Hard to say really. You would need to know the impedance graph ("nominal" values are just about worthless) along with the phase graph to make an accurate guess.
Also, in some cases, published sensitivity specs are far from the true value.
But assuming your numbers are hypothetical and the phase angles are similar for both, #1 would require a little less power than #2.
Arthur 
Also depends on the crossover. 1st order, 4th order? More crossover parts make it harder for low power amps to drive. But based on the info given, I agree with Aball. See what the lowest imedance is on the speakers. If one drops to 2ohms, it will be a challenge to drive. Speaker Sensitivity 
Assuming the numbers given are accurate, the 94 dBwatt 4 ohm nominal speaker is 5 dB more efficient than the 89 dBwatt 6 ohm speaker. In this case, the 4 ohm speaker will also have a higher 2.83 volt sensitivity, at 97 dB for a 2.83 volt input vs roughly 90.5 dB for a 2.83 volt input for the 6 ohm nominal speaker.
I have heard that the convention is to quote a "nominal impedance" value roughly 1.25 times the minimum impedance, rounded to the nearest ohm. There's no guarantee that a given manufacturer follows this convention, though.
Note that a few amplifiers might still be happier with the lower efficiency 6 ohm speaker.
Duke 
Thanks guys. I suspected the answer would not be easy.

Duke notes the convention is to quote a "nominal impedance" value roughly 1.25 times the minimum impedance Many manufacturers quote the impedance @ 1kHz instead. Not so illuminating (but things are getting better). On topic  No1 spkr of course, as all above note. 
Duke's qualifier is interesting: Note that a few amplifiers might still be happier with the lower efficiency 6 ohm speake. This is telling, and why I posted the question. The 94db speaker is the obvious choice, but the more difficult load of the 94db speakers kaes the answer not so cut and dried, IMO. 
The 94 db/w speaker is obviously more efficient, even when both are converted to a common basis for comparison (2.83 volt sensitivity measurement). For most solid state amps, it would be much easier to drive to high volume; the 4 ohm nominal impedance would not be a problem. But, with a tube amp, it is harder to say which would work better. Single ended triodes, in particular, tend to have higher output impedances which translates to less ability control the movement of the cone.
As a pure guess, provided that the 94 db speaker does not have a severe drop below 4 ohms, particularly at very energetic bass frequencies, and if the phase angle is not severe at a frequency where the impedance dips, the 94 db speaker would be easier to drive for most tube amps. 
The question of imedance drop is important. The 6 ohm speaker dips to 3 ohms. I don't know about the 4 ohm nominal speaker, which conceivably could dip lower than 3 ohms.
The idea is to use a PP triode tube amp with a power output of 50wpc or less.

If one were to have a tube amp with say an output impedence peak of 3.5 ohms at 200 hz and speakers which have an impedence droop to 3.5 ohms at 200 hz would the two cancel the potential negative effect of either, everything else being equal? 
Looking at this from the other end, how loudly do you like to listen to music? How large is your room? These answers may have more impact than going strictly by speaker specs.
I auditioned a pair of huge JM Lab Electra 946 with a 3.5W ASL 2A3 amp. It worked great as long as you didn't turn the volume up too much. As an owner of a McIntosh amp with power meters, I can tell you that most of the time, I use less than 5W of power. Maybe you are the same way? 
I sit 8 feet from the loudspeakers in a 19x16x8 foot room. I listen around 8990db, and occasionally crank to ~101db. Classical to jazz to heavy metal. At the loudest volume, I'm using around 32 watts. My present speakers are the 89db/1w/1m in the example above.
It seems 91db, 8 ohm speakers would be a more significant change from what I presently own in terms of efficiency and driving ease than would 94db, 4 ohm speakers. 
Depends on the amp also. Most solid state amps will double power output when going from 8 ohms to 4. While most tube amps won't increase output ratings at all. Try this chart also: @8 ohms for 89db speaker 1 watt = 89db output 2 watts = 92db 4 watts = 95db 8 watts = 98db 16 watts = 101db and so on..... every increase of 10db will give you a perceived doulbing of volume. Compare your amp rating for 4, 6 & 8ohms if you can. This will let you know what the actual output will be. 
Regardless of the speaker nominal impedance, impedance curves, or phase angles, what you will need at your listening position of 8 feet for an SPL level of 89 db is 90 watts for the 94 db/WM speakers and 285 watts for the 89 db/WM speakers. This is assuming 15 db peak headroom which is reasonable for classical and jazz.
But which speaker requires more power? Can't tell. If the 89 db speaker stays flat, then all it needs is 285 WPC. But if the 94 dB speaker drops down to 1ohm at some frequencies, then it may need 360 WPC or more to prevent clipping while delivering the same SPL as the 89 db speaker.
Nominal ratings are a guide, but minimum and maximum power requirements are spec'd out by the manufacturer for these reasons. 
