Locality, it is observed, is the most efficacious me dium of reminiscence ; and that system of memory will be the most serviceable which brings this principle into the most extensive operation. For this reason, localilu, of the connection of our ideas with places, is made the foundation of the present system.
A room having generally four walls, the most obvious division of it is into four sides, and each wall or side may be subdivided into panncls or compartments. Accord ingly, the ancient system divided a wall into five spaces; and this plan was applied to as many rooms as were found necessary to the extent of each particular scheme —every room being similarly divided into four sides, and every side being subdivided into five compartments, Thus, any idea which, according to this method, had been associated in the mind with the forty-eighth corn partment, would be placed in the third compartment of the second wall, in the third room. But, as few com partments could be obtained on each wall by these means, the calculation of high numbers would be exceedingly difficult. To remedy this defect, each wall might be divided into nine or ten compartments. If a wall be divided into nine parts, there Nv il I be 36 compartments in every room. In order to ascertain the situation of any particular number, it is to be considered in relation to the total number of the subdivisions. For example, if the situation of number 48 be required; according to the last mentioned division of the rooms, it is to be found by considering the proportion NI hich that number hears to 36, the total number of the compartments in this ar rangement. If the number in question be less than this total, the place inquired after will be obvious; thus, 12 being within the number 36, must, of necessity, be in the first room ; being above 9, it is equally clear that in can not be on the first wall, and being less than 18, it must, necessarily, be in some part of the second wall ; and as it exceeds the number of the first wall by 3, it fol lows, of course, that its place must be in the third com partment of the second wall. If the number in question be higher than the number of the compartments in one room, its place will be readily found by dividing it by that number. Thus, suppose 48 to be the number whose
place is required: As 48 exceeds 36, we know that it can not be in the first room, the 1 is therefore changed into two ; and the fraction remain ing shews it to be in the twelfth compart ment. There being nine compartments in every wall, this remainder, or number of the compartments, is divided by 9, for the pur pose of ascertaining the wall. Nov, as the divisor is contained more than once, but not twice, in the dividend, it follows, that the compartment sought must be on the second wall ; the remainder gives the specific compart ment. This operation, then, shews that 48 is in the third compartment, on the second wall, in the second room. This was the plan adopted by the ancients, when they divided their rooms into parts ; but being both complicated and difficult, Mr. Feinaigle has rejected it in his system ; and another scheme has been introduced in its place, which he conceives to be more simple in its construction, less difficult in its application, and much more extensive in its powers.
He divides a wall in the following manner : These figures are arranged from left to right, in the usual manner of wi iting ; and for the more easily remembering their situation, it will be found, that if two lines be drawn diagonally from the four corners of the figure, they will intersect all the odd numbers. The nine squares, or compartments, are termed /daces, and are called first place, second place, &c. The same mode must be pursued with the three remaining walls in this room, by which means four walls are obtained, each divided into nine places. Li order to find the number 36 in this room, we should na turally say four times nine will be 36, and should, of course, conclude that 36 would be in the last place of the last side or fourth wall of the room. But this cal culation is erroneous ; 6 must ever be in the same situ ation, which will be that occupied by the point in the following figure : And the numbers 16, 26 and 36, will be in the corresponding situation on their respedtive walls.