[Taxacom] Dispersal clarifications

[Michael Heads]

Hi Jason,
 
Thanks for the long letter. I've added my replies below.
 
Michael
 

Wellington, New Zealand.

My papers on biogeography are at: http://tiny.cc/RiUE0

--- On Sat, 11/6/11, Jason Mate <jfmate at hotmail.com> wrote:


From: Jason Mate <jfmate at hotmail.com>
Subject: [Taxacom] Dispersal clarifications
To: "Taxacom" <taxacom at mailman.nhm.ku.edu>
Received: Saturday, 11 June, 2011, 12:04 PM



Sorry for the delay Michael.  In regards to your last email:
I am still waiting for a working definition of your ´simple movement´.  
 
MH: I haven't offered a definition of physical movement, but is this really necessary? (In any case, I don't believe definitions are so important, it's concepts I'm interested in). 

 
Til now you haven´t really defined it, rather you have offered a gut-feeling of what you think it is. Movement, IMO, is a range of distances travelled and probabilities, for each individual organism out there. So why make a special, compartmentalised category for the long distance ones? 
 
MH: I'm not. You're confusing two very different concepts. Albatrosses move long distances but that is nothing to do with the concept of 'Long-distance dispersal' as used in dispersal theory, which is a mode of speciation.
 
 
You seem to define it based on your personal observations, e.g.: "One of the 'weeds' that visits my garden is a native honeyeater (meliphagid) bird, Prosthemadera. There is a seasonal migration on a local scale, probably over 3-4 km. I haven't worked out the details yet, but you could. Wikipedia says: 'The movements of honeyeaters are poorly understood. Most are at least partially mobile but many movements seem to be local, possibly between favourite haunts... It seems probable that no single explanation will emerge: the general rule for honeyeater movements is that there is no general rule'.  I don't really believe this (it sounds like chance dispersal!), the problem just hasn't been worked on enough. A meliphagid in southern New Zealand (Dunedin) seems to migrate in a similar way to the Prosthemadera." I have no doubt that if you were to look at the movements of all the Tui´s in your neighbourhood you would find, with a sufficiently large dataset,
 the odd long range trips.
 
MH: Of course, it's a normal distribution. I'm more interested in the overall pattern than odd individual data points. 

Dispersal per se is not a single step but is dependent on several factors, both internal and external to the organism. Internal ones would be the vagility of the species or population and the physical condition of each individual. 
 
MH: it depends what you mean by dispersal. Are you referring to normal ecological movement or a mode of speciation? If the latter, evolvability would be the mkey interbal factor. 
 
External ones would be the size of the source population, the distance to the target (not intentional of course, there is no grand plan to disperse to greener pastures) the size of the target (hitting Madagascar vs the Canary Islands), atmospheric variables (main currents, weather conditions, etc). And even if you get there there are yet more variables: lack of mates, habitat heterogeneity, competition. So colonising a new area (an island for example) involves a great number of variables on which ultimate success depends, surely a small chance. The probability for any given species if the aggregate of all the individual probabilities, not just the behaviours that you may observe outside your window, although in a way it is a beginning. On the other hand you mention: "I don't think 'chance dispersal', with speciation, sensu Mayr etc., exists." Do you mean that a species that colonizes an island cannot diverge, in time, to the point of becoming a distinct
 species, even an endemic? 
 
MH: Of course not, it happens all the time. But I don't think it happens by 'chance dispersal' sensu Mayr etc. as this would produce individual, idiosyncratic patterns rather than what we see in practice on islands: common patterns shared by many different taxa with completely different ecology and means of dispersal.
 
What particular mechanism would prevent this from happening? To which you add "all clades are endemic to some area or other, but the weed that makes it to my garden (an island of suitable habitat, just like an island in the sea) doesn't become a local endemic there." Well obviously not, dispersal to your garden is easy, genetic exchange has been hardly disrupted. 
 
MH: That's the whole point. What is the mechanism disrupting genetic exchange in chance dispersal theory? Nothing - just 'chance', an accident: 'dispersal over a barrier'. 
 
 
But still you continue "Range expansions don't usually involve one-off dispersal events by single individuals, but populations or, usually, whole communities. Range expansions occur for a reason - change in climate or whatever - and these factors affect the whole community." There are so many ad hoc assumptions in this remark it is difficult to discuss it. In the first place, communities are not monolithic blocks that respond equally to external pressures. 
 
MH: I didn't say they were. The point is that it is not just one individual weed that colonizes a garden or a new volcanic island. It's a whole community of weeds.  
 
 
I think that the biodiversity conservation end ecological literature has sufficient examples to show that although closely related species often respond the same to external pressures communities don´t. As for how many individuals are needed to start a colony, tramp species don´t appear to need many, but I guess you can never tel exactly how many. Nevertheless I agree that the vast majority of dispersals fail, simply because they end in some sort of sink (ocean, non-vegetated atoll) or an island where the ecological conditions of the species are not met. But we can make some guesses as to which dispersals are more likely. Polar bears would have a tough time settling in the Bahamas but I doubt Drosophila would be as challenged. Hence some groups will be better at dispersal than others. In fact if you look at the faunas of oceanic islands you see some groups, i.e. mammals, that are simply no good at it whereas others like birds are. 
 
