^{page 107 note 2} e.g. Kubitchek, Rückgang des lat. im Orient.

^{page 107 note 3} See Mrs. Hasluck's (Miss Hardie's) satisfactory statement of the evidence in J.H.S. 1912, p. 148 f. The much extended evidence now published fully confirms her general view.

^{page 108 note 1} τὸ πρὀς Άιτιοχείᾳ τῇ πρὸς Πισιδίᾳ (ὶερόν), p. 557: compare p. 577. He describes Antioch among the Phrygian cities, along with Kolossai, Lysias, Apameia, ect.

^{page 109 note 1} There is no room to restore ωρ. Mrs. Hasluck by a slip prints ω [ρ], but there is only I on the stone (as her epigraphic text shows). The first three letters are blurred, but the eye accustomed to these texts reads HCY as certain.

^{page 109 note 2} From this grew a paper in J.H.S. 1918, on the right use of old epigraphic copies.

^{page 109 note 3} Anderson, Calder and myself.

^{page 109 note 4} Journal Royal Asiatic Society, 1882.

^{page 110 note 1} Athen. Mit. 1898, p. 108.

^{page 110 note 2} It is easy to complete Mrs. Hasluck's description and correct it in certain points: e.g. on p. 12, II. 6–7 it would be necessary to read ‘to approach the stadium’ instead of ‘to pass round the theatre’.

^{page 110 note 3} Measurements vary, and no wall is exactly parallel, or at right-angles, to another.

^{page 111 note 1} The Moslem population of Asia Minor continue to feel marked veneration for the old sacred places, and sometimes actually participate in Christian religious festivals. [This statement, true until recently, may have ceased to be so as mutual goodwill was destroyed gradually.] It is recognised by most if not all travellers in Asia Minor that the so-called Turkish population is in large degree the old Anatolian people turned Mohammedan.

^{page 113 note 1} On the name and situation see J.H.S. 1918, pp. 144–6.

^{page 113 note 2} Strabo evidently ranked both equally among the great ancient Anatolian religious centres.

^{page 114 note 1} It is described in Annual B.S.A. 1911–2, pp. with photographs.

^{page 114 note 2} It was destroyed along with the sanctuary.

^{page 115 note 1} Two inscriptions of very late character (Anderson, J.H.S. 1913, p. 281 f.) are built into the coarse reconstruction of the wall beside this little exit.

^{page 115 note 2} The building is rough and uneven, and measurements vary considerably. The N.E. wall is not straight.

^{page 115 note 3} The N.E. wall varies from 3 ft. 8 in. to 6 ft. 2 in.; it is by far the most irregular.

^{page 117 note 1} In C.R. 1919, pp. 1 ff. reasons are stated to prove that the family took the name Gaios Kalpournios about A.D. 70, and resided at Gemen.

^{page 117 note 2} In this rough Sanctuary measurements vary greatly at different points. The walls were of dry stone laid without mortar.

^{page 117 note 3} See especially Anderson loc. cit.; also my paper in Annual B.S.A. 1912, p. 29 ff.

^{page 118 note 1} Not a scrap of the Temple, above the stylobate, was found, except two or three blocks of the lowest course. The stones (marble ?) were carried away.

^{page 120 note 1} If any visitor finds one Byzantine coin lying on the ground a little north of the central sanctuary he may understand that this was brought to me from the lower town, and was accidentally lost.

^{page 121 note 1} See Anderson, J.R.S. 1913, p. 268 f.

^{page 121 note 2} The letters are those used in Mrs. Hasluck's plan.

^{page 121 note 3} The name is given only to springs of potable water. At Apameia-Kelainai there is only one Huda-verdi, for the water of all the other springs is undrinkable: see C.B. Phr. ii, p. 400.

^{page 121 note 4} In Cont. Rev. 1912, Sept., p. 372 f. the view and scenery and fountain are described as dictating the choice of this home for the god. Perhaps also the new bieron had some resemblance to the old Phrygo-Pisidian theocratic centre.

