^{page 6 note 1} The incompatibility of the conception of a sun-god with that of a fertility-god is fully discussed in Baudissin's subtle and penetrating study, Adonis und Ešmun, 106 ff., 169 ff.

^{page 6 note 2} In Acta Orientalia, VII (Leiden, 1928)Google Scholar.

^{page 6 note 3} Contenau in Revue Biblique, 1917, 3 Google Scholar.

^{page 7 note 1} Justification for this procedure is furthermore found in the material discussed by Mr.Smith, Sidney, J.R.A.S., 1932, 295 ff., esp. 304 Google Scholar.

^{page 7 note 2} In the Second Preliminary Report of our expedition: Oriental Institute Communications, No. 16, pp. 40-6.

^{page 8 note 1} See The Times of July 10th, 1933, and the Illustrated London News of July 15th and 22nd, 1933.

^{page 8 note 2} Zimmern, , Tamūz, 718 Google Scholar; Langdon, , Sem. Myth., 131 Google Scholar; Meissner, , Bab. u. Ass., II. 31 Google Scholar.

^{page 8 note 3} The identity of the composition of the group on the seals from Khafajah and Tell Asmar, and on the stela, establishes the scene as a distinct ideogram which should neither be confused with some obscene terra-cottas nor with another group of representations in which Dr. Campbell Thompson sees a rendering of the vampire ( Liverpool Annals, XIX. 92 Google Scholar). This vampire-scene occurs on seals and on a polychrome vase from Tepe Musyan, but differs entirely from the one we are here discussing, for the priest standing beside the couch is absent, and the two figures upon it are not stretched side by side but a female figure crouches at some distance above the male one. Mr. Gadd points out to me that the scorpion underneath the couch in Plate I, fig. b may stand as a symbol for Išḫara. See Thompson, Campbell, The Epic of Gilgamesh, 74 Google Scholar, on the role of this goddess. But on the seal from Khafajah there is a dog underneath the couch; however, as a symbol of Gula, the wife of Ninurta ( = Ab-ù), it is not inappropriate.

^{page 9 note 1} Jeremias, , Handbuch der altorientalischen Geisteskulturen, 247, fig. 142Google Scholar, and Weidner, in O.L.Z., 1919, 11 ff.Google Scholar

^{page 10 note 1} Heidenreich has recently published (Z.A., N.F. VII) a seal which belongs to the Jamdat Nasr period, and upon which we find a deity indicated by a mere symbol, so that the usage which prevails on the Kudurru has an immemorial tradition behind it which is also manifest in the Gudea vase.

^{page 10 note 2} This is shown to be the meaning of the ‘Bügel-schaft’ by Andrae, , Das Gotteshaus und die Urformen des Bauens im alten Orient, 55 f.Google Scholar So also Contenau, , Manuel, II. 588 f.Google Scholar

^{page 10 note 3} A.f.O. V. 219 Google Scholar. The identity in type and function of the ‘twin’ from the plaque and those from seals like our Plate IV, b relieves us of the necessity of identifying them with any of the well-known gods who are sometimes referred to as twins, whom Ebeling enumerates. The seals show that the twins are entirely subsidiary figures and therefore probably nameless. Heidenreich, , Beiträge zur Geschichte der vorderasiatischen Steinschneidekunst, 24 Google Scholar, refers to Ningizzida and Tammuz, mentioned as twin-guards in the Adapa myth, because he sees Tammuz in the locked and bearded nude figure. The evidence for that identification, however, is very frail; and it certainly seems impossible to identify the ‘twins’ in their subservient position with those two important deities. But Heidenreich's evidence makes it probable that the nude figures with hair in locks were considered ‘sons of Ea’.

^{page 10 note 4} Commonly as standard-bearer or foot-rest of the sun-god in the Hammurabi dynasty. Instances enumerated by Prinz, , Altorientalische Symbolik, 91 Google Scholar (cf. 60) and 98. As opponent of the sun-god the bull-man occurs: Louvre, A. 142; De Clercq, 181 bis; Southesk Collection, Qe 5; P.S.B.A. XIV. 5 Google Scholar; Philadelphia, 150.

