A Miniature Ivory Sundial with Equinox Indicator from Ptolemaic Tanis,Egypt

Petrie W. M. F. , Tanis, I: 1883–4 (London, 1885), 20 and 34.

This is hippopotamus ivory, as kindly confirmed by Caroline Cartwright on close analysis. Even apart from the Greek corpus, the use of ivory for sundials is rarely attested. However, one apparent example is of ancient Egyptian manufacture and was discovered at Gezer (Palestine). A small, half-circular, vertical dial, it is divided into 12 equal pie-wedges and therefore not based on geometrical gnomonics. The cartouches on the back date it to the reign of Merenptah (c. 1224–1214 b.c.). See MacAlister R. A. S. , The excavation of Gezer 1902–1905 and 1907–1909 , i (London, 1912), 15, and ii (London, 1912), 331; Borchardt Ludwig , Die altägyptische Zeitmessung (Berlin, 1920), 48; and Clagett Marshall , Ancient Egyptian science, ii: Calendars, clocks, and astronomy (Memoirs of the American Philosophical Society, ccxiv; Philadelphia, 1995), 95–98. Unfortunately, the present location of this dial is not recorded in the literature. In the Landesmuseum Mainz (inv. R 2321) there is a small (68-mm diameter), disk-like, portable dial, inscribed with month names in Latin. This dial was found at Mainz (latitude 50°), but seems to have been made for a latitude of about 44°. Because it uses the month August, this dial is of the first century a.d. or later, but could be as late as the fifth century. The Mainz dial was described as ivory by Derek J. de Solla Price, “Portable sundials in Antiquity, including an account of a new example from Aphrodisias”, Centaurus, xiv (1969), 1969–66, p. 246. Earlier Joseph Drecker , Die Theorie der Sonnenuhren (Munich, 1925), 61, also had it down as being of ivory. In actual fact, this object is not of ivory but antler. See Mikler Hubertus , Die römischen Funde aus Bein im Landesmuseum Mainz (Montagnac, 1997), 24–25, 125, and plate 14.9. The Mainz dial is discussed most fully by Schaldach Karlheinz , Römische Sonnenuhren: Eine Einführung in die antike Gnomonik (Frankfurt am Main, 2001), 114–23. There is also a Roman cylinder dial of the first century a.d. made of bone, in the Museo Nazionale Atestino, in Este near Padua. See Arnaldi Mario Schaldach Karlheinz , “A Roman cylinder dial: Witness to a forgotten tradition”, Journal for the history of astronomy, xxviii (1997), 1997–17. In the Renaissance, of course, ivory became a popular material for small, portable sundials. See Gouk Penelope , The ivory sundials of Nuremberg, 1500–1700 (Cambridge, 1988).

Of these, British Museum 1886,0401.1477 (also registered as 1909,0216.10) was found by Petrie at Naukratis in 1884–85: See Petrie W. M. F. , Naukratis , i (London, 1886), 16, with drawing on plate 18.6. This dial is no. 1040 in Gibbs Sharon L. , Greek and Roman sundials (New Haven, 1976). Gibbs gives its dimensions as 78 mm wide × 65 high × 34 to 50 deep (thicker at the top). According to Gibbs (p. 155), traces of engraved letters are visible below the spherical surface, a remark that was perhaps based on Petrie's rough drawing. However, we have examined this dial carefully in raking light and find no trace of any inscription. Two other, rather inexactly made miniature spherical dials in stone can be mentioned. One, in the Archaeological Museum of Rhodes (inv. APX1113), is broken, but its original dimensions were roughly 90 mm wide × 55 high × 70 deep. The other, in Syracuse (Museo Archeologico Regionale Paolo Orsi, inv. 35080), is almost complete, with dimensions 40 mm wide × 34 high × 31 deep. We thank Karlheinz Schaldach for these measurements and for sharing his electronic photographs of these two dials. A photograph of the Syracuse dial appeared in Bellina Giovanni , Su alcune misure di tempo degli Iblei: Circulo didattico “Paolo Vetri” (Ragusa, 2002), 34.

