IKi PLANTS OF NEAV ZEALAND 



revevse to be true. Though the plant is weh adapted for 

 cross-polhnation, insects rarely visit it. Mr. Cheeseman 

 states, " For the last seven years I have made it a practice to 

 watch beds of this orchid, and, save on two occasions, I have 

 never seen winged insects enter the flower, and in both these 

 cases the pollinia were not removed." Out of '218 flowers 

 examined, only seven had lost their pollinia. Thus, it would 

 appear, that the plant is occasionally, though very rarely, 

 cross-pollmated. On the other hand, as the species is a 

 predominant one, it must obviously be largely self-fertilized 

 and often cleistogamic, thus apparently forming an 

 exception to the general rule that it is a disadvantage to 

 a flower to be self-pollmated. Earina, Dendrohitim, and 

 CorysantJics, on the other hand, are apparently completely 

 dependent upon insects to secure pollination. The following 

 description is abstracted from Mr. G. M. Thomson's account 

 of the pollination of Eari)ia siiaveoleus : " The flowers are only 

 about one-third of an inch in diameter, white in colour, with a 

 yell(jw centre, and with an almost overpoweringly sweet 

 perfume. The labellum is 3-lobed, and stands nearly erect in 

 front of the column. There is no nectary, but the tissue at the 

 base of the labellum is easily punctured, and exudes beads of 

 moisture. The column is short and erect, the stigmatic 

 surface very concave, with the viscid rostellum projecting 

 prominently forward above it. The anther is terminal and 

 deciduous, and encloses four pollinia attached in pairs to a 

 short stem, and resting on the rostellum. From the position 

 of the parts it appears to be impossible that self-pollination 

 could take place. The pollinia are very coherent, and lie 

 closely ensconced in the anther case. For pollination by 

 insects, however, the parts are very simply fitted. An insect, 

 visitmg the flower, would insert its head or proboscis into the 

 small square apertiu'e between the labellum and the column, 

 and in withdrawing it would inevitably touch the viscid surface 

 of tlie rostellum, and bring away the pollinia. Were these to 