^{page 11 note 1} Langdon, , Ishtar and Tammuz, 116, n. IGoogle Scholar, quoting C.T. 33, 2, Rev. 8: ‘(Star) Hydra = (god) Ningiz-zida, Lord of the earth.’

^{page 11 note 2} Langdon, , Sem. Myth., 119 Google Scholar.

^{page 11 note 3} Bab. und Ass. 11. 410 Google Scholar.

^{page 11 note 4} See above, p. 8, n. 2.

^{page 11 note 5} Jacobsen, in Oriental Institute Communications, 13 (Tell Asmar and Khafaje), 58 Google Scholar.

^{page 11 note 6} The other impression, As. '32-682, shows only the caduceus and the ‘twin’ gate-keepers.

^{page 11 note 7} Jastrow, , Religion, 11. 900 Google Scholar.

^{page 12 note 1} Mrs. Van Buren, in this journal pp. 74-5.

^{page 12 note 2} Mr. Parker refers to a figure on p. 722 of Schreiber, Herpetologia Europaea.

^{page 12 note 3} It is very interesting to note that the fertility-god of Plate II, b drinks through a tube from a large vessel. Now we have found just such a tube in a hoard of vessels, lamps, &c., which are identical in form with those from the so-called Royal Tombs of Ur, but are, at Eshnunna, parts of a temple service dedicated in an inscription to the ‘Lord of Vegetation’. And a similar jar, with three drinking-tubes standing in it, is pictured in the scenes of the divine connubium (Plate I, b). All this evidence supports the view of Mr.Smith, (J.R.A.S., 1928, 849 ff.Google Scholar) and of Professor Bohl (Z.A., N.F. v. 83 ff.) that the interments at Ur do not represent royal burials at all, but contain victims of a ritual connected with the god of fertility; exceptional possibly because some special circumstance, such as a famine, or the founding of a new state temple, made it essential to achieve the highest efficacy, only obtainable through the actual sacrifice of human life. The main actors would then be the ‘Meskalamdug’, no doubt the priest who acted the god for the king in the fatal ritual, and the divine bride. It has not, to my knowledge, been noted that Meskalamdug and Meskalamduglugal were each buried high up in the shaft of a separate tomb which at least in the case of the latter contained that of an adorned woman, who would be the adorned bride. For Meskalamdug see Fig. 1 on p. 421 of A.J. VIII;for Meskalamduglugal see A.J. IX. 314 and pl. XXIXGoogle Scholar. It would seem likely then that the name of the male actor was only an epithet which he bore in the ritual drama; and if this can be translated, as has been suggested, ‘Warrior of the Nether World’, we shall see that this title is perfectly in keeping with certain myths which have influenced the ritual of the New Year's Festival as known in later times (but which our seals allow us to trace back to the period of Sargon of Akkad) when the king is identified with the warrior Ninurta who avenges his father ( Ebeling, , Tod und Leben, 36, II. 20–2Google Scholar). The union of Ningirsu and Bau took place on New Year's Day according to Gudea, Statue G, 5; Cylinder B describes, as Böhl has seen, a ceremony such as may have been the occasion of the interments at Ur. The association of the jars and drinking tubes with the connubium which we see in Plate 1, b finds its explanation from the same text. After the union in Cylinder B 5, 11 f. ‘the Anunnaki of Lagash, together with the lord Ningirsu, feasted in that place’ (after Barton). And similarly, when a renewed union of Bau and Ningirsu had caused satisfactorily to grow all that was needful for the city and the temple (in 17, 2 ff.) we read that Gudea rejoiced and provided drink and food. It may well be that the prevalence, on Early Dynastic cylinders, of the feasting scene in which two people are shown drinking through tubes refers specifically to the feast held after the connubium of the gods which was thought to be of such a predominant importance for the well-being of the community.

^{page 12 note 4} Weber, 392.

^{page 13 note 1} e.g. As. '31-281; Mus. Guimet, pl. III, 30; Philadelphia, XI. 164 Google Scholar.

