The Calendar of this month is almost a blank. Only the Kalends, Nones and Ides are marked in the large letters with which we have become familiar; no other festival is here associated with a special deity. But the greater part of the month is occupied with the ludi Romani (5th to 19th)^[930], and the 13th (Ides), as we know from two Calendars, was not only, like all Ides, sacred to Jupiter, but was distinguished as the day of the famous ‘epulum Jovis,’ and also as the dies natalis of the great Capitoline temple.

The explanation of the absence of great festivals in this month is comparatively simple. September was for the Italian farmer, and therefore for the primitive Roman agricultural community, a period of comparative rest from urgent labour and from religious duties; for no operations were then going on which called for the invocation of special deities to favour and protect. A glance at the rustic calendars will show this well enough^[931]. The messes which figure in July and August have come to an end, and the vintage does not appear until October. There is of course work to be done, as always, but it is the easy work of the garden and orchard. ‘Dolia picantur: poma legunt: arborum oblaqueatio.’ Varro, who divides the year for agricultural purposes into eight irregular periods, has little to say of the fifth of these, i. e. that which preceded the autumn equinox. ‘Quinto intervallo inter caniculam et aequinoctium autumnale oportet stramenta desecari, et acervos construi, aratro offringi, frondem caedi, prata irrigua iterum secari^[932].’

This was also the time when military work would be coming to an end. In early times there were of course no lengthy campaigns; and such fighting as there was, the object of which would be to destroy your enemies’ crops and harvest, would as a rule be over in August. Even in later times, when campaigns were longer, the same would usually be the case; and the performance of vows made by the generals in the field, and also their vacation of office, would naturally fall in this month. We find, in fact, that the ludi which occupied so large a number of September days, had their origin in the performance of the vota of kings or consuls after the close of the wars^[933]; and we have evidence that the Ides of September was the day on which the earliest consuls laid down their office^[934]. There was, in fact, every opportunity for a lengthened time of ease; the people were at leisure and in good temper after harvest and victory; even the horses which took part in the games were home from war service or resting from their labours on the farm^[935].

It is not strictly within the scope of this work to describe the ludi Romani, which in their fully organized form were of comparatively late date; but their close connexion with the cult of Jupiter affords an opportunity for some remarks on that most imposing of all the Roman worships.

The ludi Romani came in course of time, as has been said above, to extend from the 5th to the 19th; they spread out in fact on each side of the Ides^[936], the day on which took place the ‘epulum Jovis’ in the Capitoline temple. As this day was also the dies natalis of the same temple, and that on which the nail was driven into the wall of the cella Jovis^[937], we have a very close connexion between the ludi and the cult of Jupiter. The link is to be found in the fact that in the ludi votivi, which were developed into ludi Romani, the vows were made and paid to the supreme god of the State^[938]. We have from a later time the formula of such a vow preserved by Livy^[939]. ‘Si duellum quod cum rege Antiocho sumi populus iussit id ex sententia senatus populique Romani confectum erit, tum tibi, Iuppiter, populus Romanus ludos magnos dies decem continuos faciet, donaque ad omnia pulvinaria dabuntur de pecunia, quantam senatus decreverit: quisquis magistratus eos ludos quando ubique faxit, hi ludi recte facti donaque data recte sunto.’

The epulum Jovis, thus occurring in the middle of the ludi, is believed by some writers to have originally belonged to the Ides of November and to the ludi plebeii, as it does not happen to be alluded to by Livy in connexion with the ludi Romani, and our first notice of it in September is in the Augustan calendars^[940]. But it is surely earlier than B.C. 230, the received date of the ludi plebeii, and of the circus Flaminius in which they took place. We may agree with the latest investigator of the Jupiter-cult that the origin of the epulum is to be looked for in a form of thanksgiving to Jupiter for the preservation of the state from the perils of the war season, and that no better day could be found for it than the foundation-day of the Capitoline temple^[941]. This epulum was one of the most singular and striking scenes in Roman public life. It began with a sacrifice; the victim is not mentioned, but was no doubt a heifer, and probably a white one^[942]. Then took place the epulum proper^[943], which the three deities of the Capitol seem to have shared in visible form with the magistrates and senate. The images of the gods were decked out as for a feast, and the face of Jupiter painted red with minium, like that of the triumphator. Jupiter had a couch, and Juno and Minerva each a sella, and the meal went on in their presence^[944].

