Note that while ISO 15919 indeed mandates ṁ for anusvara, it makes use of ṃ for older Gurmukhi: „Bindi shall be transliterated as ṁ (Hex 1E41) and Tippi as ṃ (Hex 1E43).“

For searching, I would typically transliterate the query into the native script and search in the native script.* That allows me to support several transliteration conventions at the same time (e.g. ū, U, oo for long u). Luckily the conflicts in this direction are rare.

Thanks,

Jan Kučera

ल Institute of South and Central Asia Students, Prague

* Usually care also needs to be taken of the special case where if the query ends with a consonant, it is expected to match any syllables with the consonant.

From: INDOLOGY <indology-bounces@list.indology.info> On Behalf Of Harry Spier via INDOLOGY
Sent: Monday, March 2, 2026 1:48 PM
To: Walter Slaje <walter.slaje@gmail.com>
Cc: indology@list.indology.info
Subject: Re: [INDOLOGY] Anusvara in IAST transliteration

Dear everyone,

Daniel BAlogh wrote:

I dare say that as far as IAST can be considered a standard, the "correct" IAST anusvāra is ṃ

Walter Slaje wrote:

 I believe that whether one uses ṃ or ṁ, there will be no undesirable consequences.

The reason I asked the question was because of search engines.

The Searchable Sanskrit Library has a search engine covering multiple collections so all anusvara's (like all other letters in it) need to follow the same convention (either overdot or underdot) .  But it seems that since all the collections I've seen use m underdot thats been chosen as the de facto standard. 

Thanks,

Harry Spier