A cell count is an infection reading, not a milk-quality score. Here is how to read one for a single cow and for the tank.
Somatic cells are mostly white blood cells that the udder sends to fight infection, plus a small number of shed lining cells. A healthy uninfected quarter sits low, often under 100,000 cells per millilitre. When the count climbs, it is because the immune system is responding to something — which is why a cell count is best read as an infection signal rather than a hygiene score.
Above roughly 200,000 cells per millilitre, a cow is generally considered to have a subclinical infection: no clots, no swelling, no visible sign, but an infected quarter that is producing less than it should and shedding bacteria at milking. This is the number worth acting on, because subclinical cases are both the most common and the easiest to miss.
Bulk tank limits are legal, not advisory. In the European Union raw cow milk must hold a rolling geometric average at or below 400,000 cells per millilitre. Under the United States Pasteurized Milk Ordinance the Grade A limit is 750,000. Many buyers pay a quality premium well below whichever limit applies, so the commercially useful target is usually a good deal lower than the legal one.
Cell counts are awkward to average because the scale is multiplicative: the step from 50,000 to 100,000 is the same biological step as 400,000 to 800,000. Linear score fixes that by taking the logarithm, using LS = log₂(SCC ÷ 100,000) + 3. Each whole point is a doubling, scores average sensibly, and milk loss rises roughly evenly with the score rather than with the raw count.
| Linear score | Somatic cell count (cells/mL) | What it usually means |
|---|---|---|
| 2 | 100,000 | Healthy, uninfected |
| 4 | 400,000 | Watch — early or minor infection |
| 5 | 800,000 | Subclinical infection likely |
| 7 | 3,200,000 | Established infection, real milk loss |
No. A high count means an udder is fighting infection, and the milk from that cow yields less cheese and keeps less well, but it is not in itself a safety question the way an antibiotic residue is. Safety and cell count are separate tests with separate limits, and passing one says nothing about the other.
Easily, in a small herd. The tank is a volume-weighted average, so a high-yielding cow at two million cells contributes far more than a low-yielding one at the same count. Finding the few cows driving a tank result is usually faster than changing anything herd-wide, which is what individual cow testing is for.
Monthly individual testing is the common rhythm, because it catches a rising cow before she becomes a clinical case and gives you a trend rather than a snapshot. Between tests, the California Mastitis Test at the cow's side costs almost nothing and answers the same question for a single quarter in under a minute.
DocDairy works these dates out for every cow the moment you record a service, and puts the ones due today on one printable list. Ten languages, offline in the barn.
Try the live demo See pricing