Franois Denis

The geometric principles

of string instrument-making in Brescia
Franois Denis

We know how craftsmen of former ages used geometry from the evidence of various practical
handbooks and treatises, one of which, written by Henri Arnaud de Zwolle in the 15th century, is par tly
devoted to the lute.1 Detailed study of these documents bears witness to the ancient ar t of drawing with
ruler and compass. Long nur tured by the oral traditions of craftsmanship, the skill fell into disuse during
the 17th century.
To reveal these now vanished processes, it is sufficient to compare the measurements of an instruments
different dimensions.2 We should remember, however, that such relations are legion and their limits often
ill-defined. Consequently, interpreting them is a much more complex business than might be supposed
from the primary evidence of their analysis. Too much data can conceal the bigger picture; a selection
must be made, informed by familiarity with the history of science, technology and the ar ts.
To complicate matters fur ther, the notions of measurement and geometry are not the same now as
they were in the Renaissance. An objective account of the techniques for ruler-and-compass drawing
used in those bygone days therefore requires some prior explanation.
Once these precautions have been taken, the instruments dimensions are like a forgotten alphabet
whose surprising elegance questions the whole idea of progress in instrument-making.

Picture of a lute in Spain c. 1260.

In Islamic culture it was the
instrument of the elite and was
the first to benefit from elaborate
construction techniques.

The only plan of a musical instrument to have survived

from the 15th century, we owe this drawing of a lute,
together with a description of its measurements and
the principles for building it, to the astronomer Henri
Arnaut de Zwolle. It contains information essential to an
understanding of all instrument-making until the 17th
century.

317

Franois Denis

FRAMEWORKS
Three ages of the history of forms

During the geometric period

compass in pictures commonly
symbolises reasoning and
measuring.
(Melancholy - Drer)

Three periods may be distinguished in the history of the forms of bowed and plucked string instruments
from the Middle Ages to the present day. From the 13th to the mid-17th century, craftsmen made their
models with ruler and compass according to the art of measurement.3 This first, so-called geometric
period was followed by a transitional period lasting about a hundred years, corresponding to the golden
age of Italian instrument-making. Despite its glorious association, it was during this post-geometric
period that the techniques of compass drawing started to decline with the spread of increasingly empirical
methods.4 The transition ended in decadence as the ancient knowledge and skills vanished completely. It
was followed by the romantic period, which extends from the mid-18th century to the present day. This
period, in which the creation of forms depends entirely on the outlines of existing instruments, also saw
the emergence of the mythical (and rival) figures who would henceforth inform both the imagination
and the organisation of the profession.
The basics
At the time when compasses were still widely used, designing an object consisted in pragmatically
defining the limits of the whole and its parts.5 The outline of a bowed or plucked string instrument was
inscribed within an organised surface called a framework. This organisation considered only the ratios of
adjacent dimensions6 of length and width.

Each of these relations corresponds to a single manipulation of the ruler or compass and forms part of a logical
sequence. All the dimensions therefore depend on a clearly defined order of operations. It is these various procedures
that the Ancients called measuring.

318

Franois Denis

The different formats of a form have been allocated

between the square and the double square according to
the measurement of the added part.
After S. Serlio

The quadrature of the square represents the archetypal

vertical and horizontal division of a surface. It is also the
basis for proportional symmetry, a notion which assumes
various meanings that in turn provided the basis for a
balanced composition of form.
The sketch-book ofVillars de Honnecourt, 13th century.

The Rheims palimpsest is one of the few surviving plans

of a medieval cathedral. It should be read from right to
left and from top to bottom. It shows the different stages
of design, beginning with the most general relations and
ending with details of the decoration.
Rheims, City Library.

319

Franois Denis

Violins 7
Surface and vertical divisions
The proportions of Brescian instruments apply to the dimensions of the outline of the edges, whereas
in Cremona the significant dimensions are those within the ribs. As we shall see below (Notes on making
process), this difference is probably due to the construction method. The violins were made in two sizes
according to an identical geometrical model.Violins N14 and N11 are good illustrations of this principle.
The quotients of the different ratios of length and width and of the lengths of the lower, middle and
upper parts (Table 2 in the annex) produce a definition of so-called harmonic surfaces.8

The construction of violin N14 starts with a square

with sides equal to XQ divided into five parts. Rotating
the half-diagonal downwards defines the rectangle that
frames the form and the dimension of XP, the height of
the lower part of the form. The midpart XZ measures
two parts of the square (i.e. a ratio of 2 : 3 between the
midpart XZ and the upper part ZQ).

