170 



PRINCIPLES OF PLANT-TERATOLOGY. 



midrib, whether on the upper or the lower surface, 

 may be ascribed to fusion with, or separation from, 

 these veins of lateral lobes of the basal infolded part 

 of the leaf, or also to dichotomy or apical infolding 

 (see above) ; but these lateral lamellae can have had 

 nothing to do with modern anther-structure. They 

 are related to, and the result of, the great laminar 

 extent of surface obtaining in the foliage-leaf as com- 

 pared with that in the anther, hence the lamellae of 

 the midrib naturally tend to be repeated along the 

 strong lateral veins. The same kind of structure has 

 been seen, though here in the form of detached, almost 



Fig. 128. — Transverse section of extrorse anther taken at the dotted 

 line in PL XLIV, figs. 4 and 5, and supposing these leaves trans- 

 formed into anthers, ma, marginal ; me, median loculi. (Diagram.) 



saucer-shaped formations, on the lateral veins of the 

 lower side of the leaf of Ficus glomerata. 



Nor can we compare directly with anther-structure 

 the basal pocket or pockets occurring on the dorsal 

 side of the Polygonum-le&Yes, but they may be regarded 

 as stereotyped stages in the formation of extrorse 

 anthers ; such pockets are a natural result of the 

 presence of two basal infolded leaf-lobes in that 

 position ; and it is highly important to note that these 

 basal lobes correspond to the basal extension in a 

 versatile anther. 



One may here emphasize the importance of theoreti- 

 cal deduction when based on the proper arrangement 

 of facts for the solution of such a problem as that of 

 anther-structure. 



