² For the reasons why publication of these fragments was postponed, see Smith H 282 n. 5.

³ HK fr. 42, the main part of which (YF 019A) was rediscovered in 1968, was intact when it was recorded by Cousin and HK. The weathered appearance of the surface where the fracture occurred shows that the stone was not broken recently. The smaller part (YF 019B) bears the last letters of II.13–14 and most of III.13–14. Its measurements are: H. 15·5 cm., W. 56·5 cm. (surface 43·5 cm.), D. 13 cm.

⁴ Hall 194.

⁵ I am grateful to Mr. D. Madge for photographing the squeezes.

⁶ In a letter to the author dated 13th August, 1974.

⁷ By “location” I mean the place of discovery. During the 1975 season we collected together the smallest fragments, in order to ensure their safety.

⁸ It should be noted, however, that fig. 3 makes the site and its buildings look twice as large as fig. 2 shows them. Fig. 2 is correct.

⁹ Hall 194 n. 22 says that “the continued use of the term ‘Great Wall’, to describe that part of the city-wall which cuts off the Esplanade, is thoroughly misleading”. It is true that the valuable survey directed by Mr. Hall has now shown that the ‘Great Wall’ is an integral part of the late (probably latter half of the third century A.D.) defences of the city. However, the term “Great Wall” has always been used by Diogenes scholars, and their continued use of it—in inverted commas—seems justified not only by tradition and considerations of convenience, but also by the fact that the richest finds of Diogenes fragments have been made in or close to this stretch of wall. Mr. Hall himself rightly draws attention (Hall 194) to the dense concentration of fragments along the wall and suggests (Hall 196) “that the stoa of Diogenes was situated at this point, and that it was dismantled to make way for the city-wall”. Whether this plausible suggestion is proved correct or not, this part of the city-walls is very great in importance, so far as Diogenes is concerned, as well as great in size.

¹⁰ Cousin 57.

¹¹ NF 39 and 40 alone contribute about 270 words.

¹² Smith K 843.

¹³ Including the introductory fragments HK fr. 59, 57–58.

¹⁴ Including HK fr. 51, which probably belongs to the Letter to Dionysius (Smith J 377–378).

¹⁵ Letter to Antipater 3·705 m., Letter to Dionysius 3·07 m. It is impossible to say to which letter NF 11 (width 75 cm.) belongs.

¹⁶ This block probably stood in the topmost course of the inscription, in the same course as the group-A blocks, though it should be noted that its height is 37 cm., not 35 cm. as stated by HK 356.

¹⁷ HK fr. 2 bears 11 lines and has a height of 47 cm. Therefore it must not be grouped with the rest of the “Private Writings”, which are carved on stones 38·5–41·5 cm. high (on HK's incorrect measurements of the height of HK fr. 24 and 26, see n. 50, 51 below). NF 38, most of which is illegible, is 46 cm. high and probably bore 11 lines.

¹⁸ These maxims are not to be confused with the maxims inscribed on separate stones.

¹⁹ It is true that the maxims, though almost certainly in the lowest course, bear letters larger than those of the Ethics, Physics, and letters to Dionysius and Antipater, but the explanation for this may be that it was thought desirable to carve them in letters similar in size to those in the line of maxims beneath the Ethics.

²⁰ NF 104 may have borne 6 lines.

²¹ These fragments have been re-edited by A. Laks and C. Millot in Bollack-Laks 341–353, Barigazzi 2–13.

²² In fr. 35.II.2–4 Chilton (B 15) takes ἁπλῶς with οὔ φασι, “absolutely deny”. However, it must be taken with ἀφθάρτους, as in fr. 34.II.4–5 (so Grilli A 398: “assolutamente immortali”).

²³ “The Plague under Marcus Aurelius”, AJP LXXXII (1961) 225–251Google Scholar. Whether one agrees with Gilliam's conclusion or not, his full discussion is very valuable.

²⁴ op. cit. 249.

²⁵ The height of HK fr. 39 is 58 cm., not 56 cm. as given by HK.

²⁶ In fact, the height of the upper margins of the fragments found in the nineteenth century is 7·5–8 cm.; but the upper margins of NF 9 (9 cm.) and NF 58 (7 cm.) fall outside these limits.

