MINISTRY OF FINANCE, EGYPT.

Survey of Egypt.

Determination of the Exact

Size and Orientation of the
Great Pyramid of Giza.

By

J. H. ~OLE B.A., (Cantab) F.R.G.S

Inspector, Oomputation OffWe.

SURVEY OF EGYPT PAPER No. 39.

Government Press, Cairo, 1925.

To be obtained, either directly or through lInr Bookseller, f!'Olll
the GOI'ER:>MEST PUBLICATIO~S OFFICE, Ministrr of Finance
(Daw,iwin P.O.), Cairo.
Price . . . . . P.T. 10.
ES
 MINISTRY OF FINANCE, EGYPT.

Determination of the Exact

, Size and Orientation of the Great
Pyramid of Giza.
t In the beginning of April 1925 Prof. Borchardt of the German Institute of
Egyptian Archreology asked the Survey of Egypt for the loan of surveying
instruments in order to carry out a new accurate survey of the Great
Pyramid of Giza.
. After consultation with Mr. R. Engelbach, Chief Inspector Antiquities
Department, it was agreed that the Survey of Egypt would carry out the
necessary measurements after Prof. Borchardt had done the necessary exca-
vations and clearing to find the original base of the Pyramid on all four sides.
The construction of the Pyramid on the outside was as follows: The
desert was cleared down to solid rock and on this rock was built a pavement
which was accurately levelled. The actual base of the Pyramid was laid out
on this pavement leaving about 40 centimetres width of pavement all round
the bottom edge of the casing blocks. This width is, however, not exactly
the same on all four sides, it being 38 centimetres on the western side, 42
centimetres on the northern side, and 48 centimetres on the eastern side, at
the places where it could be measured. At the four corners of the Pyramid
the rock was cut away, giving a greater depth for the foundations of these
points. These excavations are rectangular in form and are called the corner
sockets.
The purpose of this present survey is to determine as accurately as possible
the exact size, shape and orientation of the original base of the Pyramid on
the pavement.

METHOD OF SURVEY.

FRAME WORK.

Eight brass bolts were cemented into the rock round the base, one near
each of the four corners, and, as these were not intervisible, four more v~ere
placed, one at about the middle of each side, in such a position that each point
was visible from the adjacent points on each side of it.
These bolts are numbered from 1 to 8 clockwise from No.1 which is at
the S.E. corner of the Pyramid.
 -2-

The positions of these points were determined by means of precise traverse

observations.
The distances were measured with the Base Line Apparatus designed
by M. ~1. Benoit and Guillaume using 24 metre standardized invar wires. Each
length was measured twice with different wires with an accuracy of about
1 in 500,000. . fou
The angles were measured on 4 arcs with a 6 inch Troughton and Simms des
micrometer theodolite. exe
The top of the flagstaff on the top of the Pyramid was visible at points for
1, 6 and 7 and was included in the round of angles at these stations. 15
The closure in angle of this traverse was found to be 9'6N which was
adjusted by adding 1'r' to each angle.
Using these adjusted angles, the co-ordinates were computed and a clos-
ing error of 8 millimetres in North direction and 0 millimetre in East direction
was found. These co-ordinates were adjusted by the normal traverse method len~
for closure.
The position of the top of the flagstaff was computed from the intersec-
tion of the sights at points 6 and 7. The computed horizontal angle at 1 to
the top of the flagstaff and point 2 was found to check with the observed
angle within 3'0N of arc.
a UE
dete
AZIMUTHS.
bloc
The azimuth of the line 6 to 7 was obtained by observations taken on run
Polaris near elongation against time. The final azimuth was the mean of at t
two arcs taken both face left and face right. The time was taken on a Zenith abo,
watch rated ,to keep sidereal time.
The error of the watch was determined by vertical circle readings on both
faces for both an East and a West star against time. The latitude of station
6 was computed from the known geodetic latitude of E 1, on the Great Pyramid,
and the distance of 6 North of E 1.
The azimuth of 7 from 6 was found to be
119 11' 07'I N East of true North.

EXCAVATION OF PAVEMENT.

NORTH SIDE.-The whole of the pavement, where it existed on the

North Side, was cleared and the original line of the edge of the casing blocks
on it was found for a distance of 55 metres. On 20 metres of this, the casing
blocks are still in position.

