The mensis Martius stands alone among the Roman months. Not only was it the first in matters both civil and religious down to the time of Julius Caesar, but it is more closely associated with a single deity than any other, and that deity the protector and ancestor of the legendary founder of the city. It bears too the name of the god, which is not the case with any other month except January; and it is less certain that January was named after Janus than that March was named after Mars. The cult of Janus is not specially obvious in January except on a single day; but the cult of Mars is paramount all through March, and gives a peculiar character to the month’s worship.

It follows on a period which we may call one of purification, or the performance of piacular duties towards dead ancestors and towards the gods; and this has itself succeeded a time of general festivity in the homestead, the group of homesteads, the market, and the cross-roads. The rites of December and January are for the most part festive and social, those of February mystic and melancholy—characteristics which have their counterpart in the Christian Christmas, New Year, and Lent. The rites of March are distinct from those of either period, as we shall see. They again are followed by those of April, the opening month, which are gay and apt to be licentious; then comes the mensis Maius or month of growth, which is a time of peril for the crops, and has a certain character of doubt and darkness in its rites; lastly comes June, the month of maturity, when harvest is close at hand, and life begins to brighten up once more. After this the Roman months cease to denote by their names those workings of nature on which the husbandman’s fortune for the year depends.

By a process of elimination we can make a guess at the kind of ideas which must have been associated with the month which the Romans called Martius, even before examining its rites in detail. It is the time when the spring, whose first breath has been felt in February, begins to show its power upon the land^[61]. Some great numen is at work, quickening vegetation, and calling into life the powers of reproduction in man and the animals. The way in which this quickening Power or Spirit was regarded by primitive man has been very carefully investigated of recent years, and though the variation is endless both in myth and in ritual, we may now safely say that he was looked on as coming to new life after a period of death, or as returning after an absence in the winter, or as conquering the hostile powers that would hinder his activity. Among civilized peoples these ideas only survive in legend or poetry, or in some quaint bit of rural custom, often semi-dramatic, which may or may not have found its way into the organized cults of a city state of Greece or Italy, or even into the calendar of a Christian Church. But when these survivals have been collected in vast numbers both from modern Europe and from classical antiquity, and compared with the existing ideas and practices of savage peoples, they can leave no doubt in our minds as to the general character of the primitive husbandman’s conception of the mysterious power at work in spring-time.

It was this Power, we can hardly doubt, that the Latins knew by the name of Mars, the god whose cult is so prominent throughout the critical period of the quickening processes. We know him in Roman literature as a full-grown deity, with characteristics partly taken from the Greeks, partly extended and developed by a state priesthood and the usage of a growing and cosmopolitan city. We cannot trace him back, step by step, to his earliest vague form as an undefined Spirit, Power, or numen; it is very doubtful whether we can identify him, as mythologists have often done, with anything so obvious and definite as the sun, which by itself does not seem to have been held responsible by primitive peoples for the workings of nature at this time of year. We do not even know for certain the meaning of his name, and can get no sure help from comparative philology. Nevertheless there is a good deal of cumulative evidence which suggests a comparatively humble origin for this great god, some points of which we shall meet with in studying his cult during the month. The whole subject has been worked up by Roscher in the article on Mars in his Mythological Lexicon, which has the great advantage of being based on an entire re-examination of the Mars-cult, which he had handled in an earlier essay on Apollo and Mars.

Kal. Mart. (March 1). NP.

FERIAE MARTI. (PRAEN.)
N MARTIS. (PHILOC.)
IUN[O]NI LUCINAE E[S]QUILIIS QUOD EO DIE AEDES EI [DEDICA]TA EST PER MATRONAS QUAM VOVERAT ALBI[NIA] ... VEL UXOR ... SI PUERUM ... [AT]QUE IPSA[M].... (PRAEN.)

This was the New Year’s day of the Roman religious calendar. From Macrobius^[62] we learn that in his day the sacred fire of Vesta was now renewed, and fresh laurels fixed on the Regia, the Curiae, and the houses of the flamens; the custom therefore was kept up long after the first of March had ceased to be the civil New Year. Ovid alludes to the same rites, and adds the Aedes Vestae as also freshly decorated^[63]:

The mention of these buildings carries us back to the very earliest Rome, when the rex and his sons and daughters^[64] (Flamines and Vestales, in their later form) performed between them the whole religious duty of the community; to these we may perhaps add the warrior-priests of Mars (Salii). The connexion of the decoration with the Mars-cult is probable, if not certain; the laurel was sacred to Mars, for in front of his sacrarium in the regia there grew two laurels^[65], and it has been conjectured that they supplied the boughs used on this day^[66].

March 1 is also marked in the calendar of Philocalus as the birthday of Mars (N = natalis Martis). This appears in no other calendar as yet discovered, and is conspicuously absent in the Fasti Praenestini; it is therefore very doubtful whether any weight should be given to a fourth-century writer whose calendar had certainly an urban and not a rustic basis^[67]. There is no trace of allusion to a birth of Mars on this day in Latin literature, though the day is often mentioned. There was indeed a pretty legend of such a birth, told by Ovid under May 2^[68], which has its parallels in other mythologies; Juno became pregnant of Mars by touching a certain flower of which the secret was told her by Flora:

Of this tale Preller remarked long ago that it has a Greek setting: it is in fact in its Ovidian form a reflex from stories such as those of the birth of Athena and of Kora. Yet it has been stoutly maintained^[69] that it sprang from a real Italian germ, and is a fragment of the lost Italian mythology. Now, though it is certainly untrue that the Italians had no native mythology, and though there are faint traces, as we shall see, of tales about Mars himself, yet the Latins at least so rarely took these liberties with their deities^[70], that every apparent case of a divine myth needs to be carefully examined and well supported. In this case we must conclude that there is hardly any evidence for a general belief that March 1 was the birthday of Mars; and that Ovid’s story of Juno and Mars must be looked on with suspicion so far as these deities are concerned.

