l 30 James Small. 
when they occur are very thin, usually deciduous and not at all like 
bracts. Bentham (I, 7, p. 482) agrees that the Cichorieae are 
probably the most recent tribe, so that the sporadic appearance of 
paleaceous bracts is strong evidence of their atavistic nature. 
Calendulece. Practically without exception the receptacle here 
has no appendages. 
Arctotidece. This tribe has the alveolate type well-developed 
the margin of the alveole may be entire or variously divided. In 
Ursinia the alveole is divided almost or quite to the base so that 
each achene is surrounded by a number of narrow paleae, free or 
fused in a ring at the base. Three small genera show the naked or 
foveolate types, with, in Haplocarpha, a tendency to develop the 
fimbrillate form. The Gundeliinae are markedly setiferous, so that 
the relative positions of the sub-tribes and the primitive position 
of Ursinia are confirmed by the characters of the receptacle. 
Anthemidece. The only distinction, and it is admittedly artificial, 
between the Anthemidinae and Chrysanthemidinae is the presence 
of receptacular paleae in the former and their absence in the latter. 
The latter sub-tribe has, as a rule, no appendages on the receptacle, 
but foveoles. fimbrillae, setae, alveoles and even a few short paleae 
occur in some of the genera, so that even this artificial distinction 
breaks down in a few cases. Reclassification on a natural basis is 
required. 
Inulece. The receptacle here has no appendages as a rule, but 
paleae occur in a few species in almost every sub-tribe and the other 
types occur so sporadically that few phyletic data can be obtained. 
It may be noted, however, that Hdichrysum is naked, foveolate or 
alveolate and Gnaphalium naked or foveolate, so that the primitive 
receptacle occurs in the chief genera of the primitive sub-tribe. 
The only other point of interest is the predominance of the 
setiferous type in the Buphthalminae where every genus shows an 
approach to this character, some with paleaceous bracts which 
surround the florets, others with the true setiferous form, cp. 
Cynareae below. Bentham’s remarks (I, 8, p. 337 sqq.) on two of 
these genera are of interest, especially as he considered the Inuleae 
to have no real affinity with the Cynareae (see Chap. I, Fig. 2) 
Geigeria, he says, has the receptacle and corolla of the Cynareae; 
Rhanterium is stated to have the habit of Centaurea with the excep¬ 
tion of the ray florets. 
Cynarece. The receptacle in all but three genera is setiferous, 
the setae being slender, or flat and more or less paleaceous. The 
exceptions are Berardia and Wariouia, removed to the Mutisieae by 
Hoffmann, and Onopordon in the Carduinae, where the receptacle 
has foveoles or shallow alveoles with the margin more or less 
lacerate. This is so exceptional in the tribe that it is better regarded 
