the floods had swept away three of his poor comrades), "What a grand place for curling!" But the grass had begun to re-appear by the end of the week before a slight frost congealed the surface of the diminished pools. For ten days the Cathedral Library, over the eastern cloister, could not be heated because the stoke-hole was flooded, and the pump and hose were fully occupied in pumping from 2 to 4 inches of water from the aisles and nave into the cloisters, and for several successive days first through the cloister door and then through a window (from which some of the glass had to be taken out) into the N.E. channel of the Close. The water rose again to a height of 2ft. Sin. in the stoke-hole, already mentioned, near the cloister and chapter-house, in February, 1915; but it has been pumped out a second time, and by March 4th the fire was once more burning merrily. The cathedral organ (one of the greatest sufferers from the recent floods) was in part used again on Sunday, February 28th, and the whole of it on March 4th.

The effect of the reflection of the roof and windows of the interior of the nave in the sheet of water within was portrayed in the illustrated papers, and brought many visitors to the west and north doors and across the planks on Wednesday, January 6th. The paving about the middle of the nave was observed to be some 2 inches higher than its level at either end. The removal of the chairs from the nave has shown the true architectural proportions of the building as seen in eighteenth century engravings, a fact which was not lost to some artistic eyes, and we may hope that sketches were satisfactorily completed, to show the arcades more nearly as Elias de Derham planned them, than they have been seen for several generations. James Biddlecombe's drawing of 1754, engraved for Peter Hall's Picturesque Memorials of Salisbury, 1834, plate vi., shows the nave devoid of seats; and the font, near the west end. I wonder whether this place at the west end of the cathedral nave between the tombs or effigy of the younger W. Longespé and the

1 The Fabrick Account for July, 1662, shows that £2 12s. was paid after the Restoration for "carrying ye Font from London." There were further expenses connected with the font in Aug., 1662—“ nullibi autem comparet quid pro Fonte sit solutum." (Walton's Quarto, p. 297.)

so-called "Boy Bishop" to the north and the tombs or coffin slabs of Bishops Roger and Josceline de Bohun (brought from Old Sarum) on the south, was the site of the medieval "fontes" also. The printed Sarum Processionale rubric (1502, &c.) directs that the Easter Fire was to be blessed on Easter Even "at the column on the south side near the Font," the illustrated editions, more particularly those of 1517-1558, show the little fire blazing, in a shallow brazier with a foot, to the right hand of the celebrating priest, and the acolyte with holy-water and the thurifer next it, as shown in Salisbury Ceremonies (Camb., 1901), p. 82.

HUISH AND THE DOYNELS.

The following notes may be of service towards a history of Huish when it comes to be written. They consist of such references to the place or to its successive owners as are readily discoverable, for the most part in print, without discussion of the many difficulties in their interpretation and without any pretence to completeness. On the printed Pipe Roll (ed. Rev. Joseph Hunter) of 31 Henry I, that is to say, for the period from Michaelmas, 1129, to Michaelmas, 1130, occurs the entry (p. 22):

The name is right, and the county is right, but there is a further entry on the same roll (p. 41) which may render it less certain that, in the person of this Robert, we have found a direct ancestor of the name at Huish :

The reason for hesitation is as follows. In that truly delightful book "The King's Serjeants & Officers of State" Mr. J. H. Round explains (pp. 92 et seq.) that " About the middle of the 12th century, if not indeed rather earlier, Robert Doisnel was holding at least five manors by the service of performing its duties," viz., of the office of a marshalship at court; that by his carta of 1166 William Fitz Audelin claimed to hold, by the same service, "all the land of Robert Doisnel which had not been subinfeudated" (viz., less a knight's fee in Essex and a quarter fee held by John Gernun) as given to him in marriage with Juliana, Robert's daughter and heir; that Juliana died childless, whereupon, in 1199, her collateral heirs (Warbleton and Munceaus) fined for the inheritance; and that, in 1205, the senior of them did so obtain the lands which belonged to the serjeanty, in Hampshire, Cambridgeshire and Essex, less land (in Essex) which Juliana had given to the Hospitallers.

