* Abbreviations used in this article are as follows. LXX =Alan E. Brooke and Norman McLean, The Old Testament in Greek, vol. 1: The Octateuch (Cambridge: Cambridge University, 1917). Individual LXX MSS and hexaplaric readings are cited from this source, using its sigla. G^B=B a Ethiopic;G^L=gndpt (G^L1=gn; G^L2=dpt); G^o=G ckx Syhex; G^A=AFM etc. These groupings are not proposed as definitive or exhaustive, but as pointing to identifiable currents of LXX evidence. α′σ′θ′=Aquila, Symmachus, Theodotion (medieval citations). MT=Rudolf Kittel, Paul Kahle, et al., Biblia hebraica(3d ed.; Stuttgart: Württ. Bibelanstalt, 1937 and later imprints). Sam. = August Freiherr von Gall, Der hebräische Pentateuch der Samaritaner (Giessen: Töpelmann, 1918). Syhex = Paul de Lagarde, Bibliothecae syriacae…quae ad philologiam sacram pertinent (Göttingen: Dieterich, 1892). PAM = Palestine Archaeological Museum (photographs).

¹ Preliminary notice in my article “The Qumran Manuscripts and Textual Criticism,” VTSup 4: Volume du Congrès:Strasbourg 1956(Leiden: Brill, 1957) 155–57Google Scholar; see the discussion by Kahle, Paul, “The Greek Bible and the Gospels: Fragments from the Judaean Desert,” SEI (TU 73; Berlin: Akademie-Verlag, 1959) 615–18.Google Scholar

² See the opinion of C. H. Roberts, quoted by Kahle, “Greek Bible,”616: “[to] the end of the first century B.C. or [to] the opening years of the first century A.D.”

³ “That a somewhat awkward Greek rendering of Numbers has been reworked anciently to yield the recension contained in our later codices” (“Qumran Manuscripts,” 157).

⁴ As discussed by Shenkel, James Donald, Chronology and Recensional Development in the Greek Text of Kings (Cambridge: Harvard University, 1968), esp. 5–21, 113–20, 124–29Google Scholar, following on Barthélemy, Dominique, Les Devanciers d'Aquila (VTSup 10; Leiden: Brill, 1963)Google Scholar and Thackeray, Henry StJohn, The Septuagint and Jewish Worship (2d ed.; London: Oxford University, 1923).Google Scholar Nor does 4QLXXNum relate directly to the “Theodotionic” version of Exodus, which is also καίγε material. See O'Connell, Kevin, The Theodotionic Revision of the Book of Exodus (Cambridge: Harvard University, 1972) 292–93.Google Scholar

⁵ Compare for the Old Greek 1 Sam 11:8; 13:15; 14:17 (twice); 15:4; 20:6,18, 25,27; 1 Kgs 21 (MT 20): 15 (twice), 26. For the καίγε sections, 2 Sam 18:1; 24:2, 4; 2 Kgs 3:6; 9:34; 10:19; 11:15.

⁶ Except for a hexaplaric note to 30:12 in MS j, attached to ἐν τῇ ἐπισκοπῄ αὐτῶν, reading ἄλλος ἤτοι κατὰ τὸν ἀριθμὸν ἀυτῶν. This ἄλλος is not ɑ′ σ′ or θ′: see above. Also, under 30:13 MS M has a hexaplaric note including ἀριθμόν, giving no source. Syhex, which in 30:12 has bĕmenyānāʾ dîlěhōn 3 times, once with a σ′ θ′ attribution, and which gives bĕmenyānāʾ again in 30:13, must be rated a doubtful witness. In 38:25–26 it uses ʾestĕʿar for ἐπισκέπτεσθαι, as regularly in Numbers (see the discussion with 3:42 below) and elsewhere. If it does not do the same in Exod 30:12–13 the reason may easily have been the presence of ἀριθμόν in one or two places in its prototype, though surely not in all four.

⁷ Contrary to what is supposed in “Qumran Manuscripts,” 155.

⁸ In “Qumran Manuscripts,” 156, the writer, influenced by LXX Exodus, tried to read διωστρας in vv 6 and 8; he cited ρτρας as the reading of v 11. At the end of v 12, the Hebrew and 4Q's Greek (ρτηρός, singular) are not the same as in vv 6, 8, 11; and it is from v 12 that the missing v 10 in 4Q must be reconstructed. In v 14a LXX ναϕορεῖς again renders and 4Q needs to be reconstructed with ρτρας. The extension of v 14, withναϕορεῖς in LXX, also found in Sam., was not present in 4QNum, see below. Kahle therefore states the case inaccurately when he affirms (“Greek Bible,” 617): “In 5 places in these short fragments (4, 6, 8, 11, 12) Hebrew is translated by αρτηρας (not διωστηρας, as Skehan supposes).”

⁹ The μοχλοί for in Exod 38:24 (MT 7) is sheer confusion; it normally renders .

¹⁰ The δι᾽αὐτς of G^B in v 8 preceding τοὺς ρτρας serves the same function, and is required by the letter count in col. ii, line 28.

¹¹ This is also the reading of G^L in Exod 38:13 LXX (37:17 MT) as against ἢ ϕωτίζει of G^BAO etc.

¹² This reading parallels Exod 38:9 LXX (37:10 MT) in both G^BLA etc. and G^O.

¹³ This form is cited from q′ at Exod 39:34 in G^O.

¹⁴ The αὐτῇ of G^O ( = MT ) near the end of v 9 was probably not present in 4Q.

¹⁵ On this hopelessly wrong rendering and its parallel (and source) in Exod 27:3 LXX, see Gooding, David W., The Account of the Tabernacle (Cambridge: University Press, 1959) 60–61.Google Scholar

¹⁶ Compare Gooding, The Account of the Tabernacle, 61.