/U
(I, t
. NATIONAL, MUSEtJSnsCTT"
,/i.
THE SMITHSONIAN INSTITUTION
Washington, D.C.
^a,nfl Report" Eor,Grant N. Fe-6.~66.619, 1976 , - Special Studies/Resjearfch
^Development, of»$on~jpestr;uetive Techniques.to .Search for, a,Lost
Mural By Leonardo,da Vinci"
By
H. Travers Newton, J,r. ,
Maurizio Seraci-ni
0 copyright 197-6. -AO-l -rights -^asf^.y-ed..  No-pcti-c of the -contents .of
this report may be presented or reproduced-without the written per-
ittissiqn of the authors, -H. Tr avers Newton, Jr. and MaurizJto iSacacini...
Photograph of the east wall, Hall of the Five Hundred, Palazzo Vecchio, Florence, showing the two scaffold-
ing arrangements used in our research.  It is believed that fragments of the Leonardo mural painting the
"Battle of Anghiari" (1505), may still exist beneath some area of the mural decoration by Giorgio Vasari.
TABLE OF CONTENTS
Page
Section                                                                                                      —"*—'
II
Acknowledgements ...............,.,.......»«.........   .....
TTT
Preface — ..... — • -............• •¦.........
..„ ... ...IV
Introduction.....................••»•••...........
.                                                                   ... ..3
The Historical Discussion................---*........
Ultrasounds................•............•........
.......6
I.. Techniques..................................
;................10
II. Apparatus............• • •......•.........
.„ ,                                                            ......____. .12
III. Display Modes..............................
.... ...21
IV. Applications................•......•....... ..
.....51
V. Conclusion.....................................
. .. ,........54
Thermovision......... •.....«..............*......
. .55
I. Introduction.........<•.....•.....¦--...........................
II. The Apparatus..........................................,.........57
III.   The  East  Wall   Investigation...........................53
IV. ' Interpretation of Thermal Pictures....................75
v. .' The West Wall Investigation..........................-93
VI.   Conclusion. ...... . ........... •. • •..r...r.r..r_r_r.v.':J'..*..*_?-*^.-?-t-^-'—-*--* *¦ *1|5
Appendices
I. Chronology of Leonardo * s Activity....................J-17
II. Materials Purchased by Leonardo................• • •----118
III. The "Last Supper" and "Sala delle Asse"
11*3
Research Projects..................-...................^^
IV. Other Methodologies With Possible Application
To The Search for the "Battle of Anghiari"---.........129
The General Conclusion...................•..............* ""^2
¦II.
ACKNOWLEDGEMENTS
¦.'¦<.
First, we wish to express our deep gratitude to Mr, Paul Perrot
and the Smithsonian Institution for their support during this crucial,
initial phase of the project.  Also, we would like to thank Dr. Franklin'
Murphy of the Samuel Kress Foundation for his personstl interest in- '
our research.
. The ultrasonic equipment used in our research was loaned from the
University of California, San Diego; the Institute for Research on
Electromagnetic Waves (IROE) of the National Center for Research
(CNR), directed by Prof. Toraldo di Francia; the Institute of
Electronics of the University of Florence, directed by Prof. Masotti;
UNIRAD, Europa; and DISI, Italia.
Permission to undertake this project was granted by the Super-
intendents of Florence, Dr. Baldini, Prof. Berti, Arch. Beroporad;
the City Council; the Mayor; and the Minister of Culture of Italy,
Senator Spadolini.  None of the work would have been possible with-
out the constant support of the City Architect, Piero Micheli, and
his immensely valuable assistant. Arch. Duccio Toti.
The preliminary "Last Supper" research in Milan was made possible by
the Superintendent of Galleries, Prof. Franco Russoli.  Work in the
Sforza Castle was encouraged by the museum's director, Dottoressa
Mercedes Garberi.
We are indebted to the following persons for technical assistance:
Dr. Giorgio Accardo (Istituto Centrale del Restauro)„ Prof. John
Asmus (University of California), Arch. Barbini, Dr. Cavalchini
(Alitalia Airlines), Engineer Gabriele dementi, Prof. De Waal
(Archifoto, Holland), Dr. Robert Feller (Carnegie-Mellon Research
Institute), Pr6f. Govert Kooiman (and the Central Research Institute,
Amsterdam), Prof. Tony Moretti, Prof. Carlo Pedretti, Sir John
Pope-Hennessy, Arch. Ruffa, Arch. Selvatico, Arch. Stefanutti; and
with great respect for our good friend, Prof. Leonetto Tintori.
Special direction and logistic assistance was provided by
Arch. Guerrieri (of the Superintendency of Monuments), Prof. Enzo
Ferroni (Chancellor of the University of Florence), and Dr. Giovanni
Urbani (Director of the Central Restoration Institution, Rome).
Tante grazie Mr. Mayor, Elio Gabbuggiani, foe,allowing us to
work in your beautiful home, the Palazzo Vecchio.
III.
FOREWARD
Although several preliminary reports have appeared concerning
ultrasonic tests on hypothetical wall situations, this is the first
technical report to appear as a result of research conducted.at the
actual site in Palazzo Vecchio.  As such, it must also be considered
preliminary.  Hence the evidence set. forth must be interpreted as in-
complete.  Nevertheless, if the findings presented here are followed
as guidelines, they will serve truthfully.
In order to undertake the next and more delicate phase of the pro-
ject, it will be necessary to modify complex and costly equipment.
This will only be possible through full collaboration among the inter-
national scientific community.  Such cooperation requires a new
•global5 mentality towards conservation, which recognizes the great
urgency for action.  Thus, hopefully, the longer-term benefits of
this project will come from the development of new, non-destructive
diagnostic equipment, capable of assembling a comprehensive 'clinical
card1 for any art object, be it a fresco or an entire church; thus
permitting effective measures of conservation.  Science must apply its
knowledge gathered through medical, space, industrial research for
the benefit of art.  We are destroying our environment and our art-
istic patrimony at such a rate that the craftsman's limited approach
to conservation is not enough.  If we continue at the current rate
of pollution and urban expansion, within 50 years the major part of
the world's artistic patrimony, and particularly the Italian patrimony,
will disappear.  One only need consider the example of Venice where
over a period of 500 years, 98% of the terrible damage to exposed
stone has taken place during the past 30 years, or since the establish-
ment of the Mestre petrochemical refineries in 1946.
Faced with the bureaucratic and financial bankruptcy of most
national cultural affairs departments, the final decision concerning
the future of much of mankind's artistic patrimony will be levied
by world industry.
_-.-..-_.-'... •
Mioral painting by Giorgio Va.sari, Hall of the Five Hundred, Palazzo
Vecchio, Florence. The detail of the area indicated by the arrow,
and the explanation, appears on the following page.
This is the only inscription of the entire mural cycle in the Hall. At
a height of over 14 meters from the floor, it is invisible to any view-
er; only when our scaffolding was raised was the motto discovered.  It
is the imperative tense and means "seek, find". Could this be an ironic
piece of evidence left for us by Giorgio Vasari?
¦X-
DEVELOPMENT OF NON-DESTRUCTIVE METHODOLOGIES TO SEARCH FOR
A LOST MURAL PAINTING BY LEONARDO DA VINCI
INTRODUCTION
The mural painting is one of the oldest forms of artistic
expression in man's brief history.  Beginning with roughly in-
cised lines on rock surfaces, the process gradually became more
sophisticated so to include animal or vegetable black and a few
earth colors.  Eventually, organic binding media were introduced
to permit greater freedom of expression.
The practice of superimposing one's painting upon another is
equally ancient.  For example, amongst the rich heritage of mural
paintings in North Africa, one will find forms incised into the
rock, dated imprecisely at 25,000 B.C. These figures were sub-
sequently overpainted with increasingly complex works.  The later
examples become more stylized and abstract, and include varioxis
types of pigments in casein or gum arable binding media.
As in the case of the ancient North African mural-paintings^
the works are limited to specific areas (i.e. unusual geological
formations, particularly smooth rocks ) which suggest magical
associations.  The centuries of superimposed paintings testify that
the action itself of painting, rather than a didactic intent,
was the motivating factor.
Mural expression of the ancient world reached such a wonder-
ful level of sophisticated freedom (e.g. Altamira, Spain;and
Lascaux, France; Palermo, Italy, all datable about 15,00.0-10,000 B.C.),
that the oldest known painting on a man-made flat surface, the
Hierakonpolis mural (Egypt, c. 3200 B.C.), appears primitive in
comparison.
Z    ft	o	O	0		0*	cr	cr		ft)	0)	Pi		13	S3	lp»	<	)-j	cr	H-	0		i-i-	It	>	H-	3	Ui	W	fts		
cr   CD	d	ft)	3		fD	CD	a*		3	n	CD		O	CD	»-»	ft)	s»**	CD	3	H		N	cr	03	3	d	<^>	H>	3		
0)     Cfl	tl	2	(D		3	CD	CO		O)	CD	ft		H	Hi		t-1		cr	03			ft)	CD	CD	Hi	ti	o	3	pi		
rt   rt		CD			CD	3					CD		H-	0	H-	d	ft)	H-		0		ft		CO	ti	ft)		03			
H-	P.		£		ft)		G*	O	H*	CO	H		rt	n	3	CD	H>	3	rt	ft		H-	¦3		fD	M	• tfl	Pi	13		
H     3	CO	rt	ft)	a	rt	to	CD	3-	3	rt	3	H	H-	3		CO	ft	Pi	0	cr	CO	0	d	d	»Q	to	•	O	ft)	¦¦3	
fD    03	ft	0	CO	d	cr	ft)	CD	CD	§	H-	H-	0	O	ft)	H	«•»	CD			CD	H-	3	ii	13	d		o	3	H-	P4	
H	CD			n		<	3	i':1	ft)	H	3	P.	ft)	rt	CD		H	rt	rt	H.	3		ft)		CD	O	«	to	3	H-	
5   o	ii	H»	ft)	H-	t->	CD		Z	05	H»	H-	ft)	M	H-	ft)	U)	ft)	cr	cr		0	ft)	M	d	3	Hi	s«*»		ft	ffi	
H-     Hi	3	H-	en	3	ft)	H?	, P-	0	H-		3	*<		0	0	^—•	rt	H-	CD	3	CD	CO		3	rt.		>	»—«.	H-		
3	H-	03	rt	05	k:		CD	3	3	t-i	U3	"*	CO	3	rt		H-	CO		d			13	rt		*t)	w	CO	3	ft	
0)      Z	3	3*	0		CD	Hi	cn	Pi	ft)	H-			H-	'«	H-	H-	0		cr	H	It	ft)	ft)	H»'	CD	0	rt	o	o5	*<	
u>	ft)	rr	d	s	H		rt	CD	ft	3	rt	rt	rt		0	3	3	T3	H-	ft)	cr		H-	M	X	3	H	o		13	
Hi	ft		3	0	CO	rt	H	H	H-	H-	3*	3*	d	O	3			3"	CO	H>	H-	'd	3		ft).	13	d	o	rt	CD	
H	H-	ft)	p.	l-i		3"	0	CO	0	rt	CD	CD	ft)	3*		0	O	CD	rt	01	jn	ti	rt	W	3	CD	to	I	CD		
in	0	Hi	CD	X	0	CD	•<:		3	CD		H	rt	d	ri-	ti	Hi	3	0			H-	H-	O	•a	H-	n	Ul	a	0	
	3	rt	P.		Hi		CD	t_i.	»	Pi	13	CD	H-	ti	0	P.		O	H	s:	rt	3	3	3	H1	H-	H-	o	3"	Hi	
w		CD		H-		ft)	p.	d			H		0	n		CD	rt	3	. p.	ft)	H-	ft)	iQ'	0	CD		ft)	o	3		
0	S	4	tr	3	s.	H		CO		tr	CD	CD	3	3*	rt	ti	pr	CD	0	CO	3	ti		3	to	ft)	3		O	O	
	ft)		*<		3'	0	H-	rt		"<	w	X	CO	CD	cr		CD	3	ft)		(D	^<	0	CD		3		tc	(_a	•<	
rt	to	0		ft	H-	3"	3				CD	H-	*	CO	CD	rt		0	M	ft)	^		3	to	O	Pi	3	•	0	o	
0		<	rt	CT	rt	H-		*		rt	3	CO			0	O	ft)	1-J		CO		Hi	O	tQ	Hi		d	o	05	I-i	
	i-i	CD	3*	CD	CD	rt	rt	3*		3*	O	rt		O	h-1		ii		O		It	0	CD	d		ffi	h	•	^<	M-	
P>	CD	*i	CD		3	CD	3*	Q)		CD	CD	CO		3"	0	P.	o	0	0	o	cr	H		CD	CD	CD	ft)	•^^		h	
0	3	(-•		CD	ft)	O	H-	n-						ft)	v£!	H-	cr	ft)	3	0	CD	3	ft)		ft)	Ii	H1	<*	n	ft)	
o	CD	><	03	ft)	CO	rt	CO			3*	0	3		3	H-	CO	H-	3	rt	3			05	ft)	ti	O	CO		ft)	H1	
c	z	H-	d	i-S	3*			rt		ft)	Hi	0		iQ	O	H-	rt		CD	3	0	0	ft)	ti	H»	d		rt	3		
3	CD	3	ft)	ft		ft)	o	3"		3				CD	ft)	3	CD	cr	"A	0	0	Hi	. H-	rt	^<	(-¦;	i~.	cr		13	
CD	a.	03	3	3*	ft)	3	CD	CD		g	3-	^d		Pi	H	Hi	n	CD	ft	3	<		3			ft)	Ln	CD	tr	ti	
3			it	H3	3	a.	3			CD	H-	H				CD	rt				CD	ft)		Z	O	3	O		CD	'0	
rt	rt	13	H-	d	Pi		rt	CD		h	Pi	CD		ft	3	0	d	H.	ft).	ft)	H	ii	rt	V	Cr	"CD	o	a		05	1
	0	H"	rt	ft)		o	d	X		•«	Pi	0		0	0	fl-	H	0	3	CO	H-	rt	0	CD	H	d	1	H	rt	ti	to
ft)		ft)	•<:	tf	13	0	H	ft			CD	H-			<		ft)	d	Pi		3	H-	O	3	H-	'3	co	CD	ii	CD	1
13	&	CO		CD	M	3	^<	CD		CO	3	10		3	CD	it	M	05		rt	03	. CO	X	¦*	CO		o	CD	0)	to	
