an abbot of a lay tribe, by holding a monastery was not in the Celtic church guilty of an act of lay usurpation over the church, but was only exercising his regular legal rights.

It would be beyond the limits of this paper to discuss the peculiar custom of the rule of the selection of the fittest "ecclesiastical tanistry", as it may be called, as to how and when an existing abbot was displaced by another and fitter person making his appearance. The chief point of interest is that the Irish laws here give us an account that is obviously genuine of the organisation of the Celtic monasteries. We see here, as we see nowhere else in the same degree, the ecclesiastical system as it existed under the Celtic rule. No part of the organisation is brought out more strongly than this, that the whole system was based on kinship, or on the relation of the churches to each other by the tie of kinship. The idea that the monasteries were related because they belonged to the same order, Benedictine or Cluniac, never entered the Celtic mind. As in the lay tribe, kinship, descent in theory from a legendary ancestor, united the tribe, and bound it together, so in the tribe of the saint, kinship, descent in theory from a legendary saint, was the basis on which the tribe was united, the link that bound it together. This fact may furnish one of the reasons for the great care and attention that were paid to the genealogies of the Welsh saints.

The rules also bring out another very important point. The succession to the Celtic abbacy was hereditary, not elective, or, more accurately, elective out of an hereditary class, the descendants of the founder. While any one might become a bishop, priest, or deacon, no one could become an abbot except, to use a modern phrase, he was of "founder's kin". From among the founder's kin the fittest, in theory, succeeded. abbots of the mother church and the abbots of the offshoots were all called "conharbas", "coarbs" (joint heirs or coheirs). From among those who represented the founder's kin the abbot was elected according to cer

The

tain definite rules. If one of the "coarbs" happened to be a bishop he might be elected to the abbacy; he would, at least, be eligible for election. But unless he was a "coarb" (one of the founder's kin) an apostle would have had no chance of being elected a Celtic abbot.

It is worthy of notice that in the whole of this legal account of the Celtic Church organisation there is no mention of a bishop. Throughout the Corus Bescna it is doubtful if the word bishop occurs; it may, therefore, fairly be inferred that the Irish Celtic Church, as then constituted, knew nothing of episcopal government or episcopal rights. Neither bishop nor pope had any right of electing, or interfering with, or opposing, the election of any abbot to a Celtic monastery. When in later times we meet with episcopal and papal claims to rights and jurisdiction over monasteries, we may feel sure that such claims are traces of the rule of the alien over the national church. The great features of the Celtic Church, as shown by the Irish laws, were, therefore, three,(1), that it was not independent of, but mixed up with, the civil organisation of the tribe; (2), that it had monastic rule by abbots; (3), that there is no trace of anything like a claim to or exercise of episcopal jurisdiction or supremacy. These are all the theories and ideas of a later age and another Church. This tribal as opposed to a national character of the Celtic Church is well described by an Irish writer :"The nation was split into independent tribes, the Church consisted of independent monasteries. The civil chaos out of which society had not yet escaped was faithfully reproduced in a Church devoid of hierarchical government; intensely national as faithfully reflecting the ideas of the nation; but not national in the ordinary acceptance of the term, as possessing an organisation co-extensive with the territory occupied by the nation."

1 Introduction to the Rolls Edition of the Ancient Laws of Ireland, vol. iii, p. lxxvi.

(To be continued.)

EVIDENCES OF THE

BARRI FAMILY OF MANORBEER, PEN-
ALLY, AND BIGELLY.

WITH OTHER EARLY OWNERS OF THE FORMER,
IN PEMBROKESHIRE.

BY SIR GEORGE DUCKETT, BART.,

Knight of the Order of Merit of Saxe-Coburg-Gotha, Officer of Public
Instruction in France, and Corresponding Member of the
Society of Antiquaries of Normandy.

(Continued from p. 206.)

IN tracing the descent and record-evidences of the Barrys of Manorbeer and Olethan we arrive at the conclusion that up to about the year 1325 they were unmistakeably the same people, and that the Pembrokeshire possessions continued with the same descendants (English and Anglo-Irish) down to that time, for irrespective of documentary evidence, it is hardly to be supposed that such belief should arise from a mere coincidence of names. We shall assume, therefore, that up to that date the lords of both properties were the same persons.

