588 GREEK HORIZONTAL CURVES IN MAISON CARREE AT NIMES. 



It is generally conceded that Vitruvins drew his matter from earlier 

 Greek authors whose works have perished, and that he <lid not always 

 fully comprehend the ideas of his sources. I have omitted any argu- 

 ment concerning his direction that the stylobate curves are to prevent 

 an effect of "alveolation" (i. e., downward deflection) at the center of 

 the stylobate. The only modern author who has attempted to explaiii 

 this direction by optical theories is Thiersch. This author gives his 

 reasons for supposing that a spectator standing near an angle of the 

 stylobate and below the level of its platform might experience an 

 optical effect of downward deflection in the lines of the stylobate which 

 an upward curve would correct, but masmuch as a bulging curve in 

 plan could not correct this effect for the standpoint near the angle, I 

 have not considered his theory in tins i)ai)er, and 1 only mention it as 

 giving' one more illustration of the new light thrown on the Greek 

 refinements by the discovery of curves in horizontal planes. There are 

 very valuable remarks m Boutmy's work as to the general unreliability 

 of Vitruvins for a comprehension of the Greek curves, and one pnrpose 

 of this i>aper is to accent the value of Boutmy's contribution to the 

 philosophy of Greek architectuie. I lis work also contains quotations 

 from Greek authors on the oi)tics of architecture, showing that inten- 

 tional optical illusions and intentional optical corrections were alike 

 familiar to them. 



There is one thing more to be said before 1 close. The credit for the 

 original suggestion that there is a historic connection between the 

 Greek curves and those of Medinet Habou belongs to Mr. Pennethorne, 

 as does the credit for both discoveries. The wholly original part of 

 this paper as regards historic facts is that whi(;h i)oints to the fact that 

 two classic buildings — one early Greek atPaestum and one latelioman 

 at Nimes — show convex curves in plan which are identical in character 

 with the curves in Egypt. The wholly originsil part of this paper as 

 regards observations is that which relates to Nimes, Karnak, Luxor, 

 and Edfou. The wholly original part of this paper as regards the 

 effect of the Greek horizontal curves is that which shows the optical 

 results in actual historic buildings of convex curves in horizontal 

 l)lanes. I am willing to leave the question of purpose to the expert 

 and to the general reader. 



