Reproductive Success and Fitness are not the same thing

Reproductive Success (RS) is defined in many ways in different places by different people, but one of the most common definitions is simply the number of offspring an individual produces. This definition is further modified in most cases to mean only those individuals that will be fertile, i.e., capable of producing further offspring. RS is important in understanding Natural Selection (NS). In the simplest model, a heritable feature that increases RS will be selected for over time in a population because individuals with higher RS will contribute more offspring to future generations and this, in turn, causes the frequency of the RS-enhancing allele (gene variant) to become more common over generational time in that population.

Fitness is a property of an allele that refers to its relative likelihood of representation in future generations in a population. An allele with higher fitness will be more likely to be represented in future generations within a population than an allele for the same gene with lower fitness. The important thing here is that the likelihood of future representation has to be due to a feature of that allele, and not random effects.

At first glance RS and Fitness are the same, or similar, but one might immediately notice that RS is a feature of an individual (that has offspring) while fitness is a feature of an allele. So, it is possible that a given individual will have a relatively high RS but contain a particular allele with low fitness. Presumably the higher RS of that individual is due to high-fitness alleles of other genes. In this way, fitness and RS are different, but when considering large scales of time and large populations, the two can be (perhaps) safely conflated because things average out over time and the different alleles are being independently assorted over generational time, so each allele gets to have its day, sometimes, independently of other lower-fitness alleles. By this way of thinking, RS and fitness can be safely considered as measures of roughly the same thing, but with caveats.

RS is usually measured, in actual experimental work or field observations, as the number of offspring observed for an individual, but to make sure that RS is correlated with fitness, one might measure grand-offspring in order to factor out infertile offspring and other factors that may affect one generation but that do not apply over the long term. Again, RS and fitness are then, it would seem, equatable but with caveats.

RS is the number of offspring or grand-offspring but kin selection may apply as well. This is where an individual foregoes some of its own reproduction for the benefit of a relative, causing indirect fitness, a measure of this contribution devalued by the probability of the two individuals sharing the same allele by common descent. One can state that a measure of RS is still a measure of fitness because over the long term, again, things average out, but equating fitness and RS is done, again, with caveats.

There may be an optimal number of offspring an individual may have, above which longer term reproduction is reduced. A litter that is too large may result in a set of adults that are smaller than ideal and will thus have fewer offspring, or in the case of serial reproduction, if parental investment is spread out over several offspring, having too many offspring in a row may cause a deficit for all of the offspring, or for the later offspring that get less care because less energy is available, or earlier offspring may get short changed by being left on their own sooner. Putting this another way, the ultimate long-germ fitness strategy may be to have X offspring, where having more or fewer than X results in a suboptimal outcome. In this way, increasing RS from zero towards X increases fitness, but increasing RS beyond X decreases fitness.

So, RS equals fitness except:

• RS is a measure applied to an individual while fitness is ideally applied to alleles for a gene or some other genetic construct;
• The offspring-fertility link can be misleading. A queen bee with an allele that allows her to produces more sterile offspring may also produce more fertile offspring;
• RS is fitness plus or minus random effects;
• RS usually does not consider indirect fitness;
• RS is selected to be optimized while fitness is selected to be maximized.

Equating RS and fitness is therefore only a rough approximation. When initially learning about Natural Selection students are often led to believe that RS and fitness are the same, which is only true with these (and possibly other) caveats. Equating RS and fitness in pedagogy risks skipping past and perhaps never understanding the caveats, and these caveats are very far from trivial. They are, in many cases, the point of specific evolutionary research projects.