Now an investigator of the Roman religious system is here confronted with a difficult problem. Was this simply a Greek practice like that of the lectisternium, and one which began with the Etruscan dynasty and the foundation of the Capitoline temple with its triad of deities? Or is it possible that in the cult of the Roman Jupiter there was of old a common feast of some kind, shared by gods and worshippers, on which this gorgeous ritual was eventually grafted?

Marquardt has gone so far as to separate the epulum Jovis altogether from the lectisternia, and apparently also from the inundation of Greek influence^[945]. It answers rather, he says, to such domestic rites as the offering to Jupiter Dapalis described thus by Cato in the De Re Rustica^[946]: ‘Dapem hoc modo fieri oportet. Iovi dapali culignam vini quantum vis polluceto. Eo die feriae bubus et bubulcis, et qui dapem facient. Cum pollucere oportebit, sic facies. Iupiter dapalis, quod tibi fieri oportet, in domo familia mea culignam vini dapi, eius rei ergo macte hac illace dape pollucenda esto. Manus interluito. Postea vinum sumito. Iupiter dapalis, macte istace dape pollucenda esto. Macte vino inferio esto. Vestae, si voles, dato^[947]. Daps Iovi assaria pecuina, urna vini lovis caste.’

I confess that I do not see wherein lies the point of the comparison of this passage with the ceremony of the epulum; and Marquardt himself does not attempt to elaborate it. There is no mention here of a visible presence of Jupiter in the form of an image, which is the one striking feature of the epulum. Marquardt, as it seems to me, might better have adduced some example from old Italian usage of the belief that the gods were spiritually present at a common religious meal—a belief on which might easily be engrafted the practice of presenting them there in actual iconic form. Ovid, for example, writes thus of the cult of the Sabine Vacuna^[948]:

Or again in the sacra of the curiae, if Dionysius reports them rightly^[949], we find a clear case of a common meal in which the gods took part. He tells us that he saw tables in the ‘sacred houses’ of the curiae spread for the gods with simple food in very primitive earthenware dishes. He does not mention the presence of any images of the gods, but it is probable from his interesting description that each curia partook with its gods of a common meal of a religious character, and one not likely to have come under Greek influence^[950].

This last example may suggest a hypothesis which is at least not likely to do any serious harm. Let it be remembered that each curia was a constituent part of the whole Roman community. We might naturally expect to find a common religious meal of the same kind in which the whole state took part through its magistrates and senate. This is just what we do find in the epulum Jovis, though the character of its ceremonial is different; and it is certainly possible that this epulum had its origin in a feast like that which Dionysius saw, but one which afterwards underwent vital changes at the hands of the Etruscan dynasty of Roman kings. I am strongly inclined to believe that it was under the influence of these kings that the meal came to take place on the Capitol, and in the temple of Jupiter, Juno, and Minerva, which they intended to be the new centre of the Roman dominion^[951]; and to them also I would ascribe the presence at the feast of the three deities in iconic form. It may be that before that critical era in Roman history the epulum took place not on the Capitol but in the Regia, which with the temple of Vesta hard by formed the oldest centre of the united Rome; and that the presence of Jupiter^[952] or any other god was there a matter of belief, like that of Vacuna with the Sabines, and not of the actual evidence of eyesight.

But this conjecture is a somewhat bold one; and it seems worth while to take this opportunity of examining more closely into the cult of Jupiter, with the object of determining whether the great god was apt, in any part of Italy but Etruria, to lend himself easily to anthropomorphic ideas and practices^[953].

