Chris Kennard
Well-known member
That is where things get hairy.
The percentages are per gamete pairing (or embryo/hatchling since it is comfortably conceptualized that way).
Speaking to the caramel piece alone and ignoring the albino piece -
While a homozygous X heterozygous pairing should yield a 50/50 shot of homozygous for caramel versus heterozygous for caramel, it is new gamble of chance with every embryo and every embryo's chances are independent of the others. You can calculate a string of chances, but it is not really predictive. Also, no matter how many chances of/in series are calculated when not proving out with a homozygous caramel, that calculation does not ever reach a chance of zero. By test-breeding, we can prove hets out but not technically disprove hets because there is always that chance of poor luck taking place. Since the last time I got into this discussion, I came across an instance of someone's animals taking over forty ( 40!!! ) offspring until the het parent proved out. That is very, very far from the norm. If you get enough offspring, the stats start to link up with the results, but sample size is key. When I had something (of a different species, but the math works the same way) that was supposed to be a het for me produce a season's worth of hets only and no morphs (at over forty-five hatchlings), I figured she was not going to prove out. I used to draw the line personally (for making decisions) at 20 results from homozygous to heterozygous pairings. I am giving that animal this season to be "functionally disproven" (not mathematically disproven) based on the freakish streaks. She can never be mathematically disproven because the chance of some absurdly long string of non-morph results still is and always will be greater than zero by calculation (because you are multiplying a percentage by a percentage with both values greater than zero; 0.0000000000001 x 0.0000000000001 is still not 0 by the math).
If the call is made by reasonable assumption at some point, then we could label it as disproven despite it not necessarily being accurate. Everyone has a different threshold, though. I have seen many cases where keepers flipped out on sellers because they hatched out four, six, or ten babies and were dissatisfied because they had no visuals and they believed it should have been half visual and half het. Life does not always work like that, though. I have had 10-long streaks in both directions. That is why I made my calls traditionally at 20. With the existence of at least one 40-long streak, it even gives me pause because I thought my 20-long threshold was adequate for making a decision. Pain in the cloaca, right there. I think the 10-streak is a little less than 0.0098% probable, but they have happened multiple times for me with hets that proved out despite that chance being perceived as very small. A 40-streak is something like slightly less than 9.095x10^-11% probable. Insane, yeah? Yet it has happened.
I often get 4-6 if they are small-bodied females of this species. It can be three times that for me per clutch if large-bodied, but that is not what we would be seeing here in my experience. There should be several clutches in the season. Three to six clutches per season would be relatively common (depending on the individual animal) for me. Not all of these eggs would necessarily be viable and that would draw out the evaluation.
I would not bet the farm on the first clutch. I might even be wary with only a single season (depending on how many hatchlings I would actually get). I have had other animals not prove out with smallish clutches one season that then proved out the following season.
Thanks Nick! Lol. Your last paragraph was the only one I needed. You are truly an asset and an ace here. Your knowledge is an awesome tool. I was just wondering if the offspring from the first breeding would be proof of whether or not the turtles in question are hets, or not. I'm sure that the investigator, Chris, who hired investigators to look into his case, will present your wisdom to the judge.
