Now, as they begin to write, they will naturally pause before writing each name, to think of a name to write.  The pause may be but slight, yet there is some pause.  Of course, when they write the selected name, no pause will be necessary; and if hurried properly at that time they will make none.  This is the object of the incessant talking during the experiment.  If left to themselves, the subjects will, in about one half of the cases, write the selected name in the third space from the top.  In about half of the remaining cases the selected name will be written in the fourth space from the top.  This is especially true if in your instructions you direct the subject to “mix the dead person’s name somewhere in among the others where you cannot know where it is.”  In the remaining cases the subjects are liable to write the selected name anywhere, generally first or last.  Now my object is to so manipulate my subjects as to cause them to write the selected name when I want them to do so.  This is done by continuous talking, and distracting their attention until the proper moment.  I choose the third space, since this, being the one they are most liable to choose of their own accord, is the easiest to force.  Just as they begin to write the first name, before they make a mark, I say suddenly, “Now be sure and select names of living persons that I could not possibly know.”  This is almost certain to insure a pause, and the name of a living person to be written first.  I continue my talking in a natural manner, taking the attention to a great extent from the writing, and nearly always observing another pause just before writing the second name.  When the second name is almost finished I exclaim suddenly, “Now write as rapidly as possible!” If the subjects have been properly impressed with the seriousness of the experiment, they will almost invariably, on finishing the second name (in obedience to my command “to be as rapid as possible,” and in their desire to please me), hurry into the name already in their minds, thus writing the selected name in the third place.  If such is the case they will now most surely pause to think of a fourth name.  If so, I am certain that I now know the selected name.  However, if they should rapidly pass into the fourth name, it is then uncertain whether the selected name is in the third or fourth space.  This, however, seldom happens if worked in an expert manner.

In rare cases the subject cannot be manipulated by the performer, in which case it is purely guesswork; even in such cases, however, I stand one chance in six of succeeding; and if I make a second trial on failing (not uncommon with mediums), I stand one chance in three of succeeding.

It is hardly worth while to say that as I fold the billets, I fold the third one slightly different from the rest, so that while it will not attract attention, I can see at a glance what it is when thrown on the table.  I memorize the name; also, if in doubt, I fold a second choice in a still different manner for a second trial.  Frequently I memorize more of the names, folding so I can pick them out.  Then, after giving the dead person’s name with proper effect, I pick up the others, hold them to my head and call out the names.  The effect of this on a subject is very impressive.