Actually, from the article:

> Experiments showed that not to be the case. Instead, trials with healthy human subjects showed that the difference was due to the amount of gas in a given fecal sample.

A book I read as a child (ca. 1990) claimed it was gas content; good to know it was correct (even if by accident).

Who would have said that it was all the Archimedes Principle.

I’m impressed that you read the whole thing. I came away with fat content, as gas permeates away.

The article is pretty short and it mainly says that it is due gas content. The amount of gas seems to be dependent on the presence of certain gut bacteria.

According to the study, naturally about half of the population of mice produce floaters. When you collect stool samples of that half and inject it on the other half, they also start producing floaters. On the other hand, when you sterelize the gut of the floater group, they start producing feces that sink.

Fat content seems not to be related.

nope, it's fart content

There needs to be another "fart content" dimension to the Bristol Stool Scale.

https://en.wikipedia.org/wiki/Bristol_stool_scale