On the kinds of temples

Contents of this chapter

  1. Text and translation
  2. Comment
    1. About the text
    2. About the old editions
    3. The sources of Vitruvius
    4. Historical and archaeological evidence
      1. The three temples of Fortune near Porta Collina
      2. The temple(s?) of Jupiter and Faunus on the Tiberine Island
      3. The temple of Iupiter Stator in the porticus Metelli
      4. The temple of honor and virtus
      5. The temple of Diana at Magnesia
      6. The temple of Diana at Ephesos
      7. The Olympieion at Athens
      8. Hypaethral temples
  3. Bibliography
  4. Illustrations of the Cesariano edition, 1521

Text and translation

1. Aedium autem principia sunt, e quibus constat figurarum aspectus; et primum in antis, quod graece naos en parastasin dicitur, deinde prostylos, amphiprostylos, peripteros, pseudodipteros, hypaethros. Horum exprimuntur formationes his rationibus.


1. It is from the plan of a temple that the effect of its design arises. And first in antis, which in Greek is called naos en parastasin; next, prostyle, amphiprostyle, peripteral, pseudodipteral, hypaethral. The designs of these are formulated in the following manner.

1.In antis erit aedes, cum habebit in fronte antas parietum qui cellam circumcludunt, et inter antas in medio columnas duas supraque fastigium symmetria ea conlocatum, quae in hoc libro fuerit perscripta. Huius autem exemplar erit ad tres Fortunas ex tribus quod est proxime portam Collinam.


2.A temple will be in antis when it has in front pilasters terminating the walls which enclose the shrine, and in the middle, between the pilasters, two columns, and above a gable, built with the symmetry to be set forth in this book. An example of this will be the Temple of Fortune, nearest of the three to the Colline Gate.

3.Prostylos omnia habet quemadmodum in antis, columnas autem contra antas singulares duas supraque epistylia, quemadmodum et in antis, et dextra ac sinistra in versuris singula. Huius exemplar est in insula Tiberina in aede Iovis et Fauni.


3. The prostyle has everything like the temple in antis, except two angle columns over against the pilasters; and above, entablatures as in antis; which return at the angles on either side. An example of this is on the island in the Tiber, namely, the Temple of Jupiter and Faunus.

4. Amphiprostylos omnia habet ea, quae prostylos, praetereaque habet in postico ad eundem modum columnas et fastigium.


4. The amphiprostylos has everything like the prostyle, and besides has columns and a pediment at the back.

5. Peripteros autem erit, quae habebit in fronte et postico senas columnas, in lateribus cum angularibus undenas. Ita autem sint hae columnae conlocatae, ut intercolumnii latitudinis intervallum sit a parietibus circum ad extremos ordines columnarum, habeatque ambulationem circa cellam aedis, quemadmodum est in porticu Metelli Iovis Statoris Hermodori et ad Mariana Honoris et Virtutis sine postico a Mucio facta.


5. The peripteral will be that which shall have six columns in front and six at the back, and on either side eleven, counting in the angle columns. Now these columns are to be so placed that there is all round a distance the width of an intercolumniation, between the walls and the outer rows of the columns. This provides a walk round the cell of the temple, such as there is at the temple of Jupiter Stator by Hermodoros in the Portico of Metellus, and the temple of Honor and Virtus built without a posticum, by Mucius near the Monument of Marius.

6. Pseudodipteros autem sic conlocatur, ut in fronte et postico sint columnae octonae, in lateribus cum angularibus quinae denae. Sint autem parietes cellae contra quaternas columnas medianas in fronte et postico. Ita duorum intercolumniorum et unae crassitudinis columnae spatium erit ab parietibus circa ad extremos ordines columnarum. Huius exemplar Romae non est, sed Magnesiae Dianae Hermogenis Alabandei et Apollinis a Menesthe facta.


