a thousand. From this Varro concludes that of words derived in various ways from these there were about fifty thousand. He then goes on to explain the sense he and his predecessor give to primigenia verba. 'Primigenia dicuntur verba ut lego, scribo, sto, sedeo, et cetera quae non sunt ab alio quo verbo, sed suas habent radices', i. e. they are not grafted on or propagated from other words, but grow from their own root. Contra verba declinata sunt quae ab alio quo oriuntur, ut ab lego legis legit legam, et sic indidem hinc permulta.' He does not mean that primigenia verba are what we call roots, of which of course he knew nothing; but what he believed to be the original forms, later modified by the ingenuity of man. As with the primigenia semina, the notion is unscientific, but we can understand clearly enough what he means. I may add here that the pedantic Emperor Claudius, when he doubled two of his legions (XV and XXII), gave to the original ones of these numbers the title primigenia.

Festus, representing Verrius Flaccus, had a gloss on primigenius sulcus, now hopelessly mutilated (Lindsay, p. 270), but Paulus preserved it in this form: Primigenius sulcus dicitur, qui in condenda nova urbe tauro et vacca designationis causa imprimitur.' '1 This must mean the first furrow made by the plough round the urbs that was to be, which furrow was to be enlarged afterwards into a foss. This, though not so useful as the other usages, tends to confirm what was said above of the active or dynamical meaning in the word; seeds, words, foss, are all developed out of an original to which the epithet can be applied.

Lastly, Arnobius in the fourth century A. D. writes of primigenios ortus (ii. 61 and 70); but his text is uncertain and his meaning not very clear. I do not think that he throws any light on the meaning of the word; and I know of no other occurrence of it in the least likely to help us. The meaning as Varro understood it is perfectly clear. If we apply it to Fortuna primigenia, the meaning of the title would seem to be, the Fortuna who (or whose cult) was the 1 1 I do not know why Jordan, p. 6, says that this 'e Varronis libris fluxit '.

original of all the Fortunae afterwards suggested or devised by man's experience, the one that gave rise to the whole series. And it is not impossible that this may really have been the meaning. There were many Fortunae in Latium, among which she of Praeneste always retained her supremacy, or at any rate her claim to it. At Antium, where as at Praeneste she presided over something in the nature of an oracle, she had, so far as we know, no cult-titles. But since Jordan wrote his dissertation every one has followed him in translating (Fortuna) primigenia 'firstborn ’. I must now explain why this was.

I feel fairly sure that Jordan would not have thus translated the word, if it had not been for the discovery, three years before he wrote, of an inscription from Praeneste which runs thus (Dessau, no. 3684; C. I. L. xiv. 2863):

Orcevia Numeri nationu gratia Fortuna Diovo fileia Primogenia donom dedi.

Here Fortuna appears to our astonishment as the daughter of Jupiter, a position she claims nowhere else in Italy. There are indeed two later inscriptions at Praeneste in which she appears as Iovis puero (i. e. the daughter of Jupiter, by an old usage of the word puer, as it is generally understood); in one of these the word primigeniae follows, in the other it is absent. The Praenestines must have been most inconsistent people. The older inscription has the familiar filia, the later ones have the antique use of puer, meaning daughter. One man calls Fortuna primigenia, another does not. Still more astonishing is it that Cicero, who seems to have known something about the cults of Praeneste, declares that there actually was among them one of Jupiter puer, who was seated with Juno in the lap of his mother Fortuna, and was fondly worshipped by mothers (de Div. ii. 85; cf. Roman Festivals, p. 224). But this seems to be a case of confusion arising from a misinterpretation by Praenestines of statues and inscriptions, none of which were really primitive or of pure Italian origin. Even the oldest, the first quoted above, cannot well be earlier than the fourth century B. C., unless I am greatly mistaken,

and is probably still younger. It is hardly safe, where there is such confusion as this, to conclude that in Orcevia's dedication primigenia is to be taken with Diovo fileia, and understood in the sense of firstborn. Nor even if it must be so taken, as some will argue, should it be considered authoritative evidence of a common belief. Except this particular dedication, there is not only no ground for such a belief, but two special points which make against it.

1. It is certain that Fortuna is never mentioned or suggested by any Latin author as the daughter of Jupiter, and that no Latin deity was thought of, except after becoming Graecized, as the son or daughter of another deity. Praeneste was saturated with Greek influence, and Greek works of art were early introduced there; after which, we know not how, some notion must have got abroad such as is expressed in these inscriptions. But it never spread beyond Praeneste ; when the great goddess of that city was taken to dwell in a Roman temple in 194 B. C., she went as Fortuna primigenia simply,1 and no Roman ever suspected her of being Jupiter's child.

