SuttaCentral
I was just reading an introductory text and found in the offered glossary:
ARAHAT An individual who has eliminated the ‘fetters’ by…
At first I thought this was just a typo, but then found that SC has dictionary listings for both “arahat” and “arahant”. What’s that about? 
Sanskrit?
Much thanks, but to the best of my knowledge, nope, that’s arhat (also, the glossary intro specifically states that all terms are Pali, unless otherwise mentioned).
They are both anglicisms based upon the Pali stem but not corresponding to any of the Pali inflectional forms of the word:
arahaṃ | arahanto
arahantaṃ | arahante
arahatā arahantena | arahante(b)hi
arahato arahantassa | arahantānaṃ arahataṃ
arahatā arahantā arahantasmā arahantamhā | arahante(b)hi
arahato arahantassa | arahantānaṃ arahataṃ
arahante arahantasmiṃ arahantamhi | arahantesu
bho arahanta | bhavanto arahanto
The form arahat was used by pioneering scholars (e.g., Thomas Rhys Davids), while nowadays arahant seems to be more favoured.
I think this has to do with the fact that Pali has been traditionally and mostly rendered in abugida scripts - i.e. each letter represents a consonant with an inherent vowel - most of time related to the Brahmi script
This means that in Thai script, for example, it is อรหต. The first letter is a mute basis for vowels, the second and third are equivalent to R and H. Hence, it could be rendered in English as (A)-R-H-T, hence arhat.
In the Jain scriptures, they rendered a slightly different version of the Magadhi dialect. They also use the term in a similar way to Buddhist, and in the English rendition of their Navkar mantra they usually render it as arihanta, अरिहंता.
It’s a universal phonetic phenomenon, it exists in all languages (in linguistic jargon called “conditioned merger”, “phonetic alternation or fusion” & other terms). It usually occurs in words involving nasal phonemes such as the ‘an’ in arahant. The tendency is for the omission of either nasal sound or consonant. The nasal ‘n’ is omitted in “arahant” but the consonant ‘t’ is omitted in the Latin “annus”, which was once “atnos”.
There are similar cases of phonetic alternation that is preserved in the Pali text, which is curious because it is believed that the text follows from an oral memorisation where one expects pronunciation to be homogeneous, but obviously it wasn’t! Very curious!