As a result of the work of archaeologists such as Theodor Wiegand, Martin Schede, Ernst Buschor, and Hans Schleif, we have a clear picture of the development of this site from the early archaic period to early Christian times.
The great temple was built and rebuilt before the sixth century B.C. To the east of the temple lay the great altar of Hera, which was rebuilt seven times at seven different periods. In the late seventh century there were also several treasuries, a stoa, and some other buildings. Figure 68 is based on Schleif’s reconstruction of the site during this period.
About 550 B.C., the architects Rhoikos and Theodoros demolished nearly all the earlier buildings and erected the first great temple of Hera with its accompanying great altar (Fig. 70). A small building (G on the plan) also dates from this period.
Only a few years later, this temple was burned down, and the Samians, probably still under the rule of Polykrates, planned and began construction of another great temple, although it was never fully completed. As long as this great temple of Hera remained in use, the appearance of the forecourt remained unchanged, except that more votive offerings accumulated around it.1 Figure 71 shows the site as it was from the fifth to the second century B.C.
By the Augustan era the temple had lost its significance and was used as a storehouse for votive offerings. Gradually, the site became completely altered through the erection of many small buildings, until by the early Christian period it was a mass of small structures. Figure 74 shows the area surrounding the great altar in the first century A.D.2
I divided the development of the site into four phases:
The period just before the Rhoikos temple (Fig. 68). I have assumed this to be representative of the layout throughout the early periods, as there seems to have been only a slight difference between them.
The Rhoikos period, mid-sixth century B.C. (Fig. 70).
The period of the new great temple, from the end of the sixth century B.C. until the second century B.C. (Fig. 71).
The early Roman period, first century A.D. (Fig. 74).
No entrance has been definitely established on the site at any of these four periods. It seems that for a short time only the building southeast of the temple served as a propylon, but this was demolished before the great altar was built over it in the time of Rhoikos. It therefore cannot be considered as acting as an entrance in later periods. In endeavoring to discover from the important sightlines where an entrance might have been, I was obliged to concentrate upon the third and fourth periods, as too few buildings remained from the earlier periods to allow a test of the system to be made. As a result of my calculations, I found that it was possible to establish a point A between the two side walls (antae) of a small building, not fully excavated and thus not fully described, which, according to a verbal statement by Hans Schleif, might well have been a propylon. I have therefore postulated that a propylon or at least an entrance existed on this site, and I have investigated all four site plans to determine whether or not an entrance could have existed at this point.3
Organization of the Site. This is the period of temple II, altar VII, and the southern temenos (the great southwestern building). The layout had evolved gradually from the geometric period until the sixth century (Fig. 68). Although our supposition that point A represents the entrance is not contradicted (by any features of the layout, on the other hand, there is no strong evidence to support it, since no explicit use is made of specific angles and distances.
If we take the southeast building to represent the propylon at this time, as do the excavators, we find that no aspects of the system apply. We must therefore conclude that this was not the entrance, or that the system was not known in Samos, or that this entrance, which was the last building to be erected, could not be placed in accordance with the system.
Organization of the Site. Point A lies in the center of the edge of the top step of building I, assumed to be the propylon (Fig. 69).
a to right corner of the small building to the east; left (northeast) corner of the great altar
b to left (northeast) corner of the temple of Hera; right (southwest) corner of the great altar (B on the plan)
c to right (northwest) corner of the temple of Hera
c′ to extension of line c to the east.
Angle c′a = 36° = 180°/5.
Angle ab = 36° = 180°/5.
Angle bc = 108° = 3 × 180°/5.
The distance from point A along line b to the southwest corner of the great altar (B on plan) = 69.80 m.
The distance from point A to the nearest (northwest) corner of the great altar (C on the plan) = 43.13 m.
Hence or the golden section.
Also, AB = 69.80 = 200 × 0.349 or 200 Ionic feet.4
The hypothesis that point A represents an entrance, planned at the same time as the Rhoikos layout of the site, is supported by the following factors:
the angles within which the main buildings of the site are seen from point A are 180°/5, 180°/5, 3 × 180°/5;
the relation of the distances from point A to the great altar (ie., the golden section) and their unit of measure (ie., the Ionic foot);
the prolongation of line c eastward lies along the front line of the assumed propylon, thus perpendicular to its axis.