Now there seems to be sufficient evidence that the Doynels of later date in Wiltshire held no lands outside the county, and it is reasonable accordingly to suppose-since both the entries above must be taken to refer to one and the same person-that it is Robert Doisnel, the marshal himself, who is pardoned, in 1130, 10s. in respect of his lands in Hampshire, and one sixth that amount in respect of his land in Wiltshire, of which latter, however, there is, frankly, no further trace.

On the printed Pipe Roll (ed. Pipe Roll Soc., vol. 6) of 9 Henry II., that is to say for the period from Michaelmas, 1162, to Michaelmas, 1163, occurs the entry (p. 47) :-

Wiltescira .

Idem vicecomes reddit compotum de firma de Hiwis terra Roberti Doihnel (in Chancellor's Roll Dunnel). In thesauro 1.m. argenti.

The entry apparently does not recur. I am unable to explain it. It is the last entry on the roll for Wiltshire. In the previous entry the sheriff accounts for the "farm of Westberi. terra Escaett'." In the case of the farm of Hiwis, however, it is not described as an escheat but as the laud of Robert Doihuel. The sum accounted for, one mark, looks like a fixed payment. It might refer to the scutage of the previous year, imposed on all knights and assessed at one mark; but in an entry, two from the end of the same roll, the sheriff does account for the scutage, and calls it so, of Adam de Port, and it is not even certain that land, held as we have reason to believe Hiwis was held, was liable to scutage at all. If it were a regular payment due in respect of Hiwis it would recur, which it does not, on subsequent rolls; nor was it a payment on account, for although the words Et quietus est are absent (as they are in the entry relating to the farm of Westberi) the sheriff does not confess to further indebtedness with the phrase Et debet It might

be suggested that Robert was recently dead, leaving an heir in minority; that the wardship of this heir was granted away; and that the sheriff accounts for the issues in the interval between the death and the grant. It is consistent with the dates, as reckoned from the lower part of the pedigree, where the filiation is ascertainable and dates are known, that there should be a Robert, born

about 1128, dying in 1163, and leaving an heir born about 1158: -and the pedigree of course stands, whatever becomes of the explanation, seeing that it is quite immaterial-for the pedigree— once the heir is born, how long the father may survive the birth; but upon this theory the period of the sheriff's possession must have been brief, for the sum is small,

Thus all that this entry, in the absence of reliable interpretation, tells us, is that in 1162-3 the connexion, in some form, of Doynel with Huish had begun. It does not seem worth while labouring the point that here at any rate we cannot be encountering the Marshal. He was in all probability already long since dead, even though the entry ante-date Fitz Audelins' carta by a couple of years, and whereas a long line of Doynels held Huish, the Marshal left only female issue to survive him.

In the Liber Feodorum, or Book of Fees, there occurs (at p. 1486. of the printed edition of it known as Testa de Nevill) a list of Wiltshire Serjeanties, of which no original can be found (that is to say the original document from which the list was transcribed into the Liber Feodorum is no longer in existence) which can with great confidence be assigned to the year 1198.

In this List there are two entries which concern us:

Robertus Duinel [tenet] in Hywis ij carucatas et valent lxs.

Robertus Doynel [tenet] in Estgrafton dimidiam carucatam et valet xvs.

Here for the first time we find the tenure of Huish distinctly stated. The tenant there is included in the list of Wiltshire Serjeants, while the nature of his serjeanty will shortly appear. Further we discover by these entries that not only was a Robert Doynel holding in Huish by serjeanty-service, but in East Grafton also there is a serjeant of the same name. But in Domesday both these places are entered and each of them was then held by a Richard Sturmid, as one of the King's Serjeants! It is surely very remarkable and of no slight interest-even if the details escape us, though we fail to ascertain how for Sturmid there comes to be substituted Doynel, and though the nature of Sturmid's