Campanales.] 



ASTERACE^. 



'03 



CCCCLXXIV. 



ing Orders, also \\nth the flowers in heads, are readily distinguished hv their 



ovule, and the anthers being either wholly or partially distinct. 



In proportion to its strict natural 1 



limits, depending upon the unifor- 

 mity of its characters, is the diffi- 

 culty of separating it into sections. 

 Jussieu has three : Corymbifera;, 

 the florets of which ai"e flosculous 

 in the disk, and ligulate at the cir- 

 cumference ; Cichoracea;, the florets 

 of which are all ligulate ; and Cy- 

 narocephalse, all whose florets arc 

 flosculous ; to which has since been 

 added another division called Bila- 

 biate. Linnaeus employed the sexes 

 of the florets for the purpose of 

 defining groups, but this, like all 

 other parts of the great Swedish 

 Naturahst's Botanical System, is 



now abandoned ; and yet it was not without much merit. 



The condition of the Order had at one time, — thanks to the 



neglect of Lrnnean Botanists and the unmethodical improve- 

 ments of more careful observers, — become a chaos, the like of 



which had not been seen since the days of the Baul.ins ; but 



in 1830 an arrangement of much merit was proposed by the 



German Botanist Lessing, and at a later period De Candolle 



the elder appUed his acute and logical mind to the elucidation 



of the Order. At the present day the method of the latter, 



essentially founded on that of Lessmg, is universally followed. 



De Candolle himself stated it thus : — 



Suborder I. Tubuliflor^ ; that is to say, those in which 

 the hermaphrodite florets, which alone can be regarded as 

 normal, are tubular, mth 5, or rarely 4, equal teeth. Of 

 these the following are distinguished by their stigmas : — 



1. Vernoniaceee, Style cylmdrical, its arms generally 



long and subulate, occasionally short and blmit, 

 always covered all over with bristles. 



2. Eupatoriaceae. Style cylindrical ; its anns long and 



clavate, with a papillose surface on the outside near 

 the end. 



3. Asteroidese. Style cyhndrical ; its amis linear, flat on 



the outside, equally and finally downy on the inside. 



4. Senecioidese. Style cylindrical ; its arms linear, 



fringed at the point, generally truncate, but some- 

 times extended beyond the fringe into a cone or 

 appendage of some sort. 



5. Cynareae. Style tliickened upwards, and often fringed 



at the tumour. 



Suborder II. Labiatiflor^ ; that is to say, those in which 

 the hennaphrodite florets, or at least the unisexual ones, 

 are divided into two Ups. Of these the following are dis- 

 tinguished by their stigmas : — 



6. Mutisiaceae. Style cylindrical or somewhat tumid; its arms usually blunt, or 



ti'uncate, very convex on the outside, and covered at the upper part by a fine 

 uniform hairiness, or absolutely bald. 



7. Nassau viacece. Style never tumid ; the branches long, linear, truncate, fringed 



only at the point. 



Suborder III. Liguliflor^; that is to say, those whose corollas are sht or ligulate, viz. 



8. Cichoraceae. 



Fig. CCCCLXXIV.— Involucre, receptacle, and half fruit of Amoseris pusilla.— G'^rr/n. 



Fig. CCCCLXXV.— Stigmas of Composites, illustrating De Candolle 's tribes (after Ileyland). 1 . Alber- 

 tinia erythropappa ( Vemoniacea?) ; 2. AnisochJEta mikanioides ( Eupatoriacenc) ; ^^. Hlumea senecioidea 

 (AsteroideEe); 4. Mendezia bicolor(Senecionid€a?); 5. Lipochietaumbellata (Senecionidea?) ; «5. Aplotaxis 

 nepalensis (Cynarea;) ; 7. Leucomeris spectabilis (Mutisiaceae) ; 8. Leuceria tenuis (Nassauviacese.) 



Fig. CCCCLXXV. 



