DICHOGAMY. 311 



In protogyuous dichogamy it is no uncommon thing for the stigma to project 

 from the flower ah-eady fitted to receive pollen whilst the petals ai-e still closely 

 shut, the whole flower having the appearance of a bud. This happens in the 

 Curled Pond weed (Potamogeton crispus, figured on p. 148), in Asphodels (e.g. 

 Asphodelus albus), in Woodrushes (e.g. Luzula nivea), in Elms (e.g. Ulmus cavipes- 

 tris), in the Plantain (e.g. Plantago media), in several Rhododendrons (e.g. Rhodo- 

 dendron ChaTnoEcistus), in Cortusa, Deutzia, and many other plants. On the other 

 hand, many plants with protaudrous dichogamous flowers are known where the 

 pollen is shed from the anthers while the petals are still folded as in the bud. On 

 opening the ripe bud of Grucianella stylosa (figured on p. 265), it can be seen at 

 a glance that the anthers have already dehisced some little time, and have 

 deposited their pollen under the dome of the closed bud on the thickened warty 

 surface at the end of the style. In the flowers of Rhododendron hirsutum, the 

 pollen falls from the anthers while still in the bud, and the same may be observed 

 also in many Composites, Campanulacese, and Papilionaceous flowers. 



We are not yet in a position to say whether protandrous or protogynous species 

 are the more abundant, although the dichogamy of thousands of plants has been 

 investigated. By generalizing on this subject one is liable to fall into very great 

 error. It would be particularly dangerous to generalize prematurely on the results 

 which have been obtained from the examination of many species of a genus, or 

 many genera of a family, and to consider them to hold good for the whole group, 

 for most genera contain some protandrous species, even when the majority of them 

 are protogynous, and vice versa. Liliaceous plants are described as protandrous in 

 most Botanical books, but in reality many of the genera and species are incom- 

 pletely protogynous {Amaryllis, Asphodelus, Colchicum, Erythronium, Leucojum, 

 Lilium Martagon, Narcissus poeticus, Ornithogalum umbellatum, Scilla, Trillium, 

 &c.). Among the Umbelliferse, which are usually stated to be all protandrous, 

 there are quite a number of protogynous genera and species, as, for example, 

 j:Ethusa, Astra'ntia, Gaucalis, Eryngium, Hacquetia, Pachypleurum, Sanicida, 

 Scandix, and Tnrgenia. This also applies to the Saxifrages: the majority, of 

 course, are protandrous, but some of them, viz. Saxifraga androsacea and peltata 

 are decidedly protogynous. The large-flowered species of Crane's Bill (Geranium 

 argenteum, lividum, pratense, sylvaticum) are protandrous, the small-flowered 

 {Geranium, coluinbinum, lucidum, pusillum, Robertianum) are protogynous. In 

 the Scrophulariaceae the genera Digitalis and Penstemon are protandrous, and the 

 genera Linaria, Pcederota, Phygelius, Scrophularia, Veronica protogynous. Of 

 species belonging to Boraginaceae some are protandrous (e.g. Borago, Echium), 

 others protogynous (e.g. Gynoglossum, Lithospermum). In Ranunculacese the 

 genus Acoyiitum is protandrous, while the genera Adonis, Anemone, Atragene, 

 Clematis, and Poionia are protogyuous. In the Gentian family some are pro- 

 tandrous, viz. Swertia perennis, Gentiana asclepiadea, ciliata, cruciata, Frcelichii, 

 Pannonica, Pneumonanthe, punctata, and prostrata; others, es]^ecia,\ly Menyanthes 

 trifoliata, Gentiana Bavarica, Germanica, glacialis, Rhoetica, and vcrna are 



