Chapter XIII — 219 — Utricularia, Biovularia 



are produced which are evidently partly shoot in structure, the dis- 

 tinction between leaf and shoot being here vague. The apex of the 

 dome never develops {21 — 1-5)- 



In U. exoleta and U. emarginata the embryo is laterally compressed 

 so that the growth pole is on the edge. Here two leaves are produced. 

 These may occasionally become shoots {21 — 7D). Near their bases, 

 but displaced to one side are usually two buds, sometimes a third. 

 These become dorsiventral shoots. 



Aberrancies of development are not at all unusual. For example, 

 one embryo of U. vulgaris produced only two leaves and two shoots 

 which became at once dorsiventral "water shoots." 



In U. nelmnbifolia, otherwise similar to U. vulgaris, the primary 

 leaves are widely forked (Merl) and those of U. reniformis are broadly 

 spatulate (Goebel). According to Goebel, the embryo of U. mon- 

 tana has no primordia before germination, and on it there are pro- 

 duced two primordia (the cotyledonoids) of which one becomes a 

 spatulate leaf, the other a trap, on the side of the "vegetation point" 

 which grows directly into a radial structure from which stolons etc. 

 grow. The interpretation here is to be questioned, as it is doubt- 

 ful if Goebel's identification of the vegetation point is correct. 



The third type of germination is seen in U. purpurea. The embryo 

 is ovate, with a flattened broad end. From this there normally arise, 

 in succession, three shoot buds standing in a triangle with respect to 

 each other. The first is dominant in growth; the second to appear 

 may form at first a close second, but at length lags behind in growth; 

 while the third may not do more than put in an appearance, and then 

 fail to develop (Text fig. 6, 1-7). In one case a fourth shoot bud 

 appeared early (Text fig. 6, 4 and 5), but its subsequent development 

 showed it to be a branch at the base of the first bud, and in other 

 seedlings it appeared later, but constantly. The development of the 

 first shoot has been followed until it produced five whorls of branches 

 (leaves in the taxonomic texts). The earlier two whorls have but two 

 members bearing traps, placed symmetrically attached to the upper 

 moiety of the stem. On the upper surface of the stem as a member of 

 the whorl may appear a bud of unlimited growth, destined in its turn 

 to produce whorls of secondary members (Text fig. 6, 7). In the 

 third to fifth whorls, two additional members, placed in the wide dorsal 

 space, appear. These, however, do not bear traps. In the fourth 

 whorl, in one case observed, one of the ventral pair produced a branch 

 which, however, bore no trap. In one instance a seedling was found in 

 which the first and second buds were fasciated (Text fig. 6, 6). It is 

 evident that in this type the seedling at once takes on the morphologi- 

 cal character of the mature form, which, however, has whorls of four 

 {U. elephas) or six {U. purpurea) branched members. 



It would be most interesting to know the course of germination of 

 U. tubidata, but of this we are yet ignorant. 



Types of Utricularia. — In the following account, the various spe- 

 cies, so far as they are here included, are arranged according to the 

 character of their traps. 



I . The Utricularia vulgaris type. — All the plants of this type have 

 traps which are closely similar in the mechanical details of structure 



