Methinks it is like a weasel!

This is some output from a recreation of the METHINKS IT IS LIKE A WEASEL thought experiment described in the blind watchmaker by Richard Dawkins. Its a hisotgram of 1000 repeats of the program and presents the number of solutions tried before reaching the phrase after starting at a random point.

At first the program wouldn't work and would only produce the correct phrase after I artificially allowed it keep a letter if it was correct, this was incorrect.

Luckily i got it working properly, each generation generates 10 solutions based on the two best from the last generation. These are then mutated and the best two breed. Evolution is much cooler than intelligent design. Whats exciting about an intelligent entity being able to generate a shakespearean phrase? However a computer program written by a student on a rubbish home computer that doesn't have any idea what its doing can reach the same phrase employing just two basic rules, you pass on your genes and you are more likely to pass these on the better you are. Sorted respect due.

A meeting

Discussed the problems in relation to choosiness values and how this translates into how likely the female is to be choosy.

Also this some notes i made

Neff, B.D. and Pitcher, T.E., 2005, Genetic quality and sexual selection: an integrated framework for good genes and compatible genes.

Not many notes on this.

A good review table of examples of increased fecundity due to multiple matings, the assumption being that multiple matings allows better genetic matching between sperm and eggs. In effect this is saying there is not as much need for mate selection. Males are selected by the process of fertilisation. The more compatible they are the more eggs they fertilise.

Meeting up

Discussed how to improve the model so, that choosiness could be a real choice for the population. Im still changing things on this.

Also decided that a choosiness value of 100 should count as 0 choosiness. This increases linearly up to 200 choosiness where the probability of being choosing in the mating stage is 1.

This change was implemented because i found that the population would tend slightly towards choosiness at the start until the cost started to kick in. Also a value of 100 choosiness is what you get by random mutation anyway. I wanted selection to be able to act on choosiness loci if it was useful to do so, but not act if it was too costly. I felt this change gave the best balance.

More notes on papers

I promised to post notes, so here we are.

Panhuis et al, 2001, Sexual selection and speciation, TREE, pp364-371.

Useful article, although not entirely relevant. Certainly gives some good evidence of how sexual selection can act to speed the evolution of populations and also how this can be in non-adaptive and/or quite variable directions (leading to reproductive isolation and the possibility of speciation).

It also has a very good summary box of examples of sexual selection, the traits that are selected and whether these produce reproductive isolation, example: female Drosophila melanogaster from Zimbabwe prefer males only from Zimbabwe.