The idea that Mars was born on March 1 might arise simply from the fact that the day was the first of his month and also the first of the year. It is possible however to account for it in another way. It was the dies natalis of the temple of Juno Lucina on the Esquiline, as we learn from the note in the Fasti Praenestini; and this Juno had a special power in childbirth. The temple itself was not of very ancient date^[71], but Juno had no doubt always been especially the matrons’ deity, and in a sense represented the female principle of life^[72]. To her all kalends were sacred, and more especially the first kalends of the year, on which we find that wives received presents from their husbands^[73], and entertained their slaves. In fact the day was sometimes called the Matronalia^[74], though the name has no technical or religious sense. Surely, if a mother was to be found for Mars, no one could be more suitable that Juno Lucina; and if a day were to be fixed for his birth, no day could be better than the first kalends of the year, which was also the dedication-day of the temple of the goddess. At what date the mother and the birthday were found for him it is impossible to discover. The latter may be as late as the Empire; the former may have been an older invention, since Mars seems to have been apt to lend himself, under Greek or Etruscan influence, somewhat more easily to legendary treatment than some other deities^[75] But we may at any rate feel pretty sure that it was the Matronalia on March 1 that suggested the motherhood of Juno and the birth of Mars; and we cannot, as Roscher does, use the Matronalia to show that these myths were old and native^[76].

Yet another legend was attached to this day. It was said that the original ancile, or sacred shield of Mars, fell down from heaven^[77], or was found in the house of Numa^[78], on March 1. This was the type from which were copied the other eleven belonging to the collegium of Salii Palatini; in the legend the smith who did this work was named Mamurius, and was commemorated in the Salian hymn^[79]. These are simply fragments of a tangle of myth which grew up out of the mystery attaching to the Salii, or dancing priests of Mars, and to the curious shields which they carried, and the hymns which they sang^[80]; in the latter we know that the word Mamuri often occurred, which is now generally recognized as being only a variant of the name Mars^[81]. We shall meet with the word again later in the month. This also was the first day on which the shields were ‘moved,’ as it was called; i. e. taken by the Salii from the sacrarium Martis in the Regia^[82], and carried through the city in procession. Dionysius (ii. 70) has left us a valuable description of these processions, which continued till the 24th of the month; the Salii leaped and danced, reminding the writer of the Greek Curetes, and continually struck the shields with a short spear or staff^[83] as they sang their ancient hymns and performed their rhythmical dances.

The original object and meaning of all these strange performances is now fairly well made out, thanks to the researches of MÜllenhoff, Mannhardt, Roscher, Frazer and others. Roscher, in his comparison of Apollo and Mars^[84], pointed out the likeness in the spring festivals of the two gods. At Delphi, at the Theophania (7th of Bysios = March), there were decorations, sacrifices, dances, and songs; and of these last, some were ???? ???t????, or invocations to the god to appear, some pa???e?, or shouts of encouragement in his great fight with the dragon, or perhaps intended to scare the dragon away. For Apollo was believed to return in the spring, to be born anew, and to struggle in his infancy with the demon of evil. At other places in Greece similar performances are found; at Delos^[85], at Ortygia^[86] near Ephesus, at Tegyra, and elsewhere. At Ortygia the ?????te? stood and clashed their arms to frighten away Hera the enemy of Apollo’s mother Leto, in the annual dramatic representation of the perilous labour of the mother and the birth of her son. These practices (and similar ones among northern peoples) seem to be the result of the poetical mythology of an imaginative race acting on still more primitive ideas. From all parts of the world Mr. Frazer has collected examples of rites of this kind occurring at some period of real or supposed peril, and often at the opening of a new year, in which dances, howling, the beating of pots and pans, brandishing of arms, and even firing of guns are thought efficacious in driving out evil spirits which bring hurt of some kind to mankind or to the crops which are the fruits of his labour^[87]. This notion of evil spirits and the possibility of expelling them is at the root of the whole series of practices, which in the hands of the Greeks became adorned with a beautiful mythical colouring, while the Romans after their fashion embodied them in the cult of their city with a special priesthood to perform them, and connected them with the name of their great priest king.

In an elaborate note^[88] Mr. Frazer has attempted to explain the rites of the Salii in the light of the material he has collected. He is inclined to see two objects in their performances: (1) the routing out of demons of all kinds in order to collect them for transference to the human scapegoat, Mamurius Veturius (see below on March 14), who was driven out a fortnight later; and (2) to make the corn grow, by a charm consisting in leaping and dancing, which is known in many parts of the world. It will perhaps be safer to keep to generalities in matters of which we have but slender knowledge; and to conclude that the old Latins believed that the Spirit which was beginning to make the crops grow must at this time be protected from hostile demons, in order that he might be free to perform his own friendly functions for the community. Though the few words preserved of the Salian hymns are too obscure to be of much use^[89], we seem to see in them a trace of a deity of vegetation; and the prayer to Mars, which is given in Cato’s agricultural treatise, is most instructive on this point^[90].

The Salii in these processions were clothed in a trabea and tunica picta^[91], the ‘full dress’ of the warrior inspired by some special religious zeal, wearing helmet, breastplate, and sword. They carried the ancile on the left arm, and a staff or club of some kind to strike it with^[92]. At certain sacred places they stopped and danced, their praesul giving them the step and rhythm; and here we may suppose that they also sang the song of which a few fragments have come down to us, where the recurring word Mamurius seems beyond doubt to be a variant of Mars^[93]. Each evening they rested at a different place—mansiones Saliorum, as they were called—and here the sacred arms were hung up till the next day, and the Salii feasted. They were twenty-four in number, twelve Palatini and twelve Collini (originally Agonales or Agonenses), the former specially devoted to the worship of Mars Gradivus, the latter to that of Quirinus^[94]. The antiquity of the priesthood is proved by the fact that the Salii must be of patrician birth, and patrimi and matrimi (i. e. with both parents living) according to the ancient rule which descended from the worship of the household^[95].

