104 A. F. Rudolf Hoernle —Essays on the Gaurian Languages. [No. 1, 
NOTES TO APPENDIX. 
Note 1.—M. = Marathi; B. = Braj Bhasha ; A. = Alwar dialect; 
N. = Naipali; G. = Gujarati; Mr. = Marwarf; H. — High Hindi. 
Note 2.—The Prakrit grammars allow only those forms of this gerund, 
which change the compound consonant W into (see Pr. Prak. YII, 38) ; as 
or to which Hemachandra adds also efrfy®# and ^ry«i But 
the Gaurian dialects seem to postulate two more Prakrit forms of that gerund ; 
viz. such as change the comp. cons. into or^sf; and such as change the 
connecting vowel ^ into ^ (see my note on p. 83, 84) ; e. g., besides 
also or or Now since writing the present essay, 
I have found that my conjectures are supported by the Pali of the rock 
inscriptions ; e. g., in the Dliauli inscription occurs the form and in 
the ordinary Pali or ^t]® 4 besides (see Dr. Muir’s Skr. Texts, 
Yol. II., p. 113, and Dr. Mason’s Pali Grammar, p. 90). This is all the 
more important, as, no doubt, the Pali of the inscriptions represents much 
more closely the spoken language than the Prakrit of the grammars, which 
may have sacrificed sometimes the established hut irregular forms of popular 
usage to the uniformity and regularity of a fancied rule. 
Note 3.—The forms and I have given on the analogy of two 
sutras in Subha Chandra’s Prakrit Grammar (Adhy. I, Pada I, sutra 14. 
15.) : f^cT || i. e., whenever the technical letter w is added, 
an anunasika must he pronounced ; and II 1 
i. e. in the (four) words yamuna, cliamunda, atimuktaka, kamuka ip must 
be pronounced as an anunasika; e. g., ^#WT, etc. Perhaps we may 
assume,that in later and vulgar Prakrit the elision of consonants generally was 
compensated by the pronunciation of anunasika ; and this conjecture might 
afford us another explanation of the puzzling final anunasika of the neuter 
oblique form in Gujarati and Panjabi. E. g. if the elision of ^ should be 
compensated by anunasika, we should have the Gen. for ; 
and would change to %T^Nt, and finally to This explanation, 
perhaps, appears less forced than that given above in the text p. 85, SO. 
Note 4.—In the text (see above p. 60) I have explained that the Prakrit 
Genitives in ^lyr, as drop the final and change to In 
support of this theory compare the remarks of Beames in his Comp. Gram, 
of the Modern Aryan languages of India p. 259., which I have received in 
the meanwhile. The only example given there is Skr. cffqw which in 
Panjabi is WY, but in Orlya 3\qT. A still more apposite evidence of my 
theory has since occurred to me in the Ganwari (Hindi) oblique form of the 
near demonstrative pronoun which is y or and corresponds to the Braj 