The red pigment was analysed by Raman spectography using a Dilor Infinity with a near-infrared (785 nm) laser. The spectra produced were compared with an in-house data base, and were found to correspond to the naturally occurring iron oxide, hematite (α-Fe₂O₃). This suggests that the red came from an ochre pigment. Shearman F. Ambers J. C. , and Hook D. R. , British Museum Department of Conservation, Documentation and Science, Analytical Report 2005/60.

Greek dials preserving traces of red in the incised lines include Gibbs nos. 1002 and 3031 (in the day curves). The recent publication of Karlheinz Schaldach, Die antiken Sonnenuhren Griechenlands: Festland und Peloponnes (Frankfurt am Main, 2006), provides an important extension of Gibbs's work. A number of newly published dials are included, all the sundials are illustrated by good-quality photographs and, for those for which it is possible, new analyses are undertaken, which sometimes provide corrections or alternatives to Gibbs's. Moreover, colour JPEG images are included on the CD that accompanies the book. The colour photographs of Schaldach's Objekt Nr. 25 (= Gibbs no. 3023, a conical sundial from Piraeus) shows red colour in the meridian line and the three day curves and black colour in other hour lines.

Gibbs , Greek and Roman sundials (ref. 3), nos. 1023–27. On 1024 and 1026, all eleven hour lines and all three day curves were reddened. On 1022, only the third, sixth and ninth hour lines were reddened.

This is quite different from a label, “equinox”, on the equinoctial day circle. Quite a number of Greek spherical and conical dials have labels on the equinoctial and solstitial day circles. See, e.g., Gibbs , Greek and Roman sundials (ref. 3), nos. 1001, 1072, 3047, 3058, 3060, as well as the conical sundial in the Palmyra Museum (no. 25/92 G/1982 914) described by Kurt Locher, “Three further Greco-Roman conical sundials from Palmyra, Naples and Abû Mînâ”, Journal for the history of astronomy, xxvi (1995), 159–63.

For example, see Gibbs , Greek and Roman sundials (ref. 3), nos. 3022, 3027, 3043, 3044, 3048, 3075, 3076.

These instruments are mentioned twice by Ptolemy in Almagest III 1: Toomer G. J. , Ptolemy's Almagest (London, 1984), 133 and 134. For a discussion see Evans James , The history and practice of ancient astronomy (New York, 1998), 206–7.

A white patch on the nail shank and a small white metal fragment associated with the sundial fragments were examined using X-ray fluorescence spectroscopy and found to be of lead-rich composition. Shearman, Ambers and Hook, Analytical Report 2005/60 (ref. 4).

On the theory of ancient conical sundials, see Gibbs , Greek and Roman sundials (ref. 3), 30–39, and Schaldach , Die antiken Sonnenuhren Griechenlands (ref. 5), 163–4. An excellent treatment of sundial theory from a modern perspective is given by Denis Savoie, La gnomonique (Paris, 2007).

Gibbs , Greek and Roman sundials (ref. 3), 31.

Ptolemy , Almagest I 12; see Toomer , Ptolemy's Almagest (ref. 9), 63.

Gibbs , Greek and Roman sundials (ref. 3), 33.

As in Gibbs, Greek and Roman sundials (ref. 3), 32.

If the gnomon hole at b were preserved, the line through this hole and point g would have given an independent way to establish the horizontal, which would permit determining φ without using ψ. In this case, φ would be determined entirely by eb, ω, and r_e.

Gibbs , Greek and Roman sundials (ref. 3), Eq. 2.5 and 2.6, on p. 35. According to Gibbs, these formulae are due to Constantine Ionescu-Carligel, “Contributions à l'étude des cadrans solaires antiques”, Dacia, xiv (1970), 1970–1. Note that there should be a π in the denominator of the right side of Gibbs's Eq. 2.1, on p. 33; however, the final equations (2.5 and 2.6) are correct. Using only Eq. 2.5 (involving L_w) we would get φ = 16°; using only Eq. 2.6 (involving Ls), φ = 46°. This provides an example of the well-known instability of analyses based on L or L alone. See Gibbs , Greek and Roman sundials (ref. 3), 75, as well as the scatter plots in Kościuk Jacek , “A conical sundial from Abû Mînâ”, Bulletin de la Société d'Archéologie Copte, xxxi (1992), 43–55.

Petrie , Tanis (ref. 1), 34.

