The idea of _stretching_ signified by the original root _tan_ has no _direct_ or _immediate_ connection with any of the conceptions expressed by the derivative words. But by stretching an object it is diminished in _breadth_ and _depth_, while it increases in _length_; hence it becomes _thinner_; so that the Mind readily makes the transition from the primitive conception of _stretch_ to that of _thinness_, indicated by the English word, and by the Sanskrit _tanu_, and the Latin _tener_, _tenuis_. _Thinness_, again, is allied to _slimness_, _slenderness_, _fineness_, etc.; ideas which are involved in the conception of _delicate_, and furnish an easy transition to it. But it is also from the notion of _stretching_, though in a still less direct manner, that we gain an idea of sound as conveyed by musical tones; '_tone_,' as Professor Mueller remarks, 'being produced by the _stretching_ and vibrating of cords.' Still further: if we cause a heavy piece of cord to vibrate, or, what is better, the bass string of a violin or guitar, or strike a very low key on the piano, and pronounce the word _tone_ in a full voice at the same time, the remarkable similarity of the two sounds thus produced will be clearly apparent. Thus the root _tan_, to stretch, becomes also expressive of the idea of _sound_ as seen in the words _tonos_, _tone_, _tonitru_, _thunder_, etc. But what is especially to be noticed is this: that in those derivatives of _tan_, to stretch, which are _not_ indicative of ideas of sound (as _tenuis_, thin, etc.), the sounds of the words do _not_ cause us to imagine that we hear the imitation of noise; while in those derivatives which _are_ expressive of it, we not only imagine that we _do_ hear it, but, in the case of _tonos_ and _tone_ at least, have an instance in which we _know_ that the word employed to convey the idea is a proximately perfect representation of the sound out of which the idea arose. Even in _tanyu_, _tanyatu_, _tanayitnu_, thundering, in which Professor Mueller affirms that 'we perceive no trace of the rumbling noise which we imagined we perceived in the Latin _tonitru_ and the English _thunder_'--although he seems to admit that it is perceptible in the Sanskrit word for thunder expressed by the same root _tan_--the reason why we cannot trace it may be because of the terminations, which, as it were, absorb the sound that is there, although less obviously, in the _tan_, or shade it off so that it becomes diluted and hardly traceable.