It has been suggested that the shields (ancilia) which the Salii carried, being twelve in number for each of the two guilds, represented the twelve months of the year, either as twelve suns^[96] (the sun being renewed each month), or as twelve moons, which is a little more reasonable. This idea implies that the number of the Salii (which was the same as that of the Fratres Arvales) was based on the number of months in the year, which is very far from likely; it would seem also to assume that the shape of the shields was round, like sun or moon, which was almost certainly not the case. According to the legend, the original shield fell on the first new moon of the year; but it is quite unnecessary to jump to the conclusion that the others represent eleven other new moons. It would rather seem probable to a cautious inquirer that though an incrustation of late myth may have grown upon the Salii and their carmen and their curious arms, no amount of ingenious combination has as yet succeeded in proving that such myths had their origin in any really ancient belief of the Romans. What we know for certain is that there were twelve warrior-priests of the old Palatine city, and that they carried twelve shields of an antique type, which Varro compares to the Thracian peltae (L. L. 7. 43); shaped not unlike the body of a violin, with a curved indentation on each side^[97], which, when the shield was slung on the back, would leave space for the arms to move freely. In this respect, as in the rest of his equipment, the Salius simply represented the old Italian warrior in his ‘war-paint.’ In the examples of expulsion of evils referred to above as collected by Mr. Frazer, it is interesting to notice how often the expellers use military arms, or are dressed in military fashion. This may perhaps help us to understand how attributes apparently so distinct as the military and the agricultural should be found united in Mars and his cult.

Non. Mart. (March 7). F.

... [VEDI]OVI. ARTIS VEDIOVIS INTER DUOS LUCOS. (PRAEN.)

Various conjectures have been made for correcting this note. We may take it that the first word is rightly completed: some letters seem to have preceded it, and feriae has been suggested^[98], but not generally accepted. The next word, Artis, must be a slip of the stone-cutter. That it was not Martis we are sure, as Ovid says that there was no note in the Fasti for this day except on the cult of Vediovis^[99]. Even Mommsen is in despair, but suggests Aedis as a possibility, and that dedicata was accidentally omitted after it.

We do not know when the temple was dedicated^[100]. The cult of Vediovis seems to have no special connexion with other March rites: and it seems as well to postpone consideration of it till May 21, the dedication-day of the temple in arce. See also on Jan. 1.

vii Id. Mart. (March 9). C.

ARMA ANCILIA MOVENT. (PHILOC.)

As we have seen, the first ‘moving’ of the ancilia was on the 1st. This is the second mentioned in the calendars; the third, according to Lydus (4. 42), was on the 23rd (Tubilustrium, q.v.). As the Salii seem to have danced with the shields all through the month up to the 24th^[101], it has been supposed that these were the three principal days of ‘moving’; and Mr. Marindin suggests that they correspond to the three most important mansiones Saliorum, of which two were probably the Curia Saliorum on the Palatine and the Sacrarium Martis in the Regia^[102].

PRID. ID. MART. (MARCH 14). NP.

EQUIRR[IA]. (MAFF. VAT. ESQ.)
FERIAE MARTI. (VAT.)
SACRUM MAMURIO. (RUSTIC CALENDARS^[103].)
MAMURALIA. (PHILOC.)

These notes involve several difficulties. To begin with, this day is an even number, and there is no other instance in the calendar of a festival occurring on such a day. Wissowa^[104], usually a very cautious inquirer, here boldly cuts the knot by conjecturing that the Mars festival of this day had originally been on the next, i.e. the Ides, but was put back one day to enable the people to frequent both the horse-races (Equirria) and the festival of Anna Perenna^[105]. The latter, he might have added, was obviously extremely popular with the lower classes, as we shall see from Ovid’s description; and though the scene of it was close to that of the Equirria, or certainly not far away, it is not impossible that it may have diverted attention from the nobler and more manly amusement. Wissowa strengthens his argument by pointing out an apparent parallel between the festival dates of March and October. Here, as elsewhere, in the calendar, we find an interval of three days between two festivals, viz. between March 19 (Quinquatrus) and March 23 (Tubilustrium), and between Oct. 15 (‘October horse’) and Oct. 19 (Armilustrium). Now, as we shall see, the rites of March 19 and Oct. 19 seem to correspond to each other^[106]; and if there were a chariot-race on March 15, it would also answer to the race on the day of the ‘October horse,’ Oct. 15, with a three days’ interval as in October. The argument is not a very strong one, but there is a good deal to be said for it.

A much more serious difficulty lies in the discrepancy between the three older calendars in which we have notes for this day and the almanacs of the later Empire, viz. that of Philocalus (A.D. 354) and the rustic calendars. The former tell us of a Mars-festival, with a horse-race; the latter know nothing of these, but note a festival of Mamurius, a name which, as we saw, occurred in the Saliare Carmen apparently as a variant of Mars, and came to be affixed to the legendary smith who made the eleven copies of the ancile. How are we to account for the change of Mars into Mamurius, and of feriae Marti into Mamuralia? And are we to suppose that the later calendars here indicate a late growth of legend, based on the name Mamurius as occurring in the Carmen Saliare, or that they have preserved the shadow of an earlier and popular side of the March rites, which the State-calendars left out of account?

