Excavations of the sacred precinct of Delphi are still incomplete, and the published reports on them do not present a clear picture of the entire site; there are even differences of opinion concerning the structures that have been fully excavated. I have therefore ventured to draw only one general conclusion regarding the site as a whole, which is supported by the plans that have so far been published. (Rechecking of measurements on the spot would be a tremendously difficult and elaborate operation, as the site is on a steep mountain side.) A visual contrast between the two largest building masses, the temple of Apollo and the theater, was prevented. From both the main entrance to the site and from the southeast entrance, the theater was completely hidden by the temple; and from other entrances, views of it were blocked by retaining walls and similar structures (see Fig. 14).
The temple terrace is undoubtedly the most important feature of the site, but my studies of it are merely preliminary. Scholars have established that it passed through two great building phases. The first layout included the temple said to have been built by the architects Trophonios and Agamedes,1 probably in the first half of the sixth century B.C. (unfortunately there are too few traces of this building to permit examination), and the second comprised the great temple built with the aid of the Alkmaeonid family about 530 B.C.2 The exact points of access to this temple terrace are uncertain, but it seems likely that the first general view of the temple from the sacred way was obtained from point A (Fig. 8), just south of the great altar erected by the Greeks of Chios, even though it is possible that there was no direct access to the temple from this point. The second view was from point B, the only entrance of which we are positive (this was not framed by a gateway). At this point the sacred way turned to enter the temple terrace itself. A door in the western wall of the precinct may have served as a third entrance to the other end of the temple. The great altar was dedicated in 476 B.C.
Several monuments were added in the Hellenistic period, including the double columns (D on the plan);3 a statue of Eumenes II, King of Pergamon, 182 B.C. (E on the plan; and see Fig. 11); a statue of Aemilius Paullus, 168 B.C. (F on the plan; and see Fig. 12).
The precinct developed gradually over the years. The great temple of Apollo was built about 530 B.C.
SIGHT LINES FROM POINT A
Point A lies in the center of the opening just south of the great altar.
a to left (northeast) corner of the temple.
b to left corner of the great altar, right (northwest) corner of temple.
Angle ab = 60° = 180°/3.
SIGHT LINES FROM POINT B
Point B, though more important, can be located with less certainty, as it cannot be placed centrally in a clearly marked entrance.
c to right corner of great altar, left (northeast) corner of the temple
d to right corner of the temple
Angle cd = 60° = 180°/3.
The accuracy of lines c and d is confirmed by the placing of the later Hellenistic monuments D, E, and F, which shows regard for these lines. In fact lines c and d explain the positions of D and F, which (unaccountably, as it may seem to us) bear no relation to the axis of the temple. The sight lines may also account for the position of the column south of the temple terrace.4 All these monuments were placed so that they would form a symmetrical composition with the façade of the temple when seen from one or other of the main viewpoints; no account was taken of their position in relation to the geometrical axis of the temple.
SIGHT LINES FROM POINT C
The position of point C is somewhat uncertain: it lies on the direct axis of the temple and forms an isosceles triangle with the corners of the west façade, whose angles measure 180°/4, 3 × 180°/8, 3 × 180°/8.
Bourguet, Emile. Les Ruines de Delphes. Paris: Fontemoing, 1914.
Courby, Fernand. La Terrasse du temple, pt. 1. Fouilles de Delphes, vol. 2. Paris: De Boccard, 1927.
Pomtow, H. “Delphoi,” pt. 1. In A. F. von Pauly, ed., Real-Encyclopädie der classischen Altertumswissenschaft, suppl. 4, Stuttgart: Metzler, 1924, pp. 1189–1432.
———. “Die Paionios-Nike in Delphi.” Deutsches Archäologisches Institut Jahrbuch 37, 1922.
Schober, Friedrich. “Delphoi,” pt. 2. In A. F. von Pauly, ed., Real-Encyclopädie der classischen Altertumswissenschaft, suppl. 5, 1931, pp. 59–152.
Tournaire, Albert. Relevés et restaurations, pt. 1. Fouilles de Delphes, vol. 2. Paris: Fontemoing, 1902.
Amandry, Pierre. “Chronique des fouilles de 1943 à 1945,” Bulletin de correspondance hellénique 68–69, 1944–1945, pp. 439–441.
———. “Chronique des fouilles en 1947,” BCH 71–72, 1947–1948, pp. 445–452.
———. “Recherches à Delphes, 1938–1953,” International Congress of Classical Studies, 2nd, 1954, Acta Congressus Madvigiani, Copenhagen: Munksgaard, 1957–1958, 1, pp. 325–340.
Daux, Georges. Inscriptions, pt. 3, 2. Fouilles de Delphes, vol. 3, Paris, 1943.
———. “Inscriptions et monuments archaiques de Delphes,” BCH 61, 1937, pp. 67–72.
———. Pausanias à Delphes. Paris: Picard, 1936.
Kähler, Heinz. Der Fries vom Reiterdenkmal des Aemilius Paullus in Delphi. Berlin: Mann, 1965.
La Coste-Messelière, Pierre de, Au Musée de Delphes. Paris: De Boccard, 1936.
———. Delphes. Paris: Editions du Chêne, 1943.
The remains of buildings in the sacred precinct of Aphaia (Fig. 18) show that there were three major periods of construction, but only for the last period is it possible to reconstruct an accurate plan (Fig. 19), as traces of the layout during the two earlier periods—especially indications of the position of the temples—are insufficient.5 The final temple of Aphaia and its contemporary precinct were “built roughly between 500 and 470 B.C., and were probably started nearer to the earlier than the later date.”6 The structures erected at the same time as the temple—they can be recognized, as they follow the same orientation—all form part of a single cohesive plan. They comprise the large altar (Fig. 21), the propylon, three of the terrace walls, and three bases for monuments on the east part of the terrace.7 These three monument bases are shown as S, M, and N in Figure 19. The only monument of an earlier period that might possibly have remained in its original position is an ancient Ionic votive column,8 the site of which is shown as E in Figure 19.
The precinct was laid out between 500 and 470 B.C. Entry was through the propylon, and point A, center of the inner edge of the propylon floor, represents the geometric center of the complex.
SIGHT LINES FROM POINT A
a to left (southwest) corner of temple (D″ on the plan, Fig. 19)
b to middle (southeast) corner of temple (B on the plan)
c to right corner of pedestal S; right (northeast) corner of temple (C on the plan)
d to left corner of monument M
e to left corner of altar steps* (C′ on the plan); right corner of monument M
f to left corner of monument N; right corner of altar steps
g to right corner of monument N.
