Very properly these are objects of art, and they afford the earliest illustrations in histories of art as they do in histories of the book. Thus the printer who questions what art has to do with his business stamps himself as two hundred thousand years behind the times. They are pictures, and the book of to-day has descended as directly from them as the printer of to-day has descended from the man who made them. They are, moreover, in some instances, works of very high art. The picture of the mammoth, scratched on a fragment of the mammoth's tusk, is a piece of drawing so skillful that only the greatest living masters can equal it. Not even Rembrandt's drawing of the elephant, which Dr. Holmes celebrates in one of his poems, is more expressive or wrought with more economy of effort. In the same district of southwestern France, Dordogne, that yielded the drawings are found long cave galleries of paintings representing the creatures of that period, all executed with great spirit and ability. But what are the steps in the descent from these ancient pictures to the printed book?

Primitive man had one more string to his conversational bow than most civilized people have, namely, sign language. But gesture and speech alike prevail but little against space and time. Each is possible only at short range, and each dies on the eye or ear that receives it. Pictures may be carried to any distance and may be preserved for any length of time. They were probably made first in response to an instinct rather for art than for the communication of ideas; but their great advantage for communication must have been perceived very early, and, as we find picture writing employed by primitive races to-day, we have the right to infer that prehistoric peoples at the same stage of culture also employed it. Pure picture writing, however, does not suffice for all that men have to say. It is easy to represent a house, but how shall we represent a home? It is easy to represent a woman, but how shall we add the idea of wife? To do this we must pass from simple pictures to symbols. Chinese writing has never advanced beyond this stage. Its prodigious type-case of more than forty-two thousand characters contains, therefore, only a series of pictures, direct and symbolic, all highly conventionalized, but recognizable in their earlier forms. To represent "wife" the Chinaman combines the two signs for "woman" and "broom"; to represent "home" he makes a picture of a pig under a roof! The Egyptian and Mexican systems of writing, though very different to the eye, were both of this nature and represented ideas rather than words. Yet all true alphabets, which are representations of sound, have been derived from such primitive ideograms or pictures of ideas. What was the process?

The rebus is the bridge from the writing of thoughts to the writing of sounds, and it came into use through the necessity of writing proper names. Every ancient name, like many modern ones, had a meaning. A king's name might be Wolf, and it would be indicated by the picture of a wolf. Ordinarily the picture would be named by everyone who saw it according to his language; he might call it "wolf," or "lupus," or "lykos"; but when it meant a man's name he must call it Wolf, whatever his own language. So such names as Long Knife and Strong Arm would be represented, and these pictures would thus be associated with the sound rather than the thing. By and by it was found convenient, where the word had several syllables, to use its picture to represent the sound of only the first syllable, and, still later, of only the first sound or letter. Thus the Egyptian symbol for F was originally a picture of the horned asp, later it stood for the Egyptian name of this venomous creature, and finally for the first sound in the name, being used as the letter F itself; and the reason why we have the barred cross-piece in the F, the two horns in U, V, and Y, and the four in W (VV) is because the Egyptian asp had two horns, as may be seen from the illustration in the Century Dictionary under the word cerastes; and every time that we write one of these letters we are making a faded copy of the old picture. We find systems of writing in all the stages from pure pictures to the phonetic alphabet; in Egyptian hieroglyphics we find a mixture of all the stages. So much for the background of the book as the bringer of a message to the eye, but the outward form or wrapping of that message has also a long and interesting history.

No objects could be much more unlike than a Babylonian tablet, an Egyptian papyrus roll, and a Mexican book. They are as different as a brick, a narrow window-shade, and a lady's fan; they have nothing common in their development, yet they were used for the same purpose and might bring identically the same message to the mind. Inwardly, as regards writing or printing, all books have a parallel development; but outwardly, in their material and its form, they are the results of local conditions. In Babylonia, which was a fertile river-bottom, bricks were the only building material, and clay was therefore a familiar substance. Nothing was more natural than that the Babylonian should scratch his record or message on a little pat of clay, which he could afterwards bake and render permanent. Some day all other books in the world will have crumbled into dust, their records being saved only when reproduced; but at that remote time there will still exist Babylonian books, even now five thousand years old, apparently no nearer destruction than when they were first made.

The Babylonian book carried its message all on the outside; the Egyptian book went to the opposite extreme, and we should find our chief objection to it in the difficulty of getting readily at its contents. There flourished on the banks of the Nile a stout reed, six feet high, called by the Egyptians "p-apa" and by the Greeks "papyros" or "byblos." It was the great source of raw material for Egyptian manufactures. Its tufted head was used for garlands; its woody root for various purposes; its tough rind for ropes, shoes, and similar articles—the basket of Moses, for instance; and its cellular pith for a surface to write on. As the stem was jointed, the pith came in lengths, the best from eight to ten inches. These lengths were sliced through from top to bottom, and the thin slices laid side by side. Another layer was pasted crosswise above these, the whole pressed, dried in the sun, and rubbed smooth, thus giving a single sheet of papyrus. As the grain ran differently on the two surfaces of the papyrus sheet, only one side was written on. Other sheets were added to this by pasting them edge to edge until enough for a roll had been made, usually twenty, a roller being fastened to the last edge and a protecting strip of wood to the front. The manuscript was unrolled by the right hand and rolled up by the left. It is obvious that a book of reference in this form would be subjected to great wear. In our dictionaries it is as easy to find Z as A; but in a papyrus book, to find the end meant to unroll the whole. The Latin word for roll was "volumen," hence our "volume." A long work could obviously not be produced conveniently in a single roll, therefore Homer's "Iliad" and "Odyssey," for instance, were each divided into twenty-four books, and that is why the divisions of an epic poem are still called books, though they are really chapters. The rolls composing a single work were kept together in a case something like a bandbox. The roll was the book form of the Greek and Roman as well as the Egyptian world, but it left no descendants. Our book form was derived from a different source, which we will now consider.

