[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Arguments Against Mixing Locations (was Re: N. rachovii)

On this very interesting thread, Doug Karpa-Wilson wrote: 

>>  On the balance it is far more likely that a species 
>>  is lost through too much inbreeding than through 
>>  too much mixing.

That's an excellent summary of my leanings on this subject, coupled with the "fact" that our original collection populations are typically 
  - very small (from a few pairs up to a few dozen pairs, perhaps) 
  - non-random samples (as is true for most predation situations, there's a systematic collection bias towards the slow, stupid and/or situationally unaware specimens)

Then what often happens when collectors return from exotic locations is that, to circumvent the "all the eggs in one basket" problem, they farm off a pair or two of the fish in question to Great_KIllie_Breeder_A_in_California, and a pair or two to Great_Killie_Breeder_B_in_Wisconsin, and a pair or two to Great_Killie_Breeder_C_in_Dusseldorf ... each of which most likely becomes   the F0 generation of a (relatively) "reproductively isolated" sub-population that, if things work out well (i.e., the great breeders are successful) then persist in the hobby for a long time.

If you are a Great_Breeder in the above example, what is the probability that the four fish you received represent a good sample ... i.e., statistically representative ... of the full-spectrum genetic diversity that was present in the species population present at the named collection location ? 
 
My feeling is that the probability is very low, near zero. And that because the original collection samples of so many of our species in the hobby are really quite small, "genetic drift in the hobby" is a huge issue that we really don't understand very well. 
 
Way back a quarter-century ago when I was an undergrad, I did a term paper on (+/-) "The impact of genetic drift on small, reproductively isolated populations of hunter-gatherer hominids," which was based on a few thousand computer simulations of those kinds of populations over 500-1000 generations. And it turns out that randomness in small pops is huge, even a single random mutation that results in a +50% survival advantage to the individual with it has a +99% probability of being snuffed out within 100 generations. (I called this the "wobbling your bicycle along the edge of the cliff" effect.) To my amateur way of thinking about population genetics, the population dynamics of the world of killies in the hobby is very similar. I have no doubt the science has progressed a great deal since then, and the tools for running these kinds of tools are undoubtedly vastly better and friendlier. 

What we need, I think, is a mathematically-based understanding of genetic diversity in killie species/locations, from which we could develop "Statistically-based guidelines for maintaining genetic diversity in captive populations of a species across multiple locations." That's an editorial "we", clearly the academics in the hobby would be the most likely people to be able to do that. Or it could be that "zoo-science" types already have that kind of knowledge well-researched and described in their professional literature, but it's not yet known by fish hobbyists. How many genes does a typical killifish have? What's a typical mix of alleles for a gene? How big does a sample have to be to be reasonably confident it's representative of the source population? How much redundancy is needed to be reasonably sure that F-50 generation is similar to the F-0 sample?  (Ignoring "selection for traits especially advantageous in the aquarium environment", a bias that is probably impossible to eliminate.) If you're collecting in the Cameroons, is it better to bring back 10 specimens of 20 species/locations, 20 specimens each of 10, or 50 specimens of just four ... which strategy is more likely to have those fish still in the hobby ten years later? All of these are questions that can be explored, and could perhaps influence how we manage our "fish resources." 

Questions to ponder:

How many kinds of fish have been "lost to the hobby" over the years ?

How much of that can be attributed to the fact that we're basically clueless about how to maintain genetic diversity ?

How much of that can be attributed to the fact that as a hobby we don't have any systematic plan or effort to keep a "new fish" in the hobby in perpetuity ?

Could we do better ? Should we ? Would it be worth the effort ? How could we do it? (IMO, a near real-time killie census would be a key resource, and that's conceivably do-able with today's technology ... but it'd still be work.)

Questions to ponder.


Doug Dame
Interlachen/Gainesville FL


---------------
See http://www.aka.org/AKA/subkillietalk.html to unsubscribe
Follow-Ups: