In such examples as the following, the result-idea involved is very clear: LS. 2. 230. 166 Began da to bodigenne pa godspellican lare swa lange pam sceaðan op þæt he ge-lyfde on god; ÆH. 1. 304. 29 swa lange he æteowde his wundra dam hæðenum folce odpat hi geleaffulle wæron.

NOTE 1. I have not been able to find a parallel to this construction in the poetry.

NOTE 2. I have noted a similar construction in Middle English: Brut 3. 200. 13 Bruttes þat long heolden wel swiðe longe, a p Aðestan þe stronge, þe king of þissen londe, heom binom þas londes alle.

1h. swa lange oð.

This connective owes its origin to the same confusion of constructions that gave us swa lange . . oð dæt.

I have noted only five examples, all of which I quote: Chron. 169. 8 he forweornde swide swa lange of his sciperes ge fengon hine; Chron. 180. 8 Da wið læg se cyng sume hwile peah, swa lange od pet folc pe mid þam eorle wes, weard swide astyred ongean pone cyng; Bo. 121. 20 Fordæm hi ne lyst spyrian æfter ælcre spæce swa longe of hip riht witen; Cart. 3. 284. 19 þa spræc hit fæstlice Byrhsige Dyrincg swa lange oð þa witan þe þa wæron gerehton Eadgife, þæt heo sceolde hire fædres hand ge clænsian be swa myclan feo; Jos. 7. 16 and eode þæt gehlot swa lange od hit becom to pam ilcan men þe þæt man gefremode.

NOTE 1. With this construction is to be compared Modern German so lange bis, thus: er quälte mich so lange, bis ich that, &c.

This conjunction is the result of the same sort of confusion between the result- and temporal ideas, or

fusion of them, that produced swa lange oð dæt. I have noted only three instances, all of which I quote: Dial. 220.9 se hit me sæde 7 pus cwæð, þæt nu for fif gærum hit gelumpe on Romebyrig, þæt Tifrestream was upp gangende 7 swa swide weaxende, o pat hyre wæter 7 yða fleowen ofer þa weallas; Dial. 248. 23 witodlice hit gelamp, þæt þa yþa reðgodon in heora þeawe 7 wæron upp ahafene for para winda mycelnessum swa swide, op þæt of þam scipe wæron þa næglas forlorene; Wulf. 206. 22 and seo eorde weoll ongean þam heafonlican flode swa swyde, oð ðæt þæt water was heahre, ponne ænig munt æfre wære.

1j. swa lange æt.

The clauses with this connective so approach the result-force that it is difficult to decide to which category they should be assigned. Both connotations must have been in the mind of the writer in most of the cases.

Five of the six examples are in Chron., the other in Laws. I quote all, since they are so few, and so doubtful in some cases: Chron. 117. 24 7 was þær þa swa lange pse arcebiscop Oswald of Eoforwic was forð ge witon. In the following example probably the result-idea is predominant: Chron. 117. 30 was þær swa lange man sette him to biscop on Wintan ceastre ; Chron. 177. 23 da lengde hit man swa lange p seo scip fyrd eall belaf. The next example is the only one in which does not follow swa lange directly: Chron. 178.1 7 eodon þær up 7 hergodon swa lange þær p Þ folc geald heom swa mycel swa hi heom on legden; Chron. 179. 25 þa for æft ongean to Wiht 7 þær abutan be þam sæ riman swa lange p hi comon togædere Harold eorl his suna; Laws 226. 7 Da geræddan witan, þæt hit betere wære, þæt man æure tymde, ðæs hit ærest befangen wære, swa longe pat man wiste,

hwær hit ætstandan wolde, dy læs de mon unmihtigne man to feor 7 to lange for his agenan swencte. The Latin for this reads: Unde consuluerunt sapientes, quod melius erat, ut saltem advocaretur ubi deprehendebatur, donec innotesceret, &c.

NOTE 1. I have noted a parallel in Middle English: Brut 2. 116. 19 Heo eoden swa longe forð ward, þat heo comen in ænne orchard.

1k. Öæt.

The ideas of time and result lie very close to one another in all the clauses of the until-division; but, in the case of those with dat as the connective, it is most often impossible to assign them definitely to either one of these categories. In most of the examples from Chron. probably the primary notion is that of time, though the result-element is usually present; but in some few cases I cannot see that there is any connotation of result at all.

