It was a short-lived lenticular process, with a striped (!) filter which necessitated a rectlinear focusing mount in which the front element did not revolve: hence the Summar and Hektor (which was of course primarily a taking lens, not a projection lens). The film was embossed with lenticles and the filter had to be kept parallel.
It was introduced in 1932 and worked only with fast lenses (because of the striped filter) and the film speed was the equivalent of about 0.2 ASA (of course the ASA speed scale came much later, after WW2, based on Kodak film speeds), so fast lenses had a second advantage as well: 1/20 wide open on a sunny day! The arrival of (non-substantive) 35mm Kodachrome in October 1936, and Willmann's (substantive) Agfacolor Neue in the same year, both much faster, explain the short life.
Dufacolor was also available from about 1934/5 -- a ruled-screen additive process -- and was faster than either Kodachrome (ASA 10) or Agfacolor Neue (ASA 8) at around ASA 25. It could also be user-processed.
As far as I am aware -- and I could be wrong -- Autochrome was never available in 35mm; that was the one with the dyed starch grains, and well antedated World War One.
The Summar was also introduced in 1932 and one source (Wright and Glanfield) refers to a hexagonal iris for use with Agfacolor Neue; the Hektor was slightly earlier, 1931, but appeared in rectlinear mount only in 1933/4.
Cheers,
Roger