½ Half-Relationship cM Range Reference
Half relationships (half-sibling, half-aunt/uncle, half-cousin) share roughly half the DNA of their full counterparts, which means their cM ranges often overlap with the next more distant full relationship category. This chart shows common overlaps to watch for.
Why a half relationship lands on the next row down
Full relatives descend from a couple; half relatives from one shared person, so only one line carries DNA between them and the expected share halves. A generation of removal halves it too, which lands the two on the same figures:
half of X ≈ X once removed
half sibling ≈ aunt or uncle ~1,300 - 2,300 cM
half 1st cousin ≈ 1st cousin 1x removed ~200 - 850 cM
half 2nd cousin ≈ 2nd cousin 1x removed ~45 - 235 cM
Read the chart across, then down: each half range repeats as the full range on the next line. That is not a rounding artefact. The two relationships have the same expected share, near 25% for the first pair and near 1.5% for the last.
Separating relationships the total cannot
- Ages do most of the work. A 1,700 cM match twenty years older than you is far more likely a half aunt or uncle than a half sibling, though the cM total alone cannot say.
- Look at the segments, not just the sum. Closer relationships tend to show fewer, longer segments, and more distant ones more, shorter segments at the same total.
- Shared matches place the link. A half sibling shares matches on one parent's side only, usually the fastest way to tell that case from an aunt, uncle or grandparent.
Frequently asked questions
How much DNA do half siblings share?
Roughly 1,300 to 2,300 cM, centred near 1,700, about 25% of the autosomal genome. Full siblings sit higher, around 2,300 to 3,400 cM.
Can DNA tell a half sibling from an aunt or uncle?
Not from the shared total, since both occupy the same band. Ages, which side the shared matches fall on, and X-chromosome sharing are what settle it.
Could my half first cousin look like a second cousin?
Yes. Half first cousins run about 200 to 850 cM and second cousins about 90 to 480, so anything from 200 to 480 is consistent with either. A cM figure narrows the possibilities; it does not prove the relationship.