^{page 13 note 2} Zimmern, , Tamūz, 716 Google Scholar; Dhorme, , Rel. Ass.-Bab., 106 Google Scholar; Langdon, , Tammuz and Ishtar, 7, n. 2Google Scholar. Further proof of a relationship (which to my mind derives from original identity) between Tammuz and Ningizzida will be found in Mrs. van Buren's article in this journal, viz. Ningizzida's marital relation with Geshtinanna and the use of the name Umunmuzida for Tammuz.

^{page 13 note 3} Langdon, , Sem. Myth., 104 Google Scholar. ‘He (Anu) created the corn-goddess the goddess of flocks and wine, Ningiszid, Ninsar … as those who enrich the fixed sacrifices.’

^{page 13 note 4} Legrain, , M.J., 1929 Google Scholar, interprets as Ashnan the corresponding figure of No. 77 on pl. XXXVI.

^{page 13 note 5} The same conjunction of plants and goats appears on the ‘Kultrelief’ from Ashur published by Andrae (W.V.D.O.G., No. 53). Mrs.van Buren, , Flowing Vase, 102–4Google Scholar, may well be right in seeing in the figure of the main god ‘Tammuz’, if we take this name simply as the most widely spread designation of the fertility-god with whom we are dealing in this section of our article. This does not exclude at all the possibility that he represents Ashur whose characteristics include many of the same nature which need not be due to borrowings from Marduk, but may, in the case of Ashur as in that of Marduk, belong to the oldest elements in the composite character of those gods. See Sidney Smith, in J.E.A. VIII (1922), 41 ff.Google Scholar, and Bulletin of the School of Oriental Languages, IV. 70 ffGoogle Scholar. It seems to me that Tallqvist's objections to this view (Der Assyrische Gott, 110 f.) are vitiated by the fact that all texts referring to Ashur are late and that the character of State-god in the case of Marduk also causes a relegation to the background of ‘Tammuz’ characteristics.

^{page 14 note 1} Weber 440. Dr. A. Moortgat, of the Vorderasiatische Abteilung, Berlin Museum, has been so kind as to verify for me that the seat of the god ends behind in an animal's head; the design is not clear enough to decide whether it is the head of a serpent or that of the ‘lion-bird’; he inclines, with reserve, to the latter view. The god of fertility of our group III seems in any case on this cylinder to be identified with the Ningizzida of our group I.

^{page 14 note 2} Deimel, , Pantheon, 202 Google Scholar.

^{page 14 note 3} Gudea, Cyl. B, 10, 4 ff.

^{page 14 note 4} Die Ionische Säule, Bauform oder Symbol? (Studien zur Bauforschung herausgegeben von der Koldewey Gesellschaft, Berlin 1933), pl. VGoogle Scholar.

^{page 14 note 5} Z.A. N.F. VII. 200 ff.

^{page 14 note 6} Ball, , Light from the East, 15 Google Scholar.

^{page 14 note 7} ‘“Labbu” strictly speaking means “the raging one” and is often employed for lion’ ( Langdon, , Sem. Myth., 287 Google Scholar); tentatively so, Ungnad, , Religion der Bab. und Ass., 61 Google Scholar.

^{page 14 note 8} Langdon, , Sem. Myth., 119 Google Scholar.

^{page 15 note 1} Ninurta as conqueror of Zu: Zimmern, , Zum Bab. Neujahrsfest, II. 8 Google Scholar; Langdon, , Sumerian and Babylonian Psalms, 232, l. 26Google Scholar.

^{page 15 note 2} Philadelphia, 177.

^{page 15 note 3} Revue d'Assyriologie, VIII. 94 Google Scholar.

^{page 15 note 4} Gudea, Cyl. B, 5, 11 ff.; ibid. XVII. 2 ff.

^{page 15 note 5} See Langdon's interpretation of Ningirsu's name: Epic of Creation, 216. Also Zimmern, , Tamūz, 717 Google Scholar, and Deimel, , Pantheon, 201 fGoogle Scholar.