^{page 123 note 1} Antioch was situated on the right bank of the river, which runs in a deep, narrow gorge. The rocks on the left side (south-east) are higher than on the right; but they are equally precipitous.

^{page 124 note 1} The letters of Mrs. Hasluck's plan are adopted.

^{page 124 note 2} Corrupted to δέρμα in Souidas, a meaningless saying, although there are some scholiastic explanations of it. The correction, which is convincing, was made by Calder: the ‘glance of Epimenides’ was needed in the case of hidden mysterious things (ἐπὶ τῶν ὰποθέτων): Ramsay in Quarterly Review, 1919, p. 389.

^{page 125 note 1} I cannot doubt that the dedication ‘an Unknown God’ (Acts xvii, 23) was attributed to Epimenides: the speech of Paul to the Council of Areopagus is full of the spirit of Athens and of Epimenides, and a verse from a poem attributed to the Cretan prophet is a marked feature in Paul's argument, as Rendel Harris detected: see my paper in Quarterly Review, 1919.

^{page 125 note 2} The dedications on marble, found mostly inside the central sanctuary, can be dated with much greater confidence than those cut on the wall, and often furnish a norm for the latter.

^{page 126 note 1} That this word indicates quite late date may be taken as certain.

^{page 126 note 2} The number is now very much greater, but Aur. as a pseudo-praenomen is still almost as rare as ever.

^{page 126 note 3} The inscriptions of this society are collected in Studies in the Eastern Provinces, p. 300 ff. Annual, 1912, loc. cit.

^{page 126 note 4} On imperial estates in the East the pseudo-praenomen Aur. is almost universal.

^{page 128 note 1} See on this the Teaching of Paul in terms of the present day, pp. 10, 122, 285.

^{page 129 note 1} The Roman custom of dedicating a tenth of their gains to Hercules probably springs from an old Aegean and West-Asian custom.

^{page 129 note 2} The church was built mainly of stones from the sanctuary. So perhaps was the monastery, which has not been excavated.

^{page 129 note 3} Many stelai were destroyed, as the small fragments prove. Some may yet be restored, A certain number of dedications may be found in the monastery and elsewhere. Still those already found adequately represent the probable character of the whole series.

^{page 130 note 1} Biometrika, 1913, pp. 366–386: he proves that the expectation of life for young women was less, for elderly women greater, than for men. The reason is obvious. See Bearing of Research on N.T. p. 380.

^{page 131 note 1} See Anderson in J.R.S. 1913, p. 299 f.

^{page 132 note 1} It is different in the case of ‘Confessions.’ Their purpose was to serve as ‘exemplaria’ and warnings of the god's power and man's duty. They therefore recite the circumstances of their origin, the sin and the punishment and the expiation. We found only one ‘Confession’ near Antioch, and this not at the hieron, but in the ridge called the Snake (p. 113).

^{page 132 note 2} κατὰ ὅναρ, κατὰ τελετήν.

^{page 133 note 1} It may be taken for granted that ο and ω were interchangeable in late dedications. This is certain in the case of the important word Tekmor, which will be treated in the sequel; and the confusion of the two letters was widespread in the third century and later, and occurs even earlier. Moreover, Ἰόνιοϛ occurs often beside the more correct Ἰώνιοϛ, etc. Catullus speaks of the nuntius horribilis Ionios fluctus … iam non Ionios esse, sed Hionios.

^{page 133 note 2} Of these we possess several: one at Troketta in Lydia (best in Keil and Fremerstein, Reise I in Lydien, p. 8); one in North Phrygia (see my Studies in the Eastern Roman Provinces, p. 128).

^{page 133 note 3} See the art. ‘Mysteria’ in Daremberg, and Saglio, , Dict. des Antiquités, iii, p. 2142 AGoogle Scholar, note 6: also a chapter on the Relation of Paul to the Greek Mysteries and his contemptuous use of the word ἑμβατεύσας, in my Teaching of Paul in Terms of the Present Day. pp. 287–301. The inscriptions of Klaros are published in Oest. Jahresh. 1906 and 1912.

