What goes wrong

It did not arrive intact

On this page11

The seal failed at altitude The absorbent was decorative The secondary was a bag, not a containment The outer was not rigid The parcel was too small to mark The classification was chosen for convenience The person packing had not been trained The parts came from three suppliers The laboratory rejected it anyway The fix is to buy the system and train the bench References

The absorbent has to take the entire contents. A pad in the bag is not a calculation.

The seal failed at altitude

A closure that holds on a bench and not against a 95 kPa differential. The parcel flew, the primary leaked, and the containment moved to whatever was outside it.

The absorbent was decorative

One pad, chosen because a pad is what goes in. The requirement is enough to absorb the entire contents of the primary receptacles, which with several primaries in one parcel is a sum nobody did.

The secondary was a bag, not a containment

A zip bag is a secondary packaging only if it is leakproof. A bag that holds a tube and vents at the seam has not contained anything.

The outer was not rigid

A padded envelope is not a rigid outer packaging, and the drop test is performed on the outer. This is the substitution that happens when somebody is shipping one sample and the kit feels excessive.

The parcel was too small to mark

At least one surface of the outer must be 100 mm by 100 mm. A small box cannot carry the UN3373 mark legibly, and a parcel nobody can identify is a parcel nobody can handle correctly.

The classification was chosen for convenience

The exception requires routine testing, unrelated to infectious disease diagnosis, and a low probability of infection. All three, assessed, not assumed because the packaging is easier.

The person packing had not been trained

Which is where almost all of this actually fails. The classification is made correctly upstream and applied by somebody at a bench at the end of a shift.

The parts came from three suppliers

And were assembled into something that had never been tested as a system. The test result belongs to the assembly, not to the components.

The laboratory rejected it anyway

Wrong tube, no label, insufficient volume, broken seal. More samples are lost to the laboratory's acceptance criteria than to any regulator.

The fix is to buy the system and train the bench

A certified packaging from a supplier who sells it as one, and twenty minutes with whoever packs it. Neither of those is a container decision, and both are cheaper than a repeat collection.

References

49 CFR 173.199, Category B infectious substances. "A Category B infectious substance must be packaged in a triple packaging consisting of a primary receptacle, a secondary packaging, and a rigid outer packaging." Absorbent material "must be placed between the primary receptacle and secondary packaging" and "must be of sufficient quantity to absorb the entire contents of the primary receptacles". "The primary receptacle or the secondary packaging must be capable of withstanding without leakage an internal pressure producing a pressure differential of not less than 95 kPa." "The packaging must be capable of successfully passing the drop test in 178.609(d) of this subchapter at a drop height of at least 1.2 meters (3.9 feet)." "At least one surface of the outer packaging must have a minimum dimension of 100 mm by 100 mm (3.9 inches)." By aircraft, each primary receptacle may not exceed 1 L and each outer packaging 4 L for liquids, or 4 kg for solids. law.cornell.edu

49 CFR 173.134(a)(1)(ii) defines Category B as "an infectious substance that is not in a form generally capable of causing permanent disability or life-threatening or fatal disease in otherwise healthy humans or animals when exposure to it occurs", to be described as "Biological substance, Category B" and assigned identification number UN3373. Paragraph (b)(11) excepts human or animal samples transported for routine testing unrelated to the diagnosis of infectious disease where there is a low probability the sample is infectious. law.cornell.edu

Written from 49 CFR 173.199 and 173.134, and from the catalog as it stood on 19 October 2026. Nothing here is offered as a compliant primary receptacle for a regulated shipment: the pressure and drop requirements attach to a tested assembly, not to a bottle. Category A infectious substances are a different packaging instruction entirely and are not summarized. Sterility is not claimed for any product, and what a laboratory will accept is the laboratory's to say.