I found an interesting article on the matrilineal descent options that might link Banquo to the Stewart line (rather than going with assumptions that it was fictitious) here: http://www.mckinneyandstewart.com/genealogy/getperson.php?personID=..., and I've also pasted it below.
I'm putting it here as a record; and for anybody so inclined to talk back to; or research further:
THE STEWARTS, VOLUME 21 No. 2 (2001), pages 97 to 100.
By Henry Stewart Fothringham
Walter Fitz Alan
The family tree of the
First High Steward of Scotland
The problem of who exactly was Walter Fitz Alan, Ist High Steward of Scotland, has perplexed many minds over the years and several different conclusions have been reached by different genealogists. Part of the problem has been the sparseness of the data and the different interpretations capable of being put upon them, compounded by what looks to some like deliberate obfuscation of the facts. In this brief paper the writer throws down the ancestral gauntlet for historians to pick up and challenge their own pre-conceived ideas by re-examining all the available sources with an open mind. The editor looks forward to receiving informed responses on both sides of the debate.
Genealogists up to the eighteenth century seemed happy to accept that the Stewarts were descended from Banquo, Thane of Lochaber. This was derived principally from Hector Boece's History of Scotland and the Chronicles of Holinshed but they both drew on older material for their narratives. The latter was the source used by Shakespeare for Macbeth.. These notions, which accorded with the 'Secret Knowledge' preserved by only a very few persons at any one time, were also supported by a 'folk memory' of the events and by ancient lore and legend, but by very little documentary evidence. The idea was subsequently discredited by the discovery of seemingly conflicting evidence.
The idea that supplanted the Thanes of Lochaber thesis was that the Stewarts were descended from the Senescalls of Dol in Brittany and there is certainly evidence to support this interpretation. The evidence was brought together and published by George Chalmers in his Caledonia (1807-24), the manuscripts of which are now in the Advocates' Library in Edinburgh. What has not been generally realised, however, is that there is room for both interpretations, not forgetting that Walter son of Alan had a mother as well as a father. Victorian genealogists seemed to have had a fixation about patrilinear descent, quite overlooking the fact that the maternal lines were of equal, or even greater, importance in early times. The principle by which a cousin or nephew on the mother's side usually succeeded as king persisted in the Scottish monarchy for hundreds of years. Further, the new theory propounded in Chalmers' Caledonia so busied the minds of genealogists trying to prove it to be true that they forgot about trying to disprove the old theory; had they tried to do so they would have found that it was not going to lie down and die as quietly as they expected. At about this point also, conspiracy theory kicks in, suggesting that the House of Hanover wished deliberately to suppress the Scottishness of the Stewart ancestry in favour of a European descent.
The genuine confusion was begun by mixing up two people called Alan, who turn out to be Walter's father and his maternal grandfather, and treating them as one and the same person. Because of this it was assumed that the older Alan was his father, whereas he was actually his mother's father. Walter's father was Alan, Thane of Lochaber, who died in about 1155. Walter's mother was Adalina of Oswestry, daughter of Alan Fitz Flaald de Hesdin, Sheriff of Shropshire (d. c.1122). It is therefore through his mother and not his father that Walter was descended from the Senescalls of Dol. That line of ancestry can be traced back an astonishingly long way, to the lst century AD and, if we are to believe the 'Secret Knowledge', even further. According to what you choose to believe, the descent goes back, like the famous Macleod family tree at Dunvegan with which it merges, at least as far as Adam and Eve.
On his father's side Walter's male line of descent leads back through, among other people, Fleance, Banquo (his great-great-grandfather) and King Aedh to Kenneth MacAlpin (10 generations) and thence back through that family tree, once again arriving rather optimistically at Adam and Eve. The reason for the descent from Banquo having been discredited since the late eighteenth century was that it was assumed to conflict with the Dol descent, whereas the two are perfectly compatible, as shown in the accompanying genealogical table.
Adelina's father, Alan de Hesdin, was Flaald, hereditary Steward of Dol in Brittany. In the early 1100s Alan was Baron of St Florent, Saumur; his early forebears were the Counts of Brittany, who were kin to the Merovingian Kings of the Franks. It is with Flaald and his wife that the genealogical confusion usually stems. Flaald was married to Aveline, daughter of Arnulf, Seigneur de Hesdin, but some peerage registers (including Burke's and the 1858 History of Shropshire) erroneously show Aveline as having been the wife of Flaald's son, Alan. The fact appears to be, however, that Alan Fitz Flaald was born with the 'de Hesdin distinction inherited from his mother, Aveline (Ava) of Picady. Her status is confirmed in the Cliartulary of St George, Hesdin. When Aveline's father, Arnulf (brother to Count Enguerrand de Hesdin), joined the crusade in 1090, Aveline became his nominated deputy in England. She was known as the 'Domina de Norton' (the Lady of Norton) and her son, Alan Fitz Flaald, was Baron of Oswestry during the reign of Henry I. As correctly detailed by Chalmers in his Caledonia, Alan was married to Adeliza, the daughter and heiress of SheriffWarine of Shropshire, thereby inheriting that office. Warine's arms consisted of a field azure and argent, the same tinctures as the fesse adopted by his Stewart descendants. This, and not the supposed chequer-board theory, was the true origin of the Stewart arms.
