25 
-for w by Cockneys, like the immortal Weller. Some Australian 
dialects have only eight consonants in their alphabets ; the American 
Tinnes have 63. Prince Lucien Buonaparte gives 390 as the maximum 
of phonetic sounds in all the languages of the world. The sibilants 
offer difficulties to many persons, but these lispers do not emulate 
certain modern savages, whose lisp results from the wilful filing away 
of front upper teeth and the removal of four lower ones. Among 
Pacific Coast tribes the adoption of nose and lip ornaments interferes 
with the enunciation of the nasal and labial consonants. These are but 
trifling instances of the effects of custom on language. The is no d in 
Chinese, and no classical Chinaman ever pronounces the letter 7; he 
says “ Eulopa” instead of Europe. Mazdarin, therefore, is not a true 
Chinese word, being derived from the Sanskrit mantrin, a counsellor. 
The Huron and Mohawk languages of North America possess no labial 
consonants whatever. The gutturals, entirely absent in the dialects of 
the Society Islands, abound in Semitic tongues, and often form the 
basis of the triliteral or three-letter consonantal roots characterising 
that ancient group of languages. 
There is no Zin Zend or Old Persian, and the initial s is always re- 
placed by A, as in the Sanskrit word samd, for summer, which becomes 
the Zend hama. Hence, in transferring the Sanskrit name of river 
sindu,it becomes Hindu. The Greeks passed it on with the soft aspirate 
and so we get our name India as the equivalent of sindhu, or the 
country of the seven rivers. The Aryan group of languages is also 
characterised by a regular alternation of consonants, which is known as 
Grimm’s law, after the philologist who formulated the important dis- 
covery. No etymologies can be regarded as sound if these phonetic 
changes have not been fully considered. They are, however, quite 
distinct from the irregular “ phonetic decay ” to which the consonautal 
sounds are especially subject in all languages, and which result mainly 
from laziness and a desire to pronounce words with the least effort. 
For example, our word “speech” should be spreech, as it is derived 
from the A. S. sprecan, O. H. G. spreckan. It lost the “r” about the 
llth century and is certainly pronounced more easily without it. The 
laws affecting consonantal change and vowel variation have not yet 
been fully worked out in the Asiatic, American, Polynesian, and 
Australian languages. In the African Kaffir a certain regularity has 
been noted already by Bleek. It is possible that consonantal variation 
may be somewhat affected by physiological causes. Experiments have 
recently been made at Boston, U.S.A., among the boys of the Latin and 
High Schools, ranging in age from 12 years to 20, which tend to shew the 
occurrence of a large percentage of deficiency of sound perception, which 
has been rather absurdly named “sound blindness.” The fact that a 
good ear is as essential to a linguist as to a musician or a vocalist has 
_ scarcely received the attention it merits from the scholastic profession. 
We must now dismiss from our minds the idea of language as we 
are accustomed to conceive it ; for the investigations of philologists 
reveal the indubitable fact that primitive language was scarcely in- 
telligible. Its phonetics were certainly variable in the extreme, and it 
mattered little how sounds were pronounced. There was, moreover, a 
stage in its development when the word did not clearly define the idea,or 
