Mr. E.C. R. ARMSTRONG has earned the thanks of all students of early Irish art and epigraphy by rescuing the remarkable series of slabs at Gallen Priory from the oblivion into which they had fallen, and by the very clear descriptions and illustrations which he has given of them (antea, pp. 61, 173). In studying these, a few points occurred to me which I have ventured to put together in the hope that they may be of interest.

SlabNo. 1. In type the cross on this slab resembles those of a large series of the oldest (eighth and ninth centuries) of the Clonmacnois slabs; and the style of the lettering, with a ? shaped like the minuscular Roman d, accords approximately with the same dating. I hardly think it is necessary to see anything but the first four letters of a proper name in the inscription; analogy would be altogether against any descriptive adjective, such as Petrie attempts to find.

I have said that in type this cross resembles certain of the Clonmacnois slabs; but in detail it is absolutely different. The first thing that strikes the eye is the bifurcation of the inner lines of the cross, which is altogether extraordinary, and unlike anything that I can recollect. The nearest parallel that I can call to mind is the lozenge panel surrounding the cross on the lost Orthanach slab at Clonmacnois; but this is by no means "on all fours" with the example under consideration. In Irish crosses and interlacing patterns, bifurcation of lines is very rare indeed, and every example is worth careful consideration.

A second point that this apparently simple, but really very complex and suggestive, slab presents for consideration is the large number of lines of which the cross and corner panels are composed - no fewer than eight in number in the horizontal bar. There is only one cross of the hundreds at Clonmacnois (and that one is now reduced by fracture to a small and insignificant fragment) presenting so many; and this belongs to a much later type than the cross at present before us.

Thirdly, we must notice the singular want of symmetry in the design, which Mr. Armstrong's careful drawing enables us to study in detail. Even though only half of the slab remains, there are so many obvious divergences from the exact balance that the type of pattern demands, that we must, I think, ascribe the slab to a rather unskilled workman, notwithstanding the striving after originality that his treatment of the centre of the cross indicates. Of course even the Clonmacnois sculptors are sometimes found nodding, and there are about half a dozen slabs in the cemetery there that we must pronounce a disgrace to that incomparable series if we could succeed in divesting them of the absorbing interest attaching to every relic of the past of Ireland. But I do not think there is any one of the Clonmacnois slabs displaying so many anomalies as this from Gallen Priory. No Clomnacnois sculptor would have added the little cross-bar to the base of his cross. Nor would he have prolonged the lower line of the inscribed panel to the right, so as to spoil the mitre-joint at the lower right-hand corner. There are two parallels at Clonmacnois for the unequal number of lines in the vertical and horizontal bars (here eight lines in the horizontal and, apparently, six in the vertical), but not in crosses of this type; and indeed, owing to the fracture of the slab, it is by no means easy to see how the sculptor got over the difficulties he had created for himself by this piece of perversity. Another anomaly is the mitred joint in the top limb of the cross, corresponding to a butting joint in the side and bottom limb; it makes one hope that some day the other half of the slab may come to light; for otherwise it is impossible to guess what happened to the innermost line of the circle, in the upper left-hand corner, when the design was complete. These remarks will be sufficient to show that though the design on this slab is almost of the simplest possible - a plain cross in a square panel - it really is one of very great interest.

No.2. Evidently, as Mr. Armstrong says, the shaft of a standing cross. Probably it was meant to stand against a wall, which would account for the reverse face being blank. The panels on this cross are in many respects very noteworthy. In the upper panel, the plain voided circular centre to the cross in the middle is unique. Was it suggested to the sculptor by a cruciform reliquary? In the middle panel, the opposing animals appear to me to have a distinct Scandinavian "feeling" about them, though it would require very minute examination of details - such as the shape of the eyes, and the nature and treatment of the interlaced tails - to determine with certainty the art-school to which they properly belong. The little panel in the centre of a diaper-field is also unusual in the Irish crosses, though there are examples in Wales. Some of the Iniscaltra slabs are covered with diaper in a similar way to the lower panel of this cross-stem; but in none, if I recollect aright, is the diaper founded as here on the triangular key-pattern.¹ In short, this extremely interesting stone raises almost as many questions as its neighbour.

No. 3. Here again we have a well-established Clonmacnois type. The cross with circular expansions, the terminals being rather larger than a semicircle, and with eared loops at the angles, the stem formed of seven lines, two, three, and two, and the terminals containing a key-pattern, is a common late tenth-century type at Clonmacnois. But in none of the Clonmacnois stones of this type does interlacing work occur in the central expansion; and very few of them bear inscriptions written in the elegant alphabet that shows an angle at the bottom of O, ?, and the looped letters. As Mr. Armstrong observes, the notation of the palatalized n in the dative case is not common in these early inscriptions, which adds a philological importance to the artistic interest of this slab.² The drawing is not quite large enough to allow us to be sure about the design of the central interlacement. It looks like a simple fret of four closed loops; but usually, when such patterns are studied minutely, subtle little complications make their appearance, turning the banal groundwork into a design full of ingenuity. It would be interesting to know if this be the case on the stone of Bran. May we hope that Mr. Armstrong will give us a detail drawing of the centre pattern of the cross? The slab is so remarkable in every way that one hungers for the fullest possible information about it.