Organization of the Site. The layout was altered by several new constructions: the great new temple, a small structure surrounded by a single row of columns (the monopteros), and a row of votive offerings, which were being erected when the great temple was being built.
The assumed entrance has been maintained at point A.
a to right corner of the small building to the east (G on the plan) left (northeast) corner of the great altar (F on the plan)
b to right (southwest) corner of the great altar (B on the plan)
d to left (southeast) corner of the “monopteros” (F″ on the plan); right (western) end of the row of votive offerings (P″ on the plan)
e to right (northwest) corner of the “monopteros”; left (southeast) corner of the great temple (D on the plan).
Angle ab remains the same = 36° = 180°/5.
Angle bd = 37° = ca. 36° = 180°/5.
The distance from point A along line e to point D′ at the southeast corner of the temple = 78.50 m.
If we describe an arc with radius AD, we find it touches the western end of the row of votive offerings (D″ on the plan) and the southeast corner of the great altar (D on the plan).
According to measurements taken on the plans, AD′ = 78.50 m; AD″ = 78.50 m; AD = 78.00 m.
The distance from point A along line a to point F at the northeast corner of the great altar = 56.00 m.
If we describe an arc with radius AF, we find it approximately meets the southeast corner of the monopteros at F′ along line d. According to measurements taken on the plans, AF′ = 57.00 m.
From point A the monopteros is linked optically with the temple.
Although line cc′ now meets the temple at the corner of its second row of columns, our assumption that point A represents the entrance is not necessarily invalidated, since the perpendicular of the temple exactly bisects the angle bd, and the line of votive offerings starting from point D″ runs parallel to cc′. It may be added, without attaching much importance to the observation, that the temple’s northern line of columns was never actually built.
Organization of the Site. Considerable changes were made in the early Roman period. As has already been noted at other sites, the Romans ignored the earlier organization of the layout and concentrated on the use of the right-angled triangle.
The site now contains a wide central pathway (attested by a broad strip of land devoid of buildings, with a water channel running through its center) and short paths that meet it at right angles. This new Roman route starts from point A, which we have taken to represent the entrance in earlier periods. That point A now marks the start of an important north-south axis greatly strengthens our case, especially since the route runs parallel to the façades of several existing structures such as the temple and the altar. The organization of the site is no longer based on lines of sight, angles of vision, or relations between distances but is entirely determined by right-angled axes.
1 I have studied this site only from the plans and reports prepared by the excavators.
Buschor, Ernst. “Heraion von Samos: frühe Bauten.” Deutsches Archäologisches Institut. Mitteilungen. Athenische Abteilung 55, 1930, pp. 1–99.
———, and Schleif, Hans. “Heraion von Samos: der Altarplatz der Frühzeit.” AthMitt. 55, 1930, pp. 146–173.
Schede, Martin. “Zweiter vorläufiger Bericht über die Ausgrabungen auf Samos.” Preussische Akademie der Wissenschaften. Abhandlungen. Philosophische-Historische Klasse, 1929, no. 3, pp. 1–26.
Schleif, Hans. “Der grosse Altar der Hera von Samos.” AthMitt. 58, 1933, pp. 174–210.
Schleif, Hans. “Heraion von Samos: das Vorgelände des Tempels.” AthMitt. 58, 1933, pp. 211–247.
Wiegand, Theodor. “Erster vorläufiger Bericht über die Ausgrabungen in Samos.” AbhPreuss., 1911, no. 5, pp. 1–24.
Buschor, Ernst. “Imbraso.” Deutsches Archäologisches Institut. Mitteilungen. Athenische Abteilung 68, 1953, pp. 1–10.
———, and Ziegenaus, Oscar. “Heraion 1959.” AthMitt. 72, 1957, pp. 52–64.
Johannes, Heinz. “Die Säulenbasen vom Heratempel des Rhoikos.” AthMitt. 62, 1937, pp. 13–37.
Ohly, Dieter. “Die Göttin und ihre Basis.” AthMitt. 68, 1953, pp. 25–50.
Reuther, Oscar. Der Heratempel von Samos: der Bau seit der Zeit des Polykrates. Berlin: Mann, 1957.
Walter, Hans. Das griechische Heiligtum: Heraion von Samos. Munich: Piper, 1965.
Wrede, Walther. “Vorgeschichtliches in der Stadt Samos: Fundtatsachen.” AthMitt. 60–61, 1935–1936, pp. 112–124.
