440 PROCEEDINGS OF THE AMERICAN ACADEMY. 



This may be its real affinity, but it is to be noted that Phihppi repre- 

 sents the style-branches as being recoiled through more than 360°. 

 This would be highly exceptional in the Eujxitorieae. Furthermore, 

 the style-branches as shown in Philippi's figures are unusually short 

 for the tribe. Until material of this rare and local plant can be ob- 

 tained and subjected to further study the species would better be left 

 here, where Bentham placed it, but it would seem almost as likely to 

 prove one of the Helianthcae, perhaps near Isocarpha. 



As explained elsewhere in this paper, the rudimentary annular 

 or coroniform "pappus," accredited in the past to several species 

 hitherto placed in Agcratum, proves not to be a true pappus, i. e. a 

 calycular structure exterior to the corolla, but only an annulus upon 

 which the corolla itself is borne and which after the disarticulation and 

 fall of the corolla sometimes is slightly accrescent. These species, 

 destitute of a true pappus, must certainly be transferred to Alomia 

 if that genus is to be kept distinct from Agcratum and they are so 

 treated in the following revision. 



While in nearly all of the species concerned the presence or absence 

 of a pappus is upon careful observation sufficiently evident and con- 

 stant to permit a pretty ready separation of species into those which 

 should be referred to Alomia on the one hand and those which would 

 better be placed in Agcratum on the other, there are two exceptional 

 species, Agcratum littorale Gray and A. maritimimi HBK., in which 

 the distinction breaks down absolutely. Here the pappus may be 

 entirely wanting, it may consist of minute teeth, very short and slightly 

 exterior to the corolla, or finally it may develop into a perfectly definite 

 and conspicuous scale-pappus. These marked variations in pappus 

 occur in individuals of precisely similar habit, and so far as can be 

 ascertained are accompanied by no concomitant changes of structure. 

 This wide intra-specific variation presents, of course, a technical 

 difficulty in delimiting the genera Ageratum and Alomia. However, 

 it is to be remembered that calvous forms in normally pappus-bearing 

 Compositae are by no means rare and must be accepted as one of the 

 inherent difficulties of the group. It would be highly artificial to 

 transfer to Alornia the calvous forms of Agcratum littorale and A. 

 maritivium, nor does it seem best to unite with the otherwise consis- 

 tently pappus-bearing genus Ageratum the consistently calvous genus 

 Alomia because in certain exceptional species an abortion of the pappus 

 occurs inconstantly, as it does also in Trichogonia menthacfolia Gardn., 

 Calca pcthmcularis HBK., and various other Compositae in which 

 the loss of pappus can in no sense be regarded as having generic 

 significance. 



