64 NATUIiAL HISTORY OF PLANTS. 



regularity or irregularity of their form ;' the absolute number of 

 them in each verticil, or pseudo- verticil, and the number of these 

 verticils themselves ;- the aspect of the anther ;* the number of 

 carpels, and the number of ovules in each ; the direction of these 

 ovules* and of the seeds ; the consistency of the pericarp.* 



To LiNN.EUs is due the first foundation of this family. Bernard 

 UE JussiED, in his arrangement of the garden of tlie Trianon," only 

 borrowed from the author of the " Frnynirntd Bofanica'' and A. L. 

 DE Jdssieu' simply reproduced the work of his uncle, adding to his 

 Ranunculi the genus Podophijllum, whose right to a place in this group 

 has been much disputed, besides four small genera of little im- 

 portance.^ Adanson has been accused of having destroyed the homo- 

 geneity of this group in his great work," by adding the greater number 

 of the Alismacece. We have said elsewhere,'" and we repeat it, that this 

 course appears to us thoroughly rational. As arranged in Jussieu's 

 " Genera Planiarum' the BaunnculacecB include" twenty-three genera, 

 studied by A. L. de Jussieu himself in several detached memoirs,'- 

 and after him by most of his successors, with peculiar predilection 

 and attention, as representing on the whole a group of vegetables 

 fittest to afibrd types for the most important principles of taxonomy." 



for example, that in the two sections established terior. We shall see that if the DiUeniacea hud 



in the genus Caltha, the one has persistent, the a descending ovule it would be like that of Oi/- 



other caducous sepals, &c. (See ^rfrtH*o»j«, iv. 36). lianthemum ; but when the ovules are few in 



' Describers have usually confused the irregu- number, they arc ascending, 

 larity of the corolla, and that of the individual ^ We have not attachetl much importance to 



sepals or petals. Sepals of very strange form, this character. Can a thoroughly ripe Almond 



helmet-shaped or spurred, may be very regular. with its pericarp quite dry, be separated gene- 



On the other hand spurless sepals may be truly rically from a Teach with its mesocarp succulent ? 



irregular, their halves being unsymmetrical {op. We think not, and we would here recal the in- 



cit., iv. 9). stance of the Adonids where the fruit, to-day a 



^ The proof that the number of verticils has dru])e, will bo to-morrow an achene. We have 



no importance is the facility with which in the been unable to found generic divisions on this 



Paoniu, Aiumoiie, &.C., we pass from one quin- character. 



cuncial verticil to two alternately trimerous vcr- " In A. L. de JrssiEr, Oe»., Ixviii. 



ticils (see pp. 37, 42, 61). ' Genera Planlarum, sec. ord. nat. diitpot. 



'" Till the time of I)F. (Jandolle it was be- (178!»), 231. 

 lievcd tliat the Ilnnunciilacffr had their anthers '* Jli/dra.tfi.f. llaniadnfiis, Xanthorh'ua, Cimi- 



generally extrorse, and the Dilleniacece had them eifiiqti. 



introrse. A. UK St. Hii.aiuk was the first to * Ftimilles des Plantes {\7(i'Ji), u. \^\. 



rectify this eiTor (see Adaiis„ni't, iv. 14, note). '" Adansonin, iv. U>. 



There are far more Hanunciilarefp with introrso " Not counting Pohtphyllum, with which we 



anthers than has been generally allowed. Nii/elln, are not now concerned. 



Delphinium, Eranlhix, &c., described as ii.iving " Chielly in that which he points out himself 



extrorse anthers, have them decidedly introrse. (op. r;7., 235) : "Apia flrnenim flifnis ttume- 



* Callianthemum is the oidy Uanunculad with rosi.i affinium conjunHio at' disitusitio jam in 



a pendulous ovule and the micropyle exterior. Art. Acad. Paris. 1773, slaluta." 

 Otherwise every ascending ovule has itnmicroi)ylo " We have seen that on tlic other hand sevc- 



cxterior, and every descending ovule has it in- nil authors consider this family as being of a low 



