f 'M'^'% ^'hl'^A'W ■^:.-<-''^ THE BRITISH ACADEMY English Place-Name Study Its present condition and future possibilities By Professor Allen Mawer \_Froin the Proceedings of the British Academy, Vol. JT] London Published for the British Academy By Huniplirey MOfbrd, Oxford University Press Amen Corner, E.G. Price One Shilling and Sirpence net ENGLISH PLACE NAME STUDY ITS PRESENT CONDITION* AND FUTURE POSSIBILITIES By Professor ALLEN MAWER Read January 26, 1921 1. Introduction. 2. Essential principles as laid down by Professor Skeat. 3. The extent to whicli these principles are now observed. 4. Tlie weaknesses of place-name study as now carried on : (a) AVorkiii^ on is(dated aieas. (b) Varying and imperfect principles of selection of names for treat- ment. (c) In relation to Topography. (d) In relation to History. 6. Principles for the future suggested by these weaknesses. 6. Advantages to be gained by organized work ou these lines. (a) To Linguistic Studies. (6) To Historical Studies. (e) Practical. 7. Examples from other countries. 8. Urgency of tlie matter. 9. Conclusion. § 1. Introduction. It is now some twenty years since the scientific study of English place-names may be said to have begun with the publication of the little volume on the place-names of Cambridgeshire written by Professor Skeat for the Cambridge Antit|uarian Society. In the interval, monographs have been published dealing with the history of nineteen other counties — four l)y Professor Skeat himself, four by Professor ^Vyld and his pupils, three by Mr. Duignan, and single volumes by other scholars. The time has perhaps come when we may fittingly take stock of what has been accomplished and in the light of the experience we have gained plan our course for the future. § 2. Essential principles as laid dozen by Professor Skeat. In his prefatory remarks in the Cambridgeshire volume, Skeat laid down the cardinal principles upon which all place-name study must be based, viz. (1) that the first step in the process of determining the history of a place-name is to make as wide a collection as possible of the early forms of the name. 440i>27 a"'" I'RtrCEKDIN-GS OF THE IHUTISH ACADEMY (2) The phonetic laws tli.-it govern the history oi' place-names being precisely the same as those that govern the history of other words, the student of place-names must deal witli the special problems before him in the light of a thorough knowledge of the history of English sounds from the earliest times to the present day. § 3. The extent to ivh'uh these pr'imiples are iwio observed. Ucccnt writers on place-names have, with more or less fiiithfulness, observed Skeat's first principle, so far at least as old forms can be or have been found, but many forget the principle entirely when they attempt to deal with names for which such forms are not available. Even Skeat him- self at times, especially in his more 'popular' book on Hertfordshire place-names, speculates on the history of certain names on the basis of modern evidence alone, when all the rest of his work shows be- yond question that, here as elsewhere, 'things are not what they seem', and that it is more than useless to draw inferences from the modern forms unless we have earlier ones by which to check our con- clusions. If such names are to be dealt with at all, and for my own part I much doubt if the game is worth the candle, they should be relegated to a separate chapter which might well be headed 'Specu- lations on the history of certain names'.' The second principle laid down by Skeat has been reinforced and expanded by the work of Professor Wyld and his pupils. Wyld did inestimable service in his Place-names of Lancashire by a rigid adherence to this principle, but still more by an extension of it to include a study of the development of the pronunciation of place- names from the earliest times down to the present-day local form. On this side he has made the study of place-names one of living and vital interest to all students of our language. Other writers have at least paid lip-service to the importance of this principle, but unluckily a good many w ho undertake work on place-names still do so with an entirely inadequate outfit of philological knowledge. § 4. The weaknesses of place-name sttidy as noio carried on. (a) Working on isolated areas. At the same time that Skeat lays down the.se principles which we have been considering, he