ANGLES OF VISION FROM POINT A
Angle ac = 60° = 180°/3.
Angle cd = 30° = 180°/6.
Angle dg = 60° = 180°/3.
Angle ac is subdivided as follows: angle ab = 36°, and angle bc = 24°; i.e., angle ac is subdivided in a ratio of 3:2.
DISTANCES FROM POINT A
A is equidistant from the southeast corner of the temple (B) and the southwest corner of the altar base (B′), i.e., 11.30 m.
A is equidistant from the northeast corner of the temple (C) and the northwest corner o the altar (C′), i.e., 25.60 m (possibly 50 Egyptian ells; see Chapter 3, note 4).
A is equidistant from the southwest corner of the temple (D″) and a point (D) on the northern terrace wall, i.e., 38.55 m (possibly 75 Egyptian ells).
The distance from A to the northeast corner of the temple (C) is therefore 2/3 AD, because AC/DD = 25.60/38.55 = 2/3.
In other words, construction of the equilateral triangle AD″D determines the position of the enclosure wall.
From point A, there were two equal fields of vision, to right and left, each 60°, separated by an open view down the center, 30°. The paved path from altar to temple was too low to affect this. The view to the left exactly encompassed the temple within an angle of 60°. To the right, the complex of monument M, the altar, and monument N again fell within a 60° angle. The spectator thus saw two completely symmetrical masses to right and left from a narrow central line of vision, 30°, whose open view followed the direct axis of the propylon.
A mathematical symmetry exists, but it is a visual symmetry, relating to the building masses as seen from this point, not as they actually exist. The most important relationship is the ratio 2:1:2 (building mass, free space, building mass). But the ratio 3:2 also determines the organization of the space, e.g., the relationship of the two sides of the temple and the distances seen from A to its left and right corners. These relationships between angles and sight lines determine the disposition of all buildings and monuments in the precinct; they are based on a division of the space into six parts and follow conditions laid down by the equilateral triangle AD″D. If the Ionic votive column (E) still existed at this time, it too would have been incorporated in the system, as its distance from A (AE) is equal to the height of the triangle AD″D. Line a is oriented to the west, and line d to the north, so that two of the most important cardinal points are stressed by a sharp line between solid and void. The axis of the propylon and the clear field of vision are both directed toward the north (15° west of north).
The ground plan of the temple of Aphaia has a ratio of 1:2; its dimensions (at the lowest step) are 15.50 by 30.30 m.
Cockerell, Charles Robert. The Temples of Jupiter Panhellenius at Aegina and of Apollo Epicurius at Bassae. London: Weale, 1860.
Furtwängler, Adolf. Aegina: das Heiligtum der Aphaia. 2 vols. Munich: Franz, 1906.
Invernizzi, Antonio. I frontoni del tempio di Aphaia ad Egina. Turin: Giappichelli, 1965.
Welter, Gabriel. Aigina. Berlin: Mann, 1938.
Basing my study on plans prepared by the excavators,9 I examined the organization of the layout of the Delphineion at four different periods: Delphineion I, fifth and fourth centuries B.C.; Delphineion II, third and second centuries B.C.; Delphineion III, first century B.C. and first century A.D.; Delphineion IV, after the first century A.D.
Nothing is definitely known of the original form or layout of the Delphineion.10 From the existing ruins it appears that it was first built on this site when Miletus was refounded shortly after 479 B.C. The first Delphineion occupied only about half the area of the later one; it was the size of one of the city blocks and was surrounded by roads (Fig. 23). According to Gerkan, it is not possible to accept that the altar base formed part of the earliest altar, dating from before the Persian invasion, although the existing base does show signs of the altar having been reconstructed.11
In the time of Alexander the Great, the Delphineion was redeveloped, and the precinct was more than doubled in size through the addition of a large horseshoe-shaped stoa. The various monuments shown in Figure 24 date from the early Hellenistic period.
The erection of a western stoa (Fig. 26) marked yet another stage of development.12 Although its exact date is unknown, it could not have been built before the late Hellenistic period. There is no clear evidence as to whether the central tholos was built in Hellenistic or Roman times.13 I have shown it in my plan of the late Hellenistic period (Fig. 26); Gerkan believed this date to be the most likely, although in the official publication of the excavations the question is left open.
The Delphineion underwent a third radical transformation in the early Roman period (at the end of the first or beginning of the second century A.D.), when it was surrounded on all four sides by a reconstructed Roman stoa (Fig. 27).
From the inscriptions on the structures still extant, we realize that the sacred precinct must have been embellished with very many more inscribed monuments. There are now very few remains from which one can gain a notion of the richness of its former state,14 but fortunately enough structures have survived to enable us to make important observations regarding the organization of the layout.
An entrance is presumed in the center of the west wall of the Delphineion (Fig. 23), and point B is located in the middle of its inner face.
a to right corner of the northern stoa
b to left (northern) side of the great central altar
c to right corner of the great central altar
d to left corner of the southern stoa
The angle between the left-hand part of the western wall of the Delphineion and line a = 61° = 180°/3.
Angle ad = 58° = 180°/3.
The angle between line d and the right-hand part of the western wall = 61° = 180°/3.
If a semicircle with radius 15.20 m (x) is described from B, it passes through the following points:
the west corner of the northern stoa
the southeast corner of the great central altar
the outermost of a row of three circular altars
the west corner of the southern stoa.
We also find that the length of lines a and d, extending from B to the farther corners of the northern and southern stoas, is equal and measures 30.80 m (y).
Hence 2x = y.
The distance of 30.80 m may be taken as equivalent to 100 Attic feet, as used in Athens before the Persian invasion.15 The central altar is set slightly diagonally in the space, so that the line of its northern side, if projected, would meet the western wall of the Delphineion at a point slightly south of B. One may perhaps be justified in believing that this was done so that the northeast corner of the altar would be clearly visible from B, for it is this corner that is at the critical distance from B of 15.20 m (x).
In the reconstruction of the Delphineion (Fig. 24), which was begun in 334 B.C., careful respect was paid to all existing monuments. There were three entrances, all in the western wall. Points A, B, and C lie in the middle of each opening, on the line of the inner face of the wall.
a to left tangent of the large west-facing exedra
b to right corner of the small south-facing exedra
d to left corner of the north-facing exedra
e to right tangent of a cylindrical pedestal; right corner of the great central altar; right corner of the inscribed stele
f to left corner of the square pedestal
g to right corner of the same square pedestal; left tangent of the nearest of the row of three circular altars. The field between lines b and d is left open.
b to left corner of the inscribed stele
c to right corner of the south-facing exedra
d to left corner of the large west-facing exedra
e to right corner of the great central altar; left corner of the north-facing exedra
f to right corner of the inscribed stele.