6. The pseudodipteros is so planned that here are eight columns both in front and at the back, and fifteen on each side, including the angle columns. But the walls of the cella are to face the four middle columns in front and at the back. Thus there will be a space all round, from the walls to the outside rows of the columns, of two intercolumniations and the thickness of one column. There is no example of this at Rome; but there is at Magnesia the temple of Diana built by Hermogenes of Alabanda, and the temple of Apollo by Menesthes.

7. Dipteros autem octastylos et pronao et postico, sed circa aedem duplices habet ordines columnarum, uti est aedis Quirini dorica et Ephesi Dianae ionica a Chersiphrone constituta.


7. The dipteros has eight columns in front and at the back, but it has double rows of columns round the sanctuary, like the Doric temple of Quirinus and the Ionic temple of Diana at Ephesus built by Chersiphron.

8.Hypaethros vero decastylos est in pronao et postico. Reliqua omnia eadem habet quae dipteros, sed interiore parte columnas in altitudine duplices, remotas a parietibus ad circumitionem ut porticus peristyliorum. Medium autem sub divo est sine tecto. Aditus valvarum et utraqua parte in pronao et postico. Huius item exemplar Romae non est, sed Athenis octastylos et templo Olympio.


8. The hypaethral temple has ten columns in front and at the back. For the rest it has everything like the dipteral, except that in the interior it will have two stories of columns, at a distance from the walls all round like a colonnade of a peristyle. The centre as no roof and is open to the sky. There are folding doors in front and at the back. Of this there is no example at Rome; but there is the Octastyle at Athens, in the Olympian temple.


1. About the text

This text contains a great deal of technical terms like anta, prostyle etc. which can not be translated in English. I am not going to try to explain all these words. I hope this will become clear further on and on the basis of the illustrations. If there remain some questions let me know.

In sentence 6 there is a fragment that can lead to some confusion: Huius exemplar Romae non est, sed Magnesiae Dianae Hermogenis Alabandei et Apollinis a Menesthe facta. In the first printed editions of Vitruvius Hermogenis Alabandei was understood as a single name: Hermogenes of Alabanda, while in reality Hermogenes was from Priene. Some recent translations (like the one I am using) still persevere in this error. The error finds its roots in the genitivus of the word Alabandae which was seen as an adjectif for Hermogenis (also genitivus), followed by the word 'et' which was read as 'and'. In reality we must read a locativus (which has the same ending as the genitivus) and translate the word 'et' as 'also'. Doing so the text becomes more evident: Hermogenes was one of the leading architects of his time who wrote a treatise about architecture which Vitruvius must have known very well. We can not admit that Vitruvius should have made such a mistake in situating his great predecessor in the wrong town; furthermore he is very meticulous in situating the examples of each style in their context; suddenly he should have ommitted this in his mention of 'a' temple of Apollo built by Menesthes. If we consider all this we can read the fragment as follows: 'There is no example of this at Rome; but there is at Magnesia the temple of Diana built by Hermogenes (and) at Alabanda also the temple of Apollo built by Menesthes.' I think this makes more sense.

2. About the old editions

The illustrations you have found in the Latin text are taken from the printed edition of 1567. The plans and elevations of the types in antis and prostyle/amphiprostyle are completely wrong. The illustration shows the exact typology. This error was clearly due to the fact that these old editors didn't know the exact meaning of the word 'anta'. In the French translation by Claude Perrault (1684) we read the following definition of 'anta' Les mots Latin Anta & Antes signifient la mesme chose parmy la pluspart des Grammairiens, & ils viennent tous deux du mot Antè qui signifie devant: Quelques-uns y mettent cette difference que Antes sont les premiers seps qui bordent les pieces de vignes, & Anta les colonnes quarrées qui font les coins des Edifices, ou mesme les Pillastres qui sont aux costez des portes<. (Translation: The Latin words Anta and Antes have the same meaning for most Grammarians, both are derived from the word Antè which means before: some (grammarians) indicate the difference that Antes is used for the first plants bordering the rows of vineyards while Anta means the squared columns forming the corner of a building, or even the pilasters bordering doors.) From this text we may conclude that there was some uncertainty about this word. Now we know that the anta is a pilaster or corner post of slight projection terminating the end of the lateral walls of a cella, and usually serving as respond to a column (definition taken from W.B.DINSMOOR, The architecture of ancient Greece). This uncertainty lead to the misunderstanding in the illustration of sentence 2: the antae were seen as corner pilasters, almost not projecting from the cella wall; the columns described by Vitruvius stood between these pilasters. This resulted in the above given scheme, with two columns standing before the temple. It was impossible to cover this kind of elevation with one single roof. For this reason the design gives a pediment only over the two central columns.

