Of course, association is the best means of remembrance. We remember your words "Salle-à-manger" by _Sara-mongi_, that is, a "plate and written characters," and the English word "Minister" as _me-no-shita_, that is, "below the eyes." In fact, _me-no-shita_ is used very commonly in corrupt English at the open ports of Japan. _Frans-Me-no-shita_ is "French Minister," and _Igiris-Me-no-Shita_ is "English Minister." A dozen years ago there was in Japan an enterprising man who advertised that he had invented a good system of memory, and even opened a school. It was no other than remembering things by association, and I think there is a good deal in it. A little time ago, as you know, their Highnesses Prince and Princess Arisugawa were on a visit to Europe. Lord Lansdowne had great difficulty in remembering the name. Our _Me-no-shita_ in London asked him if there were not a Princess Alice in England, to which his Lordship replied "Yes." He then asked if there were not a street called "Gower Street," to which his Lordship similarly answered "Yes." Thereupon our minister said: "Very well, Princess Alice and Gower Street, that makes Arisugawa." After that his lordship, the Foreign Secretary of Great Britain, remembered the name of our Prince and Princess very well.' --'That's all very well,' said she: 'but you mix up _r_ and _l_ altogether.' --'That's true,' I answered, 'it is the weakest point for us in speaking European tongues. We cannot, or at least without the greatest difficulty, make any difference in pronouncing _r_ and _l_. Thus "right" and "light" become alike when we speak. It is very necessary to think of that fact when you talk with a Japanese. All this arises from the fact that in the Chinese and Japanese tongues there are not two different sounds of _r_ and _l_; those sounds in Japanese are more like your _ra_, _ri_, _ru_, _re_, _ro_, whilst they are _la_, _li_, _lu_, _le_, _lo_ in Chinese.' --'I have noticed it very often,' she said, 'even while you talk you do the same. Your allusion to remembering things by association is very true; we do the same very often. But it often produces curious errors.'