^{page 15 note 6} Compare the suggestion of Mrs.van Buren, (Clay Figurines, 233, No. 1137)Google Scholar that certain ornaments worn by priests might be ‘badges of office, indications of their rank in the hierarchy’, comparable with the symbols of those who attained certain grades in the Mithraic worship.

^{page 16 note 1} Oriental Institute Communications, 13, 51 ff.

^{page 16 note 2} Oriental Institute Communications, 13, 58, n. 2 must not read, of course, ‘Hardly before’, but ‘Hardly after the Dynasty of Akkad for the personal name Tishpakkum occurs on an unpublished tablet (now in Chicago) of that period’.

^{page 16 note 3} Oriental Institute Communications, 16, 53 ff.

^{page 16 note 4} Zimmern, , Tamūz, 716 f.Google Scholar

^{page 16 note 5} Oriental Institute Communications, 13, 19, fig. 14.

^{page 16 note 6} Jacobsen, loc. cit., 42 ff.

^{page 17 note 1} The cylinder in Sarzec, De, Découvertes en Chaldée, I. 293 Google Scholar; the stela in Schäfer-Andrae, , Die Kunst des alten Orients, 474 Google Scholar. See this journal, Plate xi, c.

^{page 18 note 1} Thompson, R. Campbell, Devils and Evil Spirits of Babylonia, I. 200 ff.Google Scholar Tablet K, II. 183-95. On pp. liii ff. of the same work the passage is discussed in full, mainly with references to various attempts, fashionable at the time, to read in it a description of the Garden of Eden. We may well admit that these attempts have failed, and that kiškanu refers to a plant used as medicine for a sick man (ibid., p. lix); but that does not imply that kiškanu is ‘never used with a mythological meaning’. As medicine it is gathered by three minor gods of Eridu; but in the preceding lines of the text, which form our quotation, its mythological prototype and the place where it is to be found are described. And in this description the point which interests us above all is how Shamash and Tammuz came to be mentioned here together. Incidentally an allusion to the victory over death won by Tammuz and Shamash each in his own way would be most appropriate in a magical text dealing with medicine.

^{page 18 note 2} The main figure in the boat is characterized as a sun-god by the flames rising from his shoulders. One might think of the moon-god, who, as Dr. Campbell Thompson has shown, was thought to be travelling through the sky in a boat. But in the first place there is no instance of which I know, in which the moon-god is represented with rays rising from his shoulders. And in the second place our god holds in one instance the saw with which the sun-god ‘cuts decisions’ (Weber 406; VA 2952; on the saw, Meissner, , in M.V.A.G., IX. 234 Google Scholar). Since few pre-Sargonid seals are pictured in this article it is perhaps not superfluous to point out that the god in the boat of Plate III, g does not wear the crescent on his head, but that this is the normal way in which the horned crown of the gods is rendered on pre-Sargonid seals. It is true that in Plate III, g moon and stars appear above the boat, but this merely shows that the sun-god's journey ‘during the night’ is depicted.It is obviously impossible to express the notion which I have put between quotation-marks in any other way. That the sun passes during the night through the earth or through waters underneath the earth is, of course, a widely spread idea.

^{page 19 note 1} Illustrated London News, 07 15th, 1933, 97, fig. 3Google Scholar, and Smith, Sidney, Historical Texts, Frontispiece, fig. 4, and pp. 72 ffGoogle Scholar.

^{page 19 note 2} Khafajah II-276 and VA 2952 (Weber 406).

^{page 19 note 3} After Gadd, , in Myth and Ritual …, edited by Hooke, S. H., 43 Google Scholar; Thureau-Dangin, , Rituels Accadiens, 82 Google Scholar.

^{page 19 note 4} Plate II, i where it is carried by the god on the right; further VA 242 (Weber 440), Southesk Qa 22, Ward 374.

^{page 19 note 5} Plate III, e and Khafajah III-279; As. '31-25.

^{page 20 note 1} M.J., 1929, pl- xxi, 54 Google ScholarPubMed.

^{page 20 note 2} See p. 19, n. 1, above.

