The inbreeding coefficient, written F, is the probability that the two copies of a gene at a random locus in a cat are identical by descent, meaning both came from the same ancestor. Full siblings mated together give F = 25%. Half siblings give 12.5%. First cousins give 6.25%. For a working cattery the ceiling should be 10% per generation, and closer to 6% if the line already has a fertility or heart problem in it.
What F measures, and what it does not
F is a statement about ancestry, not about health. A 25% kitten is not automatically sick; it is a kitten in which a quarter of its gene pairs are expected to be identical, so any recessive fault carried by the shared ancestor has a much better chance of meeting itself. That is the whole mechanism. F does not tell you which faults, and it does not replace a DNA panel or an echo. Two cats with the same F can differ completely in outcome, because one line carries a clean set of recessives and the other carries a lethal one.
The coefficient is also an average across the genome. A cat can be 8% overall and still be highly homozygous across the region around a major gene, which is why linebreeding to one outstanding cat sometimes produces both the best and the worst kitten in the litter.
F values for the matings breeders actually consider
| Mating | F of the kitten | Practical reading |
|---|---|---|
| Parent to offspring | 25% | Not a breeding strategy; use only as a one-off test mating with a plan to outcross |
| Full siblings | 25% | Same genetic cost as parent-offspring, rarely justified |
| Half siblings | 12.5% | Above the working ceiling; acceptable only if both lines are known clean |
| Grandparent to grandkitten | 12.5% | Common in older stud books, watch litter size closely |
| Uncle or aunt to niece or nephew | 12.5% | Typical linebreeding, one generation only |
| First cousins | 6.25% | Comfortable for most breeds, sustainable across several generations |
| Second cousins | 3.1% | Effectively an outcross for practical purposes |
| Unrelated cats, no common ancestor in 5 generations | 0% | The reference point; gives you room to linebreed later |
Why 10% is the working ceiling
Fertility and litter size fall slowly and then suddenly. Catteries that stay near 6–8% per generation usually hold litter size steady for eight or ten generations. Catteries that run 20–25% see the drop within two or three. The 10% figure is not a law of nature; it is the level at which most breeders can still select for type without paying for it in kittens, and it leaves headroom for the accidental repeat mating that every cattery has once.
Reading a five-generation pedigree
A five-generation pedigree lists 31 ancestors: 2 parents, 4 grandparents, 8 great-grandparents, 16 great-great-grandparents. Print it, then highlight every name that appears more than once. One name appearing on both the sire's and dam's side is the ancestor doing the damage, and its position tells you how strong the pull is. A dog or cat repeated in the third and fourth generation is mild; the same animal repeated as a grandparent on both sides is where F climbs fast.
Do not count names by eye and guess. Most registry software and the free online COI calculators will compute F from a five-generation file, and the number they return is the one to record, because hand-counting misses paths through two or three common ancestors at once. If you keep paper, use the path method: for each common ancestor, count the generations up from the sire, the generations down to the dam, and add them; each path contributes (0.5) to the power of that sum, adjusted for the ancestor's own F.
What inbreeding depression looks like in the nursery
The signs are boring and cumulative, which is why they get missed. Track four numbers per litter and per queen:
- Litter size at birth. A line sliding from 4.5 to 3.5 over four generations is telling you something the pedigree already knows.
- Birth weight. Mean 90–110 g is normal for most breeds; a line averaging 80 g with more kittens under 75 g is under strain.
- Deaths before 3 weeks. Any repeat of fading kitten syndrome in the same line deserves a look at F before it gets a look at infectious disease.
- Conception rate and interval. Queens that need two or three matings to conceive, or that slip litters, are the first place inbreeding depression shows.
Add sperm quality in the stud: volume, motility and percentage of normal heads fall earlier than libido, so a tom that still mates eagerly can already be a fertility problem. None of these markers is specific. Their value is in the trend, and the trend only exists if you wrote the numbers down before you needed them.
Where a line has a known cardiac problem, pair the pedigree work with the screening described in heart murmur and HCM screening, because a tight line plus a dominant heart condition is the combination that ends catteries. Plan the outcross at least one cycle ahead of the mating date, using the cat pregnancy timeline to work back from a due date, and keep the coefficient definitions in the glossary handy when you explain a pedigree to a buyer.
Rules that keep a closed line alive
- Cap F at 10% per generation, 6% if the line has any known defect.
- Outcross at least every third generation, and outcross to a cat with a lower F than your average, not a higher one.
- Keep two parallel sub-lines so a single loss does not end the programme.
- Record F on every litter sheet next to litter size and birth weight, so the trend is visible without a spreadsheet.
- Freeze semen from your best low-F males; it buys you an outcross years later without importing.
Frequently asked
Is a 25% inbred cat automatically unhealthy?
No. F=25% means about a quarter of the cat's gene pairs are expected to be identical by descent, which raises the chance that a recessive fault from the shared ancestor meets a copy of itself. Whether that happens depends entirely on what the ancestor carried, so the number sets risk, not diagnosis.
How many generations should I count?
Five is the registry standard and the figure most software reports. It captures 31 ancestors, which is enough to see the repeating names that drive the coefficient, while going deeper adds little because distant ancestors contribute halved influence each generation and are often unknown in older pedigrees.
Can I linebreed and still stay under 10%?
Yes, if you do it in one generation and then outcross. Uncle-to-niece or grandparent-to-grandkitten matings sit at 12.5%, so a single such litter is already above the ceiling; use it once for a specific trait, then bring in an unrelated cat with a lower coefficient before repeating anything.