320

Though longer by 8mm and wider by 3.5mm, the

proportions of violin N11 are identical to those of the
previous instrument.

Franois Denis

Horizontal divisions - Placing and measuring the f-holes

The relative dimensions of the width are determined according to a geometrical procedure similar to
the one used by the Amati at the same period. The maximum width of the lower part pp is the starting
dimension, from which the maximum spacing between the f-holes aa is derived. It is therefore always
relative to a division of pp. The minimum spacing between the f-holes bb is equal to half the difference
between pp and bb. It is determined by the simple construction shown in the diagram below. The logic
of this ordering of widths is apparent from a study of the ratios given in Table 3 in the annex.

The methods for determining

the widths and the openings of
musical instruments were related
to those used in architecture at the
same time.

With violin N14, the width is divided into 8 parts, 5

parts being given to aa, the maximum spacing between
the f-holes, and 1.5 parts to bb, the minimum spacing
between the f-holes.

The relative dimensions of the f-holes are taken

from Table 4 in the annex.The geometrical method
is borrowed from the architecture of faades. The
openings are placed on a plane according firstly to
their two principal dimensions, i.e. the height and
width of the surfaces on which they are inscribed,
and not the measurements made from the centres
of the holes. Thus, the two f-holes of violin N11
are situated in a 3 by 5 rectangle, while those of
violin N14 are situated in a 7 by 10 rectangle.
It is likely that each f-hole was initially inscribed
in a given surface, but the scale of the asymmetries
restricts any interpretation on this point.
More generally, these relations deriving from
the initial division into 5 or 8 of the instruments
greatest width appear to owe much to the
proportional relations attaching to the Fibonacci
sequence. 9

Using a proportional section to

measure a door.
(After S. Serlio).

321

Franois Denis

The 3 to 5 proportion between pp and aa is found again

in the measurement of the main dimensions of the f-holes,
since they are inscribed in a rectangle with the same 3
by 5 relation.

The height of the f-holes is measured by 3.5 of these

parts, such that the two f-holes are inscribed in a 5 by
3.5 (10 by 7) rectangle. As before, each f-hole is itself
inscribed in a double square.

The division of widths continues in an elegant manner. The maximum width of the upper part qq
derives from a division into 2 of the space between the vertical lines passing through p and a (and
symmetrically for the space between a and p), lastly ee, the minimum width of the midpart, is equal
to two-thirds of qq.

Geometrical construction of qq, the maximum width

of the upper part, and ee, the minimum width of the
midpart (violin N14). In accordance with the principles
of symmetry, still perceptible even though the making
process is often approximate, bb (the minimum spacing
between the f-holes) is a measurement close to the
difference between pp (the maximum width of the lower
part) and qq (the maximum width of the upper part).
bb is also equal to half the difference between pp and
aa (the maximum spacing between the f-holes). This
ordering of the framework, identical to that used by the
Amati, may be regarded as one of the oldest of the violin
family.
322

Franois Denis

Two violas
Violas N5 and N6 of this exhibition are two large instruments which have not been recut, a relatively
rare occurrence. Superposing the two outlines shows the similarity of their model.
Surface and vertical divisions

Construction by the diagonal of

the square.
(After S. Serlio)

Comparison of the outlines of violas N5 and N6

The outline of viola N6 is inscribed in a 7 by 4 rectangle

(table 5 in the annex). The remarkable feature of this
framework is the placement of X (XP : XQ = 0.414).
Diagonals are used to construct the figure.The point Z is
situated by dividing XQ into 5.

Example of a drawing process similar to that of the viola

applied to the construction of facades. (After Francesco di
Giorgio Martini)
323

Franois Denis

The outlines of violas N17 and N18 (and N10) are also inscribed in harmonic rectangles (see Table 7
in the annex). They form a group whose similarity derives from the symmetrical placement of the points
X and N, respectively the limit of the lower corners and the minimum width of the midpart (see Table
8 in the annex).

It is known that diagonals were

used to set type in the earliest days
of printing. This Gutenberg bible
is also based on a division of the
height into nine parts.
(After Adolf Wild, Cahiers de
Gutenberg no. 22, 09/1995)

The width pp being divided into 10 parts, construct a

harmonic rectangle (i.e. PQ = 17 parts), q is placed one
part in, Qq is half the width of the upper part. Draw a
diagonal joining p to q (at the level of the horizontal
lines passing through P and Q) to place N, the midpoint
of the minimum width of the midpart. As a result, the
length is divided into 9 parts at the level of the line PQ
and the point N is situated 4 of those parts below Q.
Place X below N such that PX equals NQ.

