Milan where it has inventory number 392.

Figure 1a

Colour photograph of a wooden model of a design by Leonardo da vinci for testing friction

Models displayed in the 2016 exhibition at the Science Museum

Figure 1b

Sketch of friction experiment models from designs by Da Vinci

Sketch of models comprising the main table A and the three devices labelled B, C and D. The table A is 1.88 m long and 0.81 m high

Figure 2a

Colour photograph of a friction experiment model from a design by Da Vinci

Model A constructed for the 1953 Milan exhibition

Figures 2b-d

Triptych showing three of the wooden models from designs by da vinci on display in the science museum

Models B, C and D

The exhibit consisted of four separate models, although these components all have the same MUST inventory number. The largest was the ‘worktable’, 1.88 m long, 0.81 m high and 0.85 m wide, which we shall call Model A. It supported at one end a rectangular wooden box attached to a string which passed over a cylinder (195 mm in diameter and 320 mm long) on an inclined plane to a hanging weight, and at the other end a horizontal cylinder (237 mm in diameter and 450 mm long) lying in a hemi-cylindrical cavity, again attached to a weight by a string. On the rear left corner of the table lay a wooden plank supporting a stack of shorter wooden blocks, of which one was attached to a long T-shaped handle. The view of these blocks is obscured in Figure 1(a) and they are not depicted in Figure 1(b). On the table also rested three separate devices: Models B, C and D. The models are all labelled in Figure 1(b).

These four models were created by Canestrini for the major exhibition in Milan in 1953, ‘Scienza e tecnica di Leonardo’, and were subsequently donated to MUST where they form part of the permanent collection (Giorgione, 2015). Separate images of the four models are shown in Figure 2. All are based, with varying degrees of fidelity, on drawings contained in Leonardo’s notebooks. Model A is based on sketches from folio 41r of Codex Arundel, which is reproduced in Figure 3. Models B, C and D are based on diagrams found on folio 11v of MS L of the Bibliothèque de l’Institut de France (commonly referred to as ‘Paris’ manuscript L), shown in Figure 4.

Figure 3

Colour photograph of a page of Leonardo da Vincis Codex Arundel sketch book

Codex Arundel f. 41r

Figure 4

Colour photograph of a page of Leonardo da Vincis Codex Arundel sketch book

Paris MS L f. 11v. The text reads ‘Quale di queste fia di più facile moto e quanto: o a o b o c

Giovanni Canestrini (1893–1975) had a long career as a motor racing journalist, and is famous as one of the founders of the Mille Miglia road race in 1926. He also wrote three substantial pieces on the contributions of Leonardo da Vinci to mechanics, particularly in the context of the development of the motor car. His first essay (Canestrini, 1938) was contained in a volume on the Italian contribution to the evolution and development of the motor vehicle, published by the Reale Automobile Club d’Italia (RACI). It was a detailed account which showed familiarity with much of Leonardo’s writing as well as evidence of wide reading of other sources, and despite the apparent narrowness of the chapter title (‘Leonardo da Vinci and the problems of locomotion’) attempted to show that Leonardo’s contributions had pre-dated and indeed informed the work of later inventors in many fields ranging from geometry and optics, to statics and dynamics, fluid mechanics and hydraulics, military engineering, mechanical devices, metalworking and other areas. In discussing Leonardo’s studies on friction, Canestrini reproduced MS L f. 11v and identified it as showing ‘studies of rolling friction’[1], but while he quoted a statement about friction from Codex Arundel f. 41r[2], he did not reproduce any sketches from that folio.

In the following year Canestrini published two fuller accounts of Leonardo’s work on friction. One (Canestrini, 1939a) was contained within a contribution entitled ‘Leonardo’s machines’ in a collection of essays published in conjunction with the major exhibition of Leonardo’s work and Italian invention held in Milan in 1939. The other (Canestrini, 1939b) was a book entitled ‘Leonardo constructor of machines and vehicles’ in which one of the three chapters was on ‘Friction and traction’. In both works Canestrini reproduced the whole of folio 41r of Codex Arundel. He wrote in the first ‘Leonardo is the first who braved the systematic study of the causes of friction, both for the case of flat surfaces and for the case of axles, preceding by two centuries Amontons (1699) and Coulomb (1781), who reproduced, for his own experiments, the bench which we find drawn on folio 41r of Codex Arundel’[3]. In the second the image of the folio was captioned ‘bench for experiments on friction’[4]. He also reproduced MS L f. 11v, describing it in the two publications as ‘axles rotating on rollers’[5] and ‘studies on the friction in axles’[6].