Tvad wrote:
"It seems 91db, 8 ohm speakers would be a more significant change from what I presently own in terms of efficiency and driving ease than would 94db, 4 ohm speakers."
Efficiency no; driving ease probably.
I assume you're quoting the 1watt efficiency figures instead of "sensitivity" figures (sensitivity is referenced to 2.83 volts input; 2.83 volts into 8 ohms = 1 watt but 2.83 volts into 4 ohms = 2 watts. Sometimes manufacturers give the 2.83 volt sensitivity and call it "efficiency". Read closely to see if this is the case, especially with speakers whose nominal impedance is below 8 ohms). Assuming your numbers are efficiency and not sensitivity, then the 94 dB/watt, 4 ohm speaker would be 3 dB more efficient than the 91 dB/watt, 8 ohm speaker.
Now if the 4 ohm speaker manufacturer is actually quoting the 2.83 volt sensitivity instead of the 1 watt efficiency, then its actual 1 watt efficiency is only 91 dB, in which case both speakers have the same efficiency.
In either case, the 8 ohm speaker will most likely be the easier load. I say "most likely" because I sell a supposedly "8 ohm" speaker that's a more difficult load than most 4 ohm speakers due to the unusual nature of the load (it's a fullrange electrostat).
Now assuming that your tube amp is a transformer coupled pushpull type, as long as the manufacturer doesn't caution against 4 ohm loads you probably wouldn't have any problems. I would think that at a given SPL your amp would be less taxed by the 94 dB, 4 ohm speaker than by your present 89 dB, 6 ohm speaker. But I'm not sure about the comparison between the 94 dB, 4 ohm speaker and the 91 dB, 8 ohm speaker.
As you can see there are so many exceptions and caveats and whatif's that a generic answer probably won't suffice (see post by Gs5556 above as an example of how far off my generic answer could be). We might be better able to give useful information if you let us know the specific speakers and amp.
Duke 
050406: Gs5556 Regardless of the speaker nominal impedance, impedance curves, or phase angles, what you will need at your listening position of 8 feet for an SPL level of 89 db is 90 watts for the 94 db/WM speakers and 285 watts for the 89 db/WM speakers. This is assuming 15 db peak headroom which is reasonable for classical and jazz. I don't follow. I have measured the SPL from my listening position, and according to the chart Elevick provides, I am using 1 watt to produce 89db and 36 watts to produce 101db...far below your 90watt/285watt requirement. Furthermore, I reach 101db with the volume control on my preamp at about 2 o'clock. My speakers are 89db/1w/1m 6 ohm nominal. My tube amp is 110 wpc. 
Tvad, you have to take into account that as the distance from the speaker increases the SPL decreases. That one watt of power is at a distance of one meter, not the listening position (about 3 meters). Doubling the distance decreases SPL by 3db. Also, the headroom required to reproduce the type of music listened to is a factor. The SPL db at a given distance is (desired SPL  sensitivity) + 20 LOG(Distance to speaker/1 watt) + Headroom dB. The power required is 10^(SPL dB/10).
Your numbers are correct  but only at one meter with a 3 dB headroom. At 3 meters with 15 db headroom, it's 90/285.
The position of the volume pot is indicative of gain, not power. Power consumption is only done by the speaker as it draws current from the source amplifier which has a constant voltage, power supply permitting. The draw is determined by the impedance at any given frequency and the amp has to be rated for the power the speaker calls for. 
My real world experience does not agree with your numbers, Gs5556, because when I'm listening to Metallica measured at 101db (not at peak) at 8 feet from the drivers with an SPL meter, I have plenty of headroom remaining, and my PP tube amp is 110 wpc. Your numbers would indicate this is an impossibility, since according to your calculations, I require 285 watts to produce only 89db at 8 feet. Yet not only is it possible, it's real well beyond what your numbers tell you.
Perhaps it's the 15db headroom allowance that is the difference. I have never experienced a peak of 104db when the average listening level was 89db. I can say for certain that with an average reading of 101db on the SPL meter, peaks are generally no higher than 105db. With readings around 90db, peaks are no higher than about 95 or 96db. 
After reading Tvad's question in the post just above this one, I ran my own estimates of the power required at an 8 foot listening distance. Assumptions are a semireverberant room, and we're listening to stereo (both channels driven).
Anechoically, sound pressure level falls off by 6 dB for each doubling of distance from a point source. Eight feet is about 2 1/2 meters, so we could expect a falloff of about 8 dB relative to the 1 meter SPL. But we have two speakers, each driven by its own separate amplifier channel. Double the total cone area equals +3 dB, and double the total amplifier power equals another +3 dB, so we gain back 6 dB right there. And this is a semireverberant room, so we can expect a dB or two of room contribution. Let's be conservative and estimate 1 dB additional from room contribution at 8 feet (at greater distances, the relative room contribution would be more). Plugging in the numbers, we have 8 dB distance falloff, +6 dB for double the cone area and double the amp power, and +1 dB for reverberant field contribution, for a net change of 1 dB compared to the singlespeaker 1 meter SPL. So back at 8 feet, from a stereo pair of 89 dB/watt efficient speakers we could expect 88 dB SPL at the listening position for a 1 watt input. Similarly, with 94 dB efficient speakers, we could expect 93 dB at the listening position.