Ziegenaus, Oscar. “Der Südbau: Ergänzende Untersuchungen.” AthMitt. 72, 1957, pp. 65–76.
———. “Die Tempelgruppe im Norden des Altarplatzes.” AthMitt. 72, 1957, pp. 87–151.
This site bears evidence of building activity at several different periods.5 Although the only remains extant from before the third century B.C. are those of the first altar, built between 350 and 330 B.C., there is ample evidence of extensive construction during the first golden age of the island of Cos, between 300 and 205 B.C.,6 and in later centuries.
Structures dating from the first active period include the lower terrace (terrace I) with its stoas and chambers (300–250 B.C.) and the earliest buildings on the middle terrace (terrace II), such as building £ in Figure 77 and the buildings shown as C and D. Temple B and the exedra on terrace II were built a little later, the former dating probably from about 280 B.C.
It seems likely that the magnificent development of the site was begun soon after 190 B.C. following the island’s great victories in 197 and 190 B.C., although evidence from the structures shows that the project was not completed until about 160 or 150 B.C.7
The late Hellenistic altar on the middle terrace (K on the plan) was built later than temple B and the exedra, between 160 and 150 B.C. Temple C and building D were built in Roman times, the former dating from late in the Antonine period (the second half of the second century B.C.)
Organization of the Site. This northern and lowest terrace has a central entrance from the north. Point F lies midway between two columns of the stoa on the axis of the outer stairway.
a to right (southwest) corner of the eastern stoa on terrace I
b to left (northeast) corner of temple C on terrace II
c to right corner of the exedra on terrace II
d to right corner of the steps leading up to terrace II; middle corner (northeast) of temple B on terrace II
e to right (northwest) corner of temple B on terrace II; right (northwest) comer of the western stoa on terrace III
f to left (southeast) corner of the western stoa on terrace I.
Angle ab = 34° = ca. 36° = 180°/5.
Angle bd = 36° = 180°/5.
Angle df = 36° = 180°/5.
Hence angle af = 34° + 36° + 36° = 106° = ca. 108° = 3 × 180°/5.
The distance from F to either end of the northern stoa (G and G″ on the plan) = 47.50 m. If a semicircle is described with radius FG, it touches the supporting wall of terrace II at point G′. The ground plan of terrace I thus has a ratio of 1:2, although it is not a mathematical rectangle.
Organization of the Site. Although the middle terrace had three distinct forms in three consecutive periods, I have been obliged to base my investigations on the form of the last period, as there was not sufficiently precise information concerning the earlier ones.
From 300 to 250 B.C. the terrace contained the first altar (K on the plan, Fig. 77), building E, earlier buildings below C and D, temple B, and the exedra. It is not possible to show whether or not these were disposed according to a prearranged plan.
From about 150 B.C. until the first century A.D. the terrace contained the later altar, all the former buildings, and the steps leading to the upper terrace (terrace III).
At the end of the second century A.D. building E, temple B, the exedra, and the later altar (K on the plan) remained from earlier periods, and temple C and building D were added.
It can be recognized that the main entrance would lie on the axis at the head of the central flight of steps leading up from the lower terrace (terrace I), but, as the upper part of these steps are missing, I have had to assume a probable location for this point: H on the plan.
g to left (northeast) corner of the building group E on the plan
h to right (southwest) corner of building E; left (northeast) corner of temple C
i to left (northeast) corner of the altar (K on the plan); right (southwest) corner of temple C; left corner of the exedra
j to left (northeast) corner of temple A on terrace III
k to left (southeast) corner of temple B on terrace II
l to right (northwest) corner of temple B on terrace II.
Angle lk = 36° = 180°/5.
Angle kj = 36° = 180°/5.
Angle jg = 72° = 4 × 180°/10.
The entire layout as seen from point H thus falls within an angle of 144° = 36° + 36° + 72° = 4 × 180°/5.
The distance from H along line i to the northeast corner of the altar (K on the plan) = 10.50 m.
The distance from H to the nearest (northeast) corner of temple B (L′ on the plan) = 17.00 m.
If an arc is described with radius HL′, it touches the nearest (northwest) corner of temple C at L″.
It is found that HK/HL = 10.5/17 = 0.618/1 = golden section.
