^{page 97 note 1} To compile even a select bibliography of writings on this topic for the last hundred years would take much more time and trouble than I am prepared to spend on it. The subject is beyond the scope of the standard bibliographical publications: it is hard to discover what has been written; and then it is often still harder to obtain the books. It was by mere chance that I found one of the two treatises which proved most useful—Die Krankheit zu Athen nach Thucy-dides, by Dr. H. Brandeis, Kais.-russ. Hof-rath, a pamphlet published at Stuttgart in 1845; it is not mentioned by any other work which I have seen on this subject. Gleanings from the last forty years of Bursian are small and generally unfruitful. Schmid-Staehlin, 1. v, p. 75, n. 3, refers to two useful works: Hagen, B. von, ‘Die sogenannte Pest des Thuk.’, Gymnasium, xlix, 1938, pp. 120 ff.Google Scholar (I am obliged to the University Librarian at Cambridge for providing me with microfilms of this elusive article); and Ehlert, J., de verborum copia Thuc., diss. Berlin, 1910.Google Scholar Classen-Steup mention only the agnostic Ebstein, W., Die Pest des Thuk., Stuttgart, 1899Google Scholar, and ‘Nochmals die Pest des Thuk.’, Deutsche Mediz. Wochenschr. xxxvi, 1899, pp. 594 ff.Google Scholar Valuable notes and comments, such as those of Finley, , Thucydides, 1942, p. 158Google Scholar, n. 2, and SirAllbutt, Clifford, Greek Medicine in Rome, 1921, pp. 340 f.Google Scholar, are to be found in numerous places, likely and unlikely. Useful introductions to the medical literature are provided most recently by Shrewsbury, J. F. D., Bulletin of the History of Medicine, xxiv, 1950, pp. 1 ff.Google Scholar (mostly British and American), and by Hagen, B. von, op. cit. (mostly German). I have read a great deal: but I expect and hope that my attention will be drawn to serious omissions.Google Scholar

^{page 98 note 1} Nestlé, W. in Hermes, Ixxiii, 1938, pp. 28 ff.Google Scholar, gives some Hippocratic examples of a few Thucydidean terms; such obiter dicta on this difficult subject are misleading, and Ehlert had already rendered them super-fluous.

^{page 98 note 2} Cochrane, C. N., Thucydides and the Science of History, 1929.Google Scholar

^{page 99 note 1} Thuc. 2. 48. 3. Cf. Epid.¹ 11 (i. 164 J., i. 189–90 K.)

^{page 99 note 2} I suppose that Thuc. refers especially to the numerous shades of colour named by the doctors in this connexion: cf. Prog. 13.4 (ii. 28 J., i, p. 91 K.)

^{page 99 note 3} For the purpose of what follows, I have admitted evidence from schools other than the Hippocratic, but seldom unless there ap peared to be no reason to doubt that the terminology in question was more or less uniform. I have further admitted the evidence of treatises written probably in the fourth century B.C., on the ground that a high proportion of the terms standard in that era were probably established in medical parlance long before. The dating of the treatises opens a wide field for research: differences and resemblances in thought and style between one work and another are often obvious to the most casual inspection; and very different levels of medical science are represented. Perusal of Gossen, RE viii. 1802 ff., and Edelstein, RE suppl. vi. 1290 ff., suggests that a great deal remains to be done. The confident dating of a large number of the treatises to the second half of the fifth century B.C. surprises me: but I see no reason to dispute it in some cases (esp. Prog., V.M., Epid. ^1.3, Aër., Acut., and a few others; of these I have made most use), or to doubt that the majority of the remainder were composed before the end of the fourth century.

^{page 99 note 4} In the sequel, an asterisk signifies that the word occurs nowhere else in Thuc, a dagger that it does not recur with the same meaning. In quoting from the Hippocratica, I have thought to serve die reader's con venience by adopting the following tedious procedure: The excellent text of Dr. W. H. S. Jones in the Loeb Series is quoted first (by chapter and line, followed by number of volume and page+‘J.’) for all treatises in cluded in it. If these treatises are found also in the Teubner text of H. Kuehlewein (vol. i, 1894; vol. ii, 1902) or in the Corpus Medicorum Graecorum (1. i, ed. I. L. Heiberg, 1927), references to these works (by volume and page+‘K.’ for Kuehlewein, by page+‘H.’ for Heiberg) are added to the Jones-references. (The only treatises in K. and H. which are not in Jones are Liqu. and Medic. 2-end, both in H. only.) Treatises which are not included in Jones's edition are quoted (by chapter, followed by volume and page+‘L.’) from the great work of Littre (Paris, 1839–61). Abbreviated treatise-titles conform throughout to the list in LSJ.

^{page 99 note 5} Loc. Hom. 45 (vi. 340 L.): is much commoner in the doctors.

^{page 99 note 6} Thuc. uses in 2. 11. 7; is much commoner than throughout the doctors.

