Over the past few days I've been studying genetic concepts and theories. I won't claim to have any deep understanding of genetics because it's not my field of study. Like most, I'm going to have adjusted what I read to fit my field; so I'll certainly make some mistakes in terminology.

As I study the various hints and clues given by the game designers, I see three main themes:

1) Starters are not required.
2) Random mutation is common and explained as throw-backs.
3) Genetic drift is central to the genetics system.

Actually, in my way of thinking, points 1 and 2 are actually the same subject. The game designers desire to eliminate the dependence upon starters for the distribution of new (or, at least, undiscovered) traits. The purpose of starters is to provide a means of INITIAL random distribution of traits. By striving to eliminate any dependence upon starters, they naturally needed to add strong random mutation to distribute traits during the normal course of game play.

While I might question the wisdom of this avoidance of starters (it smacks to me to be based mainly upon logical fallacy) that is not where the principle problem seems to lie.

Much has been said about the desirable practice of "breeding down the generations". At the same time, there have been rather bold hints that the genetic heritage of a family line strongly effect the offspring from that line. To me, this is an attempt to state that Genetic Drift have been pre-programmed into the system.

It has been stated that, if a family line "is strong" in a given trait, one should expect that trait to remain strong. Under the theory of Genetic Drift, this, then, tends to either eliminate, or fully distribute, those traits which do not have a strong relation to procreation or survivability. For Meeroos, gender and personality are the prime traits which effect procreation and survival is not an issue. All other traits, then, are most likely subject to Genetic Drift.

In the literature, Genetic Drift is stressed as having the strongest effect on small or fractured populations. Meeroos certainly fit the description of small populations and, even with the Fellowship system, their populations are also, generally, well-separated.

In other words, without any additional factors .. that is, if mating is completely random, some traits will be distributed until they appear in 100% of a population, and others (most, in fact) will be eliminated. Given the breeding requirements for Meeroos, the pre-programmed effects of Genetic Drift would seem likely to rather quickly take effect; with the most likely result, as in the Real World, being the elimination of the more rare genetic variations.

Genetic Drift, however, is most strongly caused when breeding choices are influenced rather than left to random chance. In fact, it seems plain the principle goal of most breeding programs is to re-enforce Genetic Drift in carefully selected breeding choices.

Looking at history, the game designers have fairly consistently under-estimated player psychology. They have even admitted to this in some cases. It is my belief that the effect of intentional breeding is another case of under-estimation. In this case, though, the problem is compounded by the game's already programmed tendency toward a strong Generic Drift effect. Players will, for any number of reasons, naturally choose to pair up traits. Whether this is an intentional choice to re-enforce those traits, or just a heuristic used to simply the breeding process, the effect will be the same: genetic variation will tend to be eliminated.

The game designers obviously recognized that Genetic Drift would need to be overcome. Their chosen method to do so was to increase the rate of Genetic Mutation. They make the case that this is not really "mutation" but the effect of "throw-back" to earlier generations. While there is some discussion in the literature about genetic throwbacks, it's clear the effect is generally accepted to be either a manifestation of certain types of mutation (since the rate of throwback generally follows the rate of mutation), or simply random chance taking effect in the last stages of Genetic Drift. While the scientific basis of this approach to offsetting the effects of Genetic Drift is questionable, it's a valid choice in game design.

Two questions then arise:

1) Can the rate of mutation, or throw-backs if one prefers, be adjusted sufficiently high to allow the players to feel they're making acceptable progress for the resources expended. At the same time, can the rate of mutation be kept sufficiently low that the game does not devolve into a simply lottery.

2) Is there an approach players can take to reduce the effects of Genetic Drift.

The first question can only be answered by the game designers, and then, most likely only by trial and error.

In answer to the second question, though, yes, there are actions players can take. These actions, though, may seem counter-intuitive. Basically, to counter-act the games pre-disposition to Genetic Drift, players should take care to ensure their breeding choice exhibit the greatest genetic variation. In other words, players should, rather than match like to like, should strive to ensure their family lines represent the greatest variation possible. Extensive cross-breeding between many family lines, taking care to ensure as many trait variations as possible are represented, should have the best chance at counter-acting the game's natural tendency toward Genetic Drift.

Simply stated: the best breeding choice a player can make is the one which strives the hardest to show any attempt at intentional breeding.

Real World models of such breeding programs are readily available. Zoos and conservation organizations, for example, provide many ready examples.