^{page 134 note 1} Treated in the writer's article in the Annual B.S.A. 1912, p. 46. Cp. p. 133, n. 3.

^{page 134 note 2} Naturally, expense to the initiated was caused, for dues were levied by the priests for performance.

^{page 135 note 1} And through the mystic entrance way, shown in the photographs, Annual, plate i, 1 and 4.

^{page 135 note 2} Examples need not ne quoted: κατά ὅναρ below in no. 237.

^{page 135 note 3} τέκμορ is a false from instead of τέκμωρ. It cannot be taken as abbreviation of τεκμορεύσας as it is once used with ποι(ἤ)σας. Moreover, there is no reasonable justification for abbreviating the word. The use of o for w is common in the late Phrygian inscriptions. The omission of a verb is common in dedicatory inscriptions; and in fact with εὐχήν the omission is almost invariable. This archaic and poetic word was re-vitalised by readers of epic poetry at school, and after some tentatives a verb τεκμορεύω was coined. On the meaning ποιεῖν used with Tekmor see below on p. 138.

^{page 135 note 4} The religion on these estates took the form of a cultus of Artemis and Mèn. Strabo, p. 557, mentions only the hieron of Mên in the region of (subject to) the Antiochian State. In the local inscriptions of the Tekmoreian Guest-friends Artemis takes first place, and Mên is not mentioned under that name, but as Zeus Eurudamenos, or Ourudamenos, or the more strictly Pisidian form Mannes or Manes Ourammoes (Orumaios). See J.H.S. 1918, loc. cit.

^{page 136 note 1} See note on p. 138.

^{page 136 note 2} Subsequent literature: Ramsay, J.H.S. 1912, pp. 151–170: Annual B.S.A. 1911–2, p. 63: M. M. Hardie (Mrs. Hasluck, loc. cit., p. 123).

^{page 136 note 3} See Annual, p. 63, but the name Tekmoreian does not occur. It may be restored in the mutilated beginning.

^{page 137 note 1} On this remarkable title, ‘he who reclines first at the Table,’ see Annual, loc. cit. p. 153.

^{page 137 note 2} See the growing list of homes (Sterrett, W.E. p. 271), complete to 1905, Stud. in E. Roman Prov. pp. 362 ff., later additions J.H.S. 1912, p. 169.

^{page 138 note 1} Sterrett in a Prelim. Report, followed by Ziebarth, Gr. Vereinwesen, p. 67. and Judeich, Alt. von Hierapolis, p. 120, regarded Tekmoreioi as a topographical term. Later Sterrett, W.E. p. 432. mentioned my interpretation, apparently preferring it. Pagan Revival in Pauline and other Studies, p. 103 ff.

^{page 138 note 2} ιερὰ ποιεῖν, or ῥέζειν, is perhaps a sufficient defence, for the Tekmor was a sacred object. Compare also θυσίαν ποιεῖν. This sacrifical or ritual use of ποιεῖν became commoner in the later period, and particularly among the Christians.

^{page 139 note 1} έπήκοος, a common epithet of the supreme Anatolian god or goddess.

^{page 140 note 1} I think it was his re-publication of Lebas's Monuments Figurés.

^{page 141 note 1} See Annual, loc. cit.

^{page 142 note 1} No. (3) is practically identical identical with no. (4).

^{page 142 note 2} These horseshoes are described in the Proceedings of the Society of Antiquaries of Scotland, 1914–15, p. 198, with figure 45, 9, and 1915–16, p. 121: see also Pitt Rivers, Excavations in Cranborne Chase, iv, plate 258, fig. 24.

^{page 144 note 1} The empty divine throne is common: see The Thousand and One Churches, p. 505 ff

^{page 145 note 1} δάος has been taken as the analogue of θώς. The word in the sense ‘torch’ occurs in the Tekmoreian lists : J.H.S. 1912, p. 163: also Hesych. l.c.