Alan de Hesdin's uncle (Flaald's brother) was Alan, hereditary Seceschal of Dol, who was a Crusade Commander and died on crusade in 1097. From his son, William Fitz Alan, the Fitzalan earls of Arundel descended, while his daughter, Emma, married Walter, Mormaer (or Thane) of Lochaber, the son of Fleance and grandson of Banquo. Walter of Lochaber died fighting at the side of Malcolm Canmore at Alnwick in 1093. Some sources make the error of showing him as being High Steward, which is either a blunder or, more probably, a 'pious' invention to enhance status.
From this initial marriage between the Scots and Breton families (c.1085) emerged Alan, Thane of Lochaber. Born c. 1088, he cemented a further alliance with the Breton house by marrying Adelina, daughter of Alan Fitz Flaald. It was their son, Walter, who succeeded to the Shropshire inheritance. By virtue of his Lochaber heritage and responsibilities, Walter was summoned to Scotland by his friend, David I, in c. 1136 and one of his first task was to guard the western coast from Loch Linnhe to the Firth of Clyde against the Norse invasions. On 22 August 1138 he fought at the Battle of the Standard, near Northallerton in Yorkshire, when the Scots were heavily defeated.
Banquo's son, Fleance, was the first husband of Princess Nesta of Gwynedd but, after Fleance's death, Nesta married Osbern Fitz Richard, grandson of Guiomarc, Comtede L6on, who held substantial estates in Dol. In later life, Guiomarc became a Benedictine monk of St Florent at Saumur, where Flaald of Dol was to become the Baron. Indeed, the family ties were very close, which is how Osbern came to marry Nesta.
In 1080 Flaald and his brother, Alan, Seneschal of Dol, consecrated St Florent Abbey. Two years later their younger brother, Rhiwallon of Dol, a monk, became its Abbot. In 1102 Flaald was present at the dedication of Monmouth Priory and Flaald's son, Alan de Hesdin, founded Sporle Priory in Norfolk as a cell of St Florent Abbey. Powerful families habitually intermarried with their close cousins in order to consolidate their power and possessions; Osbern and Nesta's son, Hugh, married Eustacia de Say of Clun, while William Fitz Alan married her sister, Isabel de Say. As previously stated, William's own sister, Emma, married Walter of Lochaber, son of Fleance and Nesta. This all sounds fearsomely complicated but the essential lines are much more easily followed in the genealogical table.
The writer is grateful to Laurence Gardner for permission to use his research and to Prince Michael of Abany for permission to reproduce the genealogical table. It was the exertions of Laurence Gardner which have brought this matter to public attention once more. He researched, among many other neglected sources, the family archives of Germaine Elize Segers de la Tour dauvergne and her husband, Michael Stewart of Annandale; the pre-1792 records of Lyon Court (publicly available on request); the Diocesan Archives of Angers; the Chartulary of St George at Hesdin; and the Chrtulary of St. Florent.
The following tables give a skeletal lay-out, generation by generation, as far back as Aminadab on Walter's mother's side and on his father's side to Ere of Dalriada. Among the early names there are many variations in spelling and some people have more than one name.
I: Patrilinear Descent of Walter the High Steward
Note.- each succeeding generation is the son of the previous one.
24.Ere of Irish Dairiada (Dal naraide)
23.Fergus Mor Mae Ere, d.501
22.Domangart
21. K. Gabran of Dalriada, c.548-558
20.Aedan Mac Gabran, d.608, m. Ygerna de Acqs
19.Eochaid Buide (younger brother of the historical King Arthur)
18.Donald Brec
17.Domangart
16.Eochaid, d.696
15. Eochaid
14. Aed the White (Aed Find)
13.Eochaid the Poisonous, d.781
12. Alpin
11. K. Kenneth MacA]pin
10. K. Aed (Aeth), d.878
9.Doir, b.870-d.936
8.Murdoch, b.900-d.959
7.Ferguard, b.929-d.980
6.Kenneth, b.960-d.1030
5.Banquo, Thane of Lochaber, b.990-k.1043
4.Fleance, Thane of Lochaber, b.1020-d. c.1064
3.Walter, Thane of Lochaber, b. c.1045-d.1093
2.Alan of Lochaber, b. c. 1088-d. 1 153, father of-
1.WALTER FITZ ALAN, 1st HIGH STEWARD OF SCOTLAND, d. 1177
II: Maternal Descent of Walter the High Steward
Note.- each succeeding generation is the son of the previous one,except for 2. Adelina, who was Alan's daughter.
34.Aminadab, 2nd century A.D., m. Eurgen, dau. of Lucius, son of K. Coel I.
33.Castellors
32.Manael
31.Titurel
30.Boaz (Anfortas)
29.Frotmund, k.404
28.Faramund, Lord of Franks, d.430
27.Fredemund
26.Nascien I, K. of the Septimanian Midi
25.Celedoin
24.Nascien II, K. of the Septimanian Midi
23.Galains
22.Jonaans
21. Lancelot
20. Bars the elder
19.Bars the younger
18.Lionel
17.Alain
16.Froamidus, Count of Brittany, c.762
15.Frodaldus, Count of Brittany, c.795
14.Frotmund, b.850
13.Flotharius
12. Adetrad
11.Frotbald, c.923
10.Alirad
9.Frotmund, c.982
8.Fretaldus, c.1008
7.Frotmund Vetuies, c. 1052
6.Fratmaldus the Senechal
5.Alan, Seneschal of Doi
4.Flaald, Seneschat of Doi
3.Alan Fitz Flaald, de Hesdin
2.Adelina of Oswestry, motliet- of-.
1.WALTER FITZ ALAN, 1st HIGH STEWARD OF SCOTLAND, d. 1177