2. The word primigenius never means firstborn in Roman literature; that I hope I have already proved. If it had ever had this meaning, why was it not retained? why be content with the awkward' natu maximus'? The make of the word does not suggest this meaning to us, nor did it suggest it to Romans. When Cicero in his de Legibus, ii. 28, is discussing the various forms of Fortuna, and their particular objects, he simply says of Fortuna primigenia that she is a gignendo comes a new aspect of the deity indeed, and perhaps a fancy one, but with no relation whatever to firstborn children, or to the deity herself as any one's first born. Nor even at Praeneste (Div. ii. 85) does he know of primigenia in such a sense.

Without doubt Fortuna primigenia is the real title of the Praenestine deity, and until we have more explicit evidence we may let the Diovo fileia drop out. Without doubt too she

1 The hopelessly mutilated gloss in Festus (p. 272, Lindsay), might have solved some difficulties, as Verrius knew Praeneste.

was not called primigenia because she was a firstborn daughter. What the title really meant was, so far as I can see, unknown to the Romans, like many other such cult-titles, e. g. Gradivus as used of Mars. We ourselves can only guess at it, and I confess that I have no definite conviction. I have already said that if we hold to the meaning of the word as Varro understood it, we should think of a Fortuna who claimed to be the original deity of the name, before man began to interfere with her, adding artificial titles such as Fortuna huiusce diei, Fortuna muliebris, Fortuna virilis, and so on; as the first of the series, which gave rise to the rest, and therefore not standing in the relation of an eldest child to the younger

ones.

6

On the other hand, there may be something in the a gignendo comes '1 of Cicero. She may originally at Praeneste have been thought of as the equivalent of genius, the spiritual power which the Romans called Juno, attending on women as genius on men. This is what Mr. Mackail seems disposed to see in her. Or to use his own words in a letter to me, 'Fortuna primigenia would be the power which determines the whole movement of any life from its outset.' He refers to Iliad vi. 488–9, and Od. vii. 197-8; and there can be no objection in the case of a Praenestine cult in thus referring to Homer. This would fall in with the fact that at Praeneste and elsewhere Fortuna was specially worshipped by women, without doubt in special connexion with childbirth. Possibly she held at Praeneste the same place, or something like it, which Juno Lucina held at Rome and elsewhere; for Lucina, I believe, is not known to have inhabited Praeneste, where Juno was of secondary importance. It is worth noting that there was a Iunonarium, i. e. a cella or shrine, in the great temple of Fortuna there.

I have not yet mentioned that the cult-title primigenius is found in a few inscriptions applied to Hercules, both at Rome and in the province of Baetica (C. I. L. ii. 1436, 1545, 2463; and Dessau, no. 3433), and that Jordan claims (op. cit., p. 8) 1 If this reading be right, there is a lacuna immediately after comes.

that here too we are to understand it as firstborn (of Jupiter). I think that after what has been said above we may be sure that whatever it does mean in this instance, it certainly does not mean that.

PASSING UNDER THE YOKE

IN ancient Italy, when an army surrendered in the field, there were three alternatives before the victors for dealing with the vanquished. First, they might put them to death; there was nothing to prevent this but the feeling and tradition among Italian peoples in historical times against unnecessary bloodshed.1 Secondly, they might keep them as prisoners of war, and sell them as slaves; but in early times this was practically out of the question, partly owing to the difficulty of feeding and guarding them before they were sold, and partly because the machinery of sale, the slave-agents at hand and the slave-markets in the cities, had not then been invented. Thirdly, they might let their captives go free, with or without imposing conditions on them to be ratified by their State. This was really the simplest and easiest plan, and was adopted in the few cases recorded by Livy in which whole armies were captured. But before the vanquished were dismissed, they were made to go through the ceremony of passing under the yoke' (sub iugum missi), which Livy, when he first mentions it, explains as a kind of dramatized form of degradation. Two spears were fixed upright in the ground, and a third was fastened horizontally by each end to the tops of them; under this extemporized arch the conquered army had to pass, disarmed, and apparently wearing nothing but an under-garment, probably the subligaculum, in later times the dress rather of slaves than of soldiers or citizens. Livy's language is explicit: 'ut exprimatur

1 See Phillipson, International Law and Custom of Ancient Greece and Rome, ii. 253 ff. Liv. ix. 3 shows the feeling against bloodshed.

2 Liv. iii. 28; ix. 6; x. 36.