70 ON THE OLDER FORMS OF 



Successive invasions of the Asiatic tile, in a measure, 

 supplanted the normal flat type which seemed at the out- 

 set to be associated with monumental buildings, though 

 this purely classic form has survived in the modern flat 

 type seen in Italy to-day. The circular disc closiug the 

 imbrex points distinctly to eastern Asia, and the subse- 

 quent decoration of the eaves and ridge tiles, while 

 strongly suggesting an Eastern origin, is no sure criterion, 

 as to whatever the Greeks touched they imparted a charm 

 derived from their own matchless instinct for the beautiful. 



It seems curious to see the antitixffi attached to the 

 eaves tiles, at Assos, as late as the Roman epoch, and yet 

 600 years before, at Selinus, these elements had already 

 become detached from the roofing-tiles and were indepen- 

 dent pieces, nailed to the top of the stone coping. 



The historical sequence in the development of the early 

 Grecian, Etruscan, Roman and Sicilian tile, and the source 

 of the first form — the norm as Graeber describes it — so 

 common in China to-day, must ultimately be cleared up. 

 The material is indestructible and the character of a frag- 

 ment, even, is easily recognized. 



It has been impossible to find data indicating, even ap- 

 proximately, the first appearance of the pan tile and the 

 flat tile, though it is probable that these data exist. 



The geographical distribution of these three types of tile 

 to-day is a matter easily ascertained and I venture to pre- 

 sent the following map of Europe (fig. 85) upon which 

 are indicated by conventional lines the regions where these 

 various forms occur. These lines represent the appear- 

 ances of the tiles in section and will be readily understood. 

 The single curved lines represent the normal tile, the lines 

 of double flexure the pan tile, and the short, straight lines 

 the flat tile. 



As the normal tile is almost universally distributed in 