P.	CD	ft	o		ft)	CO		3.		o		CD		0	3	3"		cr	ft	S'		rt			rt	*—>	o	X	o	CO	
N	<	CD	Hi	Pi	to	CD	ft1	rt		ft)	3			CO	CD	CD	CO	H	cr	CD	0	H-	H-	CD	H-	H			CD	H-	
	CD	H		H-	rr	h	H			H	d	3		,Q	3		CD	^<	CD		<	o	it	ft)	ft)	CO	rjd	ft)	Pi	0	
3*	H1		3	to	CD	<	O	0		TJ	i-i	CD		d	rt	s:	rt			3	CD		CO	ti	3	ft	•	3		3	
O	o	0	ft)	ft)	ii	ft)	3	Hi		CD	ft)	rt		CD	CO	&)	rt	p.	O	0	H	CD		M			O	Pi	ft	\	
13	•c	Hi	K	CO	•*	ft	CD			H-1	h-i	3^		CO		H	H-	H-	0	ft		X	13	^	'3	o	•		Cr	P.	
CD			<	ft		0	^	3		^		O		>i	*—"V	r-1	3	<	d	H-	O	13	M		d	CD	*W	•TJ	H	CD	
Hi	ft)	tr	CD	CD		H		ft)			13	Pi			CD	CO	iQ	H-	3	<	Hi	ti	ft)	H-	ii	3	•*»	CD	0	03	
d		>i	h-1	H			ft)	3		ft)	ft)	0		CD	*		¦«•	P.	ft	CD		CD	0	3	ft)	rt.		ti	d	CD	
M	3	p-	H			*T*	3	-		3	H-	H		rt	03	ft)		CD	ii	CO	3	CO	CD		H	d	rt	to	03	3	
H	CD	o	0	ft)		ft)	P.	CO		P.	3	0		0	•	Hi	to	Pi	K		d	CO		rt	• CO	ti	cr	H-	Cr	CD	
><	ft>	X	d	H		Pi					rt	iQ		»		ft	*¦"		¦*!	5!	H	H-	H-	cr	¦^»	'¦<	CD	ft)	<-¦	ti	
¦«t	3		CO	CD			3T	3		rt	H-	•<:		*-*	H-	CD		H-		CD	ft)	0	3	CD				3	it	ft)	
	01	z		ft)		ft)	O	d		3*	3			-.	O	H	o	3	ft	ii	H	3			ft	W	ft)		Cr	ft	
en		ft)	•0				s:	»i		CD	U3	Hi			0		3"	rt	31	IT)	CO	¦	s:	m	cr	•	to	3	CD	H-	
tu	0	en	ft)	0		3		ft)			CO	0		cn	>->	rt	ft!	0	CD				CD	K)	ti	O	rt	d		0	
<!	Hi		H-	i-h		CD	3	H		H	t	ti		^-#	0	?-r"	3			<	tr		CO	ft	0	•	o	ti	td	3	
CD		en	3			ft)	d			H-					o	CD	'•Q	Hi	H	ft)	k;		ft	31	d	J*m*'	d	ft)	03		
	3	3*	rt	3		3	0	13		3		H		0	H		H-	H-	ft)	ii			CD		03	"%|	3	H	•<:	tr	
CO	0	P	H-	0.		CO	3*	ft!		H-	n	0		s-	ft)	13	3	<	ft	H-	'%'		ii	O	3*		p.	to	13	0	
0	3	W	3	Ii				ft		rt	0	O		¦ 0)	CO	i—"	uQ	CD '	H-	0	3"		3	CD		•0	H-		ft	it	
3	!	CD	05	rt		0	3	ii		CO	3	ft)		3	S3	ft)			0	d	H-			3	rt	ft)	3	^~.	H-	cr	
CD	Cb	3	to	H-		Hi	H-	H-			CO	ft		•^5	".	U3	ft!	a	3	CO	rt		0	rt	tr	co	05	N>	ft)		
rt	CD			CD			iQ	3		0	CD	H-		m.		d	CD	ft)	ft)	•	CD		H-	d	CD	ft		O	3	H-	
Sr	Cfl	H	rt	H		s	3*	0		Hi	n	3		b	H-	CD	to	rt	^—i		«		<	ii			W	o		3	
H»	r+	0	cr	3		co	ft	3			<	iQ		U2	3*	**•	ft	CD	CD	>	ft)		H-	^	s	rt	O	o			
3	H	0	ft)			CD		)-<		rt	ft)				<D		3"	uD		o	CO		M	Ni	H-	cr	3	1		10	
03	d 0	en CD	rr	H ri-		CD	3* 0)			H-3	rt 0	0 H					CD rt	0 ii		0 0	Cf		1		Pi P.	CD	ft) 3			ft	
0	rt	*-		ft)		3	<			CD	H						H-	H-		H					M					H '	
Hi	< CD			•<		03	CD				W						O	cs CO *•		P. i	13 ft) W ft (D ti				, CD					fl)	
-3-
THB HISTORICAL BACKGROUND
' I.
THE HALL OF THE GREAT COUNCIL OP FLORENCE
As a result of the French invasion of Italy in 1494, the Medici
empire was in ruins.  The populace of Florence immediately decided
to create a Great Council, following the Venetian example, which
was composed of all the responsible citizens of Florence.  A building
was thus urgently needed where all the members (at least 1,500 persons)
could meet.  On May 11, 1495, the order was given to erect the Hall
of the Great Council, today known as the Hall of the 500.  In great
part due to the urging of Savonarola, the building was constructed
in such a furious pace that only nine months after the beginning of
the work, the Great Council was able to meet in the new Hall for the
first time.
The model for the new Hall was the Sala del Maggior Consiglio at
Venice (finished in 1419).  It is indeed instructive to compare this
hall with the Florence counterpart in order to imagine what, the Hall
of the Great Council looked like at the time of Leonardo da Vinci.
Only two attempts have been made to reconstruct the Great Council
Hall".  This is not an easy task, and subject to some conjecture,
as the original structure was radically altered by Giorgio Vasari,
beginning about 1563. As in the Venice example, the ceiling was
rather low (12 meters).  On the west wall, where Savonarola ordered
an alter be placed and a panel painting by Fra Bartolomeo, four
large windows illuminated the Hall. The shorter north and south
walls held three windov/s each (some are still partially visible from
the exterior) .  The east wall was reserved for the Tx-ibuna in the
middle, e.g. a wooden structure where the elected governor and his
councellors sat.  On either side of this structure was a window which
probably also served as access to a balcony (as in the Venetian example)
There were also two doors in the east wall; one led to a room where
the votes were counted, the other led to a room where a list of town
1. These are: The Hall of the Great Council of Florencef   'Journal of
the Courtauld and Warburg Institute' 1942, by. J\" .\WiMe; and The
Battle of Anghiari; An Introduction to the Excavation, by Travers
Newton, Jr. (unpublished manuscript).
-4-
debtors was kept.  There were two doors on each end of the west
wall by which one could enter the Hall.  The door at the south end
of the west wall gave access'from the street, whereas the door at
the north end connected the Hall with the ground floor of the old
palaee.
II.
THE MURAL PAINTINGS OF THE GREAT COUNCIL HALL, FLORENCE
In an order signed by Machiavelli as Secretary of State, Leonardo
da Vinci was given the comission to paint a large mural in the new
Hall.  In March, 1504, Leonardo began the full-scale cartoon (or
drawing) of the central portion of his composition, the "Battle
for the Standard".  And just a few months later, while Leonardo
was on a brief leave of absence to Piombino in order to study the
fortifications of that city, Michelangelo, some twenty years his
junior, was awarded a contract to decorate a wall in the same Hall.
Both compositions were secular didactic pieces of the victories
of the citizen army of Florence? Leonardo depicted the "Battle of
Anghiari" against the troops from Milan; Michelangelo's "Battle of
Cascina" against the Pisan forces.
It is generally accepted that these two works were the turning
point of the High Renaissance.  For years artists came from far
off places to copy the paintings;  Benvenuto Cellini called them
"the school of the world".
However, it has never been determined where the paintings were
located in the Hall.  Indeed, there exists some doubt whether
Michelangelo even actually began painting on the wall, despite
documents which show that he purchased colors, stucco, sinopia, etc.
Leonardo, instead, worked on his mural for a full year with three
assistants.
-5-
Leonardo was called to Milan on 30 May, 1506.  He probably
returned to work in the Great Council Hall between 1507-08.  Giorgio
Vasari stated that Leonardo abandoned the project because of a
faulty painting technique which necessitated the presence of great
fires.  Other sources cite Leonardo's dissatisfaction with the
quality of the linseed oil that was supplied to him by the city
("l'olio di lino che fu falsato"); a reference which might indicate
that the oil was not properly boiled, which would lead in turn
to a lack of siccative capacity.  Whatever the reason, therecclearjry
must have been a tremendous amount of friction in the Great Hall
with Leonardo and the young Michelangelo working side by side.
The "Battle of Anghiari" mural was admired for over fifty years
until Giorgio Vasari, at the order of Cosimo Medici, altered the
Hall after 1563.  Some of the more indicative references to the
Leonardo painting are the following:
"In the new large Council Hall is a panel by fra Philippo,
the horses of Leonardo da Vinci, and the designs of Michel-
angelo" (from a guide to the sights of Florence by Francesco
Albertini, 1510).
"Leonardo was commissioned to paint in the Great Council
Hall of the Palace in Florence...and he began the work in
that place as can be seen there today; and with varnish"
(from a biography of Leonardo by the Anonimo Gaddiano, 1543) .
"Having ascended the stairs of the Sala Grande, take a diligent
view of a group of horses, a portion of the battle of Leonardo
da Vinci, which will appeas a miraculous thing to you" (from a
letter by Anton Francesco Doni, August 17, 1549).
Did Vasari destroy the "Battle of Anghiari", or did he cover it
over with his own composition?
APPLICATIONS OF ULTRASOUNDS IN SEARCHING FOR TRACES OF
THE LEONARDO MURAL "THE BATTLE OF ANGHIARI"
I. TECHNIQUES
In Italy, the most common form of mural painting is the fresco,
e.g. natural pigments fixed to a surface of lime plaster through
the carbonation of the calcium hydroxide [ca(OH)g] .  In other
countries, the murals are often executed in a tempera technique,
where pigments are fixed to the wall with an egg, gum or casein
binding media.  Since the 15th century, limited examples of oil
painting on wall are documented.
To date, the only means of localizing mural paintings hidden by
whitewash, plaster or brick, is that of scraping away the obstruct-
ing surface.  In the field of industrial non-destructive testing
(NDT) of materials, internal defects such as cracks or density
variations are usually recored by one of the following methods:
x-radiography, gamma-ray radiography, magnetic, luminescent or
ultrasonic variations.
Recently, we believe to have made major advancements towards
non-destructively describing the internal characteristics of walls.
The most promising results have been yielded by ultrasonic NDT and
by infra-red Thermovision scanning.
The techniques of non-destructive examination of materials with
ultrasound are based principally on the fact that the propagation of
a beam of ultrasonic waves depends on the acoustical impedance of
the medium in which it is propagated.
When this beam strikes the interface between two materials with
different acoustical impedance characteristics, part of the waves
are reflected and part transmitted into the other material according
to a relation that combines the acoustical impedances of the two
materials.
~7~
Some factors, such as the non-homogeneity of the materials and
the thickness of the. discontinuities present in them, produce
phenomena of dispersion and/or absorption that cause a loss of
ultrasonic intensity, thus limiting the penetrating capacity of
the beam of ultrasonic waves and, hence, the intensity of reflected
signals.    •
The Vasarian walls present non-homogeneities, porosities, and
discontinuities; moreover, the thickness of the layers of the under-
lying bricks varies from a few centimeters to more than a meter.
These are all factors that condition the choice of frequency of
the beam of ultrasonic waves.  We had to seek a balance which would
result in a good resolution of images and in sufficient penetration of
signals through the Vasari murals and the layers of bricks.  After
preliminary experimentation, it was decided to try frequencies varying
between 500 KHz and 1 MHz.
The ultrasonic signals are generated by transducers which have the
function of transferring the tension impulses at high frequency into
mechanical vibrations and vice versa.
The principal component of a transducer is the piezoelectric
element composed of a crystal (normally quartz or barium titanate)
that under the influence of an electric field expands and contracts,
thus producing ultrasonic waves in the coupled material.  Conversely,
when the crystal is struck by sound waves, it generates an electrical
impulse (figure 1, page following).
The wave travels at a constant speed through the first layer of
material and at the same time the signal traces a horizontal line
on the screen of a cathode tube.  When the wave reaches the interface
between the first material and the second, part of the energy is
reflected back, and part continues into the second material.
When the reflected wave reaches the transducer, it generates a
voltage that, upon amplification, causes a deviation of the tracing
on the screen.  The distance between the two deviations (or 'peaks')
shown is proportional to the thickness of the first layer (figure 2).
ULTRASONIC      TRANSDUCER
CABLE    CONNECTO
¦Upvsirto
S *
5  ?*8r     <«* 5* ¦r- - "    "   -  ••• •- »
*-      ''Jf^   'jut •    '.   "---  , •; - 1    .*•   *    »   - -"   .1

