Indiana University

subglobal1 link | subglobal1 link | subglobal1 link | subglobal1 link | subglobal1 link | subglobal1 link | subglobal1 link
subglobal2 link | subglobal2 link | subglobal2 link | subglobal2 link | subglobal2 link | subglobal2 link | subglobal2 link
subglobal3 link | subglobal3 link | subglobal3 link | subglobal3 link | subglobal3 link | subglobal3 link | subglobal3 link
subglobal4 link | subglobal4 link | subglobal4 link | subglobal4 link | subglobal4 link | subglobal4 link | subglobal4 link
subglobal5 link | subglobal5 link | subglobal5 link | subglobal5 link | subglobal5 link | subglobal5 link | subglobal5 link
subglobal6 link | subglobal6 link | subglobal6 link | subglobal6 link | subglobal6 link | subglobal6 link | subglobal6 link
subglobal7 link | subglobal7 link | subglobal7 link | subglobal7 link | subglobal7 link | subglobal7 link | subglobal7 link
subglobal8 link | subglobal8 link | subglobal8 link | subglobal8 link | subglobal8 link | subglobal8 link | subglobal8 link


Contents of this Section

Working With More Than One Gene

When we look at any organism, we see that it has a great many characteristics--many traits. For any particular trait, how do we know whether it depends on one gene, two genes, or many genes? At first glance, we don't know and can't tell. But we can find out by performing a series of genetic crosses.

What do we expect if the trait depends upon a single gene? We have examined a number of such traits in the previous sections, so we can describe our expectation based on what we have learned. We would expect the following:

For a trait that depends on one gene:

  • All aspects of the trait should be inherited together.
    • Example: if we think curly brown hair is one version of the trait, and straight blonde hair is another version, then we would expect curly-brown always to be separate from straight-blonde. If our crosses produce straight-brown and curly-blonde, then color must depend on different genes than curliness.
  • To the extent that one version of the trait is dominant, we should find roughly a 1:1 distribution of offspring when we cross a heterozygous individual to an individual that is homozygous recessive.
    • Example: heterozygous brown/blonde crossed to homozygous blonde/blonde should give roughly as many blonde offspring as brown.
      • (This is a test-cross, which is probably more straightforward than crossing F1 heterozygotes to produce F2 offspring.) The F2 generation is also informative, however:
  • Heterozyous F1 individuals, crossed to each other, should produce roughly a 1:2:1 distibution of F2 offspring. These would be one homozygous "dominant" to two heterozygotes to one homozygous "recessive."
    • We have put "dominant" and "recessive" in quotes here because the inheritance pattern will be the same for incomplete dominance and for complete dominance.
    • Example: red-flowered snapdragons crossed to white-flowered snapdragons produce pink-flowered F1 offspring. The F1's are red/white. Crossed among themselves, we expect red/red, red/white, white/red, and white/white. (Of course, it makes no difference to the plants whether we write white/red or red/white; these are genetically the same thing. We're written it this way only so that we can keep track of things more easily.) We should expect a 1:2:1 ratio of red:pink:white.
    • For alleles that show strict dominance, the heterozygotes are indistinguishable from the homozygous dominant, so the 1:2:1 ratio of genotypes produces a 3:1 ratio of phenotypes.

What happens when we have two traits, dependent upon two genes? Let's first consider something that looks at first as if it might be a single trait, but that turns out not to be.

back to top

Flies With Dark Eyes and Dark Bodies


Normal Drosophila
Drosophila with brown eyes and dark body

We have a fly that has dark eyes and a dark body. Is this due to an allele of a gene that makes everything dark? Or is it a combination of an allele of the gene brown (which we have studied before) and an allele of some other gene that makes the body a blackish, ebony sort of color? Let's find out.

back to top

Cross 1: normal, wild-type flies x brown, ebony flies

First, let's construct flies that are heterozygous for the wild-type allele(s) and the brown, ebony allele(s):

Because all of the offspring have red eyes and tan bodies, we can conclude tht the brown, ebony phenotype is recessive to wild-type (whether this phenotype results from variant alleles of one or two genes). These offspring are the F1 generation from this first cross.

Now that we have heterozygous flies, we can find out how the brown, ebony trait is inherited in subsequent crosses. Do brown eyes and ebony bodies always occur in the same individuals, indicating that a single allele darkens the fly? Or do some flies have brown eyes and tan bodies, while others have red eyes and ebony bodies?

back to top

Cross 2: A Test-Cross of the F1 to the Homozygous Recessive Parental Type

The data -- even as simple as the number of different phenotypes -- indicate that we are not following the inheritance of a single gene. There are two classes of offspring that do not have both dark eyes and dark bodies; brown eyes appears to be a separate trait from ebony bodies, not part of a single "darkness" trait. The separation of characteristics in this cross argues that least two genes are involved here. One is responsible for eye color, red vs brown. The other is responsible for body color, tan vs ebony.