EAST SIDE.-An excavatiDn was sunk down to the pavement, starting

from the centre of the side and. extending 30 metres to the north. In this

.1
 -3-
~re~traverse
excavation the bottom edge of the casing blocks was found, giving a clear
line for almost the entire length of the excavation.
~ratus designed
ar wires. Each SOUTH SIDE.-Five pits were dug at various places along the edge j in
uracy of about four of these it was found that the casing blocks and pavement had been
destroyed. In one, tp.e top edge of the casing block was still in place and the
lton and Simms excavation was extended as far as possible along this edge, which was found
for a distance of 18 metres, the edge being clearly defined for a distance of
visible at points 15 metres.
~ stations.
9.6" which was WEST SIDE.-Four pits were sunk, the pavement being found in 3 of
these.
mted and a clos- In the fourth, the pavement had been entirely destroyed. The maximum
.in East direction length of the line given from these three holes was 28 metres.
traverse method

rom the intersec- PRECISE LEVELLING.

ntal angle at 1 to
nth the observed
Before attempting to survey these lines, it was decided that it would be
a useful check as to whether they really were on the pavement or not, if we
determined the levels of each point.
Paint marks were made at various places on the pavement and casing-
blocks, where they were found in the holes, and a line of precise levelling was
:rva~ns taken on run connecting each point with two Precise Bench Marks already established
was the mean of at the Pyramid. The descriptions of the points and their reduced levels
: taken on a Zenith above M.S.L. at Alexandria are given in the following list.

Ie readings on both
: latitude of station
the Great Pyramid,

~ it existed on the
of the casing blocks
es of this, the casing

~ pavement, startin.g
) the north. In thIS
 -4-

PRECISE LEVELLING AT GIZA PYRAMID.
=~=======================
aver
S.E. '
BM. DESCRIPTION OF POINTS. V A.Ll'E,

North East Comer.

2
Bottom of KE. corner-socket ... ...
Rock W. of B.l\L I =foundation of pavement

North Side.
Top surface of pavement W. of 2 (beginning of pavement)
596958
598913
,
I
"
']
the Ii
follo~
']
3 at B. on diagram .. 604129 as po
4 Top surface of pavement E. of remaining casing-blocks... . 604049 paver
5 Top surface of casing block (near 4) 619011
6 Top surface of pavement W. of remaining casing-blocks at A. 604115 of thE
7 Top surface of casing block (near 6) 619077 line c
it waf
North West Co'rner. cIamI
8 Rock E. of K.W. corner socket=foundation of pavement ... 598584 the Ii
9 Bottoll of K.W. corner socket ... ... ... ... 596034 'J
']
West Side. sure 1
~nta
10 Rock in hole L=foundation of pavement ... 598841
11 Top surface of pavement in hole K . 604129 end, :
12 Top surface of casing block in hole K . 619148 faces.
13 Top surface of pavement in hole .T 604154 ']
mean;
South West Corner. end.
14 Rock N. of S.W. corner socket=foundation of pavement 598798 J
15 Bottom of S.W. corner socket ... 598485 methc
on th
South Side. botto
16 Rock in hole H=top surface of pavement ... 604222 ']
17 Top surface of casing block in hole H 619333 differ.
18 Top surface of pavement in hole G 604220 on tb
19 Top surface of pavement in hole F ... 604264 mark
']
South East Corner.
were
20 Bottom of S. E. corner socket ... ... ... 503700 read.
21 Rock W. of S.E. corner sockeet=foundation of pavement 598305
'1
East Side. trave:
22 Top surface of pavement in the southern end, of the hole C .. 604193 These
23 Top surface of pavement in the northern part of the hole C . 604210 the P
sides
 f

-:s-

From this list we can see that the pavement is practically flat, but has
a very slight slope of about 15 millimetres up from the N.W. corner to the
S.E. corner.

MEASUREMENT OF THE BASE OF THE PYRAMID.