* * * y t
ELECTRICAL
CONNECTIONS
*.*.•¦ •*.   «• v*. *•-*'•     «    , vsvjl
vz^/ssam/s/////m
.PAMPING    MATE&ii
CRYSTAL
PROTECTIVE    .B01.Fi
Ficjure   1
¦9-
STRESS WAVE PULSE
ULTRASONIC
TRANSDUCER

•••¦.(


^
o
1
fl
(I
A.
E.
Figure 2
¦10-
The nature of the discontinuities detected can be determined by
measuring in dB (decibel) the ratio between the energy sent into
the wall and that reflected.
II.
APPARATUS
A UM 715 Reflectoscope by Sperry Industries was employed to
drive two 500 KHz Aeroteck transducers.  We found that transmission
techniques, with both transducers positioned on the same side of
the wall in close proximity, yielded the least noise and the strong-
est echoes.  Coupling was assured by a thin layer of water soluble
gel which was directly applied onto the painting.  The equipment
is shown below.
Figure 3
- 11 -

-12-
III.
DISPLAY MODES
The method of visualization of ultrasonic signals to which we
have been referring until now is designated as type A.  With this
method, the echoes received are shown directly on the screen of the
C.R.T. in the form of peaks of a height proportional to the intensity
of the echoes themselves.  This intensity is directly related to the
reflecting properties of the surfaces of discontinuities encountered
by the ultrasonic beam during its travel.  In particular, the
intensity increases with greater difference between the acoustic
impedances of the two contiguous layers.  From the analysis of a
type A representation we can obtain two results:
a)  knowledge of the number and location of discontinuities
found along the path of the sound waves;
b)  information concerning the type of discontinuity.
Unfortunately, the presence of various disturbing factors such
as multiple reflections and small air pockets make such analysis rather
difficult.  In order to minimize the possibility of error, it is
necessary to do the scanning in several contiguous areas.
The major limitation of type A representations is that with every
scanning, the appearance on the screen is relative to only one direct-
ion to the interior of the mural.The type B representation, instead,
allows one to view the situation in a cross-sectional view which
allows a more direct and often more precise interpretation.
./Figure 5: Type A mode
13-
The type B visualization allows the establishment of a threshold
in the height of the echoes received, intensifying the trace of
C.R.T. whenever an echo goes beyond that threshold.  A sequence of
small dark segments corresponding to the amount of echoes coming
from that direction is then obtained.  Therefore, by moving the
transducer and accordingly'the trace on the C.R.T. and by utilizing
a memory tube, it is possible to observe the echoes coming from the new
direction and so on until the entire screen is complete.  However,
the images of the section of material thus obtained lacks the inform-
ation relative to the height of the echoes due to the presetsthresh-
old.  However, this problem has been solved by relating such height
to the gradation of grey visualized in such a way that the greater the
intensity of the echo, the darker the trace.
Among the various types of apparatus equipped with type B
representations, the Unirad Sonograf III has demonstrated, during
repeated tests performed at the Istituto Centrale del Restauro, Rome,
both an excellent rendering of images and a significant resolution
in visualizing mural stratification by means of grey tones.  The
equipment has a range of 40 dB.


Figure 6: Unirad Sonograf B mode
-14-
TM
M

TM TRADEMARK OF UNIRAD CORPORATION
SUNIRAO CORPORATION 1975
Figure  7
.  .-      .                             Lr.-...^-«*-

-15-
UNIRAD SONOGRAP III
Figure 8
Figure 9
- 16 -
Experimental study on humidity detection on
fresco and underneath masonry materials by means
of ultrasounds, performed at the Centra.1 Institute
of Restoration in Rome.
Figure 10
•17-
It should be added that the movement of the transducer
arm in the Sonograf III, as with other ultrasonic equipment with
B mode type visualization, is cumbersome and ineffective.  It would
be an enormous advantage to automatize the apparatus, thereby
permitting an electronically synchronized scanning with the effective
displacement .of the transducer along the wall.  In the following
two diagrams, the remote-controlled ultrasonic transducer scanning
apparatus is shown.  The transducer is immersed in a container of
water, and can freely scan a wall due to a ve;rtical, horizontal and
sectional axis.
To obtain maximum displacement of air, one can improve upon all
types of coupling fluids by using a light layer of Wacker VP 3522
inert silicone rubber.  This material does not appear to harm the
mural paintings tested, and can be removed without any solvent.
The scanning apparatus is attached to the wall via a membrane wetted
with an acqueous gel; the hydrostatic pressure of the water in the
tank assures a strong contact.
/ /\
WA T 19.    t EVE i.
/
TRANSDUCER
u
ZJ
COUPLING    LAYER
MEMBRANE
E
<$£? «*«£~
WATER    TAN:
co
I
j..
5c r<\
Side    View
FIGURE   11
25cm    WATER    TANK
Top    View
I
i
FIGURE   12
20   -
20
w
Id
Eh
<
K X
K <;
Eh d
ft
fa <
O
U  U
H    C.1
K
H  U
M  H
O  O
O  <
W  ffl
0?   E-i
si.   i-3
ro
O
H
Q.
N      =
					UJ
					o
			>		z
			55		<
			z		o
			UJ		UJ
			o	«	Q.
			ill X	< IT u 1-	s
			l£*		o
			o	<	
UJ o z < Q UJ Q.	CO z	o o	o O o	S < u. O >-	COUST
s II	UJ Q	UJ >	Q. UJ X	o o	< o
N	Ql	>	m	UJ >	m
			m o	o z	cc
			z		ULS
			<	^"%	H
			o		O
			a.		<
			1		sc
					< X o
Z2-Z1
Z2+ Z1
R = REFLECTION
COEFFICIENT
MEDIUM 1
Zi
REFLECTION
-21-
IV.
APPLICATIONS OF ULTRASOUNDS TO THE LEONARDO PROJECT
AND PRELIMINARY RESULTS
Several areas on both the east and west walls were studied by
means of ultrasonic non-destructive testing.  The pictures shown on
the following pages are a representative sampling of the type of
signals recorded.  The east wall showed consistent signals of the
type shown in the pictures in all areas investigated.  The same con-
sistency was shown by the west wall where the definite difference
in the shape of echoes was seen in all panels, but particularly in
the left area.
Our presentation of these pictures serves only to indicate preliminary
results.  Nevertheless, we believe that these results are indeed
significant.  Only through a complete study of every part of the
two walls, combined with laboratory testing on sample walls similar
to the ones in the Hall of the 500, can confirm or refute our
findings.
For a better understanding of the following pictures, it would
be helpful to first define acoustic impedance: the density of the
medium times the velocity of the sound wave travelling through that
medium.  One can relate acoustic impedance to the density of the
material.
As stated previously, when a sound wave reaches the interface
between two materials with different acoustic impedance, the ultra-
sonic waves are partly reflected and partly refracted.  The amount
of sound that is reflected depends upon the degree of difference between
the two media, i.e. the greater the acoustic mismatch, the greater
the amount of sound reflected.
-   22   -
-   22   -
/p ^    /p   «/P
DENSITY    SCALE
	/?	/?	/?
transducer-	¦-= >		
	='      ^		
	1=                         ^-	>	*
	1---------------tm»	^	j
		4"	
	=            «"		1 1 |
initial    pulse
A
first   echo
A,
J\
AjLA.......£.......
second
echo
FIGURE   14
CM
¦

east wall, left panel
vasari's "la presa di
siena "
picture
15
CN
detail of the panel,
showing the site of the
acoustical scanning.
picture
1 fi
-25-
I
V0
CM
SITE
Picture 17
Strong peak coming probably from the intonaco-brick interface.
Following smaller peaks may be attributed to multiple echoing
from the same interface or from small discontinuities in the
first part of the brick layer.
- 27 -
I
h-1
00
01
H-
ft
(D
>
00
CM
SITE  A
Picture 18
Showing same peak of picture 17 with following echoes
damped by turning the reject knob on slightly.
I
a-,
CM
I
site B
east wall, left panel,
picture 19
site B
o
picture 20
en
I
SITE  B
Picture 20
Peak nl falls into the initial pulse width. This might signify
the presence of a discontinuity near the surface.( air?).
Peak n 2 has the same delay as the one observed in site A,even ;-.h
though attenuated by the strong reflected signal pEBcedis£_v • it.
Peak n 3 could be attributed to an internal discontinuity within
the brick layer or representing the plaster between the briks.
- 32
Graphic Interpretation' of Surface Stratification,
East Wall                                     .
Vasari's fresco
T
R


brick
intonaco     brick    interface
Hardly any si-spial is coming from inside the bricks.  Great acousti-
cal impedance difference between the intonaco and the bricks. Almos
all the signals sent into the v/all are bounced back at the intonaco-
brick interface.
m
m
picture 22
west wall, right panel,
vasari's '-presa di pisa-
34
H-
O
r+

05    Cb
H-   0)
rt  rr
fD     W
05    H-
H
&o
rt  Hi
H
05    H-
O    O
3   ft
pJ    ni
H	M
<	to
	^
¦	
(1	05
	^r
5	Q
5	1
a	H-3
	¦s
-   35   -
WEST  WALL   -   RIGHT   PANEL