Apparently Mommsen holds the former opinion^[107]. In his note on this day he says that it is easy to understand how the second Equirria came to be known to the vulgus as Mamuralia (i.e. so distinguished from the first Equirria on Feb. 27), seeing that Mamurius who made the ancilia belongs wholly to the cult of Mars, and that this day was one of those on which the Salii and the ancilia were familiar sights in the streets of Rome. In other words, the Salian songs gave rise to the legend of Mamurius, and this in its turn gave a new name to the second Equirria or feriae Marti. And this I believe to be the most rational explanation of our difficulty, seeing that we have no mention of a feast of Mamurius earlier than the calendar of Philocalus in the fourth century A.D., which cannot be regarded as in any sense representing learning or research^[108].

But of recent years much has been written in favour of the other view, that the late calendars have here preserved for us a trace of very ancient Roman belief and ritual^[109]. This view rests almost entirely on a statement of a still later writer, Laurentius Lydus of Apamea, who wrote a work, de Mensibus, in the first half of the sixth century A.D., preserved in part in the form of two summaries or collections of extracts. Lydus was no doubt a man of learning, as is shown by his other work, de Magistratibus; but he does not give us his authority for particular statements, and his second- or third-hand knowledge must always be cautiously used.

Lydus tells us that on the Ides of March (a mistake, it is supposed^[110], for the 14th—which, however, he should not have made), a man clothed in skins was led out and driven with long peeled wands (out of the city, as we may guess from what follows) and shouted at as ‘Mamurius.’ Hence the saying, when any one is beaten, that they are ‘playing Mamurius with him.’ For the legend runs that Mamurius the smith was beaten out of the city because misfortune fell on the Romans when they substituted the new shields (made by Mamurius) for those that had fallen from heaven^[111].

This is clearly a late form of the Mamurius-myth: in all the earlier accounts^[112] only one ancile is said to have fallen from heaven. Lydus seems rather to be thinking of twelve original ones^[113], and twelve copies—perhaps of the Palatine and Colline ancilia respectively. If the form of the myth, then, is of late growth, suspicion may well be aroused as to the antiquity of the rite it was meant to explain, for with the older type of myth the rite does not seem to suit. And this suspicion is strengthened by the fact that in the whole of Latin literature there is no certain allusion to a rite so striking and peculiar, and only one that can possibly, even by forcible treatment, be taken as such. In Propertius v (iv.) 2. 61, we have the following lines, put into the mouth of the god Vertumnus:

Usener took this to mean, or to imply, that Mamurius was driven out of the city to its enemies the Oscans; but how we are to get this out of the words, which will bear very different interpretations, obscure as they are, it is not easy to see. And can we easily believe that, with this exception, no allusion should be found to the rite in either Latin or Greek writers—not in Ovid, Dionysius, Servius, Plutarch^[114], or in the fragments of Varro, Varrius, and others—if that curious rite had really been enacted year by year before the eyes of the Roman people? It certainly is not impossible that it may have slipped their notice, or have been mentioned in works that are lost to us; but it is so improbable as to justify us in hesitating to base conclusions as to the antiquity of the rite on the statement of Lydus alone.

There are indeed one or two passages which seem to prove that skins were used by the Salii, and that these skins were beaten. Servius^[115] says of Mamurius that they consecrated a day to him, on which ‘pellem virgis caedunt ad artis similitudinem,’ i. e. on which they imitate the smith’s art by beating a skin. So also Minucius Felix^[116]: ‘alii (we should probably read Salii) incedunt pileati, scuta vetera^[117] circumferunt, pelles caedunt.’ If we may judge by these passages of writers of the second century, there was something done by the Salii which involved the beating of skins; but if it was a skin-clad Mamurius who was beaten, why is he not mentioned, and why did they, as Servius says (and the context shows that he is speaking of him with all respect), set apart a day in his honour?

Yet Lydus’ account is so interesting from the point of view of folk-lore, that Usener was led by it into very far-reaching conclusions. These have been so well condensed in English by Mr. Frazer that my labour will be lightened if I may borrow his account^[118]:

‘Every year on March 14 a man clad in skins was led in procession through the streets of Rome, beaten with long white rods, and driven out of the city. He was called Mamurius Veturius^[119], that is, “the old Mars,” and as the ceremony took place on the day preceding the first full moon of the old Roman year^[120] (which began on March 1), the skin-clad man must have represented the Mars of the past year, who was driven out at the beginning of a new one. Now Mars was originally not a god of war, but of vegetation. For it was to Mars that the Roman husbandman prayed for the prosperity of his corn and vines, his fruit-trees and his copses; it was to Mars that the Arval Brothers, whose business it was to sacrifice for the growth of the crops, addressed their petitions almost exclusively.... Once more, the fact that the vernal month of March was dedicated to Mars seems to point him out as the deity of the sprouting vegetation. Thus the Roman custom of expelling the old Mars at the beginning of the New Year in spring is identical with the Slavonic custom of “carrying out Death^[121],” if the view here taken of the latter custom is correct. The similarity of the Roman and Slavonic customs has been already remarked by scholars, who appear, however, to have taken Mamurius Veturius and the corresponding figures in the Slavonic ceremonies to be representatives of the old year rather than of the old god of vegetation. It is possible that ceremonies of this kind may have come to be thus interpreted in later times even by the people who practised them. But the personification of a period of time is too abstract an idea to be primitive. However, in the Roman, as in the Slavonic ceremony, the representative of the god appears to have been treated not only as a deity of vegetation, but also as a scape-goat^[122]. His expulsion implies this; for there is no reason why the god of vegetation, as such, should be expelled the city. But it is otherwise if he is also a scape-goat; it then becomes necessary to drive him beyond the boundaries, that he may carry his sorrowful burden away to other lands. And, in fact, Mamurius Veturius appears to have been driven away to the lands of the Oscans, the enemies of Rome^[123].’