Organization of the Site. It is known that the upper terrace was created between 160 and 150 B.C. The entrance was clearly on the steps leading from terrace II; its exact position has not been definitely established, but it must have lain on the axis of the steps and the temple, approximately at point I.
n to left (northeast) corner of the temple; left inner corner of the upper terrace (N′ on the plan, Fig. 82)
o to right (northwest) corner of the temple; right inner corner of the upper terrace (N″ on the plan).
The angle between line n and the axis of the temple (IM) = 36° = 180°/5.
The angle between line o and the axis of the temple (IM) = 36° = 180°/5.
The distances from point I along lines n and o to point N′ and N″ equal 69.80 m.
The distance from point I along the axis of the temple to its far end (M on the plan) = 43.13 m.
Hence IM/IN = 43.13/69.80 = 0.618/1 = golden section.
The basic unit of the length of 69.80 m is the Ionian foot of 34.9 cm.
Hence IN = 69.80 m = 200 × 34.9 cm = 200 Ionian feet.8
Further, if the triangle IMM′ is turned on its side IM′, we find that point M′″ of the new triangle IM′M′″ determines the inner line of the eastern stoa. The inner line of the western stoa can be determined in the same way. The entire upper terrace is thus based on the 36° angle (180°/5) or on the ratio resulting from it, the golden section.
*****4 [Although these steps have since been restored, it is not certain that the restoration is correct.]
Herzog, Rudolf, ed. Kos: Ergebnisse der deutschen Ausgrabungen und Forschungen, vol. I. Berlin: Keller, 1932; in particular, Paul Schazmann, “Die bauliche Entwicklung des Asklepieions.”
The position from which I studied the layout of the site is to the west of the altar in the center of the agora at Priene. From this point, the temple of Athena appears exactly between two sight lines, a and b (Figs. 84, 86), that just touch the corners of the western and northern stoas and form an angle of 17° (18° = 180°/10). The temple stands on a higher level and, seen from here, forms a link between the ridge of the steep rocks to the right and the low buildings to the left (Fig. 85).
Although the entire agora was not built at the same time as the temple of Athena, “it is obvious that the agora was planned for in the layout of the city, which goes back to the fourth century B.C., and the leveling operations and retaining walls may well date back to that period, as well perhaps as the south stoa … and parts of the west and east stoas.”9 The temple of Athena was built in 344 B.C. Shortly after 150 B.C. a new stoa was built on the upper level, just south of the temple, H in Figure 84, shutting off the view of the temple from the central altar in the agora.
The great altar of Athena was erected to the east of the temple in 150 B.C., just to the right of sight line b (Fig. 84), perhaps to avoid blocking the view of the temple from the agora. Yet this choice of site would have been meaningless if (as indicated in Fig. 85) the terrace to the south of the temple, with its retaining wall, were at the same level as the floor of the later stoa, for then the great altar would be completely concealed from view. This means that, if sight line b were in fact taken into consideration, either the level of the original southern terrace was much lower, or its retaining wall was gradually stepped back, so that the altar could be visible from the agora.10 As the propylon to the sacred precinct of Athena was built much later than the southern stoa, it is of no interest to this investigation.
The northern stoa on the agora, which is of importance to this study, was not built until 150 B.C., but it replaced a former shorter stoa, equal in length to the parallel southern stoa, which extended from the steps leading up to the temple of Athena to the steep path leading up to the theater.11 The colonnade of this earlier stoa may lie below the Ionic columns of its successor, but it seems more likely that both had the same depth. Since the space of the earlier stoa was divided into two,12 it was perhaps similar to the central part of the southern stoa. In this case the southwestern corner of the earlier stoa would have coincided with the same corner of the later one. This appears to be self-evident, as from the very first there must have been a desire to achieve a view of the temple of Athena framed by sight lines a and b. It is impossible to believe that this occurred as a happy accident after the erection of the temple and the western stoa.
If these observations are correct, they explain why there was a gateway at the eastern entrance to the agora but not at the western entry: the latter was left open so that there would be no obstruction to the view of the temple.
This entire precinct was built to the east of the agora during the third century B.C. as a single project. The site has not been fully investigated, as a Byzantine fort was built over one of its corners, but the published plans permit us to make certain observations. Wiegand had attributed the temple to Asclepios, but Schede found that it was built to honor the Olympian Zeus.13
Organization of the Site. Entry was only through a gateway on the east. Point A lies in the center of this opening (Fig. 93).
a to left corner of the great altar; left (southwest) corner of the precinct (B on the plan)
b to left (southeast) corner of the temple
c to right (northeast) corner of the temple; perhaps the whole northern side of the temple
d to left corner of the pedestal (S on the plan); right (northwest) corner of the precinct.