K?**t%j;..;"


¦¦ >                                          «h    ..                     mat. iA4m iirflMWSHa
*     *,'*--

:*ivi*;s______


SawsBjsiaw
S-f T E
SITE      B
36
Both sites show two strong echoes.
Peak n 1 in sites A and B splits in two. This might
be caused by a discontinuity such as it might be found
at the intonaco-arriccio interface.
Second echo comes from a distance almost twice as further
in the wall than the first one. Great acoustic impedance
difference with the following layer .
site C

picture 25
SITE C
CO
Picture 2 5'
It shows a strong peak (1) coming from
a definite interface,followed by a split
type of peak.
en
PULSE  LENGTH
FIiT^
picture 26
initial pulse with enlarged
scale of distances.
site C
o
picture 27
In)!? r»
-3>
I
Picture 27
Showing the peaks coming from the same
place as picture 25 but with a larger
scale of distances.
CM
picture 28
west wall, left panel,
vasari's "presa della torre
di san vincenzo"

picture 29
detail of picture 28,
showing sites investigated.
-44-
I
(D
Ul
o
CXI
H
¦in
Tl A
Picture 30
Peak n 2 seems to indicate the presence of a discontinuity
with different acoustical impedance,compared to the
Vasarian layer and the brick layer. Other following echoes might be
attributed to small interfaces in the bricks as well as
multiple bouncing of the major peaks.
Peak n 1 shows the same delay than the strong echo recorded
on the east wall.
-46-
U
0
01
H
M
Cd
- 48 -
WEST WALL - LEFT PANEL
SITE  A
Picture 32
Picture 33
..........:..-14.9 -
We can summarize the analysis of the echoes coming from different
sites on the west wall with the following graphic reconstruction,
showing the possible layers which might have authored the echoes.
Vasari's fresco unknown,
SLXAX.......InlfetittA.
¦ - 50 -
The presence of two strong peaks in all areas investigated,
particularly on the left panel of the west wall, reveals the
existence of two distinct discontinuities, located immediately
beneath the Vasari mural.  We can identify the first echo as the
one which has originated at the interface between the Vasari mural
and an unidentified layer.  The second echo appears to have
bounced back from the intarface between the unknown layer and the
underlying brick layer.
From the analysis of trie delay, as well as the intensity of the
second echo,-..-we tend to discount the possibility that this latter
echo could be a multiple reflection of the first echo.  Neither
do we believe this to represent the interface at the eud of the
brick layer.  Furthermore, from the selection and variation of the
scanning area, we exclude the possibility of the presence of an
echo-forming layer of air under the Vasari mural.
Finally, the 'unknown strata" is characterized by a very strong
acoustic impedance, which would suggest a greater density than the
Vasari mural.
-50a-
GRAPHIC CO? TA1? ISDN OF FCHO SIGNALS DEFECTED
F**CM THE EAST AND WFST WALLS
EAST   WALL   -   LEFT   PANEL
WEST   WALL   -   RIGHT   PANEL
EAST   WALL   -   LEFT   PANEL
WEST   WALL   ¦-    LEFT   PANEL
WEST   WALL   -   LEFT   PANEL
WEST   WALL      -   RIGHT   PANEL
¦¦j - 51 -
The most striking difference which can be noted in the previous
diagrams (p.49), is the constant presence of a second strong peak
originating from the panels of the west wall; this is in contrast
to only one significant echo from the east wall.
It should be remembered that the first peak at the left extremity
of each diagram is the initial pulse, and is not to be considered.
52
V. CONCLUSION
The results presented herewithin must be verified under known
conditions.  For this purpose we have begun the construction of
wall sample^ using ancient bricks.  The materials that Leonardo
used for his mural is known.  It is possible then to reproduce
and study the-  possible cases, shown on the following page.
In addition, as previously stated, the new remote-controlled
transducer apparatus will eliminate any necessity of hand-held trans-
ducers, which has proven inadequate for a precise, pressure-free
scanning of the wall.
Summarizing our preliminary ultrasonic research, one may state:
1. There is a marked difference between the east and west wall
surface stratification characteristics (to a depth of about
20 cm.).
a. The areas "of the east wall seem to reveal the presence
of' a layer of plaster placed directly upon a brick
layer.  This is suggested by the snalysis of strong
peaks originating from a definite discontinuity
immediately below the Vasari mural.  Further penetration
of the sound wave into the brick layer was prevented
by the relatively high frequency used.
b. Conversely, the sites tested on the west wall, specific-
ally on the left panel, clearly show two echoes origin-
ating from two separate discontinuities with high
acoustical impedance variations. A third echo, seldom
detected, seems to originate from the third interface,
believed to be the brick layer.
2.  Only with a systematic acoustical analysis of both walls
using A and B mode displays, such as those presented by the Sonograf III
attached with the new scanning method, will it be possible to
identify the. true nature of these discontinuities.
53 -
1- Original bricks
2- Vasari's fresco replica
1- Original bricks
2- Leonardo's mural replica
3- Vasari's fresco replica
1- Original briks
3-
2- Leonardo's mural
replica
4- Vasari's fresco
replica       ^
. / - 54 -
3. We intend to use lower frequencies, in the range of 50-150
KHz, in order to achieve much greater penetration.  Together with
the infra-red Thermovision study, this low-frequency research should
yield valuable new^information concerning the complete stratific-
ation of walls over 1 meter thick.
-55-
THE APPLICATION OF THERMOVISION IN
THE SEARCH FOR THE "BATTLE OF ANGHIARI"
I. Introduction
To compliment ultrasonic NDT, other methodologies are needed
to describe the following aspects of our research: a) a complete
knowledge of the wall structure immediately beneath the murals by
Giorgio Vasari; b) the verification of the presence of structural
components such as doors and windows, hidden by successive alter-
ations of the walls, which would have a direct bearing on establish-
ing possible locations of the Leonardo mural;c) to establish or
discount whether Vasari had superimposed his painting immediately
upon the Leonardo mural (photograph .1) .
Photo 1, east wall, Hall of the 500, showing the large scaffolding and
the smaller, portable scaffolding used in our research. A portion of
the Vasari mural cycle is visible.
•56-
In order to answer these crucial aspects of our research, we
decided to employ infrared Thermovision apparatus.  This equip-
ment was loaned to us by Dr. Giovanni Urbani, director of the
Central Restoration Institute, Rome.  This new equipment has
already been used by Dr. Urbani in the Church of S. Francis in
Lodi, Italy, to study the effects of superficial temperature
variations in walls and fresco paintings in relation to humidity.
As well, Dr. Urbani established the usefulness of the equipment
for studying architectural reconstructions.
The Thermovision equipment, shown in photograph 2, is able to
measure the temperature distribution in an object by registering
the natural infrared emissions, which vary with the superficial
temperature, of the object itself.
Photo 2, showing the AGA Thermovision 680 color equipment and the
portable 750 (right) black and white apparatus.  Refrectory of
S. Maria delle Grazie (the "Last Supper"), Milan.
-57-
The Thermovision has revealed an exceptional ability for
revealing discontinuities in walls.  This capacity is often a
direct function of temperature gradient which itself can be a
function of density and conductivity.
Heat is passed on between adjacent stationary particles of
wall matter by the process of conduction.  The conducted heat
energy , flow     can be expressed as the number of watts flow-
ing through a square meter of material one meter thick under a
temperature gradient of 1 degree centigrade (W/mct.
The density, porosity and, in some cases, water content of
building materials determines their density.  The "bulk densities
of building materials vary approximately inversely to their air
content.  Therefore, since air disperses heat, those materials which
contain a large proportion of air in their structure (e.g. those
components which have a low bulk density, such as soft plaster),
are poor conductors of heat.
Air cavities in a structure or material tend to inhibit the
conduction of heat.  The size of the air cavity can be roughly
calculated by measuring their ability to reduce the passage of
heat flow.  The factors which mainly effect heat transfer are the
orientation (or location) of the air pocket in a substance, the
dimension of the cavity, and the thermal capacity of the material
in question.  A material of high thermal capacity requires more
heat to raise its temperature than a material of low thermal
capacity (diagram on following pages).
Also, since water is a relatively good heat conductor, it affects
the thermal conductivity of a solid material in which it is present
in the opposite manner to that of air; in other words, wet materials
permit heat to flow more easily than dry materials.
II. The Apparatus
The AGA Thermovision 750 portable infrared black and white camera
system was used for the majority of our study (photographs 3a, b, c),
The apparatus is composed of an infrared focusing system by means of
Germanium lenses and a prism scanning mechanism*' The infrared rad-
iations of an object (in this case the wall) are transmitted to an
-58-
infrared photovoltaic detector of indium antimonide with a spectral-
range of 2-5,6 /Am.     The detector, cooled by liquid nitrogen (-196*0
for optimum temperature sensitivity, converts these radiations into
an electrical signal.  After preamplifyication, the signal is processed
in the display unit via a television monitor (vidicon) where a grey
tone thermal image (Thermogram) is displayed in real time.  The
television image represents relative temperature differences of the
surface on which the camera is focused.  With this camera system,
which has a temperature sensitivity ranging from -20 C. to +900 C,
it is possible to detect a minimum temperature difference of 0.2 C.
e
at 30 C. object temperature.
-59-
1.4
1.2
1.0 ¦
Conductivity 0 (W/m°C) 0		.8 .6
	0	? 4
	0	.2
0
		/ / coupon
		
		brick
		/ f
'•'		f f /
		/'
		gypsum
		/"plaster
mmr-rr-"",\------	y —i—	—i-----------------1---------------->    ..  .. i
400 800  1200 1600 2000 2400
Density (Kg/m )
This graph illustrates the relationship between
thermal conductivity and bulk density for plaster
and brick at normal moisture content.
-60-
'Thermal Capacity
Temp.
(°C)
\<C.Q				J,-.«-«!¦-"'',*™,*"
10.0    •	?	/	if" / s i /	^ ..,1*^""
7.2     ¦	t			'
		/		
4,4     •				
1 .6			----------:----.------------------1-------:---------:----------------------------------1----------------„------	
1    ¦ '     2                                  3
Time (H.)
The heavier brick wall takes longer to heat up, and.
inversely, longer to cool, down, than the light plaster.
-61-
3a. AGA 7 50 camera.
3b. AGA 750 monitor.
-62-
ti	
rd	
ctf	
U	
0)	_p
§	-rH
r0	c
0	3
o	>1
in	rC
p*	rH
	&
rt<	03
O	•H
rfj	T3
o	
m	
¦63-
At our request in 1974, the entire area below the Vasari raurals
on the east wall was stripped of plaster.  This was fully document-
ed photographically.  Numerous architectural findings, such as a
door, a large arch and several smaller arches, were clearly rec-
ognized (photographs 4-9). The lower wall was subsequently re-
plastered.  Hence, a percise system existed for determining the
precision of the Thermovision*s ability to reveal the presence of
hidden structural features.
For maximum image sensitivity, the study.was preformed in total
darkness; as well, in the evening hours the museum iS free of
visitors, whose body heat might influence the findings (photo 10,
showing the empty museum, before the lights were turned off).
We were particularly pleased with the almost exact correlation
between the findings of the Thermovision in the lower area of the
east wall with the already known elements.
For example, we were able to identify the location of the long-
sought window/door in the central portion of the east wall (7.5 m.
to the north of the present-day door), as displayed in photographs
11-12, and the graphic rendering 13.  Due to the limited spectral
sensitivity of the polaroid film, and its inability to record the
entire range of grey tones which appear on the television screen,
this type of graphic representation can clarify otherwise in-
coherent photographs.
Human error is an element of this type of documentation, how-
ever.  Hence, in addition to photographic and video recording, we
would suggest the possibility of taping directly .' ..• the electric
signal of the detector.  A computer would then elaborate and dis-
play the data in a picture form or a print-out with a broader range
of grey tones, resulting in a more precise and faithful image. As
well, the computer would be able to measure a greater range of temp-
erature differences by means of isotherm functions.
III. jThermal Investigations of the East Wall
After preliminary tests, a systematic study of the entire east
wall was initiated (photo 14).  A wide-angle lense was used to form
composite photographic documentation of the large murals.  Beginning
with the right mural, "The Battle of Val di Chiana", one notices
¦64-
CD	
i	f«
CD	CD
Sx	U
-p	fO
X	
0)	.—.
	^
K	-P
rH	iH
rH	0
CO	c
s	"—'
4->	+J
Cfl	4-1
rd	CD
w	rH
-CP	
I
I
area in photo 4, strip-
ped of plaster.
-66-

i


Thermovision of area
4, showing the dark
outline of a door.
¦
-
I
I
.;.,
+*.¦&>'-*j -*=
¦i

7. East wall, extrene
right (south) area.
I
CD
I
8. Area in photo 7, strip-
ped of plaster layer.
IMi, Hj
I
i
I
i
9.Thermovision of area 7
shov/ing the dark out-
line of an arch.