How can we visualize what's going on here?

In the illustration above, we have shown an F1 female. She is presumably heterozygous for the alleles involved here:

or, using abbreviations for the gene names,

[For now, we will refer to the variant alleles as brown1 and ebony1. There are many other alleles of each gene (e.g. brown75, brownD, and ebonysooty, to mention a few of them.]

During egg production (i.e. during meiosis), what will happen with these two genes?

  • bw+ will segregate from bw1, so that half of the female's eggs carry bw+ and half carry bw1
  • e+ will segregate from e1, so that half of the female's eggs carry e+ and half carry e1
      • [Are there any other possibilities? Isn't this how genetics normally works?]

If each gene behaves independently of the other during meiosis, then half of the eggs bw+ should carry e+, and half should carry e1. The ebony alleles should be distributed similarly in the bw1 eggs. We can visualize it this way:

[Drosophila eggs have two little extensions from one end, which serve as little breathing tubes for the embryo. The mother fly inserts the egg into the fruit (or fly food), with most of the egg buried, and these two little "snorkels" sticking out.]

Since the male is homozygous for bw1and e1, all of his sperm cells will carry these alleles. But it gets confusing trying to keep all of this straight in our heads; let's use one of those tables that Punnett developed to make this easier. Along the left side, we'll write out the 4 genotypes of eggs that we show in the drawing above. Along the top of the table, we'll write the genotype of the sperm. There's only one, so we need only one column (the males are homozygous in this test-cross).

Now, just fill in the table.

Using Punnett's handy table, it's fairly easy to keep track of the possible combinations of alleles. When we do a test-cross, in which we cross our "unknown" genotype back to a parent that is homozygous for the trait(s) we are examining, we do a moderately simple experiment that gives a straightforward result. We expect--as indicated in the table--roughly equal numbers of the four different types of flies.

When we actually do this cross with Drosophila, we can easily examine thousands of offspring. A typical result would be 586 red eyed, tan bodied flies; 523 brown eyed, tan bodied flies, 547 red eyed, ebony bodied flies, and 518 brown eyed, ebony bodied flies. That is, we observe roughly 1:1:1:1.

We can explain the 1:1:1:1 result and the four different phenotypes by the hypothesis that two genes are involved, and that they segregate independently from each other during gamete formation (i.e. during meiosis).

back to top

Cross 3: Crossing the F1 Flies Among Each Other to Produce the F2 Generation

The test-cross described above is a good, direct test of the type of inheritance of a trait (or traits), but it is not the only way to investigate this question. One can also cross the F1 individuals among themselves. Gregor Mendel did this with his pea plants, and we can do so with flies as well (but it's not a good idea for studying human genetics). Let's take advantage of the information we gained from the test cross, and apply it to the problem of crossing F1's among one another.

The best explanation of the observations we obtained from the test cross is that the F1 flies are heterozygous at two different genes, brown and ebony. The 1:1:1:1 ratio of phenotypes among the test-cross offspring suggests that these two genes behave entirely independently during meiosis. So... many different female gametes will we have in this cross?

[Shouldn't this be the same as for cross 2, the test-cross? They are the same flies, after all.] many different male gametes will we have?

[Shouldn't these be the same distribution as for the female gametes?]

Different people use different "mental tricks" to figure out the possible gametes. Whatever trick works for you is good; all you are after is the possible combinations of alleles. Some people think of it this way:

For the first gene, the allele "on top" (or on the left, if you use the Aa type of symbols) will be paired in some gametes with the allele "on top" for the second gene. In other gametes, it will be paired with the allele "on the bottom" (or on the right). This gives you half of the possible combinations. For the allele "on the bottom," follow the same logic. This gives you the other half of the possible combinations.

To keep track of the possible combinations of gametes, we really need to use a Punnett Square. It's just to confusing without this tool to help us keep things organized:

Again, we simply fill in the square. But with a 4 x 4 table, there are 16 combinations. By the time we write in all of the allele combinations, the table is pretty crowded and hard to read:

If you can sort through all of the + symbols and 1's (or if you've used the Aa Bb nomenclature, the capital and lower-case letters), you can figure out the phenotypes of each of these 16 different squares in the table. It might be easier, however, if we use completely different symbols for these genes. [If we were planning to communicate with actual Drosophila geneticists, we'd need to use the correct terminology and symbols, like ebony+ and ebony1, or ebonyb or ebonysooty if we were working with these other alleles. But for our own work, just trying to keep track of things, we can use any symbols we want. The symbols we write in a table don't affect the flies, after all; they are merely to help us figure out what the flies are doing.]

