Respecting the division of the works
which are inside the walls and their arrangement
so that the noxious breath of the wind
may be avoided

1. Moenibus circumdatis secuntur intra murum arearum divisiones platearumque et angiportuum ad caeli regionem directiones. Dirigentur haec autem recte, si exclusi erunt ex angiportis venti prudenter. Qui si frigidi sunt, laedunt; si calidi, vitiant; si umidi, nocent. Quare vitandum videtur hoc vitium et avertendum, ne fiat quod in multis civitatibus usu solet venire. Quemadmodum in insula Lesbo oppidum Mytilenae magnificenter est aedificatum et eleganter, sed positum non prudenter. In qua civitate auster cum flat, homines aegrotant; cum corus tussiunt; cum septentrio, restituuntur in salubritatem, sed in angiportis et plateis non possunt consistere propter vehementiam frigoris.


1. When the walls are set around the city, there follow the division of the sites within the walls and the layings out of the broad streets and the alleys with a view to aspect. These will be rightly laid out if the winds are carefully shut out from the alleys. For if the winds are cold they are unpleasant; if hot, they infect, if moist, they are injurious. Wherefore this fault must be avoided and guarded against, lest there happen what in many cities is not infrequent. For example in the island of Lesbos, the town of Mytilene is magnificently and elegantly built, but not situated with prudence. For in this city when the South wind blows men fall ill; when the Nort-west, they cough; when the North, they are restored to health; but they cannot stand in the alleys and streets because of the vehemence of the cold.

2. Ventus autem est aeris fluens unda cum incerta motus redundantia. Nascitur cum fervor offendit umorem et impetus factionis exprimit vim spiritus flatus. Id autem verum esse ex aeolis aeris licet aspicere et de latentibus caeli rationibus artificiosis rerum inventionibus divinitatis exprimere veritatem. Fiunt enim aeoli pilae aereae cavae, - hae habent punctum angustissimum - quae aqua infunduntur conlocanturque ad ignem, simul autem ut fervere coeperint, efficiunt ad ignem vehementem flatum. Ita scire et iudicare licet e parvo brevissimoque spectaculo de magnis et inmanibus caeli ventorum naturae rationibus.


2. Now the wind is a wave of air flowing with uncertain currents of motion. It rises when heat strikes moisture and the onrush of the force presses out the power of the breath of the blast. That this is true we may see from Aeoluses of bronze, and by the craftsman's invention of things which express the truth of the divinity, about the causes which lurk in the heavens. Now figures of Aeolus are made of hollow bronze, and they have a very narrow point. These are filled with water and placed on the fire; before they begin to warm, they have no rush of air, but as soon as they begin to boil, they produce on the fire a vehement blast. Thus we may know and judge, from this small and very brief spectacle, about the great and immense causes of the nature of the sky and the winds.

3. Exclusi fuerint; non solum efficient corporibus valentibus locum salubrem, sed etiam si qui morbi ex aliis vitiis forte nascentur, qui in ceteris salubribus locis habent curationes medicinae contrariae, in his propter exclusiones ventorum temperatura expeditius curabuntur. Vitia autem sunt, quae difficulter curantur in regionibus, quae sunt supra scriptae, haec: gravitudo arteriace, tussis, pleuritis, pthisis, sanguinis eiectio et cetera, quae non detractionibus sed adiectionibus curantur. Haec ideo difficulter medicantur, primum quod ex frigoribus concipiuntur, deinde quod defatigatis morbo viribus eorum aer agitatus est, ventorum agitationibus extenuatur, unaque a vitiosis corporibus detrahit sucum et efficit ea exiliora. Contra vero lenis et crassus aer qui perflatus non habet neque crebras redundantias, propter inmotam stabilitatem adiciendo ad membra eorum alit eos et reficit, qui in his sunt inpliciti morbis.


3. Suppose they are excluded. Not only will this render a place healthy for sound persons; but also if any diseases shall happen to arise from other infections, those who in other healthy places find cure from counteracting medicine, in these, on account of the moderate climate and by the exclusion of the winds, will be still more quickly cured. For the diseases which are cured with difficulty in the regions which are described above are these: cold in the windpipe, cough, pleurisy, pthisis, spitting of blood, and others which are cured by strenthening remedies rather than by purgings. Tese ailments are treated with difficulty, first because they are caught from chills, secondly because when the strength is worn out by disease the air is agitated; it is thinned by the agitation of the winds; at the same time it draws the sap from diseased persons and renders them thinner. On the other hand, a smooth and thick air which is free from the passage of draughts and does not move backwards and forwards, builds up their limbs by its steadiness, and so nourishes and refreshes those who are caught by these diseases.