^{page 101 note 1} DrJones, , Malaria, pp. 21 ff., points out that Thuc. does not use the common word for a fever, : there is great probability in his inference that since ‘in popular speech… there is a tendency to limit to definite fevers, namely, to those exhibiting a certain periodicity’, Thuc. deliberately excludes this term' from his description of a wholly unfamiliar disease.Google Scholar

^{page 101 note 2} Brandeis aptly compares Dion. Hal. arch. 10. 53 (on the plague at Rome, written with many conscious imitations of Thuc.)

^{page 102 note 1} So Brandeis, , op. cit.Google Scholar

^{page 102 note 2} The truth is not to be found in either LSJ or the Dindorfs' Lexicon. Nor yet in Galen (xviii. i. 122 Kühn), who defines (stools) thus: this definition is much too narrow for many places in the Hippo-cratica (e.g. in Epid. ^1.3) where is regularly added to where appropriate.

^{page 102 note 3} In what follows I have consistently used ‘dysentery’ in its Greek sense, referring to stools of blood, mucus, pus, and the like.

^{page 108 note 1} Cochrane, , op. cit., p. 17. Ehlert observes that in 29 out of 43 examples of the singular number in the doctors, occurs in the genitive case governed by a preposition, as in Thuc.Google Scholar

^{page 109 note 1} I ought perhaps to have said something somewhere about Lucretius' imitation of Thuc. in 6. 1138 ff. The position is apparently quite simple: from 1138 to 1181 and again from 1197 onwards Lucr. follows Thuc. closely, with a few additions and embellishments (1150, 1202–3) and one or two mistakes (esp. 1209 ferro: I am not convinced by Maas's explanation in Bailey's, Lucretius, p. 1758). In the midst of all this he interpolates a passage (1182–96) based on well-known Hippocratic sources which have nothing whatever to do with the Atiienian Plague. It is an extraordinary procedure for a scientific writer; but the only point of importance at present is that there is no reason to believe that Lucr. knew anything about the Plague beyond what he found in Thuc, or that he read Thuc. in any other form than what we possess today.Google Scholar

^{page 111 note 1} Those who try to identify Thuc.'s Plague with a disease of which some sort of mental disorder is characteristic either fail to notice that this feature is wanting in his account or adopt the doubtful expedient of arguing that the patients must have been mentally deranged or they would not have thrown themselves into wells; as if Thuc. had not explicitly given an entirely different (and sufficient) reason for this action.

^{page 112 note 1} Among the defects alleged by modern medical writers the only one of any importance, which must be acknowledged, is the inadequate description of the exanthem. At what stage did it first appear? Did both the and the exist side by side, or did the former develop into the latter ? What size and shape were they? How long did they last ? What was the process of the disappearance in cases of survival—did they peel, or flake, or what? Other charges of inadequacy are less appropriate: (i) Thuc. does not mention the pulse: true, but the significance of the pulse in relation to health was not, so far as we know, appreciated by the doctors until after the lifetime of Thuc. (ii) He does not refer to the condition of the urine: but that may be because there was nothing significant to record; I notice that standard modern accounts of the disease with which we shall shortly identify Thuc.'s Plague include no reference to the urine. (iii) Thuc. gives too little detail about the development and duration of individual symptoms, and does not distinguish systematically enough between the various stages in the progress of the Plague: I think it a fair comment that descriptions of such diseases in modern medical textbooks are not much superior in these respects, (iv) Brandeis (p. 62) complains that Thuc. does not distinguish between invariable and occasional phenomena: this is plainly unjust; Thuc. states explicitly (51. 1) that he describes the invariable phenomena, omitting individual deviations from the norm.

^{page 112 note 2} At least we must continue to try until failure is proven; which is not yet. And ultimate failure need not mean that Thuc.'s description is at fault, for (1) his Plague might be a disease now apparently extinct, like the English ‘sweating-sickness’, 1485–1552, ‘suette des Picards’, 1718–1870; (2) there is no proof that the characteristics of a disease remain sufficiently constant over so long a period of time.

^{page 114 note 1} Hagen, B.Von, op. cit.Google Scholar, is the most recent pleader for smallpox. He admits, but makes no attempt to answer, the objection stated under (iii) above (he scrutinized the Naples bust of Thuc. for scars, but it gab keinen Anhaltspunkt). He admits further that gangrene is incompatible with the smallpox-theory, and suggests that this complication was introduced by a concurrent outbreak of a second plague, typhus exanthematicus; the same notion, that Thuc. has confused a plurality of simultaneous plagues, had already been expressed by Sticker, G., Festschr.für B. Nocht, 1937, p. 604 (quoted by von Hagen; I have not seen it). He does not discuss objections (i) and (ii).Google Scholar

^{page 114 note 2} I have seen it stated that it is not quite certain that the rat is the sole permanent reservoir of epidemic typhus, and that the body-louse (which was thought to convey from man to man an infection derived by man from the rat) may itself be the host. But then we should have to make the very improbable assumption that the Athenians had already in the spring of 430 B.C. sunk to such a state of filth that die disease might be generated and the infection universally transmitted in this way. The city had indeed for some months been crowded by the abnormal influx of residents from the country: but the Athenians were not a dirty people, and there is no other indication that a decent standard of cleanliness and sanitation was not maintained.

