Does genetics ever "split the difference" - ie fat + thin parents = medium child?

I know there are dominant and recessive traits, but in some cases does genetic inheritance ever “split the difference” physically, so that a fat parent + thin parent would tend to have medium weight children… or does it not work this way at all?

Yes, it happens often for many traits, though fatness is a factor of a lot of things besides genetics.

For an easy example, take a look at kids of black and white parents, who tend to have medium skin-tones.

Would racially-mixed children count?

It does somewhat work like that, but it’s way more complicated. If that were totally true then there would be so such thing as blond hair or green eyes. Everyone would just be a muddle of everything after so many generations.

Actually, splitting the difference is the rule, because so many outwarldy visible physical characteristics are the result of not just one gene, but of many, many genes.

It’s only in some very specific cases, such as eye color, where you can see the influence of a single gene (or very small number of genes). And even here, it’s not so simple. A blue-eyed daddy and a green-eyed mommy can have:[ul]
[li] a blue-green-eyed kid[/li][li] a blue-eyed kid[/li][li] a green-eyed kid[/li][/ul]
Simple dominant/recessive Mendellian inheritance is a fine model for pea plants, but in higher animals, things get very muddied very quickly.

It’s not even a fine model for most traits in pea plants. It just happens that the rules of inheritance of dominant-recessive traits was first worked out for them.

For any trait that is determined by a large set of different genes, rather than just one - that is, multifactorial traits - intermediacy is the rule.

This phenomenon, known as “regression toward the mean,” was first described by Darwin’s cousin Francis Galton. Unlike Mendel, who deliberately studied discontinous traits (such as yellow and green color in peas), Galton studied ones that varied continuously, like height.

This site has an extensive discussion of multifactorial inheritance.

A blue-eyed daddy and a green-eyed mommy can have:

Also a brown eyed kid.

Let me see how well I understand this…

We all know that genes occur in pairs, in corresponding locations on the right pairs of chromosomes. Some genes are recessive, meaning that they need to be in both locations to have an effect, whereas other genes are dominant, meaning that one copy is good enough to have an effect.

Genes have only one function: they tell the cells to build a particular protein. If one copy of a gene makes enough protein to cause something to happen, it’s dominant. Otherwise, it’s recessive.

So for traits that are caused by a single gene, there’s no “splitting the difference”. Either you have the trait, or you don’t.

But most traits are caused by several different genes, and here you can get averages, after a fashion. Some of the genes will be recessive, and some will be dominant. An individual can have any given mix of those, which can lead to any given scheme of protein production.

Make sense?

 A bit of nit-picking:  the simple explanation of eye color is that there are three alleles--brown, green, and blue.  Brown and green are both dominant to blue, and codominant to each other--meaning that an individual who has both a green and a brown allele has hazel eyes, but that an individual who has both a brown and a blue has brown eyes, and one who has both a green and a blue has green eyes.  Those adults who have blue eyes have two blue alleles.  According to these rules, brown-eyed children coming from a green-eyed parent and a blue-eyed parent indicate either marital infidelity or the hospital switching the kid.
 Reality is a bit more complicated.  There are more than 4 eye colors out there.  Most of them are probably just caused by variant versions of one of the alleles--it's not hard to see how a slight change to a blue allele can result in the rare-but-verified violet eyes of people like Elizabeth Taylor.  Not all people with a given eye color have exactly the same shade of it, after all.  I myself have a mutant eye-color allele which I inherited from my mother, which makes my eyes shift multiple times daily from green to blue, but most often being rather gray.  A simple pedigree shows it to most likely be a variant form of a green allele.  However, brown is still dominant to blue and codominant with green, so barring unusual circumstances (a novel mutation, etc.), you won't get a brown-eyed child from these hypothetical parents.
 As for dominance and recessiveness--these are relative terms.  Sometimes just one copy of the gene is enough to make something happen, but the products of a different gene can overwhelm it.  Delete the second gene, and the first one will have a visible effect, even though there's still only one copy of it.
 In a final note, I leave the caveat of "adult" in the first paragraph because many more children have blue eyes than adults.  I'm not entirely sure of the explanation.  The one that leaps to mind is that there is reduced transcription of the other alleles in young infants compared to the same individuals later in life, and thus their phenotype changes to the expectation of their genotype later in life.