Elevick's chart is for a single speaker at one meter, but as you can see it's certainly in the ballpark for what you can expect from a stereo pair at your approximate listening distance. One thing the chart necessarily leaves out is the effects of power compression, but that's a whole nother can of worms and can't be generalized in a such a chart as it varies significantly from one speaker to another.
Now Gs5556 brings in the issue of peaktoaverage SPL, and he's right in estimating 15 dB of headroom being required to avoid clipping. On some recordings the peaks can be as much as 25 dB above the average level! Fortunately, the ear is relatively forgiving of modest amounts of clipping. In fact, many popular recordings have numerous "flat tops" (clipped waveforms) already. The extra power requirments figured into his numbers are to avoid clipping on instantaneous peaks, but if you don't have that much power available doesn't mean you can't still enjoy the music. You're just missing a little bit of the liveliness and sense of effortlessness that having no amplifierinduced clipping would give you.
At least, that's my take.
Tvad, the fact that you're running a tube amp helps a great deal if you're pushing the amp into clipping sometimes. Tube amps "soft clip" (relatively little high order harmonic distortion produced), while solid state amps "hard clip" (lots of high order harmonic distortion produced). The ear is vastly more sensitive to high order harmonic distortion than to low order harmonic distortion. You've perhaps heard something like "one tube watt equals two solid state watts"? Well, the reason has to do with the audibility of the type of distortion they produce when driven into clipping.
Duke 
Interesting. I'm awaiting delivery of a pair of 200wpc SS mono blocks, and it would seem they are 50% underpowered for my needs.
Also, I wonder if it would be beneficial to run Zero Autoformers on my loudspeakers to double the 6 ohm nominal impedance to 12 ohms? Oh, and the speakers have 89db/1w/1m SENSITIVITY not efficiency. Thanks for explaining that, Duke.

Tvad, well maybe and maybe not on that 200 watt amp. I'll admit that I probably overgeneralized in my last post, and that many exceptions exist. And to a certain extent it depends on how happy the amp is with the load the speakers present, and solid state amps tend to be happier with difficult loads.
Also, the shortterm power available for transients often exceeds the rated power by a considerable margin.
And from what I've been told by amplifier manufacturers (and supported by my own experience) there are different ways to arrive at the rated power figure  some optimistic, and some conservative. If that's 200 watts per channel reproducing broadband pink noise with both channels driven, it's an extremely powerful amplifier. If that's 200 watts per channel reproducing a sine wave a 1 kHz, that's not nearly as impressive and realworld it will probably clip long before the other amp.
Just as a matter of interesting trivia, FTC rules require stereo amplifier to be rated with both channels driven. But those same rules do not apply ot home theater amplifiers  the other channels only have to be on, not driven to the same power levels. This often results in highly inflated power specifications for home theater amps. Let me illustrate:
You walk into Bust Buy and there's the dimunitive Zony X1,000,00 home theater system on sale for $299, featuring an 800 watt 7channel amplifier. They achieve this amazing engineering feat by skimping on the power supply. You see, they only have to run one channel up to 115 watts with a 1 kHz sine wave, with the other channels barely on. They are then permitted to multiply that 115 watts by the number of channels to arrive at their rated power. And never mind that a 1 kHz sine wave is much easier to reproduce than a broadband music signal, and that the actual broadband power output before clipping may be as little as 1/8 that figure (this based on inhouse testing described to me by an amplifier manufacturer).
As an extreme exammple, a decent 10 watt per channel tube amp could conceivably play louder than the "800 watt" home theater amp before audible distortion sets in.
The point of that tangent was just to illustrate one of the ways that the published specs often don't tell the whole story when it comes to amplification. It's not that measurements are inherently untrustworthy, it's that inadequate measurements provide inadequte (and often misleading) information.
Duke 
Thanks, Duke. I've only shopped once at Bust Buy. Everything was paid in $1 bills.
I'll email you privately regarding the amps in question, if you don't mind.
Would it be beneficial to double the 6 ohm nominal impedance to 12 ohms via Zero Autormers, in your opinion? Or, is this unecessary if the amps have 4 and 8 ohm taps and/or high current delivery? 
Tvad, I think it's very unlikely the Zeros would help with a solid state amp. With a pushpull tube amp I'd guess somewhat unlikely but still possible. I have customers using Zeros with their pushpull tube amps, but the speaker load is an unusually difficult one (fullrange electrostats).
Feel free to email me. I've had intermittent email trouble lately so if you don't hear back from me in a reasonable amount of time try again or post a message here.
Duke 
Thanks, Duke. I've emailed with some system specifics. 