The cult of Jupiter is found throughout Italy under several forms of the same name, with or without the suffix -piter = pater, which, so far as we can guess, points to a conception of the god as protector, if not originator, of a stock. This paternal title, which was applied to other deities also, does not necessarily imply an early advance beyond the ‘daemonistic’ conception of divine beings; it rather suggests that some one such being had been brought into peculiarly close relations with a particular stock, and does no more than indicate a possibility of further individual development in the future^[954]. The ‘father’ in this case has no wife, though we find the word ‘mater’ applied to goddesses^[955]; Juno is undoubtedly the female principle, but she is not, as has so often been imagined, the wife of Jupiter. The attempt to prove this by arguing from the Flamen Dialis and his wife the Flaminica cannot succeed: the former was the priest of Jupiter, but his wife was not the priestess of Juno^[956]. There is indeed a certain mysterious dualism of male and female among the old Italian divinities, as we know from the locus classicus in Gellius (N. A. 13. 23. 2); but we are not entitled to say that the relation was a conjugal one^[957].

Before we proceed to examine traces of the oldest Jupiter in Rome and Latium let us see what survivals are to be found in other parts of Italy.

In Umbria we find Jovis holding the first place among the gods of the great inscription of Iguvium, which beyond doubt retains the primitive features of the cult, though it dates probably from the last century B.C., and records rites which indicate a fully developed city-life^[958]. His cult-titles here are Grabovius, of which the meaning is still uncertain, and Sancius, which brings him into connexion with the Semo Sancus and Dius Fidius of the Romans. The sacrifices and prayers are elaborately recorded, but there is no trace in the ritual of anything approaching to an anthropomorphic conception of the god, unless it be the apparent mention of a temple^[959]. No image is mentioned, and there is no sign of a common meal. The titles of the deities too have the common old-Italian fluidity, i. e. the same title belongs to more than one deity^[960]. Everything points to a stage of religious thought in which the personality of gods had no distinct place. The centre-point of the cult seems to be a hill, the ocris fisius, within the town of Iguvium, which reminds us of the habits of the Greek Zeus and the physical or elemental character—unanthropomorphized—which seems to belong to that earlier stage in his worship^[961].

It is on a hill also that we find the cult among the Sabellians. An inscription from Rapino in the land of the Marrucini tells us of a festal procession in honour of ‘Iovia Ioves patres ocris Tarincris,’ i. e. Jovia (Juno?) belonging to the Jupiter of the hill Tarincris^[962].

Among the Oscan peoples the cult-title Lucetius is the most striking fact. Servius^[963] says: ‘Sane lingua Osca Lucetius est Iuppiter dictus a luce quam praestare hominibus dicitur.’ The same title is found in the hymn of the Roman Salii^[964], and is evidently connected with lux; Jupiter being beyond doubt the giver of light, whether that of sun or moon. So Macrobius^[965]: ‘Nam cum Iovem accipiamus lucis auctorem, unde et Lucetium Salii in carminibus canunt et Cretenses ??a t?? ???a? vocant, ipsi quoque Romani Diespitrem appellant ut diei patrem. Iure hic dies Iovis fiducia vocatur, cuius lux non finitur cum solis occasu, sed splendorem diei et noctem continuat inlustrante luna,’ &c. The Ides of all months, i. e. the days of full moon, were sacred to Jupiter. But in all ceremonies known to us in which the god appears in this capacity of his, there is, as we might expect, no trace whatever of a personal or anthropomorphic conception.

The Etruscan Tina, or Tinia, is now generally identified, even etymologically, with Jupiter^[966]. The attributes of the two are essentially the same, though one particular side of the Etruscan god’s activity, that of the lightning-wielder, is specially developed. But Tina is also the protector of cities, along with Juno and Minerva (Cupra and Menvra); and it is in connexion with this function of his that we first meet with a decided tendency towards an anthropomorphic conception. Even here, however, the stimulus can hardly be said to have come from Italy. ‘The one fact,’ says Aust^[967], ‘which is at present quite clear is that the oldest Etruscan representations of gods can be traced back to Greek models. Tinia was completely identified in costume and attributes with the Greek Zeus by Etruscan artists.’ The insignia of Etruscan magistrates were again copied from these, and have survived for us in the costume of the Roman triumphator^[968], and in part in the insignia of the curule magistrate, i. e. in sceptrum, sella, toga palmata, &c., and in the smearing of the face of the triumphator with minium.

