50 
SAGAETIADiE. 
There is no doubt that S. miniata and S. rosea approxi- 
mate in some of their varieties very closely ; and I have 
had many doubts about the propriety of keeping them 
separate. I have seen, in the vicinity of Tenby, specimens, 
in which some of the small tentacles of the outer row had 
a scarlet or orange core, and yet in no other respect could 
I distinguish them from the true rosea. Normal rosece and 
o 
normal miniatce were abundant on the same rock (the 
Woolhouse-rock) within a few feet; which fact suggests 
the possibility of hybridization. Besides the scarlet-cored 
tentacles, miniata may be described, in those varieties 
which come nearest to rosea, as darker externally ; as 
growing to a far larger size ; as being lower and less pillar- 
like ; and as having a much more lax, flaccid habit of 
body. 
The qucestio vexaia, — What constitutes a species? what 
a variety ? is one which it is much easier to answer theo- 
retically than practically. Some have proposed certain 
arbitrary canons, such as that assumed by Mr. Tugwell, 
i\\xtfo7'm distinguishes the species, colour only the variety. 
But this is quite untenable. In many instances colour is 
not only specific, but even generic ; — as black, white, and 
red, in well-recognised patterns and in certain fixed regions 
of the body, in the Woodpeckers ; black, yellow and red, 
again in certain patterns, in Paqnlio ; yellow, red and 
white in the Pieridce. Indeed, our entomological friends 
would be sorely puzzled to define their species, if colour 
were denied them as a distinction. In the Butterflies 
alone, hundreds of indubitable species rest exclusively on 
colouring. The fact is, anything may be a specific character, 
provided it be constant. Constancy, permanency, is wliat 
we require ; let us only indicate any mark that is invariahly 
found, — no matter whether it be colour, form, pattern, 
surface, sculpture, or any thing else ; or any combination of 
