ERASILIAN SAW-BILLED ROLLER. 
standing the impossibility he states, of convey- 
ing bywords a distin6t Idea of the intermingled 
colours of this bird, it had been previously 
done, with good efreft, by our countryman 
Edwards, whom BufFon has not noticed on this 
occasion. The variances, however, v/hicli 
he hints, as affedled by age or sex, must be 
allowed to prevail ; and, we may add, perhaps, 
some slight local distin6tions, v/liich enable us 
to reconcile the differences that occur in the 
several descriptions of Edwards, Marcgrave, 
Brisson, &c. who certainly could not have 
before them the sam.e identical bird, though 
each of them, m.ost undoubtedly, had a bird of 
the same species. 
We may conclude that, comparing the ac« 
counts of this bird with those of the Rollers, 
it will be found sufficiently similar, in many of 
the characteristics, to justify Edvv^ards in the 
opinion of it's affinity. 
For example, BufFon tells us, " the Roller 
is more wild than the Jay or the Magpye, and 
settles in the thickest and most solitary woods. 
It's plumage is beautiful ; it has an assemblage 
of 
