(I don't know anything about chemistry in high schools outside Ukraine, so maybe this is redundant): it often seemed to me, in school and after (and still after, reading upon mass-spectrometry analysis) that much of the problems we had to solve could be easily moved to arithmetics curriculum in primary classes. We studied inorganic chem for 2 years (I think), then 2 years of organic chem. Many times we had to calculate mass of a given compound, or concentrations of solutions, or other stuff like that which was necessary, of course, but still seemed misplaced and too simple. And yet we made mistakes and forgot numbers and so on. I mean, can't people studying multiplication already be given problems like 'Carbon atom weighs 12 units, oxygen atom weighs 16 units. How much does a molecule having 1C and 2O weigh?' instead of only stories about boys buying apples and cakes? By the time they get to the explanation why the molecule consists of exactly such combinations of elements, they will have some idea of how heavy it is. And it will free time for actual chemistry. I remember being very surprised to learn that there are classes of compounds built on Phosphorus-Nitrogen bonds (the problem of what valence is) and of alloys where ions of two metals form separate grids (I used to think that atoms' radii should prohibit that one).

Mainly I object to school-induced view of atoms and groups of atoms like curried functions (SO4, depending on context, means 'add 2', 'minus 2', ...), mostly because in MS analysis these functions aren't reliable any more:))