अकः सवर्णे दीर्घः

When a single vowel is followed by a homogenous vowel, the corresponding long vowel is the single substitute for both the precedent and the subsequent vowels.
Thus दण्ड + अग्रम् = दण्डाग्रम्, दधि + इन्द्रः = दधीन्द्रः, मधु + उदके = मधूदके and होतृ + ऋश्य = होतृश्यः ॥ Why do we say 'an अक् or simple vowel ?' Observe अग्ने + ए = अग्नये ॥ Why do we say 'by a homogenous vowel?' Observe दधि + अत्र = दध्यत्र ॥ The word अचि is understood here also. The word सवर्णे, therefore, qualifies the word अचि understood. The rule will not apply if a homogenous consonant follows. As कुमारी शेते ॥ The ई and श are homogenous, in spite of (l. I.10): for that prohibition does not apply to the long ई and श, because the rule of classification and inclusion contained in अणुदित्सवर्णस्य चाप्रत्ययः 1.1.69 is not brought into operation at the time when नाज् झलौ 1.1.10 rule operates, because of its being a portion of सवर्ण rule. Therefore, so long as it does not come into operation it is not accomplished. Therefore first the rule of नाज् झलौ comes into play, then the rule of सवर्ण definition 1.1.9 and then comes the ग्रहणकवाक्यं 1.1.68. Therefore in नाज् झलौ those अच् only are taken which are not included in the class of homogenous vowels i. e. only the 9 vowels contained directly in अच्, and not the सवर्ण vowels which 1.1.68 would have denoted. Therefore though short इ and श are not सवर्ण by 1.1.10: the long ई and श would be savarna.

Vart:- When ऋ short is followed by ऋ short, the long substitution is optional: so also with लृ ॥ This vartika is necessitated because (1) the two ऋ or लृ -- the precedent and the subsequent -- are not homogenous, because one is samvrita and the other vivrita, or (2) because their prosodial length is 1 ½ and so the word दीर्घ cannot be applied with consistency in their case (ऊकाल &c). Thus होतृऋकारः = होतृकारः or होतॄकारः, so also होतृ + लृकारः = होत्लृकारः or होत्लॄकारः॥ The दीर्घ of लृ is ॠ ॥