This plant seems quite rare on the international scene but is fairly common in Australian collections, comparatively speaking. When Len Colgan put a query to users of the Internet he received one reply only and one note of encouragement.
Tillandsia roseoscapa was named in 1972 by Matuda but was not noted by Smith & Downs ( 1977). However. it does appear in Pamela Koide's Bird Rock catalogue in California.
A similar plant, T. roseospicata was named in 1975 by Matuda and is described in Smith & Downs but, as far as is known, is not held in any collection world-wide. The name has been bandied around in Australia but I believe this is because T. roseospicata is included in Smith & Downs while T. roseoscapa is not.
Access to the Mexican Cactus & Succulent Journals for 1972 and 1975, where the original descriptions were published, would be helpful but these are elusive in Australia. These descriptions would probably detail where Matuda originally collected T. roseoscapa. The published description may include a drawing or photograph of the original plant and this would be helpful also as I suspect there are problems with the description of this species.
This would not be the first problem with a description by Matuda. You may recall me jumping up and down about the description of T. beutelspacheri not matching the plants we had in Australia. Eventually Harry Luther amended the description in the BS.I Journal (1994). page 117
In 1984 Wulfinghoff brought a plant to Werner Rauh's attention. As a result of Werner's investigation Matuda’s original description of T. roseoscapa was updated and published in the B.S.I. Journal (1986). pages 66-68. This article forms the basis of my investigations.
The one internet reply, apart from the encouragement, suggested that Len's plant was a hybrid. Len then asked me to CAREFULLY take his plant to bits.
Our plant differs from Rauh' s amendment as follows:
1 Plant is long-stemmed, not short-stemmed:
2 Floral bracts almost equalling sepals, not exceeding:
3 Sepals anterior free, posterior joined 5 mm, not high connate:
4 Petals citron at top 1.5 cm, mauve 1 cm. white 2 cm to base, not 4 cm long, pale yellow-green white at base:
5. Filaments and anthers exceed the flower tube not just the anthers;
6 Anthers have a mauve flush (colour absent from Rauh's description).
All in all, I think our plant is correctly named as T. roseoscapa but the 1 cm mauve section in the petal intrigues me as it did Len Colgan. This was the reason for his international S.O.S.
Certainly the hybrid Tillandsia ‘Jack Staub’ or the wild x rectifolia or the formula ionantha x schiedeana (call it what you will) does have bicoloured petals. The description says upper 2 cm cream or white, bottom 3 cm pinkish-violet to purple but sometimes you come across just a pinkish-violet band in the vicinity of the sepal which is visible at flowering. Here we have a known violet petalled flower crossed with a yellow or greenish-yellow petalled flower.
In the case of our T. roseoscapa, does the mauve section occur because of the stronger light here than in Germany? Is ours a 'one-off situation'?
Would the owners of T. roseoscapa carefully inspect and note the colours of the flowers of their plants when they next flower?