L. Hably szerk.: Studia Botanica Hungarica 19. 1986 (Budapest, 1986)
Németh, Ferenc; Iványi, E.: Morphometrical studies on the Hungarian representatives of Ophrys scolopax Cav. agg. (Orchidaceae)
and petals show the same trend in differences in each tivimlnitl \v.ir. I'he number iif flowers is consequently higher In the Kunpeszér population thun In Pécs. I inet is an inverse' trend In the length of the horn. At Kunpeszér the dominating colour factor is i»ink. white is ivr\ rare, ami green or greenish pink never occur, in contrast with ihe Pécs pt^xii.-rtion ,\iu're all kinds of colour are present in a remarkable proportion. These tJJjcrepuncics • m !»• ^plained bv gcnctical differences between the two populations. The reason for the significant dilti . ncis in the size of the sepals is not so clear, because there are also remarkable ditferenei*s la-lvvfcn the subsamples. Weather-dependence of the length of plant is not surprising but this feature is unexpected at the length of labelium. A considerable part of the characters (size of the petals, colour and shape of the appendix! follows a much more ambivalent trend, I.e. thev show high significant differences between certain subsamples, but seem to be homogeous in others of the same range. This can be partly explained bv the difficulties of scaling. The ranging of some really continuous characters into 2-3 classes was not always unambiguous and reproducable. Furthermore, the sampling error mentioned above can be a relevant explanation too. The multivariate analysis also verified the existence of wellseparated groups of homogeneous localities (Fig. 1). Two isolated groups of the Pécs population can be observed with considerably low values of linkage with the others. These groups are rather heterogeneous in themselves too. The distribution of the intermediate forms is also ordered into more or less homogeneous groups on the basis of their locality. CONC LPS IONS The taxonomical investigation of the genus Ophrys is be set by some special difficulties: - many populations are genetically heterogeneous - the original descriptions are insufficient - the garden culture is unsolved - the characters can be well observed only on living plants - some characters change during flowering - the scaling of some qualitative characters is problematic - the flowering specimens of one year do not represent the whole population. Some of these problems have remained unsolved in our investigation, e.g. the nomenclatural identification, the scaling of some qualitative characters or the correct method of sampling.Certain examined characters have proved to be of diagnostic value (length of the horn, colour of the sepals and petals, in contrast with some others) sizes of the flower parts, colour and shape of the appendix. These latter characters are unsuitable for further analysis, but we can propose some new ones for examination. The density of the inflorescence can be composed from the length of inflorescence and the number of flowers; this latter one is itself a distinguishing character. Another possible character is the pattern on the labelium, but to scale it seems to be extremely difficult. A similar examination was made on Ophrys sphegodes (SAPSAL 1985) without convincing results; it was caused probably by the scaling problems. According to the descriptions, the hairless edge of the labelium is considered to be an important distinguishing character, but we could not determine it unambiguously, because it has a continuous transition into the papi 11 ou S part of the labelium. Some other characters e.g. the length of the bractea and the position of the medial sepal may be differential in certain Mediterranean races (DANESCH, SPNDERMANN), but not in the Hungarian ones. Thev were observed in the first year and we have found them strongly dependent on the phenuphase. Acknowledgement: In the; first place we would like to express our thanks to L. KOVACS who wrote the computer programme of the multivariate analysis and helped us to run it. and J.SAPSAL, who gave a similarly valuable assistance in the field observations. Further assistence was also kindly given by B. LIPP AKT. '/s. VAl.ASKK, O. BORSOS, Gy. MA KAR A, Ê. SZŐKE, G. KOVÁCS, P. ZAMBELLY, B. KE VE Y , J. BÍ'Kl, G. RESZLER, A. DEMETER and Cs. MOSKÁT to whom we are indebted.