The field between lines c and d is free of structures. All the features noted at Delphineion I, as seen from point B, still exist.
a to left corner of the great central altar
b to right corner of the square pedestal; right corner of the great central altar; right corner of the south-facing exedra
c to left tangent of the most northerly of the three circular altars; left corner of the inscribed stele; left corner of the north-facing exedra
d to right tangent of the most southerly of the three circular altars; left tangent of the circular fountain at the far end of the enclosure.
The field between lines b and c is free of structures.
The diagonals of the enclosure divide the area into two identical right-angled triangles, each with one angle of 60° and the other of 30° (180°/3 and 180°/6). Line b from point C and line c from point A not only run parallel to these diagonals but also determine the two open fields of vision; they are therefore the most important sight lines from each of these two entrances. During the period of Delphineion I the entire space was organized within an angle of 60° from point B. With the enlargement of the precinct this organization was no longer possible from point B, but the space was divided upon the same principle from the new entrances C and A.
During the erection of the south-facing exedra it proved necessary to remove an earlier pedestal (A in Fig. 25). In the account of the excavations the following explanation is given for the selection of this position for the exedra: “The reason for the choice of this particular location can most easily be explained by the care in the placing of the south-facing exedra symmetrically opposite the north-facing one, which is consequently thought to have been constructed rather earlier.”16 But, as the matter of symmetry is not taken into account in the relations between the great central altar, the large west-facing exedra, and the north-facing exedra, it seems unlikely that it would have become so important in determining the position of the south-facing exedra: this can be explained by the importance of the sight line b from the already existing entrance C, which the new exedra would have to respect.
The construction of a stoa along the western wall did not materially alter the organization of the precinct. The entrances remained the same, and the most important lines of vision were carefully preserved. The central tholos (as mentioned earlier, we cannot be certain that it was erected at this time) is encompassed by lines b and c at 60° from entrances C and A, and its presence strengthens the importance of these sightlines. All other features noted at Delphineion II still remain.
In the early Roman period, about 100 A.D., the Delphineion was again reconstructed. In rebuilding the stoas, even greater emphasis was given to the two chief sightlines b and c from the entrances C and A, since the two farthest corners of the enclosure were now sited directly upon them. The entire central tholos could be seen within an angle of 15° from each of these entrances (i.e., 180°/12) between the columns of the western stoa. This leads me to believe that the central tholos belongs to this early Roman layout rather than to the late Hellenistic one, but, as I have said, this cannot be stated definitely. Two other assumptions seem equally probable: first, that the tholos and the new stoas were planned at the same time, but, for unknown reasons, only the tholos was actually built; second, that the stoas were both planned and constructed after the tholos had been built and that their form was derived as a result of consideration for the existing sightlines.
A complete survey of the architectural development of the agora at Miletus would be beyond the scope of this work. Many public buildings were frequently altered and enlarged during the seven centuries of the city’s existence, from the fifth century B.C. to the second century A.D. I have therefore confined my study to the four chief periods examined by the excavators.17 A plan is given for each period (Figs. 31, 32, 35, 36) so that comparison can be made of the layout of the site at each of these four phases of development.
At this period it is possible to reconstruct only the northern part of the agora (Fig. 31). According to the excavators, “The ground plan of the stoa complex indicates that the extension of the later agora was not yet contemplated.”18 We should therefore consider the layout at this period as a complete entity, not as the first stage in a development. No traces remain of definite entrances from the city into this area. The layout of the southern part of the site is uncertain: it probably consisted of a large rectangle within which several small monuments were located, but the exact positions of these are still unknown.
By the middle of the second century B.C. the agora had acquired its final, typically Hellenistic form (Fig. 32). Earlier buildings had been gradually completed, a new southern agora had been built, and the area just south of the northern entrance had been reorganized. As in the first phase, this layout does not show signs of being developed according to a prearranged plan;19 it should be considered as having evolved from the already existing layout and as independent of later development.
It was only at the end of the second century B.C. that an eastern boundary wall was built enclosing the northern agora and cutting the formed space into two parts. This area could be entered from several of the city streets, but the most important entrance was doubtless the one from the southern agora (B on the plan, Fig. 32) as this provided access for almost all the traffic from the southern part of the city (Fig. 33). Two of the other entrances are marked C and D on the plan (Fig. 32).
Point B lies in the center of the opening between two stoas on the line of their northern façades.
a to right corner of monument M on the plan (Fig. 32); left corner of the propylon in the center of the new boundary wall
b to right corner of the stoa facing the harbor; left corner of the city block nearest the harbor
c to left corner of monument M′ on the plan; left corner of the Delphineion
d to left corner of monument M″ on the plan; right corner of the propylon to the eastern building
e to right corner of monument M″; entrance D on the plan.
f to left corner of monument M′ on the plan; right corner of the Council House
g to left corner of monument M″ on the plan; right corner of the propylon in the new boundary wall
h to right corner of monument M″ on the plan; right corner of the propylon to the eastern building
i to left corner of monument M″ on the plan; entrance B
j to right corner of monument M″ on the plan; right corner of monument M′ on the plan; left corner of the propylon to the Council House
k to left corner of the propylon to the eastern building; left corner of the eastern building; right corner of monument M on the plan.
Sight lines drawn from the other entrances yield similar results.
The newly enclosed northern agora could be entered only from the propylon, which lay directly on the axis of the temple. But lines l and m seem to indicate the position of another entrance (E), as from point D the propylon would be seen in its entirety between two monuments. This entrance is also denoted by the later erection in the middle of the agora of a small altar (F on the plan), which is so placed that it in no way interferes with the view of all existing structures from point E. In fact, line n, which runs from point E to the right corner of the altar, also touches the left corner of the last of these structures, thus bringing the altar into the picture as part of a continuous sequence of objects, with no gaps or overlapping. It seems, therefore, that the position of the altar (F) was determined by the axis of the agora and line n from point E.
If we assume that point E represents an entrance to the agora, then it must also have been a point of access from the harbor or from an important space, such as a temple precinct. The space to the north is sufficiently large for either assumption to be possible, although neither is certain. Gerkan considered it probable that point E represents an entrance either from the harbor or from an earlier important architectural complex.
At the beginning of the Christian era the general layout of the site remained substantially as before, but the desire to create enclosed spaces was manifest everywhere (Fig. 35). The continued respect for the freedom of sight line n adds strength to the supposition that an entrance existed at point E, but this still cannot be confirmed.
