214 OPENING OF THE PASSAGE TO THE INTERIOR OF THE FLOWER. 



18 Days. 

 Vaceinium Oxycoccos. 



30 Days. 



Cattleya labiata. 

 Va/ida coerulea. 



40 Days. j 60 Days. 



Ci/pripedium iJisigne. 

 Odontoglossum (many). 



50 Days. 

 Epidendrum Lindleyanum. 

 Phalcenopsia grandiflora. 



Oncidium cruentum. 



70 Days. 

 Cypripedium villosum. 



80 Days. 

 Odontoglossum Rossii. 



The duration of flowers varies then, in different species, from three hours to 

 eighty days. These remarkable differences are connected with the amount of pollen 

 produced in the flowers, and with the number of flowers on each plant. They also 

 depend on whether or no the stigma is entirely dependent on insects for pollen. 

 Flowers with numerous stamens and ample pollen, as for instance, Poppies, Cistuses 

 and Portulacas, have but a brief duration, whilst on the other hand, such as have but 

 a single stamen, e.g. most Orchids, remain fresh often for weeks. In plants which 

 produce but a single flower throughout the year, as the Snowdrop, the one-flowered 

 Winter-green {Pyrola uniflora), Herb Paris and Trilliwm, or at most two or three, 

 as in the Lady's Slipper Orchid {Cypripedium Calceolus), and in tropical Orchids of 

 the genera Oncidium, Stanhopea, and Cattleya, the flowers persist fresh and open 

 for long periods. It may happen also that in consequence of unfavourable climatic 

 conditions flowers may be deprived of insect-visits for many days at a time. In the 

 case of flowers so constituted that in the absence of insects no production of seeds is 

 possible, it follows that in some years the whole object of flowering (where but a 

 single or very few flowers ai-e produced) will be jeopardized. It is obviously of 

 advantage to flowers of this kind that they should be able to hold out for a consider- 

 able period. The longer they persist the better is their chance of being visited by 

 insects bringing pollen from other plants. 



Let us now take the other extreme, a plant producing numerous flowers, one 

 after the other, in the course of a year, flowers which are able in the absence of 

 insects to pollinate themselves. Here the duration of each flower need be only very 

 short. Notwithstanding the short duration of the flowers the plant remains in 

 blossom for weeks or months. The Spiderworts (Tradescantia crassula, Virginica, 

 &c.) have ephemeral flowers, but they go on producing them for eight weeks, during 

 the whole of which time the plants are daily provided with new^ ones. The .same 

 holds good in most Crucifei-s, Cistuses, Rock-roses {Helianthermim), Droseras, &c. 

 The last-mentioned open their flowers only under very favourable conditions of 

 weather, and then only every other day. At any rate, for Drosera longifolia it has 

 been shown that, even in the finest weather, a flower-bud opens on alternate days 

 only. Thus we see that pretty much the same result is accomplished in the tw'O 

 classes; in those plants possessing numerous, ephemeral flowers, and in those with 

 solitary, long-persisting ones. 



It has already been explained (p. 107) that in localities where a heavy precipi- 

 tation of dew obtains, flowers which remain open for long periods are liable to a 

 saturation of their pollen during the night, and that many protective arrangements 

 prevail to minimize this danger. One of the most commonly occurring of these 



