¹ In his Preface to Slavery Professor Buckland remarked that he knew of no treatise expounding the principles of the Roman law of slavery as a whole; and that most of the literature he used for questions of detail was foreign. It may now be stated, after more than thirty years, that his book on the subject has not been replaced either in England or elsewhere, and is not likely to be replaced for a long time to come.

² In addition, Professor Buckland completed and edited the second volume of Monro's translation of The Digest of Justinian (1909), and corrected and saw through the press E. C. Clark'e History of Roman Private Law, Part III, on the Regal Period (1919).

³ This is a list of his articles which one hopes will soon be incomplete. Book reviews and memorial notices are not included. (1) Difficulties of abstract jurisprudence, Law Quarterly Review 1890. (2) Wardour Street Roman Law, L. Q. E. 1901. (3) Manumissio vindicta par un fils de famille, Nouvelle Revue Historique de Droit Français et Étranger 1903. (4) Transfert de la propriété par un esclave, N. R. H. 1908. (5) Les manumissions prétoriennes, N. R. H. 1908. (6) More Wardour Street Roman Law : The actio de in rem verso, L. Q. R. 1915. (7) The Comitial Will, L. Q. R. 1916. (8) L'interêt dans l'actio furti en droit classique, N. R. H. 1917. (9) Mancipatio by a slave, L. Q. E. 1918. (10) Institutio d'un postumus alienus, N. R. H. 1920. (11) Did Ulpian use Gaius? L. Q. R. 1922. (12) Cretio and connected topics, Tijdschrift voor Rechtsgeschiedenis 1922. (13) F. W. Maitland, Cambridge Law Journal 1923. (14) Libertus (G. 3. 56), Revue Historique de Droit Français et Étranger 1923. (15) Interpolations in the Digest, Yale Law Journal 1923–24. (16) Gaius and the liber singularis regularum, L. Q. R. 1924. (17) Le Constitut possessore—animus et corpus, R. H. 1925. (18) Aestimatum, Mélanges Cornil 1926, and L. Q. R. 1927. (19) Usufruct in classical law, L. Q. R. 1927. (20) Actio damni iniuriae, R. H. 1927. (21) Servitudes in Roman Law, L. Q. R. 1928. (22) Les limites de l'obligation du fideiussor, R. H. 1928. (23) D. 44. 2. 21. 3 and res iudicata, Mélanges Fournier 1929. (24) The protection of servitudes, L. Q. R. 1930. (25) D. 47. 2 and the methods of the compilers, Tijdschrift 1930. (26) Diligens paterfamilias, Studi Bonfante 1930. (27) Aestimatum, L. Q. R. 1932. (28) Culpa and bona fides in the actio ex empto, L. Q. R. 1932. (29) Exceptio rei residuae (G. 4. 122), R. H. 1932. (30) Marcian, Studi Riccobono 1932. (31) Les limites de l'obligation du fidejussor, R. H. 1933. (32) Casus and Frustration in Roman and Common Law, Harvard Law Review 1933. (33) L'Edictum Provinciale, R. H. 1934. (34) The duty to take care, L. Q. R. 1935. (35) Finium Regundorum, R. H. 1936. (36) Classical Roman Law, ch. xxi in vol. xi (The Imperial Peace) of The Cambridge Ancient History, 1936. (37) Reflections suggested by the new fragments of Gaius, Juridical Review 1936. (38) Gaius and the liber singularis again, L. Q. R. 1937. (39) Civil proceeding against Ex-Magistrates in the Eepublic, Journal of Roman Studies 1937. (40) Note on the vocabulary of Gaius, L. Q. R. 1938. (41) D. 12. 4. 16, Tijdschrift 1938. (42) Praetor and Chancellor, Tulane Law Journal 1939. (43) Ritual acts and words in Eoman law, Festschrift Koschaker 1939.

⁴ See Maitland, , Bracton, and Azo, , Selden Society, vol. viii, Introduction, p. xxxiii.Google Scholar

⁵ Pour at least of his books must here be cited. The comparative lawyer appears in two which have already been mentioned. Equity in Roman Law (1911), and Roman Law and Common Law (1936) by Buckland and McNair; the Medievalist in Glossators of the Roman Law (1938) by H. Kantorowicz in collaboration with Buckland; an'd the modern lawyer in the second edition of G. B. Pinch's Selection of Cases on the English Law of Contract (1896) by R. T. Wright and Buckland.