Coming nearer to Rome we find at Falerii, a town subject to Roman and Sabellian as well as Graeco-Etruscan influence, the curious rite of the ?e??? ???? described by Ovid (Amores, 3. 13), and found also in many parts of Greece^[969]. In this elaborate procession Juno is apparently the bride, but the bridegroom is not mentioned. At Argos, Zeus was the bridegroom, and the inference is an obvious one that Jupiter was the bridegroom at Falerii. But this cannot be proved, and is in fact supported by no real evidence as to the old-Italian relation of the god and goddess. The rite is extremely interesting as pointing to what seems to be an early penetration of Greek religious ideas and practices into the towns of Western Italy; but it has no other bearing on the Jupiter-question, nor are we enlightened by the little else we know of the Falerian Jupiter^[970].

But at Praeneste, that remarkable town perched high upon the hills which enclose the Latin Campagna to the north, we find a very remarkable form of the Jupiter-cult, and one which must be mentioned here, puzzling and even inexplicable as it certainly is. The great goddess of Praeneste was Fortuna Primigenia—a cult-title which cannot well mean anything but first-born^[971]; and that she was, or came to be thought of as, the first-born daughter of Jupiter is placed beyond a doubt by an inscription of great antiquity first published in 1882^[972]. But this is not the only anomaly in the Jupiter-worship of Praeneste. There was another cult of Fortuna, distinct, apparently, from that of Fortuna Primigenia, in which she took the form not of a daughter but of a mother, and, strange as it may seem, of the mother both of Jupiter and Juno. On this point we have the explicit evidence of Cicero (de Divinatione, 2. 85), who says, when speaking of the place where the famous ‘sortes’ of Praeneste were first found by a certain Numerius Suffustius: ‘Is est hodie locus saeptus religiose propter Iovis pueri (sacellum?) qui lactens cum Iunone Fortunae in gremio sedens, castissime colitur a matribus.’ Thus we have Fortuna worshipped in the same place as the daughter and as the mother of Jupiter; and nowhere else in Italy can we find a trace of a similar conception of the relations either of these or any other deities. We cannot well reject the evidence of Cicero, utterly unsupported though it be: we must face the difficulty that we have here to account for the occurrence of a Jupiter who is the child of Fortuna and also apparently the brother of Juno, as well as of a Jupiter who is the father of Fortuna.

As regards this last feature, the fatherhood of Jupiter, Jordan says emphatically^[973]—and no scholar was more careful in his judgements—that in the whole range of Italian religions ‘liberorum procreatio nulla est unquam’: and he would understand ‘filia’ in the inscription quoted above in a metaphorical rather than a physical sense. Yet however we choose to think of it, Mommsen is justified in remarking^[974] on the peculiarly anthropomorphic idea of Fortuna (and we may add of Jupiter) at which the Latins of Praeneste must have arrived, in comparison with the character of Italian religion generally.

Even more singular than this is the sonship of Jupiter and the fact that he appeared together with Juno in the lap of Fortuna ‘mammae appetens.’ Cicero’s language leaves no doubt that there was some work of art at Praeneste in which the three were so represented, or believed to be represented. Yet there are considerations which may suggest that we should hesitate before hastily concluding that all this is a genuine Italian development of genuine Italian ideas.

1. Italy presents us with no real parallel to this child-Jupiter though in Greece we find many. Jordan has mentioned three possible Italian parallels, but rejected them all: Caeculus Volcani, the legendary founder of Praeneste, Hercules bullatus, and the beardless Veiovis. The attributes of the last-named are explained by a late identification with Apollo^[975]; Hercules bullatus is undoubtedly Greek: the story of the birth of Caeculus is a foundation-legend, truly Italian in character, but belonging to a different class of religious ideas from that we are discussing. To these we may add that the boy-Mars found on a Praenestine cista is clearly of Etruscan origin, as is shown by Deecke in the Lexicon, s. v. Maris.