My examination of the evidence will, I hope, have made it clear why I hesitate to endorse these conclusions in their entirety (as I did for many years), interesting as they are. I rather incline to believe that the whole Mamurius-legend grew out of the Carmen Saliare, and that we may either have here one of those comparatively rare examples of later ritual growing itself out of myth, or a point of ancient ritual, such as the use of skins—perhaps those of victims—misinterpreted and possibly altered under the influence of the myth. As to Lydus’ statement, it is better to suspend our judgement; he may, for all we know, have confused some foreign custom, or that of some other Italian town where there were Salii, with the ritual of a Roman priesthood^[124]. In any case, his account is too much open to question to bear the weight of conjecture that has been piled upon it.

Id. Mart. (March 15). NP.

FERIAE^[125] ANNAE PERENNAE VIA FLAM[INIA] AD LAPIDEM PRIM[UM]. (VAT.)

ANNAE PER. (FARN.)

This is a survival of an old popular festival, as is clearly seen from Ovid’s account of it; but the absence of any mention of it in the rustic calendars or in those of Philocalus and Silvius leads us to suppose that it had died out in the early Empire. This may be accounted for by the fact that the people came to be more and more attracted by spectacles and games; and also by the ever-increasing cosmopolitanism of the city populace, which would be continually losing interest in old Roman customs which it could not understand.

On this day, Ovid tells us^[126], the ‘plebs’ streamed out to the ‘festum geniale’ of Anna Perenna, and taking up a position in the Campus Martius, not far from the Tiber^[127], and lying about on the grass in pairs of men and women, passed the day in revelry and drinking^[128]. Some lay in the open; some pitched tents and some constructed rude huts of stakes and branches, stretching their togas over them for shelter. As they drank they prayed for as many years of life as they can swallow cups of wine; meanwhile singing snatches of song with much gesticulation and dancing. The result of these performances was naturally that they returned to the city in a state of intoxication. Ovid tells us that he had seen this spectacle himself^[129].

Whether there was any sacrificial rite in immediate connexion with these revels we do not know. Macrobius indeed tells us^[130] that sacrifice was offered in the month of March to Anna Perenna ‘ut annare perannareque commode liceat’^[131]; and Lydus, that on the Ides there were e??a? d??s?a? ?p?? t?? ???e???? ?e??s?a? t?? ???a?t??; but we do not know what was the relation between these and the scene described by Ovid.

Who was the Anna Perenna in whose honour these revels, sacrifices, and prayers took place, whatever their relation to each other? Ovid and Silius Italicus^[132] tell legends about her which are hardly genuine Italian, and in which Anna Perenna is confused with the other Anna whom they knew, the sister of Dido. Hidden under such stories may sometimes be found traces of a belief or a cult of which we have no other knowledge; but in this poetical medley there seems to be only one feature that calls on us to pause. After her wanderings Anna disappears in the waters of the river Numicius:

Her companions traced her footsteps to the bank: she seemed to tell them

This tale led Klausen^[133] into some very strange fancies about the goddess, whom he regarded as a water-nymph, thinking that all her other characteristics (e.g. the year) might be explained symbolically; the running water representing the flow of time, &c. But it is probable that she only came into connexion with the river Numicius because Aeneas was there already. If Aeneas, as Jupiter Indiges, was buried on its banks^[134], what could be more natural than that another figure of the Dido legend should be brought there too? There does not indeed seem to be any reason for connecting the real Anna Perenna with water^[135]. All genuine Roman tradition seems to represent her, as we shall see directly, as an old woman; and when she appears in another shape, she must have become mixed up with other ideas and stories. It may perhaps be just possible that on this day some kind of an image of her may have been thrown into the Tiber, as was the case with the straw puppets (Argei) on May 15, and that the ceremony dropped out of practice, but just survived in the Numicius legend^[136]. But this is simply hypothesis.

The fact is that, whatever else Anna Perenna may have been, all that we can confidently say of her is that she represented in some way the circle or ring of the year. This is indicated not only by the name, which can hardly be anything but a feminine form of annus, but by the time at which her festival took place, the first full moon of the new year. The one legend preserved about her which is of undoubted Italian origin is thought to point in the same direction. Ovid, wishing to explain ‘cur cantent obscena puellae’ in that revel of the ‘plebs’ on the Tiber-bank, tells us^[137] how Mars, once in love with Minerva^[138], came to Anna and asked her aid. It was at length granted, and Mars had the nuptial couch prepared: thither a bride was led, but not the desired one; it was old Anna with her face veiled like a bride who was playing the passionate god such a trick as we may suppose not uncommon in the rude country life of old Latium.

There is no need to be startled at the rude handling of the gods in this story, which seems so unlike the stately and orderly ideas of Roman theology. It must be borne in mind that folk-tales like this need not originally have been applied to the gods at all. They are probably only ancient country stories of human beings, based on some rude marriage custom—stories such as delighted the lower farm folk and slaves on holiday evenings; and they have survived simply because they became in course of time attached to the persons of the gods, as the conception of divinities grew to be more anthropomorphic. Granted that Anna or Perenna^[139] was the old woman of the past year, that Mars was the god of the first month, and that the story as applied to human beings was a favourite one, we can easily understand how it came to attach itself to the persons of the gods^[140].

