evolve easily. In the third manner of arranging the manuscript the
page forms show very clearly, and it is still used in the scrolls of the
law in Jewish synagogues, kept on two rollers, one at each end. But this
form of writing also developed a new method for its own more convenient
preservation. A roll of this kind can be folded up, backwards and
forwards, the bend coming in the vacant spaces between the columns of
writing. When this is done it at once becomes a book, and takes the
Chinese and Japanese form known as _orihon_--all the writing on one side
of the roll or strip of paper and all the other side blank. Some books
of this kind are simply guarded by two boards, but generally they are
fastened together along one of the sides, which then becomes the back of
the book. The earliest fastening of such books consists of a lacing with
some cord or fibre run through holes stabbed right through the substance
of the roll, near the edge. Now the _orihon_ is complete, and it is the
link between the roll and the book. This "stabbed" form of binding is
the earliest method of keeping the leaves of a book together; it occurs
in the case of a Coptic papyrus of about the 8th century found at
Thebes, but it is rarely used in the case of papyrus, as the material is
too brittle to retain the threads properly.

The method of folding vellum into pages seems to have been first
followed about the 5th century. The sheets were folded once, and
gatherings of four or more folded sheets were made, so that stitches
through the fold at the back would hold all the sheets together and each
leaf could be conveniently turned over. Very soon an obvious plan of
fixing several of these gatherings, or quires, together was followed by
the simple expedient of fastening the threads at the back round a strong
strip of leather or vellum held at right angles to the line of the
backs. This early plan of "sewing" books is to-day used in the case of
valuable books; it is known as "flexible" work, and has never been
improved upon.

As soon as the method of sewing quires together in this way became well
understood, it was found that the projecting bands at the back needed
protection, so that when all the quires were joined together and, so
far, finished, strips of leather were fastened all over the back. But it
was also found that vellum leaves were apt to curl strongly, and to
counteract this tendency strong wooden boards were put on each side. The
loose ends