With 18ah per battery and small loads, I don't think Peukert is going to have much applicability in this situation. It mostly needs to be considered for systems large enough were it can make the difference in hours or days of extra run-time, or where it can mean the difference between a small depth of discharge or a large depth of discharge, or in systems where every watt counts.
There's theory, and then there's the real world.
I gotta go with Tennmogger on this one.
But then, I'm a well-known battery abuser. And I don't have solar on my truck at all - I prefer a tiny generator, a battery charger and parking in the shade.
My only nit to pick in Tennmogger's post is the bit about one battery discharging the other.
Well yea - but - all else being equal (make/model/age/type/size AND properly balanced wiring) then the discharge from one to the other is going to be small. Just a little until they are truly equal - but it won't be much since they'll have been charged and discharged together.
So yea, the slightly better battery will do slightly more work than the other battery and wear out slightly sooner. But that process will increase its internal resistance, and then it'll be pretty much the same as the other battery - resistantially speaking (yes, I just made that word up ).
With large banks and with everything else being equal, it's just something that has to be lived with.
Especially if you've got a large bank made up of 2v L-16s:
Current: 76 E-250, bubble-top, self-contained|couple of old Yamaha enduros
Previous wheelers: 41 Willys|78 FJ40|78 Bronco|84 Bronco|74 Ramcharger|78 Ramcharger|79 D150 PowerWagon|77 D100|79 D400 dually, converted to 4WD, utility bed, 10' Lance|75 Westy|69 Scout, RHD|bunch of others|bunch of bikes|couple of boats|couple of motorhomes|blah blah|so what|not my idea|just doin' what I'm told|wank wank|this space for rent|candy is dandy|but liquor is quicker