This sundial has also been mentioned by Favard-Meeks Christine , “Mise à jour des ouvrages de Flinders Petrie sur les fouilles de Tanis”, in Brissaud Philippe Zivie-Coche Christine (eds), Tanis: Travaux récents sur le Tell Sân el-Hagar (Mission Française des Fouilles de Tanis 1987–1997; Paris, 1998), 113, where the Petrie reference is correctly linked to BM EA 68475.

Petrie , Tanis (ref. 1), 20 (“house No. 15”) and 34 (“Find 15”).

Petrie's book nowhere mentions a House (or Find) 66 — At least not with the number included. Unfortunately, the discoveries that concern us here do not feature at all in Petrie's field notebooks and “Journals”, so those throw no light on the question (originals at the Griffith Institute in Oxford, but for the notebooks, see the CD-ROM publication The Petrie Museum archives, issued by that museum in 1999). For more on the Bes figures and door-hinge, see the main text below, with details in refs 26 and 38.

BM EA 29083. Petrie, Tanis (ref. 1), 34, does mention “some nails” from this find. It should be noted that the register book also identifies six items of pottery as deriving from Find 15 (“San [el-Hagar, House] 15”), while oddly none occurs in Petrie's listing. That he does not mention this material is all the more surprising in view of their considerable size and excellent preservation (for their inventory numbers, see refs 47–48 below). Favard-Meeks, “Mise à jour” (ref. 19), 114, might be right in assuming an omission on Petrie's part, but one wonders if this is not the pottery that Petrie says “was obtained from a neighbouring house”.

Favard-Meeks , “Mise à jour” (ref. 19), 113–14.

See our quotation from Petrie's book at the beginning of this discussion. In the same sense, Petrie , Tanis (ref. 1), 20, talks of “one good house”.

See Petrie , Tanis (ref. 1), 34, and Favard-Meeks, “Mise à jour” (ref. 19), 113–14.

One figure is now in the Museum of Fine Arts in Boston (86.779) and the other is in the British Museum (EA 22378). The latter was published in Walker Susan Higgs Peter (eds), Cleopatra of Egypt: From history to myth (London, 2001), 99, no. 119, with a discussion by Sally-Ann Ashton; she does not mention Petrie's book and incorrectly gives the provenance as “unknown”.

BM EA 15916. Height: 2.7 cm, width: 1.4 cm, thickness: 0.8 cm. The scorpions have been reduced to rudimentary oval shapes and are not mentioned by Petrie.

BM EA 22693.

For the hanging up of divine metal figures at temples, at least in earlier times, see Perdu Olivier , “Des pendentifs en guise d'ex-voto”, Revue d'égyptologie, liv (2003), 155–66.

BM EA 15887 and EA 15864 respectively. Both have been published; see Cooney J. D. , Catalogue of Egyptian antiquities in the British Museum, iv: Glass (London, 1976), p. 16, no. 160 (with ill.) and p. 161, no. 1854 (without ill.).

This may be BM EA 15859, as suggested both by Cooney, Glass (ref. 30), 16, no. 159 (without ill.), and, with query, by Favard-Meeks, “Mise à jour” (ref. 19), 114, although no findspot in Tanis is given by the British Museum register book. Petrie, Tanis (ref. 1), 34, describes the piece he found as “an inlaid mosaic eye in glass, the cheek part being inlaid with stripes of squares of different mosaic patterns, all fitted into a green glass frame, with strips of white glass between the stripes — Though much burnt and broken it is a fine piece”. In its present state, the decorated top surface of EA 15859 has completely broken away and apparently been lost.

BM EA 15910; Petrie , Tanis (ref. 1), plate 12, no. 44; Cooney , Glass (ref. 30), 132, no. 1638 (with ill.).

BM EA 15936; see Cooney , Glass (ref. 30), 132, no. 1641 (without ill.).

A glass workshop of contemporary date was discovered by Petrie at Gumaiyima, a few kilometres to the south of Tanis. See Petrie W. M. F. , Nebesheh (Am) and Defenneh (Tahpanhes) (London, 1888), 42–44, plate 18; Cooney , Glass (ref. 30), 112–23.

