Characteristics of this type include a handle ornament that is either figurative or in the shape of a crescent, decorated triangle, or two-lobed bud; and a usually round-tipped volute-nozzle, exceptionally with angular tip. Lamps may have one or two nozzles, but some have even more (see cat. 562). Shoulder forms vary considerably on early examples. Bailey’s groups i and ii, dated Augustan to Claudian/Neronian, show Loeschcke flat shoulder forms I, II a, and III a, as well as various rills and moldings not recorded in Loeschcke’s classification. Bailey’s group iii, dated Tiberian to Flavian, still shows the flat shoulder forms II b and IV a and hybrid forms III a, IV a, and IV b, while Bailey’s group iv has rounded shoulder forms VI a and VII a exclusively. Most lamps in Bailey’s group v, dated Claudian to Early Trajanic, have flat shoulder forms IV a and IV b, except for two examples with the rounded form VII a.
All the Getty lamps of Loeschcke type III have flat shoulders, either of unusual forms or of Loeschcke forms I a, II b, III a, IV a, or IV b. This sign—flat shoulder—that the lamps are early is supported by the presence of a base-ring on ten examples out of thirteen. The three lamps with flat base marked off by one circular groove have shoulders of Loeschcke form IV a or IV b. Out of thirty-nine BM lamps of the type here studied and recorded in Bailey BM II that still preserve their base, seventeen have a base-ring and twenty-two a flat base. The latter is the standard in Bailey group v, dated Claudian to Early Trajanic.
Figurative discus decors are scarce: out of fourteen examples, two are related to mythology, one to gladiatorial equipment, one is a rosette, three are radiated bands, and seven discuses are either plain as cat. 156 or decorated by rings or circles.
Out of the fourteen ornament handles, four are in the shape of a two-lobed leaf or lotus bud. This shape is interpreted by Bailey as the external female genitalia (vulva). Such a reading might be justified for cat. 150, but less so for cats. 147–48 and 151. In describing his no. 387, p. 98, pl. 46, Hayes 1980 speaks of “a two-lobed split leaf, i.e., a lotus bud.” Against Bailey’s interpretation, lamp no. G 183, pl. 15, in Casas Genover and Soler Fusté 2006, shows a central slit decorated with vegetal-pattern similar to a leaf. From the same authors, see also no. G 68, pl. 6, where the decor is undoubtedly only vegetal. Three ornament handles are in the shape of a crescent: cats. 153 and 158 themselves decorated with smaller crescent, the third, cat. 154, with a bust of Jupiter; five handles are in the shape of a leaf: one decorated with a bust of Serapis (cat. 155), one with a head of Bacchus (cat. 149), and three with a plain leaf (cats. 152, 156, and 157); one handle is in the shape of an eagle (cat. 159); and a last one is decorated with palmette and acanthus leaves (cat. 160).
In the introduction to his type D, which concerns Italian lamps only, Bailey sums up the archaeological data that permit assigning the start of the production to Late Republican and Early Augustan times (BM II, pp. 199–201). He considers the production to end no later than the Early Trajanic period. But outside Italy, in various provinces, the type continued to live on much longer (Bussière 2000, p. 71): until the second century in Asia Minor, where a Cnidian lamp of the type is dated by Bailey A.D. 80–120 (BM III, Q 2686); in Pannonia, where Iványi mentions locally made examples found in situ associated with coins of Hadrian (Iványi 1935, p. 12); until the end of the second and beginning of the third centuries A.D. in Libya (Joly 1974, no. 35, signed MAVRICI, no. 41 AGRI, no. 51 PUVLLAEN), in Tunisia (Deneauve 1969, no. 54.7, signed LVCCI = LVCCEI), and in Algeria (Bussière 2000, no. 161, signed MAVRICI, no. 164 CLVCSAT, no. 165 bis PUVLLAENI) (all workshops active at the time considered); and also in Austria, where examples from Lauriacum are dated to Alexander Severus (Deringer 1965, p. 120, nos. 374–75).
Some of the lamps found in the provinces may have been exported Italian products, but most of them were locally made. A close similarity in the shapes of the body, shoulder, and nozzle between cat. 156 and Bailey’s lamp Q 2686, from Cnidus, is reason to attribute the Getty lamp to the Eastern Mediterranean, possibly even to Cnidus. Following the same approach, it is equally plausible to consider an eastern origin for the three Getty examples of Loeschcke type III (cats. 148, 152, and 155). At least two of them bear a striking similarity in color of clay and slip to lamps of the same type published in various catalogues, with certified east Mediterranean place of manufacture or origin. Comparing cat. 155 with Bailey’s Q 2688–Q 2689 and the various frr. shown on pl. 77 of BM III—all found in Cnidus—it is even tempting to attribute the Getty lamp to the same Cnidian place of manufacture or origin.
The two first lamps, cats. 147 and 148, which have no parallels in BM II, are characterized by a deep cylindrical body and the presence of two hinges on the discus rim at the foot of the handle. These hinges were intended to hold a lid for the filling-hole (to prevent mice from drinking the oil). Clay lamps with such features were inspired or copied from Hellenistic bronze models, as first proposed by Loeschcke 1919 (pp. 473–74), then by Broneer 1930 (p. 74), Perlzweig 1961 (p. 73, no. 11), and others since.