By the second century A.D. the site had become completely Romanized through the erection of new stoas that cut across many of the earlier sight lines from the entrances (Fig. 36).
Berlin, Staatliche Museen. Milet: Ergebnisse der Ausgrabungen und Untersuchungen seit dem Jahre 1899, edited by Theodor Wiegand. Berlin, 1906–1926. (I referred especially to Armin von Gerkan, Der Nordmarkt und der Hafen an der Löwenbucht, 1922, and Georg Kawerau and Albert Rehm, Das Delphinion in Milet, 1914.)
Gerkan, Armin von. Griechische Städteanlagen. Berlin: De Gruyter, 1924.
Kobylina, M. M. Milet. Moscow, 1965.
Weickert, Carl. “Neue Ausgrabungen in Milet.” In Neue deutsche Ausgrabungen im Mittelmeergebiet und im vorderen Orient. Berlin: Deutsches Archäologisches Institut, 1959, pp. 181–196.
The large enclosure, or Altis, at Olympia was one of the earliest Greek sanctuaries.20 Its many buildings date from several different periods, and the layout cannot be considered as representing an organized plan (Fig. 41). In fact, the Altis is a typical example of a site that has continually been enriched by the addition of new buildings and monuments—in this case, from the earliest archaic period to late Roman times, and even into the Christian era.
The existing foundations of buildings and monuments are insufficient to permit a reconstruction of the form of the early precinct, especially as the original lines of its enclosure walls and entrances have not yet been traced. Only after the rebuilding of the temple of Zeus in the fifth century B.C. can there be some degree of certainty. It is possible that a general reorganization of the site was undertaken at that time and that careful heed was paid to the existing monuments. By the time the new temple of Zeus was completed, the Altis had acquired a fully organized layout containing the following structures:
the temple of Zeus, built between 470 and 456 B.C.
the Heraion, probably the oldest building on the site (the early temple with wooden columns was rebuilt in stone about the beginning of the sixth century B.C.)
several treasuries, sixth century B.C. onward
the Prytaneion, fifth century B.C.
the first Echo Stoa, rebuilt in the second half of the fourth century B.C.
the Pelopion, first planned as a circular precinct sacred to Pelops and later given a pentagonal enclosure wall
the Hippodameion, of which no definite traces remain, but whose outline I have sketched in, based on the system that I have described
the altar of Zeus, of which there are no definite remains
the altar of Hera, the date of which is undetermined
many small altars and votive structures of various periods.
The form given to the enclosure at this time remained substantially unchanged for a century, as the main lines of the layout were retained when the following new structures were added in the fourth century B.C.:
the southeast building, constructed in the first half of the fourth century B.C. and destroyed by Nero’s building in the first century A.D.
the new Echo Stoa, second half of the fourth century B.C.
the Metroon, temple of the Mother of the Gods, first half of the fourth century B.C.
the circular Philippeion, begun shortly after 338 B.C. by Philip II of Macedon
numerous new altars and votive structures.
The principles on which the layout was organized remained identical during both these periods (the first of which was distinctly classic, and the second, distinctly Hellenistic) even though the site acquired many new buildings during the fourth century. Throughout both periods also, the boundaries and especially the entrances to the Altis remained unchanged. Because of the similarity between the two periods, I shall discuss them together.
During the Roman period, which I treat separately, the Altis was repeatedly altered and extended: new stoas were built, some of the entrances were changed, and the exedra of Herodus Atticus was cut into the northern boundary.
There were four entrances (A, B, C, and E on the plan, Fig. 39). The organization of the site will be discussed as seen from each of these in turn. No traces have been found of a pre-Roman southeast entrance at point A. I have therefore assumed that, as was customary elsewhere, the Roman portal was built where the Greek entrance gate had formerly been situated; this would have marked the end of the sacred way and the beginning of the Altis. This assumption is supported by our knowledge that in ancient Greek times the main road ended at approximately this point. The position of the southwest entrance, just below point B, is definitely established. On entering the Altis from this entrance and moving in the most important direction, that is, toward the altars of Zeus and Hera, one must pass close to the northwest corner of the temple of Zeus. I thus consider that this point (D on the plan) could also have become an important starting point for the layout of the west side of the precinct. Entrance C in the northwest corner of the Altis is not entirely certain, but I have accepted the position determined by the excavators in their reconstruction of the site. Although the position of entrance E in the southern wall is well established, it seems to have had very little significance in the organization of the site plan.
As the analysis will show, the field of vision from each of these points consists of a central opening bounded on either side by a continuous series of structures. The position, orientation, and distance of the buildings from each point are determined on the basis of the 30° angle. Throughout, one can sense the desire to connect the outlines of the different structures with one another and with the lines of the landscape, to form a continuous unity, and within this unity to emphasize one opening: one clear and unobstructed path leading out into the landscape. That this was the general purpose seems certain. With regard to the details, however, because of the complexity of the site and the need for consideration of the many existing structures at every new stage of development it was inevitable that in some instances the relationships established should be only approximately correct.
Measurements taken at the site show that point A (Fig. 39) is on the axis of the Roman entry, one meter back from the inner edge of the Roman foundations.
a to left (southwest) comer of the temple of Zeus; middle (southeast) corner of the Theocoleon, the priests’ residence outside the Altis (see Fig. 41).
b to left side of the Victory of Paeonios; right (northeast) comer of the temple of Zeus (F on the plan)
c to right side of the Victory statue; left (southwest) corner of the Heraion (M on the plan)
d to left corner of the altar of Zeus*;21 right (northeast) comer of the Heraion
e to right corner of the altar of Zeus*; left corner of the altar of Hera
f to right comer of the altar of Hera
g to left end of the series of altars and treasuries on the upper terrace
h to left (southwest) corner of the Metroon (G on the plan); right corner of the first treasury
i to right (northeast) comer of the Metroon; left corner of the propylon of the Hippodameion*
j to left corner of the nearest monument
k to left (northwest) corner of the Hellenistic Echo Stoa
l to right comer of the nearest monument; left comer of the nearest altar
m to right comer of the façade of the Hellenistic Echo Stoa (southwest corner)
n to left corner of the southeast building; right (southeast) comer of the Hellenistic Echo Stoa.
(This was the right, i.e., southwest, corner of the façade of the former stoa.)
Angle ab = 31° (30° = 180°/6).
Angle bh = 30° = 180°/6.
Angle hk = 30° = 180°/6.
Angle km = 30° = 180°/6.
