The upper image is of the inflorescence of Manfreda undulata 'Chocolate Chips'. The lower image is of the inflorescence of Manfreda virginica. Manfreda are so closely related to tuberoses (Polianthes of various species) and agaves that they will form viable hybrids. Manfreda virginica is hardy here; I keep 'Chocolate Chips' in a cold frame during the winter for now; I'll try it in the open after it pups a bit.
It took Manfreda virginica years to build up to blooming size; when it finally did the scape was seven or eight feet high. The inconspicuous flowers were not fragrant as far as I could tell - that was a big disappointment. I had been growing it with the idea that it would be a sort of hardy tuberose.
'Chocolate Chips' was obtained in 2010, so it has been by comparison quick to bloom, especially for a pot grown plant. The rosette of foliage is only about a foot across, so this is not a big plant. But the scape quickly pushed up to the five foot level. My experience with the flowers of M. virginica did not prepare me for the flowers of 'ChocolateChips': they''re not showy, but they are definitely unusual and in their way eye-catching.
When I described the flowers of this species on the Pacific Bulb Society forum this week, another contributor, Nan, responded with the information that she thought the inflorescence looked like a cluster of daddy long legs (aka crane flies). What a great description!
'Chocolate Chips' does not smell like its namesake. It has a very odd scent - my first impression was that it smelled like hot metal. But there is also a fermented fruit quality to the scent, too. What pollinates the wild forms of Manfreda undulata? I'm guessing bats.