[No. 2, 
164 Hoernle — Essays on the Gaurian Languages. 
principles of Gaurian, which adds the case-sign only to an inflexional 
base (or rather inflected base, viz., the Prakrit nom. sing.) ; hut strictly 
according to the principles of Prakrit. Hence those nouns in Bangali 
which add yy constitute the Prdkritic element of Bangali in opposition 
to its Gaurian element which adds simply y. We may assume, that 
at the time of the formation of the Gaurian principles of declension, 
yy, (i. e., the modified form of yry) had become finally established as 
the gen.-sign., and its real nature was forgotten. The Gaurian then 
added yy as an enclitic part of speech to its inflexional base to signify 
the genitive ; e. g., (irrfT or) ir?T (iuflex. base) -f yy (enclitic gen. particle) ; 
and in contact with the inflexional base which always ended in a vowel, the 
enclitic yy lost its initial y, according to a regular tendency of such enclitic 
words ; e. g., the Sanskrit now is in Prakrit regularly only , 
again yvry in Pr. is vjyr, in the Gaurian only vf for yr (as in Hindi ari^r 
for 5fTT = y?s y»rx ). Hence ■iiPfr + yy becomes hi^fr + y or argiy. 
There still remains a class of Bangali nouns which require an expla- 
nation, as they seem to contravene the ordinary rule of the gen. forma- 
tion ; viz. the Bangali adjectives in as small, 'yy large, good, 
etc. They add not yy as might be expected, but y ; e. g., WT«y, not wrSy. 
If we compare these Bangali adjectives with those nouns, that add yy, 
two facts become at once apparent which distinguish them one from the other, 
and which stand in the relation of cause and effect to each other. Those 
t vo facts are ; 1, those nouns which add yy (having dropped their final y? 
in Sandhi with yy) end in a consonant and are pronounced accordingly ; 
thus yiy tiger is read vagh and not vagha (or vagho). On the other 
hand those adjectives which add y, have retained and are pronounced with 
a final yj, thus WT® small is road chhota (or chhotd), but not clihot. 
The other fact which is the cause and explanation of the first one is this, 
(2), that those Bangali nouns which add yy occur in an identical form 
in the Hindi-class Gaurian and in Marathi; while those which add y 
correspond to nouns in y|f or yjT in the Hindi-class Gaurian and in Mara- 
thi ; e. g., yin tiger is yra in Hindi, Naipili, Panjabi, Siudhi, Gujarati and 
Marathi ; but iie small is WT^T or ilig ; W good is wyn or wyu, yy great 
is or <h%t, etc. in those languages. Now, as has been already shown, all 
such nouns ending in yjr or % and admitting an obligue form (in y[T or y 
in the Hindi-class Gaurian and yT in Marathi) belong to the Prdkritic 
element of the Gaurian and are formed from the 'particular Prakrit base in 
yr. Hence it follows that those Bangali nouns whose final yf is pronounced, 
are formed from the particular Prakrit base in yf ; while those whose final 
^ is not pronounced are formed from the general base ; and the final yj of 
the former is pronounced for the very reason because it is the remnant of 
the original ending yjyr. Take for instance the Bangdli adjective noun 
