I took a look at the midpoint between Dorset and Hanbury on my Jack the Ripper - significant points map, and it didn't look like it was a midpoint. So, I checked it with MB Ruler, and found that it doesn't satisfy my criteria for a midpoint, i.e. half of the way between the points on a line drawn between them. I performed the calculation again, and came up with the same results, and I checked the positions of all of the points, and they were all plotted accurately. So, I concluded that the formula which I used for calculating the midpoints (an output of the same formula used for calculating bearings) is incorrect, and that a simple average would be appropriate for such short distances. The result was 51.519563, -0.073828, and when I plotted it, it ended up precisely on the line between Dorset and Hanbury, and halfway between them, meaning that the problem was indeed the previously-used formula. I drew a precise line between this new point and the Henriques Street site (bottom of the pentagram) and noted that it runs almost precisely through the Emma Smith attack location.
This naturally shook my faith in the accuracy of the bearing-calculations, although I have been comparing the results with the results obtained from MB Ruler, and I've been assuming that any discrepancies are due to the physical limitations in making the measurements. I still assume that to be the case, because of the nature of such calculations. They don't get so close, and then develop some error - they keep converging on a more precise result.
This also had an effect on the center point of the pentagram. For the time being, I used Google Earth to draw precise lines between the four points which define the pentagram's center, and arrived at 51.516687°, -0.069764°, and the calculated bearing of this point from the center of St. Paul's dome is 9.046, which corresponds to a sunrise date of August 30th, 1888.