This is true in a lot of cases, but not true with the fruit flies I’ve been working with in lab. Drosophila melanogaster have only 4 pairs of chromosomes, one of which are the sex chromosomes.

One of the traits we’ve been looking at in lab involves eye color and eye shape. A female fly with two copies of the normal gene has round, red eyes. A female fly with two copies of the Bar mutation has apricot colored eyes that look like a thin bar. A female fly with one copy of each has kidney-shaped eyes with a color between red and apricot.

This trait is sex-linked (located on the X chromosome), so males only exhibit the wild-type characteristic or the Bar characteristic.

I tried to google pictures of the eyes to link to, but my computer isn’t being very cooperative right now.

Close. There are three known genes which affect eye color in humans. One can contain dominant brown or recessive blue alleles, the second can contain dominant green or recessive blue alleles, etc. A detailed description can be found here. Brown is not codominant to green, but is in fact dominant over it in this model.

Note the following:

well my father has black hair/blue eyes and is 6’2" and my mother has brown hair/eyes and is 5’4" one of my brothers is 5’10", the other is 6’ and I am 6’2" (female) and we all three have brown hair/eyes


The correct genetic term for a monogenic “split the difference” trait is called incomplete dominance. We see this with many traits in humans, because we can analyze humans so closely. So sickle cell anemia, cystic fibrosis, and other disorders traditionally thought of as recessive are actually incompletely dominant because we can see small phenotypes in the carriers.

With more than one gene, this becomes the rule rather than an exception.


On/off is not the rule even in single gene genetics. You can change the level by enhancer changes, you can change the activity of the protein with coding region changes.

We call a mutation which leads to a decrease in activity (either in expression or protein activity) a hypomorph. One which increases is a hypermorph. If the level of the gene is tightly controlled, these mutations can be dominant. For most genes, though, they are recessive – there is a lot of play in the level of expression. In rare circumstances (haploinsufficiency), hypomorphs can be dominant.

There are strange genetic situations which are neither hypo- nor hypermorphs. These make our lives interesting. Examples are neomorphs (a mutation gives the protein a new function) and antimorphs (a loss-of-function/gain-of-function mutation which acts against normal protein). We call antimorphs dominant negatives. You can imagine if proteins are polymerizing to make a chain, a dominant negative would “poison” the chain.


Nice to see another fly person here. So I’ll nitpick something in your post. I assume you are talking about the X chromosome Bar mutation. This is not associated with eye color phenotype – look at a balancer like FM3, which is Bar, w[sup]+[/sup]. Homozygotes have red eyes. FM7, which almost everyone uses as an X chromosome balancer, also has an allele of white on it, w[sup]ap[/sup], which gives the eyes the apricot color.

A female with one FM7 over a w[sup]+[/sup] chromosome has regular red eyes. Two copies of w[sup]ap[/sup] give a relatively dark apricot. One copy over w[sup]-[/sup] is slightly lighter, but Bar homozygotes have such a small eye that it is hard to tell the difference.

Just thought you might like to know… :slight_smile:

Several of the disease-causing mutations I test for in my lab are considered dominant mutations, because even one copy of the mutation is enough to give you the disease. However, in pretty much every case, a homozygote, with two copies of the mutation, has a much more severe form of the disease than a heterozygote, with one copy. In fact, in at least one of them, homozygotes are never seen, presumably because it’s lethal early in life.

Thanks edwino. I’m rather a novice at all this, since I’m just taking a genetics lab class and not actually working with the flies. What I know is just based on what I’ve learned in class and what I’ve seen on thousands of flies.

Close. Brown would be said to be epistatic; the term dominance is only applied to alleles of the same locus.