2. Cicero’s statement is not confirmed by any inscription from Praeneste. Those which were formerly thought to refer to Iupiter Puer^[976] are now proved to belong to Fortuna as Iovis puer (= filia). It is most singular that Fortuna should be thus styled Iovis puer in the same place where Jupiter himself was worshipped as puer; still more so that in one inscription (2868) the cutter should have dropped out the ‘s’ in Iovis, so that we actually read Iovi puero. It may seem tempting to guess that the name Jupiter Puer arose from a misunderstanding of the word puer as applied to Fortuna: but the evidence as it stands supplies no safe ground for this.

3. The fact that Cicero describes a statue is itself suspicious, in the absence of corroborative evidence of any other kind^[977]: for it suggests that the cult may have arisen, and have taken its peculiar form, as a result of the introduction of Greek or Graeco-Etruscan works of art. In Praeneste itself, and in other parts of Latium and of Campania, innumerable terra-cottas have been found^[978], of the type of the Greek ?????t??f??, i. e. a mother, sitting or standing, with a child, and occasionally two children^[979] in her lap. These may, indeed, be simply votive offerings, to Fortuna and other deities of childbirth: but such objects may quite well have served as the foundation from which the idea of Fortuna and her infants arose. There is a passage in Servius which seems to me to show a trace of a similar confusion elsewhere in this region of Italy. ‘Circa hunc tractum Campaniae colebatur puer Iuppiter qui Anxyrus dicebatur quasi ??e? ?????, id est sine novacula: quia barbam nunquam rasisset: et Iuno virgo quae Feronia dicebatur^[980].’ True, the Jupiter of Anxur is a boy or youth^[981], not an infant: but the passage serves well to show the fluidity of Italian deities, at any rate in regard to the names attached to them. That this puer Iuppiter was originally some other deity, and very possibly a Greek one, I have little doubt: while Juno Virgo, Feronia, Fortuna, Proserpina, all seem to slide into each other in a way which is very bewildering to the investigator^[982]. This is no doubt owing to two chief causes—the daemonistic character of the early Italian religion, in which many of the spiritual conceptions were even unnamed; and, secondly, the confusion which arose when Greek artistic types were first introduced into Italy. Two currents of religious thought met at this point, perhaps in the eighth and following centuries B.C.; and the result was a whirlpool, in which the deities were tossed about, lost such shape as they possessed, or got inextricably entangled with each other. The French student of Praenestine antiquities writes with reason of ‘the negligence with which the Praenestine artists placed the names of divinities and heroes on designs borrowed from Greek models, and often representing a subject which they did not understand^[983].’

4. And lastly, there is no doubt that Praeneste, in spite of its lofty position on the hills, was at an early stage of its existence subject to foreign influences, like so many other towns on or near the western coast of central Italy. This has been made certain by works of art found in its oldest tombs^[984]. Whether these objects came from Greece, Phoenicia, Carthage, or Etruria, the story they tell is for us the same, and may well make us careful in accepting a statement like that of Cicero’s without some hesitation. There was even a Greek foundation-legend of Praeneste, as well as the pure Italian one of Caeculus^[985]. Evidence is slowly gathering which points to a certain basis of fact in these foundation-stories—of fact, at least, in so far as they seem to indicate that the transformation of the early Italian community into a city and a centre of civilization was coincident with the era of the introduction of foreign trade.

