Tweeter size: larger vs. smaller radiating area

I'm in the process of selecting drivers for a floorstanding 3-way speaker with an 8" bass, 4" mid and some tweeter. As for the tweeter I'm unsure what to use. I have two candidates in particular, because I already have them on hand. It's the Peerless XT25BG60 (1" ring radiator) and the SB Acoustics SB19ST (3/4" dome). My mids (Dayton RS125) are quite well behaved as far as cone breakup is concerned, so I don't have to cross too low.

In general smaller domes tend to have better off axis behavior, but need to be crossed somewhat higher to avoid excessive distortion. I haven't heard the SB19ST play yet, nor did I find many posts about it on various forums. Also, some favor larger tweeters over smaller ones, like Troels Gravesen. The XT25BG60 is a generally well regarded tweeter, yet also known for it's relatively bad off axis response. For aesthetic reasons I don't want to toe-in the speakers, even though my listening position is relatively close to the speakers (at an angle of about 30 degrees).

Which one would you use? Other recommendations are also welcome.
 
any idea for max SPL, directivity, tweeter high pass frequency and filter steepness?
I don't have a need for high SPL. Directivity is linked to off-axis response, and is somewhat important, as my listening position is about 30 degrees off axis. High pass frequency and slope are to be determined, based on measurements to be performed once I built the boxes and mounted the drivers.
 
as my listening position is about 30 degrees off axis
oh - did not see this in your initial post!
this will probably define the max. crossover frequency for midrange (2,5 kHz max?) and consequently the max. tweeter resonance frequency (depending on slope, about 1,25 kHz?).
both tweeters are ok with that but SB seems to have better off axis response!
 
My current desktops are in fact, RS125's and XT25G30's. The VIfa's are very well behaved, so much in fact the lack of bad anomalies makes some call them boring. But, yes, not quite as good off-axis as a dome. Oh, the G60 is slightly smoother, but really slightly. I have my '25'in a diagonal baffle so I sit on-axis but the box sits square.

Even though the 125 is by far the best behaved of the series, I would still go 4th order electrical. If you go shallower or lower, you need a notch for the breakup and it takes more parts and messes up the phase far more than just doing the L-R4 to start with. For the VIfa, going 4th, you can get away with as low as 2500 if you don't play too loud. If you do, best push it up 2800 or 3K. I see so many cross tweeters way to low and they get muddy highly distorted midrange when they crank it. I think I might consider the larger SB tweeters or some of the Seas in the similar price that you can cross down as low as 2K. Even easier on the Dayton.

FWIW, I have given up on metal cones and domes. The slight edge in in-band just does not overcome the breakup issues.

Yea, Allen, the smaller the dome, the higher the beaming frequency. Ring radiators are a little "beamier". The phase plug has something to do with it, but I am not sure what. Zaph or Elliot mentioned something about why.

PS: No, not sharing my crossover as I missed the phase a little and will be re-doing them. They looked good on paper, but I have a new measurement method that was informative. Just proved it on little FE85's with a 3/4 TB. Second order @ 3300. Beautiful phase match. Quite improved imaging especially in the vertical plane.
 
Thanks for all the insights guys. Didn't get around to reading the replies for a while due to home improvement related hassle. I decided I'm gonna try the 3/4" SB dome first, mounting the baffle in a non-permanent way. If measurements disappoint, I'll make another baffle with the 1" ring radiator and try again.