Meeroo Genetics
Part 1. The basics

Meeroo Genetics are based upon Mendelian Genetics. Before we can dive into the differences, let's start with the basics. Mendel proposed two laws of genetics. For this, I'll copy from Wikipedia:

---

Law of Segregation

The Law of Segregation states that every individual possesses a pair of genes for any particular trait and that each parent passes a randomly selected copy of only one of these to its offspring. The offspring then receives its own pair of genes for that trait. Whichever of the two genes in the offspring is dominant determines how the offspring expresses that trait (e.g. the color of a plant, the color of an animal's fur, the color of a person's eyes).

Law of Independent Assortment

The Law of Independent Assortment, also known as "Inheritance Law" states that separate genes for separate traits are passed independently of one another from parents to offspring. That is, the selection of a particular gene in the gene pair for one trait to be passed to the offspring has nothing to do with the selection of the gene for any other trait. While Mendel's experiments with mixing one trait always resulted in a 3:1 ratio between dominant and recessive phenotypes, his experiments with mixing two traits (dihybrid cross) showed 9:3:3:1 ratios. But the 9:3:3:1 table shows that each of the two genes are independently inherited with a 3:1 phenotypic ratio. Mendel concluded that different traits are inherited independently of each other, so that there is no relation, for example, between a cat's color and tail length. This is actually only true for genes that are not linked to each other.

---

Armed with these definitions, let's examine a few examples. Remember, these examples will be using only the Laws. To apply the Laws, however, we need to know something about 'dominance' so we'll make one up. The intention here is to see how it would be "with any other breedable".

We'll start with a mating pair of Meeroos:
Aggressive(shy)
Aggressive(lazy)

First, we'll use simple linear dominance, like most other breedables. We'll state, simply for this example, that the order, from most- to least-dominant is: Aggressive, Shy, Lazy. Applying the Laws and this made-up dominance order, we get the following possible outcomes from our mating pair:

Aggressive(aggressive)
Aggressive(lazy)
Aggressive(shy)
Shy(lazy)

Since we started with a pair of "looks the same" Meeroos (both Aggressive) we see the predicted 3:1 ratio. With all other Second Life breedables, this is exactly the results we observe.

Meeroos, however, rarely use simple linear dominance. For most traits, dominance, that is which trait is expressed and becomes visible, is determined by the patent's gender. So, for the next example we'll use the rule that the trait passed from the Mother is expressed. We'll use the same pairs as before, with the (shy) partner being male and (lazy) being female. We get the following possible offspring:

Aggressive(aggressive)
Aggressive(shy)
Lazy(aggressive)
Lazy(shy)

Note that using gender dominance instead of linear dominance, we see a 2:2 pattern. We'll see this pattern a lot since most Meeroo genetics are gender-related. It's still Mendelian genetics; the difference simply is the ordering (dominance) rules applied.

Part 2. Dominance
Before we can go further, we need to determine the dominance rules. Above, it was stated that Meeroo genetics are gender-related, not linear as we see with most other breedables. How do we know that? The answer is in the Eyes.

The eye color traits were easily observed to appear to be female-dominated. But they offer one more clue: Clarity. And Clear eyes are how we prove, without doubt, that it's female-dominated.

With some experimentation, using the Mendelian Laws, it's easy to produce Clear eyes using the assumption that the Mother's eye colors are what determine the expressed (visible) color in the offspring. In the example above, the Aggressive(aggressive) offspring is Clear Personality (sic); all others are Dusty.

We breed until we have a few males and females showing Clear eyes. Then we cross them. For example, let's take an Eclipse(eclipse)[Clear] female and a Lunar(lunar)[Clear] male. When we cross them we note that all offspring are Eclipse(lunar)[Dusty] .. clearly (ha!) the female's genes are dominant over the male's.

Further experimentation, then, showed that Personality, Species, and Coat were, like Eye Color, determined by the female parent.

Maximum Size also appears to be female-parent-dominated, but a randomization occurs after the genetics are determined. That randomization overshadows the genetics sufficiently that it has proven hard to prove the genetics.

