PALEOGRAPHY 149 Soc., pi. 127), and the Psalter of St Nicholas of Cusa (pi. 128) and the Codex Sangallensis and Boernerianus of the 10th century (pi. 179). The same imitative characters are used in quotations of Greek words in Latin MSS. of the same periods. Cursive. The materials for the study of early Greek cursive writing are found in papyri discovered in Egypt and now deposited in the British Museum, the Louvre, the library of Leyden, and the Vatican. The earliest of these to which an exact date can be assigned are contained in the collection of documents of a certain Ptolemy, son of Glaucias, a Macedonian Greek, who became a recluse of the Serapseum at Memphis in 173-172 B.C., and collected or wrote these documents relating to himself and others connected with the service of the temple in the middle of the 2d century B.C. A series of these and other documents can be selected so as to give a fairly continuous course of cursive handwriting from that period for several centuries. The papyri are supplemented by the ostraka or potsherds on which were written the receipts for payment of taxes, itc., in Egypt under the Roman empire, and which have been found in large quantities. Lastly there are still extant a few specimens of Greek cursive writing on waxen tablets ; and in documents of the 6th and 7th cen turies from Naples and Ravenna there are found subscrip tions in Latin written in Greek characters (Marini, I papiri diplom., 90, 92, 121; Cod. Dipt. Cavensis, vol. ii., No. 250). Facsimiles of the cursively written papyri are found scattered in different works, some dealing specially with the subject. By far the most plentiful and best executed are those which reproduce the specimens preserved at Paris in the atlas accompanying Notices et Extraits des Manuscrits, vol. xviii. In the earliest examples of cursive writing we find the uncial character in use, and, as has been already remarked, many of the specimens fluctuate between the more formal or set book-hand and the cursive. As time goes on the two styles diverge more widely. The uncial book-hand had, as we have seen, a disposition to become more formal ; cursive writing naturally has the opposite tendency, to become more flowing and disintegrated, the more exten sively it is used. But the fact that there existed in Egypt in the 2d century B.C. a cursive hand not differing very material!}" from a more formal contemporary hand seems to indicate that the two styles had diverged at no very long time before. It cannot, however, be supposed that a cursive form of Greek writing did not exist still earlier. The highly developed calligraphy of the earliest examples proves that Greek writing, as we there see it, was then no newly-discovered art. Judging by the analogy of later reforms, it is perhaps not going too far to conjecture that in the papyri under consideration we see the results of a calligraphic reform, in which a new model was perfected from earlier styles. The cursive hand in its best style (e.g., N. et Extr., pis. xxviii., xxix.) is very graceful and exact. This elegance is indeed characteristic of most of the writings of the 2d century B.C., and if a criterion can be established for assist ing in the difficult problem of dating the early papyri, this simplicity and evenness of writing appears to be the best. XTP )TnX7~j A^f^r^ Greek Cursive, 163-162 B.C. In the course of successive centuries the cursive hand becomes slacker and more sloping. There is more com bination of letters, and a continual disintegration, so to say, of the forms of the letters themselves. Naturally the letters which undergo most change are those which lend themselves most readily to combination with others. Alplia, for example, a letter in constant use, and appearing in frequently recurring words (as KCU), quickly altered its shape. In the earliest papyri it is seen more cursively written than most of its fellows. Epsilon, again, is a letter which soon took a second form. It was found easier to make the cross-bar in conjunction with the upper half of the curve of the letter than by a separate stroke after the formation of the full curve S. The upper half of the letter naturally linked itself with the next following letter; and the epsilon thus broken is found as early as a hundred years B.C., and runs through succeeding centuries. The tau was treated in the same way. In the specimen given above it may be seen how the scribe first made half the horizontal stroke and attached it to the main limb by one action of the pen 1, and then added the other half separately. By this device he avoided moving his hand far back. Next, to write the letter in one stroke, some thing like a y, was a natural development. The transforma tion of pi follows on the same lines ; and the ?i-shaped nu comes from the capital letter quickly written, just as tfee same shape was derived in the Roman alphabet. Such a form as the sickle-shaped rko j> is one that would be expected ; but the system of breaking-up is in no form better illustrated than in the case of delta. This letter, it might be thought, would, from its original shape, resist combination more than any other, yet even in the 2d century B.C. this combination is accomplished, and delta occasionally appears open on the right side and linked with the following letter < Minuscule. The gradual disintegration of the pure forms of the early uncials by this progressive development of more cursive characters led eventually to the formation of minuscule letters. By the beginning of the 6th cen tury most of the letters which are afterwards recognized as minuscules in form had become individually developed. For example, the three letters B, H, and K, which in their capital or uncial shapes are quite distinct, had, at this period, acquired alternative shapes which are not very dissimilar from one another, and which by a careless reader may be confused. The letter B in cursive writing lost its loops and was joined by a tag to the following letter a process by which it became very like the Latin u. So the H readily passed through the form D to la ; and K became U. The A developed at the apex an elongation of the right side of the triangle, which, for junction with the next letter, was bent over, and hence resulted the small 8. The transformation of M through m to p., and of N through U to fj, is obvious. This development, however, of minuscules from the old uncials was a work of time. The incipient changes in individual letters can be detected in papyri of the 2d and 1st centuries B.C. ; but a fully developed minuscule hand, used as an independent form of writing, had no existence for some centuries to come. Arrived, however, at the end of the 6th century, we find a document of 600 A.D. given in facsimile in the Notices et Extraits (pi. xxiii., No. 20), the writing of which is so full of the .smaller letters that the hand is practically a minuscule one. This document and six others which are extant formed part of the business papers of one Aurelius Pachymius, a dealer in purple dye, and, ranging in date from 592 to 616 A.D., are valuable material for elucidating the history of the Greek minuscule character. After an interval of eighty years another important document presents itself, in which the two styles of writing, the old uncial and the new minuscule, are seen on the same page. This is the frag-
