^{page 304 note 1} See Müller's Rig-veda, vol. i., p. 4,Google Scholar lines 21ff.

^{page 308 note 1} [Though Prof. Róth does not consider it necessary to give instances in proof of his assertions, I may allude to the way in which Sâyaṇa considers the dwarfincarnation of Viṣhṇu to be referred to in R.V. i. 22Google Scholar 16ff., and identifies the Rudra of the hymns with the husband of Pârvatî; see his note on R.V. i. 114, 6;Google Scholar and Sanskrit Texts, iv. 57 and 257.Google Scholar Yâska, however, and the older authors referred to by him, Nir. xii. 19,Google Scholar seem to know, or, at least, they say, nothing of the dwarf-incarnation.—J.M.]

^{page 309 note 1} [The Nighaṇṭu ii., 14,Google Scholar contains no less than 122 verbs, to which the sense of going is assigned.—J.M.]

^{page 316 note 1} Compare “Sanskrit Texts,” iv. 13, note.Google Scholar

^{page 317 note 1} See “Contributions to a Knowledge of Vedic Theogony,” etc. in this Journal, for 1864, p. 116, note.Google Scholar

^{page 318 note 1} [As, for instance, in the cases viṣo vai pastyáḥ (Ṣ. P. Br. v. 4, 4, 5)Google ScholarPrajâ;patir vai Kaḥ, Ṣ. P. Br. vii. 4, 1, 19.Google Scholar—J.M],

^{page 319 note 1} Prof. Roth considers this vocabulary to be older than Yâska. (Introduction to Nirukta, , p. xii. f.)Google Scholar Müller, too, (Anc. Ind. Lit. 154), says, “probably these lists existed in his family long before his time.”

^{page 320 note 1} See all this more fully stated in Roth's Illustrations of the Nirukta, , p. 3.Google Scholar

^{page 320 note 2} See Roth's Illustrations of the Nirukta, pp. 220 ff.

^{page 320 note 3} Sâyaṇa also mentions some of these different schools of interpreters in different parts of his commentary. Thus on R.V. i. 64, 8,Google Scholar he says: Pṛshatyaḥ, the Maruts' instruments of conyeyance, are does marked with white spots according to the Aitihâsikas, and a line of variously coloured clouds according to the Nairuktas.” Again, he tells us that writers of the former class understood R.V. i. 174, 2,Google Scholar of the cities belonging to Vṛttra's Asuras, whilst those of the latter class understood it of the clouds. In like manner, on viii. 66, 10, he gives us two separate interpretations of that verse, the first that of the Nairuktas, who expounded it of natural phenomena, of showers brought by the sun (represented by Vishṇu), and the second that of the Aitihâsikas, who explained it mythologically in conformity with a story drawn from the Brâhmaṇa of the Charakas.

^{page 321 note 1} The ascetics, influenced, perhaps, by their own feelings of estrangement from family life, gave to the words in question the meaning “The father of many children surfers distress.” The Etymologists understood the same clause of the fructifying effects of rain.

^{page 321 note 1} Roth, Illustrations, pp. 221 f.

^{page 321 note 1} In Nir. iii. 15,Google Scholar several different derivations of the word vidhavâ, “widow,” are given. It is said to be either=vidhâtṛkâ, “without a supporter;” or, according to Charmaṣiras (one of Yâska's predecessors), to come from vidhavana or vidhâvana; or to be derived from vi + dhava, “without a man.”

^{page 323 note 1} See Nirukta, , i. 15 f.;Google Scholar Roth's Illustrations, pp. 11 f.; and “Sanskrit Texts,” ii. 181 ff.Google Scholar

^{page 323 note 2} “Sanskrit Texts,” ii. 184.Google Scholar

^{page 324 note 1} “Last Results of Sanskrit Researches” in “Bunsen's Christianity and Mankind,” vol. iii. p. 137.Google Scholar

^{page 324 note 2} Pâṇini, , pp. 225, 227.Google Scholar

^{page 324 note 3} Ibid 231, 233.

^{page 325 note 1} Prof. Wilson has misunderstood the latter of the two words when he translates it, “not spreading or dispersing.” See his note in loco.

^{page 326 note 1} It also occurs in ix. 72, 5; but I have no access to Sâyaṇa's Comm. on that passage.

^{page 326 note 2} See also R.V. i. 165, 5;Google Scholar vii. 56, 13.

