BY ALEX. G. HAMILTON, ESQ. 159 



to permit the indusium to come out of the tube, but contracts 

 again beyond that, so that the edges are close together, and each 

 half of the lobe bulges forward so as to make conjointly, a hemi- 

 spherical cavity just in front of the indusium, to which it forms a 

 cover, completely cutting the indusium off from the outside of the 

 flower, as shown in the diagrammatic section (fig. 10). On 

 examining the mouth of the indusium, it will be found to be 

 narrower and more slit-like than in the last specimen. Now 

 examine a series of full-blown flowers of increasing ages, and the 

 mouth of the indusium v/ill be found to close more and more on 

 its load of pollen, while it still i-emains behind the barrier formed 

 by the upper lobe. At last it will be found quite closed, and then 

 the flower withers and the edges of the lobe shrink away from each 

 other allowing the indusium to once more project into the flower, 

 but its mouth is closed against ail intrusion. In this stage the 

 indusium has a flattened shape as in fig. 9. 



From examining a very large number of the buds and flowers 

 in this way, I have come to the conclusion that fertilization is 

 effected in the following manner : — When the anthers are full 

 sized, and ready to burst, they bend towards each other over the 

 indusium so that their bases form a ring above the mouth of it, the 

 ring being a trifle less in diameter than the mouth. The style at 

 this time begins to lengthen rapidly forcing its way up through 

 the anthers, and by means of the fringe of hairs on the edge 

 scraping and brushing all the pollen out of the anthei's, the pollen 

 drops into the cup as the latter quickly grows upwards. This 

 packing of the pollen puzzled me greatly at first, for insects could 

 not possibly do it, and neither could the mere dropping in of the 

 pollen cause it to cohere so tightly as it does. As soon as the 

 anthers have discharged all their pollen, they wither and twist 

 outwards through the slit in the corolla ; and when the indusium 

 is filled with pollen, it begins to close and assume a flattened shape, 

 while at the same time the fringe of hairs slope over the mouth so 

 as to cover the pollen and retain it within the cup. Insect inter- 

 ference is provided against by the cover formed by the back lobe 

 11 



