REPRODUCTION. 225 



and other relations of the anthers. It need, of course, not be 

 stated that insects perform the act of poUination of plants uncon- 

 sciously. The place w^here they are engaged in securing honey is 



Fig. 153. — Cross-sections of an anther. (After Sachs.) 



also the appropriate place to come in contact witli pollen. The 

 following is from Jordan's communication. 



1. Anthers introrse. Nectaries intrastaminal, that is, between 

 the androecium and the gynoecium ; as in Dianthus Cartliusiano- 

 rum, Lychnis dioica^ Nymphaea alba^ Comarum palustra^ Allium 

 Schoenoprasuin, etc. 



2. Anthers extrorse. I^ectaries extrastaminal^ that is, be- 

 tween the androecium and the corolla or between corolla and calyx ; 

 as in Ranunculus acer, Tilia grandifolia, Parnassia palustrisy 

 etc. 



Most interesting are those cases in which the anthers appear to 

 be unsuitably related to the nectaries as, for examjjle, in Convolvu- 

 lus arvensis. We cannot take time to discuss these relations. 

 (Compare Fig. 143.) 



We will now pass from the forms of dehiscence to the mecha- 

 nism of dehiscence. The important factor is the anatomical struc- 

 ture of the anther- wall (Fig. 153, E). The epidermis does not 

 assist materially in the opening of the anther, the librous layer {x) 

 (endothecium) is looked upon as the mechanically active tissue. 

 As has already been stated, the weakest point is where the two 

 valves are attached to the pillar of the connective. Thin- walled 

 cells form the connecting tissue. The question arises. Whence the 

 tension which causes the margins of the valves to separate from the 

 connective? We can actually observe a shortening of the fibrous 

 layer (in the mature anther) during gradual drying, so that the wall 

 curls back. Let us consider the individual cell of the fibrous layer 



