End Notes
1 Called chándas in the RV. itself.
2 Except the two metres Aryà and Vaitàlãya which are measured by moræ.
This seems to have been the only metrical principle in the Indo-Iranian period, because in the Avesta the character of a verse depends solely on the number of syllables it contains, there being no quantitative restriction in any part of it.
3 A figurative sense (derived from foot = quarter of a quadruped) applicable because the typical stanza has four lines.
4 [I.e. »g Veda.]
5 There are also several longer stanzas formed by adding more verses and consisting of 52, 56, 60, 64, 68, and 72 syllables; but all these are rare: only two stanzas of 68 and one of 72 are found in the RV.
6 No infringement of this rule occurs in any metre of the RV. but the comparatively rare Dvipadà Viràj (4 x 5), in which three exceptions are met with.
7 The vowels ã, å, e when PragÔhya (25, 26), however, remain long before vowels. When a final long vowel is the result of Sandhi, it also remains long; tásmà adàt for tásmai adàt.
8 [I follow Macdonell in always marking the end syllable according to its natural weight. When placed in square brackets the metrical markings have been added in for clarity's sake.]
9 Next to the Triùñubh this is the commonest metre in the RV., nearly one-fourth of that Saühità being composed in it; yet it has entirely disappeared in Classical Sanskrit. The Avesta has a parallel stanza of 8 é 8 syllables.
10 The first two Pàdas of the Gàyatrã are treated as a hemistich in the Saühità text, probably in imitation of the hemistich of the Anuùñubh and the Triùñubh; but there is no reason to believe that in the original text the second verse was more sharply divided from the third than from the first.
11 By far the commonest variation from the normal type is that in which the second syllable of the cadence is short (ÛÛÛÅ). This occurs about as often in the first verse of Gàyatrãs as in the second and third combined.
12 The only long series of such trochaic Gàyatrãs occurs in RV. viii. 2, 1-39.
13 The trochtaic Gàyatrã is commonest in Maõóalas i and viii, which taken together contain about two-thirds of the total number of examples in the RV.
14 [In Vedic visarjanãya (þ), counts as does one consonant, therefore if there is a short vowel preceeding it, and no consonant following it, the syllable is light/short. Cf. also the 3rd line in the next verse.]
15 The frequency of this metre is about one-third that of Gàyatrã in the RV., but in the post-Vedic period it has become the predominant metre. The Avesta has a parallel stanza of 4 x 8 syllables.
16 Where the iambic cadence in the first verse has entirely disappeared.
17 This is the regular type of the Anuùñubh in the AV.
18 The Avesta hits a parallel stanza of 5 x 8 syllables.
19 The only irregularity here is that time first syllable of the cadence may be short when it coincides with the end of a word.
20 This appears to have been the original position of the cæsura because the parallel verse of the Avesta has it there and never after the fifth syllable.
21 Identical with the opening of the octosyl1abic verse.
22 The fourth syllable here is sometimes short: the fifth is then always long.
23 The first of those two syllables is sometimes, but rarely, long in the old hymns of the RV., still more rarely in the later hymns, and hardly ever in B.
24 This anomaly also appears in the metre of later Vedic texts and of Pàli poetry.
25 The extra syllable in such cases is perhaps due to the verse being inadvertently continued after a fifth syllable cæsura as if it were a fourth syllable cæsura.
26 The deficiency of a syllable in such cases may have been partly due to the similarity of the decasyllabic Dvipadà Viràj (8) with which Triùñubh verses not infrequently interchange.
27 About two-fifths of the RV. are composed in this metre.
28 The Avesta has a parallel stanza of 4 x 11 syllables with cæsura after the fourth syllable.
29 These are accounted Atijagatã (52) or øakvarã stanzas by the ancient metricians when the fifth verse is a repetition of the fourth. If it is not a repetition it is treated in the Saühità text as a separate verse (as v. 41, 20; vi. 63, 11) and is called an ekapadà by the metricians.
30 It is probably not Indo-Iranian, because though a verse of 12 syllables occurs in the Avesta, it is there differently divided (7 +5).
31 As the Gayatrã verse is never normally found in combination with the Triùñubh, but often with the Jagatã verse, it seems likely that the iambic influence of the Gàyatrã led to the creation of the Jagatã, with which it could form a homogeneous combination.
32 [This further supports the theory that in verses with replacement, the light/short syllables that are replaced are the ones following the cæsura; see The Prosody of the Dhammapada for more on this phenomena.]
33 That is, its first syllable is less often long than short.
34 This stanza is somewhat rare, occurring in the RV. not much more than a hundred times.
35 The otherwise universal rule that the end of a verse must coincide with the end of a word is three times ignored in this metre (at this end of the first and third verses).
36 With this metre compare the defective Triùñubh verse of ten syllables (4 a).
37 This interchange occurs especially in RV. vii. 34 and 56.
38 Here the verb, though the first word of the verse (App. III, 19 b), is unaccented. This is because the end of the first and the third verse in this metre has a tendency to be treated like a cæsura rather than a division of the stanza. Cp. note 2.
39 These three verses are treated as a hemistich, in the Saühità text.
40 The verb is accented because in the Saühità text it is treated as the first word of a separate verse.
41 [This appears to be a Triùñubh line, rather than the required Jagatã.]
42 These are the compositions of a very few individual poets.
43 Only about ten examples of this metre occur in the RV.
44 [Presumably we should read ½ but £ is written.]
45 This is the only comparatively common long metre (of more than 48 syllables) in the RV., where more than 80 Atyaùñi stanzas occur.
46 [Should we read àsiái in the 5th; and èµvasà in the last line?]
47 RV. viii. 29.
48 RV. ix. 110.
49 RV. x. 98.
50 RV. viii. 35.
51 RV. v. 87.
52 But the intrusion of Jagatã verses in a Triùñubh hymn is exceptional in the RV., though very common in the AV. and later.
53 Except stanzas 7 and 15, which are pure Anuùñubh and Triùñubh respectively.
54 [Macdonell places the division one syllable too early in this line.]