The entire field of vision is encompassed within an angle of 121° (120° = 2 × 180°/3). If we examine the angles that encompass each of the major buildings we find
the Heraion lies within an angle of
15° = 180°/12;
the Metroon lies within an angle of
10° = 180°/18 = 1/3(180°/6);
the Echo Stoa lies within an angle of
30° = 180°/6;
the temple of Zeus lies within an angle of
31° = ca. 180°/6.
It is noteworthy that the Parthenon at Athens, which has certain similarities with the temple of Zeus, also falls within an angle of 30° when viewed from the propylon.
If an equilateral triangle AFF′ is described on the base AF (F is the northeast corner of the temple of Zeus; see Fig. 39), we find that F′ falls along the line k, and that the sides of the triangle measure 80.5 m, or approximately 250 Olympian feet (250 × 0.328 = 82.0 m).22 If an arc is described from point A with radius AF (80.5 m), the point at which it crosses line i may indicate the left corner of the propylon to the Hippodameion. If the triangle AFF′ is inverted on its side FF′ to form the triangle FGF′, we find that G falls on line h at the southwest corner of the Metroon.
or AG = 80.5 × 1.732 = 139.4 m.
Measurements taken on the site show that AG = 139.2 m.
In other words, the significant measurements are . It may also be noted that the basic dimension used in the reorganization of the Acropolis at Athens, which was roughly contemporary with the Olympian Altis, was 79.6 m; this is very near the dimension 80.5 m that we find in the Altis.
From left to right, we see from point A (Fig. 40)
the temple of Zeus
the Victory of Paeonios
the altar of Zeus
the altar of Hera
an open field of vision looking toward the small Hill of Gaia23
a fountain (very low and perhaps not visible), some altars, and the first treasury
the Hill of Kronos with the Hippodameion
the Echo Stoa.
The differing orientation of these buildings and the distances between them are determined by the equilateral triangle AFF′, in particular the angle of 30° (180°/6), or a twelfth part of the total field of 360°.
Figure 40 shows how the mass of the temple of Zeus is balanced symmetrically by the Hill of Kronos, and the Metroon by the Heraion. Both are symmetrically placed on either side of the axis leading to the small Hill of Gaia, which rises only slightly higher than the Heraion and the Metroon. This axial symmetry is clearly strengthened by two balancing groves of trees, one within the Pelopion to the left, and the other in the Hippodameion to the right.
The Victory statue was apparently placed to occupy exactly the small angle of vision between the northeast corner of the temple of Zeus and the southwest corner of the Heraion, perhaps to emphasize the difference in volume of these two buildings. Its position is very similar to that of the statue of Athena Promachos on the Acropolis at Athens, which, when seen from the Propylaea, stands exactly between the Erechtheion and the smaller mass of the altar of Athena (Fig. 2). The top of the statue of Victory and the tip of the acroterion on the temple of Zeus were on the same horizontal level. This may also have been intentional.
It seems clear that a principal aim of this symmetrically organized layout, in which the landscape is incorporated, was to maintain the importance of the central axial opening. This marks the processional route of the people through the sacred precinct from entrance A to the altars. Also, from this entrance the peak of the Hill of Kronos lies directly to the north. Thus, one of the cardinal compass points is made an integral part of the composition.
Although this is the only instance in which we have found a spatially symmetrical layout, the conscious use of symmetry by the Greeks is not precluded. They did not shun symmetry when its use suited their purpose. For example, on leaving the Altis through entrance A, going directly south, one has the impression that the landscape ahead opens up axially and symmetrically. The broad band of the river Alpheios lies directly across the path, and the background is occupied by a balanced line of hills, the tallest in the center and two lower ones on either side (Fig. 42). The outline of this mountain chain, which dominates the whole valley, is immediately impressive. This was felt by the ancient Greeks and re-echoed in their layout. The outline of the mountains is still visible today, but the view of the river is now obstructed by alluvial deposits; one must imagine it stretching in a straight line across the foreground.
Point B (Fig. 43) has been located a short way in from the entrance, since the position of the monument immediately to the right of the entrance would have impeded a clear view of the temple of Zeus. Figure 46 shows a perspective view of the Altis from this point.
a to left tangent of the Philippeion
b to right tangent of the Philippeion; left corner of the entry to the Prytaneion
c to left side of the small altar before the Heraion; right corner of the entry to the Prytaneion
d to left (southwest) corner of the Heraion, approximately along the line of the west façade
e to left corner of the propylon to the Pelopion; west side of the Pelopion*
f to right (northeast) corner of the propylon to the Pelopion (H on the plan)
g to right (southeast) corner of the Heraion
h to right (northwest) corner of the temple of Zeus (D on the plan); left corner of the first of the row of treasuries on the upper terrace
i to right (southeast) corner of the temple of Zeus (H′ on the plan)
j to left side of the Victory statue; right corner of the Echo Stoa*
k to right side of the Victory statue; left comer of the southeast building.
Angle af = 30° = 180°/6.
Angle fi = 60° = 180°/3.
From point B, an equilateral triangle BHH′ can be formed, with H lying on line f at the corner of the propylon to the Pelopion and H′ lying on line i at the southeast corner of the temple of Zeus, the sides measuring 88.0 m.
The distance from point B to point I—the ramp to the propylon of the Pelopion—measures 76.2 m.
Hence the height of the equilateral triangle BHH′.
This implies that the significant measurements from this point were .
Distances to other important points from point B cannot be precisely determined without a more accurate field survey. When studied on the existing plans they show small deviations from the usual system that cannot be explained without more complete information.
Figure 46 shows the view from point B during the first period, before the construction of the Philippeion in the first half of the fourth century B.C.
From left to right we see
the Heraion, with the propylon to the Pelopion appearing in the center of its façade, giving
the impression that this axial and symmetrical position was intentional
an open field of vision between the Heraion
and the temple of Zeus, toward the Hill of Kronos
the temple of Zeus
the Echo Stoa
the statue of Victory of Paeonios
the southeast building, to the extreme right
and closing the picture.
Seen from point B the Victory statue exactly occupies the space between the Echo Stoa and the southeast building. Just as when viewed from point A, its height is related to the temple of Zeus, although in a different way: from point A it emphasizes the upward thrust of the temple summit; from point B it punctuates the far end of the long pediment.
The outline of the Hill of Kronos appears to continue on the line of the architrave of the temple of Zeus and to link it with the Heraion. The hill on the left and the Victory statue on the right thus form a visual unity with the temple of Zeus.
From point B the north point lies directly along line d, which leads to the west end of the Heraion, so that this principal direction is again emphasized in the layout.
