and that the notes
lack purity of tone.

If now, instead of Reed-Buntings in a marsh, I watch Yellow Buntings on
a furze-covered common, I find that, establishing themselves early in
February, they sing persistently, and in a few weeks are paired. But
what arrests my attention more particularly is the quality of the song;
for although the voice is unmistakably the voice of the Yellow Bunting,
yet it is incomplete and lacks the variety of phrases and musical
notation which we customarily associate with the bird. Nevertheless, as
the season advances, there is a progressive development in both these
directions, and by the end of March or the beginning of April the song
possesses all those qualities which appeal to us so forcibly.

There is one other fact to which attention must be drawn--the variation
in the song of the same species in different districts. As an
illustration let us take the case of the Chaffinch. In Worcestershire
the bird sings what I imagine to be a normal song--the notes are clear
and the phrases are distinct and combined in numerous ways. With the
notes fresh in mind I leave them and go to the west of Donegal, where I
am at once conscious of a difference; not a subtle difference that
perplexes the mind and is difficult to trace, but a change so remarkable
that one is conscious of a passing doubt as to whether after all the
voice is the voice of the Chaffinch; the song is pitched in a lower key,
certain phrases are absent, the notes lack tone and are sometimes even
harsh, and the bird seems wholly incapable of reaching the higher notes
to which I am accustomed.

Now the immature Reed-Bunting, though to our ears its song is but a poor
representation of that of the adult, gains a mate; the Yellow Bunting
pairs, and the discharge of the sexual function may even have taken
place before its voice attains what we judge to be its full development;
and there are no grounds for supposing that the Donegal Chaffinch, with
its less musical notes, has on that account any the less chance of
procreating its kind--facts which demonstrate that the biological value
of song is neither to be sought in the purity of tone, nor in the
variety and combination of phrases, nor, indeed, in any of those
qualities by which the human voice gains or loses merit, and which leave
us with no alternative but to dismiss from our minds all aesthetic
considerations in the attempt to estimate its true significance.

What, then, determin