A visible marker tells you where to look. It does not certify the stock.
A balancer chromosome helps maintain a chromosome carrying an allele that would otherwise be difficult to keep in a stock. It combines features that discourage the recovery of recombinant chromosomes with markers that help people recognise carriers. In a familiar Drosophila stock, the visible marker may let you sort flies quickly, but it does not directly tell you which allele remains on the other chromosome.
This guide follows a fictional recessive-lethal allele through a balanced cross, then asks what would happen if a marker were misread, a stock were mixed up or a chromosome exchanged a segment. The aim is to make the evidence clear: what a cross predicts, what a marker shows, and what needs a separate test.
Genetics subject guide · Learning that transfers into life · How X Works hub
Choose your reading route
Route 1 · Chapters 1–3: Build the stock model
Draw both chromosomes before naming the stock
Name the chromosomes, follow the cross, and keep zygotes separate from scored adults.
Route 2 · Chapters 4–6: Read marker evidence carefully
Ask which chromosome the marker is actually reporting
Distinguish marker expression, chromosome identity and the intended allele.
Route 3 · Chapters 7–9: Find out what changed
Use crosses to separate competing explanations
Use control crosses, direct assays and independent samples to diagnose a suspect stock.
Route 4 · Chapters 10–12: Transfer and document
Change the survival model and recalculate
Handle altered viability, multiple chromosomes and a verifiable handoff.
Full chapter index · Independent worked case · Questions and practical checks
Chapter index
Build the stock model
Draw both chromosomes before naming the stock
Follow a fictional balanced cross to adulthood
Keep the observation stage in every ratio
Read marker evidence carefully
Ask which chromosome the marker is actually reporting
Understand what inversion suppression does and does not promise
Read an unexpected fly without naming a cause too early
CHAPTER 1 OF 12 · Build the stock model
1. Draw both chromosomes before naming the stock
Start with two homologues, the pair of corresponding chromosomes in an individual. On one homologue put an allele we will call m. On the other put B, a particular balancer chromosome. The notation m/B says which chromosome is opposite which. It does not say that every other locus on those chromosomes is identical. In a real stock there may be additional variants on both homologues.
A balancer is not just a special allele at the m locus. In Drosophila, balancers commonly carry rearrangements, often multiple inversions, along with markers and other genetic features. Their design helps reduce the recovery of certain recombination products across regions of interest. Visible dominant markers can help identify flies carrying the balancer; recessive lethal or sterile effects are common in particular balancers, but their precise behaviour is stock-specific.
Why go to this trouble? A recessive-lethal allele can be carried by an individual with another functional copy of the relevant gene, yet fail to survive when paired with itself under the conditions in question. The task is to maintain carriers across generations. The balancer helps make that task trackable. It does not make the lethal genotype viable, repair the allele or identify the allele for you.
Write the chromosome, the intended allele and the balancer identifier in different fields. A label saying simply “balanced” does not name any of them. If the next reader cannot reconstruct which chromosome was meant to be preserved, the first failure has already happened before anyone looks at a fly.
CHAPTER 2 OF 12 · Build the stock model
2. Follow a fictional balanced cross to adulthood
Assume both parents are m/B. For this teaching example, each transmits m or B with equal probability, and we ignore recombination for the first calculation. The four equally likely parental combinations yield one m/m, two m/B and one B/B among four expected zygotes. The two m/B combinations differ by which parent supplied each chromosome, not by the pair they contain.
Now assume m/m and B/B die before scoring, while m/B survives and shows the dominant balancer marker. Under those assumptions, about half of the zygotes survive to the adult scoring stage and all surviving adults are marked m/B. The observed marked fraction is one among the survivors, not one among all zygotes. If you only count adults, you have not measured the number of fertilised eggs.
A typical small vial will not deliver exactly those expected numbers. Chance matters, and unrelated losses can alter the number of adults. The model makes a prediction about classes and proportions across repeated observations under its assumptions. It does not license converting a small adult count into a precise statement about embryo mortality.
