Page:Memory (1913).djvu/45

 {|
 * | Within the limits
 * | i.e. within the deviation
 * | Number of deviations
 * | By actual count
 * | Calculated from theory
 * | P.E.
 * | 4
 * | 4
 * | 4
 * | .5
 * | P.E.
 * | 8
 * | 7
 * | 7
 * | .6
 * | P.E.
 * | 12
 * | 12
 * | 11
 * | .3
 * | P.E.
 * | 24
 * | 23
 * | 22
 * | .2
 * | P.E.
 * | 48
 * | 44
 * | 42
 * | .0
 * | 1
 * | P.E.
 * | 72
 * | 57
 * | 57
 * | .8
 * | 2
 * | P.E.
 * | 96
 * | 68
 * | 69
 * | .0
 * | 2
 * | P.E.
 * | 121
 * | 75
 * | 76
 * | .0
 * | 3
 * | P.E.
 * | 145
 * | 81
 * | 80
 * | .0
 * }
 * | 2
 * | P.E.
 * | 96
 * | 68
 * | 69
 * | .0
 * | 2
 * | P.E.
 * | 121
 * | 75
 * | 76
 * | .0
 * | 3
 * | P.E.
 * | 145
 * | 81
 * | 80
 * | .0
 * }
 * | 3
 * | P.E.
 * | 145
 * | 81
 * | 80
 * | .0
 * }
 * | .0
 * }

The symmetry of distribution is here satisfactorily maintained apart from the numbers, which are unimportant on account of their smallness.

The deviation which is greatest absolutely is toward the lower limit.

If several of our series of syllables were combined into groups and then memorised separately, the length of time necessary to memorise a whole group varied greatly, to be sure, when repeated tests were taken; but, in spite of this, when taken as a whole they varied in a manner similar to that of the measures of the ideally homogeneous processes of natural science, which also vary from each other. So, at least in experimental fashion, it is allowable to use the mean values obtained from the numerical results for the various tests in order to establish the existence of causal relations just as natural science does that by means of its constants.

The number of series of syllables which is to be combined into