324

A diagonal drawn from p to Q or from A to q cuts the

diagonal through X at the level of the horizontal line
through Z, the limit of the upper corners (with XZ is to
ZQ as 5 is to 9).

Franois Denis

Commentary
One interesting aesthetic consequence of this geometric method is that ee, the narrowest width of
the midpart, descends relative to P when the width of the upper part qq increases relative to that of the
lower part pp. Consequently, the overall aspect of the form is directly linked to the relative widths of
the upper and lower parts.

Example of the use of diagonals to

apply proportion to figures.
(AfterVillard de Honnecourt,
13th century)

Superposing the outline of viola N6 (in black) with that of the Charles IX tenor viola (Ashmolean Museum) reduced
to the same length (PQ) shows a similarity of contour and proportionality.The great width of the upper part relative
to the lower part descends the point N towards P. Consequently, the upper part of the midpart curve is more open.

325

Franois Denis

Horizontal divisions - Placing and measuring the fholes

The same geometric processes already described for violins are used here. The organisation starts
with a division of the maximum width of the lower part pp, on which the measurements of the f-holes
depend, and continues with the relative widths of the upper and midpart.

The widths of violas N5 and N6 (table 6 in annex)

are calculated from a division into 8 parts of the width
pp. The maximum spacing of the f-holes aa is equal to
5 of these parts. The ratio of 5 to 8 echoes the famous
Fibonacci sequence, which is elegantly associated here
with the homothetic relations of the triangles.

This recommendation for placing the opening of a faade

(after Serlio) is an illustration of the principles applied
to the construction of musical instruments. The surface
(here the double square) is divided by a proportional
section (here the arithmetic section).

326

Franois Denis
The horizontal organisation continues with a division of the remaining adjacent and symmetrical
parts. Only the spaces between the verticals of the previously defined widths meet these two conditions.
Various possibilities exist, including the one shown below for the two violas N5 and N6.

The space between the vertical lines passing through b

and p is divided into 5 parts: the space between b and
q is equal to 4 of those parts and qq is the maximum
width of the upper part. The space between the vertical
lines passing through q and q is divided into 3 parts.
ee, the minimum width of the midpart, is equal to 2 of
those parts.

These symmetrical processes (in the proportional sense of the term) could give a paradoxical explanation of the troubling setting in f-holes to be found on many instruments of the Brescia school.
The variable orientation and length of the f-holes seem to depend on the quality of the parallelism
between the lines connecting the corners.

327

Franois Denis

The model for the organisation of widths of the N17 and N18 violas is deduced from the relations
given in Table 9 in the annex.
The maximum spacing of the f-holes aais to the maximum
width of the lower part pp as 5 is to 9.This ratio echoes
the similar divisions of the length and corresponds to the
geometrical process illustrated opposite.

As previously, ee, the minimum width of the midpart

of these two instruments, is equal to two-thirds of qq,
the maximum width of the upper part. Note that with
viola N18, the maximum spacing of the f-holes aa is
also equal to ee, the minimum width of the midpart.

Commentary
The latter two instruments, remarkably constructed on the interval of the major third (4 to 5), seem
to pay visual tribute to mean-tone. They provide evidence that instrument-makers at the time were keen
to associate harmony of form with that of music.
328

Franois Denis

Drawing outlines
A reminder of techniques for drawing with a compass
The outlines of instruments (except for the lute, which uses the ellipse) consist of a succession of
circles. The radii of these circles correspond to the spaces between the divisions marked on the axes. The
limits of the drawing are set by the framework.
The radii form a set of articulated segments that operate like the members of a body. In this sense, the
line is the trajectory of a point and drawing the outline is a geometry of movement.11
The radii of the arcs (R1) that draw the top and bottom
blocks have the points X, N or Z as their centre.The radii
of the arcs (R2) that extend those of the blocks measure
half the height XP (or ZQ for the upper part) or a third
of the width (i.e.1/3 of qq or 1/3 of pp). The radius
of arc R3 tangent to R2 that rises towards the corner is
double that of R2. The dotted segments are the hidden
radii that govern the construction.

Drer uses the ovum to illustrate

the principle of drawing arcs.
The figure is constructed from the
divisions of an orthonormal basis.

Drer devised a way of drawing

any curve using an articulated
system that mimics the working of
a human arm. The construction of
a violin outline uses geometrical
principles identical to those
applied by this peculiar device.

