additional merit of being practically useful; while, on the other hand, every subdivision, if compelled to devote years to the acquirement of a competing knowledge of these dead languages, would have to abstain from learning something which is indispensable to efficient International Agents.

Again, with reference to the above rule, it is clear that while diplomatists should confine their attention to a few languages, in order to attain a thorough mastery of them both as to reading and speaking, pursers of the navy should, on the other hand, endeavour to attain a limited mercantile knowledge of the greatest possible number of languages. The same holds, though in a lesser degree, of a certain amount of knowledge of as many languages as possible on the part of naval and military officers generally; provided such knowledge is in addition to the essential professional qualifications expected in them. But in the competitions for the staff subdivisions of the army, a great proficiency in French and German at least, should be made to tell considerably in the passing, as these officers have at times to carry on military negotiations.

All candidates for the mounted departments of the army should pass a sufficing examination in horsemanship, i.e. have to ride over a fixed tract of country more or less rough,-say over a staked course on Aldershott. And all candidates should pass a sufficing examination, proportioned to their ages and size, in running, and lifting and throwing weights,-as at the District Examinations. It has quite astonished me to read the amount of nonsense that has been uttered about "pale faced students," in the discussions on Examinations. Physical qualities are more easily tested than the intellectual. And as every really good measure brings with it collateral benefits, so the plan now proposed would have the effect of inducing great numbers of young men (and their parents) to pay much more attention to their health than they otherwise would. I know that the Military Examinations in China have that effect, though they are otherwise of little value, because not requiring intellectual military acquirements.

§ 18. In the original arranging and subsequent improving of the detailed methods of examination, it should be steadily kept in view that the first object is to guard against faults of feeling and of head on the part of the Examiners-against emotional partiality

and intellectual error. Each naturally distinct qualification should form the subject of a separate examination; even French speaking and interpreting, for instance, being competed in apart from French translating. There should always be at least five Examiners, in order to have a sufficient security against indolence or against idiosyncratic eccentricity. The written examinations in each subject should be finished before the oral commence. The signatures on each student's paper should be completely hidden by some covering sealed over it, and have a number attached to it. All should then be passed into a room of copyists; five copies made of each with its number; the originals laid by; and the copies only handed in to the Examiners. The Examination Buildings should contain five separate suits of apartments, each composed of the number of rooms, &c. necessary for the comfortable accommodation of an Examiner, and wherein the Examiners should be shut up, without possibility of communication with each other or with the public, till each had fixed the orders of the papers according to the degree of their excellence.

The following will give an idea of the circumstances under which all papers should be prepared. We will suppose the Examination to be in translating from French into English and from English into French. As this would be one of those attended by the greatest number of candidates, the latter could be divided into two or three sets by lot. As many as the Examination Hall could accommodate should be let into it at one time, and each candidate take possession of one of the boxes into which the whole of its floor should be divided. These boxes should have sides so high as to prevent the candidates communicating with each other, yet leave the motions of each open to observation from a gallery running round the Hall. Each candidate would bring his own ink and pens, but would find blank paper on the desk in his box. Each would there also find, in a closed envelope, the two papers which were to be translated. These would be selected by lot in the morning in the Examiners' common room from various books, and would each consist of a page or two on different subjects. As soon as selected, as many copies would be printed in the Examiner's room as there were candidates, and then closed in the envelopes, the printers not being allowed to leave till the last set of candidates had finished their translations. Each candidate on entering his box would hold up the envelope above his head till all were placed, when, on a bell

being struck, each would open his envelope and set to work, the time of commencement being publicly announced and noted. As each candidate finished his translations, he would sign them, seal the cover over his name and then proceed from his box to, and put them through, a hole in the wall, of which there should be one at the end of all the aisles between the boxes. At each of these holes, on the other side of the wall, would be officials who under public eye would write the hour and minute on each paper. At the end of an amply sufficient time, all the papers, whether finished or unfinished, should be put through the holes; and the whole number taken to the copyist's hall. The second set of candidates would be admitted as soon as the Hall was prepared as for the first; and, as two hours would be quite enough to allow for each set, in one day the whole of the candidates' work in this French examination would be done. That of the Examiners would commence so soon as the first copies were handed into them, and might continue for two or three days. But practice in the work would enable them to get through it with great rapidity. The proper translations of each task would be agreed upon by the Examiners before each repaired to his own apartments, and the business of each would only be to settle which papers differed least from it. As, in practice, it is, in ninety-nine cases out of a hundred, of far greater importance that a translation should be done accurately than rapidly, the time would only be considered where the papers were, in point of accuracy, alike. And if, after judgment had been passed on the copies, it was found by inspection of the originals that rapidity had been attained by bad writing, then a more than full proportion of time should be added,―bad writing being, in practical affairs, very objectionable.

man.