In this first sentence the temporal force is clear; moreover MS. C. has oỡ and MS. D. oððe: Chron. 92.5 þa rad se cyning mid firde he gewicode æt Baddan byrig wið Win burnan; Chron. 143. 14 7 wende swyde rade abutan East Englum in to Humbran muðan 7 swa upp weard andlang Trentan þet he com to Gegnes burh; Chron. 213. 4 7 þa Brytta hine heoldon se cyng com of Franc land. The idea of result is most prominent in this example: Chron. 267.1 7 besæt heom þer wæs inne mikel hungær; O. 160. 31 hie peah swa ondrædendlice gebidon þæt se ege ofergongen wæs, 7 þær sippan wælgrimlice gefuhton; Lch. 1. 340. 25 Wið lip adle, genim cwicenne fox 7 seoð þa ban ane beon læfed; Inst. 486. 52 gyf þonne hwa mid hwylcum unamtan genyd sy, he to þære mæssan cuman ne mæge ne to þæm

æfensange, þonne swa-peah gewunige he fæstende, he wite seo mæsse 7 se æfen-sang sy gesungen; Byrh. 300. 11 þonne beoð þi geare preo hund daga 7 feower 7 fiftig daga fram easter tide þat he eft cume.

NOTE 1. The following lines from Layamon's Brut present a parallel use of pat in Early Middle English: Brut 3. 207. 14 And þus heo wuneden here, wel feole gere, þat þæ children weoren muchele iwaxen.

2. hwonne.

The use of hwonne in the sense of until doubtless developed through its regular use in indirect questions. Indeed, in many of the cases in which I have construed it in this sense, it still lies close to the ordinary use in such clauses.

The nine examples I have noted occur in texts ranging from BH. to Wulf., though five are in the writings of Elfred.

I quote freely, believing that the peculiarities of meaning and use will be brought out thus more clearly than by discussion: BH. 178. 22 p alihte se eoredmonn 7 þæt gebæte of ateah, 7 þær hwile bad, hwonne his horse bet wurde oope he hit þær dead forlete. The Latin for this is: coepit expectare horam qua. BH. 186. 23 þa was sume dæge, þætte he sorgende bæd hwonne seo adl to him cwome; Bo. 14. 10 Walawa da ungesæligan menn ne magon gebidon hwonne he him to cume, ac forsceotað hine foran; CP. 121. 11 Se yfela deow cuio on his mode: Hit bio long hwonne se hlaford cume; ÆH. 1. 140. 19 and ðinco him to lang hwanne he beo genumen of dyses lifes earfoonyssum; BIH. 109. 32 ah he þær on moldan gemolsnap & þær wyrde bidep, hwonne se ælmihtiga God wille pisse worlde ende gewyricean; Wulf. 236. 11 swa oft, swa he pærinne wæs, him

puhte æfre to lang, huonne he moste beon ymbe þæs lichaman oferfylle and ymbe his agene unþeawas.

NOTE 1. Hwonne is occasionally so employed in the poetry, thus: Christ 147 Nu hie softe þæs bidon in bendum Hwonne Bearn Godes cwome to cearigum.

NOTE 2. With this use of hwonne in OE. we may compare the OS. hwan in combination with er, as in this example: Heliand 105 That werod oðar bed umbi thana alah utan, Ebreo-liudi, hwan er the frodo man gifrumid habdi waldandes willeon.

3a. fort(e).

I have observed this connective only in Lch. 3. It is made up of for+to, which becomes forte, and then fort. In one of the examples we have the spelling forte, thus: Lch. 3. 102. 17 Nim þanne swa hætte swa he hættest forbere mæge 7 habban an dæl on hys mupe, forte acoled beo. The examples are all similar, and one or two will suffice for illustration: Lch. 3. 118. 26 do þar piper to 7 lege to þan sare, fort þe man wearmie; Lch. 3. 130. 15 ete panne a morgen, fort he full sy.

NOTE 1. I have noted forte in Middle English, and quote examples: Brut 2. 171. 17 þe name stondeþ þare, forte pat Dence men driuen vt þe cnihtes; Piers Plowman A. 11. 119 Leve him on thi luft half a large myle on more, Forte thou come to a court Kep-wel-thi-tonge-From-lesynges-andlygeres speche.

3b. swa lange fortban.

I have noted only one example with this connective, and this I quote: Lch. 3. 88. 23 læt hyne liggen swa lange fortpan eara hit habben eal gedrucan.

4. til.

Only two instances of the use of this conjunction have been noted. These occur in the entries of the