I agree with SFHAZE, that @75 degrees F you really can't go wrong. A 20 degree thermal gradient from top to bottom is a little extreme, especially since most of the active growth zone for the roots will be towards the outside and bottom of the pots. What I do to minimize the temperature differential top to bottom is this. If you're using a bottom heating pad and you have the room vertically, you can put the pad in the bottom of a tray, [cat boxes work great for small grows, otherwise some sort of plastic tub that would fit your grow with a depth of at least 2/3 of the height of the growing pot] put a few inches of redwood compost or other suitable material on top of the pad, then put your grow pots on top of this and then fill the tray up to about 2/3 of the top of the grow pots with your suitable material. What I used to do was instead of a bottom heating pad, use an agricultural grade thermocoupled heating cable. This has two advantages: 1) you can bury the cable midway up the pots with minimal thermal differential top to bottom as the heat flow both directions rather than just up from the bottom. 2) you get more precise control of temps provided you use the correct thermocouple. I like 75 degrees F. The only problem is that I have been having a hard time findings this type of cable lately. If anyone out there knows where you can get this item, please let me know. This system has to be tweeked and tuned, but once you get it set it is almost Zero Mantainence. To SS's point about strains ' That strain makes a difference seems very logical to me, and the strain I'm growing tolerates a wide range of temperatures.' This is the single most important variable in this equation given the fact that we have every thing from Matanuska strains growing in the proximity of the arctic circle to equitorial strains like The Dagga from Malawi. I think that a clinical trial using a single variable on identical clones under equal conditions would help answer this question for a single strain. You could have an unheated control and another[others] with root heating. SS, you could also test the heated room hypothesis with budding plants in the same way, i.e., a norml temp grow room [say 75 D] and a variable grow at say 85D, all other parameters being equal. I agree with you that I would rather use science that anecdotal evidence, but in the absence of that, anecdotal evidence is better than none at all, especially when coming from reputable growers.