While, then, we cannot hope as yet to account for the singular anomaly in the Jupiter-cult, which is presented to us at Praeneste, we may at least hesitate to make use of it in answering the main question with which we set out—viz. how far we can find in the cult of the genuine Italian Jupiter any tendency towards an anthropomorphic conception of the god. Before we return to Rome a word is needed about the Latin Jupiter. The Latin festival has already been described^[986]: it will be sufficient here to point out that none of its features show any advance towards an anthropomorphic conception of Jupiter Latiaris. The god here is of the same type as at Iguvium, one whose sanctuary—whatever it may originally have been—is in a grove on a hill-top^[987], the conspicuous religious centre of the whole Latin stock inhabiting the plain below. Of this stock he is the uniting and protecting deity; and when once a year his sacred victim is slain, after offerings have been made to him by the representatives of each member of the league, it is essential that each should also receive (and probably consume through its deputies) a portion of the sacrificial flesh (carnem petere). This, the main feature, and other details of the ritual, point to a survival from a very early stage of religious culture, and one that we may fairly call aniconic. The victim, a white heifer, the absence of wine in the libations, and the mention of milk and cheese among the offerings, all suggest an origin in the pastoral age; and it would seem that foreign ideas never really penetrated into this worship of a pastoral race. The objects that have been found during excavations near the site of the ancient temple^[988] show that, as in the worship of the Fratres Arvales and in that of the curiae, so here, the most antique type of sacred vessels remained in use. Undoubtedly there was in later times a temple, and also a statue of the god^[989]: and it is just possible that, as Niebuhr supposed^[990], these were the goal of an ancient Alban triumphal procession, older than the later magnificent rite of the Capitol. But we know for certain that the ancient cult here suggests neither gorgeous ritual nor iconic usage. We see nothing but the unadorned practices of a simple cattle-breeding people.

Coming now once more to Rome itself, where of course we have fuller information, fragmentary though it be, we find sufficiently clear indications of an ancient cult of Jupiter showing characteristics of much the same kind as those we have already noticed as being genuine Italian.

In the first place the cult is associated with hills and also with trees. It is found on that part of the Esquiline which was known as lucus Fagutalis or Fagutal: here there was a sacellum Iovis ‘in quo fuit fagus arbor quae Iovis (so MSS.) sacra habebatur^[991]‘: and the god himself was called Fagutalis. Not far off on the Viminal, or hill of the osiers, there was also an altar of Jupiter Viminius, which we may suppose to have been ancient^[992]. The mysterious Capitolium vetus on the Quirinal may be assumed as telling the same tale, though in historical times the memory of the cult there included Minerva and Juno with Jupiter, i. e. the Etruscan ‘Trias.’ Lastly, on the Capitol itself was the temple of Jupiter Feretrius, reputed to be the oldest in Rome^[993]. It was attributed to Romulus, who, after slaying the king of the Caeninenses, dedicated the first spolia opima on an ancient oak ‘pastoribus sacram,’ and at the same time ‘designavit templo Iovis fines cognomenque addidit deo.’ The oak, we may assume, was the original dwelling of the god, and upon it were fixed the arms taken from the conquered enemy as a thank-offering for his aid^[994]. In this case we seem to be able to guess the development of the cult from this beginning in the tree-worship of primitive ‘pastores.’ The next step would be the erection of an altar below the tree, in a small enclosure, i. e. a sacellum of the same kind as those of the Argei or the Sacellum Larum^[995]. The third stage would be the building of the aedes known to us in history, which Cornelius Nepos says had fallen into decay in his time, and was rebuilt by Augustus on the suggestion of Atticus. Even this was a very small building, for Dionysius saw the foundations of it and found them only fifteen feet wide. This oldest cult of Jupiter was completely overshadowed by the later one of the Etruscan Trias—the aniconic by the iconic, the pure Italian by the mongrel ritual from Etruria.

