satisfactory, a correct copy of them was made for record, of which I prepared a very full abstract in English to provide against possible accident to the native returns. The payments due to the agents employed were made in my presence and into their own hands.
One source of error to which the results are liable is inseparable from the nature of the investigation. I was instructed that the only mode in which the desired information should be sought must be with the full consent and good will of the parties with whom I might come into communication, and that the employment of authoritative or compulsory means was to be avoided. I was fully disposed to act up to these instructions, which were indeed given at my own suggestion and were dictated by the obvious spirit and intent of the inquiry. Adherence to them, however, made me and my agents dependent on third parties for the correctness of certain details; for instance, the number of persons, male and female, of the teachable age in a family. It was, of course, not permitted to enter the houses and count the females or the children, and on these and similar points the statements of heads of families and of the headmen of villages were necessarily received; but in such cases there was generally a check against inaccuracy by the presence of many of the villagers whose curiosity drew them together to listen, and who often corrected each other in the answers that were made. Notwithstanding this partial check, the discrepancy in the returns of males and females between fourteen and five years of age, that is, the much less number of females than of males of that age, seems to prove that concealment was systematically practised. I cannot adequately account in any other way for the difference that exist in the returns, and which will afterwards more fully appear.
Another source of error belongs to the plan of employing agents under me to collect information. I have already explained how I was induced to adopt this plan; and I am satisfied that by means of it the inquiry has been made far more extensive in its scope, and probably even more complete and accurate in its details, than if I had attempted to see every thing with my own eyes and do every thing with my own hands. The efficiency of such agency must depend on the efficiency of the supervision to which it is subjected; but although I laboured to render my superintendence vigilant and searching, and although I believe that the returns I received are in general worthy of confidence as far as they go, yet I have no security that they are not defective. In traversing a district, my agents could not visit all the villages it contained, amounting to several thousands. This was physically impossible without protracting the inquiry beyond ail reasonable limits. They were, therefore, compelled to depend either upon their personal knowledge, or upon the information that could be gathered from others as to the places possessing schools, every one of which was
