Westermarck, Origin etc. of Moral Ideas, ii. 584.
 Jevons, Introduction, p. 33.
 A useful summary of the whole subject, embodying
the results and terminology of Tylor, Frazer, and other
anthropologists, is Dr. Haddon's Magic and Fetishism,
in Messrs. Constable's series, Religions Ancient and
Modern. See also Marett, On the Threshold of
 Lectures on the Early History of the Kingship, p.
89 foll. For an example not mentioned in the text
(devotio) see below, p. 206 foll. This may have been
originally practised by the Latin kings. I may here
draw attention to the almost dogmatic conclusions of the
modern French sociological school of research; e.g. M.
Huvelin, in L'Année sociologique for 1907, begins by
asserting as a fundamental law, proved by MM. Hubert et
Mauss, that magic is just as much a social fact as
religion: "Les uns et les autres sont des produits de
l'activité collective" (Magie et droit individuel, p.
1). But M. Huvelin's paper is to some extent a
modification of this dogma. He seeks to explain the fact
that magic is both secret and private, not public and
social, in historical times; and in the domain of law,
with which he is specially concerned, he concludes that
"a magical rite is only a religious rite twisted from
its proper social end, and employed to realise the will
or belief of an individual" (p. 46). This is the only
form in which we shall find magic at Rome, except in so
far as a few of its forms survive in the ritual of
religion with their meaning changed. In early Roman law,
as a quasi-religious body of rules and practices, there
are a few magical survivals which will be found
mentioned by M. Huvelin in this article; but they are of
no importance for our present subject.
 Primitive Culture, vol. i. ch. iv. See also
Jevons, Introduction, p. 36 foll.
 See Schürer, Jewish People in the Time of Christ
(Eng. trans.), Division II. vol. iii. p. 151 foll.
 Fowler, R.F. p. 232; Wissowa, R.K. p. 106. The
most careful examination of the rite and the evidence
for it is that of Aust in Mythological Lexicon, s.v.
"Iuppiter," p. 656 foll. See also M.H. Morgan in vol.
xxxii. of Transactions of the American Philological
Association, p. 104.
 Tertullian, de Jejun. 16. Petronius, Sat. 44,
adds that the matrons went in the procession with bare
feet and streaming hair (cp. Pliny xvii. 266); but this
seems rather Greek than Roman in character, and
Petronius is plainly thinking of the town (colonia he
calls it) in southern Italy where the scene of
Trimalchio's supper is laid; probably a Greek city by
origin, Croton or Cumae. A translation of this passage
will be found in Dill's Roman Society from Nero to
Marcus Aurelius, p. 133. The most useful words in it
for our purpose are "Jovem aquam exorabant."
 This suggestion was originally made by O. Gilbert,
Röm. Topographie, ii. 184.
 p. 204 foll.
 p. 657. The story is mixed up with Greek fables,
e.g. that of Proteus, as Wissowa has pointed out,
R.K. p. 106, note 10.
 See Schanz, Gesch. der röm. Literatur, vol. i.
(ed. 3) p. 270 foll.
 This fragment of Piso is preserved by Gellius, xi.
 See, e.g., Schanz, Gesch. der röm. Literatur,
vol. ii. p. 106.
 Wissowa, l.c. Aust in Roscher's Lexicon, s.v.
"Iuppiter," p. 657.
 Cumont, Religions Orientales dans le paganisme
romain, ch. 5. I shall return to this subject in my
second course of lectures.
 Müller-Deecke, Etrusker, ii. ch. vii., especially
p. 176 foll.
 Cp. below, Lecture XV.
 Pliny, N.H. xxviii. 13: "Vestales nostras hodie
credimus nondum egressa urbe mancipia fugitiva retinere
in loco precationibus."
 Plutarch, Numa, 10. Virginity would increase the
power of the spell; see Fehrle, Die kultische
Keuschheit im Altertum, p. 54 foll.
 See, e.g., Frazer, G.B. i. 360 foll.
