THE STAMEN 117 



theory received little support. At that time, type of dehiscence was con- 

 sidered important in determining sporangium position. Extrorse and in- 

 trorse anthers (that is, with abaxial and adaxial dehiscence, respectively) 

 were considered derived from sporophylls with abaxial and adaxial 

 sporangia, respectively. The position of the supposed margins of the fila- 

 ment and anther was also used as evidence of sporangium position. Even 

 in the 1820s, it was believed that the margins of stamen and leaf are 

 equivalent. On the basis of similarities in color and vesture, it was 

 argued that the margin of the anther is "in the furrow into which the 

 pollen is discharged." But this interpretation was refuted by the fact 

 that, in some flowers, all the sporangia are on one side. Further, double 

 flowers also demonstrated, in their petaloid stamens, that all four 

 sporangia belong on one side, that the margins are on the "back" of the 

 anther, not in the dehiscence furrows. In 1845, this theory was called 

 "a false interpretation, twenty-five years old." Yet the present-day view 

 that the pairs of sporangia are marginal — the margin between the mem- 

 bers of a pair — is continuing that interpretation. But the evidence from 

 the woody Ranales, the Nymphaeaceae, and other taxa that all the 

 sporangia belong on one side must be accepted and the anther inter- 

 preted on this basis. 



Major modification of sporangium position has occurred as a part of 

 the reduction of the broad, ancestral stamen (Fig. 52). The lamina 

 was reduced in area by progressive lateral narrowing, with the bringing 

 of the sporangia closer to the sporophyll margin (Fig. 52b, c, d) and 

 then to the ridges ("corners") of the thickened, four-angled anther, the 

 position in latrorse dehiscence (Fig. 52c). From the lati'orse position, 

 the two outer sporangia were further moved — "over the edge" to the 

 side opposite the original position ( Fig. 52/ ) . Associated with this change 

 in sporangium position is the bringing close together of the sporophyll 

 margins (Fig. 52x x) on the originally sterile side (Fig. 52/, g). These 

 positional changes, brought about by more or less massive, differential 

 growth — contraction and thickening — suggest a partial inrolling of the 

 lamina (Fig. 53), but one basically of tissue rearrangement rather than 

 folding of little modified wings of the lamina, as in carpel closing. 

 This remarkable, evolutionary change in position of sporangia from 

 one side of the sporophyll to the other was, of course, at first ontogenetic; 

 some ontogenetic changes are known — anthers introrse in the bud, 

 extrorse in the flower — in some palms, Polygonaceae. 



The Nymphaeaceae and Cabombaceae provide excellent examples of 

 the evolutionary modification of the stamen, showing the change in 

 position of microsporangia from sunken to protuberant and from adaxial 

 to lateral and abaxial. (The sporangia in these families, like those of 

 the Magnoliaceae, are distal on the sporophyll, in contrast with those 