⁶ See the chapter on Delicts by Slaves in The Roman Law of Slavery, pp. 98et seq. As the only part of the subject which, for reasons stated above, the Professor has not exhausted is the early, conjectural period, the following pages must mainly be devoted to pre-classical law.Google Scholar

⁷ See Mommsen, , Römisches Strafrecht, pp. 8et seq.Google Scholar and Lenel, , Actiones noxales, Zeitschrift der Savigny-Stiftung 1927, p. 26.Google Scholar

⁸ See De Visscher, , Vindicta et noxa, Études de Droit Romain, pp. 111et seq.Google Scholar; Les origines de l'obligation ex delicto, Etudes, pp. 285et seq.Google Scholar; and L'action noxale d'iniures, Études, pp. 329et seq.Google Scholar

⁹ I shall not, in this article, distinguish between noxa and noxia.

¹⁰ See Vanicek, A., Griechisch-Lateinisches Etymologisches Wörterbuch, vol. i, pp. 423et seq.Google Scholar

¹¹ Lewis, and Short, , 1927.Google Scholar

¹² De Agri Cultura 92.Google Scholar

¹³ De Be Bustica 1. 6. 15.Google Scholar

¹⁴ De Officiis 3. 5. 23Google Scholar

¹⁵ E.g., VIII 14 of the XII Tables.

¹⁶ XII 2. In D. 9. 4. 2. 1 we are told by Ulpian what Celsus thinks about the case where servus sciente domino furtum fecit vel aliam noxam commisit. The clause furtum … commisit is evidently modelled after XII 2, and yet furtum is represented as one of several possibilities of noxa. This divergence seems to result from the fact that, with Celsus and Ulpian, the emphasis lies not on the concrete acts performed by the slave, but on the attitude of his owner : the owner consents that his slave commits a delict, a noxa, be it furtum or any alia noxa. Possibly it is more than accidental in this connexion that the vivid expression of XII 2, noxiam nocere, to damage a damage, is replaced in 9. 4. 2. 1 by the general noxam committere.

¹⁷ Op. cit. pp. 284et seq.Google Scholar

¹⁸ De Visscher, , op. cit. p. 286, n. 3, takes a different view; he says that an impubes is not necessarily alieni iuris. But it is highly probable, I should think, that the XII Tables, when speaking of an impubes, mean the normal case where he does have a father. The point does not affect the main line of my argument.Google Scholar

¹⁹ VIII 24.

²⁰ Fortuita incendia quae evadunt, Coll. 12. 2. 2 and Paul 5. 20. 3. Si stipulam incenderis ignisque evagatus ad praedium pervenerit et illud exusserit, Coll. 12. 7. 4. Si stipulam incendentis ignis effugerit (compare the above cited si telum, manu fugit) in factum, agendum, quia non principaliter hie exussit sed, dum aliud egit, sic ignis processit, Coll. 12. 7. 5.

²¹ I cannot here justify this assumption at length : suffice it to advance some points. (1) Gaius in D. 47. 9. 9 no doubt gives a highly modernized version of the old rules. (2) The field case is again and again discussed in legal writings : it is an ancient, traditional problem. (3) The field case occurs as the typical example of accidental fire in other ancient systems. See, on Germanic law, v. Amira, Nordgermanisches Obligationenrecht, vol. i, pp. 388et seq. and vol. ii, pp. 408 and 866; and as for Hebrew law, Exodus 22. 5. (4) In classical texts, a second type of accidental fire is frequent : the fire in an insula. It is dealt with on much more modern lines. The chief reason is that it did not become important till Borne was a real city.Google Scholar

²² Comp. e.g. Coll. 12. 7. 4 et seq. partly quoted in note 20.

²³ Naturalis Historia 2. 107 (111). 239.Google Scholar

²⁴ Comp. above, p. 31.

²⁵ Some very special cases only are mentioned, the nocturnal cutting of another's crops, the cutting down of another's tree, etc.

²⁶ Modern codes usually try to be complete : they are made by conscientious scholars; the principle nulla poena sine lege is transplanted into private law; and perfect systems are popular in a rationalistic age.

²⁷ 21. 35 et seq., 22. 4.

²⁸ 22. 5. It is the field case : see above, p. 31, n. 22.Google Scholar

²⁹ 21. 33 et seq.