 See R.F. p. 320, notes 6 and 7.
 Within the last thirty years or so the Lupercalia
has been discussed (apart from writers on classical
subjects exclusively) by Mannhardt in his Mythologische
Studien, p. 72 foll.; Robertson Smith, Semites, p.
459; Deubner in Archiv, 1910, p. 481 foll.; and at the
moment of writing by E. S. Hartland, Primitive
Paternity, i. ch. ii. R.F. p. 310 foll. See Appendix
 This view was originally stated in Pauly-Wissowa,
s.v. "Argei." I endeavoured to confute it in the
Classical Review, 1902, p. 115 foll., and Wissowa
replied in Gesammelte Abhandlungen, p. 211 foll. Since
then my conviction has become stronger that this great
scholar is for once wrong. Ennius alluded to the Argei
as an institution of Numa, i.e. as primitive (frag.
121, Vahlen, from Festus p. 355, and Varro, L.L. vii.
44), yet Ennius was a youth at the very time when
Wissowa insists that the rite originated. Wissowa makes
no attempt to explain this. See below, p. 321 foll.
 R.F. p. 111 foll.
 e.g. the October horse, which also occurred on
the Ides; see R.F. p. 241 foll.; and the festival of
Anna Perenna, also on Ides (March 15), R.F. p. 50
foll. It is just possible that all the three festivals
were originally in the old calendar, and dropped out
because the mark of the Ides had to be affixed to the
day in the first place. See Wissowa, Gesammelte
Abhandlungen, p. 164 foll.; R.F. p. 241.
 Thus Messrs. Hubert et Mauss (Mélanges d'histoire
des religions, Preface, p. xxiv.) maintain that there
is no real antinomy between "les faits du système
magique et les faits du système religieux." There is in
every rite, they insist, a magical as well as a
religious element. Yet on the same page we find that
they exclude magic from all organised cult, because it
is not obligatory, and cannot (if I understand them
rightly) be laid down in a code, like religious
practice. I think it would have been simpler to consider
the magical element in religious rites as surviving,
with its original meaning lost, from an earlier stage of
thought. M. van Gennep, in his interesting work Les
Rites de passage, p. 17, goes so far as to call all
religious ceremonies magical, as distinguished from
the theories (e.g. animism) which constitute
religion. This seems to me apt to bring confusion into
the discussion; for all rites are the outward expression
of thought, and it is by the thought (or, as he calls
it, theories) that we must trace the sociological
development of mankind, the rites being used as indexes
only. I cannot but think that (as indeed in these days
is quite natural) this French school lays too much
stress upon the outward acts, and that this tendency has
led them to find real living magic where it is present
only in a fossil state.
 e.g. Tylor, article "Magic" in Encycl. Brit.,
and Primitive Culture, 1. ch. iv.; Marett, Threshold
of Religion, 83. See below, p. 180.
 Pliny, N.H. xxviii. 17 and 18. For the singing
or murmuring of spells in many countries, see Jevons,
Anthropology and the Classics, p. 93 foll.
 Bruns, Fontes Iuris Romani, note on this
 Civ. Dei, viii. 19.
 See, e.g., Wordsworth, Fragments and Specimens
of Early Latin, p. 446, for an account of simple land
measurement which will suffice to illustrate the point
 The carmina famosa sung at a triumph by the
soldiers had the same origin, but were used to avert
evil from the triumphator. The best exposition of this
is in H. A. J. Munro's Elucidations of Catullus, p. 76
 Pliny, N.H. xxviii. 19. For the technical sense
of defigere, defixio, see Jevons in Anthropology
and the Classics, p. 108 foll.
 The most familiar examples are Virgil's eighth
Eclogue, 95 foll.; Ovid, Met. vii. 167, and
elsewhere; Fasti, iv. 551; Horace, Epode v. 72; cp.
article "Magia" in Daremberg-Saglio; Falz, De poet.
