Primitive A ngiospemiks . 1 7 7 
in a similar way, and some of these species are aquatic. But sheathing 
petioles are also found in terrestrial Dicotyledons : for instance, in many 
Umbellifers. 
In Monocotyledons the connexion is clear between the formation of 
a sheathing petiole and the suppression of stem internodes. Both characters 
are correlated with the formation of a large bud on a squat subterranean 
axis — a character so common in this class. The bud is commonly dormant 
throughout the winter, and sends up leaves and flowers in the spring. In 
such an axis the internodes are disk-shaped, and for the adequate insertion 
of radical leaves a broad base is necessary. In a large underground bud, 
moreover, the protection of one leaf by the expanded base of that outside 
it is very convenient. 
The aquatic Dicotyledons with sheathing petioles are forms such as 
Nymphaea , having rhizomes which creep in the mud at the bottom of the 
pond and send green shoots upwards into the water. The green shoots die 
down in the winter like the aerial shoots of geophytes. Moreover the 
terrestrial Dicotyledons with sheathing petioles are mainly geophytes ; that 
is, species with underground stems which throw up aerial shoots each spring 
to die down in the winter. 
In short, the character which all these Dicotyledons with sheathing 
petioles have in common is not the aquatic but the geophilous habit. 
Aquatics, such as Nymphaea, which have a permanent axis beneath the mud 
at the bottom of the ponds where they grow, have some characters in 
common with geophytes. 
(iii) Stem anatomy . 
Similar criticism applies to Professor Henslows remarks on the anatomy 
of the stem. He explains with great clearness how a number of broad-based 
leaves inserted on a squat axis introduce more traces into that axis than 
can be accommodated in a single circle. Hence the formation of several 
concentric circles of traces. Secondary thickening is not usually required 
in such a stem. The intrafascicular cambium soon disappears, particularly 
when starch and other food-stuffs are packed in the conjunctive tissue which 
separates the bundles. But traces of cambium within each bundle are 
commonly found in geophilous Dicotyledons. 
But though this explanation may — and probably does — account for the 
scattered arrangement of bundles in the rhizomes of many aquatic Dicoty- 
ledons, the character belongs exclusively to the permanent axis rooted in 
the soil beneath the water. The vascular tissue of the axes which grow 
upwards into the water is always very much reduced in quantity. The stele 
of completely submerged stems commonly becomes a slender cylinder in 
which the leaf-traces lose their identity. 
Thus the likeness between the stem anatomy of aquatic Dicotyledons 
and that characteristic of Monocotyledons in general is confined to the 
