92 Trigonantlieae 



I cannot close this memoir without some mention of j)i'evious 

 attempts at a tribal arrangement o± the plants I have been discussing. 

 Dumortier's tribes are mostly founded on such vague notions of structure 

 and affinity, and show such eccentric combinations of genera, that their 

 adoption becomes impossible. Thus, his Chiloscyphe^e comprises — 

 besides Chiloscyphus and Coleochila Dum. {=Mylia Gray) — Pleuroschima 

 Dum. [^Bazzania Gvsij), Odontoschisma, Lejndozia ; while Cephalozia is 

 relegated to his tribe JungermaniecB, along with Jungermania, Lophocolea, 

 &c.! 



Lindberg, the latest systematiser of Hepatic®, has proposed a tribe 

 Lepidozie^*, whose first half, comprising Lejndozia, Bazzania, Odonto- 

 schisma, and Cephalozia, is a natural group of genera ; but the second 

 half : Lophocolea, Pedinoj^hyllum, Chiloscyphus, and Harpanthus, belongs 

 to a distinct tribe, differing from the former in habit, ramification, and 

 especially in the perianth being laterally (and not frontally) compressed ; 

 as I have already shewn at greater length {ante, p. p. 2 — 5). 



Nees's TRicnoMANOiDEffl (Hep. Eur. I, III et IV, and Syn. Hep. p. p. 

 XIX et 197), consisting of Calypogeia N. (^Kantia Gray), Lepidozia, 

 Mastigobryum, Micropterygium and Fhysiotinm, is really (^when the last 

 genus, Fhysiotium ,'i\ is eliminated) a natural group, if it can only be 

 proved that his Calyjjogeia (=Kantia Gray) is a marsupial extension of 

 Mastigobryum (= Bazzania Gray). For it is probable that there is not in 

 Nature any separate tribe of pouch-fruited Juiigermaniacece (= Marsupio- 

 carpese = Geocalycese = Saccogynese^), but that almost every tribe may 

 have a genus (ot genera) of marsupial species, and that, where none 

 such is known to exist, it is either because it has hitherto eluded our 

 search, or has succumbed to other plants in the struggle for place, or 

 has not yet been evolved. The transitional stage, between suprater- 

 raneous and subterraneovis perianths, is to be found in those genera 

 whose floral whorls are more or more or less adnate to each other into 



* "Hepaticse in Hibernia lectae," in 'Acta Soc. Sc. Fennica' a. 1875, p. 539. 



IT This curious genus, the Pleurozia of Dumortier, has not yet found its true place in the system. 

 With Radula and Madotheca, where it is sometimes placed, it has little real affinity. In the form and 

 structure of the perianth, and its included organs, there is very great similarity to Jungermania § 

 Afiastrophyllmn. Even the blood-red foliage is a frequent feature in both groups. Yet in the inser- 

 tion and structure of their leaves they are (apparently) so very difterent as to preclude the idea of 

 their juxtaposition in the same tiibe. 



