196 THE AGRARIAN SYSTEM OF MOSLEM INDIA 
to which he refers in general terms; I have not seen 
these, and I cannot trace any later reference to show whether 
or not any of them still exist. It is certain, however, that 
Grant's starting-point was wrong. His statement that Todar 
Mal made a detailed assessment of the province is his- 
torically impossible, as Shore pointed out, and it is directly 
at variance with the official record in the Ain, that Akbar 
maintained the method of assessment (nasaq) which he found 
in force; whether the word nasag denotes Group-assessment, 
or Farming, or both, it excludes the possibility of such a 
detailed assessment as Grant asserted. His statement 
that the basis of the assessment was one-fourth of the 
produce must also be incorrect, for in Todar Mal’s time the 
State’s claim was uniformly one-third; the figure of one- 
fourth was obviously derived from Grant's early studies of 
the Deccan assessment, which he was led to believe was a 
servile copy of Todar Mal’s work. Grant’s account, there- 
fore, cannot be accepted in its entirety, and the initial 
misapprehension affects the whole of his argument. 
In my opinion, the most probable reading of Grant's 
earlier figures is that the documents which he used referred 
to Valuation, not Demand. I have given in Appendix G 
my reasons for holding that the statistics in the Ain, for 
Bengal as for the other provinces, probably represent the 
Valuation in force at the time when the record was compiled. 
The Bengal figures, which Grant took as showing Todar 
Mal’s assessment of Demand, would on this view be in fact 
the first and summary Valuation of a newly acquired 
province made by Todar Mal, or under his orders, on the 
basis of whatever data were available at the time of an- 
nexation, probably the records maintained by the former 
Government. This view clears up the obvious difficulty 
that Todar Mal could not possibly have assessed in detail 
the Demand on those portions of eastern Bengal, which had 
not fallen into Akbar’s hands; it is easy to understand that, 
finding Chittagong, for instance, shown in the old records 
as still part of the kingdom of Bengal, he should have in- 
cluded it in the Valuation, pending the time when its pos- 
session should be obtained; while it is quite certain that, 
in this region. at least. he could not have carried out the