Culling Rabbits and Rabbitry Records: Deciding Which Animals Stay

In a rabbitry, culling means removing an animal from the breeding herd — by selling it, by rehoming it, or by slaughtering it. The word names the decision, not the method: a doe who leaves in a buyer's carrier has been culled just as surely as one who does not leave alive.

The distinction matters because readers arrive at the word carrying different meanings, and that confusion makes people avoid a routine task. In FAO's model weekly work plan for a rational rabbitry, "culling sick and unproductive females" sits on two fixed days of the week, beside filling in doe cards and weaning litters. It is scheduled bookkeeping, not a crisis.

Where the route out is slaughter, it should follow recognised humane-handling guidance rather than improvisation: the AVMA publishes separate guidelines for euthanasia and for humane slaughter, the latter with a dedicated rabbit chapter. The UK welfare code for farmed rabbits states plainly that rabbits killed on the farm must be killed humanely, and that ailing or injured animals should be segregated and treated or killed humanely without delay. This article is about the decision that comes before that, not the method.

• Sale — as proven breeding stock, to another breeder or through a marketplace
• Rehoming — as a pet, where temperament and health allow
• Slaughter — for meat or on welfare grounds, under the humane rules that apply locally
• Not culling — leaving an unproductive animal in a cage a productive one could fill

One figure sits above the rest: young weaned per doe per year. Litters per year, kits born alive, kits weaned, weaning weight, pre-weaning mortality and the interval between kindlings are all inputs to it, and a doe can fail on one input and still be worth her cage. The benchmarks below come from 100 rational French production units monitored for technical management, and they are useful because they report the spread, not only the average.

Each line below reads as general average, then the best nine units, then the poorest nine. A doe near the general average stays. A doe landing in the poorest column on two or more metrics across three consecutive litters is not a marginal call — she is consuming a cage, feed and labour a replacement would turn into weaned kits.

Two figures deserve extra care. Pre-weaning mortality is the one breeders most often write off as luck: FAO records birth-to-weaning losses near 20 percent in European rabbitries and notes that getting below 12 percent is very difficult, so a doe at 15 percent is not your problem and a doe losing a third of every litter is. Weaning weight is also a doe trait — eight kits weaned at 0.45 kg (1 lb) is less saleable rabbit than six at 0.7 kg (1.5 lb), and only a weighed record shows it.

• Kindlings per mating — 67 percent; 81 percent best, 49 percent poorest
• Interval between kindlings — 53 days; 42 best, 75 poorest
• Live births per kindling — 7.9; 8.9 best, 6.6 poorest
• Kits weaned per weaned litter — 7.1; 8.1 best, 6.0 poorest
• Litters not weaned at all — 10 percent; 3 percent best, 20 percent poorest
• Birth-to-weaning mortality — 17.6 percent; 10.4 percent best, 28 percent poorest
• Young weaned per doe per year — 47; 61 best, 29 poorest
• Weaning — 29 days at 0.603 kg (1.3 lb) liveweight

Doe fertility follows a curve steady enough to plan around. In a long-running commercial dataset from Spain and Portugal, apparent fertility in first-parity rebred does was 69.8 percent, rising to 75.6 percent at the second parity, 78 percent at the third and 80 percent at the fourth, then declining from roughly the sixth onward — 75 percent by the eighth insemination and 73 percent by the ninth. Early misses are normal; late misses are a trend.

That supports two defensible rules. Give a young doe about two litters to prove her mothering instinct and litter size, because first litters are unrepresentative and first-litter does are often simply nervous. Then treat repeated failure as a decision rather than bad luck: a herd at or below a 70 percent pregnancy rate counts as apparently infertile against a farm average nearer 85 percent, and one study of 20,000 does culled on two negative services and still averaged a 5.7-service career.

To rank does against each other rather than against a threshold, FAO uses a simple productivity index: after the second litter, take her average young weaned per litter and divide by the days between her first and latest kindling, then compare only does with the same number of litters. Performance across the first three kindlings correlates closely with lifetime output, so a fourth and fifth chance rarely changes the answer.

• Two consecutive failed services — replace rather than try a third time
• Weaned-kit count below your floor across three consecutive litters — replace
• Repeated failure to raise a litter, or cannibalism recurring after the first — replace
• Poor body condition or recurring illness such as mastitis — veterinary assessment before any breeding decision
• Age alone — median doe career is five parities, mean six, so treat age as a prompt to read the numbers, not a verdict

A buck is judged entirely on other animals' records, which is why he needs a card of his own. FAO's buck card carries the date of each service, the number of does served, the outcome of each pregnancy check and the number of live and stillborn young — yielding the two things you want: how many of his services produced a gestation, and how prolific the does he served turned out to be.

Rule out overwork first, because an overworked buck looks infertile. FAO puts one buck to seven or eight does under intensive management and one to ten or fifteen under extensive management, used no more than three or four days a week and two or three times a day. Michigan State Extension gives 1:10 to 1:15 for commercial herds, notes small producers may need 1:5, and observes that mature bucks can be used daily for single matings over long periods without losing fertility. Keep at least two bucks even in a ten-doe herd.

The other half of the evaluation is growth in his offspring: weaning weight and date, then the number and weight of his progeny still present at day 70, which is FAO's second selection point. Rabbits from rational systems reach market at about 70–90 days and 2.3–2.5 kg (5.1–5.5 lb) in strains such as the New Zealand White and the Californian, so a buck whose does conceive readily but whose kits are consistently light at day 70 is doing half his job.

Roughly the whole doe herd, every year. The 100 monitored French units averaged a female-stock culling rate of 112 percent per year, and the nine most productive units ran higher at 126 percent — better herds cull harder, not less. FAO notes monthly culling rates of 8–10 percent are not uncommon under intensive reproduction, and a large commercial dataset puts total monthly removal at about 3 percent mortality plus 7 percent culling.