What if we use colored circles to represent the brown gene, with red circles for brown+ and brown circles for the mutant allele, brown1? Let's also use colored symbols--perhaps triangles--to represent the ebony gene, with tan triangles for ebony+ and dark gray triangles for ebony1. Now, when we redraw the Punnett Square, the colors sort of leap out at us, almost showing us how the genotypes relate to the characteristics of the flies themselves.

This looks like it might be somewhat easier to interpret. We have the different classes of gametes at the top and on the left; we can easily tell from the shapes of the symbols and the colors of the symbols that we have all four possible combinations. When we combine male gametes with female gametes, in the table itself, we can tell at a glance that the colors in the upper left of the table are different from those in the lower right.

We already know that red-eyes (brown+) is dominant to brown eyes (brown1). Therefore, any square in the table with one red circle represents a class of offspring with red eyes. Only the 4 squares with two brown circles represent flies with brown eyes. Similarly, we already know that tan-bodies (ebony+) is dominant to dark bodies (ebony1). So, any square with one tan triangle represents a class of flies with tan bodies. Let's draw in some background colors in the table to help us see this:

red eyes vs brown eyes
tan bodies vs ebony bodies

It's a little more difficult to highlight the table with both sets of colors, to illustrate the distribution of all possible phenotypes, but we'll try it this way: we'll represent body color with a thick border around each box in the table, and eye color with the central part of the box, like this:

This is starting to look messy, with so much information, but we might need all of this information to figure things out. Let's see here...

How we filled in the table
How it relates to the flies
Any time we have a red circle, we've filled in the box with red. Red-eyes is dominant to brown-eyes. So any fly with at least one wildtype allele (brown+) will have red eyes.
Only when we have two brown circles have we filled in the box with brown. Flies must be homozygous for the mutant allele (brown1) to have brown eyes.
Any time we have a tan triangle, we've colored the border of the box tan.

Tan-bodies is dominant to ebony-eyes. So any fly with at least one wildtype allele (ebony+) will have a tan body.

Only when we have two gray triangles have we colored the border of the box gray. Flies must be homozygous for the mutant allele (ebony1) to have an ebony body.

Genotypes and Phenotypes

Any time we study genetic inheritance, we can see only the characteristics of the individual organisms--their phenotypes. We must infer their actual genetic makeup--their genotypes--without actually seeing the genes or their alleles. Reginald Punnett's ingenious table enables us to keep track of both genotype and phenotype, provided we have some understanding of the alleles involved. Here, we do understand the alleles well enough--we have already examined the test-cross of the F1's to the homozygous-rececessive brown ebony parental type of flies. So, we already know which alleles are dominant over which others, and that we are dealing with two genes. To be very clear about it, let's separate the complicated, multi-colored Punnett Square into two separate squares, one for genotype, and one for phenotype:


We built the Punnett Square on the left by knowing (actually, inferring) the genotypes of the parents; the boxes of the square represent predictions of the possible combinations of gametes. The square on the right is another prediction, this time using our knowledge of the contributions of the alleles to the final organismal phenotype.

Note, however, that several of the genotypes in the table are identical. For example, there are four different combinations of gametes that can produce flies that are heterozygous for both alleles of brown and for both alleles of ebony. (These are on a diagonal from lower left to upper right in the table.) Although these four genotypes turn out to be the same, we nonetheless keep track of them in our table.

If you compare the results of the F1 cross to the results of the test-cross, you find that both crosses produce the same four phenotypes of offspring. The relative numbers are different, however. In the test-cross, we found a ratio of 1:1:1:1. Here, there are more possible combinations of gametes (16), some of which produce the same phenotypes. If we simply count the numbers of squares of different colors in the right-hand table above, we get a predicted ratio of 9:3:3:1.

Of course, predicting a particular ratio does not mean we will actually see that ratio. We certainly will not see that ratio if we only look at a handful of offspring. We will need to count a very large number of flies, and even then, the numbers will not be perfect. Nonetheless, we should be able to arrive at the same conclusion as for the test-crosss: we are dealing with two genes (brown and ebony), and they segregate independently from each other.

Flies with White Eyes →

back to top