As when drawing an ovum, the measurements of the

arc radii that generate the outline of a violin will be
determinated by the axis divisions.

Another frequent feature of Brescian instruments is the way in which the radii of the corner arcs are
tangent to the horizontal line passing through X or Z.
This information is sufficient to draw the outline. From the following examples, we can see the
recurring patterns at the origin of a Brescian style.

The radii of the arcs that draw the outlines of the corners
and the midpart are derived from divisions of XZ.
329

Franois Denis

Examples of outlines
Violin N14
The principal dimensions having been established as indicated, it is remarkable that the approximate
symmetry of these instruments fails to mask the solidity of the geometric principles that underlie the
aesthetics of their lines. Let us remember that the process involves drawing an outline, not a mould. This
nuance has perceptible repercussions on the form itself, while also simplifying certain aspects of the
drawing technique. 12

Diagonals are used to place N, the midpoint of the

minimum width of the midpart, and to determine the
measurements of the radii R2 (here one-third of pp and
qq) tangent to the two maximum widths.This appears to
have been common practice in Brescia.

330

The lower part is drawn by four arcs with the following

radii:
R1 = pp
R2 = 1/3 of pp
R3 = pp
R4 = 1/5 of XP

Franois Denis
The midpart is drawn by three arcs with the following radii:
R1 = XZ R2 = 1/2 of XZ
R3 = 1/2 of ZN

The upper part is drawn by four arcs with the following radii:
R1 = XQ R2 = 1/3 of qq R3 = 2/3 of qq
R4 = 1/4 of ZQ

Outline of violin N14 superimposed on its geometrical

reconstruction (in red).
331

Franois Denis

Viola N6

The way the arcs are drawn is broadly the same for violas.
Again, diagonals are used to determine the radii R2 and
the point N. In the case of instrument N6, the radii R2
are equal to half PX and ZQ.

Outline of viola N6 superimposed on its geometrical

reconstruction (in red).

The lower part is drawn by four arcs with the following radii:
R1 = NP R2 = 1/2 of XP
R3 = XP R4 = 1/3 of XN
The midpart is drawn by three arcs with the following radii:
R1 = XZ
R2 = 1/2 of XN R3 = 1/2 of NZ
The upper part is drawn by four arcs with the following radii:
R1 = NQ R2 = 1/2 of ZQ R3 = ZQ R4 = 1/3 of XN

332

Franois Denis

Viola N18

In this last example, the radii R2 measure one-third of

pp below and half of ZQ above.

Outline of viola N18 superimposed on its geometrical

reconstruction (in red).

The framework having been defined, the lower part is drawn by four arcs with the following radii:
R1 = pp R2 = 1/3 of pp R3 = 2/3 of pp R4 = 1/2 of NZ
The midpart is drawn by three arcs with the following radii:
R1 = XZ R2 = 1/2 of XN R3 = 1/2 of NZ
The upper part is drawn by four arcs with the following radii:
R1 = NQ R2 = 1/2 of ZQ R3 = ZQ R4 = 1/2 of NZ

333

Franois Denis

Notes on the making process

The relic of Saint Catherine

is one of the oldest string
instruments to have survived
intact. A unique record of that
transitional period in which
the intermediate families of
instruments were made, it
combines both the form and
the playing style of two types
of instrument. Frederico
Zuccaris 1608 painting seems
to show the relic with two
bridges, and a dual playing
technique with the right hand
on the bow and the left hand
in the position of a spread
chord.

It was in 13th century Spain, in the cultural melting-pot of AlAndalus, that bowed string instruments,
omnipresent in the Christian world, started to mix with the plucked string instruments emblematic of
the Islamic tradition. From then on, musicians began to use both techniques on the same instruments.
This practice was at the origin of the cutaways that appeared on instruments inspired by Islamic models
and henceforth called vihuela del arco or da mano. At the same time, the simple design of medieval bowed
instruments (three elements, carved and assembled)
came into contact with the more elaborate technique
used by lute-makers. Lute soundboxes were not carved from a block of wood but assembled from skilfully
adjusted strips. In the 15th century, the two methods
of construction corresponding to two styles of playing
(plucked and bowed) began to mingle. These combinations gave rise in the following century to the intermediate families of viols, violins and guitars. The
pictorial record bears witness to this complex period
of transition, showing instruments suited to both plucking and bowing.13

Example of an instrument that can be played both ways

(15th century, France).

