54 ELEMENTARY BOTANY. 



with a little stem, or caudicle, which has a very viscid disk. 

 These two disks are so placed that when an insect visits the 

 flower and thrusts its proboscis into the spur for the nectar, 

 they will touch and adhere to its head, and be dragged from 

 their place when the insect departs. The pedicels dry quickly 

 and curve downward ; when, therefore, the insect approaches 

 another flower of the same kind, the pollen masses, or pol- 

 linia as they are called, strike against its viscid stigma, and 

 a portion of the pollen is retained. The pollinia of this 

 flower are in the same manner transferred to the next vis- 

 ited, and so on. When access to insects is prevented, no 

 seeds are produced, showing that self-pollination is impossible. 



13. Many tropical plants cultivated in the conservatories 

 invariably fail to produce seed. The cause of this is to be 

 found in the fact that the tropical insects which alone can 

 effect their pollination are not present. It is not at all seldom 

 that only a certain species, or at most, only a few species, of 

 insects can pollinate a particular kind of flower. Thus in 

 case of the Yucca, a certain minute insect (Pronuba), after in- 

 serting its eggs in the ovary, ascends the stamens and pro- 

 cures pollen ; it then mounts the pistil and deposits the pollen 

 on the stigma. In the absence of this species of insect no 

 seeds are formed. Many of the adaptations for crossing, it 

 should be remembered, do not absolutely prevent self-pol- 

 lination, so that if insects fail to visit the flowers a few seeds 

 may, nevertheless, be produced. Wlien the flowers are evi- 

 dently arranged to favor self-pollination and prevent crossing, 

 they are said to be cleistogamous (Gr. Ucistos, closed ; gamos, 

 union). But no known species is altogether cleistogamous. 



14. Examples of cleistogamy are furnished by one set of 

 flowers of Viola, Oxalis, some Grasses, etc. " Their petals are 

 rudimentary, or quite aborted ; their stamens are often re- 

 duced in number with anthers of very small size, containing 

 very few pollen-grains, which have remarkably thin, trans- 

 parent coats, and generally emit their tubes while still en- 

 closed within the anther-cells; and. lastly, the pistil is much 

 reduced in size, with the stigma in some cases hardly at all 



