That’s what I was thinking; something to do with the temperature / density of the charge and the burn-rate. What plays against that is how much ‘swirl’ there would be in the charge especially considering how advanced the spark is timed; 45 BTDC seems quite advanced to me (but I suppose it isn’t really).

As you say this timing may just have been arrived at by experimentation or it may be a method of ensuring consistency; if the two sparks were synchronised any slight variation could mean inlet or exhaust sparking first. At least with a five degree difference the exhaust plug is always going to spark first?