EXPERIMENTS ON THE ASSOCIATION OF IDEAS. 69

the picture in a foreign language can be measured. I must refer the reader to that paper for a detailed account of apparatus and methods. '001 sec. is taken as the unit of time, a being used as its symbol. B (Dr. G. O. Berger) and C (the writer) are the two subjects; after these designations there is given the average time taken in all the experiments made, and the mean variation of these measurements from the average ; after this is given a second average and mean variation, found by dropping the most irregular times in accordance with the method I have described. 1 The number of experiments made 011 each subject is given in parenthesis. The experiments were made at Leipsic during the first half of the year 1885. I give first the time it took the subjects to recognise the pictures of twenty-six familiar objects, and name them in a foreign language B in English, C in German. Pictures named in Foreign Language (78). B 649 104 632 49 C 694 87 682 43 It has been shown - that B took 477, C 545<r to see and name these same pictures in their native languages. B consequently needed 172, C 1490- in addition to find the name in a foreign language. C talks German readily, B English less so. These should be compared with other experiments I have made showing that the rate at which a person can read a foreign language is proportional to his familiarity with the language. 3 We go a step further when a word must be translated from one language into another. The mental operation is again obscure, the processes of translating and naming not being sharply defined; but if we subtract the time it takes to see and name a word from the time it takes to see a word, to translate it into a foreign language and name it, we get approximately the time of translation. This time I give for translating from a foreign into the native language, and in the reverse direction. I have sub- tracted the time it takes the subjects to see and name words (B 390, C 428<r), and the mean variation (B 28, C 20 ; in the cor- rected series, B 19, C 13j). English-German: Short (Common) Words (78). B 240 77 199 36 C 258 59 237 29 1 MIND, xi. 229. It will be noticed that the corrected averages are usually smaller than the averages from all the determinations ; this is because the subject found difficulty in a few cases. The unconnected value gives the average time taken up by associations ; the corrected average more nearly the time usually taken up by associations. 2 MIND, xi. 533. 3 Phil. Studien, ii. 635 ; Abstract in MIND, xi . 63. I hope shortly to print an account of experiments showing the increasing rapidity with which the classes of a German gymnasium can read Latin.