Abstract: Eria mêdogensis S. C. Chen et Tsi was recently found in southeastern Tibet, several specimens of which have been collected by various botanists since 1980. This is a “normal” entity with its habit very similar to that of Eria coronaria, from which it differs by having a regular perianth and longer bracts. We think it probable that this new entity is a peloric form of Eria coronaria. Peloria (or pelory) is a type of floral abnormality, which is found in many zygomorphicflowered taxa. It was first detected by Linnaeus (1744) in Linaria vulgaris, and then by others in Labiatae, Orchidaceae, etc. However, it is still an open question how to explain it theoretically and how to treat it taxonomically. In Orchidaceae, so far as our knowledge is concerned, peloria has been encountered in no less than 21 genera. In most cases, peloric flowers are found sporadically on an occassional plant, as seen in Cypripedium reginae and Eria oblitterata. Sometimes, however, peloric form may occur coexisting with normal-flowered form in one and the same species, as seen in Dendrobium tetrodon and Epipogium roseum. They are both abnormally peloric forms. It would not result in naming or renaming a plant taxonomically, whether the appearance of abnormally regular flowers on a normal-flowered inflorescence, or of abnormal-flowered individuals in normal-flowered species. In Phragmipedium lindenii, however, the case is different. It is quite “normal” and even of wider distribution than its nonpeloric allies P. wallisii and P. caudatum, from which it has once been considered to be derived. This is a normally peloric form. Whether it is a reversal or not, the appearance of a “normally” peloric taxon may be taken for a leap in the process of evolution. Taxonomically, we had better treat it as a separate species, especially when its origin is uncertain. For example, the entity just mentioned had been treated as a peloric va riety of Phragmipedium caudatum (var. lindenii) until 1975, when Dressler & Williams recognized it as an independent species based on the fact that its nonpeloric flowers occassionally found in a peloric population in Jungurahua of Ecuador are dissimilar in lip to those in P. caudatum. Garay (1979) considered it to be a peloric form of P. wallisii but maintained it at the specific level. This is indeed a good example of taxonomic treatment of normally peloric form. On the other hand, however, most of the regular-flowered entities in Orchidaceae are not peloric but rather primitive forms, such as Neuwiedia, Apostasia and Thelymitra, of which no less than 50 species have been reported since the eighteen century. They have never been regarded as peloric forms. Unfortunately, this has been neglected by some botanists. For instance, a hypothetically primitive orchid flower designed by Pijl & Dodson (1966) has a distinctly specialized lip with a short spur. In fact, in addition to the aforementioned genera we have some more examples of normally regular-flowered orchids. Among them Archineottia is the most interesting. This is a genus of four species, two of which are regular-flowered. Of special interest is that in this genus and its ally, Neottia, one can find all steps of column evolution from a simple form with stamen and style not fully united to a most complicated form in which they have well fused. Archineottia has a very primitive column, on which neither rostellum nor clinandrium is found but a terminal and undifferentiated stigma (Fig.2: 2, 4, 6, 8). In addition, there exists on the back of the column a thick ridge with its upper end joining the filament with which it is of same texture. It is obviously the lower part of the filament which has been adnate to the style (column). In Neottia, however, the column is much more advanced and very typical among the family. It has a very large rostellum and most complicated stigma structure (Fig. 10, 12, 14, 16, 18). One of the most interesting examples is Neottia acuminata, in which the stigma even becomes lamellate and almost backwards clasps the erect rostellum, but the perianth is more or less regular with its lip entire and somewhat similar to, but shorter and wider than, the petals. In these two genera there are altogether three species, namely Archineottia gaudissartii, A microglottis and Neottia acuminata, possessing regular or nearly regular perianth (Fig. 2: 1, 3, 17). They are obviously not peloric forms. We can not imagine, indeed, that a complicated form like Neottia acuminata or its allies would degenerate step by step into a simple form, and finally into a peloric form. Archineottia belongs to the subtribe Listerinae, which is closely related to Limodorinae, a rather primitivs subtribe with some genera possessing single pollen grain, relatively few and long chromosomes and monocotyledonous habit. Apparently, there is nothing surprising