THE ANDRCECIUM 



27 



THE ANDROECIUM. 



It has already been shown, in consider- 



all anthers (Fig. 119). Only rarely is thi» 

 condition permanent- Ordinarily the falsa 



ing the general nature of the flower, that septa more or less completely disappear 

 the androecium typically consists of two after the formation of the pollen, leaving 



the mature anther two-celled, or the con- 

 dition is brought about in other ways. It 

 is furthermore assumed that the filament 

 is normally continuous along the back of 

 the anther in the relation of the midrib of 



stamen-circles, the stamens of each isom- 

 erous with the parts of the other circles, 

 one standing in front of each petal and se- 

 pal, that each stamen is entirely free and 

 distinct, and of characteristic form and 

 structure (Figs. 13 and 14). We have also 

 pointed out some of the forms of devia- 

 tion due to duplication, suppression, adhe- 

 sion and metamorphosis. To these the fol- 

 lowing general remarks may be added: 

 When an anther, still present, has lost its 

 function, it is called a Sterile or Imperfect 

 Anther. When the anther has become 

 suppressed, but the filament remains, the 

 latter is called a Sterile Filament. One 

 or more complete circles of sterile fila- 

 ments, changed or not by metamorphosis, 

 may be mistaken for a crown or a disk 

 (Fig, 36). Adhesion of the stamens to 

 the corolla, or even to the ovary, may in- 

 clude only one circle, the other circle being 

 entirely free, or they may be adnate to 

 different degrees (Fig. 28a). 



Before discussing other and specific 

 points of variation we shall consider the 

 typical organ more in detail. The num- 

 ber of stamens in the androecium is indi- 

 cated by joining the appropriate numeral 

 to the suffix "androus," thus,Monandrous, 

 Diandrous.Triandrous, Tetrandrous, Pent- 

 androus, &c. These terms do not neces- 

 sarily indicate the numerical plan of the 

 flower. When the number is 20 or more, 

 the term Polyandrous or Indefinite is com- 

 monly employed. 



In color the filament is commonly white 

 or whitish, and the anther j-ellow, but this 

 is not an absolute rule, as the latter is 

 often blue, brown, black or otherwise col- 

 ored. 



There are several distinct forms of at- 

 tachment of the anther to its filament 

 which are characteristic of larger or 

 smaller groups. Its origin from the leaf 

 assumes the curving forward and inward 

 of the margins of the blade to become at- 

 tached to the face of the midrib, produc- 

 ing two thecae, and the production of a 

 secondary or "false" partition separating 

 each theca longitudinally into two locelli. 

 This implies a four-locellate condition of 



formati 



form of attach 



ment is called Adnate (Fig, 103). 



FL^.ios. 



Ftg. 'OY 



^.lo-j 



Pl^. I0<1 



F.g. ro5- 



Fig. »'3 



Ftg. no. 



^ r 



Tlcj 10 S. 



Ft a. ]\Z ^ 



Fig.1'5. 



FiQ-.llf. Fig. Its. 



•s 



I 



g.MI 



g. in Fig. 



ill 



fig. Mb* 



It may be attached only at some point 

 upon the back (Dorsifixed). Of this there 

 are two forms. In one (Pig. 104) the anther 

 is rigidly fixed, its lower portion close to 

 and parallel with the upper portion of the 

 filament, the Incumbent form. In the 

 other it moves freely upon the pivotal 

 point of attachment (Fig. 106), the Versa- 

 tile form. Rarely the anther Is wrapped 



or twisted about its filament (B ig. 105). 



