Causes of arthropod endemism on Cyprus

Aim: To investigate the relative importances of isolation and underlying geology on the evolution of endemic arthropods on Cyprus.

Methods: This project blends several disciplines, ranging from geology and pedology to evolution, biogeography, and physiology. See the “Project rationale and scientific context” section, below, for background information.

Soil and plant-associated arthropod communities are being collected from the Troodos Mountains and contrasted with those on the Mesaoria Plain. Of particular interest are the communities associated with endemic plants found in both locations. The communities are being quantified using functional and phylogenetic diversity.

Collected arthropods are being dissected and chemically analysed for concentrations of soil-derived heavy metals, in order to test for the possible effect of geology. If geology plays a role, then higher heavy metal concentrations are expected in all arthropods of the Troodos Mountains. If the Troodos Mountains also have higher levels of endemics (as measured by phylogenetic diversity), then a link to underlying geology is reasonable to hypothesise.

Regardless of the role of geology, most endemics on Cyprus evolved due to isolation after migrating here and getting stuck. To assess the time of migration and isolation, genetic analyses comparing Cypriot species with closely-related species on the mainland (North Africa, Middle East, Turkey) are being planned.

In addition, I am interested in the discrepancy between the rate of evolution at the molecular level and at the phenotypic level, a question relevant to endemic formation. To assess this, novel methods to do morphological phylogenetics on very closely-related species are being developed to allow direct comparisons between molecular evolution and morphological evolution in Cypriot endemic species and their sister species on the mainland.

Project rationale and scientific context:

The map above shows Cyprus’s position in the Eastern Mediterranean. The nearest continental landmass, the southern coast of Turkey, is 70 km away, followed by the Middle East 100 km away. Then comes North Africa at 350+ km away, and a distant Greece, over 500 km away. These distances are reflected in the fauna of Cyprus, which has a very distinct Middle Eastern feel to it.

All islands are viewed as excellent places for evolutionary research, and Cyprus is no exception. Its relative isolation is more than enough to make endemics out of any insects that don’t routinely migrate over long distances.

If one looks at the map where endemic plants are found, this relationship between isolation and endemism is easily observable. In green are areas where endemic plants are found, in white are where they’re absent. The pattern reflects very closely the level of agricultural and other human development: the white areas are all developed or agrarian areas, where human-mediated invasives have taken over the landscape. Protected and mountainous areas, on the other hand, are rich in endemic flora.

Such a map doesn’t exist for arthropods due to the lack of research, but a similar pattern can be expected, with endemic richness being much, much higher in the “wild” areas of Cyprus.

Map Source: Tsintides TC. 1998The Endemic Plants of Cyprus.

For me, one of the most important questions to ask is how insects first came to the island; this is especially important to consider for insects that don’t fly very well or at all. There are generally four avenues for dispersal to an island.

The first, and most opportunistic/random, is rafting. As the name says, this is nothing more than getting stuck on afloating raft (of wood) and getting carried by the currents to the island. It may seem like a stretch, but the oceanography makes this possible, and we know that insects can survive for long periods of time on a block of wood. So it’s by all means logistically and theoretically possible, and has probably happened.

Another method is by using land bridges. This is where knowing some geololgical history comes in handy. Cyprus’s case is rather unique in that it formed completely in the ocean, with the central mountain range (Troodos) being nothing more than elevated oceanic crust (as we will see later) and having collided with the northern mountain range (Kyrenia), resulting in uplift of the central and coastal areas (and uplift that is still ongoing, albeit in a different tectonic context). So at no time during its formation was Cyprus connected to the mainland. However, 6 million years ago, the entire Mediterranean did dry up during the Messinian Salinity Crisis. The area between Cyprus, the Middle East, and Turkey was definitely crossable by any insect and spider. When the Atlantic flooded back, the organisms were left stranded on Troodos or Kyrenia.

A third method applicable to Cyprus is jump-dispersal, which is when multiple small islands form stepping stones to the large island. This, again, requires some geological knowledge. With the climate changes of the Pleistocene and Ice Ages, the Mediterranean’s sea level experienced some wild fluctuations. By this time, Cyprus had pretty much emerged from the sea and looked more or less like it does today. But with the sea levels fluctuating, the land between Cyprus, Turkey, and the Middle East became exactly as in the picture above, and some scenarios even suggest that it was completely dry, resembling a sort of salt marsh. In both cases, the way is easily traversable by all but the most immobile of arthropods, and when the sea levels became normal again, they were left all over the island.