Point D (Fig. 43) lies at the extreme northwest corner of the temple of Zeus, at the corner of the steps, between lines g and h from point B.
a to left (southwest) corner of the Heraion
b to right (southeast) corner of the Heraion
c to right (southeast) corner of the Pelopion (in Roman times [Fig. 49] this line ran along the southeast wall of the enclosure)
d to left (northwest) corner of the Metroon; right corner of the third treasury on the upper terrace
e to right (southeast) corner of the Metroon; left corner of the great altar on the upper terrace
f to left (northwest) corner of the Echo Stoa; left corner of the last treasury on the upper terrace.
The only statement that can be made with certainty is that the Metroon is seen within an angle of 10°, just as from point A. The other angles would need to be checked after completion of an accurate site survey.
I have placed point C (Fig. 39) in the center of the entrance determined by the excavators, on a line with the inner side of the wall enclosing the precinct.
a to left corner of the entrance to the Prytaneion
b to right (east) corner of the Prytaneion; left (northeast) corner of the Heraion
c to right side of the small altar in front of the Heraion; right (southwest) corner of the Heraion (M on plan)
d to central (northwest) corner of the Pelopion; left (northeast) corner of the temple of Zeus
e to left (northeast) corner of the propylon of the Pelopion; fifth column from the northeast corner of the temple of Zeus
f to right (southwest) corner of the propylon to the Pelopion; fifth column from the northwest corner of the temple of Zeus
g to right (southwest) corner of the temple of Zeus, before the erection of the Philippeion
h to right tangent of the Philippeion.
The Heraion is seen within an angle of 30° = 180°/6.
The distance from C to the far side of the entrance to the Prytaneion (K on the plan) is equal to the distance to the nearest point of the Philippeion (K′ on the plan).
CK = CK′ = 20.9 m.
Similarly, the distance to the nearest corner of the Heraion (L on the plan) is equal to the distance to the farthest point of the Philippeion (L′ on the plan).
CL = CL′ = 35.5 m.
The distance from C to the right (southwest) corner of the Heraion measures 41.6 m.
Hence CK = CM/2.
Arithmetically, CK = 20.9 = 41.8/2 (by measurement CM = 41.6 m).
Arithmetically, and CM = 41.0 m (by measurement CM = 41.6 m).
Therefore, CK, CL, and CM are the three sides of a right-angled triangle with angles of 30°, 60°, and 90°, CM being the longest side. Similarly, they represent one side, the height, and half the base of an equilateral triangle with sides of 41.6 m. In other words, they represent a twelve-part division of the total field of 360°.
Standing at point C (Fig. 48), looking from left to right we see
an open field of vision between the Heraion
and the temple of Zeus
the temple of Zeus with the propylon appearing in the center of its façade, just as, from
point B, the propylon had appeared in the center of the Heraion
(the Philippeion, at a later date).
The outline of the mountains in the background is linked to the outline of the temple of Zeus to form a visual unity.
As has been said, although this entrance (Fig. 43) is known to have existed, it seems to have played little part in the determination of the layout. Point E has been located in the center of the entrance on the line of the inner side of the precinct boundary wall. It is possible that the boundary line of the northeast side of the Pelopion was determined by the direction EF′, F being the northeast corner of the temple of Zeus.
In Roman times (Fig. 49) the Altis received a new boundary wall, several of the former entrances were closed, and the northwest corner, especially, was transformed by the construction of a new stoa. As a result of these changes, the earlier system of relationships ceased to exist. Moreover, the erection of the exedra of Herodus Atticus in the northern boundary wall influenced the entire site by completely blocking the open field of vision, so that the former impression of a path leading directly through the sanctuary into the landscape was utterly destroyed. The Roman Altis had become fully enclosed. Thus, many of the principles that had governed the composition of the site during the classical and Hellenistic periods—in particular, the use of the landscape as an integral part of the plan—had now been abandoned.
The proportions of the temple of Zeus are 1:5, and those of the Metroon, 1:2.
Curtius, Ernst, and Adler, Friedrich. Olympia: Die Ergebnisse der von dem deutschen Reich veranstalteten Ausgrabung. 5 vols. Berlin: Asher, 1890–1897.
Dörpfeld, Wilhelm. Alt-Olympia. Berlin: Mittler, 1935.
Schleif, Hans. Die neuen Ausgrabungen in Olympia und ihre bisherigen Ergebnisse für die antike Bauforschung. Berlin, 1943. (The material was available for study before publication.)
Drees, Ludwig. Olympia. Stuttgart: Kohlhammer, 1967.
Essen, Ausstellung, 1960. Olympia in der Antike. Essen, 1960.
Kontis, Ioannes D. Tό ίερόν τῆs Ὁλvμπίαs. Athens, 1958.
Schleif, Hans. Das Philippeion. Olympische Forschungen, edited by Emil Kunze and Hans Schleif, vol. 1. Berlin: De Gruyter, 1944.
This temple site shows traces of two phases of active building construction. From the first phase, which dates from before the Persian invasion, there are only some remains of the early temple, built of tufa, and of the propylon. All the remains now visible seem to date from the fifth-century reconstruction: the marble temple, the north and west stoas, the marble propylon, and the altar, whose traces have been found on the rock near the southeast corner of the temple.
Remains of the layout of the earlier precinct are insufficient to permit an investigation. We know that the dimensions of the platform of the temple that lay below the present one were 30.34 × 13.12 m24 and that those of the later marble temple were 31.15 × 13.48 m. According to Stais, an earlier tufa building lay under the fifth-century propylon, but he furnished no proof of this.25 His opinion is supported, however, by Dörpfeld’s general observation regarding the Poseidon temples that “when constructing a new building, the ancient Greek architects sometimes intentionally retained the proportions and form of an earlier building.”26 If we assume that this statement can be applied not only to the temple but to other important buildings within a sacred precinct, such as the propylon, we can perhaps justify basing the spatial layout of the fifth-century precinct at Sounion on that of the tufa temple in the archaic period.
There are various opinions about the exact date of the fifth-century marble temple. Dörpfeld considers it almost contemporary with the “Theseum” (temple of Hephaestus), beside the agora in Athens, with which it has close affinities. This would mean it is a little later than the Parthenon.27 A probable date is about 430 B.C.28 The large northern stoa, the small stoa, and the marble propylon all appear to have been built after the temple. Thus, while the precinct undoubtedly had an organized site plan, it is not possible to determine whether this was first created in the classical period or whether it followed lines laid down in the earlier archaic precinct that was destroyed by the Persians.