Try changing one premise: B/B survives to adulthood but is sterile, while m/m still dies. Adult scoring now includes m/B and B/B, both potentially marked. The marker still points to B carriage, but it does not prove an adult can found the next generation or that it carries m. The changed survival premise changes what you can infer from the same marker.
CHAPTER 3 OF 12 · Build the stock model
3. Keep the observation stage in every ratio
Say a fictional record starts with 80 zygotes under the simple model. Before losses, the expected class counts are 20 m/m, 40 m/B and 20 B/B. If the two homozygotes are fully lethal before adulthood and there are no other losses, about 40 marked adults are expected. A result such as 38 marked adults is compatible with several processes; it is not a diagnosis by itself.
Suppose instead you are handed a vial with 40 marked adults and no earlier count. You cannot infer there were 80 zygotes merely because that number was used in the illustration. More eggs could have been laid, not all may have been fertilised, and heterozygotes might have had additional losses. Start with the stage actually observed, and distinguish assumptions from measurements.
A useful notebook records the number of crosses or vials, scoring stage, scoring window, observed marker classes, and any individuals that could not be scored. The word “none” is ambiguous without that context. It can mean no unmarked adults were observed, no flies survived, no records were kept, or no genotype-positive samples were found. Those statements have different implications.
If the purpose is to test viability, choose an observation stage and a comparison that can show it. If the purpose is stock identity, surviving marker counts can be a quick check, but they do not independently establish the genotype. An observation may serve both questions only when the sampling and measurement genuinely support both.
CHAPTER 4 OF 12 · Read marker evidence carefully
4. Ask which chromosome the marker is actually reporting
In the toy stock, a dominant marker tells us about an observable trait associated with B. It does not read the m locus on the opposite homologue. Compare two adults: m/B and +/B, where + is an alternative chromosome not carrying m. Both can show the same visible marker. A person sorting by that trait alone cannot distinguish those two genotypes.
Suppose a stock is accidentally replaced by +/B. If you score one generation of marked adults selected by hand, the stock may look normal. Their unselected progeny can be informative: in a simple +/B by +/B cross, with +/+ viable, B/B lethal and complete marker expression in +/B, the expected adult composition is two marked +/B for each unmarked +/+. That differs from the ideal m/B stock, where no unmarked adult is expected.
The contrast helps choose a test, not deliver a verdict from a single fly. An apparently unmarked adult could be a scoring error or an exception to an assumption. An actual unmarked adult could reflect a changed genotype, an unexpected survival class or a mix-up at another stage. The next task is to find the earliest independent evidence that discriminates those possibilities.
Marker phenotype is a proxy. A genotype assay for m asks about m. A diagnostic marker assay for B asks about B. An appropriate cross can test how those features segregate or how viability behaves. When possible, keep those observations separate so agreement between them is meaningful rather than circular.
CHAPTER 5 OF 12 · Read marker evidence carefully
5. Understand what inversion suppression does and does not promise
Draw a stretch of chromosome in normal orientation beside a homologue with the same stretch reversed. During meiosis, pairing and exchanges in or near rearranged regions can have consequences that differ from exchanges between fully collinear homologues. Depending on the rearrangement and the exchange, some recombinant products may be less likely to be recovered among offspring. This is why observed recombination and physical exchange are not interchangeable ideas.
Balancer chromosomes can include several rearrangements so that they are useful for preserving particular regions. This does not make every base immune to change. Genetic studies of Drosophila balancers have documented rare exchange and differences among chromosomes that had been treated as the same named balancer. The practical implication is to check the relevant region and stock identity when they matter to the conclusion.
A marker may be far enough from a locus of interest that it does not report an exchange involving that locus. A marker can also be intact even when another portion of the chromosome differs from what you assumed. The evidence must match the question: a marker pattern is useful for sorting, whereas a sequence observation or an informative segregation test can address a particular region.