The method adopted for surveying the actual base of the Pyramid, i.e. the
the lines of the bottom edge of the casing blocks on the pavement, was as
follows :-
The 6 inch theodolite was set up on one edge of the excavation as near
~ as possible in the line of the edge of the casing blocks still traceable on the
9 pavement at the bottom of the excavation. It was then adjusted by means
1
5 of the sliding head on the tripod until the marks on the extreme ends of the
7 line on the pavement fell accurately on the centre hair of the telescope as
it was revolved about its horizontal axis, the horizontal plate being level and
clamped rigidly. A mark was then placed in the ground at the far end of
4 the line, so that it also fell on the centre hair of the theodolite.
4 This mark was checked, using the opposite face of the theodolite.
The theodolite was set up over' this mark, and after checking to make
sure that the vertical hair would accurately traverse the line with the hori-
1 ~. ~ntal circle clamped, a second mark was placed on the ground, at the other
9 end, in the continuation of the line. This mark was also checked on both
8 faces.
4 The line between the two marks was extended in both directions by
means of the theodolite, until it intersected the main traverse lines at each
end.
8 A mark was placed in the ground at each point of intersection. This
.5 method was used on all four sides of the Pyramid, with a slight modification
on the South side, since the top edge of the casing had to be taken, as the
bottom edge was totally destroyed.
2 The whole work was repeated two or three times ab initio; the maximum
3 difference in position of the final marks was about 3 centimetres, this being
9 on the extension of the short southern edge. The mean position for each
4
mark was taken as the best obtainable position.
The short distances between these marks and the nearest traverse points
were measured with a standardized steel band graduated in centimetres and
o read to millimetres by estimation.
!)

The co-ordinates of the points of intersection of the sides and the main
traverse were found, and the equations of the four sides were written down.
3 These equations were solved in pairs to give the co-ordinates of the corners of
o the Pyramid. From these co-ordinates the lengths and azimuths of all four
sides and the two diagonals were found.

- _ _d
~
-6-
'I
These are as folloWB:- found
this,
Side. Length. True Azimuth.
assun:
0
Metres.

North ... . 230'253 89 57 32

S:lUth '" . 2~0'454 8il 58 03
East ...
We;t ...
N.E.-S.W.
230'391
230'357
325'699
353
359
J4
54
57
55
30
30
45
f
\
North
South
N.W.-S.E. 325'868 314 57 03 !, F.ast
West

The position of the South side was found from the actual prolongation
of the whole of the existing top edge of the casing blocks, to which the bottom Me
edge .was a8sumed exactly parallel, and the horizontal distance of 5;5times --
the vertical height of the edge above the pavement. This height is given I
in the list of levels, namely No. 17 to No. 16, and is 1'511 metres. Therefore
the distance to be added is 1'511 X5'5/7=1'187 metres to the south of the
line of the top of the casing. This distance has been~ added to get the
co-ordinates of the south-east and south-west corners, and hence the actual
lengths of the sides. ?
In justification of using th~ constant of 5'5/7, reference should be made
to W. M. Flinders. The PyramM,s and Temples 01 alza in which he gives
the angle of slope as 51 0 50' 40" 1'- 05" also Borchardt, Geqen die Zahlenmystik
an der qrossen Pyramide; finally a fu,!ther check was made by measuring up
the slope of the existing casing-ston~ on the North side. This gave approxi-
mately 1'173 metres as the horizontal_distance for a difference in height of
1'496 metres (from points 4 to 5). The theoretical distance horizontally of
1'175 millimetres agrees within the limits of accidental error of the actual
measurement.

The mean dimensions of the Pyramid are thus :-

Central Axis. Length. Azimuth.

0
Metres.

North to South ... 230374 359 56 00

East to West 230'354 89 57 48
General Mean ... 230'364 3 06
 -.7 -

These should be compared with Prof. W. M. Flinders Petrie's results

found in 1880-1882 and published in the Pyramids and Temples 01 Giza. In
this work, Prof. Petrie, after making accurate measurements and certain
assumptions, obtains the following results:-

Length Petrie New Determination

Side. Difference.
1880 1925.
1m.

!
Ins. Ius.

North ... ' " ... 90694 90651 - 43

South ... ...
~ ... + 35
90695 90730
I
East ... ... ... 90677' 90705 + 28
I
i West ... . ... ... 90686 90692 + 06
ation
Ittom Mean ... ... 906S8 90694 + 06
times
From which we obtain the mean differences on the central axis:-
gIven
~efore
If the Axis. Petrie minus Cole.
t the
,ctual
North to South ... + 17
ma~
gIves East to West ... - 04
~ystik I
19 up
>roXl-
II :Mean (~imension + 06
;ht of I
tly of
Lctual
I This very close agreement of these mean dimensions shows the accuracy
II with which Prof. Flinders Petrie determined the data on which he based
his assumptions.
The comparison of azimuths is as follows ;-
I
Side. Azimuth Petrie. Azimdh Col". DileRooe.