Angle bc = 18° = 180°/10.
Angle ad = 36° = 180°/5.
The elevation of the temple as seen from point A thus occupies half the angle of vision of the entire precinct, although it is not placed symmetrically in the center of its western side.
The distance from point A along line a to point B on the plan = 35.40 m = x.
The distance from point A along line c to the façade of the temple (right corner of the upper step) = 21.90 m = y.
The distance from A to the northwest corner of the great altar = 13.40 m = z.
From this we see that
x/y = 35.40/21.90 × 1/0.618 =
y/z = 21.90/13.40 = 1/0.612 = 1/0.618 =
The greatest distance AB, which measures 35.40 m, may also equal 100 Ionian feet, since the accepted measurement (0.349 × 100 = 34.90 m) tended to vary slightly from place to place.14
The observer is presented with a completely enclosed view, dominated by the angles of 18° = 180°/10 and 36 = 180°/5 and the proportion of the golden section which derives from them. In other words, this is a space determined by the tenfold division of the total field of 360°.
Schede, Martin. Die Ruinen von Priene. Berlin: De Gruyter, 1934.
Wiegand, Theodor, and Schrader, Hans, eds. Priene: Ergebnisse der Ausgrabungen und Untersuchungen in den Jahren 1896–1898. Berlin: Reimer, 1904.
Meyer, Bruno. “Das Propylon des sogenannten Asklepieions in Priene.” Deutsches Archäologisches Institut. Jahrbuch 49, 1934.
Schede, Martin. Die Ruinen von Priene. 2nd ed. Berlin: De Gruyter, 1964.
A sacred precinct of Artemis Leukophrene existed on this site from very early times, but the first building we know of was an archaic limestone temple that endured until the second century B.C. The precinct acquired its final form at the time of the rebuilding of the city in 400 B.C. At the end of the second century B.C., the famous architect Hermogenes constructed a new temple of Artemis, in which he incorporated for the first time his new architectural concepts. The altar of Artemis, in front of the temple, can also be attributed to Hermogenes.15 Although the other structures cannot be dated accurately, it can be accepted that Hermogenes reorganized the area at the same time as he rebuilt the temple. We can therefore consider the entire precinct as a unified layout.
Organization of the Site. There is a single entrance through the propylon from the agora. Point A in Figure 95 represents the center of the inner edge of the propylon as determined by Humann’s excavations.
a to left corner of altar of Artemis; left (northwest) corner of temple of Artemis (lowest step); left corner of the raised level for statuary, before the stoas
b to left (northwest) corner of the temple platform (top step, B on the plan)
c to right (southwest) corner of the temple platform (B′ on the plan)
d to right corner of the altar of Artemis; right corner of the lowest step of the temple; right corner of the raised level.
Angle be = 18° = 180°/10.
By turning the isosceles triangle ABB′ on both AB and AB′, we find that points B″ and B′″ determine the distance of the north and south stoas from the temple.
From point A to the temple façade (AB and AB′) = 104.80 m = 300 Ionic feet16 (34.94 × 300 = 104.80 m).
From point A to the front of the altar of Artemis = AC = 64.70 m.
Hence AC/AB = 64.70/104.80 = 0.6173 = the golden section (0.618).
The temple is placed on the center axis of the precinct. Standing at point A, the observer is very conscious of this symmetry. The view is entirely enclosed.
The layout is based on the angle of 18° (180°/10) and the golden section. The entire space of 360° is thus divided into ten parts.
Although the agora of Magnesia formed part of the new city plan, prepared about 400 B.C., the small temple that stands within it, the temple of Zeus Sosipolis (Saviour of the City), was not built until the beginning of the second century B.C., possibly after Magnesia’s great victory over Miletus. The other buildings, with the exception of a Roman statue to the south of the temple, seem to date also from the second century B.C., and were perhaps built by the architect Hermogenes. The organization of space in the agora can therefore be traced back only to the early second century B.C., and even this cannot be considered definite, as the site has not been completely excavated.17 It is possible that later findings will give rise to new points of view concerning the layout.
Organization of the Site. The building of the temple of Zeus in the second century B.C. had an important influence on the layout. There are three entrances (Fig. 101). Points A and B are placed in the center of the two entrances east and west of the southern stoa. At a later date two propylaea were built over them. Point C is placed at the access through the southern stoa from the sacred precinct (H in Fig. 96).