^{page 326 note 3} Sâyaṇa here gives the following derivation of svadhâ, viz.: svam lokam dadhâtipushṇâti iti svadhâ. This word has three senses assigned to it in the Nighaṇṭtus, viz. “water” (i. 12), “food” (ii. 7), and “heaven and earth” in the dual (iii. 30).

^{page 326 note 4} The same general sense is assigned in i. 154, 4; v. 34, 1: vii. 47, 3. See also Sâyaṇa on i. 164. 38.

^{page 326 note 5} I am not aware that in any passage the chariots or horses of the gods are said to be yoked by food, as denoted by any word which certainly bears that sense. The horses of Indra are, indeed, represented as being yoked by prayer (brahmayuj) in R.V. i. 177, 2;Google Scholar iii. 35, 4; viii. 1, 24; viii. 2, 27; viii. 2, 17; and as being yoked by a hymn (vacho-yuj) in viii. 45, 39; but in these cases, generally, at least, the god is supposed to yoke his oar in consequence of this invitation to come and partake of the oblation, or libation, and not after partaking of it. It is true that the word brahman (neuter) has sometimes the sense of “food” or “oblation” ascribed to it, and that in two of the above texts, viii. 1, 24, and viii. 2, 27, one of the optional senses assigned by Sâyaṇa to brahma-yuj is, “yoked by our oblation,” two other senses, “yoked by the lord, Indra.” and “yoked by our hymn,” being proposed in the former case, and one alternative sense, “yoked by our hymn,” being proposed in the latter. In i. 177, 2; iii. 35, 4; and viii. 17, 2, however, “yoked by our hymn” (mantra) is the only rendering given. Vacho-yvj in viii. 45, 39,Google Scholar is explained, “yoked by our hymn.”

^{page 327 note 1} This and some other instances show that Sâyaṇa did occasionally resort to parallel passages for the elucidation of the text under his consideration, but he did not carry the-practice far enough.

^{page 327 note 2} Compare achakrebhiḥ in R.V. v. 42, 10,Google Scholar and nichakrayâ in viii. 7, 29.Google Scholar

^{page 333 note 1} The Sanskrit, scholar may also examine Yâska's desperate attempt (vi. 33) to explain the two words ṛdâpâ and ṛdûvṛdh, which occur in R.V. viii. 66, 11.Google Scholar Sâyaṇa merely repeats Yâska; but his text of the passage differs somewhat from Roth's.

^{page 339 note 1} Sâyaṇa's note on this verse (iv. 46, 1) affords another instance of his referring to a parallel text (R.V. vii. 92, 1)Google Scholar to prove that the first draught of soma was offered to Vâyu.

^{page 341 note 1} We have, in his comment on this verse, a further instance of Sâyaṇa quoting another passage for illustration, as he here cites i. 10, 5 as referring to Indra's function of command. See further on.

^{page 341 note 2} This verse is repeated in Sâma-veda i. 363, where Benfey renders it “vielgebietend,” “many ruling.”

^{page 345 note 1} The verb maṃh is found in Nigh. iii. 20,Google Scholar as signifying “to give.”

^{page 345 note 2} See on this word Roth”s Illust. of Nir. p. 39,Google Scholar where other passages in which it occurs are given. Roth mentions that Durga, the commentator on the Nirukta, says that, in the R.V., mehanâ is one word, whilst in the Sâmaveda it is considered to he made up of three. On the sense of the term see also Benfey's Gloss. to S.V., p. 151.Google Scholar

^{page 348 note 1} These cows belong to the Râkshasas, whom Sâyaṇa considers to be denoted by the word aktu, “night,” in which such spirits more about.

^{page 350 note 1} In vii. 61, 3, there is another instance of Sâyaṇa's making a reference back to a preceding passage, i. 61, 9, See also his notes on i. 154, 1; ii. 2, 5; iii. 17, 1; vi. 26, 4, vii. 76,4.

^{page 357 note 1} I should add that Sâyaṇa here offers alternative renderings both of asaṣchataḥ—making it a masc. pl. with the sense, “devoid of persons entering,”—and of the verb viṣrayantâm, which he says may be explained not only “let them be opened,” but “let them seek, or approach” (sevantâm).

^{page 360 note 1} Yâska quotes thia verse (Nir. iv. 4Google Scholar) and explains vidadvasu, by vittadhana, which may mean either “he by whom wealth is known,” or “by whom wealth has been obtained.”

^{page 363 note 1} on iv. 53, 4, he makes it=dhṛtakarmâ, “he by whom work is upheld.”

