[*]l. 2 B སྙག་ [*]l. 5 B པའི་ [*]l. 7 A 1 and 2 བར་ [**]l. 7 B དྲིན་ [*]l. 10 B པ་ [*]l. 16 A 1 and 2 both ངས་. Text from B [**]l. 16 B closes the line with a ༈ instead of ༎ [*]l. 17 B དེང་ [*]l. 18 B པ་ [*]l. 19 B ནས་ [*]l. 20 A 2 བརྩ་ [*]l. 22 B བས་ [*]l. 24 A 1 and 2 བཙོན་ [**]l. 24 B པ་ [*]l. 26 B ཅིང་ [*]l. 29 B ཕུང་ [**]l. 29 A 2 སླང་ [*]l. 32 B ཀྱང་ [**]l. 32 B འཁྲལ་ [*]l. 34 B བསྒྲུབ་ [*]l. 35 B ཁྲུག་ [*]l. 37 A 1 and 2 ཐཔས་ [*]l. 38 B འཆད་ [*]l. 40 B སྐྱེལ་ instead of གཏོང་ [*]l. 41 B རྣམས་ [*]l. 42 B སོམས་ [*]l. 43 B སྦྱོངས་ [*]l. 44 B last three words in B བདག་ [*]l. 45 A 1 and B བོ་ [*]l. 45 B ཀུན་ [*]l. 46 last four words in B བར་ [*]l. 47 A 2 བླ་ [*]l. 48 4 2 བོར་ [*]l. 48 B has ༈ at the end of the line instead of ༎ [*]l. 49 B ཀུན་ [*]Colophon, A 2 has no ཅད་ after ཐམས་, and has a final ས་ to གདུངས་. B has a different colophon འདི་
[a]Contents]]
C. The Variants.
The texts used were two small blockprints, nearly identical A 1 and A 2, and a large blockprint B.
On the whole A furnishes a good text and it may be used as the basis for the edition. Two curious cases of the use of བ་ for པ་ (7. 45) seem more than mere negligence of the wood-cutter [[11]]in connection with the badly printed པའི་ in l. 13 (which looks also like བའི་) and also a པ་ like བ་ in l. 23. Inversely there is a clear པ་—ཐཔས—in l. 37 and a བོར་ for པོར་ in l. 48. A 2 twice lacks the hook in རྩ (20, 24) and the naro ོ in lines 29, 47. These two latter variants may be due to deterioration in the blocks or the roughness of the paper, or defective inking. Otherwise A 1 and A 2 are practically identical, and except for the last pages (the last two of A 1 are condensed into a single one in A 2) the two copies may have been printed from the same blocks.
In 5 B writes ལྡིང་ for བའི་ as authorized by the Dicts. But the question of final particles is still far from being satisfactorily settled. The Dicts. are on the whole much at variance on this point. Desg. gives as a rule a greater variety of them than J.
Some differences in the tenses of the verb are presented by the two copies of A on one side and B on the other. In l. 2 སྙེག་ is the present tense as against the past form བསྙེག་ in A. As to the sense both would do, and though the past form in Tibetan is better rendered in English by the present we may understand the past form as ‘has begun to rise.’ In verse XI B gives imp. forms, making the sense one of command whereas A has present forms giving a mere statement. The final ས་ in སོམས་, however, is not recorded in the Dicts., nor the form སྦྱོངས་; ཐུལ་, however, is a regularly recognised imp. form.
འཁྲུགས་ in l. 35 is a correct past tense. The form ཁྲུག་ (without an initial འ་) as in B is not recorded, though འཁྲུག་, present, might do equally well. འཁྲེལ་, l. 32, is not authorized by the Dicts. which all omit the initial འ་. The substitution of འཆད་ for བཤད་ (38) seems to lack sufficient urgency, [[12]]though J. records a འཆད་ ‘to listen to an explanation’ from Sch. A འཕུང་, l. 29, is correct according to the Dicts., not ཕུང་ of B, though J. and S. Ch. D. give the alternative spelling.
In the treatment of grammatical particles A is superior to B. པར་ (10) is correct, not པ་ B. It is an adverbial construction. In 18, པར་, and 22, བར་, equally so. In 24 པར་ is a terminative dependent on བགྱིད་.
The remaining variants are all in the nature of equivalents for or against which nothing (or the same!) can be said, and which would do as well as the readings we have adopted. Many of them are, however, curious for this reason, that they are not homonymous variants at all and consequently substitutions for, not corruptions of, the text. We have to leave the question alone whether those in A or in B are likely to be the original ones.