204 Evans: Notes on genus Herberta 



The perianth of Herberta, when well developed, clearly supports 

 the theory that three floral leaves take part in the formation of this 

 organ (Plate 8, fig. 13). It is split for nearly half its length into six 

 slender laciniae, essentially like divisions of leaves, every alternate 

 split being a little deeper than the others. These deeper splits mark 

 the boundaries of the three coalescent leaves. In the undivided 

 portion of the perianth a cross section shows that six rounded 

 ridges are present, separated by six rounded grooves (Plate 8, 

 FIG. 14). Three alternate grooves are deeper than the others and 

 correspond with the three deeper splits at the mouth of the 

 perianth. The innermost bracts and bracteoles, each with its 

 two dorsal ridges separated by a groove, are closely appressed to 

 the ridges and grooves of the perianth. The laciniae of the 

 perianth bear numerous slime-papillae and often teeth. The slime 

 papillae occur not only along the margins of the laciniae toward 

 the base but also on the inner surface of the perianth. Here they 

 sometimes form short rows and sometimes show no regular ar- 

 rangement. The surface papillae are occasionally borne on the 

 tips of short teeth or along the margins of narrow and irregular 

 ridges. In some cases the perianth is not developed normally. 

 More than six laciniae, for example, may be present at the mouth, 

 or one of the splits may extend to the very base; sometimes both 

 of these abnormalities may be seen in a single perianth. An 

 increase in the number of laciniae is not surprising, since ordinary 

 vegetative leaves with three divisions sometimes occur, and the 

 deep split might plausibly be interpreted as a place where coa- 

 lescence had failed to take place. 



The wall of the capsule is described in its essential features by 

 Spruce. It is composed of from five to seven layers of cells 

 (Text fig. 9) and shows a thickness of 70-100 ijl. This is con- 

 siderably more than in most of the leafy Hepaticae studied by 

 Andreas.* In PlagiocJiila asplenioides (L.) Dumort., for example, 

 where the wall is composed of seven or eight layers, the thickness 

 is only 55 ix, while in Chiloscyphus polyanthus (L.) Corda, where 

 the wall is composed of five layers, the thickness is only 25 ijl. It 

 is less, however, in Herberta than in Pleurozia purpurea (Lightf.) 



* Ueber den Bau der Wand und die Oeffnungsweise des Lebermoossporogons. 

 Flora 86: 161-213. pi. 12 +f. 1-25. 1899. 