This interaction of playing styles and cultures also gave rise to the different types of making that existed in the early 16th century. At that time in Europe, bended ribs had entirely replaced the Gothic sawout technique, though in some places (like Brescia) for a while the overall design remained that of the
medival bowed instrument. In contrast, the techniques used by lute-makers penetrated more deeply in
Cremona, where the use of a mould and a neck nailed to a soundbox was transposed to violin-making.

Interaction of cultural and

technical factors giving rise to
new families of instruments.

334

This study shows that Brescian instruments perpetuated principles already in use in the Renaissance.
Surfaces are correctly constructed using simple measurements that incorporate geometric constructions
based on diagonals. As the tables in the annex show, the match between the actual dimensions of the instruments and their theoretical values is surprisingly close.This finding indicates that the coherence of Brescian
instruments was at least equal to that of the Amatis output and greater than that of a good deal of later
Cremonese work.The marked differences between the instruments produced in the two cities are therefore not attributable to the quality of the craftsmens knowledge but to the ways in which they put it to use.
The measurements apply to the definition of an exterior outline, not to the construction of an interior form. This point supports the argument that instruments were made without a mould, a technique
that owed more to the Gothic model of bowed instruments than was the case for Cremonese instruments
of the same period. In accordance with the Ancients particular conception of symmetry, the Brescian
instrument-makers paid more attention to coherence of measurement than to the outline as such.
A study of these models reveals a treasure-trove of geometric ingenuity and a degree of elaboration
that is in contradiction with the more archaic aspects of Brescian production. It suggests that the violin
did not suddenly appear from nowhere but was the result of three centuries of evolution in instrumentmaking, spanning the whole of Europe and its history.
This distant past, gradually revealed, gives more than the measure of a rich and little-known heritage.
It encourages instrument-makers to see the future of the violin in the continuation of the knowledge that
gave the instrument its worth and not only in sciences alien to its nature.

Franois Denis

Conclusion

Notes
1 For a comprehensive analysis of Henri Arnaud de Zwolles drawing, see Franois Denis, Trait de lutherie, ALADFI, Nice, 2006, pp 48-56.
2 If a width qq =158 mm is compared with another width pp =197.5 mm, the relation between the two widths is expressed by the ratios
158/197.5 = 0.8 or 200/160 = 1.2. These quotients, 0.8 and 1.2, mean that the relation between the two magnitudes is like that of the
numbers 4 and 5.
3 Ibid. note 1.
4 On the empirical use of Amati family forms by Stradivarius and Guarneri del Gesu, see Analyse de la forme ....(to be publish)
5 On this approach to the plan, it is instructive to read Vitruvius, On Architecture, Book III, Definition and necessity of symmetria, and
Albrecht Drer, Gomtrie, translated and introduced by Jeanne Peiffer, Seuil, Paris, 1995, p 139.
6 These ratios are also what we call comparisons. At the time, there was no conception of a ratio without a physical link between the comparable
objects. It follows that a remote comparison has no real meaning, and for that reason relative measurements always concern adjacent
dimensions. Although this principle no longer has any prevalence, it is nevertheless crucial to the conception of form in previous eras.
7 The relations discussed here have been taken from the measurements given in the attached tables and provided by Mr E. Blot and
Mr Y.Gateau. The instruments considered in this study are those whose original dimensions have not been altered.
8 In this case a rectangular surface that can be constructed geometrically by the half-diagonal of the square or one of its approximate
measurements (e.g. 4/7 = 0.571 or 7/12 = 0.583 (the most common approximations are given in Table 1 in the annex). Manipulation of
these values played a decisive role in the origin of the measurements. On these fundamental notions for the history of measurement, see
Franois Denis, op. cit., pp 25-56. Louis Frey, Donnes architecturales et hypothse sur la mathmatique pr-euclidienne, in Bulletin
Antike Beschaving (BABesch), 1989, pp 90-99.
Pierre Gros, La gomtrie platonicienne de la notice vitruvienne sur l homme parfait in Rivista del Centro internazionale di Studi di
Architectura Andrea Palladio diVicenza.
9 About the correlation between Fibonacci series and measurements see Franois Denis, Trait de lutherie, ALADFI, Nice, 2006, pp 21-56.
10 Dividing a space between two vertical lines a whole number of times is easy if you have a graduated ruler. For an illustration of the
method, see Franois Denis, op. cit., p 52.
11 This is the extension of the Quadrivium sciences as presented by Proclus. The geometry of movement par excellence is that of astronomy
and of the trajectory of the planets, while static geometry applies to surfaces and figures.These two sciences are completed by arithmetic,
which deals with the manipulation of whole numbers, and music, which expresses the rightness of the relations between whole numbers
in audible form.
12 On the compass techniques used, see Franois Denis, op. cit., pp 101-113.
13 Christian Rault, Gomtries mdivales, tracs dinstruments et proportions harmoniques, in Instruments cordes du Moyen Age, Proceedings of the
Royaumont conference, 1994, ed. Craphis, Grnes (France), pp 51-56; Los instrumentos musicales en el siglo XVI in Encuentro Toms Luis de
Victoria y la musica espaola del siglo XVI, Avila, UNED, May 1993, pp 231-242; Les modifications structurelles radicales des instruments cordes
au XVI sicle, in Pastel no. 21, September 1994, pp 30-36.

