The tests show the sensitivity at the time they are made, and give an approximate measure of the discrimination with which the operatee habitually employs his senses. It does not measure his capacity for discrimination, because the discriminative faculty admits of much education, and the test results always show increased delicacy after a little practice. However, the requirements of everyday life educate all our faculties in some degree, and I have not found the performances with test weights to improve much after a little familiarity with their use. The weights have, as it were, to be played with at first, then they must be tried carefully on three or four separate occasions. I did not at first find it at all an easy matter to make test weights so alike as to differ in no other appreciable respect than in their specific gravity, and if they differ and become known apart, the knowledge so acquired will vitiate future judgments in various indirect ways. Similarity in outward shape and touch was ensured by the use of mechanically-made cartridge cases; dissimilarity through any external stain was rendered of no hindrance to the experiment by making the operatee handle them in a bag or with his eyes shut. Two bodies may, however, be alike in weight and outward appearance and yet behave differently when otherwise mechanically tested, and, consequently, when they are handled. For example, take two eggs, one raw and the other hard boiled, and spin them on the table; press the finger for a moment upon either of them whilst it is still spinning: if it be the hard-boiled egg it will stop as dead as a stone: if it be the raw egg, after a little apparent hesitation, it will begin again to rotate. The motion of its shell had alone been stopped; the internal part was still rotating and this compelled the shell to follow it. Owing to this cause, when we handle the two eggs, the one feels "quick" and the other does not. Similarly with the cartridges, when one is rather more loosely packed than the others the difference is perceived on handling them. Or it may have one end heavier than the other, or else its weight may not be equally distributed round its axis, causing it to rest on the table with the same part always lowermost; differences due to these causes are also easily perceived when handling the cartridges. Again, one of two similar cartridges may balance perfectly in all directions, but the weight of one of them may be disposed too much towards the ends, as in a dumb-bell, or gathered too much towards the centre. The period of oscillation will differ widely in the two cases, as may be shown by suspending the cartridges by strings round their middle so that they shall hang horizontally, and then by a slight tap making them spin to and fro round the string as an axis.