^139:1 Madhyandina text, p. 1076.

^139:2 This disputation between Yagnavalkya and Vidagdha Sakalya occurs in a simpler form in the Satapatha-brahmana, XI, p. 873. He is here represented as the first who defies Yagnavalkya, and whom Yagnavalkya asks at once, whether the other Brahmans had made him the ulmukavakshayana, the cat's paw, literally one who has to take a burning piece of wood out of the fire (ardha. dagdhakashtham ulmukam; tasya vahirnirasanam avakshayanam vinasah). The end, however, is different, for on asking the nature of the one god, the Prana, he is told by Yagnavalkya that he has asked for what he ought not to ask, and that therefore he will die and thieves will carry away his bones.

^139:3 Nivid, old and short invocations of the gods; devatasankhyavakakani mantrapadni kanikid vaisvadeve sastre sasyante. Sankara, and Dvivedaganga.

^139:4 This would make 3306 devatas.

^140:1 'The glories of these are three and thirty.' Gough, p. 172.

^140:2 Trayastrimsau, i.e. trayastrimsatah puranau.

^140:3 The etymological explanation of Vasu is not quite clear, and the commentator hardly explains our text. Perhaps vasu is meant for the world or the dwellers therein. The more usual explanation occurs in the Satap. Brah. p. 1077, ete hidam sarvam vasayante tadyad idam sarvam vasayante tasmad vasava iti; or on p. 874, where we read te yad idam sarvam &c.

^141:1 Atman is here explained as manas, the common sensory.

^141:2 The life of men, and the fruits of their work.

^141:3 They are the thirty-three gods.

^142:1 I prefer to attribute this to Sakalya, who is still the questioner, and not Yagnavalkya; but I am not quite satisfied that I am right in this, or in the subsequent distribution of the parts, assigned to each speaker. If Sakalya is the questioner, then the sentence, veda va aham tam purusham sarvasyatmanah parayanam yam attha, must belong to Yagnavalkya, because he refers to the words of another speaker. Lastly, the sentence vadaiva has to be taken as addressed to Sakalya. The commentator remarks that, he being the questioner, one expects prikkha instead of vada. But Yagnavalkya may also be supposed to turn round on Sakalya and ask him a question in turn, more difficult than the question addressed by Sakalya to Yagnavalkya, and in that case the last sentence must be taken as an answer, though an imperfect one, of Sakalya's. The commentator seems to think that after Yagnavalkya told Sakalya to ask this question, Sakalya was frightened and asked it, and that then Yagnavalkya answered in turn.

^142:2 The Madhyandina text varies considerably. It has the first time, kashur lokah for agnir lokah. I keep to the same construction throughout, taking mano gyotih, not as a compound, but like agnir loko yasya, as a sentence, i.e. mano gyotir yasya.

^143:1 Ask me. Comm.

^143:2 That from which he is produced, that is his devata. Comm.