What it gets us is the ability to increment the RA by a fixed amount over all six non-look-up rows. (Here, we are using Δt = t – t_1984, signed.) Its value, for each n is equivalent to:

–24n/T ≈ –0.065710n

which amounts to a decrement of 3m 56.56s for each unit n.

Let’s take our derived sequence, complete with its three wildcards, as:

B ? T P H ? A I C ? G B I O C I

and adjust the n used in its construction from 1 to whatever value is required to get our required J at the start.

The initial B came from Δ_1 in the first row as:

Δ_1 = 44 – 17 = 27 ≡ B

We want this to give J, which requires us to have T_1 = 52, so that:

Δ_1 = 52 – 17 = 35 ≡ J

So, what value of n will give us this?

As it happens, the possible values for n form a sequence.

Anywhere from RA_1968 = 17h 52m 00s and 17h 52m 59s will get us Δ_1 = J.

For our control table, using n = 1, we derived RA_1968 as:

17h 44m 41s

to which we add 3m 56.56s to arrive at the deduced value corresponding to n = 0, to give:

17h 48m 37.56s

This, in turn equates to the range of n values capable of producing Δ_1 = J being:

–0.855782 ⇆ –1.105195

The first gives us the following when applied to the base table:

This gives us our ‘J’ at the beginning, as had been our intention, but it had also changed the deltas in rows 2 to 5 and 8 in a way that was beyond our control.

We are not able to do any better by selecting a different n to make any of the other deltas match since the resulting changes are not independent across our rows, and we have no control at all over those for 1983 and 1984.

This, then, exhausts our attempts to exploit even the permissive ‘freedoms’ granted by the astrological model. There are none remaining, and we have failed to impose the SLoI. Indeed, we haven’t even come remotely close.