Research, made readable

Why there’s no single normal birthweight for puppies

A number on the scale needs a little company: the breed, the litter, the puppy’s age—and a clear idea of what you’re comparing.

Good Harbor Pet · 10 October 2026

About 6 minutes

Checking your reading…

Explore the breed data

A weight belongs to one puppy, in one litter, at one moment.

If you’ve ever searched for a “normal puppy birthweight,” you’ve probably found a very tidy answer to a rather untidy question.

Dogs arrive in very different sizes. A useful comparison needs to say which puppies it describes. And even a carefully measured weight can’t, by itself, tell you whether an individual newborn is doing well.

This is a closer look at what published birthweight data can tell you—and what the numbers leave for your veterinarian to assess.

Start with the population

One species. Very different starting weights.

In a French study of 6,694 live-born puppies, breed medians ranged from 118 g for Chihuahuas to 640 g for Newfoundlands. Those are the middle values in the recorded groups, not suggested birthweights. Mugnier et al., 2019 ↗

Sorted breed birthweight chart: medians range from Chihuahua 118 grams to Newfoundland 640 grams. Each band shows the interquartile range, with puppy counts beside breed names. Full values follow in the data table. — Original plot from Mugnier et al. (2019), Table 1. Voluntary records from 75 French kennels, 1,202 litters and 27 pure breeds; births 1994–2017. Each n counts puppies. Bands show variation within the sample, not uncertainty around the median or a healthy range.

Birthweight summaries for all 27 breeds · Mugnier et al., Table 1
Breed	Puppies (n)	Q1 (g)	Median (g)	Q3 (g)
Chihuahua	157	102	118	133
Yorkshire Terrier	128	120	140	164.25
Pomeranian	117	130	150	166
Maltese	178	140.5	165	187
Shih Tzu	225	158	172	190
Bichon Frise	107	162	183	212.5
Dachshund	152	163.75	184	203.5
Coton de Tulear	159	165	190	210
Lhasa Apso	153	165	190	216
West Highland White Terrier	164	176.75	200	224.25
Jack Russell Terrier	122	185	205.5	225
Cavalier King Charles Spaniel	155	202	230	250
French Bulldog	111	210	240	266
Cocker Spaniel	477	223	270	310
Beagle	124	280	310	342.7
English Bulldog	123	262	310	370
Australian Shepherd	420	315	365.5	410
Golden Retriever	483	350	400	438.5
Labrador Retriever	1846	363.2	410	460
Rottweiler	111	370	410	447.5
Boxer	123	415	470	520
White Swiss Shepherd	114	420	472.5	520
Bernese Mountain dog	265	445	480	555
German Shepherd	197	436	500	570
Leonberger	216	450	520	581.25
Alaskan Malamute	104	500	580	631
Newfoundland	163	570	640	688

Two hands support a tiny pale Labrador puppy horizontally just above the blue bedding. — Small details matter. So does choosing the right comparison.

Look a little closer

“Giant breed” still isn’t a birthweight.

Adult-size labels are broad buckets. In this study’s giant group, the Rottweiler median was 410 g; the Newfoundland median was 640 g. The bands below also overlap. Sharing a category doesn’t make the distributions identical. Tables 1 & 2 ↗

You can use a breed label to ask a better question. You can’t use it to finish the assessment.

Rottweiler: 370, 410, 447.5 grams; Bernese Mountain dog: 445, 480, 555; Leonberger: 450, 520, 581.25; Newfoundland: 570, 640, 688. Values are Q1, median and Q3. — Original comparison from Mugnier et al., Table 1; “giant” follows the paper’s Table 2 classification (adult weight >45 kg). These four groups contain 755 puppies. They are the same data shown above, enlarged for comparison.

Four breeds classified as giant in this study
Breed	Puppies (n)	Q1 (g)	Median (g)	Q3 (g)
Rottweiler	111	370	410	447.5
Bernese Mountain dog	265	445	480	555
Leonberger	216	450	520	581.25
Newfoundland	163	570	640	688

A small reading skill, a big difference

The middle half isn’t the healthy half.

A median puts the ordered observations in the middle: half lie at or below it and half at or above it. The first and third quartiles—Q1 and Q3—mark the edges of the middle half. Together they describe a distribution. They don’t sort puppies into “fine” and “needs help.”

That distinction is easy to lose when a table gets passed around without its methods. Calling an interquartile band a “normal range” changes its meaning. It also hides the observations on either side.

Keep this beside a research chart

Three things a birthweight chart can mean

Lower
quarterMiddle
quarterMiddle
quarterUpper
quarter

Q1MedianQ3

A schematic of rank, not a weight axis. Tied values can sit on the boundaries.

Median: Where the middle observation sits. It isn’t a target for every puppy.
Interquartile range: Q1 to Q3: the middle half of this sample. It isn’t the smallest-to-largest span.
Clinical assessment: A different question: how is this puppy doing? A chart of quartiles can’t answer it.

The yellow Labrador mother rests with her five small closed-eye puppies gathered against her on the same blue fleece. — Look at the puppy and the mother as well as the number.

Put the evidence to work

Bring a better question, not a verdict.

Breed is part of the explanation, not the whole explanation. An Italian census of 3,293 puppies examined maternal size, body weight and body shape as well as litter size and sex. Birthweight was associated with several maternal characteristics. That doesn’t tell you why one particular puppy is small. Groppetti et al., 2017 ↗

Later French research followed both birthweight and growth during the first week. It used five separate growth intervals and survival to two months; its growth analyses didn’t all contain the same number of puppies. A starting weight and a later change are different observations. Keep the dates and age interval attached when you discuss them. Mugnier et al., 2023 ↗

When you talk to your veterinarian, bring the actual weight, unit, observation time, puppy identity and subsequent readings, along with what you’ve noticed about the puppy and mother. Ask which comparisons are appropriate for this litter and how the weight fits the wider assessment. If you’re concerned about a newborn, contact your veterinarian rather than waiting for a chart to settle the question.

The breeder sits beside the nursery reviewing handwritten notes, with a lined scale on a nearby low work surface. — Keep the reading and its context together.

Follow the evidence

Sources & a little more depth

The figures are original plots of published summaries. No individual observations have been reconstructed from the quartiles.

Mugnier et al. · 2019
Birth weight as a risk factor for neonatal mortality: Breed-specific approach to identify at-risk puppies. Preventive Veterinary Medicine 171, 104746. Table 1 supplies the plotted weights; Table 2 supplies size groups. Methods §§2.1–2.2 and Results §3.1 describe the population. Mortality was followed to Day 21; the quartile bands shown here are descriptive.
Groppetti et al. · 2017
A National Census of Birth Weight in Purebred Dogs in Italy. Animals 7, 43. Italian voluntary census, 99 breeds; useful context for maternal differences. Institutional abstract and indexed Methods consulted; no values from its tables are plotted here.
Mugnier et al. · 2023
Relative Impact of Birth Weight and Early Growth on Neonatal Mortality in Puppies. Animals 13(12), 1928. Methods §§2.1–2.3; Table 1. 8,550 puppies, 24 breeds, 127 French kennels. Voluntary records and variable available growth data limit generalization.
Read further · Chastant, 2026
Developmental programming in dogs. Biology of Reproduction 115, 379–387. A narrative review of early-life influences and remaining research questions; it does not validate these quartiles as clinical boundaries.