It is quite possible that after the acquisition of their Irish estates, over which they exercised almost regal sway, the Barrys were less troubled about their Welsh property, though the evidence is as conflicting on this as on some other points; but it is unmistakeably shown that towards the end of the reign of Edward III this last had entirely passed out of their hands. It is said of William de Barri (third in descent from the first known head of the family) that being a favourite of King John he assigned his Irish estate to his brother Robert, and lived in England, Kent being the district in which he located himself. This assertion may be true; equally that Manorbeer may have been too remote a residence for his purposes. At a later date we

also find some of the Irish estates conveyed to a brother by another of the family.

We have already spoken of William de Barri, one of Arnoul de Montgomery's adherents, as the common ancestor of the family. Robert, the eldest son,' was concerned with Fitz-Stephen in the first invasion of Ireland, being one of the advance detachment of the expeditionary force which first set foot in that country. He was slain at Lismore in 1185. Philip, the second son, went over to Ireland a few years later, as did also, in company with Prince John, as secretary, the younger son, Giraldus de Barri (or Cambrensis).

We give, in tabular form, the first four indisputable and authentic generations of the Barri family :

A.D. 1140, 1166, 1176-85. Philip de Barri occurs at some date before quitting Pembrokeshire for Ireland, in 1185, as witness to an undated inspeximus charter of Peter de Leia, Bishop

1 Walter is recorded by some to have been an elder brother by a former wife, making Robert the eldest son by the second marriage. Of this Walter nothing is otherwise recorded.

of St. David's, granting and confirming to William Fitz-Maurice Fitz-Gerald the office of Dapifer of the bishopric, in succession to his father, Maurice Fitz-Gerald (the ancestor of the Geraldines), who had held the same under his brother David. As Peter, the second Bishop of St. David's (Suffragan to the see of Canterbury), occupied the see from 1176 to 1198, it is manifest that the dignity was conferred prior to Philip's departure for Ireland. (S. Gormanston MSS., H. M. C., iv.) He and Odo de Carew (a name of great antiquity in Pembrokeshire) married two sisters, daughters of Richard Fitz-Tancred, of whom it is said (Ang. Sac., ii, 468) "tunc temporis in partibus illis magnus habebatur."

A.D. 1146-1215. Giraldus Cambrensis, or Sylvester Giraldus de Barri, the early and well-known chronicler, youngest son of William de Barri, was born at Manorbeer circa 1146, and, like his brothers, in descent maternally from Rhys ap Theodore, Prince of South Wales. His career is thoroughly authenticated by his own testimony, and leaves nothing to be questioned in that respect. It may be safely asserted that but for him we should have been in absolute ignorance of the earliest possessors of Manorbeer; and, what is of more consequence, the invasion of Ireland, in which they took a prominent part, would never have been so fully known, or its details so authentically established. It is true that we glean little from him in a genealogical point of view, beyond his immediate relations and kinsmen; but the bare record of Manorbeer as the place of his birth affords us a sure clue, and, as one may say, a safe starting-point on which to found the family history. But for him, again, the royal and

1 It was in 1176-7 that Peter de Leia, the Cluniac Prior of Wenlock, succeeded to the see of Menevia, and died in 1198, having presided over St. David's for twenty-two years. (Annales Cambriæ, p. 55.)

2 Dapiferatus.

3 To this grant Walter de Vinsor (sic) [Wyndesore] is also a wit

ness.

4 Giraldus de Barri seems to have written at least twelve or more treatises,-The Topography of Ireland (published by Camden); The Vaticinal History of Ireland, relating to its invasion by Strongbow, Fitz-Stephen, and Maurice Fitz-Gerald, and translated by Hooker in Hollingshed's Chronicle,-for which two works he collected the materials from the time he first went to that country, in 1184, in company with his brother Philip, and as secretary to Prince John, Earl of Moreton; the Itinerarium Cambria, Topographium Cambria, De Principum Instructione, Anglorum Cronicon; besides eight others, of which the treatise, De Sudoribus circa Sedem Menevensem, sets forth his troubles in respect of the bishopric of St.