John Carlos BaezMar 6

Wow - two new mammal species have just been discovered! Even better, one is of an entirely new genus.

One is a striped possum with an extraordinarily long fourth finger - twice as long as the rest - that it uses to get wood-boring insect larvae to eat. It was known to have lived in West Papua until about 6,000 years ago... but it's actually still there!

The other is a ring-tailed glider: a marsupial that can soar downwards through the air. Fossils from this genus have been found in eastern Australia and New Guinea, but those are hundreds of thousands of years old!

Both these new species were found on Bird's Head Peninsula on West Papua. And both were found by the same team, who must be feeling ecstatic right now. Of course they're not really "new": they were just flying under the radar.

More details and pictures here:

https://www.theguardian.com/environment/2026/mar/05/marsupials-discovered-new-guinea

Show threadTony Vladusich

@johncarlosbaez

"We had thought we'd found them all."

How would we ever know this?

Mar 6 at 11:16pmWeb
Show threadJohn Carlos BaezMar 6

@TonyVladusich - I was just wrong about this, so I'll change my post.

We've only found about 6,758 living mammal species. But since 2005, there have been 805 newly described mammal species (and 774 split from known species, which is rather a different matter), which is about 40 per year.

Eventually this will slow to a trickle. This is quite unlike insects, where there's a vast number of species, and some estimate 80% of them are waiting to be catalogued.

https://www.mammaldiversity.org/

ASM Mammal Diversity Database

ASM Mammal Diversity Database

Show threadJohn Carlos BaezMar 6

@TonyVladusich - However, there's a fascinating thing called the Good-Turing estimator.

"Good–Turing frequency estimation is a statistical technique for estimating the probability of encountering an object of a hitherto unseen species, given a set of past observations of objects from different species."

https://en.wikipedia.org/wiki/Good%E2%80%93Turing_frequency_estimation

Good–Turing frequency estimation - Wikipedia

Show threadDougMerritt (log😅 = 💧log😄)Mar 6

@TonyVladusich
Very good question.

There *is* a methodology that plots discoveries plus some measure of the diffficulty of the discovery over time, looks at the curve, and guesses at the number yet undiscovered, and when done right, it has had some amount of confirmation as a decent approach at least some times.

In the absence of a mention of using that methodology, though, it sounds more like a vague feeling than something evidence-based.

@johncarlosbaez

Show threadDougMerritt (log😅 = 💧log😄)Mar 6

@TonyVladusich
P.S. amusing new profile pic; kudos to your son.

@johncarlosbaez