indicates very clearly the great weakness from which all our place-name study has suffered when he says, ' I find myself at a disadvantage in one respect — the disadvantage is that I have made no wide or extended study of English place-names in general ; and it is obvious that in many * M'liat is the value, for example, of the stitoment, based on the modern form alone, that Rabley, Herts. (Skeat, p. 42) means ' Kaba's lea', when an unnoticed fourteentli century form sliows clearly that the word once had not initial r but initial ur (cf. ChaVter Rolls, 1317, Wrobhele) ? ENGLISH PL ACE-NAME STUDY 3 ail instance, one place-name is likely to throw light upon another, though the places may be in different counties.' ^ Similarly says Stenton,^ ' It is never wise to study the place-names of any county in isolation.' The same names, often in strangely disguised forms, may be found in widely scattered counties. It is no help, as some writers on the subject would seem to imagine, to give a list of simjlar names compiled with the aid of the Gazetteer. When with infinite labour one has hunted out the earlier forms, the names often cease to be identical, and certainly invalidate any inferences drawn fiom them.^ On the other hand, real identity is often hidden under the most strangely divergent forms. The difficult development of Thrislington, co. Durham, from Thtirstancs-tun, i. e. Thurstan's firm, can be explained a good deal better when we know that the name is also found in Thurstaston, Chesh., Thurston, SufF., Thrussington, Leics., Thuxton, Norf. Then we can use these various forms in elucidating the phono- logical development of Thrislington. No Gazetteer will reveal this identity : it can only be discovered when the other forms have been gathered and indexed. That no trustworthy account of the history of a name can be given until the whole of the relevant comparative evidence has been collected and sifted may be further illustrated from the history of two names. Wyld and Moorman agree in explaining Hambleton, Lanes, and Vorks., as containing an OE. personal name Hamela, whose existence they infer from the UE. form Hnmelandunce for Hambledon, Hants. Now this element Hambh: is found in combination with the suffix -don in Hambledon, Hants, and Surr., Hambleton, N. R. Yorks., Hambleton, Rutl., Hambledon Hill, Doi-s., Hambleden, Bucks., Ilumbledon Hill, co. Durham, Humbleton Hill, Nthb., Humbleton, CO. Durham, and in an unidentified Hameldun in the Newminstcr Cartulary. The only other element with which it is ever compounded ' Place-names of Cambridgeshire, \>. 1. ' Th'! Place-names of Berkshire, p. 3'J. ^ A typical example of this defective metliod is Sedgefield's treatment of Killington, Westm. He says («, jj. ) : 'We may compare Killingbecl;, Killing- hall, W. R. Yorksh. ; Killiiigholm, Lines. ; Killington, Devon ; Cliillington, Somerset and Staffs. ; and C'liiUingliam, Northld. The first element is clearly a pers. n., and it may be Vyila, Ceolwiite, Cw.lin, or some other.' Now the earliest forms of these names — Killiugbeck has not been traced — are D. B. KiUinijala, Cheliiirton, Cheliiiiigeholm , Cheneoltonc, Feudal Aids Cherlinyton, D. B. Cillentoiie, Close Roll (1231) Cheveliiigham. It is clear to any one with any knowledge of < )K. personal nomenclature that these contain the OE. names Cglla, Cylfn, Cceniculd, Cilia, or Ceulla, and only one of Scdgetield's surmises receives any support. 4 rUOCEEDINGS 01' TIIF, BRITISH ACADEMY is -Ion, and llint onl}' in Iliinihk-ton, W. U. Yorks., anil Ilambleton, J^aiics. Now, wlmlcver its force niav hu in the last two names, it is tleni" tliat no law of probability will allow that in the other eleven we have the names of hills which chanced to be owned bv a man bearinj^ a name whose very existence rests on an inference made from one of them. Examination of the evidence as a whole shows that we have to do, not with a |)ersonal name at all, but with an adjective wliich nii^lit llttingly be used to describe the shape of a hill. The story of that adjective is told by tlie present writer in an article contributed to Xttlllll 0-. . Ml^;-L... :■■ ENGLISH PLACE-NAME STUDY unfortunate that the writer had not the evidence from my county before him, it would have saved him from many blunders too. (6) Varijlng and imperfect principles of selection of names for treat- ment. But this is not the end of the trouble. In