Yet another story is told by Ovid of an Anna^[141], in writing of whom he does not add the name Perenna. The Plebs had seceded to the Mons Sacer, and were beginning to suffer from starvation, when an old woman from Bovillae, named Anna, came to the rescue with a daily supply of rustica liba. This myth seems to me to have grown out of the custom, to be described directly, of old women^[142] selling liba on the 17th (Liberalia), the custom having been transferred to that day through an etymological confusion between liba and Liberalia. Usener, however, saw here a connexion between Anna and Annona^[143]; and recently it has been suggested that a certain Egyptian Anna, who is said by Plutarch to have invented a mould for bread-baking, may have found her way to Rome through Greek channels^[144].

XVI Kal. Apr. (March 17). NP.

LIB[ERALIA]. (MAFF. FARN. RUST.)
LIB. AG[ONIA]. LIBERO LIB. (CAER.)
AG[ONIA]. (VAT.)
LIBERO IN CA[PITOLIO]. (FARN.)

This is one of the four days marked AG. or AGON. in the Fasti (Jan. 9, May 21, Dec. 11)^[145]. It is curious that on this day two of the old calendars should mark the Liberalia only, and one the Agonia only, and one both. The day was generally known as Liberalia^[146]; the other name seems to have been known to the priests only, and more especially to the Salii Collini or Agonenses^[147], who must have had charge of the sacrifice. Wissowa seems to be right in thinking (de Feriis xii) that the conjunction of Liberalia and Agonia is purely accidental, and that the day took its common name from the former simply because, as the latter occurred four times in the year, confusion would be likely to arise.

Liber is beyond doubt an old Italian deity, whose true nature, like that of so many others, came to be overgrown with Greek ideas and rites. There is no sign of any connexion between this festival and the cult of Dionysus; hence we infer that there was an old Latin Liber before the arrival of the Greek god in Italy. What this god was, however, can hardly be inferred from his cult, of which we only know a single feature, recorded by Ovid^[148]. He tells us that old women, sacerdotes Liberi, sat crowned with ivy all about the streets on this day with cakes of oil and honey (liba), and a small portable altar (foculus), on which to sacrifice for the benefit of the buyer of these cakes. This tells us nothing substantial, and we have to fall back on the name—always an uncertain method. The best authorities seem now agreed in regarding the word Liber (whatever be its etymology) as having something of the same meaning as genius, forming an adjective liberalis as genius forms genialis, and meaning a creative, productive spirit, full of blessing, and so generous, free, &c.^[149] If this were so it would not be unnatural that the characteristics and rites of Dionysus should find here a stem on which to engraft themselves, or that Liber should become the object of obscene ceremonies which need not be detailed here, and also the god of the Italian vine-growers.

It is possible that Liber may have been an ancient cult-title of Jupiter; we do in fact find a Jupiter Liber in inscriptions, though the combination is uncommon^[150]. In that case Liber may have been an emanation or off-shoot from Jupiter, as Silvanus probably was from Mars^[151]. But I am disposed to think that the characteristics of Liber, so far as we know them, are not in keeping with those of Jupiter; and that the process was rather of the opposite kind, that is, the cult of Liber in its later form became attached to that of Jupiter, who was always the presiding deity of vineyards and wine-making^[152].

This was also the usual day on which boys assumed the toga virilis (toga recta, pura, libera):

We know indeed that in the late Republic and Empire other days were used for this ceremony: Virgil took his toga on Oct. 15, Octavian on Oct. 18, Tiberius on April 24, Nero on July 7^[154]; but it is likely that this day was in earlier times the regular one, in spite of the inconvenience of a disparity of age thence resulting amongst the tirones. For whether or no the toga libera has any real connexion with the Liberalia, this was the time when the army was called out for the year, and when the tirones would be required to present themselves^[155]. Ovid tells us that on this day the rustic population flocked into the city for the Liberalia, and the opportunity was doubtless taken to make known the list of tirones, as the boys were called when the toga was assumed and they were ready for military service.

They sacrificed, it appears, before leaving home and again on the Capitol, either to Pubertas or Liber, or both^[156].

On this day also, according to Ovid, and also on the previous one, some kind of a procession ‘went to the Argei’^[157]; by which word is meant, we may be almost sure, the Argeorum sacella. There were in various parts of the four regions of the Servian city a number of sacella or sacraria, which were called Argei, Argea, or Argeorum sacella. What these were we never shall know for certain; but we may be fairly sure that their number was twenty-four, six for each region; the same number as that of the rush puppets or simulacra also called Argei, which were thrown into the Tiber by the Vestal Virgins on May 15. The identity of the name and number leads to the belief that there was a connexion between these sacella and the simulacra; but the very difficult questions which arose about both must be postponed till we have before us the whole of the ceremonial, i. e. that of May 15 as well as that of March 17. About this last we know nothing and can at best attempt to infer its character from the ceremony in May, of which we fortunately have some particulars on which we can fully rely.

Kal. xiv Apr. (March 19). NP Caer. Vat. N. Maff.

QUINQ[VATRUS]. (CAER. MAFF. PRAEN. VAT. FARN.)
QUINQUATRIA. (RUST. PHIL. SILV.)

A note is appended in Praen., which is thus completed by Mommsen with the help of a Verrian gloss (Fest. 254).

[RECTIUS TAMEN ALII PUTARUNT DICTUM AB EO QUOD HIC DIES EST POST DIEM V IDUS. QUO]D IN LATIO POST [IDUS DIES SIMILI FERE RATIONE DECLI]NARENTUR.

FERIAE MARTI (VAT.)

[SALI] FACIUNT IN COMITIO SALTUS [ADSTANTIBUS PO]NTIFICIBUS ET TRIB[UNIS] CELER[UM]. Praen., in which we find yet another note: ARTIFICUM DIES [QUOD MINERVAE] AEDIS IN AVENTINO EO DIE EST [DEDICATA].