³⁰ Some sort of liability must have existed, seeing that liability existed even where the damage was done by son, slave or quadruped. That restitution was customary rather than vengeance may be inferred from the fact that even in damage to a person's body retaliation was no longer predominant.

³¹ See Lenel, Edietum Perpetuum, 3rd ed. p. 328.

³² G. 3. 190 et seq.; Gellius, Noctes Atticae 11. 18. 15.

³³ The words may well have occurred in furtum by the paterfamilias. But I shall disregard this possibility, which would form too insecure a basis for the following argument.

³⁴ The Aquilian theory of remote causality did not penetrate into this field. As, on the one hand, sons and slaves had nothing wherewith to pay, and, on the other hand, they committed the same delicts as a paterfamilias, the system of the absolute duty of the latter (i.e. duty existing no matter whether or not he could be regarded as a cause of the wrong in question) either to make amends or to surrender the culprit was by far the best.

³⁵ The provision relating to pastus pecoris (VIII 7), which may also have contained the expression noxiam sarcire, must, I believe, here be added. In this case, as distinct from pauperies, the animal acts under the control of man rather than as a self-determining being.

³⁶ Moreover, noxae deditio was never possible in these cases. That there was, however, some sort of noxae deditio in ancient systems if a weapon had done damage, is well known.

³⁷ On the question whether paupeties gave rise to a noxal action in classical time, see Lenel, , Actiones noxales, Zeitschrift der Savigny-Stiftung, 1927, pp. 2et seq.Google Scholar

³⁸ Livy 3. 55. 5 and 7. It is true that Livy may be inaccurate. But there is no reason to trust him less than those writers from whom we learn about noxa in the XII Tables.

³⁹ See below, p. 36.Google Scholar

⁴⁰ Op. cit. p. 9.Google Scholar

⁴¹ Caligula, 27. 1 and Nero 12. 1. But the word seems to have the wide meaning of ‘criminal’ in Claudius 34. 1 and Vitellius 17. 1.

⁴² Mercator 724. An example of obnoxius is to be found in Truculentus 834. Diniarchus confesses himself obnoxius, whereupon Callicles proposes to go to law : evidently, Diniarchus is thought of as sui iuris.

⁴³ Magistrates nec oboedientem et innoxium (see Mommsen, , op. cit. p. 38, n. 1; others read noxium) civem multa vinculis verberibusve coherceto.Google Scholar

⁴⁴ 39. 41. 7.

⁴⁵ The constant interpretation by praetor and jurists of noxiam nocere as ‘actually to commit a delict’ may conceivably be one of the reasons why even in classical law a paterfamilias was noxally liable only for the real wrongdoing of a person in potestas (see Buckland, Slavery, p. 101); but not, e.g., where a person in potestas occupied a house from which something was thrown.

⁴⁶ See p. 28.

⁴⁷ The question what was the exact relation between these alternatives need not here be considered.

⁴⁸ The oldest case recorded (320 B.C.) is in Iiivy 9. 10. 9. The leaders of the Roman army had made an undesirable treaty with the Samnites. It was argued by the Romans that the treaty, concluded against the will of the people, was not absolutely binding, and that the Samnites must be content with having the leaders surrendered. These were offered by the fetial with the words : Quandoque hisce homines iniussu populi Romani Quiritium foedus ictum iri spopon-derunt etque ob eam rem noxam nocuerunt, ob eam rem quo populus Romanus scelere impio sit solutus, hosee homines vobis dedo. It would seem that the formula, at that time, was not yet a genuine common piece of international custom; that it was made up by the pontiffs ad hoc, an artificial imitation of less solemn acts of private law; and that it was a cheat. But, as is well known, the precedent was followed on several subsequent occasions.

⁴⁹ Even metaphorical applications are to be found, e.g. Livy 26. 29. 4 (the Sicilians do not want to be ‘noxally surrendered’ to the terrible Marcellus) and Columella 1 praef. (rem rusticam pessimo cuique servorum, velut carnifici noxae dedimus).

⁵⁰ The adjective occurs only in post-Augustean texts, and only in juridical Latin.

¹ E.g. D. 39. 3. 1 pr. and 1. The case was not absorbed by the lex Aquilia, and the term damnum, therefore, was only occasionally used.

² See p. 45.

