Nature of Variations : Units 1 (4 ha) and 2 (10 ha) before correction 
can in effect be considered as inclusions in the soil areas lying 
respectively to the south and west. As such this cannot be considered 
as a serious error; also the difference is only in the sub group qualifi 
cation, namely "Alfic" and "Udic" Chromusterts. 
The soil in unit 3 (23 ha) was found to be 'Fine Alfic Chro- 
musterts' as against the predicted 'Clayey Udic Haplustalfs'. This can 
be considered as a rather serious error because the difference in Taxonomy 
is at the level of order itself. But it may be mentioned that the 
dominant soils in the plain areas are "Alfisols" and "Vertisols" and 
their intergrades, all being highly clayey; in such a situation the soils 
could be considered as near taxadjuncts. The case of unit 4 (3 ha) 
is very similar to unit 3, the only difference being that instead of the 
expected "Udic Haplustalfs", the actual soil was found to be "Udic 
Chromusterts" . 
Again in unit 5 (8 ha) , while the expected soil was one with 
a gravelly sub soil (Clayey skeletal Haplustalfs), the actual soil was 
gravelfree to a considerable depth (Clayey Haplustalfs); the extent is 
only a small percentage (1.8%) of the check area. 
Summarising, it may be stated that the variations from the 
photo interpretation map account in all for an area of only 48 ha or 
about 11% of the Intensive Check Area of 448 ha. Even within this, 
the error is not of the type or degree that can be avoided in the course 
of conventional detailed soil surveys using air photos as base maps. 
Comparison with Old Soil Map 
This survey was conducted in the year 1972; the same area 
had been covered by a detailed soil survey using very large scale 
cadastral maps as bases (Scale 1:396 0) about ten years earlier. These 
two soil maps in respect of three contiguous villages (total area 1316 
ha) were superimposed to identify the differences. This was done only 
for the purpose of illustrating with a practical example the differences 
in the soil maps prepared by the two procedures, as it is fully appre 
ciated that the comparison is not a 'fair' one. The photo-interpretation 
map for the 1316 ha had 10 mapping units and a total length of 54 km 
of soil boundaries against 23 and 36 respectively in the old map. More 
importantly, the photo interpretation procedure ensured consistent and 
accurate mapping of soil erosion, for which the cadastral map is a 
poor base. 
The plain region of the study area lacked significant relief 
differences. Even so the systematic photo-interpretation procedure 
was found to be efficacious. It can, therefore, be confidently stated 
that for mapping tracts with greater relief differences, the procedure 
would offer even greater advantages. 
Time and Cost 
It was found that the adoption of systematic photo interpre 
tation procedure helped in covering about twice the area normally 
covered by conventional surveys; costwise, in respect of field survey 
only, the expenditure of the photo-interpretation procedure was 1/3 
of the conventional method (I.A.R.I., 1970).