4. Nonnullis placuit esse ventos quattuor: ab oriente aequinoctali solanum, a meridie austrum, ab occidente aequinoctali favonium, ab septentrionali septentrionem. Sed qui diligentius perquisierunt, tradiderunt eos esse octo, maxime quidem Andronicus Cyrrestes, qui autem exemplum conlocavit Athenis turrem marmoream octagonon et in singulis lateribus octagoni singulorum ventorum imagines excalptas contra suos cuiusque flatus designavit, supraque eam turrim metam marmoream perficit et insuper Tritonem aereum conlocavit dextra manu virgam porrigentem, et ita est machinatus, uti vento circumageretur et semper contra flatum consisteret supraque imaginem flantis venti indicem virgam teneret.


4. Some have held that there are four winds: the Solanus from the equinoctial east, the Auster from the south, Favonius from the equinoctial west, and Septentrio from the north. But those who have inquired more diligently lay down that there are eight: especially indeed Andronicus of Cyrrha, who also, for an example, built at Athens an octagonal marble tower, and, on the several sides of the octagon, han representations of the winds carved opposite their several currents. and above that tower he caused to be made a marble upright, and above he placed a bronze Triton holding a rod in his right hand. He so contrived that it was driven round by the wind, and always faced the current of air, and held the rod as indicator above the representation of the wind blowing.


It is clear that vitruvius gives here a description of the socalled tower of the winds on the Athenian Agora. It is an octagonal building with a diameter of ± 7,8 m and a height of ± 14,30 m. It was probably the first part of the building program of Julius Caesar for the Roman agora at Athens, begun just after the battle of Pharsalus in 48 B.C. The building is erected in marble. As architect Vitruvius mentions Andronicos Cyrrhestes. On each side was a sculpture in relief representing the winds blowing from the quarter facing it. The roof is of 24 triangular marble slabs, supporting each other by means of their radiating joints, and carved on their upper surfaces in the form of roof tiles. On the top of the roof was an octagonal Corinthian capital as a finial, supporting a huge bronze Triton working on a pivot, with a rod in his hand which pointed toward the figure representing the quarter in which the wind lay. Also on the faces of the tower were sundials, and inside a water clock or clepsydra. There were two entrances, formed by two small distyle porches on two of the sides of the tower with architraves and pediments.

5. Itaque sunt conlocati inter solanum et austrum ab oriente hiberno eurus, inter austrum en favonium ab occidnete hiberno africus, inter favonium et septentrionem caurus, quem plures vocant corum, inter septentrionem et solanum aquilo. Hoc modo videtur esse expressum, uti capiat numerus et nomina et partes, unde flatus certi ventorum spirent. Quod cum ita exploratum habeatur, ut inveniantur regiones et ortus eorum, sic erit ratiocinandum.


5. Therefore there are placed between the Solanus and the Auster, the Eurus from the winter sunrising; between the Auster and the Favonius, the Africus from the winter sunset; between the Favonius and the Septentrio the Caurus (which most people call Corus); between the Septentrio and the Solanus, the Aquilo. The diagram seems to be so arranged as to receive the names and the quarters whence the fixed currents of winds blow. Since they may be regarded as ascertained, we must calculate as follows to find the quarters and risings of the winds.

6. Conlocetur ad libellam marmoreum amusium mediis moenibus aut locus ita expoliatur ad regulam et libellam, ut amusium non desideretur, supraque eius loci centrum medium conlocetur aeneus gnomon, indagator umbrae qui graece sciotheres dicitur. Huius antemeridiana hora circiter hora quinta sumenda est extrema gnomonis umbra et puncto signanda deinde circino diducto ad punctum, quod est gnomonis umbrae longitudinis signum, ex eoque a centro circumagenda linea rotundationis. Itemque observanda postmeridiana istius gnomonis crescens umbra, et cum tetigerit circinationis lineam et fecerit parem antemeridianae umbrae postmeridianam, signanda puncto.


