The Mercury diagram is probably the best example of an ancient “figure-plan” used for three centuries to organize the sacred spaces of Angkor Proper.
The ecliptic longitude of Mercury on 22 / 04 / 967, probably determined at the beginning  of Angkorian history (9th century) as explained elsewhere, was “engraved” in the Khmer soil by linking the highest neighbouring sacred hill (Phnom Bok) to a particular location used as the observer of the inner planets.

Phnom Bakheng : the prime meridian of Angkor

The observer location was determined by the intersection of the “Prime Meridian” of Angkor, that of Phnom Bakheng (another sacred hill near the centre of Angkor ) with a solar apsides line (22/04) drawn from the centre of the  Bakong  temple, the principal monument of the first Khmer capital in the Angkor region (near today’s Roluos).
Moreover, this observer (called “ancient observer of Angkor Thom”) was located at distances from the sacred hills (Phnom Bakheng and Phnom Dei) and from the Bakong temple which are integers of  two Khmer units:
- the krta yuga (abbreviated here as KY) defined by Mannikka (1996) as equal to 752.4576 m
-  the Khmer arcminute (abbreviated here as kam) defined by the authors as equal to 1881.144 m
Although the centre of the Mercury “Midnight” diagram was the observer, the key location of the diagram was the central tower of the Takeo temple. We’ll see that several dimensions of Takeo were  determined by the Midnight parameters of Mercury.  When we fix the centre of the Manda epicycle of Mercury on the tower, the centre of the Sighra epicycle lies (in accordance with the rules of epicycle theory) on the parallel of the observer, which is by convention, also the parallel of the Sun.
The radius of the Manda epicycle determined the size of the external enclosure of Takeo: the former was equal to a quarter of the latter’s perimeter.
The Mercury diagram was later corroborated by a number of structures constructed by the Khmer Kings during the 11th and 12th centuries :
- During the 11th century, the Phimeanakas temple was built on the geographical parallel tangent to the northern limit of the Manda epicycle (previous diagram).
- Early in the 12th century, the king Suryavarman II built the magnificent temple of Angkor Wat and used the north-west corner of its terrace as the observer location of a Moon-Sun diagram. During the same period, he built the small twin temples of Thomannon and Chau Say Tevoda, one on either side of the east-west axis of the Victory Gate of Angkor Thom; (although this gate, in its present state, was erected later it was probably built on the location of a more ancient structure).

The original photo (without diagram) is in the public domain




The original photo (without diagram) is in the public domain  because it was created by NASA


The original photo (without arrow and labels) is in the public domain


        MANDA epicycle


    Coordinates of centre:




    Radius = 0.11057 km

                 = 0.14694 KY

    Azimuth =183.273°

Radius = quarter of  perimeter of  takeo’s external  enclosure.

(Glaize: The external enclosure wall forms a rectangle of 120 metres by 100)




Takeo temple     This file is licensed under Creative Commons Attribution 2.0 License
Full resolution (2,272 × 1,704 pixels, file size: 1.28 MB, MIME type: image/jpeg)




Centre of Manda epicycle on central tower

- During the second half of the 12th century, the king Jaryavarman VII constructed the Krol Ko temple, near the northern bank of his North Baray, on the continuation of the loxodrome (line of constant bearing) from the Angkor Wat terrace corner to Takeo.    
We have discovered that the centre of the Takeo lies at the vertex of a right angle whose sides are formed from lines running from the north-west corner of the naga terrace of Angkor Wat and the centre of the Thommanon temple to the Takeo.
AAt first sight, this plan might be mistaken for a non-astronomical construction, but we have determined that there is an underlying Mercury “Midnight” ratio: the parallel of the Victory Gate determines the Mercury Sighra epicycle radius to deferent radius ratio, expressed as the separation distances between the parallels of the twin temples.
The same king also built the Ta Prohm temple to the south of the older (circa 900 AD) East Baray. Now, the loxodrome from Krol Ko to Ta Prohm is precisely tangent to the Sighra epicycle of Mercury.

             Ta Prohm           Courtesy of Moyra Farrington

The loxodrome running from the centre of Ta Prohm to Mercury (-91.545°) is perpendicular to the loxodrome running through the observer and the centre of the Sun’s epicycle (-1.571°) and probably parallel to the temple's axis. More simply, the loxodrome from Ta Prohm to Mercury is perpendicular to the main axis of the East Baray ( between -1.5° and -1.6°)

- Jaryavarman VII (or one of his predecessors) built the eastern enclosure of his city (Angkor Thom) midway between the western bank of the East Baray (more precisely its western shoreline) and the meridian of another huge hydraulic construction, the Great North Channel.
The length of each side of the Takeo’s upper platform (47 m) was determined, (circa 1000 AD) by the product of the distance from the platform’s centre  to the western bank of the East Baray (603 m), with the Mercury equant eccentricity.
603 x 2 x 0.0389 = 46.9
Furthermore the  diameter of the Manda epicycle (221.14 m), divided by the distance (603m) = 0.3667.
Diameter of deferent / 603 = Sighra ratio / Manda ratio
Jayavarman VII therefore appears (as late as the 12th century) to have considered the site of the western bank an important location in his city’s planning.

The original photo (without diagram) is in the public domain  because it was created by NASA



Radius of deferent (Rd) = 2.84253 km   

Radius of Manda epicycle(Rm) = 0.11057 km

Rm/Rd =  0.0389  = Midnight eccentricity


Azimuth of radius of Manda epicycle

=183.273° from Sun (East) on 22 / 04 / CE 967  (Julian calendar) at midnight between 22nd and 23rd.

= 220° from Vernal equinox (Midnight apsides line)



Radius of deferent (Rd) = 2.84253 km  

Radius of Sighra epicycle (Rs) = 1.04234 km

Rs/Rd = 0.3667 = Midnight ratio



Last modification : 15 / 12 / 2008



Copyright 2008  Jean-Pierre Lacroix & Robert Bywater




Links :   Sun of Beng Mealea

            Sun of Chau Srei Vibol

             Methods of cartography

We have discovered that the azimuth (i.e. ecliptic longitude) of Mercury from the observer location was determined by a loxodrome running from the observer of Angkor Thom to a particular point located on the Phnom Bok’s crest (error = 0.1°). We’ll see this location,  which is one of the main “crossroads” of the planetary diagrams, was used as the observer of the outer planets of the hinterland. The Mercury loxodrome runs through  a giant Linga erected near to the crest of Phnom Bok. The Linga and its platform were proportioned on the “Midnight” eccentricity for Mercury. The location of this Linga shows the goal of the Khmer was to include natural structures in their diagrams. A circle whose centre is located on the Linga and crossing the centre of the Phnom Dei temple runs indeed  at 31 m from the centre of the Phnom Bakheng temple and precisely through the intersection of the latter’s meridian with the crest (a location labeled “Bakheng summit” here above). The radius of this circle is 14379 m long (22.93 X 360 phyeam).  It is worth noting the Phnom Bok temple was built at 23.03X360 phyeam from the Bakong temple (centre to centre).