Mathewson : Pollen-tube in Houstonia 



491 



no reason why it should not make its way into the cavity of the 



ovary directly after reaching 

 this tissue, but it appears that 

 it never does so. 



At the place where the 

 stylar elements blend with the 

 basal elements of the partition 

 the tube encounters a third 

 kind of tissue which is in sharp 

 contrast with those heretofore 

 P traversed. Its course appears 



to be interrupted by relatively 



Fig. 2. Contact of stylar tissue (S) 

 with the basal element of the ovary-par- 

 tition (P)y showing the turn made by the 

 pollen-tube on reaching the latter. 



shorter, columnar cells with 

 thicker walls beneath which are 



* 



parenchymatous cells. On 

 reaching this tissue the tube 

 invariably makes a sharp turn to 

 the right or left (figure 2), and 

 thus enters the stalk of one of 



the two placentae. It often Fig. 3. Basal portion of placental 



happens that this turn takes the stalk ( Pi ) and ad J ac ent tissue (/>) of the 



basal partition, with a part of the stylar 

 tissue (S) abutting on the latter. C', cellu- 



tube directly toward the surface 



of the placenta, and in such a lose plug in the pollen-tube, which at this 

 case the tube makes' a further point turns in order to penetrate the pla- 



turn which carries it beneath the 



cental stalk. Z, locule 



epidermis (figure 3). In no case does it approach nearer than 

 within one cell-layer of the surface. Large cellulose plugs (figure 

 3), twice the diameter of the pollen-tube itself, are often seen in 

 this region. They also occur quite often in the style. 



The pollen-tube continues its course 



through the placenta 



either immediately below the epidermis or from one to several 

 cell-rows below it. This tissue is composed of approximately 