^{page 115 note 1} There is apparently no doubt about the existence of the rat in Italy in the first century A.D. See the evidence assembled by SirMacArthur, W. P. in Transactions of the Royal Society of Tropical Medicine and Hygiene, xlvi, 1952, pp. 209 ff.CrossRefGoogle Scholar, with references on p. 212 to modern treatments of the subject. I am very much obliged to the author for sending me a copy of this paper, and also of another, ibid. p. 464, where it is reported that the skeleton of a rat, indistinguishable from Rattus Rattus, has been found by Prof. Haas in a neolithic site on Mt. Carmel, another (of unidentified species) in a palaeolithic site in the desert of Judaea. I am indebted to Sir William also for a description (which I have been careful not to go beyond) of the symptoms of typhus fever in relation to Thuc.'s Plague. The disagreement between us on the main issue here remains unfortunately absolute; but none of my numerous correspondents has helped me nearly so much.

^{page 115 note 2} Hagen, B.Von quotes Schröder, Mü. Med. Wochenschr. 1916, as a supporter of pneumonic plague. The discrepancies seem to me so numerous and large that I have not thought it worth while to pursue the matter farther here.Google Scholar

^{page 116 note 1} Since this paper was written I have seen in typescript an article by Mr. P. Salway and Miss W. Dell, arguing the case for ergotism. I had rejected this possibility for the reasons given by Finley in his Thucydides, p. 158, n. 2Google Scholar(compare esp. Kobert, R., Zur Gesch. des Mutterkorns, 1889, pp. 1 ff.Google Scholar, with the objections of Ebstein, W., Deutsche Med. Wochenschr. xxxvi, 1899, pp. 594 ff.CrossRefGoogle Scholar The resemblance between the Athenian Plague and ergotism is in many respects most striking: the apparently insuperable objection was that it would be necessary to prove that rye was used in the making of food, yet we know, so surely as such things can be known, that ‘rye was not used for bread in the Mediterranean region throughout antiquity’ (Finley, I.c., with authorities). I learn from the above-mentioned article (due for publication soon, I hope) that this objection is not founded on fact-that claviceps purpurea may attack other grains (including wheat) as well as rye. There remain, however, at least two further obstacles: (1) we should have to suppose that Thuc. was mistaken in thinking that the Plague was infectious: a very bad blunder, if it was one; (2) delirium and similar mental disturbances are said to be characteristic of ergotism. I say no more about it at present, in the expectation that Mr. Salway and Miss Dell will throw new light on these and other points.

SirAllbutt, Clifford, Greek Medicine in Rome, pp. 340 f., inclines to favour scarlatina maligna. Again, there is much general similarity in the symptoms, but again the discrepancies are numerous and important (sc. mal. is normally accompanied by prostration and delirium in the early stages, and in fatal cases death normally ensues within 48 hours, or, at least, long before the ‘seventh or ninth day’ of Thuc.; moreover, I cannot find that gangrene is a complication of this disease. Brandeis, p. 24, absolutely rejects the possibility of this identification).Google Scholar

^{page 117 note 1} Corney, B. C., Trans. Epidem. Soc., London, n.s. iii, 1884, pp. 76 ff.Google Scholar

^{page 117 note 2} Thurston, J. B., Report to the Governor; for this I have depended wholly on Shrewsbury.Google Scholar

^{page 117 note 3} Esp. Layard, H. L., Missionary Notices, xxi, Methodist Mission House, London, 1875–1877.Google Scholar

^{page 118 note 1} These quotations are directly borrowed from Shrewsbury's article, though I have been able through the kindness of Dr. Williamson to read in full the sources of most Dr.

^{page 119 note 1} I should add, fifthly, a question raised and answered by Dr. W. H. S. Jones: if the Plague was measles, it should have become endemic; yet there is no later reference to measles in Greek (or Roman) medical (or other) writers. The strength of this negative argument is broken by the parallel example of mumps, described at Thasos in the fifth century B.C. but ignored by all subsequent Greek medical writers (though there may be a reference in Celsus).

^{page 119 note 2} The foregoing is a revised version of a paper read to the Philological Society at Cambridge and to the Classical Association at Oxford in 1952. I have done my best to improve it in the light of the considerable correspondence which followed those occasions. Medical opinions, in which of course I was most interested, were fairly evenly divided for and against. The measure of agreement was such as to encourage me, perhaps against my better judgement, to publish this; the expressions of disagreement were such as to lead me to expect no mercy for having done so.