³ In some texts a distinction is made between furtum on the one hand and noxa, as denoting the remaining delicts, on the other hand.

⁴ The lex de Gallia Cisalpina also must here be adduced, if De Visscher's interpretation is correct : comp. above, p. 29.

⁵ 23. 14. 3.

⁶ De Legibue 3. 4. 11.

⁷ D. 50. 16. 131, ad legem Juliam et Papiam.

⁸ See above, p. 28.

⁹ D. 21. 1. 17. 18.

¹⁰ P. 174 and Pauli Excerpta 175.

¹¹ D. 50. 16. 238. 3, libro VI ad legem duodecim tabularum.

¹² It is no doubt possible that both Servius, when explaining noxa as damnum, and Gaius, when explaining it as omne delictum, had in mind only those rules of the XII Tables which related to noxal liability (as for Gaius, context, and thereby generalized, by Pestus (or Verrius Flaccus), the latter's by the compilers. There is, however, no proof for this assumption : the jurists may well have taken into account also the provisions about aqua pluvia and the like, which all contained the word. Moreover, even if they did think only of noxal actions, the fact remains that, according to their definitions, the term noxa in itself does not imply indirect responsibility.

¹³ See below, p. 54.

¹⁴ See Vanicek, A.op. cit. vol. i, pp. 326et seq.Google Scholar

¹⁵ The word is frequent in Plautus and always has this sense. Occasionally some lady is represented, not very politely, as the personified damnum : e.g. Menaechmi 133.

¹⁶ D. 9. 2. 29. 2.

¹⁷ 3. 219.

¹⁸ It no doubt comes to us from Gotkofredus on I. 4. 3. 16; but comp. already the Lex Eomana canonice compta.

¹⁰ A characteristic example is to be found in D. 39. 2. 24. 5 (Ulpian with reference to Servius) : si ex aedibus … tegulae deiectae damnum vicino (not aedibus vicinis) dederint. The etymology of damnum in 39. 2. 3, originally given by Paul when commenting upon legacy per damnationem (Lenel, Palin-genesia, vol. i, p. 1054), seems to be very much in favour of my suggestion. Damnum, as explained by Paul, is not the harm done to the thing, but the loss suffered by its owner : damnum et damnatio ab ademptione et quasi deminutione patrimonii dicta sunt.

²⁰ There are, of course, a good many texts showing a tendency in this direction and thus preparing the medieval use of damnum. In the field of the lex Aquilia itself, I. 4. 3. 16 is to be mentioned : si non corpore damnum fuerit datum neque corpus laesum fuerit. Outside the lex Aquilia, damnum occasionally signifies something like the concrete harm inflicted in the case where a free man is wounded : D. 9. 3. 1. 6, 9. 3. 5. 5, 9. 3. 5. 12. Evidently, the concrete harm and the expenditure resulting from it meet here in the same object, the wounded person : in other words, of the three steps, harmful act, effect on the object harmed, and ensuing loss, the latter two more or less coincide and are not, therefore, clearly distinguished by the lawyers. Moreover, the common extra-legal use of damnum in the sense of ‘natural defect’ (Livy 7. 4. 6) may have been of some influence in these passages : it may also account for damnum as actual harm in D. 11. 3. 9. 3, on servus corruptus. Several other cases (all outside the lex Aquilia) ought to be added, but I cannot here go into them.

²¹ Livy, 3. 55. 7.Google Scholar

²² Livy, 3. 55. 5.Google Scholar

²³ In two (D. 9. 3. 1 pr. and 21. 1. 40 et seq.) the expression damnum datum factumve is used, which must have been created by the interpreters of the lex Aquilia; and damage to a slave falls under damnum (as in the lex Aquilia), not (as in the XII Tables) under the same category as damage to a free man.

²⁴ D. 9. 3. 1 pr.

²⁵ D. 9. 3. 5. 6.

²⁶ D. 9. 3. 5. 12.

²⁷ D. 21. 1. 40 et seq.

²⁸ This rule indicates that the edict is later than the two others : it is modelled after the actio iniuriarum aestimatoria.

²⁹ Damage to a slave may constitute an iniuria under the ‘edictum generate’ : on this, see below.