The last, and most easily-detectable, way for insects getting onto the island is through humans, either purposely introduced as pest control (as is the case with some coccinellids, for example) or accidentally. And this has a very rich history in Cyprus’s case, given that Cyprus has been conquered by just about every major seafaring civiliation from Europe, North Africa and the Middle East at some point in its history. From living on the ships, to the products in the ships (timber or food), to the clothes of the sailors, this is by far the easiest way for any small insects to make their way around the world. Luckily, it can also be very easily accounted for using genetic tests.

The point here is that merely noting the current biodiversity of the island isn’t enough. In order to fully understand how the insect communities of the island developed (and thus how endemism came to be), we need to have this historical aspect in mind.

So, now the insects are on the island. How do the new species form? I already wrote a post on speciation, useful if you need some summary background on the processes involved. On Cyprus, there are three ways in which speciation can happen: by cladogenesis, by anagenesis, or by anacladogenesis.

Cladogenesis is a case of classic sympatric speciation. Population B arrives on Cyprus from the mainland and encounters different habitats and different ecologies than what they are adapted to. One part of the population colonises one habitat, the other colonises a different habitat (either by chance of where they land, or “on purpose”). Over generations, they will adapt to the new plants and abiotic conditions, and they will be new species, distinct both from each other and from the parent population.

Anagenesis can be summed up as “speciation without branching”, occurring as change accumulates in a single lineage. For our case in Cyprus, this can happen if the migrating population lands somewhere where the identical conditions and plants are found, which is likely considering the similarities in climate and flora of the closest landmasses. Additionally, there needs to be some amount of secondary remixing by travelling back and forth or occasional waves of migration – if they get geographically isolated with their own little gene pool, they are treated as a branched-off lineage arising by cladogenesis. But even with some remixing, there will be some isolation, and by genetic drift, mutations will slowly but surely accumulate within the Cypriot population, initially resulting in some reproductive isolation (as indicated by the i in the diagram). Eventually, this will result in complete speciation.

Anacladogenesis, as the name suggests, is a mixture of the two, with part of the population undergoing anagenesis by remaining in a place with similar ecologies and abiotic conditions, while another segment of the population colonises new ground and buds off to form a new species.

There is no way of predicting, in the absence of data, which of these has played a more important role in Cyprus. I would hypothesise the obvious: insects with good dispersal ability undergo anagenesis, while more immobile insects undergo cladogenesis. This will all be tested once I have baseline information on the taxonomy of the endemic insects.

What we’ve talked about so far has been general and more or less applicable to any landscape affected by invasion or to any island, but Cyprus still provides an extra experimental and data-gathering arena. What will now follow is a look at the scientific basis for my research.

First off, a purely theoretical look at the background of the main thesis, that increased mutation rates will lead to speciation. The flowchart above shows, generally, how macroevolution proceeds.

The foundational level in macroevolution is the genome – the entire collection of genes and sequences, both those that will be expressed at some point and those that will not. From this collection, a few select parts are activated during development of an organism. These tend to be evolutionarily conserved (indicated by the loop), as screw-ups during development are generally very deleterious.

Development leads to the building of the phenotype, of the individual. The phenotype then has a very tight interplay with ecology – which I lump here with natural history, so including things like behaviour and life cycles, not the least through functional morphology. Ecology in turn can modify the phenotype through environmental effects – more food leads to larger organisms, a preference for colder habitats leads to different phenologies, etc.

As I stressed in my natural selection lecture, the individual is where natural selection acts, because it’s the individual that survives and reproduces, not its genome. But, as the flowchart shows, the individual is itself dependent on development, which is itself derived from the genome. So, in effect, tracing the evolution of the phenotype can be done by looking at the evolution of its development. The action of natural selection on the phenotype and on development leads to an evolutionary response on the genome, making some parts of it invalid, or emphasising the roles of other sections.

This is how macroevolution works, basically. It’s nothing more than a feedback between development, the phenotype, and the genome, with the former two exerting their effect on the latter.

What my project aims to do is subvert this scheme on its head, by asking a hypothetical question: What if the diversity in the genome increases? Will that make the filter of development more porous and lead to higher phenotypical diversity, in turn leading to more ecological diversity and to speciation?

The answer known already is that yes, higher genomic diversity can lead to speciation, as evidenced by studies with gene or genome duplications. Where my project differs is that I am taking the question at a truly fine level, looking at individual mutations, and all in a well-constrained natural arena. The reason I can do this is only because of Cyprus and its geology.