^{page 364 note 1} Compare R.V. viii. 28, 4,Google Scholar where a similar idea is expressed without the employment of the word vrata: Yathâ vaṣanti devâs tathâ id asat tad eshâm nakir â minat, etc., “As the gods wish, so it comes to pass; “no one hinders that [will] of theirs,” etc.

^{page 370 note 1} So, too, Mahîdhara on Vâj. S. xii. 34.Google Scholar

^{page 370 note 2} Both explanations seem to be wrong. Compare the words Vishṇurgopâḥ pararnam pâti pâthaḥ, in iii. 55, 10,Google Scholar where Sâyaṇa himself renders the last three words “guards the highest place,” though he adds an optional rendering of pâthaḥ as the “place of water, the atmosphere.”

^{page 371 note 1} There are other instances in the hymns of the sense running on from one verse into another. See vv. 5 and 6 of this same hymn, viii. 19, and viii. 12, 32 f. in Prof. Wilson's translation.

^{page 373 note 1} Sâyaṇa adds, “For mountains belong to Vishṇu as his own, as the Veda says ‘Vishṇu, is lord of the mountains.’“

^{page 378 note 1} In v. 11 of this hymn Sâyaṇa explains the pronoun tâ, “these,” as meaning the “sins committed in another birth,”—a further instance of his ascribing more modern notions to the Vedic age.

^{page 379 note 1} Sâyaṇa here refers, in illustration of one of his views, to another passage, i. 49, 4.

^{page 381 note 1} I adopt here Sâyaṇa's rendering of mitramahas, whether it be correct or not.

^{page 386 note 1} Prof. Wilson has the following remarks in a note on R.V. v. 2, 1:Google Scholar “According to what is no doubt the most accurate interpretation of this verse, and of those which follow, they contain only a metaphorically obscure allusion to the lighting of the sacrificial fire: the mother is the two pieces of touchwood, which retain fire, the child, and will not spontaneously give it up to the father, the yajamûna, until forced by attrition: till then, also, people, the priests, do not behold it, but they see it when bursting into ignition: this, however, has not satisfied the commentators, and a curious and strange legend has been devised for the interpretation of the text, or has been, perhaps, applied to it by way of explanation, having been previously current: it is more probably, however, suggested by, than suggestive of, the verses, ” etc., etc.

^{page 391 note 1} Atavavat does not mean “not spreading or dispersing” as Prof. Wilson translates it in p. 28, note 3, but “going,” atana-vat (not a-tana-vat).

^{page 391 note 2} Prof. “Wilson proposes, in p. 92, to take ayâtuḥ as the genitive of ayâtṛ, but I know of no such word as the latter with, the sense of “one not sacrificing”

^{page 391 note 3} In p. 114, Prof. Wilson proposes a translation of a word left unexplained by Sâyaṇa, svapivâta, which is founded on an analysis not sanctioned by the Padatext, as the latter divides the compound su + apivûta, whilson Wilson would divide it svapi + vûta.

^{page 392 note 1} In a noté to p. 193, Mr. Cowell corrects part of Prof. Wilson's translation of vii. 97, 6; but I do not see that the verse contains any word which can be rendered “friendship”

^{page 394 note 1} In explanation of this legend Sâyaṇa refers to the Taittiriya Sanhitâ, vi. 1, 3, 6. The following is the passage referred to, which I qnote to show how little light it throws on the text of the R.V.:—Yajno dakshiṇûm abhyadhûyat | tûm samabhavat | tad Indro ’châyat | so ’manyata “yo vâ ito janishyale sa idaṁ bhavishyati” iti | tâṁ prâviṣat | tasyâ Indra evâjâyata | so ’mạnyata “yo vai mad ito ’paro janishyate sa idaṁ bhavishyati” iti | tasyâ anumṛṣya yonim âchhinat | sâ sâtavaṣâ ’bhavat | tat sâtavaṣâyai janma | tâm haste nyaveshṭayata | tâṁ mṛgeshu nyadadhât | sâ kṛshṇavishâṇâ ’bhavat | “Indrasya yonir asi mâ mâ hiṁsîr” iti | “Yajna (sacrifice) desired Dakshiṇâ, (largess). He consorted with her. Indra was apprehensive of this. He reflected: ‘whoever is born of her will be this.’ He entered into her. Indra himself was born of her. He reflected: ‘whoever is born of her besides roe will be this.’ Having considered, he cut open her womb. She produced a cow.” etc. No mention is made of his killing his father.