Be careful with the phrase “recombination is suppressed”. Ask what is suppressed, in which cross, across which interval, and how suppression was measured. A result about recovered recombinants in one interval does not tell you the status of every other interval. Likewise, a rare discrepancy in a stock is not proof that a recombination event occurred; several explanations remain possible.
CHAPTER 6 OF 12 · Read marker evidence carefully
6. Read an unexpected fly without naming a cause too early
Suppose 27 of 28 scored adults show the expected marker and one does not. Before calling the exception a recombinant, check whether it was scored under the same conditions, whether another observer sees the same phenotype, and whether the sample and vial identities are clear. A single observation can be valuable without telling you which explanation is correct.
There are several plausible branches. The marker might be difficult to see in that individual. The stock might contain a different genotype. A culture or label could have been mixed. The assumed lethality might not be complete under the conditions. Or an exchange could have changed the relationship between a marker and a segment. Each explanation predicts a different pattern if you choose an observation that can distinguish it.
For example, if two scorers disagree but a marker-specific genotype check is consistent across the flies, scoring conditions are a useful next place to investigate. If the intended allele is absent in multiple marker-positive individuals from a separate sample, the stock identity itself is in question. If molecular assays disagree, first check what regions they cover and whether the sample labels and controls establish that they tested what you think they did.
Keep the unexpected result rather than removing it as an inconvenience. Record its time, source, scoring conditions and interpretation status. An excluded data point should have a documented reason; a surprising animal is often the only clue that the assumptions are worth testing.
CHAPTER 7 OF 12 · Find out what changed
7. Use crosses to separate competing explanations
A diagnostic cross can answer a question about transmission or viability that a static phenotype cannot. Start by writing the predictions for two competing explanations. In the simplified model above, a suspect marked parent could be m/B or +/B. Comparing appropriate progeny classes with a verified partner can reveal a difference if the partner’s genotype and the assumptions make those classes informative.
To see the logic without treating this as a laboratory protocol, consider a model verified partner carrying m/B. A suspect m/B parent gives the familiar one-quarter m/m, one-half m/B and one-quarter B/B zygote expectation. A suspect +/B parent gives one-quarter +/m, one-quarter +/B, one-quarter m/B and one-quarter B/B. With +/m viable and unmarked, and B/B lethal, that second cross yields an expected unmarked adult class.
The test distinguishes the ideal models only when the partner is what you think it is, the scoring stage is suitable and the relevant classes are recognisable. If m has incomplete lethality, an unmarked m/m survivor could complicate the interpretation. If the marker is incompletely expressed, an unmarked adult need not lack B. Use the cross as one line of evidence, not as a universal identification test.
Record the cross purpose before seeing the result. A test designed to distinguish a suspected loss of m from the expected genotype is different from a test designed to map an exchange. Do not retrofit a mapping claim to a cross that lacks informative flanking observations. Good evidence starts with the alternatives the experiment was able to separate.
CHAPTER 8 OF 12 · Find out what changed
8. Ask the direct assay what it actually detects
An allele-specific assay can be more direct than a visible marker, but it is still an observation with a scope. A test that detects a diagnostic sequence at m supports the presence of that sequence in the sampled material. It does not automatically establish the full sequence of the chromosome, the phase of every nearby variant, or the biological effect of the allele under every condition.
A pooled sample adds another layer. Suppose a pooled test is positive for m, yet only one of several adults actually carries it. The result shows some sampled material contained the target. It does not prove every marked adult carries m. If the decision concerns individual identity or segregation, sample individuals or use a design that can support the inference you need.
A useful assay check includes known positive and negative material and any relevant controls for failure or contamination. If the assay gives no interpretable signal, report that outcome separately from “m absent”. Where a decision depends on distinguishing a complex rearrangement or unexpected genotype, a narrow assay may need confirmation by an independent method.
Link each assay result to the particular individual or sample, generation and source record. A correct result attached to the wrong sample can be more misleading than an obvious failed test. The goal is to verify a chain from the claimed stock to the observed material, not to collect impressive-looking evidence with no reliable connection between them.
