1. What is the result of uniting two unconnected "Extremes" by means of a developed Analysis? 2. What are the first steps in memorising a correlation? 3. How long are these repetitions required? 4. What will be the result in a short time? 5. What will be the final result? 6. Are correlations easy to remember? 7. What is the result of making and memorising them? 8. When does the most vivid concurrence take place? ASSIMILATIVE ASSOCIATION AND MEMORY. Probably no psychological mistake was ever fraught with greater injury to the cause of public or self-education than the too prevalent opinion amongst teachers generally that "physiological retentiveness" is the memory's sole reliance _in all stages of life_. It is nearly the sole reliance in infancy, and a partial reliance in youth. But when an accumulation of experiences and a fair command of language have been gained, new acquisitions are henceforward principally made by _the affiliation_ of one idea upon or with another or _the making of associations between ideas already established_. And, if this be so, then memory must be very greatly improvable, since no mental power is susceptible of so much improvement as assimilative association. A good memory, whether natural or acquired, belongs to quick and vivid _associability_ and _revivability_ rather than to mere inherent and perpetual physiological _record making_. After a certain number of experiences the child learns the appearance of a square. All his future experiences, however varied, of squares become affiliated upon, or connected with the record of this original square. If each new square had to be separately impressed on the brain as a distinct and independent physiological record, it would take as much time and trouble to learn every new square as it did to learn the first square. But the _instant_ recognition of every square after learning the first one shows that the old brain record is used in the case of each new experience of squares or that the new square is interpreted by the old or original record through the Laws of Association. Again: Taking the prefixes _com._, _de._, _im._, _op._, _re._, _sup._, &c., which are used in thousands of cases, and the suffixes _ment_, _sion_, _ible_, _ibility_, &c., also used in thousands of words, and using these in connexion with the root word "Press" we have compress, depress, impress, oppress, repress, suppress, and also compressible, depression, re-impress, suppression, impressment, &c. Must a new physiological record be made for each form of the sixty or more words of which Press constitutes the base, and must a new record be also made for each of the prefixes and suffixes in the thousands of combinations in which they occur? No one believes any such absurdity.