134 Bird-Song : and New Zealand Song Birds 
The top singer occasionally fell out of time in the same way as 
the bottom singer, but the time was always caught again. This 
was varied on the 4th January as shewn in (68). 
This w T as very beautiful, and apparently no more than two 
birds were singing, though a third made a sort of bass accom¬ 
paniment with tiu tiu uttered at intervals. The lower notes 
were rich, deep, and bell-like, the higher thinner, whilst still 
bell-like in quality. Their high pitch was quite evident when 
heard in contrast with the c of the lower voice. Whilst a great 
many birds might be singing at one time, they appeared to seg¬ 
regate into parties of two, three, four, or five, the song of each 
party or choir being in harmony, whilst it might not exactly 
harmonize with that of others. The time, too, was practically 
exact. At dawn the other birds also joined in,—white-heads, 
fantails;—and an English lark was heard among the others. 
Whilst the song of these did not altogether harmonize with the 
larger birds, the wild chorus resulting was by no means un¬ 
pleasing. It wms impossible to say if the singers were tuis or 
bell-birds; the notes were common to both; but the harmonies 
are considered with the song of the tui since the call-birds cer¬ 
tainly were tuis, and most of the calls ceased on the bell-notes 
beginning. On the 8th January the birds were singing in four, 
sometimes five parts. 
Two other duets recorded on the 10th and 13th January are 
here noted, the second being particularly beautiful when heard 
in the quiet of the early dawn, (69) and (70). 
The songs w T ere continued indefinitely; no repeats of parts 
that I could detect; each bird apparently sang ad lib. There 
were other singers when (70) was noted, but the ones that sang 
(70) were the chief performers among those in the immediate 
