RASC Review here: https://www.rasc.
JHA review (scanned here): http://cdsads.u-strasbg.fr/full/seri/JH ... 1.000.html
What impresses me the most about Stewart, so far, is his dedication to (forgive the pun) time-tested methods. Electronics was young back in 1920. The
It was in this same period of the early 20th century that radio - which Stewart advanced - replaced the telegraph. It was critical for survey crews in the huge Dominion who could now receive accurate astronomical fixes from their actual work locations, rather than some distant point with a telegraph line.
For all of those reasons and more I found it interesting that Stewart published a graphical method for obit determination of binary stars in Proceedings of the American Astronomical Society, 1927. (Harvard Astrophysical Data System here (not secure): http://cdsads.u-strasbg.
Graphical methods in mechanics were typically taught to draftsmen (draughtspersons) who as skilled trades workers were not expected to know the mathematics of vectors. It was also true that these compass-and-straightedge methods would allow an engineer or scientist to get a first-order approximation, similar to the 3-digit estimates of a sliderule before plunging into the more exacting arithmetic.
It speaks to Stewart's range of interests. He led the most modern efforts of his day to create and improve an accurate celestial timekeeping system for the Dominion. He also remained aligned on the methods that were known to work.