working at one county, I may, as has been shown, want the history of some name in another county. There is a book on the names of that county, but when I turn to it the name is not there. Why .'' Because its author is working on no principle — or too narrow a principle of selection — and, short of doing all his collecting over again, it will be ditficult for me to secure just the evidence I need for the solution of my proljlem. We look at times with some reasonable satisfaction at the number of books on the place-names of particular counties now appearing. What we fail to realize is that these books often do not deal with a tithe of tlie names within their particular area for which old forms caH be found, and that the selection is made on no generally accepted principle. Skeat, as a rule, tried to explain all the names which he found in Kelly's P.O. Directory and considered to be of interest. Study of the minimum of documents mentioned in the note on p. 4 gives iu Suffolk alone some hundred names which have not been dealt with. For Sussex, Roberts follows Kelly and Bartholomew's Gazetteer, and the same documents reveal some two hundred names omitted. Wyld covers the field more closely iu his book on Lancashire, but a comparison of this work with Sephton's book on the same county shows that the former is by no means exhaustive. Baddeley, in collecting his Place-names of Gloucestershire, adopted more pictur- esque methods. His collection was 'formed in the course of country walks, by wick and ridge and wold V a'lJ further, he has ' ventured to take the view that vanished names are of almost equal importance with those which have remained in use \^ We have an excellent book as a result, but it is by no means exhaustive. Goodall endeavours to cover the ground more completely in his Place-names of Soiitli-iVest Yorkshire, but unfortunately does not confine himself to those names for which early forms can be found. One of the most serious results of these defective methods of selection is seen in the treatment of river-names, which by some writers are almost entirely neglected. The consequences are unfortu- nate, as will be shown later. We must have a complete survey. The fact tliat a name appears in Kelly or Bartholomew, or even on the 6 in. Ordnance Maj), is no criterion of its value to the serious student. The only limitation upon the names dealt with should be the absence of early forms, and. ' lutroduction, p. viii. ib. , p. 6 PROCEEDINGS OF THE BIUTISII ACADEMY if we inust fix a workinj^ limit of date, it will be fouiul, with very few exceptions, tlmt no satistVictory work can be done on iiiimcs for which we have not forms earlier than 1 jOO. (c) In rehiiioH to Topography. But tliere is anotlier direction in which our place-name study needs co-ordinating and directing. Next to Dr. Skeat no one has done more for the study tlian Professor \Vvld. In his Place-names of lAinca.shire he did invaluable work in tracing and solving many of the phonological problems which have to lie faced in the study of place-names, and his example and direct inspiration have been fertile in suggesting lines of work to iiis own immediate pupils and others. While recognizing all this to the fullest extent, the present writer at least holds that a false direction is given to these studies in such a statement as the following: 'It may be stated at once that place-names are here considered as elements of language, and their development as a purely linguistic problem. The work ... is not concerned with the question wlietlier the names fit the places to which they are attached, nor whether they ever did so.' ' Iiis pupil Walker writes in similar strain. ' In the present work the subject has been approached solely from the linguistic point of view. It may be assumed that the historian, archaeologist, and topographer will disagree with many of my conclusions. They will doubtless possess information which I do not. Be that as it may, the explana- tion of place-names can only be attempted in the first instance by the trained philologist. The historian may supplement his work after- wards.' ^ These statements seem to me to involve a serious defect of principle. In the study of one place-name after another you are, as a philologist, faced with more than one possible interpretation, and yet what is phonologically possible may be absolutely ruled out on topographical grounds. Why