The entrance was through the propylon; point A has been located on the axis of this structure at the inner edge of the platform (Fig. 52).
a to left corner (northeast) of the lowest step of the temple (B on the plan)
b to left (southeast) corner of the temple platform
c to right (northwest) corner of the temple platform
d to right (northwest) corner of the lowest step of the temple (C′ on the plan)
e to left (southeast) corner of the small stoa.
Angle ad = 60° = 180°/3.
The distance AB along the line a to the northeast corner of the lowest step of the temple measures 27.2 m. We can call this distance x.
If we describe an equilateral triangle ABB′, we find that point B′ falls along the line d. If this triangle is then inverted along the side BB′ to form the triangle BCB′, with radius AC, we can describe an arc cutting the projection of the line AB′ at C′, which falls at the northwest corner of the lowest step of the temple.
Hence ; since AB = 27.2, AC = 27.2 × 1.732 = 47.11 m.
Measurements taken on the site show that AC = 46.5 m.
If another equilateral triangle ADD′ is described, with its height AC, we find that point D falls close to where traces of the altar of Poseidon have been found in the rock. Thus, it is possible that point D marks the nearest (i.e., the southwest) corner of this altar, lying at a distance of 2AB, or 2x, from point A and upon the arc AD′.
From left to right we can see
the eastern boundary wall of the precinct
an open field of vision
the broad-stepped base of the temple and
perhaps the altar of Poseidon
the temple of Poseidon
the stepped base of the temple
an open view of the sea
the western stoa.
The observer has two open fields of vision, one to the left of the temple, the other to the right, between the temple and the stoa. In general, we have found that whenever a sector of the field of vision is left open, it has a particular significance: it may, for instance, mark the processional way to the altars, as on the Acropolis at Athens and in the Altis at Olympia, or it may stress one of the cardinal points, as is the case, again, on the Acropolis at Athens and—where this is very clear—in the agora at Pergamon. At Sounion the open view to the left of the temple leads to the temple entrance and, most probably, to the altar of Poseidon. The open view to the right of the temple may have been left free to emphasize the path to a cave below the rock, which perhaps had special significance in the cult of Poseidon. It is also possible that this sector was not, in fact, left open but was closed by a structure of which no trace has yet been found (as was the case, for example, on the terrace of Athena at Pergamon). But, if we take the surrounding landscape into account, we cannot rule out the idea that the purpose of an open sector here was to offer an unobstructed view of the sea—the realm of Poseidon. Examination of the cardinal points of the compass reveals that the two limiting lines of this open sector to the right of the temple lie at 10° and 20° south of west, so that during the months of February and October there would be a direct view of the setting sun over the sea.
Clearly, the extension of the landscape played an important part in the plan for this precinct. The great temple is raised high in the center of the area in contrast to the other buildings, which are so much lower and smaller that they appear quite insignificant. Similarly, the immediate natural surroundings of the temple are unobtrusive and have no marked characteristics. The position of the northern and western stoas echoes the relationship between two low chains of hills that meet one another at a right angle, to north and west.
Dörpfeld, Wilhelm. “Der Tempel von Sunion.” Deutsches Archäologisches Institut. Mitteilungen. Athenische Abteilung 9, 1884.
Fabricus, Ernst. “Die Skulpturen vom Tempel in Sounion.” Ibid.
Orlandos, Anastasios K. “Tó ἀέτωμα τοῦ ἐv Σoυvίωvαoῦ τoῦ Πoσειδῶvco.” ’Aρχαιoλoγική ’Eφημεoίs, 1915.
———. “Toῦ ἐν Σουνίω ναοῦ τοῦ Πoσειδῶνοs τοῖχοι καί ὀροφή.” Ibid., 1917.
Staϊs, Valerios, “’Aνασκαφαί ἐν Σουνίω.” Ibid., 1900.
———. “Σουνίου ἀναοκαφαί.” Ibid., 1917.
———. Tó Σύνιον καί οἱ ναοί Ποσειδῶνοs καί ’Aϑηνᾶs. Athens: Library of the Archaeological Society, 1920.
William H. Plommer. “Three Attic Temples,” pt. 2, “The Temple of Poseidon.” British School at Athens, Annual 45, 1950, pp. 78–94. (A detailed account of the measurements of the temple.)
———. “The Temple of Poseidon on Cape Sunium: Some Further Questions.” BSA 55, 1960, pp. 218–233.
It is not known precisely when construction of the Pergamon agora began, but the existing ruins seem to indicate that the entire site dates from the same period—the era of the Seleucid kings.29 It is thought that the western half of the agora and the stoas were built at the same time30 and that the temple of Athena and the altar of Zeus were erected somewhat later.
There were two entrances, north and south, at the points where an older road crossed the agora. They are marked A and B on the plan (Fig. 56). Through entrance A passed all the traffic proceeding from the high citadel to the lower city, south of the agora (Fig. 57). Similarly, entrance B was used by all traffic from the lower city proceeding to the citadel.
Point A is situated in the center of the entrance at the first spot from which the entire western agora can be seen.
a to farthest (southern) corner of the southeast stoa (C on the plan)
b to left corner of the central group of monuments (F on the plan)
c to right corner of the central group of monuments; left corner of the temple façade (D′ on the plan)
d to left corner of the exedra; right (northeast) corner of the temple façade
e to middle corner of the exedra; right (northwest) corner of the temple (C′ on the plan)
f to right corner of the exedra; left comer of the northwest building (D′ on the plan)
g to a line parallel to the northern retaining wall
Angle ab = 30° = 180°/6.
Angle bg = 60° = 180°/3.
Thus, the northern boundary of the upper half of the agora is based on the position of line g.
AC, along line a = AC′ along line e = 52.4 m. This is exactly 100 Egyptian ells;31 i.e., 100 × 0.524 = 52.4 m.
AD′ along line c = AD″ along line f = 45.37 m.
If an arc is described with center A and radius AD′ to cut line b at D, then , and so
45.37 = 52.40 × 1.732/2.
Thus ACD is a right-angled triangle whose side AC = 52.4 m and angle ab = 30°.
The composition of the layout is determined by the angles of 30° and 60° and the ratio resulting from them (. In other words, there is a twelvefold division of the total area of 360°.
The field of vision is fully enclosed. The view of the temple façade between lines c and d is kept clear. Line AG, which points due west to the sunset, lies exactly between AE and AD′ (line c).
At point B, in the center of the southern entrance to the site, the road is 3 m lower than the surface of the western half of the agora. Only when the spectator moves to point B′ is his eye at the same height as the parapet, i.e., 0.70 m above ground level. B′ is thus the first point from which a view of the site can be gained; it can therefore be considered a visual entrance to the site. If the road were not sunk, a view of the open area without buildings that lies south of the central group of monuments might be obtained between points B and B′ (see Fig. 55). As it is, however, the layout of the site permits no open views, for point B′ the view of the open area is already closed by the most southerly of the central group of monuments. This point also determines the line B′E.
