Bromelia 1(2): 11-18, 1994 - Leme & Costa
Vriesea saundersii (Carriere) E. Morren ex Mez and V. botafogensis Mez have been known to science and raised by growers for over a century, but specimens of these two species have often been mistakenly identified in the past 40 years. Besides resembling each other morphologically, these two rock-dwelling species are both endemic to the city of Rio de Janeiro and surrounding areas, a fact that has added to the confusion. In the early 1980s, Edmundo Pereira studied this problem but his data were inconclusive.
V. saundersii was described initially as Encholirion saundersii by A. Carriere ( 1872), who gave only a brief description of the general appearance of the plant, mentioning its attractive red-hued leaves with wine-colored spots. In the following year, Andre (1873) published a more complete description of the species and included an illustration. He commented that this bromeliad from Brazil was introduced into England in the famous collection of Wilson Saunders and commercialized by Louis De Smet.
V. saundersii was included in the genus Tillandsia by K. Koch (1874). It was given its present name by C. Mez (1894), who adopted the combination suggested by E. Morren in 1875. He based his decision on what was thought to be the type specimen from the Herbarium Universitatis Leodiensis in Liege, Belgium. In the same work, Mez also described a new species, V. botafogensis, based on specimens collected in 1885 by A. Glaziou (# 16466) on the slopes of Pao de Acucar Mountain in the Botafogo area of Rio de Janeiro.
After describing the species, the author noted that he had based his description on a specimen (the holotype) which was deposited in the Herbarium of Berlin, Germany. Duplicate specimens, known to exist at the herbaria of the Paris Museum in France and the Royal Botanic Gardens at Kew, England, should be considered isotypes.
Mez maintained V. saundersii separate from V. botafogensis in his monograph on the family in 1935. He linked the former to V. muelleri Mez and V. procera (Martius ex Schultes filius) Wittmack and the latter to V. rodigasiana E. Morren and V. friburgensis Mez.
In 1955, L. B. Smith considered V. botafogensis to be a synonym of V. saundersii and retained this interpretation in his monograph on the family in the series Flora Neotropica (Smith & Downs, 1977). This interpretation has held sway since that time.
We examined a wealth of material, from live, flowering plants, both home-grown and wild, to various exsiccata (or photographs of exsiccata). including type specimens, and concluded that V. botafogensis is quite different from V. saundersii. We conclude that the interpretation of Mez (1894; 1935) is correct and should be reinstated. These species strongly resemble one another morphologically. Both have stiff leaves covered with wine-red spots and inflorescences which are similar in structure and color, and this led to their being considered synonyms in 1955. Certain traits which serve to distinguish these two bromeliads, however, have been recently overlooked by a number of authors.
In the first place, the overall appearance of the live plants is quite different. V. saundersii is larger , with an ample, loosely branched inflorescence; the stamens do not extend beyond the petals. V. botafogensis, on the other hand, forms a compact rosette, the inflorescence has shorter, more densely arranged branches, and the flowers have conspicuously exserted stamens.
The data given on page 15, which include the distinguishing features of each species and list the exsiccata which were examined by the authors were based on descriptions of V. saundersii made by Andre (1873) and Mez (1894; 1935). and on details contained in the illustration which was used by both authors. Although the rosette shape presented in this drawing does not closely resemble that which most often occurs in nature, the flowers with their included stamens are clearly visible. Because the type specimen from the University of Liege was not available, we examined a photograph of this material from the Smithsonian Institution. The isotype of V. botafogensis from the Royal Botanic Gardens at Kew was also examined as well as a photograph of the isotype from the Paris Museum. In this photograph, the exserted stamens, typical of the species, are easily observed. It was not possible to examine the holotype mentioned by Mez, which is probably still kept in the Museum of Berlin.
Populations of V. saundersii are found mostly on rocky slopes in some parts of the Tijuca mountains, The plants flower from December to March. V. botafogensis on the other hand grows on the steep granite rock races near the entrance to Guanabara Bay on both the Rio de Janeiro and the Niteroi sides of the bay. The plants flower from February to August. Because of the very limited distribution of these species, they should not be collected since enough plants are now under cultivation to ensure ready availability of both species. They are easily grown from seed or from offsets.
Although V. saundersii is easily set apart from similar species such as V. procera, some difficulty still remains when it comes to identifying dry specimens of those species that resemble V. botafogensis and which may be confused with this species. To cite an example, the following V. rodigasiana specimens of considerable historical interest, were erroneously identified by Mez in 1904 as V. botafogensis: Sao Paulo, Itapecirica, Wettstein & Schiffner 653 (WU); ibidem, Conceicao de Itanhaem, Wettstein & Schiffner 772 (WU). V. saundersii is distinguished from V. botafogensis by its less robust appearance, finer textured leaves varying in color from green to red and which are not consistently spotted, by its narrower, laxly branched inflorescence and smaller floral bracts and sepals, among other things. Furthermore, V. saundersii populations are found from Rio de Janeiro to Rio Grande do Sul, thus being much more widely distributed .
Finally, dry specimens of another species, V. friburgensis var. paludosa (L. B. Smith) L. B. Smith, which is found from Sao Paulo to Rio Grande do Sul, may also be confused with V. botafogensis. Plants of this species are either epiphytic or terrestrial and they are usually more robust than V. botafogensis. Other characteristics which serve to distinguish the former species are more pliable leaves and a longer, cylinder-shaped inflorescence with long, laxly arrayed branches. The recently described V. rastrensis Leme from the Serra do Rastro, Santa Catarina state, southern Brazil, belongs to the same species complex. High light intensities, strong winds and cold temperatures shaped this species to resemble V. botafogensis in overall appearance. However, its leaves have no spots, the upper bracts of the scape are rather narrow between the base and the orbicular tip, and the inflorescence is much longer and has sparsely arrayed, nearly reflexed branches. —SeeBromelia
