Among familiar cultivated species spent flowers forming a tube = Camassia leichtlinii. This, therefore, is the correct identification of the mystery and not C. quamash.
I have never known before how to tell Camassia quamash V Camassia leichtlinii Thanks Ron for the tip.
I'm a little baffled by this; as well as the tube formation, the key in Flora of North America gives the additional distinction: * ... flowers ... usually (4–)10–35(–58) blooming simultaneously, except on few-flowered individuals — C. quamash. * ... flowers ... usually 1–3 blooming simultaneously — C. leichtlinii [and C. howellii]. The photos above show respectively 14, 7, and 11 flowers out simultaneously. So on this character, they are C. quamash. But photos online purporting to be C. leichtlinii often show more than 10 flowers out per stem, so either those photos are misidentified, or else the FNA character is unreliable. Unfortunately, I don't know the species well enough to say which.
I don't remember anything I was supposed to have learned five years ago either, and so I've been reluctant to photograph and name Camassia, but Ron B and @Daniel Mosquin have tried to teach us this before, see https://forums.botanicalgarden.ubc.ca/threads/flowers-in-park.95273/#post-370832. The Botany Photo of the Day links three postings before that don't work. I wish they would come back; there is so much great information there, with links to them all over the place. I don't suppose there's anything I could do to help make that happen, but I'd be willing. Edited: Michael F's posting came in while I was typing; my posting is not a reply to that.
The mystery looks to be one of the commercially circulated types of "Camassia leichtlinii" showing a blending of purple and light blue in the flower colors. With particular cultivar names like 'Azurea' perhaps being applicable. Awhile back I read an account on the web saying such versions were from crossings of the often purple C. leichtlinii and the light blue C. cusickii undertaken in the Netherlands. (An account which I have not been able to find subsequently). Perhaps this parentage explains discrepancies with pure C. leichtlinii regarding the open flower counts.