4. A correlation is a successive advance, and an intermediate must not refer back to any except its immediate antecedent, never to its second or third antecedent. A pupil wrote:—Short steps … stepson … real son … more a son … Morrison. Here, “more a son” refers to the comparison between “real son” and “stepson,” but the latter is the second antecedent so the correlation is a defective one. He might have said: Short steps … stepson … Morrison.

5. A word may be used twice but never three times. Pen … pensive … gay … nosegay … Nose. Here “gay” is properly used twice, and after that, it is dropped and you can go on with the rest of the word, to wit, nose.

6. A compound phrase including a verb is rarely allowable, since the intermediates must be the simplest elements, either sensations or perceptions [relations among sensations] or abstractions [relations among relations], or one of these with either of the others, always exemplifying either In., Ex., or Con.

7. My correlations are good for me, but they may not be so vivid to others, especially where the concurrences are used. To fix the date of Magna Charta (1215), the pupil could memorise this Correlation—Magna Charta …‌ King John … Jew’s teeth … DeNTaL. But if the pupil did not know before that King John had granted that charter, and if he did not also know the story about the extraction of the Jew’s teeth to make him pay the royal exaction, there would be no concurrence as to the first word and second, or second and third, and if he learned the Correlation it would be by mere repetition without aid from Analysis. In such a case he would make and memorise his own Correlation, perhaps thus: Magna Charta … magnify … diminish … DwiNDLe. When a pupil makes his own Correlations, every concurrence he uses is a real concurrence to him, and so with his Ins. and Exs. This is a decisive reason why the Pupil should merely look upon my Correlations as models, but make and memorise his own Correlations in all cases, as being more vivid to him and, therefore, more certainly remembered, as well as more effectively strengthening the Memory in both its Stages.

8. Vivid Ins. by meaning are better than Ins. by S. (the latter when used, should be as perfect as possible). Ear … Eel makes a weak In. by S. to some persons, but it would make a much more vivid first impression to most persons to deal with them in this way: Ear … (w)ring … twist … wriggle … Eel. But “Bivouac … aqueduct” is a perfect In. by S. as to the last syllable of the former and the first syllable of the latter, since those syllables are pronounced exactly alike. We may connect Bivouac to Rain thus: “Bivouac … aqueduct … flowing water … falling water … Rain.”

9. Never—in the early stages of the study of the System—make a second Correlation until you have memorised the first.

10. Although making and memorising Correlations serves the useful purpose of fixing specific facts in the memory, yet the main object in making and memorising Correlations is to develop the latent power of the Natural Memory to such a degree that all facts are hereafter remembered without the aid of conscious Correlations.

11. Never try to find analytic date or number words until you know the material facts connected with the date or number before you. The student wishes to fix the date of‌ Voltaire’s birth, in 1694. “The Shaper” and “The Giber” occur to him. If he is ignorant of the facts of Voltaire’s life, he will correlate thus: “Voltaire … (1) … volatile … (2) … ‘fixed’ … (1) … ‘The Shaper’ The Shaper (1694);” or “Voltaire … (1) … tear to pieces … (1) … mocking dissector … (1) … The Giber (1694).” If he had known that Voltaire was a born writer, he would have found the analytic relation in “Voltaire … The Shaper (1694)” or if he had known that he was a terrible mocker, he would have said: “Voltaire … The Giber (1694).” If he wished to fix the date of the discovery of America, he might think of “Terrapin” (a large tide-water turtle, abounding in Maryland), and correlate thus: “Discovery of America … (1) … Maryland … (3) … Terrapin (1492).” But if he remembers that Con. covers all cases of Cause and Effect, Instrument or Means to End, Person by whom, &c., and if he reflects that this discovery has been a blessing to the Old as well as the New World, he would say: “Discovery of America … (3) True Boon (1492).” Or, if he considers that the moment America was made known to Europe the whole of the Western Continent was open to every new-comer, he would find analytic date-words thus: “Discovery of America … (3) … Door open (1492).” If he merely wants to fix the fraction 92, he could use the first two consonants of the name of one of his ships, and say: “Discovery of America … Pinta (1492).”

ISOLATED FACTS.

Correlate an Isolated Fact to something (to some fact in its environment or entourage that is best known and) which you are sure to think of when you wish to recall the Isolated Fact.