SOUTHWESTERN ARCHEOLOGY — ROBERTS 521 



been, it will be again and again." It has been the failure to consider 

 carefully such factors that has caused some students trouble in 

 properly evaluating their finds. 



As an illustration of the time element involved and in response 

 to oft-repeated queries concerning the age of ruins, a number of 

 writers have supplied dates for the various stages in the sequence. 

 These were not an integral part of the Pecos Classification, with 

 the exception of Pueblo V, and were not given with the idea of 

 isolating each stage between arbitrarily chosen sets of years because 

 there is no sharp break between periods. Insofar as possible, these 

 dates were based on information furnished by dendrochronology. 

 For the earlier stages, however, data from this source were not avail- 

 able and the figures were speculative. Most reports stressed this 

 factor and pointed out that there could be no hard and fast appli- 

 cation of the numerical chronology. A tendency has developed in 

 certain quarters to make these dates the horizon determinant and 

 ignore all the elements in the complex. A bare numerical tabula- 

 tion is not sufficient to make clear all of the ramifications of j)eriph- 

 eral lags and stage survivals. 



There are two peripheral precincts where the Basket Maker-Pueblo 

 pattern is not clear cut. In these outlying reaches many features, 

 common in the nuclear districts, are missing. On the other hand, 

 local developments have contributed elements which are foreign to 

 the central portions. These marginal regions are generally desig- 

 nated as the " northern and eastern peripheries." The northern 

 comprises the territory north and west of the Colorado River, rang- 

 ing along the western slopes of the Rocky Mountains into southern 

 Idaho and extending westward into eastern Nevada. The eastern 

 includes the country lying to the east of the Rio Grande drainage 

 and extends from the Oklahoma panhandle on the north through 

 western Texas to the Big Bend district on the south. The western 

 and eastern boundaries of the two peripheries, respectively, have not 

 been determined. 



The northern periphery is characterized by a progressive fading 

 of the basic pattern in proportion to the distance from the central 

 portions of the province. The general nature of the remains indi- 

 cates a Basket Maker Ill-Pueblo I origin for a complex which has 

 distinctive qualities resulting from a combination of factors. Among 

 these may be noted the survival of early elements, varying rates of 

 diffusion for important features in the main pattern, the synchronous 

 appearance of components which were chronologically distinct in the 

 nuclear districts, the adaptation of borrowed features to local needs, 

 and inventions. Except for a narrow strip along the Colorado River 

 in the southern part of the periphery where the pattern was closely 



