The manuscript has been written by more than one scribe. The first scribe appears to have two ways of writing ḍ. The less common type occurs twice in 6v1. Taling is sometimes canceled with hulu, as in 4v2, 6r2, 6r4. Sometimes the first of two talings in sequence is canceled; it would be more convenient to be able to encode that with ⌈ but we have rather deleted e (See instances marked with below). There are a few instances of pasangan -ya being used instead of paten, especially after akṣara ra (e.g., 16v1, 2); could this be influenced by the similarlity of Javanese paten and Balinese pasangan -ya? The layar sometimes slants downward, sometimes upward. New scribal hands appear to begin on folio 33r, 48v, 54r and 61v (with this one, u and ū not distinguished at all). We have hyphenated reduplications, Javanese pair-compounds, and mixed compounds such as kalāntara-putra. Layar is often (but not always) written so as to overlap the contiguous aksaras; in many instances, it is clearly written over the first, but at the very beginning of 43v3 it is attached to the second aksara in the Indian fashion. Folio numeration is extremely faulty: often missing, wrong, corrected, or wrongly corrected (f. 60). On ff. 12-16, the scribe has written 22, 33, 44, 55, 77 (sic) by mistake. Some folios at the beginning and end are unnumbered.