other hand, there can be equally little doubt that Moons corresponds more closely to the Hebrew Mōsheh. This correspondence would of course explain sufficiently the appearance of 'Moses' in the Vulgate : St Jerome may have restored it from the Hebrew, as in the case of 'Beelzebub'. But it leaves unexplained the 'Moses' of k (and, if I am right in relying here on L, of St Cyprian): are we in presence once more of a case of the original reading of the Greek Testament having to be restored from the African Latin ?1

Nabucodonosor 121. 14 (Dan. iii 16). So AW: the rest 'Nabuchodonosor', except X (and C in Ep. vi § 3 [483. 13]) 'Nabugodonosor'. In ad Fort. § 11 (337. 12) S has 'Nabucodonosor', R 'Nabucodonossor' (and so also in Ep. vi), V apparently 'Nabucchodonosor '.

This

Natham acc. 49. 7, 75. 20 (both quotations = 2 Reg. vii 4). reading is only that of R in 49. 7, A R in 75. 20, as against 'Nathan' ('Natan' 2/2 X) of the rest: but it is supported by Priscillian tract. iii, 50. 13 'Natham profetam' (cf. 50. 21 'in uerbis Nathae profetae').

Nazoreus or Nazareus 83. 2 (Act. iv 10): 165. 13 (Act. iii 6). For 'Nazoreus' we have in 83. 2 A*T* Ub, in 165. 13 A: for 'Nazareus' (-aeus L) in 83. 2 L MOP B, in 165. 13 LMOPUBT* VW': for 'Nazarenus' in both places apparently RT', and also in 165. 13 W* (yet I suspect that 'Nazoreus' may have been W's original reading) it is clear, therefore, that Hartel's 'Nazarenus' cannot stand, and the problem is to decide between the two other forms. But this cannot be done without looking somewhat further into the whole question.

Four Latin variations of the name occur: 'Nazoreus' 'Nazareus' 'Nazorenus' 'Nazarenus'. Of these the second and third are independent attempts at conflation between the other two: Nalwpaîos and Natapnyós are the only ultimate Greek originals. The triumph of the form 'Nazarene' has been so complete both in Latin and English that it is not easy to realize that not only do both forms go back to the New Testament writers themselves, but that two out of the four Evangelists used exclusively, and a third by preference, the form Ναζωραίος. Our Authorized Version, indeed, paraphrases with the noun of Nazareth', except in Matt. ii 23 'he shall be called a Nazarene' and Acts xxiv 5 'the sect of the Nazarenes': the Revised Version is less consistent, adopting 'Nazarene' also in Matt. xxvi 71, Mc. xiv 67, xvi 6, but leaving of Nazareth' elsewhere. But it is to the Vulgate that we really owe the word 'Nazarene': and St Jerome uses 'Nazarenus' to the exclusion of all other forms, save in Matt.

1 My colleague, Mr A. E. Cowley, tells me that the form Mavons may be due to the erroneous etymology from the Coptic (Mov = water: cons such as are saved) found in Josephus Ant. II ix 6, cf. contra Ap. i 31.

ii 23, 'Nazareus,' where we may suppose that his knowledge of the Hebrew bible led him to reproduce the exact form used by the Evangelist, rather than definitely to interpret the word as equal in meaning to 'Nazarene'. For the rest the conjecture may be permitted that he wished to make a distinction between 'Nazarenus'—'Nazarene' in the literal and biblical sense, 'of Nazareth'-and 'Nazareus', the form which he was in the habit of employing in reference to the contemporary Judaeo-Christian sect.1

But, as has already been indicated, there is no such uniformity in the Greek text of the Gospels. St Matthew (ii 23: xxvi 71), St John (xviii 5, 7: xix 19) as well as the Acts (ii 22: iii 6: iv 10: vi 14: [ix 5]: xxii 8: xxiv 5: xxvi 9) use invariably Nagwpaîos: St Mark as regularly (i 24 x 47 xiv 67: xvi 6) Nagapηvós. In St Luke's Gospel alone does the usage vary between the two: iv 34 Nalapηvós, xviii 37 Ναζωραίος, xxiv 19 Ναζωραίος A D Sahidic, Ναζαρηνός B L

to a certain point this diversity of usage is reflected in the Old Latin MSS, though it is complicated further by the cross-forms 'Nazareus' and 'Nazorenus'. I do not think anything short of a table will make the matter clear.