^{page 394 note 2} I should observe that the Brâmaṇas constantly speak of the gods and Asuras as being both the off spring of Prajâpati; as contending together (S. P. Br. v. 1, 1, 1;Google Scholar vi. 6, 2, 11; vi. 6, 3, 2); and even as being originally equal or alike (Sanskrit Texts, iv, 52Google Scholar). And to prove that even malignant spirits may be called “gods,” Prof. Roth, s.v. deva, quotes from the Taitt. Sanh. iii. 5, 4, 1,Google Scholar a verse to the effect: “May Agni preserve me from the gods (devâḥ), destroyers of sacrificers, stealers of sacrifices, who inhabit the earth;” and a second text from the A.V. iii. 15, 5:Google Scholar “Agni, do thou through the oblation repel the gods who are destroyers of happiness” (? sâtaghnaḥ).

^{page 395 note 1} The verse which is illustrated in this passage occurs both in R.V i 164, 50,Google Scholar and in R.V. x. 90, 16,Google Scholar as well as Vâj. S. 31, 16. The concluding words are yatra pûrve sûdhyûḥ santi derûḥ, “where (in the sky) are the former Sâdhyas, gods.” Yâska, as I mentioned above, tells us that the Nairuktas understood the Sâdhyas to be “the gods whose locality is the sky,” dyusthûno devagaṇah, whilst, according to a legend (âkhyâna), the term denoted a former age of the gods.” Prof. Wilson translates the word Sâdhyâḥ by, “who are to be propitiated,” a sense not assigned by Sâyana, who proposes, first, that of sûdhanû yajnûdi-sâdhanavantaḥ karmadevûḥ, “performers, performers of sacrifices, etc., work-gods.” These words are rendered by Prof. Wilson in his note on i. 164, 50, “divinities presiding over or giving effect to religious acts” This does not, however, appear to be the real sense, as Mahîdhara on Vâj. S. 31, 17, tells us that “there are two kinds of gods, karmadevûḥ, “work-gods,” and âjânadevûḥ, “gods by birth,” the first being those who had attained to the condition of deities by their eminent works, and the second those who were produced at the beginning of the creation. The second class is superior to the first, and, according to the Brihadâraṇyaka, a hundred enjoyments of the latter (the work-gods), “are only equal to one single enjoyment of the former.” See all this and more declared in the Bṛhadâraṇyaka Upanishad, pp. 817 ff. (p. 230 f. of translation), and Ṣatapatha Brâhmaṇa, p. 1087.Google Scholar The second sense proposed for sâdhyâḥ by Sâyaṇa on R.V. i. 164, 50,Google Scholar is that of the “deities presiding over metres,” chhaṇdo'bhimâninaḥ, who, according to a Brâhnaṇa, by worshipping Agni were exalted to heaven, and became Âdityas and Angirases. Prof. Wilson remarks in his note: “It would seem that in Sâyaṇa's day the purport of the designation Sâdhya had become uncertain” Mahîdhara on Vâj. S. 31, 16, renders the term virâḍ-upûdhi-sâdhakâḥ, “producers of the condition of Virâj”

^{page 397 note 1} In regard to Indian tradition Prof. Benfey remarks as follows in note 450 to his translation of R.V. i. 61, 5,Google Scholar in his Orient und Occident: “If we compare the Indian interpretation, we recognize, as we have so often to do, how extremely little value we ought to attach to Indian explanations of words. On the other hand the correct explanation of things seems often to have been handed down, and such appears to be the case in the present instance.”

On i. 61, 7, the same writer observes, note 614: “This is a strophe which is perhaps the best calculated to show how little use can be made of Indian tradition for the understanding of the Vedas, or rather how greatly it misunderstood them.’

^{page 399 note 1} In as far as Sâyaṇa was in the habit of confining; his view to the single text hefore him (which I admit was not always the case) the following curious (Nirukta pariṣishṭa 1, 12) which gives a just view of the principles of Vedic interpretation, might seem to have been written with a prophetic reference to his case, and conveys a lesson not altogether inapplicable even to Christian divines, who have been too much in the habit of expounding their sacred texts without reference to the connection. “This reflective deduction of the sense of the verses is effected by the help both of oral tradition and reasoning. The verses are not to be interpreted singly, but according to the context. For one who is not a rishi or a tapasvin has no intuitive insight into their meaning … When the rishis were departing, men said to the gods, ‘Who shall be our rishi?’ The gods gave them this reasoning for a rishi,” etc., etc.