¦;-;-;.        t.
I
o
I
O.Hall of the 500,south,
showing the large
scaffolding.

I
11.Hall of the 500,east
wall, showing small-
er scaffolding.
I
!\J
I



12.East wall,central area,
showing dark zone on
right side.
-73-
14. View of the AGA Therno--
vision equipment used in
the Pall of the 500.
-
-75-
through comparing pictures 15-16, several interesting structural
elements.  In the top area of the Thermovision photographs one
sees a dark line which is parallel to the frame.  To the right,
a lighter area which crosses the entire panel- is visible.  In
the lower part of the photograph three light areas (in the form ....._,
of arches) stand out in contrast with the dark background.-        (
I The central panel shows the same characteristic dark
line at the top.  Two lighter areas at the bottom are similar, in
shape to the onesobserved in the right panel (picture 12-18) .!  In
the left panel (pictures 19-22), again the same dark area appears in
the upper portion of the photograph at exactly the same height
as in the other panels.  In the lower right area, oddly-shaped
forms appear.(picture. 23)
On the left, a light area is well-
defined by a round shadow(pictures 24-26).
IV.
Interpretation of the Thermal Pictures
All of the photographs were taken using the normal mode, i.e.
the cooler the wall areas, the darker they appear on the screen.
To be precise, one should- consider external stimula which could
have an effect on the results.  In addition, internal structuring
of the walls, architectural alterations, as well as detached areas
on the Vasari murals should be taken into consideration.
For this purpose, a graphic rendering of the rooms immediately
behind the east wall was accomplished.  The temperature of the
various rooms was measured, as well as a superficial check for
humidity.  The results are shown in pictures 28-29.
Considering picture 30, one will/>notice an overlapping of two
arches.  This same characteristic appeared with the arch just under
the mural frame (picture 31), where a correspondence between bricks
and a dark contour was noticed.  The bright zones under the arches
are probably the result of the heating system present in both rooms.
The dark line which transects the top^area of the three panels would
appear to delineate the level of the Cronaca ceiling (1495) prior
of the alteration of the room in 1563.
The lower right area of the left panel presents an unusual
situation.  The lower portion of the dark area corresponds to an
unheated room behind the wall.  A surface examination of the Vasari
I
I
15. East wall, right (south)
area.
I
I
16. Thermovision of area shown
in photograph n. 15-.
¦78-
17. East wall, central mural.
-79-
		p^				
	' *					
						
		*				
				•		* •
						
						h
				¦•		*
			18. Thermovision of area			,
			in photo 17.	*		. -

I
oo
o
l
19. East wall,
(north) by
left mural
Vasari.


20.Thermovision of area
• in photo 19.
20b. Graphic rendering
of area in photo 19
-82-


as	
Q)	
U	
«	
IW	
0	
c	.
o	a^
¦H	i—1
W	
¦H	0
>	-P
o	C)
e x	
c	ft
aj	
,£	c
6h	¦H
CN
I
CO
I
23. Thermovision detail
of lower right area
in photo 21.
-85-
24. Thermovision detail
of lower-left area
in photo 22.

24b.Graphic rendering of
lower-left area in
photo 22.
I
00
I
25. Thermovision detail
of photo 24.
-87-
bn	
£	
•H	
u	
0)	
T3	•
C	m
cu	tSj
M	
	o
U	4J
-H	o
X!	,fi
Dj	a
rtf	
U	m
e>	0
.	
kO	
cm	
I
oo
CO
I
27.Overall view of the
east wall murals.




•JT12M puxqsq Suiocu:
sip BuTMoqb   'tjbm q.spg  -%z

i
CO
1
¦J! PL Ul^vfat   SHoo)|-3hlJQ    NO,iW   3H1   wlAv0HS
llV/A 15V^ 1H1   QNIH19    N"3»VX   /V01D35
001 :J   31WS

SCALE   I. loo

^
^tCTIOAJ   TAKE-N     BCHIND   THL EAH" VWU-l-
SHOWiUQ    THE   LOCATION    OF THE. f^ooHS  4DJACLUT'
29. Graphic rendering of
Therrovision, super-
imposed on photo 28.
-91-
•. -     *
¦< .• ¦   .«. ¦ •
I >•'••••   ••• '
mm
•v..-,.
Ipf&£ •   -.••••-.•''¦•. ¦: V^;:j v %¦#••¦,. .¦•• ¦•.¦• I
.   • ...      . - .\ •       —II  I            ,      ¦-     .'                    {'i*-.J.,*-l •<•¦-'        "\^\    .       .


.     I-.-•--¦.»'•*     ,   .-., ¦.* .Zi'¦:¦¦>• -"••"•-•
¦r ?   I    • .'     •-••».••  ftt/v.- •   ,' .'
K;;*'\^'-:;-::-C'-'V-^:';- QfitilA? • •'•
\    A .¦¦¦              ¦ ¦^;^-,y ^^r'\^-:
i. ;- . "X.. ¦ •¦' ¦¦:• .    •• ••*:•,-• •¦%W->'.-•'.HS
I


•Y-
l   .jl' I ir'


•- ¦        •    .", I* \  A..   •
1		
k	-p	
CO	m	
0)	(i)	
s	i—i	
&		
(C	^	
H	H	
u	I	
	+J	.
S	rn	rH
&	tt)	id 1
m	IW	
0	0	
H -H	n	5
(d	•H	s
-p	+J	Q
S;fl		C
o		
m		

'

-92-

31.  'Thermovision of arch,  east
wall,   lov/er right corner.
-9 3-
mural in the corresponding areas of loose paint and plaster.
In pictures 32-33, the upper area in this spot shows an. even
curve.  No humidity was found on the opposite side of the wall
in this spot; the corresponding room is heated, and no major
detachments appear in the Vasari mural.  One is left with the
possibility of an architectural remnant with a greater thermal
inertia than the surrounding area.  Judging from its shape and
location, one is tempted to judge it a window.
In the central panel, two large arches are clearly seen.  The
exact same forms are repeated, for at least three times, through
the right panel (pictures 34-35).  One could most easily consider
these arches as supportive elements; their symmetry and perfect
form would, however, suggest some other previous function.
In an attempt to render more legible some of these architectural
remnants, a simple experiment was undertaken.  The plaster area
below the murals was slightly heated with a tungsten 1000 watt
lamp,- after a few minutes, the lamp was removed (picture 36).
The brick layer was clearly visible under the plaster layer
(picture 37).  With improved recording arraratus, the Thermovision
equipment may allow a complete mapping of the masonry materials
beneath the plaster and mural painting.
V. The Thermal Investigation of the West Wall
First, a complete study of the area beneath the Vasari murals
on the west wall was completed (picture 38).  Several interesting
architectural remnants are visible (pictures 39-40).  On the
extreme right area, grey tonalities suggest traces of four apertures
(pictures 41-42).
Another interesting element noticed is that through Thermovision
studies, the marble statues along the east wall appear black, while
those along the west wall appear white (pictures 43-44).  The most
plausible explanation of this variation would derive from the fact
that the rooms behind the east wall are heated, hence raising the
temperature of the wall in comparison to the statues.  The west wall,
being in contact with unheated rooms, is colder than the Hall of
the 500; hence the statues have a slightly higher temperature than
the wall.                                                                                                                                         .¦¦*>¦¦..
I
I
32. Thermovision of east wall,

-95-
-96-
44 O -P	
	m
I	a
&	
¦a	1
h	ft
o	rH
•H	(H
«	in
&+-	
01	C
&	(!)
(3	O


-97-

I   .
¦¦¦'.'$ .¦¦:¦:¦¦
.¦¦:-.
I
¦' . im**                          ."¦•¦:|
•    «.•      '    •                           •                                                     •                                 ...                                                                      f
':¦;•••..••: ..    ¦- ••/•                    i'& -                    I
!



/   •      --Q". ¦'"
• '• ¦¦• Hi.- •.:•¦'¦:
•    •*¦ ?.%'.            ¦     •' .   •¦'".   •               '/•'¦¦'      ¦':'•'-  h\
•'• W-- *»   - • *'   *  . ./ •••••'•, i   ''•   .•>!
-       ,*.:....,v.>..-.-'.'/>    •.„-   v   •   . -V/.tiv
• '        '¦ '". •/ 1*:-.      »,   •>* '. '<;•
•-•.'V-'       /. .-. /¦>/.';« :-.--:.;-v

•:   «*
'/¦;-.r,',;.;/^-:r;-.^;^- :\ 5t:.\\->?'i-.--v..
^••:^v/:.^^\:;::'V-:---.v^'-': '^:'^r;'r:.':.:¦
¦ '.-?:.• :•:••;tfS.ifoO iv.V:<.•:.-:•.¦:.:•;:;¦¦>•.•//;i:¦;•.••". •.;
h > v •••:". * > * * 4*-.: ;«v*'* -
'.t'-y.'r'.-'kt}''lt-^lx
•.*/•.•"'.•*.'?--.-'.'%-.,'y'-i'*'v-.*•'¦¦*• ¦.-!.¦••"••¦•¦••:   •••.-'••• V '
-.V-      '•¦.•'- ¦   • '¦¦• ¦ vjV '¦• ¦     .'.
V iJ'  :¦•¦:¦¦   ¦ '; -.
'.:.• ;.v••.'>.%4>*:ii'>y,f""•''.,'. '¦.' •'.'¦¦'*.**» ''.'¦"ri "^ •
:vV>:;:;^:Wy^;.v;..;.v:;.:y;;;:>i .    .
•^•'">'\W '-V'.*. ,i^\'- '.•'•.'•    • ' • '•,s.'f'-*?-:-.'-,r-   ..'

.'••.'.  -Nl
Wl'      -y-. —
>¦«        *
-p	
jC	•
tjIrH	
¦H	H
5-1	£2
	:?
m	
O	+j
	Cfi
tn rd	
S	cu
•H	
M	w
fl>	rH
13	CD
S'S	
^	a
o	-->
C)	^~^
¦rH	5
^   ^	
m	O
M	in
(3	'  '
iA	
m	

-98-
S->*y
•H
>
r-t
¦a
1
CO
I
I
' ?
37. Thermovision of bricks
under 2 cm. of plaster.
o
o
I
38. Overall view, west wall.
r
O
I—1
I
39. Thermovision of area
below murals, west wall.
-102-


/
¦ ¦' V-   ¦ .¦'¦-¦.   ,-.¦¦,'.' ¦¦     ¦'.                   ....    — ¦"!. . ¦•
¦   "•    , >/   5^K /


''^^


¦¦:¦. x!:«x&$>£J    ¦¦¦¦¦¦>&:*¦¦¦¦¦¦

VJ
«f$wp
m	
0	
m	
c	
¦H	
!h	
0)	
"S3	
t^	
0)	
n	•
	c^
0	n
-H	
,£	O
&4J	
to	O
H 4:	
O	Uj
,	
O	
"*	
-103-
41. West wall, detail of
right (notth) area.
-104-
•H
C    •
O   r-l
w
-H   O
it

