GROUP [I. BRYOPHYTA: HEPATIC^ 1 . 345 



The normal mode of branching in the dorsiventral forms is that which takes 

 place at the growing-point in the plane of expansion. In the thalloid forms, 

 as also in the foliose Fossombronia and Blasia, it may be described as dichoto- 

 mous (see p. 184) although the apical cell does not undergo division so as to 

 form the apical cells of two branches ; the apical cell of the parent shoot per- 

 sists, and that of the branch is developed from an adjacent segment, either 

 before or after further division. When the two shoots develope with equal 

 vigour, the resulting branch-system resembles a dichotomy; but when the 

 parent shoot grows the more vigorously throughout, the branches are lateral 

 upon it and the branch-system is a monopodium (see p. 34). In the foliose 

 forms the mode of normal branching is generally monopodial. The apical 

 cell of a lateral branch is developed from the lower (ventral) half of a dorso- 

 lateral segment cut off from the apical cell ; either from the whole of the seg- 

 ment, or from the posterior (basiscopicj portion of it. 



In the radial Haplomitrium, the branching is monopodial and the branches 

 are borne on all sides of the stem, either in the leafy region or on the subter- 

 ranean parts. Ttiey are not developed from the apical growing-point, and are 

 probably all adventitious. 



A development of branches on the ventral surface is not uncommon either in 

 the thalloid (e.g. Metzgeria, Symphogyna, Umbraculum on the midrib) or in 

 the foliose forms (e.g. Mastigobryurn, Lepidozia, Calypogeia, Jungermannia 

 [Cephahzia] bicuspidata, Lophncolea bidentata). Since the origin of these 

 branches can in some cases be distinctly traced (e.g. Mastigobryum, Sympho- 

 gyna) to the apical growing point, and since they are generally developed in 

 acropetal succession and in definite positions (in the axils of the amphigastria 

 of the foliose forms), they are regarded as normal and not as adventitious. 

 The branch-rudiment does not, however, always immediately develope, but 

 may be quiescent for a longer or shorter time. Their origin is frequently en- 

 dogenous. These ventral branches are generally more or less modified sexual 

 branches (gametophores) ; but in Mastigobryum sterile branches occur which 

 are long and delicate, with rudimentary leaves, and are known as flagella. 

 Similar flagella occur in other foliose genera, such as Lepidozia and Calypogeia ; 

 they are not, however, ventral branches, but modified lateral branches. 



Adventitious branches are often formed on the older parts of the shoot ; in 

 Metzgeria, from marginal cells of the lamina ; in Pellia and Sphaerocarpus, 

 from single cells of all parts of the dorsal, and sometimes also of the ventral 

 surface; in Jungermannia bicuspidata and Blasia, fro'n cells of the ventral sur- 

 face of the stem : in Lophocolea bidentata, from the marginal cells of the 

 leaves. Their origin is generally exogenous, but sometimes endogenous. 



In many cases (e.g. Blasia, Sphserocarpus, LopJwcolea bidentata, Junger- 

 mannia bicuspidata) the adventitious branches become separated from the 

 parent plant, and develope into new individuals, thus subserving vegetative 

 propagation. 



Closely connected with the adventitious branching is the formation of gemmce. 

 In Aueura, certain cells of the margin and of the dorsal surface of the shoot 

 each become divided into two, and the two cells thus formed are set free as a 

 bicellular gemma, with probably a proper wall of its own, by the rupture of the 

 enclosing wall. In Blasia, the gemmae, which are solid multicellular nearly 