BM EA 15835. See Cooney , Glass (ref. 30), 69, no. 761 (with ill.); Taylor J. H. , in Coutts H. (ed.), Gold of the pharaohs: Catalogue of the exhibition of treasures from Tanis (Edinburgh, 1988), p. 96, no. 88, with photo on p. 97 (centre).

BM EA 22508; see Cooney , Glass (ref. 30), 29–30, no. 289 (without ill.).

Cooney , Glass (ref. 30), 29–30, no. 289.

BM EA 37137. Taylor, in Coutts (ed.), Gold (ref. 35), p. 95, no. 82, describes this item as “perhaps part of a tool or weapon, but its precise nature cannot be determined”. The British Museum register book calls it a “chopper”. There is no doubt, however, that this is a door element, along with the following two items. Curiously, EA 37137 is nowhere mentioned in Petrie's book.

BM EA 22713. The lock plate is wrongly cited with the number 22718 by Favard-Meeks, “Mise à jour” (ref. 19), 113. EA 22718 is in reality a square lead weight, acquired in the same year (1885) but with no documented provenance. The British Museum register book gives no provenance for EA 22713 either, but at some point it was labelled in the collection database as deriving from the house under discussion (called “House M”, as on Petrie's site map). The lock plate is briefly discussed by Taylor, in Coutts (ed.), Gold (ref. 35), 95–96, no. 82, last paragraph. Reading the catalogue entry, one should not confuse this object with one that is mentioned at the end of Taylor's preceding paragraph; there the item intended is EA 37137, which we describe as a door-hinge (cf. our previous note). Misleadingly, while EA 22713 and 37137 are both riveted fragments, only the latter is so identified at the top of this catalogue entry.

One of these being BM EA 29083 (cf. ref. 22 above).

Petrie , Tanis (ref. 1), 34.

BM EA 15898.

British Museum, Greek Coin Hoards no. 1724 at the Department of Coins and Medals.

BM EA 18348 and 22654; for the latter, see Petrie , Tanis (ref. 1), plate 12, no. 45.

BM EA 22495. Petrie, Tanis (ref. 1), 34, speaks of only one “pin-head” (with query). Also Favard-Meeks, “Mise à jour” (ref. 19), 113, fails to indicate that there are as many as seven pins, acquired together and given the same inventory number. She suggests that these pins, which are never more than 10 cm long or 1 cm wide, were used for weaving, though in what manner this would have been is hard to see.

BM EA 15819.

The jars in question are BM EA 22171, 22175, 22361, 22363 and 22384. Note that amphora EA 22171 was missed in the listing of Favard-Meeks, “Mise à jour” (ref. 19), 114, while on p. 131 she wrongly cites jar EA 22363 a second time as if it were part of Find 35, the well-known house of Ashokhi (‘Bakakhuiu’).

BM EA 68473.

BM EA 15886 and 15932, both illustrated by Petrie, Tanis (ref. 1), plate 12, nos. 13–14. Petrie's comment about the papyri occurs only on his p. 62, in an index to this plate. For more recent literature on EA 15932, see also the following note.

Stanwick P. E. , Egyptian royal sculptures of the Ptolemaic Period (Ann Arbor, 2000), plate 209; and Stanwick P. E. , Portraits of the Ptolemies: Greek kings as Egyptian pharaohs (Austin, 2002), 61 and fig. 244.

BM EA 22480; see Petrie , Tanis (ref. 1), plate 12, no. 51.

On the possible reuse of ancient scarabs in Graeco-Roman Egypt, see the remarks by Hornung Erik Staehelin Elisabeth , Skarabäen und andere Siegelamulette aus Basler Sammlungen (Mainz, 1976), 28.

For overviews of pharaonic time-keeping devices, see Borchardt , Die altägyptische Zeitmessung (ref. 2); Sloley R. W. , “Primitive methods of measuring time with special reference to Egypt”, Journal of Egyptian archaeology, xvii (1931), 166–78; and Clagett , Calendars, clocks, and astronomy (ref. 2), 83–98. A tomb inscription first documents the use of water clocks for the reign of Amenhotep I (c. 1525–1504 b.c.). The oldest actual implement is an L-shaped shadow clock from the reign of Thutmose III (c. 1479–1425 b.c.), for which see now also Lippincott Kristen ., The story of time (London, 1999), 108, no. 101.