There is still some question about the dominance for Head, Fur, Ear and Tail. Some of these have shown a trend toward male-parent-dominance; but that could be skewed results because of the relatively low number of experimental subjects. As more evidence is appearing, these traits are beginning to appear to be female-parent-dominated like all others.

Part 3. Breeding patterns
Armed with the Laws and a good working theory for dominance, we can now develop some breeding patterns. Note that we've not yet examined the effects (if any) of Regard and Mutation (sometimes called Reaching Back). So, while these patterns will, for the most part, work we need to remember there is more going on and we'll sometimes get unexpected results.

The first pattern is Gene Discovery.

With this pattern we can quickly learn which genes we have so that we can get on with the job of crossing those genes into other lines to design the Meeroos we want. For the Gene Discovery pattern, we need a minimum of one female and two males. That minimum, however, is not as efficient as if we have two females and two males.

Gene Discovery (minimum population)
Select one male, which does not matter. Breed it repeatedly with the female. Take all female offspring and breed them repeatedly with the other male.

Gene Discovery (maximum efficiency)
Select one male and one female, which does not matter. Breed them repeatedly. Select the other male and female and breed them repeatedly. Take the female offspring from each pair and breed them repeatedly with the male parent of the other pair.

Note the time requirements for both patterns are roughly equal. The second pattern, however, if run to completion, exposes 8 out of 8 genes where the first only exposes 6 out of 6, and leaves one male idle at all times. The second pattern leaves the original females idle once their daughters are crossed to the other line. Those females can be used to start a new pattern, released for regard, or held for a few days until grandsons begin appearing (which leads to the next pattern).

The next pattern is Gene Stabilization.

The goal at this point is to produce a family line which consistently expresses a desired gene. Here, again, there are two choices: based upon the gender of the offspring.

The pattern is simple: breed the grandson to his grandmother; breed the granddaughter to her grandfather.

If you're using the maximum efficiency Gene Discovery pattern, you'll have idle grandmothers long before you'll have idle grandfathers (if your grandfathers every do idle before aging out of the game). You'll most likely want to get them back into production by breeding their grandsons against them.

The question is which grandsons to choose.

If the grandsons express the same trait as their grandmothers, about 3/4 of the offspring will carry that trait. In addition, of all offspring, about 1/2 will express the trait, and about 1/4 will carry only that trait (i.e. for eyes, will be Clear eyes).

Part 4. This is all well and good, but it doesn't work!

Once you've been breeding Meeroos a while you'll note that, while all the above seems OK and usually works, at some point it fails and things fall apart.

This is Regard and Mutation rearing their heads.

Some call this effect "reaching back through the generations to find a long-lost trait". Yes, it's possible that, as you were breeding, a hidden trait remained hidden for several generations before reappearing. But even then, some traits appear which have no explanation from the Genetic Models, Breeding Patterns and Genetic History.

Where do those come from?

Well, this is what sets Meeroos apart from all other breedables. The simplest way to think of it is this: at some level, *ALL* Meeroos are 'starters'. They all have a chance to, like starters, randomly choose a new hidden trait.

For the offspring of a given pair, that trait may appear only once, or it may stick around. But whatever it does, there is no logical explanation for its appearance. We can't look back and see the trait appeared in great-great-grandpa and, due to luck and your breeding choices, has silently been carried down the generations until it reappeared. No matter where we look, we can't find the gene, and we can't even find a way for it to have been missed, and silently carried down from our true starters.

This is Random Mutation.

It's been described in terms of "reaching back" past our starters, past Levio and Dawnara to "long lost genetic history". That's all well and good as a story, but for our purposes, it's simply: Random Mutation.

Part 5. Mutation

We can't control when or how mutation occurs, but we can make some observations.

First, it's been given to us, and experimentation has shown, that the Meeroo's regard, as it increases, increases the chance of a mutation.

Second, it's been given to us that our personal regard scores effect the chances of a hidden trait appearing in our true starters. It's reasonable to assume that our regard scores also effect the chances with each new nest as well.