may you not allow topography then and there to settle the point ? Linshields (earlier Lymhsdc) on the Coquet in Nthb. may, from the phonological point of view, be either 'lime-tree shiels' or 'shiels by the lynn or pool' (with epenthetic r/). When there is a lynn there and no lime-tree could posRil)ly stand the climate, why not make the only possible choice and at the same time give your work a fair chance with the intelligent reader who happens to possess local knowledge ? Many men of sound scholarly instincts look askance at the whole study of place-names when they find books on the subject full of cxplanationsthatcontradict easilyrccognized topographical facts. Further, those who work from a purely jihiloliigical point of view often invent entirely unnecessary and mi.sleading forms from sheer ' Preface, p. viii. - Flace-numcn of Llei-lyshire.f. 1. ENGLISH PLACE-NAME STUDY negk'ct of topographical evidence. Ekblom suggests for Liddington, Wilts, ,^ a hypothetical patronymic LyJa. In an Anglo-Saxon list of boundaries dealing with this district" we find mention of a river Hbjdan, clearly the name of the stream on which Liddington stands and forming the first element in that name. Cockersand and Cockerham in Lancashire stand on a river Cocker. Wyld ^ suggests that the first element is the genitive of a hypothetical Norse name Kokr, but why not 'sands and homestead by the Cocker'.' There is a river Cocker in Cumberland on which stands Cockermouth, and a Cocker Beck in CO. Durham on which stands Cockerton. That all these cannot be back-formations is proved by the tenth-century form Cocur for the Cumberland Cocker.* Many names of towns and villages are river-names unchanged, as Dr. Bradley' has shown; still more contain river-names as their first element, often in disguised forms. Why be sceptical about the explanation of Mitton, Lanes., as ' farm at the mytlie or juncture of two rivers \^ when this place stands at the juncture of the Kibble and Hodder, and the only other places similarly named stand in every case at such a juncture, viz. Mitton, Staffs., of Severn and Stour ; Warw., of Avon and Leani ; Wore, of Avon and Carrant ; Mytton, Salop, of Severn and Perry ; Myton, Yorks., of S\vale and Ure .' All this goes to show that the conclusions of the philologist must in every case be checked by the evidence of topography derived from maps, from first-hand knowledge, or from special inquiry. The whole science will be brought into disrepute if those who deal with it deliberately shut their eyes to evidence of this kind. (il) In relation to History. There is still the question of the relation of the students of place-names to the historian, (^n this point Wyld writes : ' No attempt has been made to attack anv pro- blems of an historical, political, or racial character. No conclusions are drawn as to the wanderings or distribution of races, as to the identity of the holders of the personal names which in a large number of cases are built into the structure of local nomenclature, or as to the size of the various manors and their dependencies. These questions, and many others allied to them, must be left to those scholars whose inclination, or better still, whose training, leads them into these fields ^ Place-names of Wiltshire, p. 114. ^ iiirchj CuHularium Saxonicurti, No. 479. ' Place-names of Lancashire , p. 97. * Place-names of C'umhcrland and Westmorland , p. 36. " Essays and Studies by Menibers of the Emjlish Association, vol. i, p 32. " Op. cit., p. 191. 8 PROCEEDINGS OF THE imiTISII ACADEMY of research. It is possible that the facts here brought together may be of some servirc to those special students in their domains.'' Now all this is to the good in so far as it is a protest against much idle sjieculation on historical and cultural problems by philologists not en- dowed with the scientific knowledge needed in these particular studies, but if tliis limitation leads to ex{)lanations of jilace-names which ;ire in direct contradiction of historical fact our studies will lose their value and fall into disrepute as nuich on this account as for a neglect of topography. From tlie purely philological point of view it may be satisfactory to explain Ilamsey, Sussex, as ' OE. liammes ea, stream bordering the en- closure', or * hamiiies c^, island or marshy land in the bond of a river', on the basis of a form Hammcs Say dated 13^1,''^ but when reference to the document itself shows that the manor was then in possession of Geoffrey de Say,' these explanations cease to be of value. For llackham in the same county, Roberts assumes as the first element a personal name liaca, a shortened form of a personal name Jianilf inferred from Baculf and liaculfcestre, the AS. forms of Reculver, Kent.