The effect of temperature:

Although there is a temperature coefficient to solubility, this is not generally the worse problem that is encountered when crystallizations are set up using the vapour diffusion methods. The main problem are temperature fluctuations. This is because vapour diffusion provides a controlled but relatively slow method of equilibration by the transfer of vapour between the protein and precipitant solution in the reservoir.
How to recognize the effect of temperature fluctuations
One of the most common effect of temperature fluctuations is incosistent drop size. Hanging drops in cold rooms are particularly likely to display this problem. The drop are generally much bigger than the initial drop. The drops at the tray edges, which are more susceptible to temperature fluctuations display this problem more than the drops in the middle wells of the tray. This is because the the tempearture of the drop which in hanging drops setups is in thermal contact with the ambient and has little thermal capacity, changes rapidly, while that of the reservoir changes more slowly. During the temperature cycles there is a tendency to have condensation at the highest point in the setup (because moist warm air is lighter than dry air). The drop provides a sink for this moisture and will tend to grow more than it will shrink during cycles in which the temperature changes in the opposite direction. In the hanging drop setup, the protein drop is in the middle of the crystallization enclosure and hence it is much less affected.
Reducing the problem
The problem will be reduced by changing to a sitting drop setup. The plastic bridges supplied by Hampton Research are a first step. The glass pedestals are much better as they provide conduction between the reservoir and the drop. These are more expensive. If cost is an issue, you can make your own using pots available from Fisher cat#08-733 (ref 20.).


Mounting Crystals grown in the cold for use at room temperature.

The the influence of the temperature coefficient of solubity is likely to be felt more on crystal mounting that for growing crystals, as when growing crystals we can compensate for this by increasing or lowering the precipitant concentration accordingly.
Problems you may have encontered
Crystals dissolve when the tray is taken to room temperature
The crystals must be mounted in the cold, unless before mounting the crystals extra precipitant is added to the reservoir and allowed to equilibrate for 2 days or more to compensate for the difference in solubility at 4C compared to room temperature.
The crystals can be mounted in capillaries and diffract but later dissolve
This is likely to occur when the crystals are mounted in the cold and later brought to room temperature. It is caused by distillation of water from the mother liquor generally added to each end on the capillary. Don't use mother liquor! Remove as much solution from around the crystal as possible, without ever touching the crystals as this may dehydrate the lattice and cause loss of resolution. Add paraffin oil to each end of the capillary instead of mother liquor. This is a good method even for crystals not grown in the cold.
Cryocrystallography
The use of cryocrystallography should encounter the standard problems associated with the method. If you know of special problems because the crystals were grown at 4C instead of 20C send me e-mail.