*
*

o
Ln
I
43. Thermovision of the
central portion of the
west wall.
o
CTl
I
44.  Tlierrovision of the left
Dortion in photo 43.
H-	ft)	3	el-	3*	CO	t"1	01	S	O	•tt	rt		ft)	i«	.«¦"¦»»	tr	3	H	<	Hi		—    O	3*	13	Hi	Pi	
*	tr	ft)	s'	H-	O	CD	pf	3"	H>	ft)	3*		H.	H	13	fD	d	3	H-	d		13     ft		ft)	3"	CD	
K.	3	3	ft)	t£	3	0	ft)	CD		3	fD		fD	ft)	H-	H	H		d:	3		H-    3*	g	3	CD	ft	
•	o	3	rt	3*	fD	3	P.	3	3"	(D	H	o	ft)	Pi	n	O	ft)	ft	.{3*	O	>	O     fD	H-	CD		CD	O
	n	(t)		fD		ft)	0		H-	H	3	3	CO	H-	rt	«:	J-1	3*		rt	rt	rt    ii	Pi	I-1	O	O	Pi
CD	3	H	<	H	O	ii	s£_	01	vfl	01	ft)	fD		ft)	d			ft)	o	H-		d	H-	"*t	3	ft	Pi
3	ft)	^	0)		Hi	Pi		O	31	9	t-»		rt	ft	H.	<+	C	rt	d	O	Hi	H    ft)	rt		M		M
H-	H1		oi	H-		O		ft)	fD			H-	O	H-	CD	tr	&		ii	3	H-	fD    ii	^<	rt	^<	ft	^
01	H-	rt	ft)	•	ft		0	3	hi		cr	m		0		(D	OS	O	-	H-	H	<T>		3"		3*	
01	rt	3"	ti	l-S	3*	Hi	3	3		CO	CD		ft)	.3	*"<¦		,;¦>	ft)	Hi	3	01	^    ft)	Hi	H-	13	CD	CD
H-	*<	m	H-	*	fD	0		CD	H-	d	tr	^Q		01	00^,3			01	tr	iP	ft	C1     03	?i	03	ii	li	3
<:		3			03	n	rt	Pi	3	o	p)	H	S		'*->	d	<	fD	(D		^	•«-»	O		O	3	O
CD	3		13	CD	CD		3*		Hi	3*	<	CD	ft)	P-	t	H	H-		'ti	O		•'     0	3	Pi	3	ft)	d
	H-	fu	ft)	3		el-	CD	s:	>i		H-	ft)	H	3		ft)	01	ft)	3	Hi	rt	Hi		ft)	H-	M	U3
tr	vq		H-	H-	a	s'		H"	ft)	ft)	O	rt	3			¦ H	H-	01	0		tr		rt	ti	3		tr
ro	3"	13	3	01	0	fD	3	rt	H		H	H"		rt	^3	^»	cr		<	ft	F-	ft	3"	^	CD	<	-
3	rt	t~>	rt	01	3		ft)	3*	CD	i-i		K	ft!	3*	3"		H1	s:	H-	tr	03	tr	CD		3	ft)	
fD		ft)	CD	H-	13	3	01		Pi	CD	O		H-	fD	CD	o	fD	CD	01	CD		fD		ft)	ft	ii	rt
ft)	tr	d	&	<	0	rjO	O	rt		H	Hi	rt	ii			H1	a	!—'	H-		0		M	ti		H-	3*
ft	CD	w		H-	3	ft)	3	3*	CD	ft)		CD		3	O	CD		H	O	(D	Pi	«	3	f&	CD	ft)	CD
3*		H-	rt	rt	fD	ft	H	CD	'3'	rt	rt	3	Hi	d	3	ft)		*»	3	(Q	pi	ft!	ft)	ft!	M	rt	
	wQ	cr	•3*	^	3	rt	*<		H-	H-	3*	na	M	H	M	n	a			d		|->	H1		©	H-	Hi
rt	H-	H<	fD		ft	H1		i-3	01	O	CD	rt	O	ft)	^<		0	3	01	H-	' n-	H"	M	13	3	O	31
3*	<!	m		rt	m	fD	3	tr	0)	3		CD	s:	H"		Pi	$	O	ft	13	3*	•N		ti	CD	3	CD
CD	fD		fD	3-	^		ft)	CD	Ul.	03	3	pi			3	H-	CD		d	3	CD		^	O	3	01	>i
	3	fD	ft)	ft)		O	rt	¦H	<	3*	ir1		,—*	<	CD	Hi	<	01	P.	CD-	ii	rt	H-	cr	ri-		3
<		X	03	3	01	H)	ffi	3	H-	H-	ft)	rt	13	H-	ft)	Hi	CD	H-	K	3	¦3	¦¦'..tr	3	ft)		ft)	O
ft)	cr	13	li-		d		H	0	rt	13	01	O	K-	01	3	CD	ii	<£5		ft	ft)	m	P.	cr	ft)	3	<
03	*<	H>		3	0	>	W-	<r	••<!		rt		O	H-	01	H	' ¦»•	3	o	•	H1		O	H"	U	O	H-
ft)		ft)	fu	O	3"	3	ft!	H-		3		a.	ft	tr		CD			Hi			rt	«	k;	13	3	01
M	rt	3	3	H		U3	H	01	,—,	H-	W	H	d	H"	rt	3	Z	O			cr	3*			CD	03	H-
H-	tr	ft)	P>	3	ft)	3*	01	H-	H-	U3	d	ft)	H	CD	3*	O	tr	Hi	rt	rjd	CD	CD	ft)	o	ft)		O
	CD	rt		ft)	01	H-		O	3	tr	13	.<	CD	••	ft!	CD	CD		3"	d	3T	H	cr	0	li	ri-	3
3		H-	S	H		fti	tr	3		n-	13		CO		rt	01	3	03	fD	rt	ft)	3	0	Ii	CD	D'	
d ¦	13	O	fD		13	H	CD		H		CD	Pi		«				rt			<!	ft)	<	ii	Pu	a	fD
H	H	3	01	3	H,	H-	3	fD	CD	cr	H	H-	^	ft)	K	O	ft	H	3	rt	H-	H	fD	CD			vQ
ft)	CD		ft	ft)	rt	3	fD	?Q	01	CD	3	H	^	03	CD	H)	3*	d	It*	tr	o			M	O	rt	d
M	oi	Hi		01	a		ft)	d	13			CD					CD	o	ft)	fD	K.	¦ 13	*-—»	13	3	3*	H-
to	CD	o	S	O	3*	ftl	rt	H-	CD	Pi	3	o	cn	cr	«	rt		ft	01	3		.?:    h-	13	O		ti	13
•	3	H.	ft)	3		li	3"	13	O	d	d	ft	o	^	CD	fD	O	d	rt		^	O	H-	3	rt	(D	3
	O		H	>i	ft)	fD	*	3	ft	CD	H.		•—		Ii	3	ft)	li-		rt	ft)	ft	n	Pi	tr	fD	CD
	fD	rt	H	;<!	3			CD			ft)	o	*!	cr	fD	13	3	ft)	CO	tr	01	" ' d	rt	03	CD		3
		3*			Pi	0=		3	rt	ft	M	0		H		CD	CD	H1	d	fD		H	d			13	rt
	0	H-	3	3		O	Hi	rt	O	0		H.	IA	H-	ft)	H,	ii		13		ft)   ¦	CD	ii	rt	rt	ft)	
	Mi	m	d	ft)	l—i	o	3*	•*			ft)	H	~~.	CD	cr	ft)	ft)	CD	13	H	rt-		CD	Q	o	3	Z
			n	rt	H-	d	fD		3	o	3	CD		H»	H	rt		h-1	CD	CD	rt	£¦			13	CD	ft)
	02	a*	&	CD	3	3		n	O	o	P.	i_i		H*	CD	d	3	CD	hi	01	»i	¦V   ft)	*>•	rt		M	01
	0	H	H	H.	01	CD	3	ft)	H	3		ft)		^		H	ft!	3	3	d	H-	03	CT!	3*	Q	03	
	tf	ft)	en	H-	CD	3	ft?	01	3	T3	rt	rt			rt	CD	01	CD		I-1	cr		«_^	CD	CO		31
	01	oi		ft)	CD	rt	pj.	rt	Q)	O	3*	M',		CD	O			3	3	¦ft	d	01	»		3	O	ft)
	rt	rt	H.	H	&	(D	fD		M	3	CD	b		X		t,	13	ft	d	01	rt	.¦ c		13	ft	Hi	ii
	fa	H-	3	Cfl		Pb	n	fu		fD		3		13	i-J	CD	O	«	ii		CD	¦-',   13		(D	(D		Pi
	3	O		•	O		H-		•o	3	£	05		O	CD	H.	H-		ft)	Z	pi	H	H	3	*i	ft	(-•
	O		ft		H-	«¦—«*	ft)	«	H	ft"	CD			01	3	0	3	cr	H1	CD		H-	3	CD		3"	K
	CD	rt	3*		H	03	H	rt	ft)		01	tr		H-	Pi		ft	CD	»	ii	ft	03		ft	O	CD	
	Cfl	3*	fD	3*	¦*«	CD	01	3*	01	3	rt	fD		3	CD	•i	CD	3	,—,	CD	o	'.      H-	ft)	ti	Hi		ft)
		CD		03		fD		fD	rt	ft)		ft		lQ	ti	CD	Pi	fD	13			3	H1	ft)		S	cr
	3"	H,	CO	01	3*		d	H.	CD	rt	s:	^				O		ft)	H-	O	ft)	ip	H	rt	rt	fD	M
	H-	3	ft)	d	JU	ft!	a	3	H	CD	pi	CD		01	el-	0	ft)	rt	O	O		H-1		H-	3T	01	fl>
	iQ	ft)	3	3	<	•0	CD	ft)	•».'	n	(-»	(D		O	s'	H.	cr	3*	ft	3	3	Hn	ft	O	CD	ft	
	3*	|_i	fD	H-	CD	id	Pi	t-1	«^	H-	t-«	3		3	(D	Pi	0		d	13	ft*	!	3*	3			rt
	h*			3		CD			s:	ft!				fD	H.	CD	<	rt	ii	ft! .	H*	d	CD		H»	$	o
	*!			U3	ft)	3 *	cr		3* O tr	H 01		rt 3^ CD			3 ft) H	Pi	CD o ii	tr	fD '*» * 0Q	ii Pi	!	3 H-Hj O ii 3		O H)	CD H» ft	ft)	
-108-
45.Vest wall, left nural,show-
ing dark area possibly
caused, by humidity.
¦

L.
«
46. Thermovision of west wall,
right (north) mural. The
dark lines visible here
were the only anomalies
detected in the entire panel.
-no-
fc	H
H	¦H
fd	<-
	
R	<U
F	¦H
	N
c	(13
k{	£
a	
B<	a)
3	H
C/3	i-l
	a)
+J	T3
U]	
3	¦H
-	1
SJ	
1	cr_
r^	
w	
•111-
48.Thermovision of the "Last
Supper", showing the uni-
form emission.
-112-