* Study of Roman Britain would have shown that Raculf is not a personal name at all, but an anglicizing of the Romano-Celtic place-name Rcgidbkim. Whalley, Lanes., is mentioned in the Saxon Chronicle, s.a. 798, in the form Hweallxg. The manuscript may be of the twelfth century, but there is no reason to think that the name is an invention, and if it existed in the eighth century, Wyld's suggestion of ON. hvdll as the first element is impossible, for historical considerations do not .admit of Scandinavian influence at that date. The truth is that the student of place- names nnist not be a mere collector of forms on which to exercise his philological skill. The documents from which he derives them are after all historical docu- ments and he must read these documents with intelligence, gathering from them whatever sidelights liistory ma"y throw upon the problem of the ultimate meaning of the names with which he deals. On the other hand, he will certainly be wise not to tackle the larger problems indicated by 'Wyld, whether they, be historical, archaeological, or ethnological, unless he secures the help and co-operation of those who are expirts in these problems. When this is done, books on place- name study will at once become more accurate and scientific in their interpretation of individual names, and of real value to those who are ' Preface, p. vi Place-names of Sussex, p. 7R. ' Cf. Feudal Ai'l.i, s. a. 12!il, and InqrdsUhm Post Mortem, h. a. liUO. « U)., p. 202. ENGLISH PLACE-NAME STUDY 9 interested in the wider and more general problems involved in the history of place-names. § 5. Ptinctplcsfoi- the future study of place-names suggested hy the foregoing ivcakne.sses. Such is the present condition of place-name study. What are the future possibilities ? From a consideration of what has already been said it would seem that the whole future of place-name study depends on our recognition of four main princijjles : — (1) Tliat the place-names of a single area (countv or whatever it may be) can only be explained in satisfactory fashion when the material for the whole of England is available in ordered form. (2) That no haphazard restriction should be put on the material collected. The more the material, if it is good material, the better the final results. (3) That the interpretation of the material is, in the first instance, the work of the philologist, but his results must be checked and interpreted in the light of all available topographical and historical evidence. (4) That for the fulfilment of (3) and for inferences of a general character involving questions of topography, history, archaeology, and ethnology the writer on place-names must secure the active help and co-operation of those possessing the requisite local knowledge, and still more of those who have had the particular scientific training which alone qualifies them to speak with authority upon points outside the domain of the philologists. Now, it is quite clear that if these principles are to be followed the investigations of the history of place-names cannot be left to a few scholars worknig in isolation, but must be pursued under the authority and with the aid of some learned society. No society is so compe- tent to take up the work as the British Academy, and I would most respectfully urge it as a first step to appoint a committee of inquiry to consider the whole matter and see what can be done. § 6. Advantages to be gained hy organized xcork on these Vmes. And here the question may well be raised, ' What, apart from the value for its own sake of any increase of knowledge, are the advantages to be gained by the promotion of the particular studies for which I am appealing?' They would ■seem to be threefold: — (a) To Linguistic Studies. Material will be provided for the solution of many of our most diflacult and important linguistic problems. One such problem is that of the classification and distribution of the Old and Middle English dialects. What may be done in this domain is shown by the brilliant work of Professor W^yld on the development of OE. y in the ME. dialects, based entirely uijon 10 rilOCEEDINGS OF THE BRITISH ACADEMY the evidence of place-name forms.' Other points also may be settled by evidence of this kind, such as the dialectal distribution of various ph()nolo<,'ical features of ME. and even of OE., and the f^cncral duonology of English sound-changes. Another problem of a difi'urent kind is that of the local distribution of certain elements in common use in the formation of place-names, e. g. dv7ie (OE. dcnu, valley), hale (OE. hcalh, nook, corner), low and law (OE. hlazc), or the diliicult driiy — found in the numerous Draytons and Draycotes.