. .
1
I I

N ...... ... ... - 3 20 - :2 28 + 0I 52

E ... ... ...
... - 3 H - 1 :)7 + 44
... ... .. . ... - 3 57 .- 5 30 - 1 3.'3
"
W... .. , ...'" - 3 51 - 2 30 +1 24

- 3 43 - 3 06 + 37

__.=l. ~ d&.tL.i2tlE.!.
---~
 -8-

These differences in azimuth are due to the fact that the new azimuths
are found from the actual directions of the. sides determined from the ex-
cavated pavement, whereas those of Prof. Petrie are of a hypothetical base
obtained by computing " a square that shall pass through the points of the
casing found on each side, and having also its corners lying on the diagonals
of the sockets."*
In the lists of azimuths given above it will be seen that the most dis-
crepant side is the East side which differs by about 3' from the other three. ,
In order to check this a further excavation was made on this side 23 metres
south of the original one. In this excavation the pavement was found, and
on it the line of the casing blocks was clearly. shown. This line was found
to lie exactly in the line as already extended, thus confirming the accuracy
of the original extension.
Prof. Borchardt pointed out a small line on the pavement which projected
a few centimetres from the edge of the casing-block about the middle of the
North side. This line was neither a joint in the pavement nor in the line with
the joint of the casing block. The measurements from the two northern
corners to this line are as follows:-
to N.E. corner=1l5'161 metres.
to N.W. corner=1l5090 "

Diff.= 71 millimetres.
Thus this Fne is probably the original line of the aXIS.

MAXIMUM ERRORS TO BE EXPECTED IN THE NEW

DETERMINATION.

NORTH SIDE.-This is the extension of a sharp and clearly defined line

of 55 metres in length. The maximum error at either end should not exceed
6 millimetres.

EAST SIDE.-This is the extension of 52 metres of fairly clear line and

is not likely to be in error by more than 6 millimetres at either end.

SOUTH SIDE.-This is the extension of 15 metres of the top edge of the

casing blocks. The 15 metres chosen are unworn and give a very definite
line to work from. The final extension should be correct to 1 centimetre
on the West end and 3 centimetres on the East end, to which must be added
any error there may be in the computed horizontal distance between the top
and the bottom edges of the casing blocks.

Petrie op. tit. p. 39.

 -9-

~w
\..r
azimuths WEST SIDE.-This is the extension of points in the pavement 28 metres
rom the ex- apart. The pavement is badly worn in most places, as are also the casing
;hetical base blocks which are still in place. The points chosen are however fairly definite
points of the and there is little doubt that they are correct. The final errors at either end
he diagonals should be less than 3 centimetres.
It should be clearly understood that these possible errors are due to
he most dis- uncertainty of the absolute position of the line on the pavement since a difference
other three. of 1 millimetre only at either end of the line on the excavation will make
Ie 23 metres a difierence of 5 millimetres or more at each end of the extended side, to which
s found, and must be added the possibility of slight deviation of the original side from the
le was found straight line. Checking along the 55 metres on the northern edge has shown
the accuracy this is likely to be inappreciable.
The position of the points of intersection of the adjacent sides was also
,ich projected found on the corner sockets, and measurements were taken from these points
middle of the to the outside edges of the socket where they existed.
the line with The South Western corner socket was too broken to give any information.
jWO northern For the other three sockets, the-follo~g measurements were found:-

Socket. Edge. Distance.

ems.

'-', N.E. ~
N
E
S
85
E5
83
S.E. 1 E 85
) N 75
NEW N.W. W 76

y defined line From these measurements we see that the corners of the Pyramid ob-
Jd not exceed tained from the extension of the sides actually fall on the diagonals obtained
from the socket corners where these can be determined, thus giving additional
confirmations that the size and orientation of each side as found above is
~learline and very close to the truth.
her end.

