In considering the difference between two European Bank Rates as the cause of a transfer of funds between the two centres, the cost of remittance, as measured by the difference between the telegraphic rate of exchange outwards at the beginning of the transaction and the telegraphic rate of exchange back at the end of it, is not, of course, to be neglected. But where the two centres are near together and there is no reason to anticipate the suspension of a free market in gold, this cost is, relatively, a minor consideration. The great distance, however, between London and India makes it in their case a very significant quantity, and a brief calculation shows that, measured in terms of Bank Rate, the cost of remittance works out higher, perhaps, than uninstructed common sense would anticipate. For, under present conditions, the cost of remittance both ways can hardly be less than 1/16d. per rupee, rising in most years as between certain dates as high as 5/32d., and reaching occasionally as much as 3/16d. It would not be prudent to act on the expectation of a less cost than 3/32d. Now 3/32d. on a rupee is about ·6 per cent. If this loss on exchange (_i.e._ on remittance) is to be recouped in three months (_i.e._ in a quarter of a year), an additional rate of nearly 2½ per cent per annum must be earned in India as compared with the rate in London. If a different degree of loss in exchange is anticipated, and if the length of time for which money can be used in India at a high rate is expected to be more or less than three months, the calculation must be adjusted accordingly. In any case the reason why the Indian and London Bank Rates can differ from one another for short periods by large amounts is adequately explained. If, for example, money can be employed in India at the high rate for one month only, even if the double cost of remittance for that period is so low as 1/16d., the difference between the London and Indian rates must amount to 5 per cent per annum to make a transfer of funds _prima facie_ profitable. These illustrations show that what seems a very small fluctuation in exchange can account for a very wide difference in the rate of discount; and, apart from questions of unequal knowledge and unequal security, it is this possibility of fluctuation that makes distinct markets of the two centres. The underlying explanation is essentially the same as that of the circumstance to which I called attention in § 9 of Chapter II., namely, that a temporary premium of ¾ per cent on gold in those European countries where gold is not always freely obtainable, is as effective as a very great increase in the Bank Rate in preventing the remittance of funds abroad and even in attracting an inward flow of funds.