^ Or, on the morpho- logical side, another problem arises in connexion with the survival of iuHexionnl 71 in the JME. dialects as illustrated by the development of OE. set Pxm mzoan tuiie or hdm{e), in the various Newingtons, Ncwntons, Nauntons, Newcnhams, Newnhams or Nunehams in contrast to the Newtons and Newhams of other districts.^ (h) To Historical Sti/diex. Many vexed problems of history will be solved. The story of Anglo-Saxon Britain will never be told aright without the help of place-name study. Numerous place-niinies mentioned in the Chronicle and other documents, often associated with important events, remain unidentified. With the aid of later ME. forms, many of these places will be identified and, it may be added, a good many traditional identifications will be abandoned. What may be done in this direction by work on charters is shown in the work of Napier and Stevenson on the Crawford Charters, while 'Stevenson's edition of Asser's Life of Alfred shows liow much can be done with another type of document by one who is both philologist and historian. New light will be thrown on historical and archaeological problems of a wider character. With organized work it will be possible to secure the assistance of Celtic specialists who shall deal with the problems raised by the non-Teutonic names in England. Writers on place-names at present are, as a rule, specialists ' in Germanic philo- logy, and do not dare to handle the Celtic names which foil within their province. AVhen these names are really tackled, much light may be thrown on the dark history of pre-Saxon Britain. Further, ' Engtiscke Studien, 1913 and 1914. ' A preliminary study of tliese questions, based on the evidence of the docu- ments mentioned above, shows, for example, the followinj; results : dene, unknown in the group Cheshire, Staffs. , Salop ; ^a/e, of very doubtful occurrence \n the proup Dev., Dorset, Somerset ; law and low unknown in Dev., Dors., Som., Hants, Surrey, I. of W't., Kent, also in Herts., Hants, Nf., Hunts., as well as Cum. and Westm.; dray- unknown in the six northern counties. ' 'Hie area of distribution is fairly definitely marked. It includes Northants, Beds., Cambs., Herts., Ess., Middlesex, Kent, Surrey, Isle of ^^'ight, Devon, A\'ilt3., Glouc, Warw. , Wore, Oxf. ENGLISH PLACE-NAME STUDY 11 we need the co-operation of specialists in the Scandinavian dialects and in Anglo-French ; and here a word of tribute must be paid to the great Avork already done in these matters by such scholars as Bjiii-kmau in his work on ME. names of Scandinavian origin, and by Zachrisson on Anglo-French influence in our place-nomenclature. Place-name study will, moreover, add to our knowledge of the racial divisions of our Teutonic forefathers and the history of their settlements. One fruitful line of work would be a study of the distribution of certain types of place-name which must be associated with the actual manner of carrying out the original settlement, e. g. hampstead, hampton, ham, ing (OE. ingas, pi. patronymic), ton} Problems of social history may be illustrated from the distribution of such names as Charlton ^ and other compounds in Charl- (from OE. ceorla = of the churls) and Hardiaick? (c) Practical. Great practical service will be done to the nation if, as I hope may be the case, this work is ultimately taken up in close co-operation with representatives of the public services, more especially of the Ordnance Survey. Our place-names are being sorely mishandled by would-be educationists aiming at a false standard of correctness, by railway officials hunting for the picturesque, by map- makers who are skilled surveyors but men of no scientific under- standing in this particular matter. § 7. Examples from other countries. Finally, I