1897.] on Greek and Latin Paloeography. 385 



small d or tall a ; this development is seen pretty well completed in 

 the fourth division. The letter E, besides the capital form, is also 

 written in t^o vertical strokes, a form found in inscriptions and 

 which apj)ears in the old Faliscan alphabet. In the waxen tablets 

 this form is very generally used, no doubt because it was so very 

 easily written. In the letter F, again, the cross stroke gradually 

 drops away, and the letter is formed eventually of merely two strokes, 

 both of them vertical. The development of the tail of G can be traced 

 in the column as we descend. In the fourth division, the four 

 strokes of the letter M fall into a perpendicular arrangement. But 

 this form of the letter does not occur in the Dacian tablets ; it was 

 probably found confusing in a class of writing which contained so 

 many verticals. The letter N g )es through the same course, falling 

 into three vertical lines. The breaking up of the letter is very 

 interesting : when it is formed by the double action of two curves 

 meeting, the second curve tends to become concave like the first, the 

 letter tinis assuming the form of a badly made cursive a. In the 

 letter P we see the gradual loss of the bow — or rather its change 

 from a curve to a mere oblique tag or stroke. Important changes 

 pass over the letter E ; first comes the opening of the bow, then the 

 gradual change in the direction of the stroke, which becomes a mere 

 waved line. 



The second table of alphabets represents the forms of letters found 

 in the Dacian waxen tablets of the second century. Here is a still 

 further development of the letters of the graffiti., and in writing on 

 such a material as wax there would be even more temptation to get 

 rid of superfluities in the letters, than when writing on a plaster- 

 covered wall. Further, the tendency of the action of the hand 

 would be to write letters sloping rather to the left, the curves would 

 all tend to become concave, the stilus being held with its point inwards. 

 The princijjal difficulty in reading the writing on the waxen tablets is 

 caused by the linking of the letters, many of the combinations form- 

 ing almost monograms ; these are all collected in the lower division of 

 the plate. Accurate facsimiles of the wall inscriptions are collected 

 in the ' Corpus Inscriptionum ' and may there be studied in all their 

 details. 



From the tables of alphabets it is seen how the cursive hand of 

 everyday life developes from the capital letters ; and those capital 

 letters are of course nothing more than the later development of the 

 archaic alphabet. To find the Roman literary hand, we must start 

 again from the capitals, but move in a diftereut direction from that 

 followed by the cursive writing. For public inscriptions a refined 

 and artistic form of letters w^as naturally soon required ; and the 

 creation of very perfect alphabets of capital letters, both square aud 

 rustic, resulted. To apply this large style to literary purposes may 

 appear to us a costly aud cumbersome method ; and it is certainly 

 remarkable that the practice of producing manuscripts in large letters, 

 or majuscules, should have endured so many centuries as it did. On 