^

_
_
49. Thermovision of the vault-
ed ceiling just above the
"Last Supper':.
¦113-
50.Thermogram of figures in
the "Last Supper' ,close-up.
-114-
k
51. Thermogram of the head
of Christ, the "Last
Supper", Milan.
-115-
VI. , Conclusion
¦* ¦
The use of the Thermovision apparatus as an instrument of NDT
has yielded valuable information both in our research on the
"Last Supper" mural, and in -the search for the "Battle of
Anghiari".  Although considerable research with this apparatus
has yet to be completed in the Hall of the 500, a wealth of
information concerning the architectural history of the Hall has
already been collected.  The presence of all the major structural
elements, hidden by subsequent alterations of the Hall, has been
documented.  In come cases, the detection of windows or arches
has precluded the possibility of finding traces of the Leonardo
mural.  As well, the Thermovision apparatus may allow us to come
to a more complete understanding of the types of masonry materials
beneath the Vasari rturals.
The preliminary Thermovision scanning of the "Last Supper" mural
was of primary importance to understanding the situation in the
Hall of the 500, west wall.  No structural elements were detected
beneath the "Last Supper"; probably due to the high i.r. emissivity
of the unusual components of that painting.  Indeed, the only means
of revealing thermal gradients in that painting was by exposing
areas to a warm air flow; clearly, unusual properties of density,
conductivity and i.r. emission are inherent to these painting
materials.
The west wall presents similar thermal characteristics; just as
the "Last Supper", almost no structural elements were visible beneath
the Vasari murals.  This is in sharp contrast with the findings on
the east wall, where numerous arches, windows and other elements
were clearly evident.                                                 '
Further studies concerning external factors which might effect
the findings, the influence of variations in the wall structure, etc.
are needed in order to come to a complete understanding of the causes
of such anomalous behavior.  Nevertheless, the causes for such thermal
homogeneity in the west wall areas beneath the Vasari mural must be
determined.  One means of achieving this end may be by employing
panels which evenly radiate the wall with i.r. radiation.
¦116-
APPENDICES
I- A Brief Chronology of Leonardo's Activity 1494-1508.
II. List of the Materials Purchased by Leonardo for the
"Battle of Anghiari."
III. The "Last Supper5' and "Sala delle Asse" Research
A. Photographic Documentation
B. Therraovision Study
C. Photogrammetry of Surface Refief
D.  Infra-red Reflectography Study
E. Analysis of the Cross-Sections
F. The Future and Dissemination of the Research
¦IV. Other Methodologies With Possible Applications to the
Search for the "Battle of Anghiari"
A. X-Radiography Transmission Techniques
B. Electron Emission
C. X-Ray Fluorescence   -
D. Gamma-ray Radiography
E. Neutron Activation
F. Electro-Magnetic Variations
-117-
A BRIEF CHRONOLOGY OF LEONARDO'S ACTIVITY FROM
1494 TO 1508
DATE
LOCATION
ACTIVITY
1, November,1494
2.    1494
3. May 23-June 16
1495
4.  1495-96
5.  1493-98
6.  1498-99
7.
1500
Milan
Milan
Florence
Milan
Milan
Florence
Florence
8. 9-	1502 1503	Urbino Florence
10.	May 1,1504	Florence
11.	April 30,1505	Florence
12. 13. 14.	June 30, 1505 May 30, 1506 1507-08	Florence Florence Florence
Leonardo abandons the possibility
of casting the great bronze horse
for the Sf/orza monument.
Decoration of rooms in the Sforza
castle.
Brief visit to Florence. Note that
on May 23 Antonio da Sangallo was
nominated architect for the Sala
del Gran Consiglio.
Decoration of the ' Carrierini' of the
Sforza Castle.
Studies and paints the "Last Supper
May have studied a project for the
fortification of the city.
May have participated with Baccio
d'Agnolo in the designs for the
wooden decoration of the Sala del
Gran Consiglio.
In the service of Cesare Borgia.
Project for diverting the Arno
to the sea.
Life-size drawing for the "Battle o
Anghiari"   ali~eady begun.
Prepares the plaster for the mural
in the Sala del Gran Consiglio.
9:30 a.m. begins to paint.
Departs for Milan.
May have continued working on the
"Battle of Anghiari".
•118-
LIS^ 2£ MATERIALS PURCHASED BY LEONARDO FOR THE "BATTLE OF ANGHIARI"
On April 30, 1505, Leonardo and his three helpers prepared the
plaster for the "Battle of Anghiari" mural in the Sala del Gran
Consiglio.  Although the exact painting technique has never been
determined, most of the constituent materials are known.  In the
following list, the weight and relative volume refers onlv to the
sum oi; the materials purchased, and does not necessarily indicate
the proportion of the materials used for the painting.
Also, this list refers only to documented evidence.  Other
materials may well have been used, of which no records exist.
.  W£±2h;t                                     Material                       Relative Volume
1. 343 libre............gesso volterrano.................74%
2. v:'8a.-.l±br§ 8".oace. . . . .pece greca............. . ..___...20%
3.  20 libre............biacca alexandrina. ...............4%
4.  11 libre 4 once. ... .olio di lino sema.................2%
On August 30, 1505, the following was also purchased;
60 libre...,.....,..gesso
4 libre 4 once.....cera bianca2
11 once.............olio di noce
10 once.............biacca
1. In 1505, 1 l'ibra=. 3395 Kg.
2. The cera bianca (white wax) was used "per impannare le finestre'8
(to cover the windows with waxed sheets).
-119-
THE ASSOCIATED "LAST SUPPER" AND "SALA DELLE ASSE" RESEARCH
In order to better understand the mural painting technique of
Leonardo,i.e. to know what to look for under the Vasari murals
in Palazzo Vecchio, research projects were launched on the two
other existing examples of mural painting by Leonardo, the wSala
delle Asse" in the Sforza Castle, Milan, and the "Last Supper"
mural in Santa Maria delle Grazie, Milan.
Particularly frustrating is the complete-absence of definitive
research concerning the painting technique of the "Last Supper".
The presence of elevated quantities of lead white, for example,
could have definite bearings on our Florence research.
Somewhat inadvertently, we were requested to extend our research
so to document the current state of conservation of the masterpiece.
Although it is much too early to present any type of finalized
report, it is pertinent to this document to briefly list the types
of research accomplished to date, and those which are anticipated.
These findings will be presented to the public via exhibition in
the refrectory of Santa Maria delle Grazie in the autumn of 1978;
A- The photographic documentation.  The entire work has been
documented in small sections with normal light, black and
white,and color, in 4x5" frames.  Color raking-light
exposures have been made of selected details.  This technique
is particularly useful in this case, as the "Last Supper"
has an unusually elevated paint layer; it appears like the
skin of an alligator.  The elevated islands indicate areas of
original color, and the surrounding vallies, although often
containing pigmentation, can be recognized via this technique
as not original areas.
Ultraviolet black and white and color exposures were made
of all the figures.  Since the entire surface of the painting
was covered by a thick layer of shellac in 1950, much fluor-
escence limited the effectiveness of the color ultraviolet
study.  The b£ack and white work, however, yielded some important
clues concerning * fogging1, e.g. repainted,,areas
•120-
Also, color infra-red exposures have been made of all     '
the figures.  Through this technique,, different pigments
which may appear the same innDrmal illumination may register
entirely differently -on color infra-red film.  For example,
compare the two tonalities of the blue robe of Matthew
(photograph page   ) in normal light.  They appear the
same.  In the infra-red photograph, however, one entire
area appears red.  One can conclude, therefore., that the
two blue areas which appear the same, in normal illumination,
are not the same.  The detail of John's robe (photograph
page   ) is even more revealing.  The small islands of color
which appear red in the infra-red photograph correspond to
areas of original blue, probably azurite„  The surrounding
blue is repaint.  These two blues are not easily distinguished
in normal illumination.  Also, in the window area behind
John, the two hills and the tree-top appear red, indicating
a correspondence with the original blue in John's robe.
Color infra-red photography may well yield the most valuable
information concerning the originality of certain areas of
the "Last Supper".  Further studies need to be accomplished,
especially laboratory work whereby pure samples of all the
natural pigments would be studied via color infra-red photo-:
graphy, to more fully comprehend the material gathered to
date.
B- The infra-red Thermovision study. As previously discussed in
the chapter concerning our Thermovision research in Florence,
the preliminary scanning of the "Last Supper" proved crucial
to understanding the confusing situation in the Palazzo
Vecchio.  Indeed, what at first we considered a technical
failure (the inability of the equipment to reveal structural
abnormalities in the wall behind the "Last Suppex*"), proved
an advantage to our research in' Florence.
q.   The photogrammetry of the painting surface relief. As previous-
ly stated, the islands of original color of the "Last Supper"
tend to stand out in relief, while the repaint often lies
in the depressed areas of the painting. ^Hence, if a means
could be developed for creating a 'map' of the entire paint-
Ill
n
A. The "Last Supper",
normal color film.
&C
B.Same area, color infra-
red film. Note reaction
in "fltthpu1 c Kino v-^k^
-122-    |
ing surface, one would surely have a good understanding of   4
just how much of the painting is original, and how much is
repaint (we calculate that only about 25-30% of what the
visitor sees is original).  Photogrammetry, or stereo-photo-
graphy, may be one possible means of creating just such a
map.  With a sensitivity for relief/depression in the order
of 0.1 mm., a point system diagram can be created, as seen
on the following page.  A point source, believed to be an
area of original paint, is assigned the reference number 0
by the operator of the graphic interpretation equipment.
All the other points of the painting can then be assigned
numbers, accoring to the degree (in mm.) of relief or depress-
ion in relation to the original point 0.  Imprecisions are
inherent to the system.  For example, an area which extends
9.1 mm. in relief does not necessarily mean that the paint
layer is that thick;  more probably, it reflects a bulge in
the plaster layer.  However, together with optically-creat-
ed diagrams and laser contourgrams, this system will prove
invaluable to understanding and differentiating between
original and repainted areas of the painting.
D• The infra-red reflectography study.  A Hamamatsu infra-red
vidicon system, attached to a black and white television
display system,was used to scan the "Last Supper".  A Kodak
87C filter was used to minimize the visible light.  This
system, with a sensitivity up to 2.2 microns, can be most
useful for revealing underdrawings in oil paintings when
the drawings have been executed in animal or vegetable black.
The method that Leonardo used to "rough-out" the figures
for the "last Supper" has been a continual source of spec-
ulation.  He may have used a full-scale cartoon, in which
case the figures could be traced via "spolvero" (outlining
the forms with pin-holes in the cartoon, then passing over
the holes with powered charcoal thus leaving a series of
dotted lines on the wall).  However, this techniques pertains
more to the rigid mural concepts of the Ouattrocento than
to the spontaneous nature of Leonardo.  It is more likely
J. 4& A«f
t
Droftle A-A
f\
Detail  of  the  photograirahetric  study  of   the  relief  of  the  paint
layer of  tte   "Last  Supper".     The  numbers ' (in mm.,)   indicate relief
or depression   (-)   iii relation to original  point  0.
-124-    I
i
that Leonardo "roughed-ouf' the figures in a free-hand
style, following indications from a smaller drawing.  If
the undermodelling was executed in a black pigment, there
exists the possibility of detecting traces via an improved
infra-red vidicon system.
One advantage of the vidicon system is that it operates
in 'real time', and the results can be recorded on tapes.
A limiting factor is that photographic documentation must
be taken directly from the television monitor.  Hence, even
with the highest resolution screens (in the order of 800
lines), part of the image is lost.  Furthermore, no special
photographic film has ever been developed for faithfully
recording the image of a television monitor.  Although the
first study of the "Last Supper" has yielded some truly
interesting data, major improvements in the system must
be effected before a coherent presentation of the material
can be anticipated.  In particular, difficulties in determin-
ing whether the camera lense of the vidicon system is in
maximum focus or not, and problems associated with assembl-
ing the composite of the entire image, present considerable
obstacles to the research.
The analysis of the cross-sections.  The precise painting
technique of the "Last Supper" has never been established.
And as previously stated, the detection of certain elements,
such as an elevated quantity of lead, would have a direct
bearing on our researh.uin.-. E-locence'.
As well, an analysis of contemporary samples may shed new
information concerning the state of conservation of the work.
How has the shellac, applied in 1950, altered the painting?
Can traces of tiny spores be confirmed?  What are the types
of repaint, and are they soluble?  Momentum is getting under
way, as a result of our research, for a new project of con-
servation of the painting.  These analyses will be the found-
ation for any new intervention.
On the following page is a theoretical reconstruction of
what a cross sectional analysis may show.  The numbers in-
dicate: 1) the paint layer (tempera, oil,"or both?); 2) a
white layer; 3) white layer (one. may be lead white); 4) a
resinous layer; 5)gesso layer; 6) plaster layer; 7) wall.
The "last Supper", g. Maria delle Grasie, Milan. Theoretical reconstruction
of the painting technique, as viewed in cross-section,      '"
1
2
3
4

