¹ I am indebted to Mr I. R. Rosin, a leading surgeon of Salisbury, who initially discussed some of the anatomical and physiological aspects of this article with me. Dr A. E. Strover, formerly Lecturer in Anatomy in the University College of Rhodesia and Drs P. Bunning and J. Firth, of the Royal Free Hospital Medical School, London, dealt with further difficulties I encountered in connexion with this and other investigations, demonstrated in cadavers the vessels and organs mentioned in this article, and made a number of helpful suggestions. I am most grateful to Dr Strover, as I am to the Librarian of the U.C.R. Library, Mr A. Harrison, who patiently obtained books and articles for me. Clearly, I alone am responsible for any absurdities which may persist.

² Tatius, Achilles, Leucippe and Clitophon iv 9.1–2.Google Scholar

³ iv 15.4. Mr Rosin has identified this from the symptoms as cantharides, which was known and used in the ancient world. I have not been able to discover a Greek or Latin parallel.

⁴ Vilborg, E., ‘Achilles Tatius, “Leucippe and Clitophon”. A Commentary’ in Studia Graeca et Latina Gothoburgensia xv (1962) 85Google Scholar writes, ‘It is hard to say whether this alludes to more than one person. Menelaus has some companions in the tent’. I feel more confident than Vilborg that means ‘those with Menelaus’. It is true that in the Greek of the second century A.D. the expression need mean no more than ‘Menelaus’. Tatius preserves the classical usage at iii 20.4, for example, where the sense must be ‘those with him’. λύσατε in 9.4 (to which Vilborg, ibid., draws attention) points the same way, odd though it sounds after ἐδεόμψ Μενελάον and the words τῶν πολλῶν ἀπηλλαγμένων can mean that some people remained, and may even imply this.

⁵ Achilles Tatius' historical setting is very vague, and one can only suppose that these bystanders together with their opinions must be appraised in terms of Tatius' contemporaries. For a discussion of the mise-en-scène, see Vilborg, op. cit. 9–10.

⁶ Tatius, ibid. 9.2.

⁷ In contrast to the bystanders and the doctor, Clitophon, in manner typical of himself and the hero of the Greek novel in general, ascribes Leucippe's madness to the δαίμων and Τύχη (iv 9.7).

⁸ These terms will be discussed below. Medical terms were used for similar, and frequently humorous effect, in Comedy. Cf. Miller, H. W., ‘Aristophanes and Medical Language’ in TAPA lxxvi (1945) 74 ff.Google Scholar For a brief discussion of Homer's (and his audiences') interest in precise details of surgery and anatomy, cf. E. T. Withington's introduction to Volume iii of the Loeb edition of Hippocrates, xi–xiii.

⁹ Vilborg, op. at. 10.1–2.

¹⁰ For similar phonetic effects in Tatius cf. e.g. iii 2.5 and T. F. Carney's note ad loc. in Achilles Tatius, Leucippe and Clitophon Book Three [Classical Association of Rhodesia and Nyasaland, now Classical Association of Central Africa, 1960] 42–3. For further examples cf. Carney's index s.v. anaphora, chiastic expression, hiatus, onomatopoeia and verse rhythms.

¹¹ iv 10.3–4.

¹² If, that is, Celsus was a doctor and not just a compiler and translator: cf. W. H. S. Jones' introduction to Celsus in the Loeb edition, viii ff.

¹³ Jones, ibid. ix: Celsus, , Prooemium 54 ff. and 74–5.Google Scholar

¹⁴ v 25.1.

¹⁵ v 25.5.

¹⁶ v 25.4 and 9.

¹⁷ v 25.6 and 16.

¹⁸ The pills Celsus lists were with one exception used internally. The exception (v 25.4), made from a decoction of wild poppy-heads and raisin wine, could be used inter alia mixed with water and smeared on the forehead to stop phlegm flowing into the eyes.

¹⁹ In the early Hippocratic Sacred Disease Loeb ch. xviii sleeplessness is attributed to the brain.

