presumably, a common ancestor of AbC and L (not to be confounded with the original version of the saga) already contained some brief account, or hint, of Hrosskell taken from the Hrólfssaga; and that the copyists, both of AbC and L, independently helped themselves to the fuller account of the Hrólfssaga which, frequently being in the same codex with the saga to be copied, was easy enough of access.

On the subject of the quick death of Ref, Neri, and the queen, at the end of the saga, Ranisch has the following observation 1): "Man sieht nur zu deutlich, weshalb alle drei dem frühen Tode verfallen. Refr und Neri kamen in der Hrólfssaga nicht vor; am meisten aber war die Königin im Wege, da die Hrólfssaga mit einer neuen Werbung des Königs begann; auch war der Tod einer unbenannten Königin schon als notwendige Vorbedingung für die Werbung des alternden Königs in den einleitenden Worten der älteren Hrólfssaga erwähnt".

essen

Again, I certainly admit that these deaths are suspicious. But alas! The very look of the last lines tially the same in all versions is suspicious also, the end being, indeed, the likeliest spot for later hands to insert obvious explanations 2). They would be inserted, I surmise, even if the priority of a story was entirely beyond doubt. As to the prolonged existence of Neri and Ref, it would have troubled the original author of the Hrólfssaga but little. His story was set in the times of Romance, his scene is the whole Northern world, the whole conceived in a different key. The interpolator betrays himself by failing to grasp that; and, prosily fearing lest the reader might ask why Hrólf seeks no aid from his kinsmen Neri and Ref, he hastens to inform us that they are dead. The fact is, it looks as if the author of the Hrólfssaga had left Norway

1) L. c. xxi.

2) Cf. Finnur Jónsson, above.

serves, the queen was most in the way. Ay, truly; but then, the Hrólfssaga is just based on the death of Gautreks queen!

The argument ex omisso '), that the Hrólfssaga occurs in MSS. not containing the Gautrekssaga, but the latter never without the former, and that, hence, the Gautrekssaga had no independent existence, is a most unsafe one. At best its force can be cumulative. Applying it without discretion, we might arrive at the most absurd results. Sounder it is to suppose that the Hrólfssaga was, or became, the more popular of the two naturally suiting the later taste better and has thus come down to us in a greater number of MSS.

To the above, chiefly negative, arguments I would add the following considerations.

In the Gautrekssaga, it is the purpose of the author to put the generosity of Gautrek on a line with that of two other mighty kings of old. The Danish king is here made to be the famed Hrólf kraki. If the Hrólfssaga has substituted the indistinguishable and stereotype Hring, it is clearly in order to avoid the collision between two world-conquering Hrólfs. The reverse is far more unlikely. King Ella of England, of course, did not need to be changed in the Hrólfssaga.

The loss of his queen plunges Gautrek into deep sorrow. He has his throne set on her burial mound. In the Hjálmþérssaga ok Ölvés 2), king Yngvi likewise has his royal seat placed on the how of his queen 3) and lives to

1) L. c. xix.

2) Fornaldarsögur Nordrlanda III, chap. ii.

*) Nothing actually proves this to be a late motif. Lehmann. Zfdph. 1910, 1, has examined (not all) occurrences of 'sitja á haugi', without being able to shed much additional information on the subject. He is not aware of Ranisch's observations; see below. Nor of A. Olrik's article in Danske

his sorrow.

When his son Hjálmþér asks whether he means to sit thus longer and yearn after the departed, instead of seeking him another queen, he answers naught. One day, he strolls down to the shore, "með mikilli sturlan ok áhyggju, hugsandi hversu veröldin mátti brigð verða”, from which heavy thoughts he is roused and rid by the lovely sorceress Lúda who lands there in a boat. In the same saga, another king loses his queen 1). "Var haugr orpinn eftir hana, ok þráði konungr mjök drótning sína”. -- In the Göngu-Hrólfssaga, it is Jarl þorgnýr who laments the loss of his consort. The burial-mound for her was thrown up near the castle 2). "Sat Jarl þar oftlega í góðum veðrum, eðr þá hann hafði málstefnur, edr lét leika fyrir sér". And one fine day 3), as the Jarl sat "á haugi drotningar, ok var leikit fyrir honum", a swallow drops a golden hair of wondrous length, which sight kindless new love in him. With reawakened energy, he makes the solemn vow to win the woman or die in the attempt.

