On Thursday, I presented a species problem taken from a post over at my old blog. I presented data from experimental matings that were carried out among three insect populations, added a little bit of information about the appearance, behavior, and location of the populations. I asked you to tell me how many species these three populations represented, and promised that I'd give you the "official" answer today. I've decided, though, that it wouldn't be totally fair to answer the question just yet. You see, I withheld relevant data when I presented the original version of the question.
When I gave you the information on Thursday, I just gave you data for some of the populations that were involved. There are two more populations that I didn't tell you about at the time. We'll call them Population D and Population E. Individuals from these populations were included in the original crossing experiments, and we also have data about their reproductive behavior and appearance. I'm going to throw these populations into the mix.
What I'd like to know now is this: how many species exist here in total, and which populations fall into which species? Here's the more complete set of data:
Geographically, the populations all fall more or less along a single line, with individual populations separated by a minimum of 5 km. They're arranged in the following order:
E -> D -> A -> B -> C
In appearance, the populations can be split into two groups. The members of each group are identical in appearance to each other, and differ slightly from the other group:
Appearance 1: Populations A, B, D, and E
Appearance 2: Population C
In behavior, the populations can also be split into two groups. One group will only deposit eggs in a very specific type of plant debris. The other group will deposit eggs in many different types of plant debris.
Specific egg depositing: Populations C, D, and E
Generalist depositing: Populations A and B
Reproductive isolation. Here's where it gets really interesting, and not just because the most commonly recited definition of "species" depends on reproductive isolation:
If you cross an individual from Population C with one from either Population A or Population B, the fertility of the male hybrid offspring is greatly reduced, as is the number of viable female offspring. When Population C is crossed with either Population D or E, the hybrid males are still almost all sterile, but there's much less of a reduction in the viability of hybrid females. However, when the hybrids of those crosses reproduce, many of their offspring are infertile or non-viable.
When Population A is crossed with Population B, there's no significant difference in the hybrid fertility or survival than there is when they're bred within their own population. The same is true when Population D is crossed with Population E.
When Population A or B is crossed with Population D or E, there is a slight drop in male fertility, and no obvious drop in female viability. However, when these hybrids were crossed with each other or with an individual from either parental population, there was a sharp decline in the fertility of the male offspring.
How many species do you think there are now? If you're not sure, what do you think the most reasonable choices would be? On Friday, I'll post my own views of the situation, and provide some references to the peer-reviewed literature that will show how experts have viewed this.
- Log in to post comments
With this new information, I'm now suspicious whether this is actually a "ring" clade (or whatever the technical expression is), in which the different populations' ranges are geographically arranged in a circle -- either around the globe or around some utterly uninhabitable space such as a lake. If so, I'd guess that populations E and C are are close together geographically, and that C is a little separated from the others.
Alternative possibility: the ring clade is actually E through B, with C sharing a common ancestor with the ring clade but always being geographically separated from it from the time it diverged.
The populations are really arranged along a single line. E is the population that's at the NW end of the line; C is located at the SE end. There's absolutely no ring involved here.
Mayr had a way out: call it a "species cluster" or a "superspecies". It sounds very like the chromosomal races of the Israeli Naked Mole Rat (the other kind). And this is not uncommon.
I reckon you have three phylospecies :-)
I'm saying that A and B are one species and D and E are one species. That makes C a separate species, and means that you have three species. But I'm sure it could be argued that you have as many as five or as few as one species here, depending on how badly the lab reporting it want to find a new species :-P
- JS
Thank you for the response. In that case, I'd say that C is still a separate species. I don't know how easily insect species can change the specificity of their egg-laying, or how big a genetic jump that is . . . so it might well be that all the other populations constitute one species or superspecies with several subspecies. You don't say whether there's a geographic barrier of any sort between populations D and A, or if so, how temporary it might have been.
One thing no one has brought up yet: What is the sexual determination schema for the species? Sexual determination varies greatly among various insect taxa, which may account for the differential in viability by gender. For example, in some species, males hatch from unfertilized eggs (this is called haplodiploidy).
I'll take three species for $0.02
E-D as one, A-B as another and C as the last based on reproduction primarily...
Though I am strongly tempted to say one species.