²⁰ Celsus, ii 14.1.

²¹ Cf. Celsus, iii 23.7 et passim.

²² Celsus iii 23.7. Cf. too 23.3; iii 18.6 and especially 12.

²³ iii 18.12.

²⁴ iii 18.7.

²⁵ Cf. e.g. iii 23.3.

²⁶ Cf. e.g. Dobson, , ‘Erasistratus’ in Proc. Roy. Soc. Med. xx (1927)Google Scholar Section of the History of Medicine, 27. Galen xi 226, etc. (volume and page references to C G. Kühn's edition).

²⁷ Cf. Celsus, , Prooemium 16.Google Scholar

²⁸ Asclepiades perinde esse dixit his sanguinem mitti, ac si trucidentur Celsus, iii 18.6.

²⁹ Celsus, iii 23.2.

³⁰ iii 18.4.

³¹ Vilborg, op. cit. 85.

³² Cf. for example Galen, xiv 728 for the divergent views of ‘Hippocrates’, Erasistratus and Asclepiades on the cause of disease. I shall discuss below points of detail which could possibly be deliberate mis applications of what one may perhaps call ‘technical’ terms. Cf. further Galen xix 342–4, etc.

³³ Fl. c. 260 B.C. The best general discussion of Erasistratus known to me is M. Wellmann's article in Pauly-Wissowa.

³⁴ The blood is commonly described as ‘boiling’ (cf. e.g. Galen, ii 287), and specifically in the left ventricle of the heart (Galen, v 573). Galen frequently describes blood boiling in the young (v 693, 703; xvi 57; xvii B 638; xix 374, etc.). At iv 810 he writes: and:

³⁵ Vilborg, op. cit. 9.

³⁶ It is possible that the term φλέβες is being used untechnically to mean simply ‘blood-vessels’, as is clearly the case in, for example, Quintus Smyrnaeus, xi 192 (to this passage I shall return below). Nowadays laymen often refer to the blood-vessels simply as ‘veins’. Nevertheless one might not expect this to be the case in even a humorous medical diagnosis, and as will be seen, the concept of blood displayed indicates that in fact specifically veins and not blood-vessels are meant.

³⁷ Cf. Galen, xi 175 et passim. To Erasistratus' views I shall shortly return.

³⁸ One of the reasons given by Galen for blood's corrupting when it becomes unnaturally compressed in any part of the body is τῷ μὴ διαπνεῖσθαι: cf. vii 375. Galen conceives of blood thickened by phlegm and other humours, or even of blood compressed and compacted arresting the flow of pneuma, but never of blood in its normal state doing so (cf. vi 260; viii 176 ff., 181–2; x 617, 640; xv 781; xvi 17 etc.). viii 176 ff. is particularly instructive. Melancholic humour is not, in Galen's view, to be identified with black bile, but can perhaps be termed μελαγχολικὸν αἷμα. It is οἷον τρὺξ αἵματος and, like phlegm, can in the brain cause epilepsy by interfering with the psychic pneuma: but it is not normal blood. Blood is commonly thought of as fervent (cf. n. 35), and this per se does not indicate an abnormal state of the blood in Tatius.

³⁹ Cf. the Loeb edition of this work ch. vi et passim.

⁴⁰ To my knowledge the latest discussion of this problem in the is contained in my thesis ‘Anatomy and physiology in the Hippocratic treatise submitted to London in 1966: cf. 116 ff.

⁴¹ Cf. n. 43.

⁴² Galen, iv 721 ff. et passim, Celsus, , Prooemium 60–1.Google ScholarCf. too Rufus Ephesius (ed. Daremberg and Ruelle) 183. Praxagoras, or possibly his father, a contemporary of Aristotle (cf. Steckerl, F., The Fragments of Praxagoras of Cos and his School [Leiden, 1958] 1, 17 et passim)Google Scholar, first postulated this remarkable dogma (which was soon attacked by his famous pupil Herophilus), and according to Steckerl its exponents are limited to Praxagoras, Erasistratus and the author of the pseudo-Aristotelian περὶ πνεύματος (op. cit. 24). Praxagoras, however, considered the heart to be ‘the centre of the intellect, of feeling, and of life in general’ (Steckerl, op. cit. 35–8 et passim). The view was also held by the Erasistrateans, and, Galen suggests (and as was certainly the case), by others unnamed (iii 492). Cf. further iv 664, where Galen says many were deceived, and especially philosophers, who did not understand what they saw in dissection. Erasistratus himself was a pupil of Chrysippus, who, however, believed that the heart was the centre of the intelligence (Galen, v 278).

