A litter is the smallest unit a breeder can learn anything from, and still far too small to prove anything — which is why it has to be read as a pattern rather than as a result.
The most important first
- Evaluate the litter as one unit: how uniform it was, what recurred in several puppies, and how much it varied, rather than collecting anecdotes about individuals.
- In a longitudinal study of Border collies, litter membership affected every component of the seven-week behavior test, and only one of roughly ten puppy traits still predicted adult behavior at 18 to 24 months.
- Assessors can agree closely on what they see in a puppy and still be wrong about what it becomes: inter-rater agreement in one seven-week test reached 0.76 to 0.89, which says nothing about predictive accuracy.
- Follow every puppy, including the ones that were never shown, never worked and never mentioned again; an evaluation built on the successful half is not an evaluation.
- Your eight-week notes are the comparison material, so they have to exist before you need them. See observing and documenting puppies and which puppy to keep.
A litter is a sample, not a verdict
Everything that follows rests on one uncomfortable fact: a litter of six puppies from two parents is a single draw from a very large set of possible outcomes. The same pair mated again would give you six different puppies, and some of what you see in front of you is the result of which sperm met which egg rather than anything about the combination. Evaluating a litter therefore means describing a sample honestly, not declaring a verdict on the parents.
That does not make the exercise pointless. It makes it a different exercise. What a litter can tell you reliably is what happened: how the puppies grew, how uniform they were, what conditions appeared, how they behaved in the environment you provided, and how far the result matched what you planned. What it cannot tell you is whether a difference between this litter and your last one means anything. Both statements are true at once, and holding them together is the whole skill.
The practical consequence is that you should write down patterns and decisions, not conclusions dressed up as findings. “Four of six puppies were slow to recover from the vacuum cleaner, and two still were at twelve weeks” is a pattern. “This combination produces sound nerves” is a conclusion no single litter can support.
Variation within the litter is the data
The most useful thing in a litter evaluation is usually not the average but the spread. A litter of six that was uniform in size, growth, structure and temperament tells you something different from a litter of six where two puppies were half a kilo behind at eight weeks and one was markedly bolder than the rest, even if the averages match.
Some of that variation has known causes, and they are worth separating from the rest before you interpret anything. Birth weight falls as litter size rises: in a dataset of 1,013 Labrador litters, each additional puppy was associated with about 11 grams less weight at birth, and the dam’s own genetic contribution to birth weight was as large as the puppies’ own, with a maternal additive heritability of 0.22. These figures come from a single guide-dog breeding colony, and the authors caution that they may not apply to Labradors in general. Parity matters too; in the same dataset birth weight rose to around the second or third litter and then declined. A smaller, lighter litter from a maiden bitch in a large litter year is not evidence of anything about the sire.
What remains after the known causes are accounted for is the part worth your attention, and the honest description of it is comparative: which puppies sat at the edges of the litter, in what, and whether they stayed there. The growth records described in weighing and tracking growth make this visible in a way memory never does.
Variation is information, but a puppy that falls behind early is first a health question. Weigh every puppy daily for the first two weeks. A healthy puppy should reach at least 1.5 times its birth weight by day 7 and about double it by day 10–14. A puppy whose gain slows or stops, or who will not nurse, needs your veterinarian promptly, not a note in the evaluation. Record the outcome either way. It belongs in the litter’s history.
What your eight-week notes can carry forward
The hardest finding to accept in this area is how little early assessment predicts. In a longitudinal study of Border collies, 99 puppies from 18 litters were tested as neonates, 134 were tested at 40 to 50 days, and 50 were retested as adults at 18 to 24 months. The authors found little correspondence between the three stages. Of roughly ten predicted associations between puppy and adult traits, only one survived correction for multiple comparisons: puppies that explored less were somewhat less active and exploratory as adults. Nothing measured in the first ten days of life predicted anything at all once the statistics were done properly.
It is tempting to read that as a reason to stop assessing puppies, which would be the wrong conclusion. The same study found that litter membership affected every component of the seven-week test, while adult results showed no significant litter effect after correction. In other words, what you can measure at seven weeks is substantially a property of the litter and its environment during those weeks — which is precisely the thing your evaluation is about. Your notes are good evidence about the litter you raised. They are weak evidence about the adults those puppies became.
Agreement between observers does not rescue this. A study of a seven-week test with three independent video assessors reported agreement statistics of 0.76 to 0.89 across all five subtests (a stricter measure gave 0.44 to 0.70), which shows the test can be scored consistently. Consistency and predictive validity are separate properties, and the limits set out in the predictive validity of puppy tests apply to your own observations as much as to any published protocol.
Follow the whole litter, not the successful ones
The single most common flaw in litter evaluations is that they are built on the puppies who stayed visible. The one in a competition home sends results; the one in a pet home two hours away sends a photograph at Christmas; the one who was returned at eighteen months and rehomed quietly tends to vanish from the account altogether. An evaluation assembled from that material will always look better than the litter was.
Decide at the outset that the unit of evaluation is every puppy born, including those that died. If seven were born and five are traceable at three years, write down that the figure is five of seven and what you know about the other two, rather than describing the litter on the basis of five. The gap is part of the result, and it limits what you can claim.
Pet puppies matter as much as the ones that stayed in breeding homes. For conditions such as hip and elbow dysplasia, how a dog’s littermates turn out says a lot about the genes it carries, and littermates’ breeding values only diverge once their own results come in. Offering pet owners an easy route to the breed’s recommended screening is how a litter’s evaluation becomes complete.
