A lower bound, stated as one
On this page11
What we publish
Height and diameter, on 147 of the 163 containers in this group. Both fields together give the cylindrical side wall.
The computation
Pi times diameter times height. For an 88.7 mL jar 2.875 inches across and 1.335 tall, that is 12.06 square inches.
What the regulation measures
Total surface area available to bear labeling, which is a broader measure. A lid face, a shoulder and a base can all carry labeling.
So our figure is always smaller
Which makes it a lower bound rather than an estimate, and a lower bound is a usable thing as long as it is called one.
It answers one direction cleanly
If the side wall alone is over 40 square inches, the container is over 40. 56 of our 147 are in that position and the answer is settled.
And it leaves the other direction open
A container at 35 square inches on the side wall may be over 40 once the rest is counted. 91 of 147 sit in that uncertain position.
Which is why the figure is published as a bound
An estimate of the regulation's own measure would need assumptions about which surfaces are available, and those are properties of your label and not of our container.
The same discipline as the topicals property
There we publish the character count and the panel dimensions and refuse to compute a fit, because character width is a typeface property. Here we publish a bound and say it is one.
And 16 containers cannot be computed at all
Height is recorded on 152 and diameter on 147, so 16 of the 163 lack one or both. For those the question cannot be answered from our data.
A blank field is a reason to ask
Not a reason to assume. We do not fill dimensions in with a guess.