^{page 20 note 3} It is extremely interesting to note that similarly the seat of the fertility-god on the seal, Weber 440 (see above, p. 14) ends in a serpent's or dragon's head.

^{page 20 note 4} See Tenner, in Z.A., N.F. IV, 1929, 186 ff.; Götze, , Kleinasien, in Müller's, Iwan Handbuch der klassischen Altertumswissenschaft, III. 1. 3 Google Scholar; p. 130.

^{page 21 note 1} Pantheon, p. 204.

^{page 21 note 2} Khafajah II-276; III-279; Mackay, , Report on the Excavation of the ‘A’ Cemetery at Kish (Chicago, 1925), Part I, pl. VI, 15, 17 Google Scholar; and our Plate III, g.

^{page 22 note 1} In Acta Orientalia, I. 155–98Google Scholar.

^{page 22 note 2} A.J.O., VI, pl. III, 2 Google Scholar.

^{page 23 note 1} e.g. Bibl. Nat. 65, 66; Louvre A. 127, 128, 129, 130; S. 471.

^{page 23 note 2} e.g. De Clercq 181, 181 bis; Philadelphia 150.

^{page 23 note 3} We therefore interpret his weapon as a firebrand rather than as lightning, though it is of the same shape as the lightning held by the storm-god on ‘Hinke’ monuments of Senjirli and elsewhere.

^{page 23 note 4} De Clercq 176.

^{page 23 note 5} Zimmern, , Zum. Bab. Neujahrsfest, I. 131 Google Scholar; compare also Ebeling, , Tod und Leben, 36 Google Scholar.

^{page 24 note 1} Weber 440.

^{page 24 note 2} Zimmern, , Zum Bab. Neujahrsfest, II. 3 Google Scholar.

^{page 24 note 3} The god does not emerge from the mountainside as is often supposed and as our photograph might suggest. The separation between god and mountain is clear in such figures as Prinz, , Altor. Symbolik, pl. XI, 12 Google Scholar, or Schäfer-Andrae, loc. cit., 452. The erroneous connexion of the emerging god with the outline for the mountain is no doubt largely responsible for the interpretation of the tree-clad mountain as a bent tree. Opitz ( A.f.O. VIII. 329 ff.Google Scholar) sees in our Plate IV, c an illustration of the new episode of the Gilgamesh legend which Mr. Gadd discovered among the texts found at Ur, but a comparison of the translation (R.A., XXX. 127-43) with the seal shows that the seal and the text have very little in common even if one interprets as a tree what we believe to be a mountain covered with vegetation.

^{page 24 note 4} Langdon, , Epic of Creation, p. 36, l. 16Google Scholar; p. 37, n. 5.

^{page 25 note 1} Langdon, , Tammuz and Ishtar, 8, noteGoogle Scholar.

^{page 25 note 2} Ebeling, , Tod und Leben, 36, l. 20Google Scholar.

^{page 25 note 3} See also Speleers, No. 590.

^{page 25 note 4} The last view is held by Mrs.van Buren, E. Douglas, The Flowing Vase, 37 Google Scholar; she sees in this structure the sanctuary built for Marduk by the grateful gods after his victory over Tiamat. If Mrs. Van Buren sees Marduk in the sun-gods on such cylinders as Bibl. Nat. 73 or B.M. 120,540 (her Figs. 7 and 8) we can only follow her with reserve. Since the occurrence of Marduk in Sargonid times remained to be proved, the objection could be made that the sun-gods on the seals just quoted might well be other deities. On the other hand, the indubitable evidence from the seals in our Plate iv, a and c adds probability to her view, and so would her contention that on Bibl. Nat. 73 and B.M. 120,540 Ea is ‘an interested onlooker’, if only we were sure that there was always a connexion between the various scenes on one cylinder. Since Mrs. van Buren's exhaustive work and the present article often cover the same ground with results which sometimes agree and sometimes diverge, it is well to state here that the book appeared while the present writer was still in the field, only becoming available to him in August 1933, by which time this article was already written.

^{page 26 note 1} Ebeling, , Tod und Leben, 25, Rs. 1. 4Google Scholar.