This is also where the uncomfortable symmetry appears. A litter that turned out badly in several homes may be telling you about the combination, about the dam’s maternal behavior, about the way the litter was reared as described in raising a healthy litter, or about how you matched puppies to homes. Those explanations are not distinguishable from one litter, and the one you reach for first is usually the one that costs you least. The decisions recorded in placing each puppy are part of what is under evaluation here.
Reading the litter against the plan you made
A litter evaluation is only interpretable against the intention it started from. If the plan was to improve hind angulation without losing substance, the question is whether the litter did that, in how many puppies, and at what cost elsewhere. Without a written plan you are left comparing the litter against your memory of what you hoped, which adjusts itself to the outcome.
Work through the plan point by point and mark each one achieved, partly achieved, not achieved, or not assessable yet. That last category is larger than most breeders expect: structure at eight weeks is a weak guide to the adult, and the recognized limits in evaluating puppy conformation mean that several questions simply cannot be answered until the puppies are two or three. A litter evaluation done once, at eight weeks, is an evaluation of a plan’s early stages and nothing more.
Then record the things that were never on the plan but happened anyway: the cesarean, the one puppy with a heart murmur that resolved, the two weeks when the dam’s milk dropped. Those become context the next time you read these notes, and they often explain more than the intended traits do.
Writing it down so it is still useful in five years
An evaluation you cannot find is an evaluation you did not do. Our recommendation is to revisit the same litter at least three times — at eight weeks, at around a year, and at three to five years — with each round added to the same document rather than replacing it, so you can see how your own early judgment held up. Researchers who followed puppies into adulthood recommend repeated behavior assessments from about six weeks to a year and a half, and most hip and elbow schemes give results between 12 and 24 months, which is why the second pass falls around a year.
Keep the fields compatible with your ordinary records, so a litter evaluation can be read alongside the whelping log, the weight charts and the health results without translation. The structures described in record-keeping and documentation for breeders are built for this, and a litter evaluation that uses different headings from your kennel records will quietly stop being filled in. Finish each round with one sentence naming what this litter changes in your plans, even when the answer is that it changes nothing.
In practice
A breeder reviews a litter of seven whelped four years earlier. Her eight-week notes record two puppies she considered outstanding, three she thought ordinary, and two she had worried about for soft reactions to noise. At four years, five of the seven are traceable. One of the two she rated highest has a title and a good hip result; the other was neutered early after an orthopedic injury unrelated to breeding. Of the three she called ordinary, two are working steadily in active homes and one has a recurrent skin problem. Of the two she worried about, one lives quietly and reportedly startles easily; the other was placed in a busy household and her owner describes no such problem. She records that noise sensitivity appeared in two of seven and persisted in one, that her own ranking at eight weeks predicted little, that two puppies are untraceable, and that the skin problem is worth watching in the dam’s next litter. She does not conclude anything about the combination from seven puppies.
(Illustrative, composite scenario.)
When to call your vet
- A puppy is not gaining weight steadily, has not reached about 1.5 times its birth weight by day 7, or loses weight.
- A puppy will not nurse, cries persistently, feels cold, or is much weaker than its littermates.
- A puppy has diarrhea, vomiting, a swollen or painful belly, or labored breathing.
- One or more puppies in the litter lag clearly behind the others in size or development at any age.
- An adult offspring develops lameness, an eye problem, or another sign that could be inherited. The owner should see their vet, and you should ask to hear the result.
- If in doubt, call — your vet would rather hear from you early.
Work on it
A few minutes here will teach you more than re-reading the article.
Ask yourself
- For my most recent litter, how many puppies can I still describe accurately, and how many have effectively disappeared from my account of it?
- Did I write down my assessment of each puppy at eight weeks in a form I could now compare against what they became?
- When I describe that litter to someone else, am I describing all of it or the best of it?
Find out & check
- Find out what became of the one puppy in your last litter you have heard least about, and record the answer whether or not it is flattering.
Do one thing
- Take one past litter, list every puppy born including any that died, and write one line per puppy covering health, temperament and what it is doing now.
Was this article helpful?
Anonymous. We do not store your name, email or IP address.
Sources & further reading
- Riemer S., Müller C., Virányi Z., Huber L. & Range F. — The predictive value of early behavioural assessments in pet dogs: a longitudinal study from neonates to adults. journals.plos.org
- Alberghina D., Giunta F., Gioè M. & Panzera M. — Behaviour test for seven-week old puppies: inter-rater reliability and factors associated with test performance. iris.unime.it
- Schelling C., Gaillard C., Russenberger J., Moseley L. & Dolf G. — Heritabilities for the puppy weight at birth in Labrador retrievers. pmc.ncbi.nlm.nih.gov
- Fédération Cynologique Internationale — FCI international breeding strategies. fci.be
- Merck Veterinary Manual — Management of the Neonate in Dogs and Cats. merckvetmanual.com
- Jerold S. Bell, DVM (Tufts, via VIN) — Breeding Strategies for the Management of Genetic Disorders. vin.com
- Wilson et al. (2013), PLOS ONE — Estimated Breeding Values for Canine Hip Dysplasia Radiographic Traits in a Cohort of Australian German Shepherd Dogs. pmc.ncbi.nlm.nih.gov
Where evidence is limited, recommendations rest on established practice and expert judgment, and may be updated as knowledge develops. A single litter is far too small a sample for statistical inference of any kind, so the purpose of a litter evaluation is to describe patterns and record decisions, not to establish that a difference is real.
Next steps
- Read next: Evaluating the combination.
- Related: Which puppy to keep.
- Return to the area hub: Closing the loop: follow-up and the next generation.