¦<??'-
to
J


•126-
LIST OF SAMPLES REMOVED FROM THE
"IAST SUPPER" MURAL
1. Section of the tablecloth with blue decoration, immediately
below Mattheus. Is the grey original?  What is the nature of
the blue pigment?  what is the yellow preparation laver?
2.. Section of the tablecloth, apparently without blue decoration.
This sample was removed from an area immediatelybgtole n.l,
but in an area where no shellac was applied (1950). ^What
xs the nature of the grey pigment?  Comparing with sample
n.l, how has the shellac affected the paint laver?
3. Section of Christ's mantle, composed of a dark'blue over-
paint on a lighter blue original,  what is the nature of
of the overpaint?  what is the nature of the original blue?
(Compare findings with color infra-red photographs}..
4. Section of sky, removed from the viewer's upper'left area
of central window. Sample composed of a white on blue, with
a grey preparation. what is the nature of the white, and is
it original?  what is the nature of the blue and grey?
5. Section removed from panel of wall, architecture on the
viewer's left of Christ. Sample composed of a dark brown,
shiny overpaint with some traces of a. light red (possibly
original) pigment underneath. What is the nature of the
overpaint?  Solubility?  Are there any traces of an original
pigment underneath?
6.  Section removed from Christ's right sleeve, just before His
hand.  The sample is composed of a bright-red original layer,
covered by some darker red overpaint.  The area is quite shiny,
suggesting an elevated quantity of shellac was applied.  What
is the nature of the two reds?
7. Section removed from the table area immediately below Christ's
right hand, composed, of grey and white pigments, preparation
and intonaco.  What is the nature of the pigments and prep-
aration? Are there traces of lead white?
8. Section removed from Peter's right sleeve, just before his
hand holding the knife.  Sample composed of ^rown pigment and
preparation layer.  Are there traces of another pigment strata
under the brown?  Or is the brown original?  If not, is it
soluble?
¦127-
^ "^ST. SUPPER" SAMPLES (CONTINUED)
9. Section removed from the tablecloth under the right elbow
of Judas, composed of a dark grey pigraent (believed to be
original) on two layers of off-white preparation.  What is
the nature of the grey, and its binding medium?  What is
the nature of the two preparation layers?  Traces of lead
white?  Tempera?
10. Sample removed from area under the table, viewer's extreme
left of painting, at the border of the painting.  Section
composed of brown overpaint, traces of red underneath, and
yellow preparation layer.  Is the underlying red original?
Solubility of brown overpaint?
F" The Future and MsseMnation of the Researd,.Concerning the
Last Supper" and "Sala delle Asse" mural paintings.  As pre™
vxously stated, the primary motivation for launching the "Last
Supper" and "Sala delle Asse" research projects was to gain a
quantitative understanding of Leonardo's mural painting technique.
We bexieve.that the technique of the "Last Supper" is quite
sxmilar, if not the same, as that used for the "Battle of Anghiari".
For example, the verification of the presence of lead white in
the "Last Supper" would encourage the application of electrical-
field technology in searching for the "Battle of Anghiari".
As a by-product of our studies, we now possess the most complete
documentation concerning painting technique and current state of
conservation of the "Last Supper" that exists.  Undoubtedby, this
documentation will aid the Italian authorities to examine the possib,
llity of future interventions of conservation.
As a part of this documentation, we are interested in examining
. means of non-destructively describing the original and repainted
areas of the painting.  Color infra-red photography has yielded
important information; together with maps of the relief of the
painting (photogrammetry and"laser contourgrams), we should be able
to indicated the original areas with exactness.  As well, it is
beleived that Leonardo, used an underdrawing for the "Last Supper*\
¦¦ We hope to continue experimentation with an improved infrared
vidicon system, to veirify the presence of an underdrawing.
The "Sala delle Asse" research, project is also an important
element in understanding Leonardo's mural technique.  One must
also remember that this composition was discovered under a' layer
of whitewash only 80 years ago. ' During the course of•our work, we
too found traces of Leonardo's original charcoal drawings still"under
whitewash.  We have proposed to¦use this occasion to test various
methodologies (e.g. neutron activation), and thereby establish their
relative capacity for detecting a mural hidden by .whitewash or
plaster.  This research, in turn, .might find a direct' application .'.
to our work in the Palazzo Vecchio.
On the basis of our findings in the "Sala delle Asse", yet an-
other problem comes to light: How to remove whitewash and plaster
from the surface of a mural?  To date, the only means is via the . .:' / -
hammer and scalpel.  We wish to use this particular example for
encouraging research in this difficult area (e.g. laser cleaning,
chemical, ultrasonic etc.).
-129-
' IV. OT§ER METHODOLOGIES WITH TO§SIB^ APPLICATIONS '
22 THE SEARCH FOR THE _^BATTLE OF ANGHIARI"
A, X-Radiography Transmission Techniques "
Generally, mural paintings are covered over either by layers of
whitewash (1-5 mm.), layers of plaster (1.-4 cm.), or layers of brick
and plaster (over .10 cm.),  m the case of the mural paintings  of
Leonardo da Vinci, the first two cases are believed to exist today
in the Sforz-a Castle, Milan; the latter two cases may well be applic-
able to his "Battle of Anghiari" mural painting in the Palazzo Vecchio.
Together with ultrasonics and infra-red thermovision transmission
techniques, x-radiography may yield important information concern-
ing wall structure.  This technique .is generally associated in art
conservation with its ability to describe stratification and the
distribution of lead white in oil paintings.  In view of the documented
evxdence that Leonardo used a considerable quantity of lead..white
for the "Battle of Anghiari" mural, x-radiography might detect traces
of that pigment through a layer of plaster or brick.
In addition, by placing the x-ray film near the surface of the
Vasari mural, and the x-ray source on the opposite side of the wall,
using an energy source in the order of 300-400 KeV, it might be
possible to distinguish the outline of the masonry materials immediately
beneath the Vasari murals.
B. Electron Emission
To date, there exists no method for detecting the presence of
pigmentation beneath whitewash, plaster, brick.  In order to come
to terms with this problem, one would probably leave the various
transmission techniques for some type of excitation/emission meth-
adology.  However, the obstacles to developing such.a system are
indeed formidable.
One possibility might be to employ electron emission radiography,
whereby an image is created by electrons emitted by the pigments
themselves, upon irradiation by high energy x-rays.  The theory is
simple.  A sheet of Industrial x-ray film would be placed between
the x-ray tube and the mural, with the emulsion side down on the
painting.  One would expect to use a minimum exposure of 250 KeV
-130-
at 10 mA for several minnt-oc   o~ 4. ¦, • .
teS'  So to llm±t the 'fogging' of the
image by direct exposure, only hard v M(n +-¦
Thprefnro                                '   iy hard x~radxatxon would be desired.
Therefore, a copper filter (8 mm. minimum thickness) would be
Placed at the tube to remove the soft x-radiation.
One of the major obstacles to this method that can be anticipated
would be an inability to rollprf =n 4-w    ¦                                        p
0  ,„  .                          Y   c°llect all the emitted electrons due to
scattering and a high level of absorption.
C. X-Ray Fluorescence
,Jl  ^  flU°reSCenCe ^hniques, the medium and heavy elementE
can be identified ncn-destructively in real time.  The apparatus,
however, is effective for surface measurements up to about 600, m.
The penetrative resolution is a function of the radiated energy
and of the elements present in the mural examined.  Further ex-
periments to determine the applicability of this technique are needed.
D. Gamma-Ray Radiography
Factors limiting the application of this technology to the ex-
ploration of wall structure are the long exposure period required,
and the need to implant a radioactive source nearby or within the
wall.  The health hazards represented Vw fh^ +.^„u •    /
^cj-^a itpi,.faentea by  this technique (radium or
cobalt capsules) are considerable obstacles.
E. Neutron Activation
Accelerated charged particles, photons, epithermal or fast neutrons
may well yield new possibilities for inducing and detecting radiation
in walls.
The msot promising techniques would be to expose the wall to thermal
and/or fast neutrons, thereby inducing limited radioactivity.  At various
intervals after the activation, industrial x-ray film is placed against
the wall.  Numerous pigments may possibly be identified in this manner,
as their component elements have different rates of decay.  Studies
to date show that the elements most easily activated and identified
by means of thermal neutron activation are aluminum, antimony, arsenic,
cobalt, copper, gold, manganese, mercury, potassium, silver, sodium,
and several other possibilities.1  Iron, a component of all the earth
Sayre and Lechtman, "Neutron Activation Autoradiography of Oil
Paintings-", Studies in Conservation, 13 (1968), 161-185.'
¦131-
colors and perhaps the mast commonly found piament ^     ,
ings, is detected by fast neutrons.                     "     " "U"1 ^^ '
plasBteefrr:ia00rnderng a S6arCh f°r mUralS hldde" ^  «*"~"* «
IZtZZ       \ !U  " aCtlVatl0n ^hnique, some of the many
vaiiables must be considered:
a) What is the nature of the pigments sought (i.e. the number
of susceptible atomq nPr ^m 2   *   ^
absorption d  ¦                                          hS probabili^ of neutron
absorption durxng time T of irradiation)?
b) What-is the ability of this technique to penetrate *h„
material /.a-i*.   ,   ,                                      penetrate the covering
material (whitewash, plaster)?
in'™'5 the ener9Y °fthe emittea beta particlss »*» *««*>¦
ed porLbT                          fleM °f 3eOChei»i-1 eXpioration has develop-
ed portable neutron activation equipment for work in the field l
The M]or obstacle for research in thi= =
,„i1aM1i, ¦   ,_     research m tins area appears to be the
availability of the equipment and cost factors.
F.Other Electro-Magnetic Variations
Preliminary experimentation suggests that electrical field
measurements may yield important information concerning the
wall structure.  in particular, the "Battle of Anghiari» mural,
which contained a considerable quantity of lead white, may be
detected by variations of an electrical field placed across the
wall.  Two plate electrodes would be placed directly upon the
Vasari murals, which would act as an insulator.  Excluding the
presence of moisture in the wall, if a current flow is established
a given area, one might determine the presence of a metal such
as lead.
1. Tolmie and Thompson, "Field. Equipment for Neutron Activation
Analysis", Nucl. Tech. Miner. Resour., (i969), pp. 489-505.
-132-
GENERAL CONCLUSION
Th1S report has described interrelated data collected by ultra-
sonic NDT and Thermovision techniques.  Their relative application
to the search for traces of Leonardo's "Battle of Anghiari" mural
painting has also been discussed.
Concerning the east wall of the Hall of the 500, ultrasonic
scanning has shown that the Vasari murals are composed of an in-
tonaco and arricio layer; these plaster layers are immediately
superimposed upon a brick wall.  The Thermovision scanning of the
east wall has revealedj complex architectural history of arches,
windows, doors hidden beneath the plaster.
It has also been shown that both ultrasonic NDT and Thermo-
vision techniques have agreed upon the presence of a serious an-
omaly in the structure of the west wall, strong echoes indicate
an unusual layer of material immediately beneath the intonaco layer
of the Vasari murals.  The returning sound waves appear to indicate
that this unknown layer is much denser than normal plaster.
Thermovision has clearly described this anomaly in the west wall,
although in different terms.  The unknown layer has such a high level
of infrared emissivity that no structural elements of the wall
beneath are visible.  The only precedent for such "hiding ability"
of a substance is the case of the Leonardo "Last Supper" mural.
Further refinement of these diagnostic methodologies (e.g. the
use of ultrasonic B mode scanning; thermal panels; improvement
of the Thermovision recording apparatus), combined with the applic-
ation of other, pertinent means of non-destructive testing, should
more clearly define the. perameters of this unknown layer and, hope-
fully, allow identification of the presence of pigmentation.
Furthermore, our work has indicated that ultrasonic NDT may be
used for describing comparitive density, porosity, etc. of build-
ing materials.  Also, through further refinement, ultrasound may
prove to be an effective tood for describing interfaces (or strat-
ification) , detecting cracks, areas of lesser density (with a
133-
d^rect application to locating areas of poor.adhesion in mural
paintings), and for defining areas of elevated humidity in walls.   "
Furthermore, it has. already been shown that some of the results
of this preliminary research can be extrapolated by the conservator
and scientist and applied to other related problems:
1. The Superintendent of Monuments of Florence has requested
us to apply our experience and diagnostic methodologies in
the cathedral, of that city, s. Maria del Fiore, to describe
the transept floor density. The equipment could also be used
for detecting cracks in the cupola.
2. The Simone Martini "Maesta"' fresco in Siena, has been afflict-
ed by elevated levels of.humidity and salts, the provenance
of which has remained a mystery for over 20 years (similar "
problems apply to the Luca Signorelli frescoes in Orvleto,
the Brancacci Chapel murals.by Masaccio, etc.). We expect
the union of ultrasound and Thermovision to provide some
answers to these problems.
3. Some other application may lie in detecting cracks in stone
and sculpture, normal and underwater archaeological explor-
ation, etc.
Finally, the exceptional level of cooperation among various inter-
national institutions towards the advancement of this project should
be underlined.  This occasion must be used not only to unite the
scientist, art historian and conservator, but also to affirm the
need for developing and integrating new methodologies capable of
diagnostics and describing a clinical card for any art object, whether
a mural or an entire church.