The examination in speaking and interpreting would require much more of the Examiners' time. They would each be seated in a box open in front. On one side of a table, on a lower level before them, would be a Frenchman, on the other side an EnglishBetween these two, and facing the Examiners, each candidate would seat himself and interpret a fixed set of questions and answers between them. The two interlocutors would speak at fixed intervals, irrespective of the candidate's interpretations. Every word uttered by him would be taken down by the Examination short-hand writers; and the Examiners would each make notes on

a paper with that candidate's number on it. Either before or after the interpreting, the candidate would have to read, in a loud voice, a passage from a French book ;-the Reporters and Examiners taking notes as before. On the printed Reports and on his own notes, each Examiner would subsequently make out his list of candidates. This oral Examination might last ten or twenty days, according to the number of the candidates. It would, therefore, be necessary to have a new conversation, and a new passage to read, for every day, (care being, of course, taken that they should be alike in point of difficulty,) as it would be impossible to keep one conversation and passage secret beyond a single day from the candidates who were to be examined. After each sitting, the Examiners should be conducted to their own apartments, and should hold no communications with each other or the public till after making out their lists. Altogether the written and oral Examinations would occupy the five Examiners in French for several weeks. In China the Examiners are always occupied for some such period. But the candidates would each only be occupied for two days; before and after which they would be severally undergoing the other Examinations, appointed for that subdivision of the Executive which they competed for.

I have given the above details because many who would not otherwise object to the proposed system of Examinations give up the idea of instituting them because they cannot conceive how it could be possible in practice to conduct examinations in so many different qualifications of so many candidates. It is, however, evident from the above that, after two or three years' experience and modification of details, the work would be done rapidly and with great order as well as with impartiality and accuracy. As everything would be printed after each Examination, the Examiners and the public together would soon discover what was, with reference to each qualification, the smallest quantity of work that would afford sufficient scope for distinguishing between the degrees of proficiency in each candidate, as also how to get, in the most speedy way, at the essentials of each particular branch of knowledge. It may appear to some readers that I have projected an unnecessary amount of precautions to secure impartiality on the part of the Examiners. But it must be remembered that entrance

d

to the Public Offices at home, to the Diplomatic and Consular Services, and to the Army, being only possible through these Examinations, every conceivable agency of corruption will be brought to bear on the Examiners, and that, to all the right-minded among them, it would be a relief to be put beyond every suspicion. Besides, we have to guard against what might be called incorrupt, because unconscious impartialities and the suspicion of them. When a small proportion only of Scotch passed at one of our recent examinations (one of the first I believe), it was immediately pointed out that there were no Scots among the Examiners.

§ 19. As I understand the present method of passing candidates by means of marks, it appears to me to involve a risk of considerable inaccuracy. It requires the Examiners to refer to an imaginary standard. Speaking of the Indian Civil Service Examinations, we find, for instance, "Composition" put down at 500, and we hear that none of the candidates attained this highest number. The number 500 represents, therefore, some imaginary degree of excellence, the conception of which must manifestly vary considerably in the minds of the different Examiners, and even in the mind of each Examiner at different times. If they affix their marks separately, there is certain to be a wide range in those attached to one paper. My plan requires no comparison of a real thing with an imaginary one, but of one (candidate's) paper with another. Given five papers of really different degrees of excellence, it is easy, by comparing and recomparing them with each other, to number them 1, 2, 3, 4 and 5; and there will be little, if any difference in the order adopted by separately-judging Examiners. In all this matter, we should never forget what is the practical object. The practical object is to select yearly, from the young men who present themselves for examinations, the required fixed number of the very ablest. Whether or not the graduates of this year stand higher than the graduates of last year, is undoubtedly an interesting question, and it is one which can be solved. But it is not the practical question, and can hardly be solved during each year's examinations. It is extremely doubtful if the judging faculties of any Examiners (supposing them to be the same men) would remain from year to year sufficiently consistent to enable them to solve it directly; and