1916] LEVINE—CHLORANTHY IN DROSERA 391 
followed by another leaf rudiment opposite the first, and the next 
leaves follow in succession, forming a rosette-like structure in 
the axils of the leaves first formed. As against Miss Roprnson, 
KIRSCHLEGER and WINKLER, SALISBURY maintains that sections 
through these buds show that they are connected with the fibro- 
vascular bundle of the leaf on which they appear. Salisbury 
believes that these buds form a definite reproductive mechanism 
in these species of plants. 
GOEBEL (5) reports on D. pygmaea the appearance of a highly 
specialized type of brood body resembling in shape the gemmae 
of Marchantia. These bodies appear in rosettes and are dorsi- 
ventral. The under surface of each body is smooth, while the 
upper surface is horseshoe-shaped. Stomata and vascular bundles 
are present. GOEBEL believes that each brood body arises from a 
leaf anlage. 
PLANCHON (16) described an anomaly in the flowers of D. inter- 
media. He asserts that the ovary becomes elongated and the 
carpels bear both tentacles and ovules. All gradations between an 
ovule and a tentacle were found. The petals of these flowers were 
_ likewise modified, being leaflike in nature and bearing tentacles. 
PLANCHON asserts that all organs become chloranthic except the 
calyx. FERNALD (3), while collecting bog plants, found a form of 
D. rotundifolia in which the inflorescence was subcapitate and 
consisted of a few flowers. The petals and sepals were greenish 
to crimson in color, and the ovary instead of producing a normal 
capsule formed a rosette of glandular foliage leaves. This tendency 
was noted in other floral organs. Leavitt (11) studied these 
flowers and states that they are quite sterile. FERNALD makes 
no mention of stamens. It seems that they are generally absent, 
and this constitutes the main difference between this form and the 
one described later. *The constancy of this abnormality in D. 
rotundifolia led FERNALD to make a new variety which he called 
D. rotundifolia comosa. Neither FERNALD nor LEAVITT gives 
figures of these modified flowers. FERNALD regards them as a 
means of vegetative propagation. 
In studying the development of pollen my attention was 
attracted by the appearance of a number of plants of D. intermedia. 
