do well, before trying to alter it, to make sure that there is not already in it some point where it may actually end strongly. But the rule tells the whole story. I have yet to find the composition that may not to advantage end with words that deserve distinction.

So we come to the third phase of the principle of Mass in whole compositions. In the textbooks I have found this somewhat dryly formulated thus: Due proportion should obtain between principal and subordinate matters. In simple English I conceive this to mean that, generally speaking, what is most important may conveniently be treated at most length. In biography, for example, a kind of writing that students often have to try, the first question is why the subject is worth writing about at all. During the past ten years it has been my misfortune to read, I should guess, from five hundred to a thousand undergraduate accounts of the life of Daniel Webster. Now, Webster, I conceive, is worth writing about for three different reasons he was a great orator, and a very notable lawyer, and a great statesman. Any or all of these phases of his character might properly occupy the greater part of any account of his life. But what in my opinion should be passed over hastily is what in a great number of the undergraduate compositions is treated at the greatest length; namely, the not very exceptional circumstances of his childhood and youth. I remember one theme which covered perhaps a dozen pages of carefully written manuscript, of which all but two were devoted to an elaborate account - re

ferred to no recognized authority of how the infant Daniel, engaged in ploughing with his father, plied the old gentleman with many edifying questions concerning the rights and duties of American citizens, and received answers that might have been copied from "Sandford and Merton." I remember another theme, entitled "John the Baptist," which told of nothing but the extremely picturesque and very highly colored misconduct of Herod. And only a short time ago I had occasion to study a life of Sir Richard Steele, in which a great many pages were devoted to discussions - illustrated by legal documents quoted at length-as to who were, and who were not, related to his wives. Yet really what a reader wanted to know in each of these cases really, I think, what the writer wished to tell — was why Webster, or John the Baptist, or Steele, was worth the attention we were called upon to give him.

Of course, even in writing of this kind, our purpose may be different from the general one. Last year I read a Life of Abraham Lincoln, by a Mr. Herndon who was an intimate friend of his in early life. Whether Herndon's book is authentic or not, I do not pretend to decide. It purports to give an astonishingly complete account of what Lincoln did and what manner of world he lived in up to the time when he emerged into the sight of the nation. It purports, indeed, to tell the whole story of his life; but after Lincoln was in national politics, Herndon saw and knew comparatively little of him, and other people

saw and knew a great deal. Of Herndon's three volumes, then, almost if not quite two are devoted to the earlier part of Lincoln's career. Into the third volume is compressed, in very general form, all that makes Lincoln's name a household word; but this massing of Herndon's book, far from being faulty, seems to me admirable. What Herndon had to tell, what nobody else knew, was precisely that personal detail of early life which the other books and other writers, for want of knowledge, passed over. A truer title would have been the "Early Life of Lincoln.”

A better book

might have ended at the moment when Lincoln became a public character. But, given Herndon's purpose, Herndon's book is, in its main masses, very well composed, for the very reason that it gives most space, and so attracts most notice, to what most deserves distinction.

An interesting composition from this point of view is the chapter in "Vanity Fair" which tells of the battle of Waterloo. In point of fact, I rather think Thackeray had never seen a great battle, and was too prudent an artist to venture on the description of a very notable kind of thing which he knew only from hearsay. He lays his scene in Brussels, then, and tells with great vividness and detail the story of the panic there, -not essentially a different thing from any other scene of general excitement and confusion and terror; a great deal nearer the ordinary experience of human beings than any form of battle, murder, or sudden death. But he never lets you forget that what has

made this panic is Waterloo: every now and then you hear the growling of the cannon, and feel, hovering not far off, the dreadful shadow of Bonaparte. Soin my little Tauchnitz edition - he writes for twentytwo pages, dwelling at greatest length on that part of his subject which he was best able to treat, and leaving in the reader's mind what every writer really wishes to leave there a deep sense of reality and of power. But this has not told his whole story. In the last page and a half he tells very briefly what had been doing in the field all this time; and in his very last paragraph—and the very last words of it— he tells the fact which makes the passage an essential part of his story. Here is the paragraph, and it is so placed that in the total effect of the chapter it remains the chief point of the whole:

"No more firing was heard at Brussels: the pursuit rolled miles away. The darkness came down on the field and city; and Amelia was praying for George, who was lying on his face, dead, with a bullet through his heart." For skilful massing that chapter has always impressed me as notable. It is the space given to Brussels that emphasizes the part of the story which Thackeray could tell best; it is the placing of that single sentence about George Osborne - not even a sentence, only a relative clause-which leaves it once for all inevitably in the reader's memory.

In whole compositions, then, the question of mass of how we should begin, how end, how arrange the

proportions of our work becomes more important and more delicate than before. On our management of it depends to an amazing degree what effects we produce with given material. It cannot be considered too carefully. And nothing has so assisted my con sideration of it as that simple device with cards that show me, as I arrange them in different orders, what different effects are at any moment within my power.

So we come to the principle of Coherence: that the relation of each part of a composition to its neighbors should be unmistakable. In sentences and in paragraphs, we shall remember, we found that this matter of coherence depended on one or more of three devices: the actual order in which we arranged the parts of our compositions; uniformity of constructions; and the use of connectives. In whole compositions these three devices remain important; but the first and the third are more so than the second. The simplest way of considering them, perhaps, is to revert to the little packs of cards that I have said are so useful in deciding questions of mass. In arranging these it is not enough that we should give most space to what we wish most to impress on the reader, or put at the beginning and the end the matters we wish chiefly to emphasize. It is almost equally important that we arrange the separate parts of our compositions in this case, the separate paragraphs in an order that shall as far as possible indicate their mutual relations.