This addendum discusses linear dispersion of prisms and gratings, something that the solar observers of about 140 years do not even hint about. Once you realize how much dispersion they are using, it gives you a better idea of the Zeeman effect on a spectral line)
Discussion on linear dispersion with prisms and gratings is from the following books: Experimental Spectroscopy, R. A. Sawyer, pages 96 to 97, and Amateur Astronomers Handbook, J. Sidgwick, chapter on spectroscopy.
A single dense flint prism, n = 1.625, with 60 degree angles of 95mm refractive face. Littrow spectroscope of 1827mm f.l. at 5893A wavelength.
Linear Dispersion of a Dense Flint Prism
wavelength one prism four prisms four prisms
1.8 meters 1.8 meters 0.9 meter
4896A 8.9A/mm 2.2A/mm 4.5A/mm
5593A 13.8A/mm 3.4A/mm 6.9A/mm
6315A 20.3A/mm 5.1A/mm 10.2A/mm
A single flint prism of 10mm base passing yellow light is equal to 1000 grating lines resolution.
A single prism with 95mm base is equal to 9500 grating lines.
Four prisms with 95mm base are equal to 38,000 grating lines.
Or five prisms with 75mm base are equal to 38,000 grating lines.
Linear Dispersion of a Reflection Grating
focal length. 1200 gr/mm grating 600 gr/mm grating
3.8 meters 2.0A/mm 1st order 0.7A/mm 2nd 4.0A/mm 1st order 2.0A/mm 2nd
0.05A=24 mic 0.05A= 72 mic
1.9 meters 4.0A/mm 1st 1.3A/mm 2nd 8.0A/.mm 1st 4.0A/mm 2nd
0.05A= 12 mic 0.05A= 36 mic
0.95 meter 8.0A/mm 1st 2.6A/mm 2nd 16A/mm 1st 8.0A/mm 2nd
0.05A = 6 mic 0.05A= 18 mic
The photospheric yellow nickel line between the two sodium lines is 0.03A dark core. The sodium lines have a dark core of about 0.1A. The green magnesium line of b 4 is 0.05A dark core.Detecting the Zeeman widening effect is not real obvious with a medium sized spectrohelioscpe. You must look carefully and go slowly from spectral line to line.
3.8 meters f.l. is 12 ft 6 inches; 1.9 meters f.l. is 6 ft 3 inches; 0.95 meters is 3 ft one inch.
Visual photospheric lines of 0.03A dark core width that are slightly Zeeman sensitive will widen to about 0.05A to 0.06A. Very sensitive Zeeman lines will widen to 0.2A or more. Therefore, a spectrohelioscope in the spec mode needs about 2.7 meter f.l. for the telescope, giving a 25mm sun image, and about 1.9 meters f.l. for the spectroscope, 1200 gr/mm grating, used in the second order, giving about 1.3A/mm dispersion. Cones in the human eye are about 2 microns. CCD pixels are about 10 microns. Eye has a keen advantage to detect the Zeeman effect. Of course, longer focal length optics are better yet.
I hope that this discussion clarifies the observation of the Zeeman effect. Dr Hale of Mt Wilson had about professional 30 shs built and distributed in the world after 1930.Two amateur chaps in England made such a fixed instrument about 1935, using large two prisms and a flat mirror, giving about 6A/mm. Original gratings were expensive. After WW II two more chaps in the United Kingdom made a fixed shs and two in the USA about 1957 when replica gratings were produced. Oh, yes, and one in Tasmania. SHS in the hands of amateurs were extremely rare. I had my compact, portable shs published in Sky and Telescope, 1969. Things changed thereafter.