6. Let there be placed to a level a marble dial, somewhere in the middle of the city; or let a space be so polished to rule and level that the marble dial is not wanted. Above the middle point of that place, let there be put a bronze indicator to track the shadow (which in Greek is called sciotheres. Before midday, at about the fifth hour, the end of the shadow of the indicator is to be taken and marked with a point. Then a radius being taken from the indicator to the point which marks the length of the shadow, with that, from the indicator as the centre, a circumference is to be drawn. Aften midday the growing shadow of the indicator, when it touches the line of the circle and marks a post-meridian shadow equal to the ante-meridian, is to be marked with a point.

7. Ex his duobus signis circino decuatim describendum, et per decusationem et medium centrum linea perducenda ad extremum, ut habeatur meridiana et septemntrionalis region. Tum postea sumenda est sexta decima pars circinationis lineae totius rotundationis, centrumque conlocandam in meridiana linea, qua tangit circinationem, et signandum dextra ac sinistra in circinatione et meridiana et septentrionali parte. Tunc ex signis his quattuor per centrum medium decusatim lineae ab extremis ad extremas circinationes perducendae. Ita austri et septentrionis habebitur octavae partis designatio. Reliquae partes dextra ac sinistra tres, aequales et tres his distribuendae sunt in tota rotundatione, ut aequales divisiones octo ventorum designatae sint in desciptione. Tum per angulos inter duas ventorum regiones et platearum et angiportorum videntur deberi dirigi descriptiones.


7.From these two points, two intersecting circles are to be described. Through the intersection and the centre of the circle first described, a line is to be carried through to the end so that the southern and northern quarters may be indicated. Next we take a radius the sixteenth part of the circumference of the circle. From the centres given by the meridian line at the two points where it touches the circle, and with that radius, points are to be marked rigth and left in the circle, both on the southern and on the northern part. Then from these four points, intersecting lines are to be drawn through the middle centre from one side of the circumference to the other. Thus both for the south wind and for the north wind we shall have marked out the eighth part of the circumference. the remaining parts in the whole round, three on the right and three on the left, are to be distributed equally, so that equal divisions of the eigth winds are marked out in the figure. Then the angles between two quarters of the winds will determine the laying out of both of the streets and of the alleys.

8. His enim rationibus et ea divisione exclusa erit ex habitationibus et vicis ventorum vis molesta. Cum enim plateae contra derectos ventos erunt conformatae, ex aperto caeli spatio impetus ac flatus frequens conclusus in faucibus angiportorum vehementioribus viribus pervagabitur. Quas ob res convertendae sunt ab regionibus ventorum derectiones vicorum, uti advenientes ad angulos insularum fragantur repulsique dissipentur.


8. For by these methods and this division, troublesome winds will be excluded from the dwellings and the streets. For when the quarters of the city are planned to meet the winds full, the rush of air and the frequent breezes from theopen space of the sky will move with mightier power, confined as they are in the jaws of the alleys. Wherefore the directions of the streets are to avoid the quarters of the winds, so that when the winds come up against the corners of the blocks of buildings they may be broken, driven back and dissipated.

9. Fortasse mirabuntur i qui multa ventorum nomina noverunt, quod a nobis expositi sunt tantum octo esse ventis. Si autem animadverterint orbis terrae circuitionem per solis cursum et umbras gnomonis aequinoctialis ex inclinatione caeli ab Eratosthene Cyrenaeo rationibus mathematicis et geometricis methodis esse inventam ducentorum quinquaginta duum milium stadium, quae fiunt passus trecenties et decies quinquies centena milia, huius autem octava pars quam ventus tenere videtur, est triciens nongenta triginta septem milia et passus quingenti, non debebunt mirari, si in tam magno spatio unus ventus vagando inclinationibus et recessionibus varietates mutatione flatus faciat.


9. Perhaps those who know many names of the winds, will wonder because only eight winds have been described by us to exist. But if they perceive that the circumference of the world, ascertained by the sun's course, and the equinoctial shadows of the gnomon and the inclinations of the sky, have been found by Eratosthenes of Cyrene with mathematic calculations and geometric methods to be 252.000 stades, which give 31.500.000 paces, while of this the eighth part which the wind seems to occupy is 3.937.500 paces, they ought not to wonder, if in so great a space one wind, as it moves with its inclinations and retreats, causes varieties through the change of its current.

10. Itaque dextra et sinistra austrum leuconotus et altanus flare solet, africum libonotus et subvesperus, circa favonium argestes et certis temporibus etesiae, ad latera cauri circias et corus, circa septentrionem thracias et gallicus, dextra ac sinistra aquilonem supernas et caecias, circa solanum carbas et certo tempore ornithiae, euri vero medias partes tenentis in extremis euricircias et volturnus. Sunt autem et alia plura nomina flatus que ventorum e locis aut fluminibus aut montium procellis tracta.