Although the temple of Athena was built in the fourth century B.C., when Pergamon was only a small town, its precinct was not created until the reign of King Eumenes II (197–159 B.C.), at a time when many other urban changes were taking place. The existing temple served as the starting point for the layout of the precinct, which was formed by the erection of three stoas. The only remaining monument, which stands in the center of the precinct, dates from the Roman era. We can be fairly certain that it replaced a Hellenistic monument, since its position fits into the Hellenistic layout, and since we know that in the Hellenistic period the precinct was adorned with many famous statues, of which there are now no traces.
The precinct had two entrances, a main entrance (A on the plan, Fig. 61) from the east through the propylon and a minor entrance (B on the plan) through the northern stoa.
Point A lies on the platform of the propylon, at the crossing of its axis with the line of the top step of the eastern stoa.
a to right (northwest) corner of the southern stoa (D on the plan); left corner of the altar of Athena*
b to right corner of the altar of Athena*; pronounced angle in the boundary wall
c to left (southeast) corner of the temple of Athena (E′ on the plan); another angle in the boundary wall
d to left tangent of the central monument; right corner of the temple of Athena (junction with boundary wall)
e to right tangent of the central monument; left (southeast) corner of the northern stoa F″ on the plan).
Angle ce = 30° = 180°/6.
The angle between line e and the top step of the eastern stoa = 60° = 180°/3.
Point A to the northeast corner of the precinct (C″) = 37.0 m. If an arc is described from A with radius AC″, it touches the nearest point of the central monument at C and cuts a line a at a point C.
The distance along line a from point A to the northwest corner of the southern stoa (D) = 55.5 m. An arc with radius AD touches the northeast corner of the temple (D′).
The distance along line c from point A to the southeast corner of the temple (E′) = 64.0 m.
The distance along line e from point A to the southeast corner of the northern stoa (F″) = 74.0 m. An arc with radius AF″ touches the southwest corner of the temple (F′) and perhaps denotes the northeast corner of the altar of Athena* (F).
Hence AC′ = AC″ = 37.0 m = 74.0 m/2 = AF″/2,
and AD = AD′ = 55.5 m = 3AF″/4,
or AC′ = the small side of the right-angled triangle AF″C″.
Also AD = AD′ = 55.5 m = 3AF″/4, or AD = three-quarters of the large side of the triangle AF″C″.
In other words, the organization of the space is determined by the right-angled triangle AF″C″ and the use of the angle of 30°, or a twelvefold division of the total area of 360°.
From left to right we can see (Fig. 60)
the southern stoa
the altar of Athena*
an open view, obstructed only by the low
parapet of the enclosure wall
the temple of Athena
the circular central monument
the northern stoa.
The path to the altar and to the entrance of the temple of Athena is emphasized by an open view; line c, leading to point E′ at the southeast corner of the temple, lies directly perpendicular to the main entrance to the precinct through the propylon. The open field of vision is oriented 21° south of west.
The right-angled triangle AF″C″ is similar to the triangle formed by half the base of the temple, E′D′G, which also has angles of 30°, 60°, and 90°. The triangle E′D′G lies so that its sides are either parallel or perpendicular to the sides of the triangle AF″C″. Thus, for example,
AF″ is parallel to GD′
AC″ is parallel to E′G
E′D′ is perpendicular to AF″.
As we have mentioned, the temple was built two centuries before the precinct was laid out, and we may conclude that its position and proportions determined the form, size, and orientation of the Hellenistic stoas.
This small entrance gave access to the sacred precinct from the north. Point B lies on the edge of the top step in the center of the opening.
h to left tangent of the central monument; southeast corner of the precinct
g to left (northeast) corner of the temple of Athena (D′ on the plan), right (northwest) corner of the southern stoa (D on the plan)
f to right (northwest) corner of the temple (G on the plan); and angle in the boundary wall.
Neither the exact date of the great altar of Zeus nor the outlines of its terrace are precisely known, but it is generally believed that it was built at the time of King Eumenes II (197–159 B.C.). This assumption is based on the style of the sculptures and on the inscriptions that adorn it, as well as on its enormous dimensions. The outbreak of hostilities following Eumenes’ reign also make it unlikely that a monument of this size could have been undertaken much later than this date. We have thus accepted this date.
Three entrances seem likely, although they have not been definitely proved (Fig. 64). There is slight evidence of a propylon on the east boundary, and point A has been established as its center. From this point, the altar appears within an angle of 60°. In the northwest corner of the terrace, there are signs of some steps leading down from an upper terrace, although their beginning and end have not been ascertained. If we assume that an entrance was situated at the foot of the steps, and establish point B as its center, we find that the altar again appears within an angle of 60° and, further, that the west side of the altar lies directly perpendicular to point B. A third entrance (point C) is assumed to lie at the head of a flight of steps—of which some traces have also been found—leading down to a lower terrace to the south.
We find that the sight lines from each of these three entrances (assuming that their positions have been accurately determined) coincide, that each encompasses two corners of the altar, and that together they form a single equilateral triangle ABC, within which the rectangle of the altar is precisely placed. We also find that from entrance A there is a view due west across the altar to the only visible peak of a mountain range lying on the far side of the valley of a tributary of the Caicus. The position of the altar thus acknowledges and emphasizes the dominating feature of the landscape (see Fig. 57).
Bohn, Richard. Das Heiligtum der Athena Polias Nikephoros. Berlin, Staatliche Museen. Die Altertümer von Pergamon, vol. 2. Berlin: Spemann, 1885.
Conze, Alexander, et al. Stadt und Landschaft. Berlin, Staatliche Museen. Die Altertümer von Pergamon, vol. 1. Berlin: Reimer, 1913.
Schrammen, Jakob. Der grosse Altar; der obere Markt. Berlin, Staatliche Museen. Die Altertümer von Pergamon, vol. 3, pt. 1. Berlin: Reimer, 1906.
Humann, Carl. Der Pergamon Altar. Dortmund: Ardey, 1959.
Rohde, Elisabeth. Pergamon. Berlin: Henschel, 1961. Schober, Arnold. Die Kunst von Pergamon. Vienna: Rohrer, 1951.
Zschietzschmann, Willy. “Pergamon.” In Pauly, A. F. von, ed. Real-Encyclopädie der classischen Altertumswissenschaft. Stuttgart: Metzler, 1937. XIX, pp. 1235–1264.