As discussed by Giorgione (2015), the first mechanical models to be based on Leonardo’s drawings were constructed for exhibitions in Florence in 1929, and in Chicago in 1933. These few early models related to aeronautics and textile spinning. Subsequently around 200 models were created and exhibited in 1939, in the context of a major celebration of Italian invention in Milan which included an exhibition devoted to Leonardo da Vinci and was intended by the Fascist government to demonstrate Italian achievements and supremacy in technology[7]. Giovanni Canestrini was one of the engineers involved in designing these models, many of which were built by the RACI, which was heavily involved in organising the exhibition (Giorgione, 2015). Both the official guide to the 1939 exhibition (Guida, 1939) and the catalogue which was published after the exhibition (Catalogo, 1939) list two different models relating to friction displayed within the section on the ‘mechanical arts’: ‘Reconstructed model of bench for experiments on sliding and rolling friction’[8], and ‘Reconstructed model of experimental apparatus for the study of the action of forces and friction in a rotary system’[9]. The first of these was the precursor to Model A that Canestrini created for that exhibition (which we shall call Model A*). There are photographs of Model A* both in his book (Canestrini, 1939b), captioned ‘reconstruction of bench for experiments on friction according to Leonardo’s drawings’,[10] and also in the exhibition catalogue (Catalogo, 1939) with the caption ‘bench for experiments on sliding and rolling friction. Model reconstructed from Codex Arundel f. 41r’[11]. Figure 5 shows one of these images. In all essential elements it was identical to Model A, although in Figure 5 the blocks and plank are arranged to show a block being pulled by its handle (just visible) down an inclined plane.

Figure 5

Black and white photograph of a wooden model of a design by Leonardo da vinci for testing friction

Picture of Model A* from Canestrini (1939b) captioned ‘Ricostruzione del banco per le esperienze sull’attrito secondo i disegni di Leonardo’. An identical photograph is to be found in the catalogue of the 1939 exhibition (Catalogo, 1939, tav. 19).

The second model was not illustrated in the exhibition guide or catalogue, but an image is to be found in Canestrini (1939b, p 160), shown here as Figure 6(a), with a caption that closely parallels the entries in both the exhibition guide and the catalogue: ‘Reconstructed model – from drawing by Leonardo – for the study of the action of forces and friction in a rotary system’[12]. This model, which we shall call Model E, is also shown in a leaflet advertising the exhibition (Leaflet, 1939) and depicted in Figure 6(b). The accompanying caption reads misleadingly ‘Cylinder clutch’ or more literally, ‘Clutch made from cylindrical elements’[13]. Model E was based on the sketch in the Codex Atlanticus (f. 1081v) shown in Figure 7, which Canestrini also reproduced in two of his publications. In Canestrini (1938) he described this diagram as ‘thrust rollers in a drawing by Leonardo’[14], while in Canestrini (1939a) it was a ‘system of rotating bearings with thrust rollers on spindles’[15].

There is no record in the catalogue or guide to the 1939 exhibition of any models similar to Models B, C or D, and we must therefore assume that these were created for the first time for the 1953 Milan event.

Figures 6a & b

An illustration and a black and white photograph of one of Castrinis models

Figure 6(a) shows an illustration of Model E from Canestrini (1939b) captioned ‘Modello ricostruito – su disegno di Leonardo – per lo studio dell’azione delle forze e dell’attrito in un sistema rotoide’. Figure 6(b) shows an illustration of model E from a booklet advertising the 1939 exhibition, captioned ‘Frizione a cilindri’ (Leaflet, 1939)



Figure 7

Colour photograph of a page of Leonardo da Vincis Codex Arundel sketch book

Drawings from Codex Atlanticus f. 1081v (part of page) (Biblioteca Ambrosiana, Milan). The text beneath the large diagram reads ‘Qui è peso e forza. Il peso va perpendicular e la forza per lo traverso. Domandasi se ‘l peso e la forza insieme giunti sono equali al peso del polo a o se pure la forza è il proprio eccesso sopra il naturale peso.’

Component DOI: http://dx.doi.org/10.15180/160602/002