[16] N. xviii. 89; xix. 1; AP. cccxxxvii. 18, 27. [↑]
[17] DR. i. 15; iii. 20–22. [↑]
[18] N. xix. 2–6, 25 f.; DR. i. 11, 12, 16; SD. 296 f., 323. [↑]
[19] N. xix. 23; DR. i. 13; SD. 320–3; R. iii. 13 f. [↑]
[20] N. xix. 7–13; DR. i. 18–20; SD. 324–9; R. iii. 22–5. [↑]
[21] N. xix. 19–21; DR. i. 16 f.; SD. 317–19. The parallelism is faulty: neither episode nor incident is necessary nor corresponds to Prāptyāçā and Niyatāpti nor Garbha and Vimarça; Dhanika, DR. i. 33, admits this in effect; there is no episode in Ratnāvalī, III. Cf. R. iii. 22. [↑]
[22] N. xix. 16, 35 ff.; DR. i. 22 ff.; SD. 330 ff. Hali (DR., p. 11 n.) suggests nibarhaṇa as correct (N. xix. 36), wrongly. Cf. R. iii. 26–74. The precise parallelism of the Sandhis and Avasthās in the Bālarāmāyaṇa is given in R. iii. 23–5. [↑]
[23] Abhinavagupta (Dhvanyāloka, p. 140) frankly treats the Avasthās as the Sandhis as parts of the story, and distinguishes the Arthaprakṛtis. DR. is responsible for the doctrine that each Sandhi rests on an Avasthā and an Arthaprakṛti, accepted in Pratāparudrīya, iii. 3; GGA. 1913, pp. 306–8; R. iii. 26 f. [↑]