XXIII. PALM^. 



811 



connected ■with JuncagineeB. It is thus that we have united to Potamete, Rappia, which has liitherto been 

 placed near Posidonia and Zostera. On the other hand, it is probable that when the fruits and seeds of 

 HalopMla, Lemnopais, &c. are known, a family will be made of these genera, which, by its many-ovuled 

 ovai'ies with parietal placentation, will stand in the same relation to Naiadea that Butomece do to Alis- 

 inacece, which Aponogeton and Ouvirandra approach. 



It appears superfluous to discuss the modern view, based on that of Adanson, who considers 'as very 

 rational ' the affinity between AUsmaoeee and Ranunculacece, and we shall retain our opinion until we find, 

 on examining their seeds, with or without a microscope, an albumen and a dicotyledonous embryo in 

 jSa^sito-jij, which Adanson believed that he had seen, j ust as he fancied he saw two cotyledons in the 

 seed of Reeds. 



If, in spite of the conscientious work and the sagacious observations which during the last hundred 

 and fifty years have so greatly advanced Botany, it is allowable to revive paradoxes that have been 

 absolutely condemned by science ; if mere superficial resemblance is sufficient to establish natural affinity, 

 we do not see why we should hesitate to follow Adanson in uniting, as he has done, Cyrndea with 

 Palms, Aristolochiece with Vallisnei-ia, Polygala with Tithymalece, and so forth. 



XXin. PALM^, L., Juss., Martins, Blume, &c. 



Chamcerops humUU: Polygamous inflorescence. ChauKsrops humilU. Diagram Q,. 



ChamcEi'ops humilis. Diagram ^ . 



