136 
G. A. Grierson —Essays on Bihari Declension. 
[No. 2, 
It therefore appears possible that this Bihari locative in y is the Skr. 
and classical Prakrit loc. suffix y, which has been preserved unchanged. 
The fact that it has remained unchanged in the modern languages need 
not surprise us: for it has admittedly remained unchanged in the classical 
Prakrits, while all the other case suffixes have changed in them. If there¬ 
fore the locative has retained vitality so far, it need not astonish us that 
it has retained it to the last. 
Note on the above. It will be seen that in these derivations of y and 
y, I have given an etymology different from that put forward by so high 
an authority as Dr. Hoernle in his Gaudian Grammar, §§ 367 and follow¬ 
ing. Dr. Hoernle considers f asa contraction of nf#, and ^ as a con¬ 
traction of yjfyr respectively, terminations which have already been dis¬ 
cussed and disposed of. This derivation is also plausible, but I venture to 
think that an equal amount of plausibility attaches to the derivation given 
above, on the following grounds. 
Lassen (p. 461) connects the termination y with Skr. term. y*r, but 
Dr. Hoernle considers that this is untenable because Skr. y«r cannot 
be added to feminine bases in ^H. This point has been already discus¬ 
sed by me, and I need not repeat what I have said here,—but, admit¬ 
ting for the sake of argument that Lassen is wrong, Dr. Hoernle’s 
theory is also open to objection. Dr. Hoernle takes the termination 
and supposes an elision of which gives which is contracted 
to y; hence he gets the forms in Apabhramsa Prakrit, ^y, ^Typy, 
^fjTjy, and 3 pay, which he derives from supposititious forms # ^yry)f?, 
and which he considers old genitives. 
It thus appears that the letter yj in the termination is absolutely neces¬ 
sary for the theory ; only can be contracted to y; if the termination 
ever takes the form it must be contracted to and if it takes the 
form it can only become ^3^. It must be remembered that we are 
only dealing with weak bases, for in the modern languages, y is only 
added to weak bases, and never to strong ones and it remains 
to be seen what form takes with weak bases. Really, this termination 
is f? and not y|f% and the vowel ^ is only the termination of the base, 
* This may he denied ; but the fact remains that some weak forms do undoubtedly 
take the termination y ; e. g. qjrT <a word,’ instrumental so also y t ^ and 
which are undoubtedly formed from a weak bases, and this is quite sufficient for my 
argument. If I can prove that a single weak form takes y' in the instrumental, it 
does away with the argument that ^y' or ^y can in all cases he derived from 
It may he noted here that in the Magadhi dialect of Bihari only 
masculine weak forms ending in a silent consonant take y, and y • thus, from 
‘force’, instr. hut never yfcjy, or 
