[Taxacom] Fwd: Woodpeckers, primates, as well as the Wallace Line gauntlet

[Jason Mate]

Some thoughts on your comments.


> Hi Jason,
>  
> Every individual of every plant and animal (except in colonial organisms) has dispersed to where it is now. The problem lies in integrating that process - normal physical movement - with other processes in phylogeny and geography, especially range expansion and vicariance. The physical movement of individuals may have little or nothing to do with the distribution pattern of their clade. Jim pointed out the paradox in ferns and marine groups with pelagic larvae, and many other authors have discussed the problem. Individual albatrosses fly around the world at the drop of a hat, and yet the clades have very precise, allopatric, locally endemic, breeding ranges. Why? How? 
>  
> You write: 'I am not sure even what you mean by dispersal anymore since you brought up ecological dispersal. Is there a distance threshold that you have in mind?'. Long distance dispersal could be better termed 'chance dispersal' because the distance is irelevant. People use the concept of l.d.d. to explain differentiation across a river. The term 'dispersal' has (at least) two distinct meanings, one is normal movement, one is a mode of speciation. They may or may not be related, but they are not the same thing. 
>  Likewise, 'chance' can mean very different things. Chance in the ancient sense of Fortune, i.e. 'factors we don't understand', has never been regarded as a 'scientific' explanation (outside of biogeography), but chance in the sense of calculated probabilities is a fundamental concept in science. 
>  > You say: 'The fact that the movement of living organisms has been observed should suffice to recognize that the mechanism is probable'. Obviously movement (normal ecological dispersal) occurs - no-one is denying that. What you call 'the mechanism' , i.e. chance dispersal, is something else. 
Specific examples are always cherry-picked (I´m guilty as well), yet I am not sure why you bring albatrosses to the table. Vagrant albatrosses (and many other birds for example) are commonly seen, so obviously every year a few albatrosses get lost. But why aren´t they everywhere you ask. Well, dispersals, like mutations, are costly and lethal most of the time. So whereas a small hop between islands in an archipelago (the equivalent of a neutral mutation) is not very costly (and hence survivable most of the time) a major movement would most likely lead to death (ecological mismatch (most extreme case no island found at all), no other members found, etc). So most areas are sinks and even if a honeypot is found you still have to contend with founder effects, bottlenecks.... Maybe that´s why aptery is so common on oceanic islands? Yet amongst all these possible scenarios, where death is by far the most probable, there is a small number (maybe even only one) scenario with a low (how low?) probability of success. We don´t have the hard numbers of course. Not only is it too difficult to gather the data but the most vital piece of information, how many colonists left in the first place, is unobtainable. Is that why you´d rather ignore it?  It could be interesting to look at the distribution of trip length vs frequency for selected (monophyletic maybe?) groups of species in and around archipelagos. This would be one way of objectively assessing what consitutes long dispersal and maybe infer on the probability of successful dispersal. Or more Krakataus for that matter ( by 1933 290 plants, 720 insects, 4 reptiles, 41 bird and 4 mammal species had colonized it. It is obvious that insects, plants and birds are better suited to dispersal).
Finally, IMO missing information or 'factors we don't understand' is very much part of science. We know the probability that a photon will be reflected from a sheet of glass but not why or if a particular photon will reflect. And this is after 100 years of gathering data on the simplest of particles. Hence the ultimate causes of nature can be hidden from us yet a strong, predictive scientific field is possible. On the other hand complexity can generate randomness. Neural networks generating behaviours that cannot be predicted from the individual parts is an apt example, as are ecological networks. So regardless of the minuscule probability of succesful dispersal, considering the millions of species, and the incalculable number of individuals, dispersal is a mathematical certainty, even if philosophically unpleasant.
>  
> You write: 'Dispersal is one probable mechanism to account for the presence of a taxon in a place' - and vicariance is another. Dispersal in the sense of simple movement does not explain why a taxon is different from its relatives. The question: 'Why is this organism here?' has two parts: why is it *here*, and why is it *this organism* and not another. 
>You suggest: 'What happens afterwards [after dispersal] (evolution, radiation, extinction) is besides the point'. But evolution is not besides the point - it is the main focus of investigation. If there was no endemism and everything was everywhere the subject of biogeography would not exist. 
I´ll be  honest, I have no idea what you wrote here. I can understand that the niche of a species is, in a way, an extension of the phenotype or the environmental translation of it. But you are deliberately ignoring contingency. Are you saying that biogeography should be teleological? What about history? You can ask why til you go green and not get an answer with some deep philosophical meaning.  One can see general patterns in history but you can´t pick each individual story and see in it the finality of history. It has been influenced by it of course, in the same way that biotas have been shaped by vicariance, but ignoring the noise (dispersal) is looking at half the story, albeit simpler.
> You say: 'how did the taxa arrive to those recently discovered flat-topped mounts? In the end to support your particular hypothesis you are forced to resort to the same scenario-building excercise that you criticized from Kenneth'. The difference is that Ken's ideas were ad hoc, mine are based on geology. Why has no biologist even mentioned the Musicians guyots? Because they are more concerned with making up ad hoc hypotheses than with  trawling the primary literature in geology. Why did the founder of island biogeography (the most cited book in biogeography!) completely ignore not only the seamounts between Hawaii and the mainland as a possible center of origin, but also all the atolls (former high islands)? 
>  
> To answer your question: the taxa on the Musicians islands (before they sank) arrived there by the same processes that taxa arrived in Hawaii - by normal ecological movement from other nearby islands that no longer exist.   