³⁰ Amphitryon 391, Asinaria 112, Bacchides 690. The terminology of the lex Aquilia has no influence in this domain : as a rule a slave is flogged by his owner, and the damaging of one's own slave never comes under the lex. It may be observed, however, that in everyday language even to flog another's slave can be called nocere. In Amphitryon 391, Mercury promises Sosias not to hurt him : dic si quid vis, non nocebo. It is only natural that he should put it this way : neither he nor Sosias is concerned with the legal question whether the latter's owner may possibly suffer a damnum and have an action under the statute; they are merely concerned with poor Sosias' body.

³¹ Noctes Atticae 20. 1. 13.

³² Origines 4, preserved in Priscian 6. 13. 69. It does not follow, of course, that no agreement for composition can be made : but the actio iniuriarum (existing in Cato's time) does not lie.

³³ The singular also occurs in the edict : qui iniuriarum aget certum dicat quid iniuriae factum sit (Lenel, Edictum, p. 389). The praetor emphasizes the fact that it is not sufficient for the plaintiff just to declare that iniuriae have been offered; he must specify the exact iniuria (singular) committed.

³⁴ Lenel, loc. cit.Google Scholar

³⁵ This is expressly stated by Festus, P. 181. The etymology of convicium given by Ulpian D. 47. 10. 15. 3 et seq. seems to be old : convicium … a collatione vocum … convicium appellatur quasi convocium. Even in classical times, though the disparaging speech by a single person may constitute convicium, it must be made in coetu : D. 47. 10. 15. 11 et seq.

³⁶ As pattern formula Coll. 2. 6. 5 gives quod Numerius Negidius ilium inmisit Aulo Agerio infamandi causa. Lenel (Edictum, p. 400) alters this into quod Numerius Negidius illi libellum misit Auli Agerii infamandi causa. I should suggest that the offence originally contemplated is the case which Ulpian (47. 10. 15. 27) mentions before all other examples, and which has never been satisfactorily explained : the case where a man ad invidiam alicuius veste lugubri utitur aut squalida, aut si barbum demittat vel capillos submittat. What is meant, I think, is that A, by mourning somebody whom it is not (or no longer) necessary to mourn, tries to discredit B who does not mourn : for instance, a man mourns his deceased brother after the prescribed period, in order to discredit his brother's widow who is marrying again. If this is correct, it follows that the word infamare in this edict originally had a much fuller sense than would appear from the Digest: those who neglect the rules of luctus without justification are subjected to serious disabilities, are ‘infamous’, not only in everyday life but also in litigation. Further, the Aulo Agerio in the pattern formula can, on the basis of this view, be kept. As pattern formula I should conjecture : quod Numerius Negidius capillum submisit or demisit (instead of ilium inmisit; comp. capillos submittat in 47. 10. 15. 27) Aulo Agerio (dative incommodi; comp. ad invidiam alicuius in 47. 10. 15. 27) infamandi causa.

³⁷ This possibility was first suggested by Hitzig, Iniuria, p. 55.

³⁸ D. 47. 10. 15. 26, Ulpian referring to Labeo.

³⁹ D. 47. 10. 7. 1 (compare already Cicero, De Inventione 2. 20. 59 et seq.). As the text stands, Ulpian differs from Labeo. But there may be interpolation, the compilers preferring publica animadversio.

⁴⁰ The nature of the wrong in these edicts (behaviour dangerous to the public but not, in the ordinary way, involving criminal design) is very different from the nature of iniuria (ὕβρις). The term iniuria, therefore, does not intrude into this province.

⁴¹ D. 17. 2. 52. 16 (Ulpian). Neratius does not here have in mind the purely intellectual forms of injury, convicium, and the like: similarly, nocere means only physical injury in Paul 5. 3. 1 (Quae per turbam fiunt). The question of interpolation is immaterial to my argument.

⁴² They are frequent especially where the question of iniuria does not arise. In Cicero, Pro Caecina 21, 60, the distinction is made (and ridiculed : do our international lawyers know this?) between weapons ad tegendum and weapons ad nocendum. To say ad iniuriam faciendam instead of ad nocendum would not only be clumsy but also slightly alter the meaning of the antithesis. Weapons ad nocendum are instruments essentially destined for use in war and warlike enterprises : the enterprise may constitute an iniuria or it may not.

⁴³ See Visscher, Deop. cit. p. 126, with further literature.Google Scholar

⁴⁴ Tabula Heracleensis, lines 110—111, Fragmentum Atestinum, lines 2—4.