Finally, it's been given to us that the genetic history plays a roll in directing which mutations occur.

Let's go back to the original examples. Yes, I know Personality is not effected by Regard and Mutation, but why change horses now? If it confuses you, just read Gael coat names instead of personalities.

We have two parents:
Female: Aggressive(lazy)
Male: Aggressive(shy)

But now, the (lazy) and (shy) hidden traits are not fixed values. In fact, now, we can't even prove they actually exist!

As always, mother and father randomly select a trait to pass. But now, if that trait is the hidden trait, we don't just pass it: we make some additional checks. To keep it simple, I'll only discus Mother's side of things. The same sort of thing is going on with Father's side.

The first check is to examine the parent's actual genetic history. Mother is Aggressive. That means she's quite likely to pass Aggressive. We look at her parents and see her mother was Mischievous. So we add a chance for Mischievous. Her father was Lazy, so there's a chance of Lazy. We continue backwards, noting and counting the appearance of each trait on each ancestor. When we're done we have a probability chart for ancestral traits.

But we're not done. Now we look at the Meeroo's Regard and the Player's Regard. Using those, we add one more possibility: anything can happen. It's been given to us, and we've assumed (but not yet proven), that higher scores net a larger chance of anything happening other than the genetic history.

So, if Mother randomly chooses to pass her expressed (visible) trait, all is well. The new nest will have that trait and, since it came from Mother, it'll be expressed (visible).

But ... if Mother chose instead to pass her hidden trait we need to run through all the work to determine her genetic history, calculate the chance of a random mutation, then randomly choose from those. Then, if a random mutation is selected, we randomly choose, once again, from all possible trait values. (Yes, this probably means we have a chance of 'randomly mutating' right back into the genetic history. So be it. No problem.)

Now, assume our examples are 'starters'. They don't have any genetic history other than what we see. We don't even know that Mother's (lazy) actually exists but let's assume it does. To make it easy, Father will always pass his expressed (visible) trait. We get the following outcomes:

Aggressive(aggressive)
???(aggressive)

Assuming even odds (probably wrong but easy to calculate, we expand these possibilities:

Aggressive(aggressive)
Aggressive(aggressive)
Aggressive(aggressive)

Aggressive(aggressive)
Lazy(aggressive)
???(aggressive)

I've repeated Aggressive(aggressive) three times to represent the fact Mother has a 50% chance of passing her expressed (visible) trait. The next three represent even odds of her visible trait, hidden trait, or a random trait appearing.

We now expand the table one more time to represent the even-odds of randomly choosing ANY trait.

21 x Aggressive(aggressive)
6 x Lazy(aggressive)
1 x Friendly(aggressive)
1 x Mischievous(aggressive)
1 x Shy(aggressive)

Experimental evidence suggests the above is roughly what is happening in the game. (Well, not for Personality, but for most of the other traits).

Remember, I stated Father always passed his visible trait, so all offspring here have (aggressive) as their hidden trait. I leave it as an exercise to expand the table to include all possible traits from Father.

Given that, this result table shows the expected frequency of these visible traits appearing from our example starters.

Part 6. So where do we go from here?

All this, of course, assumed even-odds distribution and the regard effect being equal to all others. Obviously, this is wrong. The distribution odds for traits are almost certainly NOT equal for all values. And the effect of Regard almost certainly varies over time.

We're left, then, at the current state of research for players

- what is the distribution odds for given traits

- how is genetic history determined

- what is the actual effect of regard upon the selection

and equivalent points which only the game developers can research, but which can be summarized as:

- is the effect of all this sufficient to produce results often enough to maintain player interest

Part 7. Wait a minute .. what about Coats?

Read back through the Laws. See that bit about "linked traits"? Well, Species and Coats are "linked traits". Individually, they behave just as any other trait.

But the words change. If we have a Gael(pine) Fawn(ursine) Mother, she can choose to pass Pine Fawn to her offspring. But, you say: there is no Pine Fawn! Well, actually, there is .. we just call it Pine Winecoat, instead.

This leaves one more point of player research:

- what are the equivalences of coats between the various species