standard by which we may very accurately measure the relative efficiency of these several organs. If, therefore, we find, that these, compared among themselves, afford, in the Examination, for a series of years, very different and still very uniform results; we shall be entitled to infer, that one House is comparatively a good, another comparatively a bad, instrument of education ;—be warranted to determine, even on an Oxford standard, what every Oxford House does, may, and should accomplish;―be enabled, in fine; to generalize the circumstances, by which such accomplishment is there furthered or impeded ;—and, consequently, to judge what are the most feasible measures, for the reform and improvement of this University. The same comparison, with the same results, may also, it is evident, be instituted between the efficiency of the same House at one period, and its efficiency at another. Taking, therefore, as the standard of academical proficiency the public Examination in its two Departments, and its four Classes of Honor; I proceed to apply this to the several Houses. And (as shown in the following Table) in two different ways: the one giving the comparative eminence of those educated in each House (there I.); the other, the comparative eminence of those who in each House act as educators (there II.) In reference to the Instructed: The Table shows of each House the number of its undergraduates (a); then the absolute number of the honors obtained by them in each department and in every class (b, c); then the absolute number of Double Firsts (d); lastly, the number of First Class Honors in either department in proportion to the number of competitors (g, h); but previously, by the same relation, the classes of each department valued from lowest to highest, as 1, 2, 3, 4 (e, f). On this proportion in L. H., proceeding only to the first decimal, I have arranged the Houses; when equal in L. H., their difference in D. M. has then determined the order. I have taken, as a sufficient period, the ten years ending with 1847; (the Calendar of 1848 being the only one within my reach when the Table was abstracted;) and I was compelled (for the same reason) to make the number of undergraduates of the last year stand for an average of the whole ten. In reference to the Instructors: The Table shows, in each House: first, absolutely, the amount and quality of the Academical Honors belonging to its several educators, whether Tutors or TABLE-Continued; Showing the comparative efficiency of the OxFORD HOUSES, as + From the Calendar of 1851, the Instructors being accidentally not marked in that of 1848. ‡‡ Until lately New College exercised its unhappy privilege of examining and passing its own members, as candidates for a degree. Readers (i, k); and secondly, the Highest Honors, in either department, in proportion to the number of these educators (1, m).-This latter part of the Table is (for the reason assigned) wholly calculated on the year 1847.1 Looking, then, to the Table, and to its first part; we here see, that one House differs marvelously from another in what it performs. The esprit de corps is fully as remarkable in Colleges as in Regiments; although individual competency and courage must, on the average, be pretty much the same in all. Thus, while one Regiment is for generations known as the "fighting," another as "the flying, -;" so (what is more intelligible), in one College a first class is merely of commonplace respectability, while in another it is a kind of secular dignity, and not to be plucked, there even confers an enviable distinction. Comparing, therefore, the Houses in Litere Humaniores :-In this department, we find that four Houses (two Colleges and two Halls), containing above a hundred undergraduates, have during the decade no First Class Honors at all.-Again, discounting these, and comparing only the Houses which have compassed this i This Table thus affords (apart from inaccuracies), not the very truth, but only a sufficiently close approximation to it. The number of Undergraduates, in the several Houses, ought to have been calculatel, not on one, but on an average of all the ten years.-The same applies to the Instructors. Their average academical eminence, for the several Colleges, ought to have been estimated by a comparison of every year, and not assumed on the last alone. But as I was unable, as stated, when the abstract was made, to accomplish this, the Table must stand as it is; for I have neither time nor patience to reconstruct it. Nor do I think, that the result would vary in any point of importance; for collegial accommodation has been long inadequate; and, at the same time, lodging out during the first four years is not allowed; while the standard of instruction in a House does not frequently nor rapidly change. It might, however, be interesting, had we Tables of the kind, adequately executed-say for every five years. In regard to the valuation of the Classes, on which I have arranged the Houses, in their educational eminence, I have a remark to make.-This valuation is unfavorable to First Classes; therefore, to the higher Colleges, which preponderate in Highest Honors. For, while the three inferior classes testify, that a candidate is above one minimum, they testify that he is below another; whereas, the First Class, while it testifies that a Candidate is above a certain minimum, takes no account of how much or how little he exceeds it. It thus contains and equalizes the most unequal proficiencies; that which is just competent, and that which is far more than competent. was, however, unwilling that any possible objection should be taken on the ground that the valuation was, in any respect, arbitrary. Accordingly, I allow Accordingly, I allow every advantage to those Houses which rejoice in their amount of respectable, though humbler honors. A Double First evidences both talent and a power of application. But it only proves that a candidate (with competent ability) has prepared himself in two complements, each equal to the amount required for a First Class. Of more it testifies nothing. distinction, we find that one College is, on this standard, eighteen times more efficient than another.--Finally, the same discount being made, the valued classes afford a similar result; some Colleges, by a full average, in this the principal department, approving themselves four and a half, and, the discount not made, ten times better instruments of education than others. In Discipline Mathematica, the difference, if less important, is hardly less signal. During the decade, seven Houses, (three Colleges and four Halls), and with an average of undergraduates considerably above two hundred, show no First Class Honors ;and of these, two (a College and a Hall) have no Honor, even of the lowest-Again, discounting these, and taking only the Houses which have attained to a first class, still we find in this respect, one College more than ten times superior to another.-Finally, making the same discount; on the criterion of the whole Honors valued, College excels College, as an educational organ by nearly a twelve-fold difference. But in the last place (discounting All Souls and the Halls), and taking the half proportion of the highest College as a mean, we have the following results: L. H.-In Valued Classes: three Colleges are of the very mean (1:0.8); eight above; and eight below it.-In First Classes: of the mean (1:8), we have one college; above it three; and below it fifteen. D. M.—In Valued Classes: we have of the mean (1:2·4) one college; above it seven; and below it eleven.--In First Classes: there are above the mean (1:18) four colleges; and below it fifteen.1 1 1 I may append the following proportions, though I see there are probably several minor inaccuracies. But I can not go through the labor of correction; more especially as they are irrelevant to my argument, and do not affect the general result. A) Literæ Humaniores. Proportion of— All classified (923), to all (here) unhonored graduates (1932?), as B) Disciplina Mathematica. Proportion of— All classified (255), to all (here) unhonored graduates (2618?), as |