revised. We make our criticisms on these in no carping spirit, but rather in the hope that he may soon have an opportunity to reconsider them in a second edition.

Mac Shimmin (p. 27) betrays, we think, a Norse rather than a Scriptural origin. Like Simmonds in English, it points to the Norse Sigmundr as its source. (See Flatey Jarbok passim.) It is not very usual among Celtic peoples to find a Scripture name preceded, in early times, immediately by a Mac. The Clucas on p. 24 is more probably for Mac Giolla Lucas than for Mac Lucas. Costain (p. 29) and its early forms, Mac Coisten, etc., are, doubtless, derived from the Norse Eysteinn, which has given the Highland clan Huisten and the name Mac Quiston, has become Justin in Ireland, and has nothing whatever to do with Augustin. (See Cleasby and Vigfusson's Icelandic Dictionary, sub voce.) As to Mylechreest and Mylvorrey, we prefer to find in them Mael (=tonsured) rather than Mac Giolla. Milroy, Milvain, Milligen (= Maolagain), Milrea, and a number of similar names in Scotland, are cognate forms, and certainly involve Mael. McGuilley Chreest, McGuilleyorrey, are not to be identified with the preceding. Mac Vorrey is more probably from Murchadh, which gives Murray in Scotland, and Murrough or Murphy in Ireland.

Mr. Moore's reading of O'Donovan's Introduction to the Poems of O'Dubhagan and of Huidhrin has led him into making numerous false analogies. Crow (p. 36) has probably nothing to do with Fiachan, but is much more likely to be the Manx remnant of Mac Ruadh; cf. Highland McCroy. Fargher (p. 37) is, no doubt, the well-known Highland Fear Char which we find in Farquharson and McErrocher (Mc Fearchair). It has nothing to do with "Ferg" (violent), but involves Car=friend, and the intensitive prefix Fer (Welsh Gwr, Gaulish Ver): cf. Sanser. Su-Caru, Gaulish VeniCarus, Armorican Hen-Car, Welsh Caratacus, Cungar, Irish Findchar. Fergus also has nothing to do with Ferg. The "gus" occurs in Aongus, Aedgus, etc., and the Fer is the same as in the preceding name.

Kinley (p. 49) and McKinley are probably the same as the Highland McKinlay, and are formed from Findlay (Finn-laech). They have nothing to do with Mac Cinfaolaidh, which would give McNeilly. Mr. Moore cites (p. 57) a Finlo from the Statute Law Book of 1504, which is, doubtless, the same name. Alan (p. 50) is not a Norman but a Breton name. McCash (p. 59) looks sufficiently like the Highland McCosh to suggest a like origin. McCalbach (p. 60) has nothing to do with the Latin Calvus, but is most probably the same name as the Highland McKelvey McSelbach (Coll. de Reb. Alb.). Cowell, Coole, and McCoil, McCowle, and McQuill (p. 61), are, like the Highland McCool, forms of McDugal. Regan

2

1

1 Mac An Iosaich gives both McIntash and McCosh.

2 McKimmy, name of Lord Lovat, Chief of the Frasers, was Ior McShimi (Coll. de Reb. Alb.); cf. McKittrick McSitric, etc.

=

(p. 64) has nothing to do with Riach (i.e., Riabhach, grey), but is a very early Irish name. Mylrea (p. 65) presents no difficulty, but, like the Highland Milrea, is simply Mulriabhach: McGilliriabhach: McGilrea and McIlreevy, are well-known Scotch and Irish names. Both Mull and Giolla (Guilley) are compounded with colouradjectives, e.g., Mul-ruadh-Milroy, Mul-dubh Mulduff, Giollaruadh-Gilroy (cf. McIlroy). The ancient signification of Giolla is not probably "youth", but akin, as it seems, to Giall, it meant probably "hostage", or one captured in fight. It is thus cognate with the Teutonic form Gisal; modern Germ. Geissel, with the same meaning. It appears in Teutonic names, e.g., Gislbert (Gilbert), with the same import. If this be correct, names formed of Giolla and various colour-adjectives indicate that their bearers were captured in war, and thus "foreigners". Quilleash, Cuilleash (p. 74), cf. McLeese (p.91), are probably the same as the Highland McCuleis, McLeish, McAleese (McGiolla iosa), son of the servant or hostage

of Jesus.

