unfortunately it's more difficult than that
Okay, let's set things straight:
2 parallel mounted motors on ONE fixed axle.
If the overall resistance (POS-lead -motor- NEG-lead) for each motor is identical,
then they both will be powered with the same Voltage.
The current is limited by the controller
(well in fact it's not but it's pulsed to throttle it down.. nevermind... limited!

)
and each motor can draw a different amount of what the controller provides, based on it's load.
The motor that takes up more load will use up more current;
the other one though can rev up faster, up to the point where it has the exact same rpm..
at that point the two motors will try to level each other out to almost 50:50 and the same exact rpm.
Motor rpms are _NOT_ a question of voltage, but of Wattage (Volt * Ampere = Watt)
So while at first one tries to do all the work (because it can rev up more easily for whatever reason) the other one can rev up more easily (since it's under less load, thus is converting more energy into rotation instead of heat than the other)
And finally both run at about the same load (manufacturing discrepancies) at the exact same rpm (because of the axle/chain restriction )
So to spread the currents as evenly as possible you want similar motors; as similar as possible.
otherwise one might turn out to get much hotter than the other one
(hotter meaning more energy is provided and converted into heat, not rotation)
So now for the motors in series.
That's a very nasty thing to say the least.
Since each motor , wire, THING in the universe...
is also a resistor it get's very strange very fast.
each motor is a resistor to the other one,
so while both only can make use of
about half of the voltage applied to the system (the bigger the difference is in the motors, the bigger the difference in the voltage will be)
there's another factor to keep in mind
R = U/I (resitance is voltage divided by current)
which basically means the following:
if you send a known voltage through a known resistance all you can measure is a current;
and vice versa.
Put 24V into a 0.667Ohm system and you'll measure 16 Amps
Allow 16 Amps through a 0.65 Ohm Resistor and you'll measure 24.64 Volts
So if the internal resistance of a motor is different from the other motor and
since changing the load will change the internal resistance and the rpm
it's nearly impossible for the motors to level each other out.
Since everytime one motor takes up more load it's going to increase it's internal resistance due to the heat drawing even more currents, allowing less Voltage to reach the other one and finally it'll pull all the load and the other one just revs freely.
They still have the same rpm, but at a very different Powerconsumption.
You can't do anything against it, since adjusting one lead (to compensate the difference in resistance) it's always a resistor to BOTH motors!
And even if by chance both internal resistances are perfectly the same...
one ungreased roller in one of the chain is enough to allow one to ever so slightly draw more power.. and the problem begins.
'sid