10. Therefore on the right and left of Auster, Leuconotus and Altanus are wont to blow; of Africus, Libonotus and Subvesperus; around Favonius, Argestes and at certain times the Etesian winds; at the sides of Caurus, Circias and Corus; about Septentrio, Thracias and Gallicus; rigth and left of Aquilo, Supernas and Caecias; around Solanus, Carbas and at a definite time Ornithiae; on the distant parts, when Eurus holds the middle, Euricircias and Volturnus. There are also many other names and breezes of winds, drawn from places or rivers, or from mountain storms.

11. Praeterea aurae matutinae, qua sol, cum emergit de subterranea parte, versando pulsat aeris umorem et impetu scandendo prudens exprimit aurarum antelucano spiritu flatus. Qui cum exorto sole permanserunt, euri venti tenent partes, et ea re, quod ex auris procreatur, ab Graecis euros videtur esse appellatus, crastinusque dies propter auras matutinas aurion fertur esse vocatus. Sunt autem nonnulli qui negant Eratosthenem potuisse veram mensuram orbis terrae colligere. Quae sive est certa sive non vera, non potest nostra scriptura non veras habere terminationes regionum, unde spiritus ventorum oriuntur.


11. Moreover there are morning airs, when the sun, emerging from the subterranean part, tosses and beats the damp in the air, and rising with a rush looks forward and thrusts forth the breezes with the breath that comes before light. And when these have remained after sunrise, they hold the region of the east wind. Because this is generated from aurae (breezes) it seems to be called euros by the Greeks, and because of morning breezes the morrow is said to have been called aurion. But there are some who deny that Eratosthenes could infer the true measure of the earth. Whether this is certain or not, our writing cannot fail to furnish true outlines of the regions whence arise the breezes of the winds.

12. Ergo si ita est, tantum erit, uti non certam mensurae rationem sed aut maiores impetus aut minores habeant singuli venti.

Quoniam haec a nobis sunt breviter exposita, ut facilius intellegatur, visum est mihi in extremo volumine formas sive ut Graeci schemata dicunt, duo explicare, unum ita deformatum, ut appareat, unde certi ventorum spiritus oriantur, alterum, quemadmodum ab impetu eorum aversis derectionibus vicorum et platearum evitentur nocentes flatus. Erit autem in exaequata planitie centrum, ubi est littera A, gnomonis autem antemeridiana umbra, ubi est B, et a centro, ubi est A, diducto circino ad id signum umbrae, ubi est B, circumagatur linea rotundationis. Reposito autem gnomone ubi antea fuerat, expectanda est, dum decrescat faciatque iterum crescendo parem antemeridianae umbrae postmeridianam tangat lineam rotundationis, ubi erit littera C. Tunc a signo, ubi est B, et a signo, ubi est C, circino decusatim describatur, ubi erit D; deinde per decusastionem et centrum, ubi est D, perducatur linea ad extremum, in qua linea erit littera E et F. Haec linea erit index meridianae et septentrionalis regionis.


12. Therefore if it is so, it will have this consequence, that the several winds will have, not a fixed and measured amount, but either greater or less impetus.
Since these matters have been briefly set forth by us, in order that it may be more easily understood I have decided at the end of the book to furnish two plans, or as the Greeks say schemata: one so mapped out that it may appear whence the certain breezes of the winds arise; the second, how by layings out of quarters and streets turned away from their violence, dangerous currents may be avoided. Now there shall be on a levelled surface a centre with the letter A; the shadow before midday of the indicator, with B; and from the centre marked A the compass is opened to the point of shadow marked B, and a circle is to be drawn. The indicator being replaced where it was before, we must wait until the shadow diminishes, and again by increasing makes the shadow after midday equal to that before midday and touches the circle at the letter C. Then from B and from C let the intersection D be described with the compasses; then through the intersection D and the centre, let a line be carried through to the furthest limit, where will be the letter E and also F, and on this line will be the index of the southern and northern regions.