Small-herd guidance arrives at the same place by another route: keep replacements coming at one young doe a month for every twelve working does, and the same for bucks — with the explicit warning that this is only a guide, and culling usually has to be heavier during a rabbitry's first two or three years.

Replacement is not free, and planning it late costs litters. FAO models a backyard system where a 100 percent annual renewal rate drops the doe average from six litters a year to about 5.5, simply because replacing an animal takes time. A rabbitry therefore needs a standing reserve of young does already at mating weight, and the most practical source is the best young out of your best does — with bucks brought in from another breeder to avoid inbreeding.

• 12 working does — about 12 replacements a year, roughly one selected a month
• 50 working does — about 55–60 replacements a year, plus one or two reserve bucks
• Expect a higher number in your first two or three years, while the stock is still being improved

Three habits do most of the work. Write the thresholds down before you need them, so the decision is made once, calmly, and applied many times. Put the review on the calendar instead of waiting for a bad week — FAO's model plan schedules culling of sick and unproductive females on fixed days, doe and buck cards on others, and selection of future breeding animals twice, at weaning and again at 70 days. And compare like with like: the index is only compared across does with the same number of litters.

Fix your selection intensity in advance too. FAO's mass-selection scheme keeps daughters from roughly the best 25 percent of does, judged on the average of at least three successive kindlings, because litter size is weakly heritable and one good litter proves very little. A herd where every doe gets one more chance has no selection pressure at all.

This is where written records beat memory outright. RabbitBreeder logs mortality and incidents against the individual animal, so a doe's lost litters and health events sit beside her weaning counts at review time — which makes the decision a reading exercise rather than an argument about what you think happened last spring.

FAO's irreducible minimum for a working nursery is a daily record book holding four things: mating dates for each doe with the sire's number, so buck fertility can be checked; the outcome of each palpation; litter size at birth; and litter size at weaning. Everything beyond that is refinement. Add expenditure and every sale as soon as you can.

The refined version is a set of cards, each covering a different unit of the operation, plus permanent identification on the animal and a card at the hutch carrying health and breeding information. Cage cards win on capture: they live at the point of work, and FAO treats the mother-cage as the management unit of the rabbitry. Cage-level identification is enough where you keep records but do not breed selectively; individual identification becomes necessary the moment you intend to select.

A central register wins on comparison, and the failure mode of cards alone shows why. Producers commonly cull does when the record card is full, at somewhere between eleven and fourteen services — the stationery, not the animal, ends the career. A ledger or database has no such ceiling and can rank every doe against every other on the same index. Run both, with a fixed day each week for transcribing one into the other, because recording data is the task breeders postpone first.

• Doe card — service dates, buck's number, palpation result, kindling date, parity, live and stillborn young, weaning date, number weaned, weaned litter weight
• Buck card — service dates, does served, palpation outcomes, live and stillborn young
• Litter card — weaning weight and date, weight and date at sale or slaughter, individual identification of each rabbit
• Genealogical card — the animal's number plus its sire and dam numbers, the raw material of a pedigree

The single most common failure in rabbitry record keeping is the grow-out group: animals destined for sale and animals destined for the breeding herd get mixed up, constantly. One litter can split three ways at weaning — future replacements, animals for the meat or pet market, and animals not yet decided — and weeks later nobody can say reliably which was which. The breeding herd is small and individually named; the grow-out group is large, uniform and changes weekly.

The fix has two halves, and the first is permanent identification applied early. FAO identifies young does selected for replacement at weaning, then runs a second selection of future breeding animals at 70 days, with every rabbit tattooed or ear-tagged at weaning as it leaves the dam. Tattooing is the only genuinely reliable method — rings and ear clips are easily lost — and the UK welfare code likewise prefers tattooing to tagging, which risks damaging the ear.

The second half is grouping. In larger units FAO makes the batch — all rabbits weaned in the same week — the core reference point for every technical figure, with the fattening record listing first and last fattening days per cage, any mortality with its apparent cause, and liveweight at sale. Scale explains why this matters: Mississippi State Extension budgets about seven grow-out cages for every ten working does, so most animals in a rabbitry at any moment sit in the group that is hardest to track. RabbitBreeder mirrors that structure by keeping young stock in their own groups, separate from the breeding herd, with an animal moving across only when you promote it — and listing the ones that are leaving in its marketplace, so the sale stays attached to the animal record.

In the European Union, keeping records is not a matter of taste. Article 102 of Regulation (EU) 2016/429 — the Animal Health Law — requires operators of registered establishments to record the species, categories and number of kept terrestrial animals on the holding; movements in and out, with dates and places of origin or destination; the documents accompanying those movements; mortality on the establishment; biosecurity measures, surveillance, treatments and test results; and the results of required animal health visits.

Three details matter in practice. Records may be kept on paper or electronically, so software satisfies the requirement. They must be immediately available to the competent authority on request. And they must be retained for a minimum period set by that authority, which cannot be less than three years. Mortality is named explicitly, making an incident log a legal record rather than a nicety — though member states may exempt low-risk establishments from parts of it.

Elsewhere the obligation often shifts onto the buyer's expectations instead. AVMA's humane slaughter guidelines note that in the United States rabbits raised for food and fur fall under neither the Humane Methods of Livestock Slaughter Act nor the Animal Welfare Act, that federal inspection of rabbit meat is voluntary, that individual states may impose their own rabbit-specific inspection requirements, and that meat rabbits not voluntarily USDA-inspected come under FDA oversight for interstate commerce. Where the law asks little, the records that matter are the ones a buyer or inspector asks to see.