in the occurrence of some normally regular-flowered taxa, such as Archineottia, Diplandrorchis, Tangtsinia and Sinorchis, in these two primitive subtribes. Another instance is Aceratorchis, a genus formerly included in Orchis, from which it is distinguished by the entire lip which is more or less similar to the petals. Strictly speaking, however, its flowers are not truly regular. Two species have been described in this genus, but they were recently considered as conspecific. Aceratorchis tschiliensis is widely distributed from Hebei through Qinghai and Sichuan to northwestern Yunnan. It is cross-pollinated and produces seeds efficiently. All these indicate its normally primitive taxon, instead of peloria. It may be noted here that Asia is rich in members of Orchidioideae, as well as its primitive representatives. The occurrence of a normally regular-flowered form in Asia, whether representing primitive form of Orchis or Orchidioideae, is imaginable. In Orchidaceae, as mentioned above, regular flowers are not only found in some primitive taxa and peloric forms, but also in a few advanced groups. For example, a close investigation by the senior author (Chen 1979) on Satyrium ciliatum revealed that this species has hermaphrodite, staminate and pistallate forms, for which no less than nine names have been published. The flowers of its pistallate form are almost regular, in which nothing is found but three similar petals and an elongate style with three stigmatic lobes at its top (Fig. 2: 19). It is interesting to note that floral reversions in Orchidaceae are not always in connection with peloria. For example, Epidendrum triandrum of North America represents another kind of reversion. It is a reversal to abnormal polymery of stamens and not to abnormal regularity of perianth. Like Phragmipedium lindenii, it is also hereditary. We may give it a new name “Polyandrism” or something else, but, in fact, there is no essential distinction of this kind of reversion from peloria. It deserves mentioning that most of the regular-flowered entities, including primitive, advanced and peloric ones, occur in Asia and Australasia, where the Orchidaceae may have originated as pointed out by some botanists. We have good reason to verify the primitiveness and normality of many regular-flowered entities, but there exists no sufficient evidence for the impossible existances of normally regular-flowered species in those like Dendrobium, Eria, Lecanorchis, etc. For instance, Lecanorchis javanica, Dendrobium atavus and the new species described here are considered to be peloric forms, but it is only a conjecture, for no reason can be given for it. It is not impossible that some so-called peloric forms may prove to be truly primitive ones in the future. Of course, a closer investigation is needed. Summarizing the above, we may come to the following conclusions: 1. Regular or nearly regular perianth is a normal characteristic of orchids. It is chiefly found in some primitive taxa and sometimes also in certain peloric forms and advanced groups. Regular-flowered entities may not necessarily be peloric forms. 2. There exist two different types of peloria in Orchidaceae. One is abnormal form, with its peloric flowers appearing at random. The other is “normal” form, with its individuals all possessing peloric flowers. The latter is inheritable and can produce seeds efficiently, It would be best to treat it as an independent species taxonomically, especially when its origin is uncertain. 3. Although peloria has been considered to he a reversal as a whole, conditions vary from plant to plant. Some peloric forms have petal-like lip, and others have labellum-like petals. Sometimes the same plant produces different kinds of peloric flowers in different years, sometimes peloric flowers do not reappear upon the same plant. A few species can produce both peloric and normal individuals, but others produce peloric forms only. Peloria is in fact a term only used to cover the phase in which lip becomes similar to the petals. It is never all-embracing. We recognize the existance of peloria in Orchidaceae, but great care must be taken to distinguish truly peloric form from normally primitive one. It must be admitted that what causes peloria and even what is peloria are still problems awaiting solution. Acknowledgments: Our heartfelt thanks are due to Dr. Leslie A. Garay, Curator of the Orchid Herbarium of Oakes Ames, Botanical Museum of Harvard University, for his valuable suggestions during the preparation of this paper. We are also indebted to the artists, Mrs. Chunrung Liu and Mr. Chao-zhen Ji of our department, for their preparing the fine drawings.

Key words: Eria mêdogensis, Eria coronaria, Orchidaceae, Phragmipedium lindenii, Archineottia gaudissartii, Archineottia microglottis, Neottia acuminata, Aceratorchis, Satyrium ciliatum