That this Jupiter Feretrius^[996] was the great Jupiter of pre-Etruscan Rome seems to be proved by his connexion with oaths and treaties, in which he resembles the god of the Latin festival. To him apparently belonged the priestly college of the Fetiales, who played so important a part in the declaring of war and the making of treaties: at any rate it was from his temple that the lapis silex and the sceptrum were taken which accompanied them on their official journeys^[997]. It has been supposed that this lapis silex was a symbol of the god himself, like the spear of Mars in the Regia, and other such objects of cult^[998]. ‘We recognize here the primitive forms of a nature-worship, in which the simple flint was sufficient to bring up in men’s minds the idea of the heavenly power of lightning and thunder,’ i. e. the flint if struck would emit sparks and remind the beholder of lightning. Unluckily the existence of a stone in this temple as an object of worship is not clearly attested. Servius (Aen. 8. 641) says that the Fetials took to using a stone instead of a sword to slay their victims with, ‘quod antiquum Iovis signum lapidem silicem putaverunt esse.’ The learned commentator makes a mistake here which will be obvious to all archaeologists, in putting the age of iron before that of stone; but it has not been equally clear to scholars that he by no means implies his belief that Jupiter was ever worshipped under the form of a stone. He only says that the Fetials fancied that this was so: and the whole passage has an aetiological colouring which should put us on our guard^[999]. It is not supported by any other statement or tradition, except an allusion in S. Augustine^[1000] to a ‘lapis Capitolinus,’ which is surely the stone of Terminus (see below): and by the oath ‘per Iovem lapidem,’ which has been interpreted by some as meaning ‘Jupiter in the form of a stone.’ But this interpretation is at least open to grave doubt; and in the absence of clearer evidence for the ‘Iuppiter lapis’ of the temple it is better to understand the oath as being sworn by the god and also by the stone, ‘two distinct aspects of the transaction being run together,’ in a way not uncommon in Latin formulae^[1001].

It only remains to conjecture what the ‘silex’ or ‘lapis’ was which the Fetials took from the temple together with the sceptrum. Helbig has attempted to prove that it was not a survival of the stone age, e. g. an axe of stone. Had that been so, he argues, the Roman antiquaries, who were acquainted with such implements^[1002], would have noticed it: and those who describe the rites of the Fetials would have stated that the stone was artificially sharpened. But this negative argument is not a strong one; and I am rather inclined to agree with the suggestion of Dr. Tylor^[1003], that it was a stone celt believed to have been a thunder-bolt. There may indeed have been more than one of these kept in the temple, for in B.C. 201 the Fetials who went to Africa took with them each a stone^[1004] (privos lapides silices) along with their ‘sagmina,’ &c. This fact seems to me to prove that the silices, like the sagmina and sceptrum, were only part of the ritualistic apparatus of the Fetials^[1005], and not objects in which the god was supposed to be manifested. The idea that he was originally worshipped in the form of a stone may well have arisen from this use of stones in the ritual, especially if those stones were believed to be in some way his handiwork^[1006]. We may think then of the cult of Jupiter Feretrius as an example of primitive tree-worship, but we are not justified in going further and finding him also in the form of a stone.

There is yet another stone that may have belonged to the earliest Roman cult of Jupiter, but the connexion is not very certain. ‘The (rite of) Aquaelicium,’ says Festus^[1007], ‘is when rain is procured (elicitur) by certain methods, as for example when the lapis manalis is carried into the city.’ This stone lay by the temple of Mars, outside the Porta Capena; we learn from other passages that it was carried by the pontifices^[1008], but we are not told what they did with it within the walls. It has been ingeniously suggested that this rain-spell, as we may call it, was a part of the cult of Jupiter Elicius, to whom there was an altar close by under the Aventine^[1009], the cult-title being identical with the latter part of the word ‘aquaelicium^[1010].’ We may agree that the stone had nothing to do with the temple of Mars, which happened to be near it, and also that any such rain-spell as this would be more likely to belong to the cult of Jupiter than of any other deity. The heaven-god, who launches the thunder-bolt, is naturally and almost everywhere also the rain-giver^[1011]: and that this was one of the functions of Jupiter is fully attested, for later times at least^[1012].

