Page:Notes and Queries - Series 11 - Volume 4.djvu/49

 ii s. iv. JULY 15, mi.] NOTES AND QUERIES.

of Gloucestershire, 1683 (under Cooke of Highnam), agrees with the Visitation of 1623 that Dorothy Lucy was daughter of Rowland, but makes her mother as per Sir E. Brydges the niece, and not daughter, of Lord Chandos.

The whole position is curious, and wants light, which perhaps some of the corre- spondents of ' N. & Q.' may be able to cast upon it. W. D. PINK.

" GIFLA" : ISLEWORTH : ISLINGTON.

THE smallest assessments in the ' Tribal Hidage ' are those of 300 hides. Five regions answer thereto. These are 9. East Wixna ; 15. Sweordona (MS. -ora) ; 16. Gifla ; 17. Hicca ; 26. Fserpinga. (I am correcting MB. BROWNBILL'S numera- tion because he has set down " Fserpinga " out of its true place.) The p of Fcerpinga I take to be a misreading of n,* and the emended word Fsern-inga correctly presents the infected form of the princely name of Farri-. The true order of the MS. is 24. Ytenaga (MS. ynetunga) ; 25. Dorsaetna (MS. aroscetna) ; 26. Faerninga ; 27. Bil- linga (MS. bilmiga) ; 28. Sutferigna (MS. widerigga). It is fairly certain from this that Faerningaland lay in Hampshire, to the north of Meanwaraland,f and the east of Gewissaland, the chief city of which was Venta Belgarum, Wintanceaster. Of these two ancient regions, the first-named was probably assessed among the 1200 hides of Ytena ga ; the second was omitted from the list.

Starting from the English Channel, on the east of Ytene, or Ytena ga, we find Billingaland (West Sussex), Faerningaland (North-East Hampshire), and SuSerigna- land (West Surrey). Still proceeding north- wards, we look in vain for an exact reflex of the nomenclature of the MS. till we come to Hitchin in Hertfordshire. The survival of that name enables us to identify the western part of the county as Hiccaland. For, just as Ytena had been weakened to Ytene before A.D. 1100, so also must Hicca (gen. pi.) have been reduced to Hicce before 1086, when the Norman assessors called it Hiz. The Norman z was pronounced like

century n, c, and p.
 * Cf. 10 S. x. 227, where I refer to eighth-

t O.E. mean=Ol& Icel. *maun=Danish mtin, Mori, Seeland, Falster, and Laland were comprised in luthes-laeth MSS. Vithes and WWies). luthes is the gen. pi. of *Iuthja.

our ts : cf. fiz, fitz ; assez, assets. Ap- parently z was the nearest representation of final O.E. palatal c (i.e., -ce) that the Nor- mans could contrive. Its use proves that at the time of the Survey Hicce was pro- nounced nearly like Hitch- : cf. bicce, fticce, wicce, bitch, flitch, witch.

Taking up our position in Hiccaland, we will now inquire where its neighbour Giflaland lay. To the westward is Ciltern- saetnaland ; eastward lies East Seaxna- land ; southward we look right across Middlesex and the Thames to SuSerigna- land ; and northward is Herefarnaland, the herefinna of the MSS., which lay partly in Northamptonshire. Sweordonaland I have not yet identified satisfactorily.

It would appear, then, that Middlesex has been ignored. But there is an ancient and frequent scribal error with which we are- all conversant, and which is due to the reading of long s as /. We have seen already that the scribe of the late tenth - century MS. misread s, together with the first minim of u in sufterignaland, as w ; and in " Gifla "" he has certainly made the mistake of writing / for s. I propose, therefore, " to alter the evidence of the MS." once again, and to- emend Gifla to Gisla. This done, I identify Gislaland as West Middlesex, and for the- following reasons.

The O.E. initial g was palatal, and was pronounced something like y in yes, yet, yare. This admits of its absorption in a following long palatal vowel i, and also explains the seventeenth - century scribal' form " Thistleworth " for *Yisleworth,. which equals the " Gistelesworde " of Domes- day Book, and the " Gislhereswyrth " of an eighth - century charter in Birch, * C. S".,' No. 87. This town is now called Isleworth (pron. Izel-). In the north of the county of London we find Islington (with correption or shortening of I). This town appears three times in the Great Survey : twice as " Isendone," and once in the truer form of "Iseldone" (130 b, col. 2). "Isendone" exhibits the French confusion of the liquids n and I : cf. Nicole for Lincoln ; O. French nivel for Latin libella, our "level"; O. French posterne = posterle, for Latin posterula. " Iseldone " = *Isladone, i.e., Gislandun, Ylsla's down, now Islington. The wavering between medial -an- and -ing- in place-names is a well-known pheno- menon : cf. Abbandun, Abingdon ; Seccan- dun, Seckington ; Niwantun, Newington. The etymon of Islington, viz. Glsla, is the- short or pet form of the name Glslhere, iiu " Gislhereswvrth."