26 May 2006

How many species?

I've got a little question to keep people busy over the long weekend.

There are three populations of an organism. The populations are physically separated from each other as a result of geographical factors. Geographically, they are arranged in a more or less linear fashion. The geographic details are as follows:

Population A is the northwestern population. Population B lies to the southeast, and is eparated from population A by a minimum of ~14km. Population C is southeast of population B, and separated by about 50km.

Populations A & B are identical to each other in appearance and in a key reproductive characteristic. Population C differs very slightly in appearance, but is substantially different in the reproductive characteristic.

The organisms (flying insects) were captured and bred in the laboratory. Experimental crosses were made for the different combinations of these three populations, with the following results:

Male from A x Female from A:
91% of male progeny fertile

Male from B x Female from B:
99% of male progeny fertile

Male from C x Female from C:
92% of male progeny fertile

Male from A x Female from B:
98% of male progeny fertile

Male from B x Female from A:
93% of male progeny fertile

Male from B x Female from C:
0% of male progeny fertile

Male from C x Female from B:
0% of male progeny fertile

Male from A x Female from C:
0% of male progeny fertile

Male from C x Female from A:
69% of male progeny fertile

Female offspring had somewhat better fertility than the males, which is not unexpected in this group, for reasons I'll discuss in another post. Female offspring were fertile in the cases where the males were sterile, but the number of offspring surviving to maturity was greatly reduced in those crosses overall (around 10% of what was seen in the control crosses) and the fertility of the female hybrids was reduced compared to the control crosses.

My question for you is this: How many different species should these three populations be grouped in? Provide an explanation for your answer. Oh, and if you hadn't guessed, this isn't a hypothetical case. I've removed the names so that you can't see what the "right" answer is.

I'll talk about the currently accepted scientific grouping sometime on Tuesday.

13 comments:

Anonymous said...

well, i must say that additional evidence is warranted before placing species status in any case....laboratory breeding is not sufficient evidence. A+B are quite similar in reproductive capacity, and the interpopulation breeding is even higher than A's intrapopulation breeding success (though no error is given), so is indicative of a potentially interbreeding unit. C appears relatively distinct, but again this is difficult to say without addiditional evidence. laboratory crosses are notoriouosly difficult to interperet, in insects particularly. i have had issues keeping similar looking species from crawling into each others' cages on occaison. molecular data from multiple markers (eg STRs) would facilitate an understanding of potential inbreeding/isolation in the field, which in my opinion is a better measure of species status in morphologically cryptic groups. my guess, unless this is a trick question which I am missing is that they should be treated as a single species until other evidence supports a split. though i am a bit of a holistic taxonomist, and am a bit conservative in species delineation unless multiple lines of evidence support it.

Anonymous said...

It's one species. This is a great demonstration of Haldane's Rule, but genes can still flow pretty easily through the population. If we taxonomists had to name every geographically isolated population that showed incomplete genetic barriers we'd never finish naming things. I'd not get too hung up on the species question and just do the science, being clear on how the term "species" is used operationally.

John S. Wilkins said...

They're still just fruitflies ;-)

Anonymous said...

As a non-zoologist, I'd say that population C is sufficiently close to a separate population, and separate species.

I expect that lab breeding is much more likely to occur than breeding in nature, so the results are best-case data. And if so, the lack of successful C crosses with A and B indicate that there won't be much reproductive success in nature, if any.

Given all that, I'd say that the C population has about as much chance of successful breeding with the A-B groups as a chihuahua with a great dane in natural conditions.

Anonymous said...

There are no species.

There are only individuals.

Om.

Anonymous said...

42. Because as a Computer Engineer I don't know enough of science to give an informed answer, but I do know the answer to the Ultimate Question about Life, the Universe and Everything.

Anonymous said...

Why do you care? Why does it matter? I have written several comments on these Science Blogs asking about "species". I point out things like ring species; is species a transitive relationship; if reproduction with fertile offspring in a natural setting is a criterion, then how can chihuhahuas and Great Danes be considered the same species - and I get no answers. So why are you asking that question? How is "species" defined. Does it really have a definition? Does it matter? Isn't "species" just a modern reaction to the biblical term "kinds"

Karl Haro von Mogel said...

An interesting question. I'm tempted to go all nominalist on this one and say that "species" don't really exist, they are just human labels that attempt to describe a reality that doesn't conform to it.

However, that there is no fine line between colors of the rainbow is no reason why we cannot, for the purposes of utility, show the distinctions between red and green and why you should get a ticket for driving through the light of one color rather than the other. So let's give it a go.

A and B seem to be fairly cross-fertile, and it seems that males born from female A parents crossed with A and B males are less fertile than males born from B females crossed with A and B males. Given that the two populations are close together, and morphologically very similar, so I would consider them to be the same species, possibly different sub-species within the same species. I don't understand the criteria for distinguishing sub-species from varieties/breeds/ecotypes apart from their mere location, so I'll go with their status as sub-species. The one characteristic that seems to distinguish A from B is the fertility of males that came from A females, and the hybrid-ability of A with C.

