310 THE CROSSING OF FLOWERS. 



humble-bee, whilst lower still are flowers which have been open for two days. In 

 the recently-opened flowers the anthers are covered with pollen, while the stigmas 

 on the end of the downwardly-curved style are as yet immature and folded together 

 into a club-like body. This plant is therefore protandrous. The inflorescence of 

 Eremurus Gaucasicus, belonging to the Liliacese, is figured beside it (293 ^). Here, 

 again, the youngest flowers are still in bud, those coming next below have just 

 opened, whilst lower down still are the oldest flowers of all. In the newly-opened 

 flowers the anthers are closed and no pollen is exposed, but the pointed stigma, 

 terminating the upwardly- curved style, is already mature, so that this plant is 

 protogynous. Both protandrous and protogynous dichogamy may be complete or 

 incomplete. It is complete when the stigma begins to ripen after the removal of 

 the pollen from the adjoining anthers by wind or by flower- visiting insects, so that 

 it can no longer fertilize its own flower; or if the stigma is withered^ dried up, or 

 fallen away as soon as the anthers of the same flower open and expose the pollen 

 or scatter it abroad, as in the Wall Pellitory (see flgs. -291 2. s.*). Dichogamy is 

 incomplete when the ripening of the two kinds of sexual organs is not, indeed, 

 simultaneous, but when the capacity for fertilization of one sex is not at an end 

 before the other sex in the same flower is mature. There are, of course, many 

 grades in incomplete dichogamy. In long-lived flowers the start which one sex 

 has over the other may amount to several days, but in short-lived flowers it may 

 be Hmited to a few hours. Cruciferse all have protogynous flowers. The already 

 mature stigma is visible in the centre of the flower as soon as the petals open, 

 but the surrounding anthers are still shut up. This only lasts for a short time; 

 soon the anthers dehisce, and then both sexes come into operation. In Lepidium 

 Draba, Sisymbrium Sophia, and numerous other species, this difference of time 

 (lasting from the moment when the stigma is accessible to the moment when the 

 anthers begin to shed the pollen) is only 2-5 hours. The same may be said of 

 numerous Rock-roses, Papaveraceous plants, Cactuses, Ranunculacese, Rosaeese, 

 Boraginacese, Gentianaceae, Ericaceae, and Valerianacese (e.g. Helianthem^um alpestre, 

 Glaucium luteum, Opuntia nana, Actcea spicata, Adonis vernalis, Atragene alpina. 

 Clematis Vitalba, Potentilla caulescens, Gynoglossumpictum, Lithospertnum arvense, 

 Menyanthes trifoliata, Arctostaphylos uva-ursi, Vacoinium Myrtillus, Valerian- 

 ella dentata). Even the majority of ephemeral flowers exhibit dichogamy. The 

 flowers of the Marvel of Peru (Mirabilis Jalapa) open between seven and eight 

 o'clock in the evening; as soon as the margins of the flower unfold, the small 

 stigma, resembling a tiny brush, is able to receive pollen, but the anthers are as yet 

 entirely closed. About 10-15 minutes later the anthers dehisce and shed their 

 pollen. The difference in the time is so slight here that it would be unnoticed by 

 most people, and this explains why such flowers have not been regarded as dicho- 

 gamous. But the very circumstance that the maturation of the two kinds of sexual 

 organs is not simultaneous, even in ephemeral flowers, is of the greatest importance 

 for the question of the significance of dichogamy and must be especially dwelt 

 upon here. 



