Einstein introduced the gravitational time dilation in 1911, in an attempt to explain the fact that the frequency as measured at different gravitational potentials is different when identical clocks are used:

http://www.relativitybook.com/resour...n_gravity.html
Einstein 1911: "Let the two material systems S1 and S2, provided with instruments of measurement, be situated on the z-axis of K at the distance h from each other, so that the gravitation potential in S2 is greater than that in S1... (...) It follows, then, that a ray of light emitted in S2 with a definite gravitational potential, and possessing at its emission the frequency v2 – compared with a clock in S2 – will, at its arrival in S1, possess a different frequency v1 – measured by an identical clock in S1. (...) v1=v2(1+phi/c^2) (...) If there is constant transmission of light from S2 to S1, how can any other number of periods per second arrive in S1 than is emitted in S2 ? But the answer is simple. We cannot regard v2 or respectively v1 simply as frequencies (as the number of periods per second) since we have not yet determined the time in system K. What v2 denotes is the number of periods with reference to the time-unit of the clock U in S2, while v1 denotes the number of periods per second with reference to the identical clock in S1. Nothing compels us to assume that the clocks U in different gravitation potentials must be regarded as going at the same rate."

On close inspection it becomes clear that this protective belt (gravitational time dilation) is quite inefficient. In the absence of additional ad hoc hypotheses, the interpretation based on gravitational time dilation implies shifts in both the MEASURED frequency and the MEASURED speed of light in S1. That is, v1=v2(1+phi/c^2) implies c1=c2(1+phi/c^2). Clever Einsteinians know that and sometimes ignore, even dismiss, gravitational time dilation when discussing the gravitational redshift:

http://www.amazon.com/Relativity-Its.../dp/0486406768
Banesh Hoffmann: "In an accelerated sky laboratory, and therefore also in the corresponding earth laboratory, the frequence of arrival of light pulses is lower than the ticking rate of the upper clocks even though all the clocks go at the same rate. (...) As a result the experimenter at the ceiling of the sky laboratory will see with his own eyes that the floor clock is going at a slower rate than the ceiling clock - even though, as I have stressed, both are going at the same rate. (...) The gravitational red shift does not arise from changes in the intrinsic rates of clocks. It arises from what befalls light signals as they traverse space and time in the presence of gravitation."

There is an alternative to the protective belt based on gravitational time dilation which is sillier but more efficient: The speed of light is constant, Divine Einstein, yes we all believe in relativity, relativity, relativity, so if the frequency changes, the wavelength MUST change accordingly:

http://www.astronomynotes.com/relativity/s4.htm
"Prediction: light escaping from a large mass should lose energy - THE WAVELENGTH MUST INCREASE SINCE THE SPEED OF LIGHT IS CONSTANT."

This protective belt (the wavelength always changes so that the speed of light can remain constant, Divine Einstein, yes we all believe in relativity, relativity, relativity) is quite universal in Divine Albert's world:

http://www.pitt.edu/~jdnorton/teachi...ved/index.html
John Norton: "Here's a light wave and an observer. If the observer were to hurry towards the source of the light, the observer would now pass wavecrests more frequently than the resting observer. That would mean that moving observer would find the frequency of the light to have increased (AND CORRESPONDINGLY FOR THE WAVELENGTH - THE DISTANCE BETWEEN CRESTS - TO HAVE DECREASED)."

http://www.lp2i-poitiers.fr/doc/aps/...oppleffet.html
"The observer moves closer to the source. The wave received has a shorter wavelength (higher frequency) than that emitted by the source. The observer moves away from the source. The wave received has a longer wavelength (lower frequency) than that emitted by the source."

http://highered.mcgraw-hill.com/olcw...oppler_Nav.swf
"EXERCISES: 2. Now click on the "Observer Approaches" button. The ship will start flying towards the source. What is the wavelength of the waves now, as the ship approaches the source? Does the frequency increase or decrease? SOLUTIONS: 2. The wavelength shrinks so that about three waves now fit within the graph. (...) The frequency increases."

Pentcho Valev