Population C, however, is distinct morphologically and biologically. The question remains as to whether or not any of these populations do cross in the wild, it seems doubtful that C would cross with A or B if any cross at all. Although C can cross with A and B, the marked reduction in fertility and their distance from the other populations suggest that their gene pool is virtually isolated. Therefore, I would consider C to be a different species, closely related enough that it could still hybridize with some fertility.
A - subspecies of species 1
B - subspecies of species 1
C - species 2

I would consider this arrangement to be the most useful for study. The other option would be for A and B to be two populations of the same subspecies, and C would be the other subspecies, all under the same species. But the distance, fertility, and morphology suggest to me that they are more distinct than this arrangement describes.

I can't wait to hear the answer!

JS said...

With the caveat that I'm not a biologist, I'd go for two species... I seem to recall other examples of different species that can produce fertile offspring 'one-way' under artificial conditions, and little more than 30 % fertile offspring (when you combine the two ways to mate them), coupled with a high mortality rate isn't precisely my definition of 'interbreeding'

On the other hand, it would have been nice to have second-generation data (oh, and errorbars/sample size wouldn't hurt either).

Do, for instance, the A/C hybrids interbreed only with themselves, with only one of their 'parent' populations, or with both?

I don't know if the answer to that question would be helpful, but that would be the direction I'd look - in no small part because we're just talking about a re-iteration of an experiment that's pretty much already set up and running.

Anonymous said...

Since this thread is proceding very slowly - waiting for Monday's answer - I am not hijacking anything to try to continue a discussion of "species". NO, I am not asking from an ID denier point-of-view. I am an Evolutionist. In fact my background is in math where, before you can decide that something is or is not "A", "A" must be well-defined.
Since the definition of "species" seems not to be well-defined, why do you spend so much time trying to make these distinctions?
I like the analogy to a rainbow, but it doesn't apply here. Colors, nowadays, can be well defined - by frequency. And, interestingly, different cultures define color names to apply to different frequencies: e.g. "RED" is different in the U.S. and in Japan.
But there is no clean definition of "species". Ring species are a prime example. Adjacent variations interbreed sucessfully and therefore are the same species (ss). So population A ss B, B ss C, ... F ss G. But A ^ss G. This seems not to bother geneticists/evolutionary biologists. So I guess my question is WHY? or WHY NOT? Why does this not bother them?
Is, perhaps, there some activity currently towards recreating the definition based on DNA? Homo.Sap and Pan.Trog are different species because there is a 4% difference in their DNA. What % difference is necessary?

TheBrummell said...

Hello Karl,

To answer your question "why?" as in "why don't evolutionary biologists care?":

We do care. The definition of a species is a very slippery thing. Most use some variant on Mayr's Biological Species Concept, which (essentially) states that two populations count as separate species if they are reproductively isolated. In this example, I'd say that C is reproductively isolated from (A+B), and hence these are 2 species.

"Reproductively Isolated" is obviously a little vague. If some interbreeding is possible, but with reduced viability and/or fertility, where do you draw the line? Any choice is probably going to be arbitrary.

Also, there's a big differenc between not able to breed in nature because of isolation-by-distance or other "geographic factors", and not able to breed when brought together in the lab. One can easily imagine a wide range of hypothetical scenarios that would prevent matings between individuals in geographically separated populations, or change the relative probabilities of various mating combinations (eg, east can move to west, but not vice-versa, males A can mate with females B but not vice-versa, etc).

In any case, this is an active area of research. The best summary of the situation I've seen was a presentation by Dr. Massimo Pigliucci of SUNY-Stony Brook at the SSE meeting 2003. The basic idea was that a species concept should be based on the biology of the organism(s) under study, and features of various species concepts can be incorporated into a unique species concept suitable for the question one is trying to answer. This is still a little vague and unsettling, but research continues apace.

Also, please refrain from using the term "evolutionist". It's a made-up term from the creationists, who refuse to acknowledge that this is BIOLOGY, and people who study evolution are BIOLOGISTS.

Anonymous said...

Nice post.

No matter what species concept people support the one they use is the morphological one

Therefore two speices! Even the crossing experiments support that :P

Anonymous said...

I apologize for "Evolutionist". That is as bad as "Darwinist" eh? Except that I'm using it to describe me - and I'm not a biologist, it means only that I believe in Evolution as opposed to ID(Creation). What should I call myself?
Erik: What does "morphological" definition mean?
And back to "species: I've seen the "reproductively isolated" definition. It doesn't make sense for two reasons: If you go to a lake and scoop up (some creatures) enough to establish a breeding colony, put them in a tank in your lab and let them grow and reproduce, they are reproductively isolated fron the original population in the lake. Does that automatically make them a separate species? To me that doesn't make sense. And the opposite case: take two members of the same species where the physical variation is so great that the smallest versions can not physically mate with the largest versions. Why are they still listed as the same species? The obvious example is dogs. Under natural conditions could a chihuahua mate with a St.Bernard? It seems to me that the only hope for the definition is by genome.