[HELICONIUS] Parallel evolution

[James Mallet]

I thought it may be of interest to publish my reply to this email from 
Roger Thorpe from Bangor, Wales.

On 01/05/2013 08:52, Thorpe,Roger Stephen wrote:
> Hi James
>
> I hope you dont mind me asking a simple-minded question.
>
> I,m at the stage of trying to write up a large scale study of parallel 
> geographic radiation. Its a fairly comprehensive study of the Lesser 
> Antillean anole radiation. I,m looking for other comprehensive studies 
> of parallels/convergence at the geographic variation level. It occurs 
> to me that mimicry rings may come close but I,m not sure how many of 
> the species in a radiation are involved. In my anoles Ive managed to 
> look at every mountainous island species but one, and there are some 
> universal parallels in GV. 

> In butterflies contributing to mimicry rings do all/most of the 
> species in the genus take part or is it just a couple per ring.

Answering this final question, I'd say that the situation is probably a 
bit different from Anoles.

The evolutionary acquisition of mimicry seems to act as a major shift 
akin to "entering a new adaptive zone," and most species that belong to 
a clade that has mimicry are indeed mimics. I'm not so sure about 
non-Heliconius groups, but I think that this general rule applies 
similarly to other butterflies.

When they gain mimicry, butterflies seem often to gain the ability to 
mimic multiple other species or races. They seem to gain evolutionary 
colour pattern flexibility. Most (but not all) of the species of 
Heliconius, and its sister genera Eueides + Laparus + Neruda are close 
mimics of some other species, or, importantly, are mimicked themselves 
by some other species. However, this doesn't seem to stop a diversity of 
patterns arising; quite the contrary, they seem to switch mimicry 
allegiances rapidly.

In the Eueides-Heliconius clade as a whole, approximately 9 species are 
putatively non-mimetic, out of 54 total species in Brown 1981 (newer 
taxonomy has added more species, but the percentage should be about 
right, i.e. 17%).

However, even in these 9 species, there are three: H. hortense, H. 
clysonymus, H. ricini, and in addition there is a newly recognized 
species, H. himera. All 4 of these today have patterns which are 
similar, but because they're allopatric, they cannot easily be explained 
by ongoing mimicry. However, these species may well retain ancestral 
patterns that were involved in mimicry, but have been superseded. They 
all have yellow forewing bands and red hindwing bar patterns.

It seems possible that mimicry explains the colour patterns even of all 
of these exceptions, although historical mimicry has not been fully 
erased or updated in some species.

I hope this helps, Jim

-- 
James Mallet
OEB & GEE Depts
Harvard University & University College London
Cambridge, MA 02138 and LONDON WC1E 6BT
USA tel: +(1)617-496-5350
www.ucl.ac.uk/taxome/jim