²⁷ HK (443) rashly supposed that Carus is Titus Lucretius Carus. Their theory was promptly refuted by Körte, A., “T. Lucretius Carus bei Diogenes von Oinoanda?”, Rh. Mus. LIII (1898) 160–165Google Scholar. All subsequent editors have agreed that the Letter to Menneas is by Diogenes, and, if further confirmation of this is needed, the address to Dionysius in NF 58 provides it.

²⁸ Smith B 379.

²⁹ Epicurus undoubtedly made the distinction elsewhere, as the scholium on Hdt. 66 (= Usener 311 pp. 216–217) shows. Cf. Aëtius 4.4.6 = Usener 312 p. 217.

³⁰ See also NF 5.III.4–6 and my discussion in Smith B 362–363.

³¹ For example, there is much doubt about the views of Panaetius and Posidonius.

³² Re-edited by Chilton B 127–130, A. Laks and C. Millot in Bollack-Laks 330–337, Barigazzi 16–18.

³³ NF 9 has been re-edited by Barigazzi 13–15; NF 12 by Smith F 45–17, A. Laks and C. Millot in Bollack-Laks 354–356, Barigazzi 18–20.

³⁴ The physical features of the group are listed in Smith H 305.

³⁵ See Chilton B 97.

³⁶ See JHS XCIII (1973) 236Google Scholar, Smith H 305.

³⁷ For my defence of placata, the reading of the manuscripts, against pacata (ed. Juntina, ), see C R N.S. XVI (1966) 265–266Google Scholar.

³⁸ On the reading, see note on NF 73.4.

³⁹ See Smith F 11–13.

⁴⁰ See Smith F 45. However, a more careful inspection of NF 13 reveals that the mark on that stone is rather than Δ.

⁴¹ In Epic. Pyth. 107, where the manuscripts read ὃ μάλιστα γίνεται ἐν τῷ ἀέρι in reference to the pressure of moist clouds on one another causing hail to be produced, Usener emends ἀέρι to ἔαρι—an attractive suggestion accepted by Bignone and Arrighetti among others.

⁴² Cf. Sen., NQ 4.4.1Google Scholar: guaeritur autem quare hieme ningat, non grandinet, vere iam frigore injracto grando cadat; Pliny, NH 2.61.152Google Scholar: per hiemem nives cadere, non grandines, ipsasque grandines interdiu saepius quam noctu, et multo celerius resolvi quam nives; Manilius 1.99–100: cur … hiberna aestiva nix grandine, mollior esset.

⁴³ Thesaurus Linguae Latinae 6.2, s.v. grando, has a long list of references under the heading “de damno quod fit grandine”. There are even cases on record of hailstorms having interfered with battles. At the river Crimisus in June 339 B.C. the Phoenicians were getting the better of the Greeks under Timoleon when a great storm of hail and thunder broke, causing the Phoenicians, who were facing the rough weather, to lose the advantage and at length flee (see Diod. Sic. 16.80.1, Plut., Timoleon 28Google Scholar). Livy (26.11) reports that on two successive days hailstorms prevented the Romans and Carthaginians from fighting, and comments: in religionem ea res apud Poenos versa est.

⁴⁴ Presumably the town in the Argolid rather than the Chalcidic Cleonae.

⁴⁵ The height of the upper margin of the maxim-bearing blocks varies from 6 cm. (NF 81) to 9 cm. (HK fr. 28, 29, 32, NF 83).

⁴⁶ See illustrations: HK fr. 27 in Smith J fig. 4, HK fr. 28 and 32 in Smith A figs. 20 and 19.

⁴⁷ Illustrated Smith H fig. 7.

⁴⁸ NF 45 also has an empty space sufficient for one line between lines 7 and 8.

⁴⁹ The reason for this fragment's very tall lower margin is that it bears only 9 lines.

⁵⁰ On the height of HK fr. 24, given incorrectly by HK 370, see Smith G 129 n. 22.

⁵¹ On the height of HK fr. 26, given incorrectly by HK 372, see Smith F 28.

⁵² On the interpretation and text of this fragment, see now Sedley, D. N., C R N.S. XXVI (1976) 218Google Scholar; Clay, D., AJP XCVII (1976) 308Google Scholar; Barigazzi 15–16; A. Laks and C. Millot in Bollack-Laks 327–329.

⁵³ Cf. Cic., Tusc. 2.8.20–22Google Scholar.