These heroes having been called originally Moliōnes or Molionidae, i. e. pounders, were afterwards fabled to have been the sons of Molione, the mill, and Aktōr, the corn-man. Some mythologists have identified these twins with thunder and lightning, and it is curious that the name of Thor's thunderbolt should be derived from the same root; for the hammer of Thor Miölnir 2 means simply the smasher. Again, among the Slavonic tribes, molnija is a name for lightning; and in the Serbian songs Munja is spoken of as the sister of Grom, the thunder, and has become a mythological personage.

Besides these heroic millers, there is another pair of Greek giants, known by the name of Aloadae, Otos and Ephialtes. In their pride they piled Ossa on Olympus, and Pelion on Ossa, like another Tower of Babel, in order to scale the abode of the gods. They were defeated by Apollo. The name of these giants has much the same meaning as that of the Moliones. It is derived from alōé', a threshing-floor, and means threshers. The question, then, is whether aloé', threshing-floor, and aleuron and tà áleura, wheatflour, can be traced back to the root mal. It is sometimes said that Greek words may assume an initial m for euphony's sake. That has never been proved. But it can be proved by several analogous cases that Greek words, originally beginning with m, occasionally drop that m. This, no doubt, is a violent change, 1 Friedreich, Realien in der Iliade und Odyssee, p. 562. Preller, Grie chische Mythologie, ii. 165.

2 Grimm, Deutsche Mythologie, 164, 1171. "The holy mawle" (maul, maillet, malleus) is referred by Grimm to the hammer of Thor. "The holy mawle, which they fancy hung behind the church-door, which, when the father was seaventie, the sonne might fetch to knock his father on the head, as effete and of no more use."- Haupt's Zeitschrift, v. 72.

and a change apparently without any physiological necessity, as there is no more difficulty in pronouncing an initial m than in pronouncing an initial vowel. However, there is no lack of analogies; and by analogies we must be guided. Thus móschos, a tender shoot, exists also as 6schos or 6sche, a young branch. Instead of mía, one, in the feminine, we find ía in Homer. Nay, instead of our very word aleuron, wheaten flour, another form, máleuron, is mentioned by Helladius.1 Again, if we compare Greek and Latin, we find that what the Romans called mola· namely, meal, or rather the grits of spelt, coarsely ground, which were mixed with salt, and thus strewed on the victims at sacrifices were called in Greek oular or olaí, though supposed to be barley instead of spelt.2 On the strength of these analogies we may, I believe, admit the possibility of an initial m being dropped in Greek, which would enable us to trace the names both of the Moliones and Aloadae back to the root mar. And if the Moliones and Aloadae3 derive their names from the root mar, we can hardly doubt that Mars and Ares, the prisoner of the Aloadae, came both from the same source. In Sanskrit the root mar yields Marut, the storm, literally the pounder or smasher; and in the

1 μwλwp, a weal, seems connected with ovλaí, scars.

2 Cf. Buttmann, Lexilogus, p. 450.

4

8 Otos and Ephialtes, the wind (vâta) and the hurricane.

4 Professor Kuhn takes Marut as a participle in at, and explains it as dying or dead. He considers the Maruts were originally conceived as the souls of the departed, and that because the souls were conceived as ghosts, or spirits, or winds, the Maruts assumed afterwards the character of stormdeities. Such a view, however, finds no support in the hymns of the Veda. In Pilumnus, the brother of Picumnus, both companions of Mars, we have a name of similar import, viz. a pounder. Jupiter Pistor, too, was originally the god who crushes with the thunderbolt (Preller, Rö

character of the Maruts, the companions of Indra in his daily battle with Vṛitra, it is easy to discover the germs of martial deities. The same root would fully explain the Latin Mars, Martis, and, considering the uncertain character of the initial m, the Greek Áres, Áreōs. Marmar and Marmor, old Latin names for Mars, are reduplicated forms; and in the Oscan Mamers the r of the reduplicated syllable is lost. Mavors is more difficult to explain,2 for there is no instance in Latin of m in the middle of a word be

ing changed into v. But although, etymologically, there is no difficulty in deriving the Indian name Marut, the Latin name Mars, and the Greek name Ares, from one and the same root,3 there is certainly neither in the legends of Mars nor in those of Ares any very distinct trace of their having been representatives of the storm. Mars at Rome and Ares in Thracia, though their worship was restricted to small territories, both assumed there the character of supreme tutelary deities. The only connecting mische Mythologie, p. 173), and the Mola Martis seem to rest on an analogous conception of the nature of Mars.

1 The suffix in Mars, Martis, is different from that in Marut. The Sanskrit Marut is Mar-vat; Mars, Martis, is formed like pars. partis, which happens to correspond with Sanskrit par-us or par-van. The Greek Árēs is again formed differently, but the Æolic form, Áreus, would come nearer to Marut. — Kuhu, Zeitschrift, i. 376.

