If we once admit the principle of onomatopoeia, the difficulty remains of drawing the line which shall define the territory within which those capable of judging would limit its operation.  Its boundary would be a movable one, like that of our own Confederacy.  Some students, from natural fineness of ear, would be quicker to recognize resemblances of sound; others would trace family likeness in spite of every disguise; others, whose exquisiteness of perception was mental, would find the scent in faint analogies of meaning, where the ordinary brain would be wholly at fault.  In the original genesis of language, also, we should infer the influence of the same idiosyncrasies.  We were struck with this the other day in a story we heard of a little boy, who, during a violent thunder-storm, asked his father what that was out there,—­all the while winking rapidly to explain his meaning.  Had his vocabulary been more complete, he would have asked what that winking out there was.  The impression made upon him by the lightning was not the ordinary one of brightness, (as in blitz, (?) eclair, fulmen, flash,) but of the rapid alternations of light and dark.  Had he been obliged to make a language for himself, like the two unfortunate children on whom King Psarnmetichus made his linguistic experiment, he would have christened the phenomenon accordingly.

Mr. Wedgwood has by no means carried out his theory fully even in reference to the words contained in his first volume, nor does the volume itself nearly exhaust the vocabulary of the letters it includes (A to D).  Sometimes, where we should have expected him to apply his system, he refrains, whether from caution or oversight it is not easy to discover.  The word cow, which is commonly referred to an imitative radical, he is provokingly reserved about; and under chew he hints at no relation between the name of the action and that of the capital ruminant animal.[a] Even where he has derived a word from an imitative radical, he sometimes fails to carry the process on to some other where it would seem equally applicable, sometimes pushes it too far.  For instance, “Crag. 1.  The neck, the throat.—­Jam.  Du. kraeghe, the throat; Pol. kark, the nape, crag, neck; Bohem. krk, the neck; Icel. krage, Dan. krave, the collar of a coat.  The origin is an imitation of the noise made by clearing the throat.  Bohem. krkati, to belch, krcati, to vomit; Pol. krzakae, to hem, to hawk.  The same root gives rise to the Fr. cracher, to spit, and It. recere, to vomit; E. reach, to strain in vomiting; Icel. hraki, spittle; A. S. hrara, cough, phlegm, the throat, jaws; G. rachen, the jaws.” (As crag is not an English word, all this should have come under the head of craw.) “Crag. 2.  A rock.  Gael. creag, a rock; W. careg, a stone; caregos, pebbles.”  We do