1 These 'Nazarenes' first emerge, so far as I know, under this name in Epiphanius Haer., xxix Na(wpaîo, and Philaster Haer. viii ‘Nazaraei'. Jerome's own references to them are frequent: de Viris § 3 about the Hebrew Gospel ' mihi quoque a Nazaraeis. . . describendi facultas fuit'; Comm. in Matt. xii 13 'in evangelio quo utuntur Nazaraeni [lege Nazaraei] et Ebionitae'; Ep. cxii ad Augustinum § 13, the Ebionites 'quos vulgo Nazaraeos nuncupant'; Comm. in Isaiam ix i 'Nazaraei hunc locum ita explanare conantur', xi I 'evangelium quod Hebraeo sermone conscriptum legunt Nazarei'. On the other hand he uses 'Nazarenus' when speaking of the inhabitants of Nazareth, Comm. in Matt. xiii 54 'mira stultitia Nazaraenorum'.

An interesting passage is the reference to Nazareth in the de situ et nominibus 'Nazareth, unde et dominus noster atque saluator Nazaraeus vocatus est; sed et nos apud veteres quasi pro opprobrio Nazaraei [one MS "Nazorei "] dicebamur, quos nunc Christianos vocant'. But we cannot tell how far in this case the form in Jerome is influenced by the original Greek of Eusebius: nor can we be at all sure that our only authority for the Greek text, Vatic. gr. 1456 saec. xii, has reproduced it correctly, Ναζαρέθ, ὅθεν ὁ Χριστὸς Ναζωραῖος ἐκλήθη καὶ Ναζαρηνοὶ τὸ παλαιὸν ἡμεῖς οἱ νῦν Χριστιανοί.

On a review of the evidence and of the probabilities as a whole, I think that the reading of A in St Cyprian, 'Nazoreus', has good claims to be considered original.

Neptalim 56. 7 (Is. ix 1 [viii 23]). VPU 'Nepthalim': and this, with remarkable regularity, is the reading of the Lyons Heptateuch [not, as Hatch-Redpath say s.v. Neplaλeíμ, 'Nephthalim']. But 'Neptalim' is also given by Primasius in Apoc. vii 6.

Noe [45. 22].

Or 89. 17 (Exod. xvii 12). So AU V W X* b (M*? R*?): and so VS in ad Fort. 8 (331. 1). This is indubitably right against 'Ur' of LM2 R2 X' and 'Hur' of BT: O omits.

Paulus 127. 13.
Petrus 165. II.

Pilatus 99. 3.

Pontici [148. 16]: [148. 23]: [149. 6]. See Prolegomena § 2 (J.T.S. vi 258).

Rachel or Rachiel [53. 2]. A has 'Rachiel', 'Racel', the rest

'Rachel'.

Lugd. has 'Rachel' [Hatch-Redpath, s. v. 'Paxyλ, wrongly Lugd. Rachael' generally indeclinable, but sometimes 'Rachelem' 'Rachelis' 'Racheli': only in Gen. xxix 6 'Rachae'.

Rafael or Rafahel [53. 16]. X has 'Rafael', AT 'Rafahel', LO PU 'Raphael', R 'Raphahel'.

Rebecca [166. 7], dat. Rebeccao 51. 22. For the dative in 51. 22 PR give 'Rebecchae'. Lugd. has regularly 'Rebecca' 'Rebeccam' 'Rebeccae': Priscillian, in the nominative, 'Rebecca.'

Sabain 68. 15 (Is. xlv 14). So AOPRTUWX and perhaps L*: 'Sabaim' L', 'Sabam' B M2, 'Sabann' M*. Zaẞaeív, N*.

Sabaoth 44. II (Is. i 9): 57. 21 (Is. iii 1): 68. 14 (Is. xlv 14): and probably elsewhere. In the first and third passages X has 'Sabahot', in the first M 'Saboth', in the third R 'Sabath'.

Samuel or Samuhel [53. 6]: [53. 9]: [84. 1]. The MSS vary as

follows

53. 6 'Samuel' ALPU: 'Samuhel' ORTX: 'Samuehel' M. 53. 9 'Samuel' LPRTUX: 'Samuhel' AO.

84. 1 Samuel' PU: 'Samuhel' ALO.

Sarra [52. 16]. So too the Lyons Heptateuch and Priscillian. Satanas, acc. Satanan. 144. I (Act. v 3): 145. 13 (1 Cor. vii 5): 173. 6 (3 Reg. xi 14: Hartel gives the reference as xi 23, but that verse is absent from the text of B, whereas xi 14 is found in both the A and B texts). The first two passages are in the nominative, and present no variant: the third is in the accusative, and here the form 'Satanan' is guaranteed by ALOPRU (V) WX (possibly 'Satana' X*), as against 'Satanin' M 'Satan' B. Hartel against all

his MSS 'Satanam'.

