992 DR THOMAS R. FRASER ON STROPHANTHUS HISPIDUS. 



the immature moist follicle, are arranged parallel to and in close contact with each other, 

 but in the mature follicle, as the process of drying advances, they acquire elasticity and 

 a tendency to diverge and become separated from each other. The basal seed-hairs, being 

 interposed between the seeds and the endocarp, thus press the seeds towards the ventral 

 or placental surface, and through the openings in this surface ; and the movement out- 

 wards of the entire seeds, as well as the extrusion of the placenta, is aided by a similar 

 elastic force acquired, during drying, by the hairs of the comose appendages. 



In order to convince oneself of the adequacy of the extruding force of these hairs to 

 produce dissemination, it is sufficient to observe a follicle from which the ventral surface 

 has been removed. In a short time the contents of the follicle protrude through the 

 vacant space, and the protruded seeds, with their appendages, expand into a large loose 

 heap, consisting of the seeds mingled with the widely separated basal hairs, and of the 

 comose appendages with their hairs now widely diverging from the stalk of the coma. 

 On moistening the large heap of seeds and hairs, it soon again shrinks into a small bulk, 

 owing to the hairs losing their elasticity, and again becoming closely approximated to 

 each other along their whole length. 



The basal seed-hairs, which separate the seeds from the endocarp and from each other, 

 and which in mature follicles are no longer attached to the seeds, seem to possess the 

 additional function of preventing fracture of the long and brittle shaft of the comose 

 appendages, by forming a soft and yielding bed for the seeds, during their changes in 

 position before they escape from the follicle. They thus insure that the seeds shall be 

 disseminated with their comose appendages attached to them. 



I have not considered it necessary to give a detailed and systematic description of 

 the histological characters of each part of the Strophantus hispidus plant. These 

 characters are fully illustrated by the figures relating to histological structure in Plates 

 III. to VII., in whose preparation I owe much to the kind assistance of Dr Macfarlane, 

 of the Botanical Department of the University. The description of the figures (pp. 

 1025-1027) will sufficiently explain their more important details. 