13. Tunc circino totius rotundationis sumenda est pars XVI, circinique centrum ponendum est in meridiana linea, qua tangit rotundationem, ubi est littera E et signandum dextra sinistra, ubi erunt litterae G H. Item in septentrionali parte centrum circini ponendum in rotundationis et septentrionali linea, ubi est littera F et signandum dextra ac sinistra, ubi sunt litterae I et K, et ab G ad K et ab H ad I per centrum lineae perducendae. Ita quod erit spatium ab G ad H, erit spatium venti austri et partis meridianae: item quod erit spatium ab I et K, erit septentrionis. Reliquae partes dextra tres ac sinistra tres dividendae sunt aequaliter, quae sunt ad orientem, in quibus litterae L M , et ab occidente, in quibus sunt litterae N et O. Ab M ad O et ab L ad N perducendae sunt lineae decusatim. Et ita erunt aequaliter ventorum octo spatia in circumitionem. Quae cum ita descripta erunt, in singulis angulis octagoni, cum a meridie incipiemus, inter eurum et austrum in angulo erit littera O, inter austrum et africum H, inter africum et favonium N, inter favonium et caurum O, inter caurum et septentrionem K, inter septentrionem et aquilonem I, inter aquilonem et solanum L, inter solanum et eurum M. Ita his confectis inter angulos octagoni gnomon ponatur, et ita dirigantur angiportorum divisiones.


13. Then the sixteenth part of the whole circle is to be taken with the compass, and the point of the compass is to be put on the meridian line where it touches the circumference at E, and a mark is tobe made right and left at G H. Also in the northern part, the point of the compass is to be placed on the circumference and the northern line where is the letter F, and a mark is to be made right and left at I and K. And from G and K and from H to I, lines are to be drawn through the centre. So the space from G to H will be the space of the Auster and of the southern region; likewise the space from I to K will be of the Septentrio. the remaining parts, on the right three, and the left three, are to be divided equally; those which are to the east at L and M, and at the west at N and O. From M to O and from L to N intersecting lines are to be drawn. And so there will be eight equal spaces of winds in the circumference? When these are so marked out, at the single angles of the octagon when we begin from the south, in the angle between Eurus and Auster there will be G, between Auster and Africus there will be H, between Africus and Favonius N, between Favonius and Caurus O, between Caurus and Septentrion K, between Septentrio and Aquilo I, between Aquilo and Solanus L, between Solanus and Eurus M. When these things are done, let the gnomon be set upon the angles of the octagon and let the division of the alleys be directed accordingly.


In this chapter Vitruvius describes a completely hypothetical layout of a city. His main purpose is to give the streets a correct direction in order to avoid the winds to blow freely through the streets. He discerns 8 wind directions; in this he follows Andronikos of Cyrrha who also built de tower of the winds at Athens.
To obtain this exact direction he uses a rather difficult method (compasses didn't exist in antiquity). This can be described in 8 phases:

  1. On a flat surface in the centre of the future town, a gnomon is installed. Some time before noon the shadow of this gnomon is measured, which gives point B.
  2. A circle is drawn for which the distance A-B is the radius.
  3. When, later in the afternoon, the shadow of the gnomon touches the circonference of the circle a new point C is marked.
  4. The distance B-C is divided in two equal parts. This gives point D.
  5. A line is drawn from C to A. This must give exactly the north - south direction.
  6. In the circle an octogon is drawn in such a way that one side of the octogon is diveded in two by the north - south axis.
  7. In the octogon a square is inscribed in such a way that the diagonals correspond with every second angle of the octogon.
  8. The square is filled up with a grid which will correspond to the directions of the streets
The result of this method is a grid of which none of the main streets is oriented to one of the eight wind directions, the deviation of the grid against the north being 22,5°.
Of course this is a purely theoretical approach. None of the cities in antiquity was oriented in that way, due to the topographical possibilities and disposition of each site.

It is not certain whether Vitruvius meant that the street grid had to fill up the complete surface of the circle or octogon. Nor is it certain whether the circle or octogon corresponds with the city wall. Anyway, many antique cities are found where not the complete surface within the walls was build up. The function of the city wall was not alone to defend the town but was also the delimitation of a wider area appartaining to the city.


Les dix livres d'architecture de Vitruve, Corrigés et traduits en 1684 par C. Perrault, Paris, 1684.
Vitruvius, De Architectura libri X, ed. F. Granger, London, 1962.
Ton Peters, Vitruvius, Handboek bouwkunde, Amsterdam, 1999.
H.Knell, Vitruvs Architekturtheorie, Darmstadt, 1991.

Chapter 7
Back to contents Book I

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