⁴⁵ Op. tit. p. 347.Google Scholar

⁴⁶ Die Gemeindeordonnanzen der Tafel von Heraclea, Sitzungeberichte der Heidelberger Akademie, Philos.—Histor. Klasse, 1916, 14.

⁴⁷ With this view that of Premerstein does not actually conflict. Premerstein says (Die Tafel von Heraclea und die Acta Caesaris, Zeitschrift der Savigny-Stiftung, 1922, pp. 45 et. seq.) that the Tabula Heracleensis contains various Acta Caesaris drafted in Caesar's lifetime and promulgated, unrevised and as one law, after his death. In fact, Premerstein seems to confirm Gradenwitz. His theory explains why there are more and stronger inconsistencies in the Tabula Heracleensis than in other statutes concerning municipal government.

⁴⁸ He is mentioned also in insolvency, dealt with in lines 114—115.

⁴⁹ See Visscher, Deop. cit. pp. 383Google Scholaret seq. with further references.

⁵⁰ Plato, Ν⋯μοι 9. 879A and 883A, has noxal liability where a slave actually wounds a free man; he has no noxal liability where a slave merely beats a free man. The distinction, which seems to reflect some positive law Plato knows (see Hitzig, Iniuria, pp. 85 et seq.), may well go back to a time when a difference was made between real bodily damage on the one hand and beating a man, as ὕβρις, on the other : a slave could not be guilty of ὕβρις; if he beat A free man, he was punished ‘extra ordinem’.

¹ D. 48. 2. 12. 4.

² Livy 8.28. 8.

³ According to De Visscher, La Lex Poetelia Papiria, Etudes, pp. 315 et seq. noxa comprises here all delicts, a most plausible interpretation. But, then, has the author of the lex quite forgotten the distinction between noxa and bodily damage? Has he quite forgotten that poena is not the proper term to use in a rule covering delicts against the property?

⁴ I very much doubt whether the term poena, at the time of the XII Tables, was used only as signifying a fixed penalty. But I accept this premise for the sake of argument.

⁵ In Table VIII 9, dealing with the impubes who cuts another's crops at night, the term noxia is used.

⁸ It would lead too far afield to examine when the person responsible for a noxa must noxiam sarcire, when damnum decidere, when poenam luere and so forth. Suffice it to say, with regard to the first two expressions, that noxiam sarcire seems to come from those cases where no more and no less is required than actually ‘to mend the damage’: and damnum decidere from those where there is either no concrete damage to mend (because nothing has actually been destroyed) or where more than just ‘mending’ is expected (vindicta falsa, possibly furtum, and the like); that is to say, where the responsible party must seek ‘an agreement concerning the loss’.

⁷ But there is exactly the same penalty in iniuria as in cutting down a tree.

⁸ Even in furtum that was not manifest the damages must frequently have amounted to more than in os fractum.

⁹ According to Festus P. 174, Servius interpreted noxia as damnum. But it is very unlikely that he had in mind the noxiam nocere of Table XII 2. Even if he had, it by no means follows that he intended to exclude bodily damage : damnum in his time was wide enough to comprise both the loss resulting from damage to property and that resulting from damage to the person. Most probably, however, Servius was referring to the phrase noxiam sarcire. This is suggested by Festus P. 322 : sarcito in XII Ser. Sulpicius ait significare damnum solvito, praestato.

¹⁰ G. 4. 75 opens the chapter on the noxal actions with the words : ex maleficio filiorum familias seroorumque, veluti si furtum fecerint aut iniuriam commiserint. … If noxiam nocere in Table XII 2 referred primarily to damage to a person's body, the clause si furtum fecerint aut iniuriam commiserint in Gaius may well be a modern version (made by Gaiug himself or before him) of si furtum faxit noxiamve noxit: in classical times, the technical term for damage to a person's body was iniuria.

¹¹ In the phrase furtis noxisque solutus, however, noxae (plural) cannot mean punishment. This clause, obviously modelled after Table XII 2, must mean : a slave who is free from thefts and other wrongs. Which of the two, if any, is older than the other, furtis noxisque solutus (Varro, Kerum Eusticarum 2. 10. 5) or noxa solutus (edict of the aediles; Gellius, Noctes Atticae 4. 2. 1, and D. 21. 1. 1. 1), is difficult to say.–I should like to thank Mr. P. W. Duff for his most valuable criticism.