Page:Journal of botany, British and foreign, Volume 34 (1896).djvu/28

 12 TJiRATOLOGICAL OBSERVATIONS ON PARNASSIA PALUSTRIS L. By I. H. BuRKiLL, M.A. Owing to the great interest which attaches to Parnassia, I have been led to examine during the past two summers a considerable number of flowers obtained from the cliffs of the Yorkshire coast near Scarborough. Here the soil is a heavy boulder clay, and tho situation in many parts somewhat sunless and exposed to cold winds from the sea. I have further examined all the flowers which I could obtain (under 200) on a peaty soil in Glen Clova and Glen Prosen, Forfarshire, at elevations between 700 and 1300 ft. above sea-level. In this short paper I propose to confine myself to statements relating to the variability of the flowers of this plant. The following table sets forth the number of flowers observed, with the variation in the number of parts in the individual whorls : — Type OF Floweb. No. OF Cases. Sepals. Petals . Stamens. Staminodes. Carpels. 9 4 4 4 4 3 4 4 4 4 4 4 201 5 5 5 6 3 Approximately 4700 Normal Flower. 1 ^ d 5 d 4 i 224 5 5 5 5 6

25 5 5 5 5 6 CD 1 6 6 6 6 3 P4 8 6 6 6 6 4 14 6 6 6 6 5 1 6 6 6 6 6 Ic 4 4 6 4 3 Ic 4 4 6 4 4 1« l^< 5 4 4 4 4 II l*c 6 5 7 5 6 PL^^ 1* 6 6 5 6 4 1 5 3 3 3 4 "^ Now, disregarding for the present the abnormalities of the last six types, about which further remarks are necessary, we notice at once that, whatever may be the number of organs in the outer whorls of the flower, the gynoecium has a tendency to produce one unit less. We may tabulate this thus : — c Chorisis of one stamen.
 * Change in symmetry of flower between calyx and other whorls