From this error we can understand the design of the prostyle temple (sentence 3). In the description of this type the old editions thought that 'in versuris' was related to the columns while in fact it is only related to the 'epistylia', architraves or lintels which made the link between the row of columns and the cella of the temple. For this reason we find in these editions two corner columns.

3. The sources of Vitruvius

Basically Vitruvius describes Greek temples but illustrates them with Roman examples which were near at hand. His temple types have an Ionic origin and are clearly derived from Greek writings. Indeed: he uses Greek words to indicate the different types, refers to Greek architects and, when there is not a good example in Rome, he quotes Greek or Middle Eastern temples.
Hermogenes was clearly one of the most influential architects of this period. He build several great temples and adhered only the Ionic style. Other Greek architects who deserve a mention in this text are Menesthes and Chersiphron. Among these architects it was a normal practice to publish in detail the principles of their buildings. Unfortunately all these texts are lost, but we find some echoes of them in later publications. Seemingly Vitruvius is one of them, but he don’t quote literally but only summarizes his sources. Throughout books III and IV he follows the scheme of Greek temple architecture. We already saw this influence in the former chapter of this book where he explains about Greek mathematical systems and the ‘perfect number’. We may also assume that he – as a former war engineer in Caesar’s army - knew about this architecture only from books and that he never leaved Rome or Italy to study Greek temple architecture on the spot. This explains his clear mistake about the Olympieion at Athens which is described as an hypaethral temple but which was in reality not finished when Vitruvius wrote and even partially in ruin, while the building had stopped.
This explains also why he gives only (when not available in Rome) Near Eastern examples of the Hellenistic period. The architects whose works he used worked there and wrote about their achievements.
On the other hand he used also Roman predecessors. In sentence 5 he mentions a certain Hermodorus who worked at Rome and a G. Mucius who build the temple of Honor and Virtus. But this Mucius, as we learn from the preface of book VII, left no writing about his work. So Vitruvius could only rely on the building.

4. The historical and archaeological evidence

This text refers to a number of buildings. I did some research and in the following notes I will explain something about these buildings.

4.1. The three temples of Fortune near Porta Collina (sentence 2)

Three temples of Fortuna on the Quirinal, just inside the Porta Collina, which gave their name to the district (ad Tres Fortunas). The principal one of these three seems to have been that of the Praenestine goddess who was known officially at Rome as Fortuna publica populi Romani Quiritium Primigenia. The temple was dedicated in 194 B.C.
The second was dedicated to Fortuna publica citerior, the third was dedicated to Fortuna Primigenia in colle.
Of these temples only scarce remains were found and it is impossible to reconstruct the plans. We do not know which of these temples is meant by Vitruvius.

4.2. The temple(s?) of Jupiter and Faunus on the Tiberine Island (sentence 3)

Vitruvius mentions this as one building (in aede) but I am not sure that we should not read in aede Iovis et (in aede) Fauni. A temple of Faunus is known on the north end of the island: it is mentionned by Ovidius (Fasti II, 193-194 Idibus agrestis fumant altaria Fauni/Hic ubi discretas insula rumpit aquas). It was the only temple of Faunus in Rome. It was dedicated in 194 B.C. and situated on the island probably because of the non-urban character of the god. There are no references to this temple and no traces were found.
Also on the north end of the island was found an inscription with a reference to a temple of Iupiter Iurarius. Whether this was an independent shrine or to be identified with some other deity (Faunus?) is uncertain. Apart from this insciption nothing has been found.

