522 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1963 



Wadi Beihan, at Marib, in Dhofar, and at Sohar (Bowen and 

 Albright, 1958; Cleveland, 1959, 1960; Van Beek 1952, 1961). The 

 results of these expeditions were considerable. Chronology for the 

 period from about the 10th century B.C. to the 7th century A.D. was 

 largely fixed and much information on the culture history was dis- 

 covered, e.g., data on irrigation techniques and installations, and on 

 successive waves of cultural influence from other parts of the ancient 

 world. 



But enormous gaps in our knowledge remained. Little systematic 

 work had been done on prehistory, with the result that the develop- 

 ment and affinities of the early cultures of the region were largely 

 unknown. Further, no systematically collected evidence had come 

 to light of human occupation between the Middle Paleolithic stage — 

 represented by artifacts picked up by Miss Caton Thompson's team 

 and others — and the 10th century B.C. in Wadi Beihan, or the 5th 

 century B.C. in Wadi Hadhramaut. There were also many unan- 

 swered questions regarding the role of southern Arabia — if indeed it 

 played a role — in the diffusion of cultural traits, the development of 

 trade, the migration of man, and the interchange of cultivated plants 

 and domestic animals between Africa and Asia. 



To obtain data that would contribute to the solution of these prob- 

 lems, an intensive archeological reconnaissance in one of the major 

 drainage systems of southern Arabia was organized and directed by 

 Van Beek. After much consideration, Wadi Hadhramaut was se- 

 lected as the survey area, because it is one of the two or tliree most 

 fertile and intensively cultivated valley systems in southern Arabia 

 today, and presumably was throughout antiquity also. Such an area 

 is likely to have been continuously inliabited, and thus to contain 

 evidence for a complete culture sequence from earliest times to the 

 present. 



Wadi Hadhramaut is located in the East Aden Protectorate, ap- 

 proximately 165 kilometers north of the southern coast of the Arabian 

 Peninsula (pi. 1:1). It rmis roughly j)arallel to this coast for a 

 distance of about 200 kilometers (fig. 1), and then gradually turns 

 southeastward to empty into the Gulf of Aden. On the west, near 

 Qarn Qaimah, the wadi is approximately 15 kilometers wide, but 

 eastward its width gradually diminishes to little more than 2 kilo- 

 meters just beyond Tarim. Except for the last few kilometers of 

 its course where it is perennial, the upper reaches of the wadi are 

 dry. Occasionally a sell (Arabic for flash flood), formed of the 

 runoff from the plateau, covers part — rarely all — of the wadi floor. 

 At present, some areas of the main wadi and virtually all silt-covered 

 tributaries are cultivated by means of seil (flash -flood) irrigation. 

 Elsewhere, and especially in the main wadi where the water table 

 ranges from about 15 to 20 meters deep, well irrigation is extensively 



