If I connect two antennas to one radio using a t connector will it work?
That depends on your t-connector.
If your tee really is just a branch in the inner conductor and connected outer conducters, than the other two answers are correct: your input impedance will be different than your output impedance.
If, however, the tee is meant to be used in a system with constant impedance, and most are, then it's going to be built in a way that matches all inputs.
In your specific case, there's good reason to simply implement such a splitter yourself, as you want to have control over the splitting/combination.
The typical way to do that here is the Wilkinson Power Divider in coaxial implementation:
With that, not only do you split the power half and half (if you transmit into P1), or combine with equal gain (if you receive on P2 and P3), but you also isolate P2 and P3, meaning that no power from P2 ends up on P3 and vice versa.
I was just assuming you're using 50Ω line; if you're not, that $\frac\lambda4$ piece needs to be $\sqrt2$ of your system's impedance.
71Ω might be hard to get, 75 Ω usually easy; the mismatch should be minimal.
If a connect 2 antennas (a mag mount + a dipole?) to the same radio with an adapter will it work normally on 2M band
What's "normally"?
I guess no, this is not going to work like you expect: you combine/split equally. That means you've just built an antenna array out of two completely different antennas. That array will have a directive gain depending on the phase, efficiency and relative position of the two antennas that will be practically impossible to predict without extensive simulation.
and will it cause extra interference?
Interference will be picked up as before, but you have a directional antenna now. So, you might be getting more or less interference, depending on where your sources of interference are relative to the antenna system pattern described above.
Also, if it works would this give me an added range
Only if the above directional gain points in the direction you want to work
/better signal
same
over using just one of the antennas?
What you want to have seems to be diversity gain. ON5MF's answer actually already describes one diversity reception method:
Selection combining, where you simply switch to the one antenna that has better SNR.
Assuming you place the two antennas far enough so that their reception paths are uncorrelated, you'd get a diversity gain of 50% (assuming independent, Rayleigh-distributed paths). That means, having the second antenna to select from gives you, on average, an SNR that you work with that's 50% better than if you had nothing to choose from.
If you'd want to have real combining gain, you'd need two independent receivers. Ideally, you'd do optimum combining (since you mention the presence of interference), but to limit complexity here, you'd normally just start with two synchronous receivers, and add up their signals, weighted with a factor that has an amplitude that reflects the SNR of that single receiver, and a phase correcting the receive phase. That would give you an SNR that is the sum of the two received SNRs.
This second solution, however, would require a dual-channel synchronous SDR solution, and I'd guess you don't have that lying around, so:
As ON5MF suggested, get a switch. A good idea would be to monitor the "unused" channel with a cheap RTL-SDR to notice if the signal strength or interference level suddenly changes and switching might make sense again.