“applies” isn’t the word I would use. It’s not like nature has a line that once you pass some threshold of mass, acceleration or distance it needs to flip the relativity switch.

Probably say “becomes noticable”.

I’d actually argue the opposite. With states of matter, we’re attempting to delineate how reality groups together sets of related properties that vary between conditions in similar ways for different substances.
Looking for the edges that nature drew.

With species though, we drew the lines. We drew them with a mind towards ensuring it’s objectively measurable but it’s still not a natural delineation. Taxonomists (biologists are actually a different field) mostly run into uncertainty with debating which categorization property takes precedence, and what observations of species have actually been made.
So while they debate which system to use, the particulars of the systems are pretty concrete.

First you make them memorize single digit subtraction X - Y where X >= Y. Then you extend that to small double digit numbers.
Then you teach “borrowing”. 351-213. Subtract the 1s column. Can’t take 3 from 1, so borrow 10 from the 5 in the 10s column, making 11 in the 1s column and 4 in the 10s.

Definitely more clear, right?

I believe that’s what happens anytime they say that we probably shouldn’t focus on memorizing a multiplication table, or try to teach anything in a way that puts more focus on understanding how numbers work than on symbolic memorization.
And that’s like… Elementary school.