Memory vanishes with extremes of resemblance or difference. Things which put us in mind of others must be neither too like nor too unlike them. It is our sense that a position is not quite the same which makes us find it so nearly the same. We remember by the aid of differences as much as by that of samenesses. If there could be no difference there would be no memory, for the two positions would become absolutely one and the same, and the universe would repeat itself for ever and ever as between these two points. When ninety-nine hundredths of one set of phenomena are presented while the hundredth is withdrawn without apparent cause, so that we can no longer do something which according to our past experience we ought to find no difficulty in doing, then we may guess what a bee must feel as it goes flying up and down a window-pane. Then we have doubts thrown upon the fundamental axiom of life, i.e. that like antecedents will be followed by like consequents. On this we go mad and die in a short time. Mistaken memory may be as potent as genuine recollection, so far as its effects go, unless it happens to come more into collision with other and not mistaken memories than it is able to contend against. Mistakes or delusions occur mainly in two ways. First, when the circumstances have changed a little but not enough to make us recognise the fact: this may happen either because of want of attention on our part or because of the hidden nature of the alteration, or because of its slightness in itself, the importance depending upon its relations to something else which make a very small change have an importance it would not otherwise have: in these cases the memory reverts to the old circumstances unmodified, a sufficient number of the associated ideas having been reproduced to make us assume the remainder without further inspection, and hence follows a want of harmony between action and circumstances which results in trouble somewhere. Secondly, through the memory not reverting in full perfection, though the circumstances are reproduced fully and accurately. Remembering When asked to remember “something” indefinitely you cannot: you look round at once for something to suggest what you shall try and remember. For thought must be always about some “thing” which thing must either be a thing by courtesy, as an air of Handel’s, or else a solid, tangible object, as a piano or an organ, but always the thing must be linked on to matter by a longer or shorter chain as the case may be. I was thinking of this once while walking by the side of the Serpentine and, looking round, saw some ducks alighting on the water; their feet reminded me of the way the sea-birds used to alight when I was going to New Zealand and I set to work recalling attendant facts. Without help from outside I should have remembered nothing. A Torn Finger-Nail