Can we quantify genetic and environmental influences on a characteristic?

There are indispensable books, books that we have to have (even if we end up reading only the book flap…). Other books, down below in a list of importance, are books that we plan to buy when a chance comes. However, there are certain books that we did not even know about, and that are discovered by chance when we wander through the books on the shelves of a bookstore —you pull a stool, if there is one around, sit beside the shelves, and start skimming through pages, book after book. These books, which we find fortuitously, bring sometimes approaches that we didn’t know, once we are so accustomed to those few authors of our predilection. I like to browse the shelves at random, when I have some bucks saved at the end of the month, of course…

One of these books, that I bought very cheap in a second-hand bookstore, is “Evolution: an introduction”, from Stephen Stearns and Rolf Hoekstra, Oxford University Press. I remember at first having examined the book indifferently, because I didn’t like the cover layout (yes, that’s how human mind works…). But, luckily, the first paragraph I read was about a concept that the vast majority of people and the media define incorrectly, and not so few scientists get wrong also, and that always bothered me. But the books explained it perfectly — unnecessary to say that I bought it and, to the date, it is one of my favourite textbooks on evolutionary biology. Talking about this concept that so many people get wrong is the aim of this brief note.

We all have learned that the characteristics of living organisms or structures can be determined by genes, but these characteristics are influenced by the environment in which the organism developed and lives. Traditionally, in my country of origin, high school books call this “norm of reaction” and define the phenomenon mathematically:

Phenotype = Genotype + Environment

I have two objections already: first, technically speaking, norm of reaction is another thing. Secondly, this is a very poor way of explaining what phenotype is. Phenotype is the set of all measurable characteristics or traits of an organism, all the measurable properties of an organism. Phenotype is not the sum of genotype plus environment, which does not make any sense. Also, if we treat this statement mathematically, we could conclude that “Environment = Phenotype – Genotype”. What the hell does this mean?

Ok, let’s stop being annoying and agree that what is meant by this “mathematical” statement is that the properties of biological organisms (its phenotype) depend in part on their genes and in part on the environment where these organisms live in. So far so good, this is an important concept, and those who work directly or indirectly with biology should have this concept always in mind. The problem comes when you want to quantify these influences: imagine, for example, that someone asks something like this: “skin colour in humans depends on your genes and depends on environmental factors such as incidence of solar radiation. What percentage of skin colour depends upon genetic constitution and what percentage depends on environmental factors?”

Many people think that a question like this one is valid from a biological viewpoint and that it can be answered. Who has some familiarity with genetics could think that heritability would account for this question: heritability tells us what percent of a certain characteristic is genetically determined, and what percent is environmentally determined. But it turns out that heritability is not about that!


Phenotypic variation in a population of snails (source:

In genetics, heritability is a measure of the genetic variation contributing to the total phenotypic variation of a given trait in a population. Explaining better, heritability is the amount of observed differences on a trait among organisms of a population that are due to genetic differences (I wrote “differences” in bold and italic, and if the norms of style and modern browsers allowed me I’d write it in a blinking text, so important is to keep it in mind). More simply still, heritability measures to what extent the differences between individuals in a population are due to genetic differences between those individuals (look, bold and italics again!). Just that. We cannot, because it makes no biological sense, to determine what percent of a given characteristic is due to genetic factors.

The excerpt from Stearns and Hoekstra book, that immediately made my mind about buying it, is this:

Heritability tells us about the contribution of genetic variation to the phenotypic variation in a population. It does not tell us to ‘what extent a trait is genetic or environmental’. Taken literally, this phrase is meaningless. Try for example explain (if you can!) the meaning of the following statement: ‘intelligence in humans is determined 70 per cent by genes and 30 per cent by environmental conditions such as upbringing and education’. Clearly, both genes and suitable environmental conditions are necessary for human to exist and to posses any trait. We can only infer the extent to which differences between individuals are caused by genetic or environmental factors.

Let me explain it in a very simple and elementary example. Suppose two identical twins (a clone, biologically speaking). Suppose they have exactly the same genotype. You perform any intelligence test you want, and they score different marks. So, what you can infer is that the differences between their intelligences is 100% due to the differences between their environment, or 0% due to the differences between their genotypes.

Despite this warning about what heritability is, that you can find in most good books on genetics and evolutionary biology, these questions are extremely common: “what percentage of this trait is genetic, and that percentage is environmentally determined?”

So, next time you hear a question like “is the intelligence genetically or environmentally determined?”, you give the person a ‘wtf’ look. If he or she improves the question a little bit, like “what percent of intelligence is due to the environment and what percent is due to genes?” you already know how to answer: “Mate, this question has no answer. Not because the answer is complicated or because it has not yet been discovered, but because the question itself makes no sense”.

It’s like asking how much is 4 divided by zero.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s