Having met several persons who claimed that they always remembered
figures by reasoning about them [whatever that may have meant], and yet
all such persons having shown an inability to remember many dates or
numbers, I inferred that they were honestly mistaken in supposing that
they could remember numbers, or else that such a method was not adapted
to their idiosyncrasies. At that time, I did not suspect that their
failure may have arisen from lack of _training_ in In., Ex., and Con.
From the circumstance that I myself could use this method with
promptitude and certainty, I determined to test it in a strictly
scientific way.
I made the experiment two years ago, and all my experience since has
corroborated the conclusion then arrived at.
I experimented with the two groups of 20 pupils each. Neither knew any
method of dealing with dates and numbers. The first group had had no
training in In., Ex., and Con.; the second group had been well practised
in those laws. I then gave each member of each group several very
difficult cases of dates and numbers to be memorised--one example
containing 24 figures. To save time and space in exposition, I have
heretofore only mentioned 12 figures, or the half of the amount. All of
the first group failed except one. He, however, could not memorise the
24 figures. All of the second group handled all the new examples with
success, and only two of them met with much difficulty in dealing with
the 24 figures.
Since this decisive experiment, I have heartily recommended the method
of finding relations amongst the numbers themselves, to all who are
proficient in the use of In., Ex., and Con.
The example of 24 figures must conclude this exposition. They represent
respectively the number of the day of the month in which the first
Saturday in each month falls in 1895 and 1896. To one without practice
in applying analysis to figures, there seems no hope of memorising this
long group of figures except by endless repetition. The 24 figures are
522641637527417426415375.
Yet reflect a moment and all will be clear. Divide the 24 figures into 2
groups of 12 figures each and number the first group, divided into four
sections, thus:--
(1) (2) (3) (4)
522, 641, 637, 527.
Now bring the first and fourth groups into relation, and you see at once
that the fourth group is larger than the first group by only _five_.
Bringing the _second_ group into relation with the _third_ group, we
find they differ only by _four_. Again: the third group is larger than
the fourth by 100 and by 10, that is 527 becomes 637, the seven alone
remaining steadfast. Beginning with the fourth group and passing to the
third group we have the fourth group with 110 added. The second group is
the third group with only four added, and the first group is the fourth
group with only five subtracted. Thinking out these relations you can
recall the groups as groups or the separate figures of each group or the
entire 12 figures either forwards or backwards--and you have achieved
this result by _Attention_ and _Thought_.
The other twelve figures are easily disposed of. They are 417426415375.
Divided into groups of three figures each we have
(1) (2) (3) (4)
417 426 415 375.
Bringing the first group into relation with the third group, we notice
that it is larger by two--and considering the second group with the
fourth group, we find that the second group is as much and one more
above 400 as the fourth is below 400. Other minor matters could be
noticed, as that the first two figures of each group are respectively
41--42--41--37, and that the last figure in each group is 7--6--5--5.
But these relations are hardly worth observing.
Coming back to the first series, we know that each figure represents
the number of the day of the month to which it belongs on which
the first Saturday in that month falls. The figures for 1895 are
522--641--637--527. The first Saturday in January, 1895, falls on the
_fifth_ day of January, hence the second Saturday must be 5 + 7 = the
12th day of January; the third Saturday the 19th, and the fourth
Saturday 26th. It is easy to know on what day of the _week_ any day in
January falls. Suppose you ask on what week day the 25th of January
falls? You know the 26th is Saturday, and hence the 25th must be the day
preceding the 26th, to wit, Friday, the 25th. Suppose you ask on what
week day the 9th of January falls. You know the 12th is Saturday (the
second Saturday). You now count backward thus: 12 is Saturday, 11 must
be Friday, 10 Thursday, 9 must be Wednesday. The _first_ Saturday in
January, 1895, is the 5th; of February, the 2nd; of March, the 2nd; of
April, the 6th; of May, the 4th, &c., &c. And we can tell on what week
day any day of any of the other months falls.