348 AUU-OOAMY, 



snooate iuterioi-s. The ovary is spiudle-shapoil, and boai-s at tho end of a sliort stvlo 

 a stigiua foveivd witJi little pipiUiv, The stanuvns. four in mn»lw, jvst with th^r 

 doi-sal svu-l'aeos in contact \\itl\ ih& ovary; thoir anthoi-s I'aco outwanis, an^ lanoeolftte 

 in shape, and each carries on its apex a sn>aH foliaeeons scale like the point of a 

 spear (,see tig-. •2!KS'»\ The tlowei-s aiv protogynons. or, in otlxer woi>ls, the p^^^iUose 

 stigma, which is heunned in In- tho fonr scales just ivt'envd to, is ahvady matuivat- 

 a time when the anthers aiv still chvsed (tig. 2!">S '\ At this stag-o the stigma inay 

 be dusted with pollen Cnnn other tlowers. The next step is tlxe dehiseenoe of tJj© 

 bilocular anthei-s, which takes place in a very peculiar n\annor. Tho wall of cju'h 

 anther-half which faces oiitwanls heconies detached, and is lifted up in tho form of 

 a flap with the whole of the pollen of the corivsponding cavity sticking to its inner 

 surface. The two tlaps then slarivel a>nd roll up, with the originally intor>\al surface 

 outside, nntil the only point of attfichtnent is just below tl»o pointed scale at the 

 apex of the anther; the fi-oe tlaps aivh over this foliaeeons anther tip (^tig, 298 '), 

 and also over the stigma, whicli is close to the scale (tig, "iS^S % The proct^ss is aiwul- 

 t-aneons in all fonr aavt hers, the result, being that tho stigma is nx^fed over by a vault 

 con\posed of eight cm-ved Haps, and as the exposed surfaces of tJio latter m^e thickly 

 coated with pollen, tlie whole awh piiiseutvs externally a continuous oovering of tliat 

 material (see tig. -208 s). Should insects now visit the flower to obtain the heuoy 

 stored in tlie slipper-sha^uHl petals they nnist brusli past this pollon-cin'OTOd vault; 

 nor can they fait to be besmeared at the very spot wheiv at a piwio\is stagv in tho 

 tlowers development they would have had to brnal\ against the stigma. This con- 

 dition of aiiairs lasts usually for two days. In the meantime chang-es atlecting the 

 pistil are talcing place which, notwithst^mding their inconspicuous naluiv. an> of tJ»o 

 greatest importance in relation to the eventual accomplishment of a.\itogamy. If til© 

 stigma is not dusted by insects with foivign pollen at tho lirst expansion of tlie 

 flower the pistil now elongates sutliciently to raise the stigma ii\to the vaulted cavity 

 just described; and as the Haps composing the vault curl up still raoiv (fig, '2!)8') 

 as the flower approaches its end, the stigma n\ust inevitably come into ooutaet 

 with the pollen adherent to them (cf. tig. iDS*). 



A similar process is observed to take place in sovoval Orucifera>, of whioli tho 

 Charlock {Sinapf^is anvnuift) is a typical inst«nco. The tlowers of this plant mv 

 protogynons. The bud opens ea.rly in the morning, revealing still closed anthers 

 witli their faces turned inwaiils (intnyrse), whilst the stigiua is already inaturo and 

 projecting somewhat abo\e the anthei's. At this stage of developn>ent no pollen 

 except what is brought by insects can be deposited on tlie stigma, A day later ihd 

 tlowor presents quite a diflbrent appeai'ance. The four longer staniena have sti'otohed 

 and curved a little away from the axis, a,nd the anthws are lifted above ih(s atigaia. 

 By a rapidly oxocnted twisting of tho lihvntents tlio ai\thers have been turned round 

 so as to face outwards, and extrorse sutural dehiscence has meantinie taken plnoo, 

 The stign\a. is completely withdra.wn from observation, and is also Ha.fe fnaa My 

 possibility of being dusted with pollen, for tbo aiUthers in the sa.n)0 tlowor have 

 turned away their pollen-coated faces, nor is it ywssiblo for any extraJieous pollon 



