column immediately to the right of it, exactly as in our own system of notation by means of Arabic numerals. Any sum in arithmetic can be done on the abacus, even to the extracting of square and cube roots; and Dr. Knott, the chief English—or, to be quite correct, the chief Scotch—writer on the subject, is of opinion that Japanese methods excel ours in rapidity. Perhaps he is a little enthusiastic. One can scarcely help thinking so of an author who refers to a new Japanese method of long division as "almost fascinating." The Japanese, it seems, have not only a multiplication table, but a division table besides. We confess that we do not understand the division table, even with Dr. Knott's explanations. Indeed we will confess more: we have never learnt the abacus at all! If we recommend others to learn it, it is because we hope that, for their own sake, they will do as we tell them and not do as we do. Personally we have found one method of ciphering enough, and a great deal more than enough, to poison the happiness of one life-time.
The use of the abacus is not the only peculiarity of this nation in matters numerical. A more irritating one to the accurate European mind is their habit of "inclusive" reckoning. An example or two will best make this clear. You arrived in April, say. It is now June. According to the Japanese, you have been here three months;—for the month of your arrival and the present month are both counted in. A child is born in December, 1901. By January, 1902, they currently talk of the child as being two years old, because it has lived through a part of two separate years. The thing may be exaggerated a degree further still, when, forgetting that the Japanese year formerly began some time in our February (as the Chinese year still does), they fail to make allowance for this in the case of births that took place in January or early February, previous to the reform of the calendar in 1873. In the case of a public man who died early in 1901, at the (real) age of 65, we noticed that all the obituaries were wrong by three years. They credited him with being 68; and of the various other dates mentioned, some were two years out, some three, according