Ad hoc doesn´t mean making it up, it means weaving a story based on something for a purpose. The mounts are there but it is still a story. As for why ignore the atolls, I would say because they didn´t fulfill the necessary ecological requirements of the species in question. Just because you have sand in the middle of the ocean it doesn´t mean that it is suitable. Also I like how movements between islands (even remote ones) can be normal. And the Musician islands are aptly named, like musical chairs. ;)
>  
> I asked for examples of chance dispersal, 'horrendous dispersers that have indeed made it', and you cited Aphodius holderi. I looked this up and apparently it was the most abundant dung beetle in Britain and other parts of Eurasia during the colder parts of the last glaciation, but now restricted to a very small area in the high plateau of Tibet. Where is the chance dispersal? Its range expands and contracts with climate (not chance). 

Chance is where it ends up. Surely the mountain ranges along the way could have been equally suitable? After all its companion in frozen Ol´Blightey found the mountains in Spain suitable (which begs the question, why not the Pyrenees, the Alps or the Caucasus or any of the ranges all the way to the Himalayas?). I just see chance all along. Any comments on the others?
> In another example, you say that in New Caledonia 'there are 3 species which appear to have arrived on their own "recently" (collected by Montreuzier in the 1850´s): Parastasia percheroni (Rutelinae), Dipelicus montrouzieri (Dynastinae) and Cyphopisthes sp (Cerathocanthidae)'.  
> I've just found the information - it's in Paulian's 1991 book on Scarabaeoidea of New Caledonia (Google books). He thought that New Caledonian endemics have usually followed the Solomons-Vanuatu route. So he argued that Dipelicus must be a relatively recent introduction because it was related to things in Australia and New Guinea, while the Solomons/Vanuatu form is distinct. The logic here is very obscure.  Cyphopisthes is 'Indo-Malayan'. Paulian seemed to think that this distribution also means dispersal - why? 

Dipelicus montrouzieiri is found In PNG and Australia , Solomon Islands, NC and New Hebrides, but three ssp occur: the montrouzieiri montrouzieiri in PNG, Australia and NC, Dipelicus m. oryctoides in New Hebrides and Dipelicus m. salomonensis. D .montrouzieiri is part of a group of 5 closely related species (genitalia very similar, males clearly different) which occur in the Bismarck (alveolatus, triangularis, integriceps), Moluccas (alveolatus), PNG (alveolatus, triangularis, integriceps, bowilli), Australia (integriceps, bowilli, optatus, duples). Another 12 species also occur from India to Java. So the logic here is that D. montrouzieiri montrouzieiri colonized NC too recently to develop racial or ssp differences.  
Regarding Cyphopisthes, only one specimen available and I haven´t examined it myself so I am going on Paulian here.
> Parastasia occurs in India, Indonesia, New Guinea, and (with P. perchoneri) New Caledonia/Vanuatu. Why did Paulian think this is dispersal? (or recent?). P. perchoneri ('bien distincte') is related by Paulian to forms of Woodlark Island and this area of endemism (New Caledonia plus the tip of the 'tail' of New Guinea, 1500 km away), is one of the most intriguing patterns anywhere (see my paper in J. Biogeogr. 37: 1179. 2010). The tectonics are incredibly dynamic, complex, and controversial, but at least the geologists are hard at work on it, while the biogeography is hardly ever mentioned, and when it is, it's just 'chance'.  Things may be changing though. Metzger et al. (2010, Mol. Phylogen. Evol.) attributed disjunction between Woodlark and the Louisiades in elapid snakes to the opening of the Woodlark Rift.      

Parastasia´s range covers China, India, Malaya, Indonesia, North America and a couple of species make it into PNG and P. percheroni PNG, NC, Australia (QLD) and Vanuatu. In addition the rest of Parastasina is Oriental, not Oceanian (Australia has its own ruteline groups which it shares with South America but not Parastasina) . I have no doubt that NC´s geology is complex, yet I look at Parastasia and I see a lucky vagrant. 
You make no comment on the Malagasy Aphodius, the Hawaiian Hydrophilidae or the butterfly examples, to which i would also like to add the Australian Onthophagus as a case of dispersal. But to make it up, you may want to look at Podotenus (Aphodiinae) and Sphaenognathus queenslandicus (Lucanidae) as some impressive South American vicariant genera.
Best
Jason