2 See Corssen, in Kuhn's Zeitschrift, ii. 1–35.

8 That Marut and Mars were radically connected, was first pointed out by Professor Kuhn, in Haupt's Zeitschrift, v. 491; but he derived both words from mar in the sense of dying. Other derivations are discussed by Corssen, in Kuhn's Zeitschrift, ii. 1. He quotes Cicero (Nat. Deor. ii. 28): "Jam qui magna verteret Mavors;" Cedrenus (Corp. Byz. Niebuhr, t. i. p. 25, 21 f.): ὅτι τὸν Μάρτεμ οἱ Ῥωμαῖοι μόρτεμ ἐ κάνουν οἱονεὶ θάνατον, ἢ κινητὴν τῶν τεχνῶν, ἢ τὸν παρ ̓ ἀῤῥένων καὶ μόνων τιμώμενων : Varro (L. L. v. § 73, ed. O. Müller). "Mars ab eo quod maribus in bello præest, aut quod ab Sabinis acceptus, ibi est Mamers." See also Leo Meyer, in Kuhn's Zeitschrift, v. 387.

link between the classical deities Mars and Ares and the Indian Maruts is their warlike character; and if we take Indra as the conqueror of winter, as the destroyer of darkness, as the constant victor in the battle against the hostile powers of nature, then he, as the leader of the Maruts, who act as his army, assumes a more marked similarity with Mars, the god of spring, the giver of fertility, the destroyer of evil.1 In Ares, Preller, without any thought of the relationship between Ares and the Maruts, discovered the personification of the sky as excited by storm.2

We have hitherto examined the direct offshoots only of the root mar, but we have not yet taken into account the different modifications to which that root itself is liable. This is a subject of considerable importance, though at the same time beset with

"Endlich deuten aber

1 See Preller, Römische Mythologie, pp. 300, seq. 2 Preller, Griechische Mythologie, pp. 202, 203. auch verschiedene bildliche Erzählungen in der Ilias eine solche Naturbeziehung an, besonders die Beschreibung der Kämpfe zwischen Ares und Athena, welche als Göttin der reinen Luft und des Aethers die natürliche Feindin des Ares ist, und gewöhnlich sehr unbarmherzig mit ihm umgeht. So II. v. 583 ff., wo sie ihn durch Diomedes verwundet, Ares aber mit solchem Getöse niederrasselt (ẞpaxe), wie neuntausend oder zehntausend Männer in der Schlacht zu lärmen pflegen, worauf er als dunkles Gewölk zum Himmel emporfährt. Ebenso II. xxi. 400 ff., wo Athena den Ares durch einen Steinwurf verwundet, er aber fällt und bedeckt sieben Morgen Landes im Fall, und seine Haare vermischen sich mit dem Staube, seine Waffen rasseln: was wieder ganz den Eindruck eines solchen alten Naturgemäldes macht, wo die Ereignisse der Natur, Donnerwetter, Wolkenbruch, gewaltiges Stürmen und Brausen in der Luft als Acte einer himmlischen Göttergeschichte erscheinen, in denen gewöhnlich Zeus, Hera, Athena, Hephästos, Ares und Hermes als die handlenden Personen auftreten. Indessen ist diese allgemeine Bedeutung des Ares bald vor der speciellen des blutigen Kriegsgottes zurückgetreten." See also Il. xx. 51.

Αίε δ' Αρης ἑτέρωθεν, ἐρεμιῇ λαίλαπι ίσος. — Il. ix. 4.

Ως δ' ἄνεμοι δύο πόντον ὀρίνετον ἰχθυόεντα,

Βορέης καὶ Ζέφυρος, τώ τε Θρήκηθεν ἄητον.

greater difficulties and uncertainties. I stated in a former Lecture that Hindu grammarians have reduced the whole wealth of their language to about 1700 roots. These roots once granted, there remained not a single word unexplained in Sanskrit. But the fact is that many of these roots are clearly themselves derivatives. Thus, besides yu, to join, we found yuj, to join, and yudh, to join in battle. Here j and dh are clearly modificatory letters, which must originally have had some meaning. Another root, yauț, in the sense of joining or gluing together, must likewise be considered as a dialectic variety of yuj.

As

Let us apply this to our root MAR. As yu forms yudh, so mar forms mardh or mṛidh, and this root exists in Sanskrit in the sense of destroying, killing; hence mridh, enemy.1

Again, as yu produces yuj, so mar produces marj or mrij. This is a root of very common occurrence. It means to rub, but not in the sense of destroying, like mridh, but in the sense of cleaning or purifying. This is its usual meaning in Sanskrit, and it explains the Sanskrit name for cat, namely, márjára, literally the animal that always rubs or cleans itself. In Greek we find omórg-ny-mi in the same sense. But this general meaning became still more defined in Greek, Latin, German, and Slavonic, and by changing r into the root malg was formed, meaning to rub or stroke the udder of the cow, i. e. to milk. Thus mélg, and amélgō, in Greek, mean to milk; in Latin, mulgere has the same meaning In Old High-German we find the substantive milchu, 1 Rr. vi. 53. 4. "ví mṛídhaḥ jahi," kill the enemies.