No. 4. Evidently a [Celtic?] cross in a Greek fret border, like the slabs of Tuathgal and Findan at Clonmacnois. But with regard to the inscription, I confess that I do not see how Bennan could be equated to Benén. One would think more willingly of Aedh Bennan, King of West Connacht at the beginning of the seventh century, or of the King of Ir-Luachair of the same name mentioned in the Annals of Ulster, A.D. 785 (Rolls Series ed., vol.1, p. 260):³ though both of these dignitaries belong to a date too remote for the slab to belong to either. The third letter, however, seems to me more like R than N. These characters, and h, often approximate in shape to one another, and have to be carefully distinguished; and this letter does seem in the drawing to be slightly differentiated from the two n's that the name contains. Bernan would be a diminutive of Bern (the name of the father of Cern, who murdered King Flann's son in 911). The same name is found in composition in such forms as Berngal, &c.⁴

No. 5. - This cross is sui generis. There is no parallel, at Clonmacnois at least, for the prolongation of one limb only outside the circle. The inscription is tantalising: would it be possible to read mailcraib[e]? This was the name of the lord of Ui Tortan in Meath, ob. 917 A.D.

Nos. 6, 8, 9. - Pictures of standing high crosses. No. 6 is, I think, unique in showing the crosses of the thieves - no doubt, Mr. Armstrong is right in this interpretation. In Nos. 6 and 8 the wheel is recessed behind the face of the cross, as in the south cross at Clonmacnois; in No. 7 the wheel and the cross are represented as being of the same thickness, and so mitred together.

No.7. - Here again we have, as in No. 3, the combination of interlacing work with the looped circular expansions; but in this example the interlacing work completely usurps the place of the key-pattern (which according to Clonmacnois canons properly belongs to the type), and occupies the terminals as well as the central expansion. It is a pity the interlacing patterns are so hopelessly worn out.

No.10. - The base and one arm of a handsome cross with a symmetrical plait of one cord upon it. It is a simple but effective pattern: the groundwork on which it is designed is obvious. Two simple interlacements of one closed loop (resembling the loops in the outer side-panels of the Draegenuch slab) are set side by side, and are cut and cross-tied at every third of the adjacent curves.

No.11. - It is lamentable that this fine slab is in such bad condition (antea, p. 174). It has apparently lain at some time in a thoroughfare, having been submitted to the tread of feet along its central axis. The design is evidently copied from a cumhdach - probably directly copied from an actual specimen. On closer examination, this becomes all the more likely. The border surrounding the cross is interrupted at the angles in a way which would be unnatural in a stone design, but is intelligible if it were copied from a metal plate, secured by corner clips to leather work, It is curious that the fret is triangular round the three upper sides, square along the base.

The square panel, with the interlacing of a single loop, at the upper left-hand corner (which is now the only clear part of the design), is evidently a sort of continuation downwards of the label bearing the inscription. I suspect that there was a similar panel in the corresponding position of the opposite side, now completely gone.

The cross itself seems to he divided into five panels - a small square space in the centre, and a panel in each arm - each panel bearing a separate knot-pattern upon it. These knots seem to be founded on a scheme of circles, broken at intervals with cuspidal interlacements:
a common pattern, though on the Gallen slab the details cannot be made out with certainty, owing to the damaged condition of the slab. Nor is it quite clear whether the withy-bands are entirely separate from each other, or whether they are cross-tied over the divisions of the panels. The latter is the more probable, especially as the outline of the cross itself is obviously made by the withy-band, interlaced on itself at the ends of each arm. It is likely that were the slab perfect it would be found that the outline of the cross, the outlines of the contained panels, and the interlacing devices ornamenting the surface, are all produced by the windings of a single ribbon.

As for the panels in the angles of the cross, that in the upper left-hand corner is, I fear, hopeless. In the right-hand corner I seem to see, not an interlacing pattern, but three cloaked figures, resembling those in some of the scenes on the high crosses - e.g. the central panel in the east face of the stem of King Flann's cross at Clonmacnois: but the figures in the Gallen slab are, I think, standing sideways.

The lower panels are more promising. In the left-hand panel is a knot of very extraordinary design. The combination of vertical and oblique lines in the groundwork of the pattern is very peculiar, and there are various irregularities of detail that make the analysis of this knot a matter of extreme difficulty - a difficulty enhanced by the worn state of the slab. In the description of the opposite panel, I venture to think there is an accidental lapsus calami in Mr. Armstrong's paper: the panel contains, not a cross, but, what is much more unusual in such a position and on such a monument, a crucifix.

These few remarks will help to indicate the great value of the monuments which Mr. Armstrong has added to our record, and will, I hope, stimulate others to follow his example. There may yet be many such cross-slabs lying hidden in some of the remoter cemeteries and monastic sites of the country.

Notes:

1. A similar diaper is, however, found in the head of the south cross at Clonmacnois.
2. Compare, however, Draegenuch, in No. 11.
3. This reference is concealed in the Index to the Rolls Series edition, in the first place by being out of its proper alphabetical order, and, in the second, by a numerical misprint in the date.
4. I may, perhaps, record a question that has crossed my mind in spite of the care with which the inscription has been copied, namely, whether it might not be possible to read: IARNAN - the name of a man murdered in 1003 A.D., "in the doorway of the oratory of Gallen