The first decade of the 20th century heralded the beginnings of the great photographic trigonometric parallax programs. These were the forerunners of the search for astrometric binaries, with selected stars covering some 5 years of observations at times of large parallax factors. The basic techniques remained essentially the same with some modifications with the increased awareness of systematic and random errors until the middle of the century. In the meantime new parallax series were added an din many cases a second series of previous determinations were made. The lengthened baseline from the two series for one star would greatly improve the relative proper motion and the potential of pointing to indications of variable proper motion.
When Peter van de Kamp came to the Sproul Observatory having been at the McCormick Observatory from 1925 to 1937, he built on the parallax program already in place since 1912, by including essentially all stars with parallaxes greater than 0.1 attainable from the latitude of 40*N. These nearby stars would include a majority of red dwarf stars. The main thrust of this program was not only to obtain the best possible parallaxes but to search for variable proper motion leading to the discovery of unseen companions to nearby stars. Discoveries of fainter companions to red dwarf would likely be, except for the occasional white dwarfs, even later spectral type and less massive objects; this would afford the extension of the lower end of the mass-luminosity relation.
The search for low-mass invisible companions to nearby stars was given impetus by the discovery in 1936 at the McCormick Observatory that the nearby dM5 star, Ross 614, exhibited variable proper motion which could be explained by Kepler motion. Further observations at the Sproul Observatory led to an orbit close to a 16 year period. In 1955 the separation of the two components of Ross 614 were advantageous and we asked Walter Baade to look at Ross 614 with the 5 meter Palomar reflector, he so did and saw the companion, and photographed the 14th magnitude companion. With a value for the separation and the difference in magnitude a scale of the AB orbit could be at least be provisionally given and thus the masses of the companions would follow. The resulting mass of .07 or .08 solar mass. This discovery gave encouragement for the possibility of other discoveries with continued extension of the existence of less massive objects. The 1960's was a great era of technical advances: telescopes, measuring devices, focal plane acquisition of stellar positions, etc.; all of these improvements aided the discovery of astrometric binaries. Thereafter, an increasing number of astrometric binaries were found many as members of M dwarf systems. Besides those which delve into the region of the cool red stars, there are a few which are primarily know otherwise in astronomical circles. One example is that of VV Cephei which had a long tenure on the Sproul astrometric program and was close to Larry Fredrick's heart for some time.
There is a great future in the discovery of astrometric binaries which is an instant process but the determination of the orbit remains a long-time interval of observations.