There is vocalisation of the affix ṣyaṅ [[4.1.78]] when the word bandhu follows in a bahuvrīhī compound.,
Thus कारीषगन्ध्याबन्धुरस्य = कारीषगन्धीबन्धुः ॥ So also कौमुदगन्धीबन्धुः ॥
Why do we say \when the compound is a Bahuvrihi\? Observe कारीषगन्ध्याया बन्धुः = कारीषगन्ध्याबन्धुः, which is a Tatpurusha compound. Like the last sutra, we have here also परमकारीषगन्धबन्धुः, but अतिकारीषगन्ध्याबन्धुः, कारीषगन्ध्याबन्धुधनः, कारीषगन्ध्यापरमबन्धुः ॥
Though the word बन्धुनि is exhibited in the sutra in the neuter gender, it is in fact a masculine word.
Vart:- There is vocalisation of ष्यङ् in a Bahuvrihi compound with मातच्, मातृक and मातृ optionally: as कारीषगन्धीमातः꣡ or कारीषगन्ध्यामातः꣡, कारीषगन्धीमातृकः or कारीषगन्ध्यामातृकः, कारीषगन्धीमाता, कारीषगन्ध्यामाता ॥ The indicatory च of मातच् makes the word take the udatta on the last syllable 6.1.163, thus debarring the especial accent of the Bahuvrihi (VI. 2. I). All Bahuvrihi compounds ending in ऋ take the samasanta affix कप्, so मातृ would have become मातृक by force of 5.4.153, so the separate enumeration of मातृ and मातृक here shows that कप् is also optional.,