4.3. The temple of Iupiter Stator in the porticus Metelli (sentence 5)

The porticus of Metellus was built in 147 B.C. by Quintus Caecilius Metellus Macedonicus around the temples of Jupiter Stator and Juno Regina. The temple of Juno was build around 187 B.C and the temple of Jupiter Stator some years earlier. The complex stood between the circus Flaminius and the theatre of Marcellus. The rests of the temple of Jupiter lie beneath the church of Santa Maria in Campitelli. According to Vitruvius the temple was the example of a peripteros with six columns across the front an rear and eleven on the sides and ascribed to Hermodorus who was clearly an architect from Greek origin but who worked at Rome (as freedman?).
The whole area was restored and reconstructed by Augustus after 14 B.C. We find a representation of this rebuilding on the 'Forma Urbis', a marble city map drawn under the reign of Septimius Severus (193-211 A.D.). In spite of the mention of Vitrivius of eleven columns at the sides, this map shows only ten columns. Whether this is the result of the rebuilding by Augustus or a failure in the map is not known. The illustration shows the fragment of the Forma Urbis

4.4. The temple of honor and virtus (sentence 5)

This temple was situated on the Capitoline hill and built by Marius after his victory in 102 and 101 B.C. over the Cimbri and Teutones (two Gallic tribes). Apart from some mentions in literature and the short description Vitruvius gives in this text, not so much is known about this temple. It stood on the lower slopes of the Velia near the house of Marius on the Via Sacra. There are no archaeological founds. The temple is ascribed to C. Mucius, the father of Marius’ daughter-in-law. This Mucius must have been one of the examples of Vitruvius.

4.5. The temple of Diana at Magnesia (sentence 6)

This is the first mention of Hermogenes. We will meet him later and shall have the occasion to write more about him. Hermogenes, who lived at the end of the third and the beginning of the second century B.C., was the most important architect of the hellenistic period in the Greek world. He codified the rules for Ionic order in his (lost) books which, as we shall see later, influenced greatly Vitruvius. In his studies of the Ionic order he worked out a series of ideal proportions which Vitruvius transmits us more than a century later. We shall discuss this later. According to Vitruvius (book III, chapter 3) the most remarkable innovation of Hermogenes in temple architecture was the ommitting of the inner row of columns in the temple of Diana at Magnesia to make it pseudo-dipteral. However, this scheme can be traced back through four centuries preceding him. His only merit is that he rebuilded the old temple of Diana in accordance with his prescriptions and added to the former proportions the thickness off a column - 'una crassitudo' - to the width of the ambulatory around the cella. This widening was the result of his prescription to give more emphasis to the entrance of a temple by widening the central intercolumniation through the moving of the two central columns with one thickness to the sides.

4.6. The temple of Diana at Ephesos (sentence 7)

The temple of Diana was not the largest but certainly the most important temple of Asia Minor. It was built on the site of an archaic temple which burned down in 356 B.C. It was replaced by a temple of which the plan is identical with that of its predecessor, and borrowing from the earlier temple the idea of the sculptured columns. It was the beauty of these sculptures which caused this temple to be classed among the seven wonders of the ancient world. Scopas is reputed to have been one of the sculptors. The architects of the new temple were Paeonius of Ephesos and Demetrius the slave of Artemis; sometimes the name of Deinocrates, the architect of Alexander the Great, is associated with this temple. The work is reported to have consumed 120 years, so that it would not have been completed until 236 B.C.
Vitruvius is clearly wrong when he attibutes the building of this temple to Chersiphron. Chersiphron was the architect of the archaic temple; so Vitruvius is mixing the names of the builders of this temple.
The plan is an octastyle peripteros with three rows of 8 columns at the front and at the rear and 21 columns at the sides.
Of this temple almost nothing is left.

