320 PHYSIOGRAPHY OF CENTRAL-ASIAN DESERTS AND OASES. 
but they may have been graves. Our sections appear to indicate the following 
sequence of events: 
(1) Uplift of spur and formation of Djillan-ooti valley. 
(2) Alluviation of wide flood-plain of rush-marshes, loess, and alluvium. 
(3) (a) Construction by man of a canal bringing in Zerafshan water with 
gravels; (b) building of mounds. 
(4) Drying up of that canal and shrinkage of Djillan-ooti water, with forma- 
tion of present channel and accumulation of 3 feet of loess on mounds, 
| 

Fig. 484.—The Citadel of Hissar. 
ANAU. 
PECULIARITIES OF ANAU AS AN OASIS OF TYPE I 8. 
(Plate 65.) 
To us it is of first importance to know what we can of the physiography of 
Anau, in whose ancient oases our shafts and excavations sank through no less 
than 10,000 years of man’s stratified débris. And though its geographical position 
is fully set forth in the beginning of this volume, it remains to sketch a few of its 
type-peculiarities. 
Anau is our best, or most familiar, example of an oasis of type Ib (delta-oases 
of small streams). This type, it may be remembered, is characteristic of the 
border of the plains where small silt-laden streams discharge from the mountains 
and are to best advantage diverted for irrigation. It is a type less exposed to 
sandstorms and overwhelming dunes than type Ia, like those of Merv, and yet 
more open to invasion by man, who may descend from the mountains or migrate 
along their base from oasis to oasis. It is, moreover, a type practically fixed in 
position, as contrasted with the inevitable shifting of type Ia through long- 
continued change of climate or rearrangement of distributaries. The oasis of Anau 
has for some 10,000 years remained so fixed that cultivation is still carried on 
over fields that bury its most ancient ruins. The greater oasis of Merv has in 100 
years changed place by 15 miles. 