The entrances from two roads that lead into the southern stoa from the south, apparently had no influence on the layout of the agora.
This seems to be supported by the fact that no attention is paid to the position of these roads in the design of the stoa.
b to left corner of the altar before the temple of Zeus; right corner of the propylon of the sacred precinct of Artemis
c to left corner of the structure in the southwest of the agora; right corner of the altar of Zeus, left (northwest) corner of the temple of Zeus
d to center corner of the southwest structure; right (southeast) corner of the temple of Zeus
e to right corner of the southwest structure.
The spectator has an entirely enclosed field of vision in which he perceives each structure in succession, each a complete entity. From left to right (Fig. 102) he sees, without any gaps between them, the propylon of the sacred precinct of Artemis, the altar of Zeus, and the temple of Zeus, with the lower structure containing stone benches in front of it. The position of the temple of Zeus must have been calculated to conceal the larger temple of Artemis, outside the agora (Fig. 95), and thus prevent the competition, in the eyes of the observer, of two equally large volumes.
The position of the southwest structure seems determined by a desire to interrupt the direct view of the Zeus temple, which would otherwise be very dominant, and lead the eye to the path to the altar and, beyond it, to the propylon of the sacred precinct of Artemis.
b to left (southwest) corner of the temple of Zeus; right corner of the altar of Zeus
c to center (southeast) corner of the temple of Zeus
d to right (northeast) corner of the temple of Zeus; left corner of the exedra opposite the propylon to the precinct of Artemis
e to right corner of the exedra.
b to left side corner of the sitting place; right corner of the exedra
c to right corner of the sitting place; left corner of the tall stele
d to right corner of the stele; left corner of the altar of Zeus
e to right corner of the altar of Zeus; left corner of the propylon to the sacred precinct of Artemis
f to left (northwest) corner of the temple of Zeus; right corner of the propylon to the sacred precinct of Artemis
g to right (southeast) corner of the temple of Zeus.
Although the space is entirely enclosed, the route to the propylon of the sacred precinct of Artemis is kept entirely clear from each vantage point, with the other structures in the agora ranged on each side of it.
Gerkan, Armin von. Der Altar des Artemis-tempels in Magnesia am Mäander. Berlin: Schoetz, 1929.
Humann, Carl, ed. Magnesia am Maeander: Bericht über die Ergebnisse der Ausgrabungen der Jahre 1891–1893. Berlin: Reimer, 1904; in particular, Julius Kohte, “Die Bauwerke,” pp. 9–172.
This precinct was founded in the Roman period, in the first century A.D.18 It has an axial entrance at A (Fig. 106) and two symmetrical side entrances at M and N. Although the two side entrances seem to have been left open, it was not possible to look from them into the interior of the precinct. Similarly, from inside the precinct, it was not possible to see out through these openings. This is effected by the unusual size and form of the comer columns, which interrupt the view to and from the side openings (see sight lines a, b, a′, and V in Fig. 106). In this way the architect’s desire to create a visually enclosed space was maintained.
Organization of the Site. Point A is located on the axis of the central entrance on a line connecting the centers of two columns of the stoa. There is no step.
c and c′ to left and right corners of the temple platform (E and E′ on the plan, Fig. 106); left and right far corners of the precinct
d and d′ to left and right corners of the beginning of the temple superstructure, i.e., the bases of the nearest columns (B and B′ on the plan).
Angle dd′ = 36° = 180°/5.
The distances from point A along lines d and d′ to points B and B′ are identical and = 25.5 m.
The distance from point A to the temple steps (point C″) = 15.75 m.
Hence AC″/AB = 15.75/25.50 = 0.618/1 = golden section.
We also find that the distance AC″ is determined by the base of the isosceles triangle ABB′:
BB′ = AC″ = 15.75 m.
The ground plan of the temple platform
The position of the temple is therefore determined by the isosceles triangle ABB′ and governed by the 36° angle = 180°/5 or the division of the whole field of 360° into ten parts.
Michalowski, Kazimierz. Palmyre: fouilles polonaises, 1959. Warsaw, 1960. (This covers only the Diocletian area of Palmyra.)
Schlumberger, Daniel. La Palmyrène du nord-ouest. Paris, 1951. (Thesis, University of Paris.)