I read Rick's answer and he is correct; however, it didn't include some of the whys...
When measuring a DC voltage, best practice is to start on a voltage scale higher than the voltage to be measured. In this experiment for your meter, that would be the 20v scale. I would expect that you will see a reading less than 2 volts for a single cell battery (the maximum for your next lower scale), at which point you can safely switch down to the 2v scale because you have confirmed that it is safe (for your meter) to do so. If you then see a consistent reading above 0.05v and below 0.2v, you could then switch down to the 200mv scale to get a more accurate reading, but USE CAUTION:
If the reading is jumping around or staying less than 0.05v on the 2v scale, you probably have a connection issue, either with your test leads or with the electrodes in your fruit/vegatables. This means that switching down to the 200mv scale is a bad idea. As you play with trying to correct the connections, you could easily go way above 200mv and cause the DVM to go into a self preservation mode. If that happens, you will have to read the user manual to see what it says about its over voltage protection and what steps you have to take if any to reset your meter.
With old analog DC volt meters, the danger was "pegging the meter", driving the meter needle into a mechanical stop and permanently bending the needle. With DVM's, the danger is driving the internal circuitry into saturation. Most meters have some protection against permanent damage; however, if the overvoltage is too high, some integrated circuits can have their substrate reverse biased and stop working. Sometimes you can disconnect the power (internal 9v battery) and short the test leads and leave it that way for a few minutes to several hours to restore its functionality.