335

Franois Denis

Annex
Table 1
Most frequent ratios with the name of their proportions
3/8 = 0,375
2/5 = 0,400
2-1 = 0,414
3/7 = 0,429
4/9 = 0,444
4/7 = 0,571
5/9 = 0,555
2-2 = 0,586
3/5 = 0,600
5/8 = 0,625
2/3 = 0,666

geometric section
harmonic or geometric section
harmonic section
harmonic section
harmonic or geometric section
harmonic section
harmonic or geometric section
harmonic section
harmonic or geometric section
geometric section
arithmetic section

Table 2
N2V

N14V

Theoritical ratios

N3V

N11V

N12V

Theoritical ratios

N13V

PX to PQ

0,413

0,413

2-1=0,414

0,415

0,414

0,429

3/7 = 0,429

0,422

ZX to XQ

0,402

0,396

2/5 = 0,400

0,394

0,406

0,375

3/5 =0,375

0,395

pp to PQ
pp to XQ

0,568
0,967

0,585
0,997

2-2 = 0,586
1,000

0,577
0,986

0,589
1,006

0,570
0,997

4/7 = 0,571
1,000

0,586
1,014

N2V

N14V

Theoritical ratios

N3V

N11V

N12V

Theoritical ratios

N13V

0,620

5/8 = 0,625

0,624

0,604

0,602

3/5 = 0,600

0,624

Table 3

aa to pp (top) 0,640
bb to pp (top) 0,222

0,187

3/16 = 0,187

0,233

0,195

0,200

1/5 =0,2

0,199

pp to PQ

0,568

0,585

2-2 = 0,586

0,577

0,589

0,570

4/7 = 0,571

0,586

pp to XQ

0,967

0,997

1,000

0,986

1,006

0,997

1,000

1,014

N2V

N14V

Theoritical ratios

N3V

N11V

N12V

Theoritical ratios

N13V

wf to hf

0,516

0,502

1/2 = 0,5

0,518

0,559

0,501

1/2 = 0,5

0,589

hf to aa

0,633

0,694

7/10 = 0,700

0,604

0,606

0,666

2/3 = 0,666

0,579

Table 4

336

Franois Denis

Table 5

N5A

Theoritical ratios

N6A

Theoritical ratios

PX to PQ

0,414

2-1 = 0,414

0,410

2-1 = 0,414

ZX to XQ

0,385

3/8 = 0,375

0,396

2/5 = 0,400

pp to PQ

0,568

4/7 = 0,571

0,573

4/7 = 0,571

Table 6
N5A

Theoritical ratios

N6A

Theoritical ratios

hf * to aa

0,668

2/3 = 0,666

0,657

2/3 = 0,666

bb to aa

0,374

3/8 = 0,375

0,387

3/8 = 0,375

aa to pp (table)

0,603

3/5 = 0,600

0,636

5/8 = 0,625

(* longitudinal lenght of the f-holes)

Table 7
N10
PX to PQ
ZX to XQ
pp to PQ

N17A
0,443
0,354
0,583

Theoritical ratios
4/9 = 0,444
5/14=0,357
2-2 = 0,586

N18A
0,439
0,358
0,591

cello
0,447
0,361
0,602

Theoritical ratios
4/9 = 0,444
5/14=0,357
3/5=0,600

Table 8
N17A
0,354
0,360

N18A
0,358
0,360

N17A
aa to pp (top) 0,558
bb to pp (top) 0,232
ee to qq
0,665

N18A
0,533
0,219
0,663

ZX to XQ
ZP to PQ

N10 cello
0,361
0,353

Theoritical ratios
5/14=0,357
5/14=0,357

Table 9
Theoritical ratios
5/9 = 0,555
2/9 = 0,222
2/3 = 0,666

337