McAvoy is not a contraction of Mac Aedha Buidhe (p. 67), but a short form of Mac Gilla Buidhe; cf. McAreavy, McAfee. The name Mac Effe, on p. 78, is one of the many forms taken by Mac Gilla Dubhtach, e.g., McAffee, McGuffie, McHaffie, Mehaffy. Mac Lynean (p. 78) is probably the same as the Scotch McLennan, and is for Mc Gilla Finnan. Mac Lolan is also, as Mr. Moore suggests, for McLellan Mac Gilla Fillan.

Mr. Moore is somewhat more fortunate with his Scandinavian etymologies; but if he had looked at Cleasby and Vigfusson's Dictionary he would have seen that Ottar involves Herr (meaning host, people), and has nothing to do with sword. If Mr. Moore had known that the Welsh Lloyd means "grey", he would not have compared it with the "ljotr" in Thor-ljotr. To trace Christian to an Icelandic Kristin requires a good deal of faith. It is to be sought more probably in some form beginning with "r", and involving the common Scandinavian suffix, "stein".

Mr. Moore has assumed (p. 93 et seq.), without evidence, a wholesale plantation of Hiberno-Norman families in the Isle of Man. On the same principle he would naturally conclude that the McWilliams, McWalters, etc., in Scotland, and the Guilliams and Guatkins of Wales are descendants of the Norman De Burgos. O'Donovan has evidently misled him here.

If Fairbrother (p. 103) is a translation of Beaufrère (=brotherin-law), of what is the personal name Fairclough a translation? A similar difficulty attends his tracing Freer, Creer, to the French frère; cf. the Irish name McCreary. Arin-biaurg (p. 114) has nothing to do with hearth, but contains the common Norse name-element Orn (poet. Ari) eagle, and Björg-defence; cf. Orn-ulfr, Arnold, etc.

Frior in personal names, as Friede in German, does not mean "fair", but "peace", or in an earlier sense, "inviolability". Doncan (p. 116), like the Scotch Duncan, is, as is well known, simply Donchadh, involving in its last element Cath-battle; ancient form,

Catu; cf. Welsh Dinocatu. It has nothing to do with the colour Dun, or with Chu or Gal. Finlo (p. 116) is, doubtless, the same as the Gaelic Fin-laech-fairhero. It gives Finlay, Findlay, McKinlay, etc. Symond (p. 116) is, doubtless, the Norse Sig-mundr, and has nothing to do with the Hebrew Simeon.

We have not much space to deal with Mr. Moore's etymologies for place-names; but a great many are questionable. Broogh (p. 132) in place-names in Ireland and Scotland means, especially when combined with personal names, a large house. Broughshane is not, as Joyce says, John's Border, but John's House. This meaning is well known locally. The "Broogh jarg mooar" (cited p. 143) means, most probably, not "the big, red brow", but "the big, red house." The word is known in Scotland with this meaning.

Garee (p. 135) is probably Garbh rough land, and takes the form Gariff and Garvy. Does Alt (p. 135) ever mean stream? Peel is not a Celtic word, at least not a possible Irish or Gaelic form. No pure Goidelic word begins with p. Braid (p. 139) is a well-known word for an opening in hills, and is, doubtless, nothing more than Braighid=the neck or throat. As Irish Drum is to Manx Dreeym, so is Irish Mull to Manx Meyll. Its meaning, when topographically applied, is bald headland; i.e., with no trees on it. It has nothing to do with any Scandinavian word, as Mr. Moore assumes, p. 41. Rhenshent (p. 143) is translated "holy ridge". In the Chronicon Mannice (quoted at p. 12) it is given as an equivalent of the Welsh Hentraeth-old strand. The "shent" must, therefore, represent the Irish "sean"=old.

We have dealt at some length with Mr. Moore's etymologies; but his book has its great value not in these, but in the array of carefully sifted and dated names which he gives us. He has rendered

the work of those who come after him easy. The labour involved in putting this book together must have been enormous, and students of personal and topographical nomenclature must feel deeply grateful to him, while all who are interested in the history of these islands will welcome the work as a source of fresh information upon many important points. It has been no pleasant task to point out defects, and it is with true rejoicing that we express our high sense of the historical value of this great collection of Manx personal and place-names.