## Measuring the Strength of Selection

Evolution by natural selection relies on heritable variation and the strength of selection. For that reason, examining those two different factors is important. Scientists can easily measure how phenotypes vary in a population. What they don't know as easily is how much of this variation is heritable. But they can find out.

Figure 7-1 shows the variation of a particular quantitative continuous pheno-typic trait in a population — in this case, height. The measured heights fall along the x-axis. Individuals on the left are shorter than individuals on the right. The y-axis shows the frequency in the population of different phenotypes. In this particular example, most of the individuals cluster in the middle range; some of them are very short, and some of them are very tall, but most of them are in between.

To find out how heritable height is in this population, you can selectively breed the tallest (or the shortest) individuals and then examine the frequency distribution of their offspring's height. Figure 7-2 is the same frequency distribution as Figure 7-1, highlighting the subset of the original population you plan to use to create the next generation in your experimental population. In this case, assume that you picked the tall ones.

Figure 7-1:

Variation in heights.

Height

Figure 7-2:

The population used in the next generation.

Height

Talle r

Height

### Talle r

So what will the offspring population look like? Specifically, what will be the frequency distribution of phenotypes of this quantitative trait? Figure 7-3 and Figure 7-4 show two of many possibilities. Each figure shows two frequency distributions. The first is the frequency distribution of the phenotypes of the original population, and the second is the distribution of the offspring population resulting from breeding only the tallest individuals. In both cases, the average height of the population has been shifted to the right. In both cases, on average the offspring population is taller. But in Figure 7-4, the population has shifted much farther to the right, meaning that the offspring population is significantly taller.

A result such as the one in Figure 7-4 tells you that the trait (in this case, height) is much more heritable in that experiment than it was for the case in Figure 7-3. The difference between the two figures can't be due to the strength of selection, because that was the same in both cases. (Remember, you were the selective agent because you picked which ones would have offspring.) Given that the strength of selection was the same, the difference in the response to selection was purely a function of the difference in heritability between these two populations.

Figure 7-3:

Phenotypes of offspring from the original population.

Original Small change

^^^^ i i — " i i

Height

Taller

Shorter

Height

Taller

Figure 7-4:

Height distribution of the offspring of the tallest people.

Large change

Height

Taller

Height

### Taller

If you increase the strength of selection, you would expect an even greater increase in the change in average height in the offspring population, which is exactly what Figure 7-5 illustrates. Here, the new population was founded with a much smaller subset of the original population, comprising only the tallest of the tall (compare the shaded area here with the shaded area in Figure 7-2). The result is a greater shift toward an even taller offspring population.

All of this — QTL mapping, continuous/non-continuous and additive/nonadditive traits, broad- and narrow-sense heritability — is pretty academic, and you probably need to be a scientist (or a very devoted reader) to grasp the fine and not-so-fine points of the topic. But anyone can appreciate the advantages that the study of quantitative genetics can bring to fields that touch us all. The closer we get to figuring out where all the genes are on the genome, how they interact, and what they actually do, the closer we get to treating diseases that confound us now.

Figure 7-5:

Increase strength of selection, and increase the height of the offspring.

 Largest change Height Talle r Shorter Height Talle r
0 0