128 
On the Letter ' k. 
dency to insert the semivowel y, as in Maria Andrews " to insert 
the semivowel r. 
Let us take first the simplest case of such hiatus. Ask a young 
child to spell get, and he will perhaps pronounce the three letters with 
the hiatus, gee ee tee ; but it is much more likely that he will say gee 
hee tee, or gee yee tee. In like manner in forming the present partici- 
ple of to see, to pity, there is a tendency unconsciously in speaking 
and especially in singing to compound see and ing into see-ying, pity 
and ing into pity-ying, and so on. 
Let us now turn to the other end of the vowel system. Again, we 
have no words ending in the pure sound of 6, which (if we have it at 
all) is found only before r, as in roar. The final o, or o followed by 
any other consonant than r, closes into oo as d does into e. (Sopho- 
cles writes our note as voovt). If then we take words that end in the 
labial vowel 66, as accrue, or in the diphthongal 6-66, like lo, how 
(almost all of which are written with ow), and let another vowel 
follow, then we find the euphonic insertion — or at least the tendency 
to the insertion — of the third or labial semivowel w between these 
vowels. For instance go out, go over, are sometimes unconsciously 
pronounced by emphatic speakers as gowout, gowover ; and in such a 
sentence as "the true ousel is the blackbird," there is as strong a 
tendency to say " true wou^el " as there is with some people to say 
Mariar Andrews. In like manner I have known a young lady who 
frequently corrected her young friends for calling her Lou-t<;isa, with 
one letter more than was agreeable to her ears. 
An instance of the addition of this euphonic r has been suggested 
to me by a friend as found in the pronoun her when used for she. 
The A. S. feminine pronoun was heo, which was probably pronounced 
with the e as in me, and the o as in move {hee-oo). Of this dyssylla- 
bic form sometimes one vowel was dropped, and sometimes the other, 
so that we find in Old English sometimes ho (or hoo), and sometimes 
he used in the sense of she. But when a word ends in 66 unaccent- 
ed; or in 6 (as has been remarked above), careless speakers are apt 
to shorten this into the d, as sounded in America. This kind of 
change hoo {— she) appears to me to undergo, and after its metamor- 
phose into hd, the adscititious r soon makes itself heard. 
But by the side of the fact that it is this sound of d (or u) to 
which the r chiefly attaches itself, comes the peculiar power which r 
seems to possess in English, of changing other vowel sounds into 
this one, especially in unaccented syllables. Compare, for instance, 
the final syllables of heggar^ father, Nadir, author, murmur, satyr. 
Nor is this the case only in final syllables. While I notice in boys a 
tendency to pronounce conditional as kinditional, procuring as procur- 
ing, or pricuring, this change does not take place if the short vowel 
is followed by an r, as in prorogue, where it either retains its proper 
sound, or takes that of w. Every one knows also that in such words 