The above picture summarises the geology of Cyprus. It’s split up into four units: the Kyrenia Mountains in the north, the Troodos Ophiolite in the center and west, with the sedimentary succession surrounding it, and the Mammonia Terrane to the west and southwest.

We’re interested in the Troodos Ophiolite. As you can see from the endemic plant distribution map, it’s clear that the endemic plants are mostly found in all the geological units besides the sedimentary one, but the Troodos Ophiolite is particularly well-inhabited – the patch in Mammonia, around the lake in Lemesos, is mostly due to the presence of unique salt marshes there, not due to geology per se.

The diagram on the left shows the structure of the Troodos Ophiolite. An ophiolite is nothing more than a piece of oceanic crust, and this discovery was done in Cyprus in the 1950s, thanks to the impeccable preservation of the Troodos sequence – other ophiolites tend to have bits and pieces of this sequence, in Troodos it’s complete. For orientation, the bottom of the diagram is actually the top of Mt. Olympus, and the top of the diagram is the lowland part of the mountains. So, in effect, when you drive up the mountain, you’re actually going back in time as far as geology is concerned. It’s quite surreal.

A detailed look at this falls outside the scope of the post. Suffice it to say that the bottom of the sequence, from the harzburgites to the plagiogranites, represent the rocks of the oceanic crust, from the deepest ones (subject to high temperature and pressure) to the shallow ones. The sheeted dykes and basalts are lava – the uplifting of the oceanic crust caused cracks to appear, and magma seeped through (sheeted dykes), spreading on the ocean floor (massive basalt) and reacting with the much colder ocean water (fractured basalt). After a certain period of uplift, volcanic action subsided and sediments accumulated on top of the whole thing. For some reason, when it was uplifted, the ophiolitic sequence of Troodos got flipped upside down, resulting in the oldest rocks (harzburgites and lherzolites) being found at the top and the sediments at the bottom.

Diagram source: Edwards S, Hudson-Edwards K, Cann J, Malpas J & Xenophontos C. 2010. Classic Geology in Europe 7: Cyprus.

Geology serves as a basement, and gets eroded and degraded by plants to form soil. Thus, the characteristics of the soil are intrinsically linked to the mineral content of the underlying geology. In this case, we have water reacting with olivine-rich rocks of the oceanic crust and hydrating them – this leads to the formation of serpentine minerals.

The soil derived from this group of rocks and minerals is correspondignly called a serpentine soil, and its main, relevant characteristics are shown on the right. First is very low concentrations of elements typically critical for plant growth, including potassium, nitrogen, phosphorus, and calcium. This means that only specific plants can actually grow here.

Making matters worse for the plants, and gloriously exciting for me, is the second characteristic: high concentrations of heavy metals. Iron and magnesium levels are especially increased, but chrome, nickel, and cobalt levels are also notable. Heavy metals are known to be hazardous as mutagens, and so the pool of plants that can grow on heavy-metal enriched soils increases.

And here Is where the geology ties back to my previous scheme of macroevolution and the question I’m asking. As mutagens, heavy metals will increase the diversity of the genome, and so will provide a natural experimental set-up to test my hypothesis.

There is no data to go on yet for Cyprus – that’s why I’m trying to get the project funded. But there is another region in the world where ecology on serpentine soils is being studied: California. California has an area of ~424000 km². Serpentine underlies 6000 km² of that, so only 1.5%.

Yet, as the graph above shows, the percentage of endemic plant species in this tiny patch is equal to or greater than the percentage of plant species in the rest of California. This is mostly due to the inhibitory nature of the soil, with the high heavy metal concentrations and low critical element concentrations.


But it is also valid to ask whether higher microevolution rates, caused by the mutagenic heavy metals, can also play a role. My not being a botanist (although with an interest in plant physiology) prevents me from investigating this with plants in Cyprus. However, one of the first lessons in ecology and environmental sciences is about bioaccumulation, that toxicity accumulates the higher up a food web one goes. And the next level is, of course, the myriad insect herbivores, followed by the carnivorous insects and spiders, and all supplemented by the omnivores and detritivores.

So, if any effect on microevolution rates are there, they can just as easily be observed in the associated animals, in which case I can inject my own disciplinary bias and look at the topic as a macroevolutionary problem.

If the effect is not observed, then an equally valid question to ask is “why?”. Why don’t the heavy metals affect the animals and their mutation rates? Do they simply not go up the food web (and if so, why?), or do they do so but physiological mechanisms prevent them from having an effect (if so, what are these mechanisms?).