HI WA, that was exactly what i was thinking and you can even see my previous post.... i was talking about some parallel combinations... I think what you say can take us in the right direction...
When the diode is forward biased... and there is no load... the whole current has to flow through the voltmeter as there is no other path and also the diode's forward resistance comes in parallel with the high resistance of a voltmeter... Now when the diode is reverse biased, we are taking the reading just after the diode and the ground....Now the reverse resistance of the diode is very high, much higher than the voltmeter... and the reading now changes...
Until here, i could guess... If this is right...can someone summarize the whole thing please....