would urge two practical considerations: — (1) that any committee taking the task in hand will not be venturing on entirely new or unexplored lines of work ; (2) that the problem is an urgent one, which brooks no delay. With reference to the first consideration, we have before us the example of the three Scandinavian kingdoms. Let me attempt in a few words to sketch what has been done there. In Norway organized work was first begun in 1896. Material for the work lay to hand in the evidence gathered by a commission appointed in 1878 to draw up a new Land Register. This commission ' Hampstead seems to be confined to Nf., Ess., Herts., Mx., Beds., Berks., Bucks., Glouc.,Suss. ,1. ofW't.jUev. ; //nm/(/o;i to Wilts. (i)),Som. (5), Dors. (6), Devon (3), Hants (9), Oxf. (6), Glouc. (8), ^Vol•c. (5), A\'arw. (4), Heref. (4), Salop (2), Berks. (1), Bucks. (1), and (doubtful) Norf. (1) ; ham is very common in Cambs., Ess., Nf. , Sf., Kent, Surr., Suss., Mx., equally uncommon in tbe M'est and ^^'est Midlands. There the most common suffix is ion, though this is, of course, fairly common everywhere, ing names are specially common in Ess. , Kent, and Surr. ' Not found in Nf., Suff., Ess., Mx., Cambs., Northts. ' Confined to Glouc, Oxf., Bucks., Leics. , Derby, Lines., Norf., Cambs., ■ Northts., Notts., Yorks., Durham. 12 rilOCEKDINGS OF THE BRITISH ACADEMY consisted of Profossors Sophus Uiigj^e and Olaf Hygli, and I'lovost Jolian Fritzncr. Their instructions had been to secure correct forms for tiic entries in that register by collecting and sifting all tlie evidence past and present, written and oral, which woidd throw liglit upon local names. \Viien the register was completed, liygh saw what an ojiportunity the material they had gathered offered for a scientific study of tlie history o{ Norwegian place-names, and in 18l)(), he, in collaboration with others, presented a scheme for such a study to the Storthing. The co-operation of the Church l)e])artmcnt was .secured, and, with the aid of a grant from the State, work was forthwith begun. According to the plan, the whole work was to be issued in a series of sej)arate volumes, each self-contained and dealing with a single a7nt or province, or, in the case of the larger provinces, with subdivisions of them. A general introduction of a provisional kind, planned to avoid much needless repetition, was issued in 1898 ; some nineteen volumes have successively appeared and the task is almost accomplished, llygh planned a general introduction to accom- pany the completed work, which should treat the question of Norwegian j)lacc-namcs as a whole in the light of the exhaustive work done on the .separate provinces. He did not live to write this, or even to complete his work as editor of the separate volumes. These fell after his death to his brother Karl Rygh, to Dr. A. Kja^r, and to Professors Hjalmar Ealk and Magnus Olsen. The work is throughout scholarly and yet popular in its treatment and, I may add, in price. A charge of three to four kroncrs for each volume has placed it within reach of that ever mimerous body of layfolk who are curious about the meaning of the names of the places about them.i The example of Norway had its effect on Sweden, where some preliminary spade-work was done between 1896 and 1901, largely by Professor Norecn and Professor Hildebrand. In 1901 a connnittee of in<(uiry was appointed by Koyal authority, and it was asked in 1902 to prepare a scheme of inf[uiry into tlie names of the towns, villages, farms, and of the more important hills, larger wood.s, lakes, &c., of some one province, and present it to the king with a statement of the probable cost of carrying it into effect. The scheme was prepared, royal approval was given, and a grant in aid made by the State in 1905. The work was planned on a more extensive scale than that which had been attempted in Norway, and progress has been slower, but the greater part of the volumes dealing witli Alvsborg province have appeared, and work on other provinces is well under way. The ' C'f. JN'orsAe Gaardnavne^ Forord og ludleduiug af O. Kygli, Kristiauia, Hi'M. ENGLISH I'LACE NAME STUDY 13 cost of tlie work has been met (1) by State niJ, (2) by grants from the provincial governnieuls, (3) by the generosity of private donors, notably of the authoress Selnia Lagerldf, (4) by the public-spirited action of the publishing house of Ljus in undertai