4.7. The Olympieion at Athens (sentence 8)

Here Vitruvius is making a mistake. When he wrote his book the Olympieion was only a ruin. Later, when the temple was finished, it was completely roofed. The Olympieion was the most notable Corinthian temple of the Hellenistic period, dedicated to Zeus Olympios and situated in the plain to the southeast of the Acropolis at Athens. The temple was built on the foundations of the earlier Doric temple begun by the sons of Peisistratos; the new building was commenced by Antiochus IV Epiphanes in 174 B.C., from the designs of the Roman architect Cossutius, as a gift to the Athenian people and to satisfy his own mania for showing lavish honours to Olympian Zeus. The plan was octastyle, with twenty columns on the flank; its dimensions on the stylobate were about 44 x 166 meter, and it was built in the centre of a peribolus mesuring 139 x 222 meter. The temple was dipteral, with two rows of columns on each side of the cella, and three rows across the front and rear. The structure as designed by Cossutius was left incomplete; and in 86 B.C. some of the capitals and shafts were transported by Sulla to Rome and used to decorate the temple on the Capitol. The work was resumed in the time of Augustus, but its completion and dedication were reserved for Hadrian in A.D. 132. It is improbable that when completed by Hadrian any portion of the temple was hypaethral, in view of the exposure to which the gold-and-ivory statue of Zeus would then have been subjected.

4.8. Hypaethral temples

This type of temple was very rare in antiquity. We know only with certainty two hypaethral temples, one of which (the temple of Hera at Samos) was completely deserted in Vitruvius' days. Therefore it is remarkable that Vitruvius doesn't mention the famous temple of Apollo at Didyma. This was one of the largest temples of Asia Minor, so large that they were unable to roof it. The temple measures 51,31 m x 109,34 m. It was dipteral and unique in that the façades were decastyle (10 columns), while on the flanks there were 21 columns.

Works started probably at about 313 B.C. and were prolonged until 41 A.D. when the temple was left incomplete. In Vitruvius' time this temple was still a building site but surely enough was ready to give a good idea of the meaning of the hypaethral construction. Therefore it is remarkable that Vitruvius has choosen the totally abandonned Olympieion at Athens as example and omits Didyma.<P>Moreover his description of the hypaethral temple doesn't correspond to the scheme of Didyma. As is shown on the illustration, after going through the entrance or pronaos, one descends by a flight of steps into an inner open courtyard in which a smaller shrine stood. The description of Vitruvius with an inner row of columns in two stories refers rather to the classic interior of a Greek temple like we still can appreciate it e.g. at Aegina or at Paestum in Italy. But these temples were never unroofed nor partially open to the sky.


Daniele Barbaro, M.Vitruvii Pollionis De Architectura Libri decem, cum commentariis, Venice, 1567.
Les dix livres d'architecture de Vitruve corrigés et traduits par Claude Perrault, Paris, 1684.
F.Granger, Vitruvius, On architcture, London, 1931.
T.Peters, Vitruvius, Handboek bouwkunde, Amsterdam, 1999.
S.B. Platner-T. Ashby, A topographical dictionary of ancient Rome, London, 1929.
W.B.Dinsmoor, The architecture of Ancient Greece, London, 1950.
G.Lugli, Itinerario di Roma antica, Milano, 1970.
F. Coarelli, Guida archeologica di Roma, s.l., 1975.
R.A.Tomlinson, Vitruvius and Hermogenes, in H.Geertman-J.J.De Jong (ed.) Munus non Ingratum, Proceedings of the International Symposium on Vitruvius' De Architectura and the Hellenstic and Republican Architecture, Leiden, 1989, pp.71-75.
B.Wesenberg, Griechisches und Römisches in der vitruvianischen Architektur: ein Beitrag zur Quellenfrage, ibid., pp. 76-84.
H.Knell, Vitruvs Architekturtheorie, Versuch einer Interpretation, Darmstadt, 1991.
K. Tuchelt, Branchidai-Didyma, Geschichte und Ausgrabung eines antiken Heiligtums, Mainz, 1992.
L.Richardson Jr., A new topographical dictionary of Ancient Rome, Baltimore-London, 1992.

Chapter 3
Back to contents Book III

Suggestions and remarks?
don't hesitate to send me a message