Wilson, L. G. has argued (in Bulletin of the History of Medicine xxxiii, No. 4 [July-August, 1959] 296, n. 18)Google Scholar that Dobson, J. F. was wrong in supposing that Herophilus believed that the arteries contained blood (‘Herophilus of Alexandria’ in Proc. Roy. Soc. Med. Land. [1925] xviii Historical Section, 19–32).Google Scholar Dobson's view is shared by Steckerl (op. cit. 35) and is, I believe, correct. In the first place, though Galen frequently mentions that Erasistratus held this view (cf. xx 228, col. 2 for references), and indeed wrote the An sanguis in arteriis natura contineatur (iv 703 ff.) to refute it, he nowhere to my knowledge associates Herophilus with Erasistratus’ famous (or rather, notorious) dogma. This he would surely have done had it been believed in by so distinguished an anatomist as Herophilus (whom he reports inter alia as having postulated that the walls of arteries were six times as thick as those of veins, etc.: iii 445). Secondly, in the passage on which Dobson based his view that Herophilus believed that there was blood in the arteries (Galen, iv 731: cf. Wilson, ibid.), Galen is arguing that, although (as Erasistratus maintained) pneuma could not be sent over the body from the heart if there were blood in the arteries, in fact peuma is not pumped over the body by the heart through the arteries but is drawn into them πανταχόθεν (as was maintained, Galen tells us, by Herophilus, Praxagoras, Philotimus, Diocles, Plistonicos, Hippocrates and numerous others); which being so, blood in the arteries does not inhibit the passage of air (Wilson, who refutes Dobson on the evidence of this passage, has misunderstood the main purpose of Galen's argument). The same problem is viewed in terms of the pulse at v 168 where Galen admits that perhaps if the arteries were empty of blood (as Erasistratus believed) it might be possible, as Erasistratus said, for the arteries to beat at the same moment as the heart as a result of being filled with pneuma by the heart. In Herophilus' view the pulsation of the arteries (viii 703) derived from the heart but was a movement of the artery walls, not the result of pneuma being forced through. In Herophilus' view, therefore, blood in the arteries would not prevent their drawing pneuma into themselves. This explanation of the puke ascribed to Herophilus contradicts that attributed to him by Rufus (ed. Daremberg and Ruelle 220–1: Rufus may not have been the author of this work) where Herophilus is said to have believed that the pulse followed from the filling and emptying of the arteries (i.e. with pneuma). Filling and emptying of the arteries from whatever source does not, however, imply per se that there can be no blood in them: one may compare ‘Rufus'’ view (op. cit. 183) that the arteries contain some blood and much more pneuma.

Praxagoras should not, perhaps, have been included in Galen's list at iv 731. Praxagoras, however, postulated arteries empty of blood (to make possible the rapid movement of pneuma) for a different reason from Erasistratus: for Praxagoras the arteries at their terminations became nerves and transmitted from the heart the pneuma which produced voluntary movement, which sometimes clearly would be instantaneous. I have argued above that Galen's ‘Hippocrates’ held that blood and pneuma were distributed through all the blood vessek in common (termed simply φλέβες). Galen makes precisely this point at iv 671–2.

Wilson appears not to realise the extent to which Erasistratus' dogma (that the arteries normally contained only pneuma) had been attacked before Galen came upon the scene.

⁴³ Cf. e.g. xiv 728.

⁴⁴ Cf. Galen, e.g. iii 493.