Ranisch), in thinking "á haugi drotningar" to be a mistaken addition of AbC, has not stopped to reflect that the king's Traum verlorenheit is the very condition for his giving Ref a golden bracelet in return for a stone to stir up his hawk with! For that matter, why' except for "dægrastytting" in his sorrow should the doughty king (in either version) sit all day on the hill and hunt with his hawk? Hence, I incline to the belief that L is the guilty party, in having omitted to mention the early death of the

Studier 1909 'At sidde paa høj'. As Olrik is not aware of the instances here mentioned, his conclusion "Den døde dronning (in the Gautrekssaga) spiller ingen rolle i sagaen; sorgen over hendes død er øiensynlig blot opfundet for at forklare kongens vane at sitja á haugi" (1. c. p. 3) is quite unwarranted.

1) chap. xxii.

2) Fas. iii, chap. v.

3) chap. x.

4) 1. c. xxx.

queen, which is the condition both for the generosity story and the very existence of the Hrólfssaga.

It is necessary to call attention to an hitherto unobserved point of contact between the two sagas. In the Gautrekssaga (chap. ii), we are told that Snótra and her son Gautrek leave their home in the forest after all of her family had plunged themselves from the ætternisstapa 1). "Fara nú þangat til, er þau finna Gauta konung, ok tekr hann vel við þeim syni sínum. Fæddizt hann þar upp með hirð feðr síns ok var bráðgjorr á allan þroska, ok lída nú svo fram nökkurir vetr, þar til er Gautrekr var mjąk fullkominn at þroska. þá bar svó til, at Gauti konungr tók sótt ok kallaði til sín vini sína. Konungr mælti: þér hafið verit mér hlýðnir ok eptirlátir í ollu, en nú þiki mér mikil vón, at þessi sótt, er ek hefi, skili vóra vingan; vil ek ríki þetta, er ek hefi átt, gefa Gautreki syni mínum ok þar með konungs nafn. þeim líkaði þetta vel, ok eptir andlát Gauta konungs var Gautrekr til konungs tekinn yfir Gautland", etc. Since no other sons are mentioned (the whole story being, in fact, an etiological fiction), the whole proceeding is but natural, and only a brief remark, such as this, in order to apprise us of Gautreks succession.

In the Hrólfssaga, however, there are two sons. The older, Ketil, is described 2), as "minnztr vexti ok þó inn hraustasti. Var hann ecki mjök vid lyndi Gautreks konungs sakir hávaða ok kappgirni", these not being altogether desirable qualities in a ruler, according to Old Norse conceptions; whereas Hrólf, the younger son, is his exact opposite, uniting in himself all chieftainly qualities. Immediately after the above description of Ketil, the order of succession is taken up: "Líðr nú svá fram, þar til er Gautrekr konungr tekr þyngð mikla ok er hann finnr, at hónum herðir,

1) L. c. p. 11.

2) ed. Detter, p. 8.

ok hann getr eigi af sér borit, þá stefnir hann þing ok segir þeim Gautum sótt sína, þakkaði þeim med fögrum orðum góða fylgd ok trúleik, er þeir höfðu hónum veitt". They thank him, in return, for his beneficial reign. "þá mælti konungr nú er svá, sem þér vituð at ek á tvá sonu;

Eru þat hér landzlög, at inn ellri konungsson taki ríki eptir föður sínn". He proposes, however, Ketil and his people consenting, that Hrólf be made king. This proves agreeable to all concerned. "Nú þröngvir konungi, svá at hann deyr, ok þótti Gautrekr verit hafa inn bezti konungr í fornum sið ok var hann mjök harmdaudi".

Comparing the two accounts, and especially the last sentences, the former clearly seems the more original. Here, as in almost every case where loan from an older source is demonstrable, the Hrólfssaga is more verbose and circumstantial, almost to fulsomeness.

Why is not Hrólf made to be the older brother, as in the sagas from which the different brothers' characters are taken? 1)

In another passage, Ketil is described 2) as "manna minnztr ok þó inn skjótligasti, hann var kallaðr Ketill Kregð (FGs: Kregđa), hávaðamaðr mikill ok þó framgjarn ok lét ecki fyrir brjósti brenna, at tala ok gera þat hónum kom í hug, hardfengr ok fullr áræðis". The meaning of the epithet Kregd (fem.) is given by the lexicographers as 'vantrivning, weakling; a pining away'; which is supported by similar meanings in Modern Scand. dialects. Ross: kreks, kriks; Jutish: kræg; Swed.: kræk, all meaning 'weak, pitiable creature'. Likewise evidenced in derivatives. Cf. Falk and Torp sub kræ.

What warrant there is for stating the word to be fe

1) Cf. the writer in The Journal of Engl. & Germ. Phil. 1912, 76 f. 2) L. c. p. 6.