But it must be confessed that the evidence is very slight^[1013]: and it is as well here to remember that the further we probe back into old Italian rites, the less distinctly can we expect to be able to connect them with particular deities. The formula ‘si deus, si dea es’ should always be borne in mind in attempting to connect gods and ceremonies. And this ceremony, like that of the Argei^[1014] (which also wants a clearly-conceived deity as its object), is obviously a survival from a very primitive class of performances which Mr. Frazer has called acts of ‘sympathetic magic^[1015].’ I am indebted to the Golden Bough for a striking parallel to the rite of the lapis manalis, among many others which more or less resemble it. ‘In a Samoan village a certain stone was carefully housed as the representative of the rain-making god: and in time of drought his priests carried the stone in procession and dipped it in a stream^[1016].’ What was done with the lapis manalis we are not told, but it is pretty plain from the word ‘manalis,’ and from the fragments of explanation which have come down to us from Roman scholars, that it was either the object of some splashing or pouring, or was itself hollow and was filled with water which was to be poured out in imitation of the desired rain^[1017]. Such rites need not necessarily be connected by us with the name of a god: and the Jupiter Elicius, with whom it is sought to connect this one, was always associated by the Romans not with this obsolete rite, but with the elaborated science of augury which was in the main Etruscan^[1018].

But this discussion has already been carried on as far as the scope of this work permits. It may be completed by any one who has the patience to work through Aust’s exhaustive article, examining his conclusions with the aid of his abundant references; but I doubt if anything will be found, beyond what I have mentioned, which bears closely on the question with which we set out. That question was, whether the distinctly anthropomorphic treatment of Jupiter in the ‘epulum Iovis’ could be explained by any native Italian practice in his cult (as Marquardt tried to explain it), or must be referred with Aust to foreign, i. e. Graeco-Etruscan, influence. I am driven to the conclusion that Aust is probably right. There is no real trace in Italy of an indigenous iconic representation of Jupiter. Trees and hills are apparently sacred to him, and possibly stones, though this last is doubtful: we find a sacrificial meal at the Latin festival, but no sign that he takes part in it as an image or statue. Elsewhere, as at Praeneste, peculiar representations of him arouse strong suspicions of foreign iconic influence. I think, on the whole, that the Italian peoples owed the sacred image to foreign works of art: and that the ‘epulum Iovis’ was introduced from Etruria by the Etruscan dynasty which built the Capitoline temple. It may, indeed, have been engrafted upon an earlier sacrificial meal like that of the feriae Latinae, or that of the curiae, or the rustic one of Jupiter dapalis: but, if so, the meal was one at which the ancient Romans were content to believe, as Ovid says, that the gods were present, and did not need, like the Greeks, the evidence of their eyes to help out their belief. Their gods were still aniconic when the wave of foreign ideas broke over them. We may say of the earliest Roman cult of Jupiter what Tacitus asserts of the Germans of his day^[1019]: ‘nec cohibere parietibus deos neque in ullam humani oris speciem adsimulare ex magnitudine caelestium arbitrantur: lucos ac nemora consecrant, deorumque nominibus appellant secretum illud quod sola reverentia vident.’

September 13 was also the day on which, according to Livy^[1020] and Verrius Flaccus^[1021], a nail (clavus) was driven annually by the ‘Praetor maximus’ into the wall of the cella of Minerva in the Capitoline temple, in obedience to an old lex which was fixed up on the wall of the temple adjoining this same cella. But Mommsen’s trenchant criticism^[1022] of the locus classicus for this subject in Livy has made it almost certain that the Roman scholars were here in error: that the ceremony was not an annual one, but took place once in a century, in commemoration of a vow made in 463 B.C., to commemorate the great pestilence of that year, which carried off both the consuls and several other magistrates^[1023]: that it had no special connexion with the cult of Jupiter, and was not intended, as is generally supposed, to mark the years as they passed. The nail is really the symbol of Fortuna or Necessitas; the rite was Etruscan, and was also celebrated at Volsinii in the temple of the Etruscan deity of Fate; when brought to Rome it was very naturally located in the great temple of the Etruscan Trias, the religious centre of the Roman state. Originally a dictator was chosen (i. e. Praetor maximus) clavi figendi causa; and when the dictatorship was dropped after the Second Punic War, the ceremony was allowed to fall into oblivion. Later on the Roman antiquarians unearthed and misinterpreted it, believing it to have been a yearly rite of which the object was to mark the succession of years. This brief account of Mommsen’s view may suffice for the purpose of this work: but the subject is one that might with advantage be reinvestigated.