Rom. doctrina magica, Giessen, 1903. There is a
collection of Roman magical spells in Appel's De
Romanorum precationibus, p. 43 foll. Many modern
Italian examples and survivals will be found in Leland's
Etruscan Roman Remains in Popular Tradition, pt. ii.
 Cato, R.R. 160; Varro, R.R. i. 3.
 Pliny, N.H. xxviii. 21.
 Ib. xxviii. 20. The following sections of this
book are the locus classicus for these popular
 See, e.g., Italian Home Life, by Lina Duff
Gordon, p. 230 foll.
 Juvenal v. 164. The idea probably arose, as a
passage of Plutarch suggests (Rom. 25), from the fact
that the triumphator, whose garb was no doubt of
Etruscan origin, wore the bulla.
 Frazer, G.B. i. 345, note 2, where we learn that
gold was taboo in some Greek worships, e.g. at the
mysteries of Andania, which sufficiently proves that it
possessed potency. Pliny, xxxiii. 84, mentions cases of
such potency as medicine, and among them its application
to children who have been poisoned.
 Pliny, N.H. xxviii. 39.
 See an article by the author on the original
meaning of the toga praetexta in Classical Review,
vol. x. (1896) p. 317.
 For the Compitalia, Macrob. i. 7. 34; Festus p.
238. For the Paganalia, Probus, ad Georg. ii. 385,
assuming the feriae Sementinae there mentioned to be
the Paganalia (see R.F. p. 294). For the feriae
Latinae, Festus, s.v. "oscillantes."
 Wissowa, R.K. p. 193, with whose view I entirely
agree. We learn of the imaginary goddess from Varro,
L.L. ix. 61. Pais, I may remark in passing, is certain
that Acca Larentia was the mater Larum; see his
Lectures on Ancient Legends of Roman History, p. 60
 46. Wissowa, R.K. p. 354, note 5.
 Georg. ii. 380 foll. It is not certain that
Virgil is describing the festival generally known as
Paganalia, which took place early in January; but it
seems probable from line 382 that he is thinking of some
festival of the pagus. The oscilla may have been used
at more than one.
 Note that Virgil writes of masks used in rude
play-acting, as well as of oscilla hung on trees, and
conjoins the two as though they had something in common.
The evidence of an engraved onyx cup in the Louvre, of
which a cut is given in the article "Oscilla" in the
Dict. of Antiquities, seems to make it probable that
masks worn by rustics on these occasions were afterwards
hung by them on trees as oscilla. Some of these masks
on the cup are adorned with horns, which may explain an
interesting passage of Apuleius (Florida, i. 1):
"neque enim iustius religiosam moram viatori obiecerit
aut ara floribus redimita ... aut quercus cornibus
onerata, aut fagus pellibus coronata," etc. See also
Gromatici veteres, ii. 241.
 See, however, Dr. Frazer's remarks in G.B. ii.
p. 454. He thinks that the air might in this way be
purged of vagrant spirits or baleful ghosts, as the
Malay medicine man swings in front of the patient's
house in order to chase away the disease. Cp. G.B. ii.
343, where a rather different explanation is attempted
of the maniae and pilae.
 Magic in the old forms, or many of them, has
survived not only into the old Roman religion, but to
the present day, in many parts of Italy. "The peasants
have recourse to the priests and the saints on great
occasions, but they use magic all the time for
everything," was said by a woman of the Romagna Toscana
to the late C.G. Leland (Etruscan Roman Remains,
Introduction, p. 9). This enterprising American's
remarkable book, though dealing only with a small region
of northern Italy, deserves more consideration than it
has received. The author may have been uncritical, but
beyond doubt he had the gift of extracting secrets from
the peasantry. He claims to have proved that "la vecchia
religione" contains much that has come down direct from
pre-Christian times; and the appearance of Mr. Lawson's
remarkable book on Modern Greek Folklore and Ancient
Greek Religion may tempt some really qualified
investigator to undertake a similar work in Italy before
it is too late.