अनुदात्त-उपदेश-वनति-तनोत्य्-आदीनाम् अनुनासिकलोपो झलि क्ङिति

Thus यमु gives us यत्वा (with क्त्वा), यतः (with क्त), यतवान् (with क्तवतु), यतिः (with क्तिन्), Similarly रमु gives us रत्वा, रतः, रतवान्, रतिः ॥ यम्, रम्, नम्, गम्, हन् and मन् which end in a nasal are to be considered as unaccented roots, though taught as accented in the Dhatupatha. So also of वन् : i. e. वतिः with क्तिन्; with क्तिच् the nasal is not elided as वन्ति; 6.4.39 : and before other jhaladi affixes व, retains न् as all those affixes take the augment इट् ॥ The तनादि roots belong to the eighth class. Thus ततः, ततवान्. The Tanadi roots are ten in number, तन्, सन्, क्षण्, क्षिण्, ऋण्, तृण्, घृण्, वन्, मन् and कृञ् ॥ Of these सन् takes long आ also 6.4.45. क्षण् -क्षतः, क्षतवान्, ऋण् - ऋतः, ऋतवान्; तृण् - तृतः, तृतवान्; घृण् — घृतः, घृतवान्; वन् - वतः, वतवान्; मन् - मतः, मतवान् ॥

Why do we say before a ङित् affix ? Observe अतत, अतथाः (1.2.11 these affixes are ङित्) So also not in शान्तः, शान्तवान्, तान्तः, तान्तवान्, दान्तः, दान्तवान्, not being anudatta in the Dhatupatha. If the root does not end in a nasal, the rule does not apply, as in पक्वः, पक्ववान् ॥ If the affix does not begin with a jhal consonant, the rule does not apply: as गम्यते, रम्यते ॥ If the affix is not कित् or ङित् the rule does not apply: as, यन्ता, यन्तव्यम् ॥ Why do we say \anudatta by upadesa\? So that the rule may apply to गम् root, as गतिः, but not to शम्, as शान्तः शान्तवान् ॥ For गतिः is formed by क्तिन् affix and is accented with udatta on the first, but it is a secondary udatta and not of upadesa or dhatupatha, and does not prevent the elision of म् of गम् ॥ Similarly शम् is udatta-upadesa though in शान्त it has become अनुदात्तः ॥ The former though taught in the Dhatupatha as udatta, has been specifically mentioned above as anudatta, and शम् is taught as udatta in the Dhatupatha, and no where else taught as anudatta.,