You've mastered how vowels merge at word boundaries, creating smooth transitions through rules like Dirgha and Guna sandhi. Now, we'll look at the more complex changes that happen to consonants and the special case of the visarga (ḥ). These rules are the key to deconstructing the long, unbroken strings of text, or Saṃhitā, found in classical Sanskrit literature. It's how you'll learn to see the individual words hiding in plain sight.

First, a key distinction: sandhi can be internal (occurring within a single word as prefixes and suffixes are added) or external (occurring between two separate words). While internal sandhi helps form words, external sandhi is what welds sentences together. We'll focus on the external rules, as they are crucial for reading.

The visarga, written as ḥ (ः), represents a voiceless breathy sound, like a faint echo of the preceding vowel. It's a chameleon. Its pronunciation, and often its written form, changes dramatically depending on the sound that follows it. Let's explore its main transformations.

What happens when the following word begins with another 'a'? The 'aḥ' still becomes 'o', but the following 'a' is elided, or dropped. To show this has happened, Sanskrit uses a special symbol called the Avagraha (ऽ).

The visarga has other tricks. When it's preceded by any vowel except 'a' or 'ā' and followed by a voiced sound (a vowel or voiced consonant), it turns into an 'r' sound, known as Repha.

Sometimes, the visarga simply disappears. This is called Lopah (elision). This happens when a visarga preceded by 'ā' is followed by a voiced sound, or when a visarga preceded by 'a' is followed by any vowel except 'a'.

Consonants also change when they meet at word boundaries, a process called Hale Sandhi. These changes usually involve assimilation, where one sound becomes more like its neighbor to make pronunciation easier.

Two common assimilation rules are Ścutva and Ṣṭutva sandhi. Ścutva occurs when dental consonants (like s, t, d, n) are followed by palatal consonants (ś, c, j, ñ). The dental consonant changes to its palatal equivalent.

Another powerful rule is Jaśtva sandhi, where a consonant at the end of a word changes to the third letter of its class (the voiced, unaspirated sound) if the next word begins with a voiced sound. For example, a 'k' (first in its class) would become a 'g' (third in its class).

Finally, let's look at the nasal sounds. The Anusvāra (ं), a dot above a letter, often represents a nasalization that precedes a consonant. In Parasavarṇa sandhi, this general nasal sound can optionally change to the specific nasal consonant of the following consonant's class. For example, an Anusvāra before a 'k' or 'g' (velar class) can be written as 'ṅ' (ङ), the velar nasal.

For example, the word saṃgatiḥ (company) can also be written as saṅgatiḥ (सङ्गतिः). This is why you sometimes see different spellings for the same word. Both are correct.

Understanding these advanced rules transforms Sanskrit from a series of characters into a living, flowing language. It takes practice, but soon you'll be able to spot these transformations and deconstruct complex sentences with ease.