Sedrac 121. 14. So A M* U, and Zedpáx LXX: 'Sidrac' LOPR' TWX, cf. Vulgate: 'Sedrach' B, 'Sidrach' R*.

Sileas 127. 13 (Act. xvi 25). So all our MSS, except P* and the second hands of MOR. So in Act. xv 27 the MSS of Irenaeus (III xii 14) have 'Sileam', the editors 'Silam'. For further evidence in favour of 'Sileas' see Souter A Study of Ambrosiaster p. 208. Sina: see 'Syna'.

Sion 44. 10: 46. 10: 46. 13: 82. 6 (omitted by A): 84. 24: 90. 17: 93.5: 95. 3: 96. 15: 97. 6. In these passages the accusative genitive and ablative cases are represented: and nowhere is there any variant in the indeclinable form.

In Tertullian a declension of the name with case-endings is found: but Priscillian has 'Sion' in dative (84. 13) and ablative (66. 8).

Sodoma nominative 44. 12 (Is. i 9)—no variants on 'Sodoma... Gomorra". Sodoma or Sodomam accusative 146. 16 (Gen, xix 24): here AUVX6 (L'?) give 'Sodoma et Gomorra', L* MOPRTW 'Sodomam et Gomorram'.

The root of the difficulty appears to lie in the fact that in the Greek Ο. Τ. Σόδομα was a neuter plural, Γόμορρα a hybrid between neuter plural and feminine singular. The declension Σόδομα, acc. Σόδομα, gen. Σοδόμων, dat. Σοδόμοις, is without exception in the Greek of both Testaments. On the other hand Tóμoppa makes its accusative invariably Tóμoppa (Gen. xiii 10, xix 24: Amos iv 11: Is. xiii 19: Hier. xxvii 40), but its genitive as invariably, at least in the Old Testament, Γομόρρας (Σοδόμων [καὶ] Γομόρρας Gen. x 19, xiv 2, 8, 10, 11,

▲ For the orthography see under 'Gomora' above, p. 74.

xviii 16, 20, xix 28: Deut. xxxii 32: Is. i 10).1 In the New Testament Rom. ix 29 and Jude 7 are nominative: 2 Pet. ii 6 follows the LXX use, Zodóμwv κai Toμóppas: but in Matt. x 15 we find, besides the traditional form, a double attempt at assimilation, Σοδόμων καὶ Γομόρρων in NB abcf Vulg., 'Sodomae et Gomorrae' in ff, k Iren. lat. (IV xxviii 1).

...

This assimilation to the feminine form is in fact characteristic of the earliest Latin version, and that not only in the genitive, but throughout the declension. Priscillian has 'Sodoma' in the ablative (7. 24): Tyconius has 'Sodomam et Gomorram' in the accusative (85. 3), and Ambrose 'Sodomae . . . Gomorrae' in the genitive de Elia et ieiunio 14. The Lyons Heptateuch is not quite consistent: but it has for the nominative 'eversa est Sodoma et Gomora' in Deut. xxix 23, and for the accusative 'super Sodomam et Gomoram' in Gen. xix 24. It seems clear that where both names occur together, the earliest translators habitually treated them together as feminine: and this is so rare in later authorities that the neuter is much more likely to have been substituted for the feminine in the Cyprianic MSS than vice versa. On the witness of the MSS alone one might have accepted 'Sodoma et Gomora': but the outside evidence, when brought into consideration, seems almost decisive for 'Sodomam et Gomoram'.

Solomon [167. 1]: 173. 7 (3 Reg. xi 14). The name is regularly declined in -em, -is, -i, -e.

With regard to the orthography, I expressed myself with some confidence in the first section of these Prolegomena (J.T.S. vi 251) as to the correctness of the form with o in St Cyprian. I am not inclined to retract that view: but it may be useful to bring into account here some notice of the evidence at large, which makes the variation between 'Solomon' and 'Salomon' almost more baffling even than that between 'Moses' and 'Moyses'. In the first place the witness of the Greek Old Testament is in direct contradiction to the witness of the Greek New Testament. In the latter oλ- is practically universal: in the former the witness of A B in the Books of Kings and of NB (though not of A) in the titles of the Sapiential Books, and of A B C (but not of N) in the text of the Song of Songs is regularly for Σaλ-. The Hebrew (I am told) offers no assistance in deciding: in the text as it was vocalized the first vowel is the weak shwa, which corresponds rather to than to a or w.

I confine myself, therefore, to a brief statement of Latin evidence. Of the Old Latin MSS of the Gospels and Acts, abdehqff, and the

1 In Hier. xxiii 14 (for womep Tóμoppa of the rest) N reads Aads Tóμopaapparently an indeclinable genitive. This is also the only occurrence of the single p in the great uncials.