MH: There are plenty of large mammals on islands. Hundreds of meters inland on Auckland Island (south of New Zealand) one time, I was stumbling through the forest  when I was shocked to be confronted with a huge, angry carnivoran - a so-called 'marine' mammal (sea elephant). I had to keep reminding myself: 'There are no large mammals on oceanic islands'...
   Many groups, such as cursorial mammals, cannot survive as a metapopulation in a dynamic archipelago (with islands coming and going), hopping from island to island, because their normal 'means of dispersal' are not up to it. ('Normal' means of dispersal are the obvious ones; this is a very different concept from 'means of dispersal', as used in dispersal theory. Those 'means of dispersal' are often so rare that they have never been observed and are very mysterious. They are so special that they often operate just once in tens of millions of years). The fact that cursorial mammals are not on remote recent islands because of their normal ecology does not mean that the groups that are there arrived by long distance dispersal from the mainland. They may represent a metapopulation surviving around an area of volcanism in situ.   
 
Still, you bring out your big guns and argue that dispersal is a lame duck because, "There are plenty of good fliers on both sides of the Atlantic - spore plants (bryophytes, fungi, ferns), flies, birds, plants with wind- or bird-dispersed seeds, lepidoptera, many beetles, etc. - and there are plenty of storms etc. that could blow organisms across the Atlantic. So after a few centuries, let alone a hundred million years you would expect all these groups to be more or less the same on both sides of the Atlantic - but they are not."  Why would you expect this? In fact, since the Atlantic ocean has been expanding for 100my you´d expect early dispersal to be vastly more common than now, progressively becoming rarer until it reaches the current situation, where only the most powerful fliers on a lucky break make it (i.e. D. plexippus). 
 
MH: But over millions of years extremely unlikely events become almost certain. This is emphasized by all the highest authorities on dispersal theory and is the statistical basis of the theory.  
 
The point of course is that the window of opportunity to disperse closes at different rates for all organisms so many (maybe the majority) rarely get the chance to do it. In fact some can´t do it even when it seems impossible not to disperse:"The same thing happens at narrow breaks like Wallace's line, or between very windy mountains in New Zealand with endemic daisies etc. In my experience, trying to correlate distributions with what is known about means of dispersal doesn't work." I cannot speak for the flora of New Zealand but I can vouch for the dozens of butterflies and scarabs species  that span over the Wallace line. The point is we can find examples to support both mechanisms. 
 
MH: A clade's presence on both sides of Wallace's line does not support dispersal, it just supports lack of differentiation.  It's like saying a group at A and B implies dispersal from one to the other, but in fact it might have got to A and B by evolvng there. If you have a very precise break like Wallace's line in many (not all) scarabs, lepidoptera, plants,  fish,  birds, mammals,etc. it's probably an interesting phenomenon, not just an accident of 'chance dispersal'. 
 
Some organisms can disperse, others can´t. I am sure that if we looked at the natural history of one group and another we would find wildly different reasons (competition, disease, temperature...).
 
MH: Those factors may explain why sister groups, for example, don't invade each other's territory. But it doesn't explain how they differentiated to start with. Dispersal or vicariance? 
 

"Nevertheless, a common argument goes: 'I saw species A in my genus move, therefore the biogeography of the species and the genus can be explainedby movement'.  This is like the sun going around the Earth - the appearance may not be the reality." Agreed, but it works both ways. Just because you have a pattern that appears vicariant it doesn´t mean it is the case. 
 
MH: You can't explain a pattern that shows very precise, complex allopatry by chance. 
 
Remember, you can´t always extrapolate.                             

Finally, you also mention that "dispersal theory" (I prefer mechanism) does not accept allopatry as evidence for vicariance. In truth allopatry is a pattern of distribution. 
 
MH: Agreed, by definition.
 
Explaining this pattern requires other facts and a hypothesis (mechanism) that fits the facts. 
 
MH: Yes.
 
Using a pattern to support a mechanism used to explain said pattern is circularity at its best.
 
MH: Yes, but no-one does that, not even dispersalists. Biogeographers look at many examples of allopatry and develop a concept of chance dispersal or vicariance. Then they test those concepts on other examples. Obviously, 'chance' can explain anything at all and I'm sure you'll be able to explain all your beetle distributions easily. But as more and more new examples of very precise, complex allopatry pile up in the molecular work, even in marine groups with pelagic larvae and now protozoans, people are starting to re-examine the old concepts. 

Best                           
_______________________________________________

Taxacom Mailing List
Taxacom at mailman.nhm.ku.edu
http://mailman.nhm.ku.edu/mailman/listinfo/taxacom

The Taxacom archive going back to 1992 may be searched with either of these methods:

(1) by visiting http://taxacom.markmail.org

(2) a Google search specified as:  site:mailman.nhm.ku.edu/pipermail/taxacom  your search terms here