CHAPTER 9 OF 12 · Find out what changed
9. Decide when a discrepancy should interrupt downstream work
Imagine two teams. One is maintaining a stock for a classroom demonstration of a visible marker. The other is using the same stock to infer whether a specific loss-of-function allele changes a quantitative trait. The second team needs stronger evidence about the allele, the comparison background and the genotype of the individuals actually measured. One marker result should not be stretched across both purposes.
A practical response to an unexplained discrepancy is to keep the suspect material identified and separate while checking a verified reference, an independent sample or archived records under appropriate local procedures. Do not silently relabel a suspect vial as the expected stock. Do not erase earlier observations just because later evidence shows they require reinterpretation.
Set the threshold for pausing the inference before you encounter a favourite result. For example, if the presence of m in the scored individuals is essential, a failed identity check means the result is not yet attributable to m. You can still document the phenotype and the uncertainty. What changes is the claim you are entitled to make.
A traceable report distinguishes “an unusual phenotype was observed”, “the intended allele was verified in the samples”, and “the allele caused the phenotype under this comparison”. Those are three different statements. The first does not imply the second, and the second alone does not establish the third. Each needs evidence fitted to its own question.
CHAPTER 10 OF 12 · Transfer and document
10. Change the survival model and recalculate
Return to the original m/B by m/B cross. Now assume that before adult scoring, 20 percent of m/m zygotes survive, all m/B survive, and no B/B survive. For 100 expected zygotes, that is 5 unmarked m/m adults and 50 marked m/B adults. The unmarked fraction among adults is 5 divided by 55, or about 9.1 percent.
If a researcher was expecting zero unmarked adults, this model offers one explanation that does not require loss of m. But the model is not itself evidence that survival occurred. Check the genotype of the unmarked class, the viability assumptions and the scoring conditions. A different mechanism could produce a similar adult ratio.
Now change the marker rather than viability. Suppose only 90 percent of the 50 heterozygotes show a detectable marker, with the two homozygotes fully lethal. This yields an expected 45 marked and 5 unmarked adults. The observed ratio resembles the first scenario, but the unmarked adults in this case are m/B. The correct follow-up depends on what you are trying to discriminate.
These teaching numbers show why a ratio alone cannot always settle the mechanism. The adult counts might be measured well, yet several models may fit approximately. Genotype the relevant classes or choose a cross with a distinct prediction. Better evidence is not always a bigger sample of the same outcome; sometimes it is a measurement of a different quantity.
CHAPTER 11 OF 12 · Transfer and document
11. Transfer the check to more than one chromosome
A stock can carry markers and intended alleles on more than one chromosome. Do not let one visible marker stand in for the entire genotype. Make a small ledger with a row for each chromosome and separate fields for the allele, the balancer or alternative homologue, the visible readout and the evidence for identity. If a marker is ambiguous, leave it ambiguous in the ledger.
Suppose a fictional stock needs m on chromosome 2 and n on chromosome 3. A clearly scored marker for the chromosome 2 balancer does not identify n. An assay for n in a pool does not say which individuals also carried m. If the result needs both alleles in the same individual, use an observation that can establish the joint state in that individual.
Do not assume that multiplying two simple probabilities always predicts the adult counts. The chromosomes may assort independently under the chosen model, but adult classes can still be filtered by genotype-specific survival, marker scoring and sample selection. Distinguish the inheritance model from the observation process, and change one assumption at a time when checking your calculation.
The same lesson transfers beyond flies. A linked marker in another organism can be useful without being a direct assay of the target. The organism, chromosome arrangement, viability and recombination mechanism may differ, so the numerical Drosophila examples do not travel unchanged. What travels is the discipline of asking what the proxy actually detects and what remains to be verified.
CHAPTER 12 OF 12 · Transfer and document
12. Leave a stock record another reader can reconstruct
A good handoff states the intended genotype, chromosome and allele identifiers, provenance, generation or date, and marker-scoring conditions. Attach the most recent evidence relevant to the use: observed cross classes, direct identity checks, and any relevant limitations of those tests. Write which conclusions still rely on assumptions rather than verification.