The original significance of this day is indicated by the note Feriae Marti in Vat., and also by that in Praen., which has been amplified with tolerable certainty. The Salii were active this day in the worship of Mars, and the scene of their activity was the Comitium. With this agrees, as Mommsen has pointed out, the statement of Varro^[158] that the Comitium was the scene of some of their performances, though he does not mention which. More light is thrown on the matter by the grammarian Charisius^[159], who, in suggesting an explanation of the name Quinquatrus by which this day was generally known, remarks that it was derived from a verb quinquare, to purify, ‘quod eo die arma ancilia lustrari sint solita.’ His etymology is undoubtedly wrong, but the reason given for it is valuable^[160]. The ancilia were purified on this day (perhaps by the Salii dancing around them), and thus it exactly answers to the Armilustrium on Oct. 19, just as the horse-races on the Ides of March, if that indeed were the original day, correspond to the ceremony of the ‘October horse’^[161].

The object and meaning of the lustratio in each case is not, however, quite clear. Since in March the season of war began, and ended, no doubt, originally in October^[162], and as the Salii seem to be a kind of link between the religious and military sides of the state’s life, we are tempted to guess that the lustration of the ancilia represented in some way the lustration of the arms of the entire host, or perhaps that the latter were all lustrated so as to be ready for use, on this day, and once again on Oct. 19 before they were put away for the winter. In this latter case the Salii would be the leaders of, as well as sharers in, a general purifying process. And that this is the right view seems to be indicated by Verrius’ note in the Praenestine calendar, from which it is clear that the tribuni celerum were present, and took some part in the ceremony. These tribuni were almost certainly the three leaders of the original cavalry force of the three ancient tribes, and they seem to have united both priestly and military characteristics^[163]; and from their presence in the Comitium may perhaps also be inferred that of the leaders of the infantry tribuni militum. In the earliest times, therefore, the arms of the whole host may have been lustrated in the presence of its leaders, the Salii, so to speak, performing the service; but in later times the Salii alone were left, and their arms alone lustrated, though possibly individuals representing the ancient tribuni celerum may have appeared as congregation.

But this day was generally known as Quinquatrus, simply because it was the fifth day after the Ides^[164]; i.e. there was a space of three days between the Ides and the festival. Such intervals of three days, either between the Ides and the festival or between one festival and another, occur several times in the Roman calendar^[165], though in this instance alone the day following the interval appears in the calendars as Quinquatrus. The term was no doubt a pontifical one, and the meaning was unknown to the common people; in any case it came to be misunderstood, and was in later times popularly applied to the four days following the festival as well as the festival itself; its first syllable being taken to indicate a five-day period instead of the fifth day after the Ides. This popular mistake led to still further confusion owing to a curious change in the religious character of these days, about the nature of which there can be no serious doubt.

The 19th came to be considered as sacred to Minerva^[166], because a temple to that goddess was consecrated on this day, on the Caelian or the Aventine, or possibly both^[167]. There is no obvious connexion between Mars and Minerva; and it is now thought probable that Minerva has here simply taken the place of another goddess, Nerio—one almost lost to sight in historical times, but of whose early connexion with Mars some faint traces are to be found. Thus where we find Minerva brought into close relation with Mars, as in the myth of Anna Perenna, it is thought that we should read Nerio instead of Minerva^[168]. This conclusion is strengthened by a note of Porphyrion on Horace Epist. ii. 2. 209 ‘Maio mense religio est nubere, et etiam Martio, in quo de nuptiis habito certamine a Minerva Mars victus est: obtenta virginitate Neriene est appellata.’ As Neriene must = Nerio^[169], this looks much like an attempt to explain the occurrence of two female names, Minerva and Nerio, in the same story; the original heroine, Nerio, having been supplanted by the later Minerva^[170].

Of this Nerio much, perhaps too much, has been made in recent years by ingenious scholars. A complete love-story has been discovered, in which Mars, at first defeated in his wooing, as Porphyrion tells us in the passage just quoted, eventually becomes victorious; for Nerio is called wife of Mars in a fragment of an old comedy by Licinius Imbrex, in a passage of Plautus, and in a prayer put into the mouth of Hersilia by Gellius the annalist, when she asked for peace at the hand of T. Tatius^[171]. And this story has been fitted on, without sufficient warrant, to the Mars-festivals of this month. Mars is supposed to have been born on the Kalends, to have grown wondrously between Kalends and Ides, to have fallen then in love with Nerio, to have been fooled as we saw by Anna Perenna, to have been rejected and defeated by his sweetheart, and finally to have won her as his wife on the 19th^[172]. Are we to find here a fragment of real Italian mythology, or an elaborate example of the Graecizing anthropomorphic tendencies of the third and second centuries B.C.?

The question is a difficult one, and lies rather outside the scope of this work. Those who have read Usener’s brilliant paper will find it hard to shake themselves free of the conviction that he has unearthed a real myth, unless they carefully study the chapter of Aulus Gellius which is its chief foundation. Such a study has brought me back to the conviction that Plautus and the others were writing in terms of the fashionable modes of thought of their day, and were not appealing to popular ideas of the relations of Italian deities to each other^[173]. Aulus Gellius begins by quoting a comprecatio from the book of the Libri sacerdotum populi Romani. ‘In his scriptum est: Luam Saturni, Salaciam Neptuni, Horam Quirini, Virites Quirini, Maiam Volcani, Heriem Iunonis, Moles Martis Nerienemque Martis.’ A glance at the names thus coupled together is enough to show that Mars is not here thought of as the husband of Neriene; the names Lua, Salacia, &c., seem rather to express some characteristic of the deity with whose name they are joined or some mode of his operation^[174]; and Gellius himself, working on an etymology of Nerio which has generally been accepted as correct, explains the name thus: ‘Nerio igitur Martis vis et potentia et maiestas quaedam esse Martis demonstratur.’ In the latter part of his chapter, after quoting Plautus, he says that he has heard the poet blamed by an eminent critic for the strange and false notion that Nerio was the wife of Mars; but he is inclined to think that there was a real tradition to that effect, and cites his namesake the annalist and Licinius Imbrex in support of his view.

