1G2 Hoemle — Essays on the Qaurian Languages. [No. 2, 
explained by the same principle by which the similar difference in the Hindi- 
class Gaurian was explained ; viz., that it is caused by a difference in their 
resp. derivative bases. The nouns in ^ with an oblique form in ^t, namely, 
are derived from the general base of the word in W, but the nouns in ^jT 
with the oblique form in from the particular (Prakrit) base formed by 
the affix eff (or more accurately as will be shown hereafter, ; e. g., VPFf 
is derived from the general base (Sanskrit vil), but hr^rr from the 
particular Prakrit base i)Tr?3F (== as = «fFT^r) or Sanskrit 
The full discussion and proof of this important principle I must defer till I 
come to the explanation of the nature of the oblique form of the inflexional 
base of nouns with which it is closely connected. Here I will only note 
that a parallel phenomenon is exhibited by the neuter nouns, some of which 
terminate in ^5, others in *3? or V or ^ . Of these the former correspond to 
the maseul. nouns in ^ ; as the final fT of the latter is a modification of 
the Prakrit % and Sanskrit ^f:, so the final "VJ of the former (the neuter) is 
a modification of the Prakrit (neuter) termination ^ and Sanskrit ; 
again both the masc. and the neuter nouns in ^ make equally their oblique 
form in "=5r. On the other hand the neuter nouns in if, gj, correspond to the 
maseul. nouns in and form like the latter their oblique form in *jt or ^t, 
and are also, like the latter, derived from the particular base in srr or rather 
T*fi (perhaps 'a^F ) ; e. g., flnff pearl. = Prakrit wfwFT ( = for Sanskrit 
Tlifi p ) 5 TTO branch of a river = Prakrit (or xfp|a|f, see Pr. Prak. 
Ill, 36. = Sanskrit ; W tear = Prakrit vksdi (cf. Pr. Prak. IV. 
15. for Sanskrit p ^f), etc., etc. 
We now proceed to the investigation of the inflexional base in Bangali 
and Uriya. These differ from the other Gaurian languages in not possessing 
an oblique form at all. Nevertheless it is probable, that these languages are 
not altogether destitute of a Priikritic element. In Bangali there are two 
post-positions for forming the gen. case ; viz. Vt and Of these vv must 
be the original one, for we can imagine vv; having been, in the cours'e of 
nouns in ijr^f it appears, that they likewise are declined like ordinary nouns in 
5E|f. As regards nouns in HTf^T Dadoba’a statement is confirmed by the Manual itself; 
or on p. 63, rule 112. it is said that all part. adj. in as those ending in fj-y ) 
■'ffT^FT, are declined like ordinary ndjoctives in i. e., making an oblique form in qp. 
If nevertheless the first statement of the Manual should be correct, the anomaly 
is to be explained thus ; the oblique form of these nouns is not identical witli the 
direct form (that is to say these nouns do not belong to the proper Qaurian element) 
but with the oblique form in ^p of nouns in m . Their case is a similar one to that 
of some nouns in q, which has been already explained. Their anomaly consists in 
this, that their direct form which originally ended in ^ p was not modified to sg as 
demanded by Gaurian law, but only to 
