On the Letter * r. 
127 
and the inconvenient gap between the vowels is at once occupied by 
an obliging r; "one winder of the house." Perhaps he even vi- 
brates the tongue, and we have Ben Jonson's dogs letter" in full 
force ; but this, so far as m}- observation goes, is the exception rather 
than the rule, and the euphonic r is almost always the semi- vowel or 
non-vibrant letter. 
I beheve however that we shall find phenomena precisely analogous 
to this in the employment of the other semi-vowels — phenomena less 
noticeable, but not the less actual. 
The pure vowel sounds of our language being the twelve which are 
exhibited in the following series of words — cheap, chip, pate, pet, Sam, 
psalm, nought^ not, nut, note, pull, pool — the insertion of the euphonic 
r occurs most frequently by far when a word beginning with a vowel 
follows one ending in the ninth of these sounds, viz. : the a of Rosa, 
Musd, &c., which is in pronunciation identical with the u of nut. 
The r is also occasionally added after each of the middle pair of long 
vowels, as : *' Is yom* papar upstairs ? " " I am going to draicr on 
card." This addition sometimes takes place where the long vowel is 
followed by a consonant, as when the A.S. lacan, to play, gives origin 
to our word lark= sport, and when St. Ann's Chapel, in the neigh- 
bourhood of Plymouth, changes its name to Turnchapel; but in both 
of these cases the word seems to have been erroneously identified 
with another totally different. - 
Returning therefore to the purely euphonic insertion of the r, we 
may notice that we have no words in the language which end in the 
f oi pit, the e of met, the d of mat, the o of not, or the u of pull, so we 
may pass these by. But turning to the long vowels we find again that 
no words end in the pure sound of d, which we have only (if we have 
it at all) before r, as in mare. The sound of the long d as a final (in 
which case we write it au), or before any other consonant than r, is 
really diphthongal, as we close in the sound towards the end into e. 
(This peculiarity of the English pronunciation has been nowhere 
more clearly stated than by Sophocles in his Romaic Grammar, where 
he represents our fate in Greek letters as ^ar, though doubtless he 
exaggerates somewhat in so doing). 
Now let us observe what happens when an hiatus follows a word 
ending in the pure sound of e, like decree, or in the diphthongal d-e, 
like day. Such an hiatus will exist whenever another e or the i follows 
such words ; and in such cases, I think we shall find as strong a ten- 
* I must leave it to those who are better versed than I am in the Semitic 
languages to deal with those Syriac aud Samaritan words which contain an r, 
that is wanting in the cognate Hebrew forms.- To such scholars, I would svig- 
gest the enquiry whether in such cases the earlier form is the one without the 
r, (as seems probable), or the other, and on what principle the insertion (or 
omission) can be accounted for. 
a 