But neither annalist nor play-writer can stand against that passage from the sacred books with which he began his chapter; and if we give the latter its due weight, the value of the others is relatively diminished. It appears to me that the one represents the true primitive Italian idea of divine powers, which with its abundance of names offered excellent opportunities to anthropomorphic tendencies of the Graecizing school, while the others show those tendencies actually producing their results. Any conclusion on the point must be of the nature of a guess; but I am strongly disposed to think (1) that Nerio was not originally an independent deity, but a name attached to Mars expressive of some aspect of his power, (2) that the name gradually became endowed with personality, and (3) that out of the combination of Mars and Nerio the Graecizing school developed a myth of which the fragments have been taken by Usener and his followers as pure Roman.

Having once been displaced by Minerva, Nerio vanished from the calendar, and with her that special aspect of Mars—whatever it may have been—which the name was intended to express. The five days, 18th to 23rd, became permanently associated with Minerva. The 19th was the dedication-day of at least one of her temples, and counted as her birthday^[175]: the 23rd was the Tubilustrium, with a sacrifice to ‘dea fortis,’ who seems to have been taken for Minerva, owing to an incorrect idea that the latter was specially the deity of trumpet-players^[176]. She was no doubt an old Italian deity of artificers and trade-guilds; but the Tubilustrium was really a Mars-festival, and Minerva had no immediate connexion with it.

x Kal. Apr. (March 23). NP.

TUBILUST[RIUM]. (CAER. MAFF. VAT. FARN. MIN. III.)
TUBILUSTRIUM. (PHILOC.)

Note in Praen.: [FERIAE] MARTI^[177]. HIC DIES APPELLATUR ITA, QUOD IN ATRIO SUTORIO TUBI LUSTRANTUR, QUIBUS IN SACRIS UTUNTUR. LUTATIUS QUIDEM CLAVAM EAM AIT ESSE IN RUINIS PALA[TI I]NCENSI A GALLIS REPERTAM, QUA ROMULUS URBEM INAUGURAVERIT.

Q. R. C. F. (VAT. CAER.)

Q. REX. C. F. (MAFF. PRAEN.)

Note in Praen.: HUNC DIEM PLERIQUE PERPERAM INTERPRETANTES PUTANT APPELLAR[I] QUOD EO DIE EX COMITIO FUGERIT [REX: N]AM NEQUE TARQUINIUS ABIIT EX COMITIO [URBIS], ET ALIO QUOQUE MENSE EADEM SUNT [IDEMQUE S]IGNIFICANT. QU[ARE COMITIIS PERACTIS IUDICI]A FIERI INDICA[RI IIS MAGIS PUTAMUS]^[178].

These two days must be taken in connexion with the 23rd and 24th of May, which are marked in the calendars in exactly the same way. The explanation suggested by Mommsen is simple and satisfactory^[179]; the 24th of March and of May were the two fixed days on which the comitia curiata met for the sanctioning of wills^[180] under the presidency of the Rex. The 23rd in each month, called Tubilustrium, would be the day of the lustration of the tubae or tubi used in summoning the assembly. The letters Q. R. C. F. (quando rex comitiavit fas) mean that on the days so marked proceedings in the courts might only begin when the king had dissolved the Comitia.

The tuba, as distinguished from the tibia, which was the typical Italian instrument, was a long straight tube of brass with a bell mouth^[181]. It was used chiefly in military^[182] and religious ceremonies; and as the comitia curiata was an assembly both for military and religious objects, this would suit well with Mommsen’s idea of the object of the lustration. The Tubilustrium was the day on which these instruments, which were to be used at the meeting of the comitia on the following day, were purified by the sacrifice of a lamb. Of the Atrium Sutorium, where the rite took place, we know nothing.

There are some words at the end of Verrius’ note in the Praenestine Calendar, which, as Mommsen has pointed out^[183], come in abruptly and look as if something had dropped out: ‘Lutatius quidem clavam eam ait esse in ruinis Pala[ti i]ncensi a Gallis repertam, qua Romulus urbem inauguraverit.’ This clava must be the lituus of Romulus, mentioned by Cicero^[184], which was found on the Palatine and kept in the Curia Saliorum. We cannot, however, see clearly what Verrius or his excerptor meant to tell us about it; there would seem to have been a confusion between lituus in the sense of baculum and lituus in the sense of a tuba incurva. The latter was in use as well as the ordinary straight tuba^[185]; in shape it closely resembled the clava of the augur, and perhaps the resemblance led to the notion that it was the clava of Romulus and not a tuba which was this day purified with the other tubae.

We can learn little or nothing from the calendar of this month about the origin of Mars, and we have no other sufficient evidence on which to base a satisfactory conjecture. But from the cults of the month, and partly also from those of October, we can see pretty clearly what ideas were prominent in his worship even in the early days of the Roman state. They were chiefly two, and the two were closely connected. He was the Power who must be specially invoked to procure the safety of crops and cattle; and secondly, in his keeping were the safety and success of the freshly-enrolled host with its armour and its trumpets. In short, he was that deity to whom the most ancient Romans looked for aid at the season when all living things, man included, broke into fresh activity. He represents the characteristics of the early Roman more exactly than any other god; for there are two things which we may believe with certainty about the Roman people in the earliest times—(1) that their life and habits of thought were those of an agricultural race; and (2) that they continually increased their cultivable land by taking forcible possession in war of that of their neighbours.