486 Note on the identity of the Mathiah Lath Inscription [Oct. 



some notion of the terminations and inflections of the words. Thus 

 where we perceive an instance, (and many such occur,) of five or six 

 consecutive words ending in the same letters, we may fairly presume 

 them to be connected in case and gender, like the long compound 

 epithet of the second inscription described by Dr. Mill, (p. 260.) 



6. The characters most often forming the termination of words in 

 the Delhi text, are £ and X > °f both of which upwards of 40 in- 

 stances occur. Next to them in frequency, come X > "H > an d At > 

 about 20 of each : then <j and <S > \r an d "^r , about a dozen each: 

 the other letters are comparatively rare as finals. It may be remarked, 

 that the vowel inflection, which has been set down as e, is affixed to 

 most of the final consonants, affording another argument in favor of 

 the language being Sanscrit. 



7. The order in which the inscription should be read is wrongly 

 given in Captain Hoare's plates, where he makes the east portion 

 follow that of the north. That the north is the proper commencement 

 is proved by its being the uppermost of the Allahabad column ; then 

 follow the west, the south, and the east respectively. 



For convenience of reference, I may here remark, that the first 

 eight lines of the Allahabad Lath inscription include to the third letter 

 19th line, Plate X., Asiatic Researches, vol. vii. They are here cut off by 

 the Persian inscription. The following half line, partially clipped on the 

 upper surface of the letters, begins with the eleventh letter of the four- 

 teenth line, Plate XIII. of Delhi. The next three lines finish the same 

 plate ; but three letters are missing from the beginning of each line 

 (owing probably to the peeling of the stone). 



The three following lines (13, 14, 15,) correspond with the com- 

 mencement of Plate XII., and also with the uppermost part of the 

 Bettiah inscription in the present plate ; the three or four initial letters 

 of each line are here also cut off by some accident. 



Line 20 of Allahabad begins with the sixth letter of Plate XI. of 

 Delhi, and the detached portions of the neighbouring lines may easily 

 be found in their respective places. 



In the second half of the Bettiah inscription (which should come 

 first in the order of reading), one circumstance tends very much 

 to perplex the comparison with that of Delhi, which is, that from 

 the last letter of the 20th line onwards, the native copyist (at 

 least I imagine the fault must be his) has transposed every half line of 

 the text, placing first what by the Delhi column should be the last half 

 of each line. This defect I have attempted to correct by placing inter- 

 mediate figures over the first letter of each transposed passage : thus, the 