Record notable anomalies and how they were resolved. If the stock was refreshed from a verified source, preserve the relationship between old and new material rather than silently merging their histories. A paper, screen or lesson that depends on a particular stock should be able to say which material was used at the time, not merely which label is on the current vial.
For a final check, ask a colleague to read just the record and sketch the expected marked and unmarked classes under the stated assumptions. Ask what would still look normal if m were lost. If the answer is “every marker-positive adult would look the same”, identify the independent evidence that carries the missing information.
The useful habit is small and repeatable: use the marker to sort, use the cross to test transmission or survival, and use the appropriate independent check to verify the claim about identity. You do not need every possible assay for every question. You do need enough evidence that a familiar label or a tidy ratio cannot quietly stand in for the genotype you meant to study.
Independent worked case and evidence checks
Two fictional parent vials look identical: every sampled adult shows the dominant balancer marker. In vial A the genotype is m/B. In vial C it is +/B. Assume full marker expression, equal transmission and adult lethality of B/B and m/m. In a within-vial cross, A yields marked surviving m/B; C yields both marked +/B and unmarked +/+.
If 60 adults are scored from vial C under the ideal model, the expected classes are 40 marked and 20 unmarked. This is a ratio prediction conditional on 60 survivors, not a claim about how many eggs were laid. If the observed classes are 54 and 6, compare sampling, scoring, viability and genotype before selecting a cause.
Now change the task: you need to decide whether a particular marked individual carries m. The vial ratio can raise or lower confidence in the stock, but it does not directly identify the allele in that individual. Select an appropriate individual-level assay or informative progeny test and state what it can establish.
Worked comparison
| Fictional scenario | Expected marked adults | Expected unmarked adults |
|---|---|---|
| m/B × m/B; both homozygotes lethal | m/B only | None expected |
| +/B × +/B; B/B lethal | +/B | +/+ |
| m/B × m/B; some m/m survive | m/B | Surviving m/m |
| m/B × m/B; incomplete marker detection | Detectably marked m/B | Some m/B can appear unmarked |
Questions and practical checks
Does a dominant balancer marker prove the target allele is present?
No. It is evidence about the marker phenotype and may support balancer carriage. The intended allele on the other homologue needs separate evidence.
Are balancers always lethal when homozygous?
No universal rule applies. Many balancers have recessive lethal or sterile effects, but check the identified chromosome and stock documentation.
Does suppressed recombination mean exchange is impossible?
No. The effects depend on the rearrangements and region, and rare exchanges and stock differences have been documented.
Why can a healthy balanced stock show no unmarked adults?
In the teaching model, m/m and B/B are lost before adult scoring. The surviving m/B carry the dominant marker. The prediction changes when viability or scoring assumptions change.
Is a pooled positive assay sufficient to verify individual genotype?
No. It can show that some sampled material contained the target. It does not show which individuals carried it.
What should I record when I cannot resolve an unexpected class?
Record the observation, scoring conditions, alternatives and what is not established. Preserve the distinction between an anomaly and a diagnosed cause.
Continue with the connected subjects
Follow inheritance and recombination
Chromosomal inversions and missing recombinant classes
Use flanking observations in three-point mapping
Separate transmission from survival before interpreting a ratio
Check evidence in a genetic screen
Suppressor screens, allele classes and missed outcomes
Conditional alleles and interpreting lethality in genetic screens
Sources and scope
Miller and colleagues: rare recombination events and sequence differences among Drosophila balancers
Miller and colleagues: molecular and genetic characterisation of second-chromosome balancers
Sequencing and characterisation of balancer breakpoints and mutations
The numerical cases in this guide are original fictional teaching examples, not measurements from a real site. This is an educational explanation, not a field safety instruction, engineering approval or an official certification. For a real project, use qualified practitioners, current product documentation and site-specific evidence. Sources were checked on 5 November 2026.