Notes by R Ehlers
T. minasgeraisensis had been imported from Brasil by various nurseries in America and Germany named T. leonamiana E. PEREIRA, but it differs greatly from the description of this dubious plant.
T. leonamiana E Pereira
The type specimen for T. leonamiana E Pereira was collected by Achim Seidel 1. October 1973 in Minas Gerais (in the extreme North of Brasil, without exact location), and E. Peireira described it in 1974. Renate Ehlers ordered T. leonamiana from Seidel in 1979. As the plants did not fit the description, Harro Heidt and Reinhard Thieken brought plants again in 1980 and 1981 from Seidel-nursery. Renate Ehlers has 3 different types of plants in her collection from Seidel, but none of them showed all the characters, mentioned in the description. Plants had been imported from Brasil by various nurseries in America and Germany named T. leonamiana E. PEREIRA, but all I examined, differed greatly from the description of this dubious plant. The inflorescence was never as narrow or laxly composed of few flowers as shown on the drawing by Pereira, the sepals were never subfree, connate for l mm, and never lepidote.
In 1979, at about the same time, Achim Seidel sold plants as species number 720, from southern Brasil, and showed a photograph in his catalogue. This plant # 720 corresponds exactly to T. recurvifolia var. subsecundifolia (Weber & Ehlers) Till, which has its habitat in the South of Brasil, Estado Parana, near Vila Velha. T. leonamiana E. Peireira was collected by Achim Seidel No. 668 in Minas Gerais, extreme North of Brazil, and he should have known the plant. But we then find, after the publication, that spec #720 is in fact treated as T. recurvifolia var. subsecundifolia by Seidel. It is clear that Seidel did know that this plant was different to T. leonamiana.
There remains many questions, because the plant has never been collected again in northern Brasil. One possibility is, that T. leonamiana is a natural hybrid, - this would explain that there are different types of plants around in the nursery of Seidel, but none of them really fitting the description. For a long time there was some confusion between T. leonamiana and T. recurvifolia var. subsecundifolia, and Harry Luther treated the latter as a synonym of T. leonamiana. However, a comparison of the types of both taxa revealed the following differences: in T. leonamiana the leaves are larger, the flowers are bigger, the sepals and petals are bigger than in T. recurvifolia var. subsecundifolia, and petals are violet in the former but white in the latter, and also the strong geographic separation is a fact.
Just now, in April 2006, some plants, which I got in 1980 from Seidel as T. leonamiana Pereira from Minas Gerais are flowering. I compared these plants with the descriptions and they fit quite well, in nearly all characters, to T. leonamiana. The few characters where they disagree are: inflorescence not lax, the internodes of the flowers are smaller at anthesis, but elongate and by postfloral they are very similar to the drawing by Pereira, the sepals much more connate and not lepidote.
They differ from T. recurvifolia var. subsecundifolia (Weber & Ehlers) Till by the following characters: leaves longer, 20 cm long, not only 15 cm, flowers 35 mm long, not 26 mm, sepals 20- 22 mm long and connate, not 17 mm and free, petals 35 mm and blue, not 27 mm and white, Anthers 3 mm, not 2 mm, ovary 4 mm long, not 2 mm.
In any case, these plants fit much better to the description of T. leonamiana Pereira, and they come from Minas Gerais, from extreme northern part of Brasil, whereas T. recurvifolia var. subsecundifolia (Weber & Ehlers) Till grows only in southern Brasil, Parana, near Vila Velha.
Until any material is collected from northern Minas Gerais, which fits the description of T. leonamiana it seems wiser to keep both taxa apart. —SeeSmith & Downs 1977