⁴⁵ Cf. Galen, e.g. vii 537 ff. For a fuller discussion cf. Wellmann's article s.v. Erasistratus in Pauly-Wissowa, especially cols. 344 ff., and J. F. Dobson, op. cit.

⁴⁶ Galen, for example, quotes him as mentioning specifically (vii 539).

⁴⁷ Cf. Galen, iii 493. To this I shall return.

⁴⁸ Cf. Wellmann, op. cit. cols. 343–4; Dobson, op. cit. 825–32; Wilson, op. cit. 310.

⁴⁹ Cf. v 602 ff. and especially 604.

⁵⁰ Cf. xix 315.

⁵¹ Cf. v 602 ff.; xviii A 86.

⁵² v 609: this Galen denied, saying in v 609–10 that Erasistratus was wrong to suppose that, were the dura damaged, a beast would collapse; and that this would not happen unless the ventricles were pierced (or indeed pressed, v 185): cf. too iv 501 ff.; v 604 ff.

⁵³ Dobson does not discuss this problem, noting merely that the manner in which zotic pneuma became psychic pneuma is obscure (op. cit. 24). Wellmann (op. cit. col. 343) gives two references to Galen (v 185 and viii 760) in support of his statement that the ventricles of the brain were filled with psychic pneuma in Erasistratus' physiology. In each case, however, Galen in discussing Erasistratus appears to be referring to the left ventricle of the heart.

It seems to me virtually certain that Erasistratus from the start knew something of the ventricles of the brain. Quite possibly, however, he did not originally attribute great importance to them. Galen tells us that the followers of Hippocrates believed that air inhaled through the nose passed directly to the brain's ventricles (as compared with the followers of Erasistratus who believed that air reached the brain via the heart and arteries—iv 502: to this I shall return).

It is even possible that the brain's ventricles (though perhaps not all four) were known to the author of the early who writes (x 21–3, Loeb ed.): Inter alia the words were deleted by Wellmann, (A.G.M. 1929, 293)Google Scholar n. 3) and by von Wilamowitz-Moellendorff (Sitzungsber. d. Ber. Ak. 1901, 7–8), and the possibility was not considered by Wellmann, despite the fact that he erroneously believed the writer had observed the nerves (ibid. 290 et passim). The writer did know something of the nature of the brain; e.g. that the cerebrum is ‘divided’ by a thin membrane (Loeb, vi 4–6). Galen, who certainly ascribed the work to Hippocrates (cf. e.g. xviii B 18: he likewise attributed the to Hippocrates; cf. iv 798), may well have thought that he did, for, as I noted above, he tells us that the followers of Hippocrates held that air was inhaled through the nose direct to the ventricles of the brain (iv 502, 504), and the writer of the insists that air on being inhaled passes first to the brain (Loeb, x 13 ff.; xix 10 ff.).

⁵⁴ Galen, iv 502. Galen informs us that Erasistratus' followers disagreed amongst themselves (viii 759) and were generally ignorant of Erasistratus' doctrines (xi 175).

⁵⁵ v 155.

⁵⁶ Galen describes the plexus reticularis in great detail at iii 696 ff. (cf. too 305; iv 323; v 155 and v 607 ff.), as inter alia lying between the cranium and the dura under the vasis cerebri. I am greatly indebted to Dr Malcolm Ross of Harari Hospital, Salisbury, who skilfully demonstrated this region to me in three fresh cadavers. Nothing exists corres ponding to what Galen describes. The region occupied by the δικτυοειδὲς πλέγμα appears to correspond to the cavernus sinus which lies below the dura immediately below the brain. In it the pituitary gland is medial: this may be the gland mentioned in Galen as being surrounded by the rete mirabile (iii 696). The cavernus sinus is traversed by the carotid arteries, which all but double back on themselves in the process. This may have prompted Galen's remarks (cf. 698) to the effect that the (carotid) arteries surprisingly do not immediately penetrate the dura (after passing so easily through the cranium), and then that almost equally sur prisingly after forming the plexus they do penetrate it. These are fibrous strands in the cavernus sinus which might possibly have suggested a very fine structure resembling a three dimensional net, but autopsy does nothing to strengthen the possibility. It must further be remembered that the cavernus sinus is normally filled with venous blood, which drains from it to the jugular veins. Lateral to each carotid artery on a line slightly behind the pituitary gland is the gasserian ganglion, which lies under the dura and above the periostium covering the temporal lobe. This, though nerve structure, may have suggested the plexus reliformis: it does have the appearance of a very fine three-dimensional net and receives filaments (which I did not detect) from the carotid plexus of the sympathetic. At v 607 ff. the plexus is described as being inside the dura. Nevertheless autopsy indicates it cannot be identified from Galen's description with the Circle of Willis and its supporting arachnoid.

