iginawa niya ang iták 'he made it with the knife (the knife was the instrument of his making it).'

The an passive indicates the case of any person or thing that can be regarded as a place. This place may be place where,' 'place whither,' or place whence,' including 'person to or from whom,' the difference in the locative relation being due to the meaning of the verbs themselves; e. g.,

sinusul-átan ko itó-ng papel 'I am writing on this paper (this paper is the place of my writing).'

ang Cebu' ang paroroonan niya he is going to Cebú (Cebú is whither he is going).'

bigyan mo kami ngayʻón' nang áming' kánin sa arawáraw 'give us this day our daily bread (let us be given to by thee today of our bread for every day).'

tinanggápan' ko ang aking amain I received it from my uncle (my uncle was received from by me).'

The possibility, then, of expressing the case relations of a noun to a verb by varying the verb, which is latent in all languages, is developed to a high degree in Tagalog. By means of its four voice forms, not only the case of the agent and the object may be expressed, as in Indo-European and Semitic, but all the ordinary case relations which may exist between noun and verb. The active expresses the case of the agent; the in passive, the case of the non-alienated object; the an passive, all local relations of whatever character; while the i passive bears the burden of the remaining case relations.

1 Proper names of places are treated in Tagalog as common nouns, taking the definite article.

* The reversed apostrophe is used to indicate the glottal catch at the beginning of a syllable after a consonant: in the older orthography this was denoted by a dash, viz., nğay-ón.

3 The final ng of these words is the result of the coalescing of a final n with the ligature ng. The different type is used to indicate this coalescing, and to show that ng is not the final consonant of the word, as it is for example in galing 'goodness.'

4 The spelling ngg is to be preferred to the older form ng, as in tinangápan, because it indicates the sound (guttural nasal plus guttural sonant stop) more accurately, and because the simpler form ng is more or less liable to be confused with ng.

Textual Notes on the Old Persian Inscriptions.-By Professor A. V. WILLIAMS JACKSON, Columbia University, New York City.

1. OP. patiyāvahyaiy (sic) Bh. 1. 55.

THE various editions of the text of Bh. 1. 55, from Rawlinson onward, have read patiyavahaiy, which Bartholomae (Altiran. Wb. 1353; cf. also Grundr. Ir. Ph. 1. 197, § 329, n. 2; ZDMG. 48. 156; and AF. 2. 106, n. 3) derives from 3 van-, ‘wünschen,' with patiy and a, and interprets as anflehen' or 'ich pries.' A similar view is given by Foy, KZ. 36. 67; 37. 518. The true reading of the radical part of the word, however, is -vahyaiy (-vahayiy", with hy, not y) as is plainly shown in a photograph which I took of it when I examined the Behistan Rock in 1903, as described in JAOS. 24. 76-95, and in Persia Past and Present, pp. 186-212. The verb patiyavahyaiy (pret. indic. mid.) is therefore naturally to be connected, as a denominative, with Av. avah-, avahya-, avah‘ya-, 'aid, assistance, support' (cf. Bartholomae, Altiran. Wb. 179, and see especially Friedrich Müller in WZKM. 1. 122, and Tolman, OP. Inse. p. 167), and it is precisely the Iranian equivalent of the Sanskrit denominative avasya-, 'seek for aid, take refuge with,' in Rig Veda 1. 116. 23 (arasyate, dat. pres. ptcpl.).

A month after my visit to Behistan I spent the morning of May 3, 1903, in examining the text of the Persian inscriptions at Persepolis, and I shall give here in brief form my memoranda with regard to the more important words as noted in my copy of Weissbach and Bang, Die altpersischen Keilinschriften, and of Spiegel, Die altpersischen Keilinschriften, 2te Aufl., both of which volumes I had with me. On my return home, after completing my notes, I compared my memoranda with the photographs that are accessible in Stolze and Andreas, Persepolis, and have added my comments in square brackets.

2. Dar. Pers. c = Spiegel L.

Regarding the short inscription (Dar. Pers. c. Sp. L) carved around the stone lintels of the windows through which

King Darius looked out upon his people, I have noted that the cuneiform letters of the oft-repeated sentence ardastāna abargaina dārayavauš xšāyaliyahya vibiya karta are deeply chiselled, and that the reading vitiya is correct.

1.

3. Dar. Pers. d. Spiegel H.

auramazdā: the letter m is slightly damaged, as the horizontal stroke is defective, but the character as a whole can still be made out quite clearly. [Compare now also the photograph by Stolze, Persepolis, 2. pl. 95.]

7. auramazdā: the letter z is imperfect, as its first vertical bar is missing because of a break (cf. Spiegel, op. cit., p. 47, n. 7). [See also Stolze's photograph, as already referred to.]

8. uvaspa: the u is perfectly distinct; the e, although damaged (cf. Spiegel, p. 47, n. 8), can be made out quite well. [Cf. also Stolze, loc. cit.]

11. haca: the latter wedge of the h is partly damaged, but is legible. Cf. also Spiegel's note.

11. aniyanā: the word in this form is clear on the stone. [So also Stolze.]

14. vioibis: each letter I have carefully noted as being correct. [Stolze's plate is cracked at this point.]

17, 19. dušiyārā: the u is not clear, but can be made out. Compare Spiegel's remarks (p. 48, n. 17) on the reading.

18. abiy: this is the correct reading as given by Weissbach and Bang as against Spiegel (p. 49, n. 18) and Westergaard (ZKM. 6. pl. 1). My notes state abiy is plainly on the stone.' [So also Foy in KZ. 37. 559, although the letter b is not quite sharply defined in Stolze's photograph.]

20. dušiyāram: there is some space between the i and the y, due apparently to an original defect in the stone before it was lettered. See also next note. [Stolze is not quite distinct

here.]

21. yan m: there is a slight space between n and m, apparently due, as in the preceding case, to an original defect in the stone, and not to any lacuna in the tablet. I could see no evidence of any letter being missing, and I believe that the reading yanam may be accepted as certain from the explanation I have given. [On gaining access to my books I was interested in seeing that Foy's conjecture yān(iy)am, in KZ. 35. 49,