Just as we speak of Russia leather, so the ancients spoke of Pergamum skins, or parchment. The story is that Eumenes II, King of Pergamum, a city of Asia Minor, tried to build up a library rivaling that of Alexandria, and the Ptolemies, seeking to thwart him, forbade the export of papyrus from Egypt. Eumenes, however, developed the manufacture of Pergamum skin, or parchment, or vellum, which not only enabled him to go on with his library, but also incidentally changed the whole character of the book for future ages. This material is not only much more serviceable than the fragile papyrus, but, being tough enough to stand folding and sewing, permitted the book to be made in its present or codex form, the original codex being two or three Roman waxed tablets of wood, fastened together like hinged slates, and thus opening very crudely in the manner of our books. This development of parchment occurred in the first half of the second century before Christ. The new material and book form gradually made their way into favor and came to constitute the book of the early Christian and medieval world. Though paper was introduced into Europe soon after the year seven hundred, it did not displace parchment until the invention of printing called for a material of its cheaper and more adaptable character.

But, though we have traced the origin of our present book form, we have not yet filled in the background of its history. Several other notable types of the book deserve our attention; first of all that of China, one of the most attractive of all book forms, to which we devote our next chapter. Though it superficially resembles our own books, it is really the product of a different line of evolution. When we examine it closely, we find that in many respects it is the exact reverse of our practice. It is printed on only one side of the paper; it is trimmed at the back and folded on the fore edge; its wide margin is at the top; its running headline is on the folded fore edge; its sewing is on the outside; its binding is limp; its lines run up and down the page; and its pages, according to Western ideas, open from the back towards the front. Yet it is a thing of beauty, and let us hope that nothing in the modern reorganization of China will change its character to prevent it from remaining a joy forever.

Just as Chinese paper is made from bamboo, which plays an even greater part in China than papyrus did in Egypt, so the book of India utilizes the leaves of that important tropical tree, the palm. The sheets of the book before me are strips of palm-leaf two inches wide and two feet long. They are written on both sides and, following the run of the grain, lengthwise. This makes an inordinate length of line, but, owing to the small number of lines on the page, the confusion of the eye is less than might be expected. The leaves composing the book are clamped between two boards of their own size, the block thus formed is pierced with two holes, through which pins are thrust, and the whole is wound with a cord. The dimensions vary, some books being larger and some much smaller. I have also before me a Burmese booklet in which the leaves are one inch wide and six inches long. Sometimes the sheets are of brass, beautifully lacquered, and the writing heavy and highly decorative. These books also vary greatly in size, some forming truly massive and sumptuous volumes. Birch bark was also employed as a book material in India, being used in what we should call quarto sheets, and in Farther India a peculiar roll is in use, made of Chinese paper, folded at the side, sewed at the top, and rolled up like a manifold banner in a cover of orange-colored or brown cotton cloth.

We do not ordinarily associate books with pre-Columbian America; yet one of the most interesting of all book forms was current in Mexico before the Conquest. As in the case of the Chinese book, it looks superficially like ours; we think it is a tiny quarto until we see that its measure is rather that of an oblong twenty-fourmo; that is, its dimensions are just scant of five inches high and six inches wide. It has thin wooden covers and is, over all, an inch thick; but between these covers is a strip of deerskin twenty-nine feet long and, of course, nearly five inches wide. This is folded in screen or fan fashion, the first and last leaves being pasted to the inside of the covers. This attachment is really the only binding; the whole strip is capable of being opened up to its full length. It is read—by those who can read its vividly colored hieroglyphics—by holding it like a modern book, turning the leaves until what seems the end is reached, and then turning the cover for the next leaf, and continuing to turn until the first cover is reached again, but from the other side. Incredible as it may seem, there is a book of India which is almost identical in structure with the ancient Mexican book. It has the shape of the palm-leaf book, but it is made of heavy paper, blackened to be written on with a chalk pencil, and it opens like a fan exactly in the Mexican fashion. Each cover is formed by a double fold of paper, and the writing runs lengthwise of the page as in the palm-leaf volume. As the writing can be erased, the book serves the purpose of a slate.

The variety of objects that men have used to write upon almost surpasses imagination, ranging from mountain walls to the ivory shoulders of Rider Haggard's heroine in his "Mr. Meeson's Will." Such unusual, if actual, writing materials belong, perhaps, rather to the penumbra than to the background of the book; but, as a final survey of our subject, running back to the time when there were no books and men must rely upon their memories, we may quote what Lane says of the sources from which the Kuran was derived after the death of Mohammed: "So Zeyd gathered the KurÁn from palm-leaves, skins, shoulder-blades (of beasts), stones, and the hearts of men."