⁵⁷ Galen well knows the protective function of the dura and pia, and mentions the fluid, but does not seem to have appreciated its purpose: cf. iii 641, 706 ff., 656 ff.; x 929, etc.

⁵⁸ Loeb, xiv 20–3: xvi 38–40: Other interesting attestations of the verb in Galen are indicated in Thes., s.v.

⁵⁹ Cranial haemorrhage results in blood being found in these regions; inexpert dissection would produce the same result.

⁶⁰ It is perhaps less likely in view of the very reduced space in which blood could be said περικλνζειν; and also, perhaps, because initially at least Erasistratus believed that ἡγεμονικόν resided in the dura mater. (Cf. n. 50).

⁶¹ Cf. v 281; Wilson, op. cit. 310 etc.; for Galen's view of the effect of thick, cold phlegm on pneuma in the ventricles (possibly shared by Erasistratus) cf. viii 174; xiv 737; xvii B 548.

⁶² Cf. iii 701; v 606–7; xv 263; cf. too x 605 ff.

⁶³ Cf. v 183.

⁶⁴ Cf. n. 55, and Galen, v 155: The term was accepted by Galen; cf. iii 305, etc.

⁶⁵ The statement in Tatius that it is the overflowing of the veins which inhibits the brain's ἀναπνοή strongly suggests that the author considered that it is (as Erasistratus believed) air from the arteries which is smothered. It is, however, clear that an Erasistratean view of blood is expressed even if in the view of Tatius' source the air (or even most of the air) reaching the brain came direct from the nose. This ‘Hippocratean’ view seems to have been revived by Galen, who supposed that smells were sensed in the two front ventricles of the brain and adduced various proofs of this view: cf. iii 646 ff.; ii 868 ff. Galen reconciles this source of pneuma for the brain with that provided by the plexus reticularis at x 839.

⁶⁶ ἐκ πηγῆς: cf. Dio Chr. xii 70 quoted in LSJ.

⁶⁷ Cf. v 324; quoted, as are the two following examples, from LSJ. One reads, ibid.

⁶⁸ xi 192: Here subject and object are reversed.

⁶⁹ Ep. 23.

⁷⁰

⁷¹ Loeb, xviii 21 ff. The resulting behaviour (experienced in sleep) ends (ibid. 30–1).

⁷² Galen, v 183; vii 542.

⁷³ vii 202.

⁷⁴ x 949.

⁷⁵ x 461.

⁷⁶ iii 492 ff., vii 542.

⁷⁷ Cf. iii 494 for the purpose Galen attributed to them.

⁷⁸ Cf. especially xi 187 ff. (Galeni de venae sections adversas Erasistrateos Romae degentes) and 250 ff. (Galeni de curandi ratione per venae sectione); vi 295, etc.

⁷⁹ Loeb, iv 10.4 and 6.

⁸⁰ Loeb, iv 9.2.

⁸¹ Loeb, iv 9.4.

⁸² The passages relating to doctors in the Greek Romance are collected by Calderini, , Cantone, Proleg. (Torino, 1913) 92.Google Scholar On a probable example of satire of the ‘ton doctoral qu'affectent parfois les practiciens’ in Heliodorus cf. Rattenbury-Lumb-Maillon, ii 12, n. 1.