^{page 26 note 2} Langdon, , Epic of Creation, IV. 135–7Google Scholar.

^{page 26 note 3} Ebeling, , Tod und Leben, 36, l. 19Google Scholar.

^{page 26 note 4} Ebeling, loc. cit., 33; Zimmern, , Zum. Bab. Neujahrsfest, II. 49 Google Scholar.

^{page 26 note 5} Ebeling, loc. cit., 29.

^{page 27 note 1} This group has been studied by Mrs.van Buren, , The Flozving Vase, 41–50 Google Scholar, in detail.

^{page 28 note 1} Through the courtesy of Mr. J. M. Upton, Acting Curator of the Department of Near Eastern Art at the Metropolitan Museum of Art, New York, I obtained photographs of the seals figured by Ward as Figs. 291 and 454 c. The seals are now in the Morgan Library. The first shows the bird-man brought to judgement before Ea, who rises from the waters of the apsu which surround him in a triple stream. Ea places his foot on land or a mountain represented by three dots. The other photograph (our Plate v, g) shows that Ward omitted in his Fig. 454 c the mountain upon which Ea places his foot and which is clearly discernible in the photograph. In the right-hand group one sees an anthropomorphic god, presumably Zu in human shape as in our Plate in, c, attacked by a god and a lion-headed eagle, while a smaller lion-headed eagle is shown in the field below. The attacking god is identified by the larger Imgi and is, of course, Ninurta or Ningirsu. The smaller Imgi may well be a mistaken drawing for an ordinary eagle which would have characterized the victim as it does in our Plate III, c. There is a seal impression in Philadelphia (No. 177) which is almost identical but which is too fragmentary to show much. The larger Imgi is there, but the smaller one is not, and of Ea we see, if anything at all, only one of the fishes which accompany him.

^{page 28 note 2} The Babylonian Akitu Festival, p. 156; Zimmern, , Zum. Bab. Neujahrsfest, 118 Google Scholar.

^{page 28 note 3} Ebeling, , Tod und Leben, 33, ll. 24–39 Google Scholar. It is interesting to remember that the mountains where the sun set and rose were considered to be the entrance and exit of the nether world and therefore of the country of the dead (see Kristensen, W. B., in Verhandelingen Kon. Akademie v. Wetenschappen, Amsterdam, 1916, 54ff.Google Scholar); on the whole, the seal-cutters do not exploit this double aspect of the mountains, and show them either as the scene of the sun-god's rising (in a large number of seals not discussed here) or as the nether world in which a god is held captive. But in one seal in the Louvre (A. 144) the sun-god is seen rising from the mountain in a manner which contrasts with the activity with which he generally steps upon it; he emerges with drooping arms from the nether world, and on this seal we see at the side the bird-man being led away as a captive.

^{page 28 note 4} Mrs. van Buren, The Flowing Vase, pl. II, 5 (plate ix, a in this journal). On this seal the figures are not all part of one scene as in our Plate v, d, but besides the contest of Ninurta and Zu there is a row of stiffly placed gods, which seems to contain the most important figures of the pantheon, without our being able to say whether they are meant to appear in the role of spectators. We may at the same time protest here against the attempt to introduce the Offord cylinder once more into scientific literature! (Flowing Vase, 30). No genuine cylinder on which such an ambitious scheme of composition was attempted was ever executed in such an uncertain manner. The various points in which the forger misunderstood the draperies of B.M. 89, 115 which was his prototype, happen to suggest to Mrs. van Buren a Syro-Hittite type of dress. But could they be bungled in any other way? No Syro-Hittite cylinder, to my knowledge, shows either this style of cutting or this scheme of spacing the decoration.

^{page 29 note 1} Ebeling in Pauly-Wissowa, 14, II, 1669.

^{page 29 note 2} See above, p. 6. King, , Chronicles of Early Babylonian Kings, 8, 11 Google Scholar, cannot be adduced as proof that Marduk was worshipped in the time of Sargon of Akkad.

^{page 29 note 3} As Langdon, does, Epic of Creation, 18 f.Google Scholar; Sem. Myth., 115.