428 BUREAU OF AMERICAN ETHNOLOGY [Bull. 187 



Beverley wrote some years after white settlements had been started, 

 this might have been an innovation. 



DOORS 



Few Southeastern houses had more than one door except among 

 the Powhatan tribes of Virginia, where there were usually two in 

 the longer houses, one at the middle of each end (Strachey, 1849, p. 

 71). According to Garcilaso de la Vega (1723, p. 52), the house of 

 the chief of Ochile in central Florida was very large and had doors 

 toward the four cardinal points, but this seems to have been excep- 

 tional, though in fact our information regarding Timucua houses 

 is very meager. Bartram (1909, p. 56) figures a Cherokee summer 

 house with two doors, but this represents a later style and, besides, 

 it is to be noted that the rooms to which these doors gave access 

 did not communicate and were of the nature of two houses set side 

 by side. Ordinarily, however, particularly in descriptions of the 

 houses along the lower Mississippi, one door is assumed. Garcilaso 

 (1723, p. 153) says specifically of the houses of Mabila that they had 

 but one door. Among the Chickasaw and Caddo this door was 

 usually turned toward the east (Adair, 1775, p. 176; Hatcher, 1928, 

 p. 52; Swanton, 1942, p. 152), but sometimes toward the southeast or 

 south, and this represents the commonest orientation throughout the 

 Gulf region, the west being the bad luck quarter. Yet it is probable 

 that circumstances altered cases, topography overriding superstition 

 or causing a reinterpretation of it. It is not likely that houses on the 

 east bank of a river fronted away from it, though this may well have 

 been true of the ceremonial houses. The door of the Natchez temple 

 was plainly east, but we are told that the door of the Head Chief's 

 house nearby was north (Swanton, 1911, pp. 59, 162). This, how- 

 ever, may have been due to an unwillingness to have it open entirely 

 away from the east and entirely away from the temple. Bushnell 

 (1909, p. 7) was informed that, in most cases, the doors of houses 

 occupied by the Choctaw of Bayou Lacomb in olden times faced south. 

 The winter houses of the Chickasaw, and presumably those of the 

 Choctaw, had short passageways before the door as a protection 

 against drafts and enemies (Adair, 1775, p. 420). The same sort of 

 approach was used before the tcokofa of the Creeks and the town 

 house of the Cherokee. It was possibly an extension or modification 

 of this sort of approach which is described as the dwelling of the keep- 

 ers of the temple among the Taensa and certain other tribes (Swan- 

 ton, 1911, p. 269). 



No De Soto chronicler indicates the materials of which Indian doors 

 were made except that the Fidalgo of Elvas refers to "a door with a 

 grating" at Mabila (Robertson, 193^, p. 133). In 1595 San Miguel 



