78 
James Small. 
serrate fashion, then the pappus is described as setose serrulate , 
(Fig. 17, C). If the lateral cilia are as long as or slightly longer 
than the diameter of the seta, the pappus is described as setose 
barbellate,( Fig. 17, D). This form leads on by the elongation of the 
lateral cilia to the setose plumose type, (Fig. 17, E), where the 
projections are considerably longer than the diameter of the seta. 
Setae of these various types occur in large numbers, in one or 
two or more rings on the top of the cypsela, and there is no 
commoner modification than the fusion of the setae at the base, to 
form a more or less continuous membranous collar of tissue. This 
modification is quite in harmony with the tendency to reduction 
and cohesion present in other floral members. It is duly noted by 
Hoffman (34) in his non-committal description of the various 
modifications of the pappus. It is included in many descriptions 
of species, such as Vernonia Kerrii (13), and is frequently a character 
of generic rank. It-occurs in the sub-genus of Senecio, Senecillis 
(19). It was noted in Tussilago by the writer(61) and occurs in a 
very large number of genera. 
This being the case it is not surprising to find that sometimes 
the collar is discontinuous and broken up to give a number of scales, 
paleae or squamellae. The setae frequently vary in length, so that 
fusion gives a form described as paleaceo-setose (Fig. 17, F). The 
paleaceous part may be comparatively short or relatively long on 
the same fruit. The addition of one or more setae to the central 
rib or the further thickening of the single central seta gives a 
similar type, which is usually ciliate at the edges, as in Gaillardia, 
and is known as aristato-paleaceous, (Fig. 17, G). By a slight 
elongation and a crowding together of the lateral cilia, as in 
Centaurea calocephala (22, Fig. 220. p. 545), we get a serrulato- 
paleaceous form (Fig. 17, H). If the constituent setae are of equal 
length and fuse completely a simple paleaceous type (Fig. 17, 1) 
results. 
It may be rather surprising that this simple fact of the 
trichome nature of the paleae has not been more widely recognised. 
Anyone can mount the simple paleae of Berkheya or Tagetes and see 
with oblique illumination under the low power of a compound 
microscope that the facts are as stated. The free ends of the 
constituent hairs can be seen distinctly and the ribbed surface is 
obviously formed by the fusion of the constituent setae. 
If the ends of the setae are free to a slight degree the result is 
a lacerate margin, giving the fimbriato-paleaceous type, (Fig. 17, L). 
