7.3.44 प्रत्ययस्थात् कात् पूर्वस्यात इदाप्यसुपः

Word by word (padaccheda)

Left context प्रत्ययस्थात् pañcamī · ekavacanaLeft context कात् pañcamī · ekavacanaTarget पूर्वस्य ṣaṣṭhī · ekavacanaTarget अतः ṣaṣṭhī · ekavacanaSubstitute / term prathamā · ekavacanaRight context आपि saptamī · ekavacanaLeft context असुपः pañcamī · ekavacana

Left context Pañcamī (5th case): "after X", the operation applies to what follows (1.1.67 तस्मादित्युत्तरस्य). Also used for limits and sources.
Target Ṣaṣṭhī (6th case): names what gets replaced, "in place of X" (1.1.49 षष्ठी स्थानेयोगा).
Substitute / term Prathamā (1st case): in operational rules, the substitute or the thing introduced; in definitions, the term or its referent.
Right context Saptamī (7th case): "when X follows", the operation applies to what precedes (1.1.66 तस्मिन्निति निर्दिष्टे पूर्वस्य). Also a condition or domain ("in the sense of").

Dotted words are technical terms. Hover or tap one to see its card.

Full reading

प्रत्ययस्थात् कात् पूर्वस्य अतः आपि असुपः

Pāṇini leaves out words that carry over from earlier sūtras. Expand to see the rule as a reader actually understands it.

Commentary

इ is substituted for the अ which stands before the क् belonging to an affix when the feminine ending आ follows, provided that, it does not come after a case-affix (i.e. when such a word in अक does not stand at the end of a बहुव्रीहि समास).
That which stands in an affix is called प्रत्ययस्थ, i.e. क् must be the part of the affix. The अक् is changed into इक् in the feminine in आ ॥ As जटिलिका, मुण्डिका, गुणिका, दत्तिका, कारिका, हारिका So also एतिका from एतद् + अकच् + आप् एतकद् + आप् ॥ The द् is then replaced by अ 7.2.102. The क् must belong to the affix, therefore, not in शका from शक शक्नोति ॥ The स्थ in प्रत्ययस्थ is for the sake of distinctness, there is no affix which is only क ॥ The rule applies to क, therefore not to नन्दना, रमणा ॥ The इ is substituted for the अ which precedes (पूर्वस्य) क्, and not the अ which follows क्, as पटुका, मृदुका ॥ The must be preceded by short अ, therefore, not here गोका, नौका, nor in राका, धाका ॥ The substitution takes place when आ follows, therefore, not in कारकी ॥ The word आपि qualifies क् ॥ The क् should be followed by आ ॥ Objection: But in कारिका from कारक + आ, क् is not directly followed by आ, but by अ? Answer: When अ + आ = आ there is ekadesa, the क् is followed by आ, there being no third letter intervening then. Objection: But an ekadesa is sthanivat to what it replaces when a purva-vidhi is to be applied, therefore there still exists the intervention. Answer: The express text of this sutra will remove the intervention. Objection: Then इ should be substituted in रथकट्या and गर्गकाम्या ॥ These words are formed by the affixes कट्यच् and काम्यच्, as रथानां समूहः, 4.2.51 and गर्गमिच्छति आत्मनः 3.1.9 = रथकट्य and गर्गकाम्य ॥ Here also before the क् of the affix, the अ should be replaced by इ, when the feminine affix आ is added: for there is no intervention between क् and आ (the intermediate letters being non-existent, as you say). Answer: We say that intervention is no intervention, when it consists of only one letter, which even does not actually exist, but only through the fiction of sthanivat. Therefore, the feminine of the above words will not take इ, as क् is not followed immediately (in the sense above expressed) by आप्, as रथकट्या and गर्गकाम्या, because in these actually many letters are heard as intervening between the क् and आप् ॥

Why do we say असुप्ः 'provided that the feminine affix आप् does not come after a case-affix? Observe बहवः परिव्राजका अस्यां मथुरायां = बहुपरिव्राजका मथुरा ॥ Here आप् comes after the noun बहुपरिव्राजक which ends in a case-affix, and hence the अ of ज is not changed to इ ॥ The case-affix is elided by 2.4.71, and it still exerts its influence by 1.1.62. The word असुपः is a प्रसज्यप्रतिषेधः (a simple prohibition of the particular matter specified without mentioning what is different from it): and not a पर्युदासः or exception. Had it been a Paryudasa, (सुपोऽन्यः = असुपः) then आप् coming after a full word not having a case-affix, would have caused the इ substitution in the last example. Nor should असुपः be explained as that in which there exists no case-affix. Had it been so, then in बहूनि चर्माण्यस्यां = बहुचर्मिका not इ ought to be added.

Vart:- मामक and नरक should be enumerated, for the क् of these is not part of the affix: as मामिका, नरिका ॥ Here ममक् is substituted for मम before the affix अण् 4.3.3, to which is then added टाप् ॥ The word ममक takes long ई in the feminine only when it is a Name or in the Vedas 4.1.30: therefore though मामक ends in अण्, it does not take ङीप् 4.1.15 but टाप् 4.1.4, नराम् कायति = नरक formed with the affix क (अ 3.2.3).

Vart:- The rule applies to the words ending in त्यक् 4.2.98 and त्यप्, 4.2.104 in spite of the prohibition in 7.3.46. As दाक्षिणात्यिका, इहत्यिका ॥

The word in the sutra is कात् the fifth case of क ending in अ ॥ If then this be the condition, that the affix must end in ka क and not in k क्, then the rule will not apply to एतिका ॥ Because here the augment is अकच् (अक्) with क्; the final अ in अकच् is for euphony only, and not a portion of the affix, as we find in भिन्धकि, छिन्धकि, रुन्धकि ॥ The word कात therefore, is construed to mean ending in the consonant क ॥

Śrīśa Chandra Vasu, The Ashtadhyayi of Panini (1897), via ashtadhyayi.com.