6 On the Power of the Letter p. [No. 9, new series. 



hard pronunciation. Thus lcswt 0 onus becomes Lcil^snLo ; and 

 Quit ear 0onm becomes QuirpjBesiLa. 



Forasmuch then as these particular changes are applicable to 

 the 4 n's and * t's alone, with only the doubtful exception of p to 

 be determined, these combinations suggest that this p also is a 

 surd of the 4 1' family, leaving its power yet to be ascertained. 



ii. When a word ending in the cerebral 4 V (or) is followed by 

 a word commencing with a dental 4 1' this latter letter is 

 changed into a 4 1' of the same power as that of the preceding 

 final in order that both letters may be pronounced consecu- 

 tively from the same 4 birth-place,' and the hiatus be avoided 

 which must otherwise occur : and for further euphonic assimila- 

 tion of the pronunciation, the final 4 1' is changed by attraction 

 into the same 4 1' as the changed initial 4 1' of the second word. 

 Thus cpsrr becomes QpiLiz.^. The same process is necessary 

 in the case of the second word having an initial 'n.' Thus qp& 

 jsear^J becomes Qpe^remesr^t. 



The same rules hold good when the final consonant of the first 

 word is the second 4 F instead of err. Thus for &®> fieSr^u we 

 have dsoa-eweSr^ ; and for &&) 0^j we have &p$jp. 



Forasmuch, therefore, as in the instance of the final cerebral 4 V 

 («r) the following initial dental 4 t' (&) was changed into a celebral 

 4 t' (*-), analogy suggests that our letter p into which the same 

 dental 4 1' is changed after the other 4 1' is likewise a letter 

 of the 4 t' family, and of similar power with this 4 1' (go). 



iii. In the case of a final 4 1' occurring before an initial surd 

 consonant in the succeeding word, this 4 1' — (the dental 4 1' (p) is 

 the only one that can well occur in such a position, and that almost 

 exclusively in words of Sanscrit origin — ) is changed euphonically 

 into p. Thus <b=jS (5(75 becomes <Fp(&jqj). This also suggests a 

 family connexion between p and p. 



4. The letters ^, and p, with their corresponding nasals, 

 are the usual sign-letters of the past tense of verbs. We have here 

 again a corroboration of the suggestion that these three letters be- 

 long to the same family. 



5. The fact of the surd consonants which correspond with the 

 dental and cerebral 4 n's respectively being 4 t's further suggests 



