42 CHICHEN ITZA. 



with a clotted circle. Between the south end of No. 26 and the commencement of the 

 ■wall at " in" there is little debris lying amongst the overturned columns, and this part 

 may have been roofed with wood and plaster. 



This colonnade runs on, broken only where the absence of four columns makes a 

 sort of courtyard or patio at "«," until it merges into the square colonnade No. 28, 

 which is enclosed on three sides by a wall about 2 feet 3 inches thick. 



In the centre of this square colonnade, also, the absence of four columns forms a 

 sort of courtyard "o," and there is again a break in the symmetry at "p," where walls 

 can be traced enclosing two columns out of line with the others. 



To the east of No. 28 was a small group of houses with columns which I regret that 

 I had not time to measure. 



From the north-west corner of No. 2S a long mound, No. 29, extends in a westerly 

 direction. It was approached by a stairway in the middle of the north side, which led 

 to a terrace, and at the back of this terrace were traces of four double-chambered 

 houses. Between these houses is a passage (q) leading into a courtyard, No. 30 

 (see Plate LXVL, a). 



The floor of this court is about 5 feet higher than the ground outside the walls. 

 The enclosing wall, seen from the outside in Plate LXVL, b, in one place measures 

 18 feet in height from the floor of the court, but is nowhere entire. There is no sign 

 of any other entrance to the court except that already mentioned at " q." The square 

 depression in the centre of the court, marked with a dotted line, was surrounded by 

 columns which measured 14 feet in height, including the capital. The corner columns 

 are the larger, each measuring 2 feet 2 inches across ; the diameter of the others is 

 1 foot 9 inches. All the columns have fallen. There is not much debris in the court, 

 and the height of the columns, their distance from the walls (24 feet), and the fact 

 that the wall where it is 18 feet in height shows no signs of the spring of a vault, 

 altogether precludes the possibility of those columns having supported a stone roof. 



Although I have already stated that it was very probable that some of the colonnades 

 were roofed with wood and cement, No. 30 affords the only example I have met 

 with where it can be confidently asserted that some material other than stone must 

 of necessity have been used for roofing. It seems most probable that wooden archi- 

 traves connected the columns, and that beams sloping downwards from the outside 

 walls rested on these architraves, the central patio being left open. The interstices 

 between the beams were probably filled with rubble and cement, as in the modern 

 houses in the Spanish-American towns. 



Stone-roofing of the type invariably found in Central-American ruins must have 

 presented great difficulties to the builders when a large space had to be covered with 

 columns alone as a support. 



