I 



670 



NATURAL CLASSES. 



Sect. XX. 4. 



that the anthers might approach the fligma 



the ftalk of the fl 



beino- fo flexible as to allow it to becoqae pend 



as 



the he 



merocallis flava, or yellow day-lily. 



In nigella, devil in the bufli, the ftyles arc very long compared 

 with the filaments, and bending down their ftigmas over the an- 

 thers in curves, give the flower a refemblance to a regal crown ; 

 which need not to have occurred, if the filaments could more eafily 

 have been lengthened. 



In 



fonia the two anthers fland widely diverging on fhort 



fligma into contaft 



In the fp 



filaments, and the tall capillary flyk bends its 



firft with one of them, and afterwards with th( 



tium fcoparium, common broom, the long flyle bends round 



circle to accommodate the ftigma to the fhort fet of anthers, 



great curvature need not 



eafily have grown longer 



related in Sed. VII. 2. 2. of this v/ork 



which 

 have exifled, if the filaments could more 



Other inflances of fim 



flrua 



It is probable, that fimilar obfervations, and a confequent reafon 

 o- on them, might be applied to many other kinds of flowers fo a 

 detedl the mofl unchangeable parts of them : 



s 



but great time, la 



f 



hour 



pport unity 



» 



nd 



ty 



9 



would be required to eflablifh 



from them the mofl- invariable and mofl 



ral clafTes of 



vege 



' ■- 



tation. 



4. Many different proportions and fituations and forms of the fila 



ments are 



merated in the Philofophia Bot 



of L 



ibme of which might poffibly have become claflical charaders, if he 

 had turned his attention to them, and given them adapted names 1 

 as he has done to thofe clafles, which he has derived from the fitu- 

 ations of the fexual organs, as didynamia, tetradynamia, fyngenefia, 

 and others,, which approach nearer to natural clafTes, and are fubjed 

 to lefs variation than the numerical ones. 



Some of thofe colle£lions of _ 

 natural orders, and fome of thof^e of Ray, and Tournefort, might 



which Linnaeus has termed 



4- 



perhap 