,p edge of the
very definite
1 centimetre
(lust be added
tween the top
 ('
..
SURVEY of the GRE.AT PYRAMID
(' r
GIZA
6
A

R.O~l~h, - ~'utc'h. ~ Stttiol\.S

\'no Va. Tio\.l~ exca.'Uo..tions.
5A K \.~O. 1.~3 ._--~
fl =:::;:~~==~-------/
5.5 I

.B
:A

A 1 1\ 8foO c

1:1 \. ~~
q: C '"11:;-

I
K.
4 4'; J
L::.8'
~ \1: 1
D ". FfoG H lfoJ

....
~
0
....
""
...,.t
/ . . ~
IJ J J.~
H G r E L t E tllq ~velt\(nt
r~ .i":.:.::.:.. . ;:..:-
..
, '.,
L!
'\J 0. 1 is toto.t\~ clutrojtd.
30. '+~* .- '''if'' '

A 3
6-
2 Scolk.. J: :l) 000
S.oFE:.25!344.
 ~ ~1L.5 of..j~ Gn.evf; 5!Y~or~:basu:iu~ & SlJrvqfT ~
!..~aGI~nu.'n-<;;5 aFfPl.Crr..v and- I:Iw .t:Jypt-i.aut- Go~-{;- ..ru..l'V~ oF 1 &f 2." 7Itu<.; ar(., ~
mnuW4:.c;l'In.6 II..a/: ha-vu dJu..'1t/ ift.,,-t/w;.s ~c.n:pt;-" ~'<tr"-1/~':I!JJ

"~
.'\'"

)
.'
~.

T~ . : 'IN,8'YPUlVr/ (;-0".' .J'urv~ Ytuiaum/

, 'P6= '!k. ~As&nf/'Ix"$'C,,~ . =7 f 5.' IJ 8 z fl!~rut ... -..... 75'5.'8 7 ~ !IJi(, oti~= +D:004
IN= II ~" If "':7SS!77Z, " ... 55.'766 " =+0.'006
nua.n- =
l! QJt.C/,.w ... 7 ~ . ~ 21 h 1I'U(1Ut; 75 S 8 2'2 " +0.005 =
~ 'naHtu~:O/JI4~sid... =1 ! 5:428 ~re.c/; .. 7 ~f.~2.s h +0. D~3 =
" ~~ " ",..; 56'o1~ ".......... 6,'084- 1/ -:: -O.'O/'J.

:Jht. ratio oF U~1t,

duo /lb.,,~ S~,;
/DOo.OO: /46bb+'
fh, SdcalU.,,~: S/j/~

OA, & luiikt ~ J./u. 'P~r~ IS ~uccJ, to 4-~ InUJUV pow eur.d. ~t:aL su0N4 ~~.
",,~~, {'f).,n.ctlll'l'}' . . . . . . . . . . " . . . . . " . . . : 481. /0'0.
, ,7k. 'fJU',:",dr.r elf FGIII , ~k, JJtU tIf flu, f.Yl"t#mid = J 0 s~s ~mv./eaI:1"./1UiU, cJ IJ.t, ~ : ') 0 '2 J:, f +
?Ai IfVA,n.., ",~~ t:!U'UJ/,14I~ tbase S'i~ -= 440 E8Ypbivt. G~c.u.hi~,~r;;tr'o:o,,:z-"" ":"7S S.f1'ZL
,~Cl is tl.u..k,tnU. fut#/r.-,f.":n/, eI'P&~~ sUJU", '7&:DOO tvUJi,6,sl/u. Ut.4Jra.( ltuiJrfM'",.,.-
ff'-iet,,.,/, , -#':000, c 1Iw ~ slan.t 1u4,1t.t, 6 1"2. ~ 0 66, t%nD/, d,S- J'k mt.alf., cF lire.- 4X.St and IrG6't' "f.,u"t
~, 6".'7 r6, Hu,n; -b'a-:: e- QJUJ/ ~'&:=.. ,~"'r(" iJ J.s a mu:ur.- 'Proporiion.od~ "" MrJl, C/
.llfIJIiF IlL is Z. /N,~n Qr/S 01 I,ypubc:la- aNi 6'h.'k CDItIuta.l:c, 1Uis-1Jwc,c. is 1'7.-& IctJiu ~ ~ ~,~ /At, ~