SEEDLING STRUCTURE IN THE LEGUMINOS.E. 49 
primary xylem plates are continuous and the third lies separate ; but in the 
older state a solid xylem star is produced, with 120° between each protoxylem. 
The phloem contains groups of stout fibres. 
This root-structure persists for about 17 mm. up the hypocotyl without 
change. 5 mm. below the cotyledons parenchyma begins to appear in the 
centre of the xylem star, and just above this two of the xylem plates begin to 
move apart and to become opposite to one another, the third protoxylem 
remaining in its former position. At the same time the phloem which lies in 
the enlarging angle between the (median cotyledonary) protoxylems extends 
tangentially and splits into two bundles, so that four phloem bundles are now 
present. 
About 2 mm. below the cotyledonary node the third (lateral) xylem divides 
into two parts, and each half attaches itself to the polar xylem plate, which is 
now becoming wedge-shaped. The polar protoxylem becomes very feeble 
and the metaxylem rounds up into two loose groups of spiral vessels at each 
end of the ellipse. On the inner edge of the two groups of metaxylem on 
one side of the ellipse there are small crushed protoxylem elements derived 
from the third root-pole. Similar elements now appear on the other side ; so 
that just below the cotyledons the structure appears to have been derived 
from a tetrarch root. 
Each cotyledon takes the two bundles of metaxylem with vestiges of the 
polar protoxylem between, and with the “lateral” half-protoxylems on their 
adaxial side. These latter are not so much crushed as the polar protoxylems. 
Two phloems, obliquely dorsal to the metaxylems, pass out to each cotyledon. 
This condition persists until the lamina is entered. 
"here is a short cotyledonary tube: no axillary buds are produced. 
The plumule contains tannin-sacs, whereas the cotyledons and hypo- 
cotyledonary axis are without them. 
The fact that the “lateral” protoxylems are much less crushed than the 
polar seems to indieate that they are differentiated slightly later ; and this is 
supported by the separation of the third xylem from the cther two in the 
young root: but the study of the early differentiation was not made. 
It is noteworthy that the fourth xylem pole is not present as such, but is 
only represented by a pair of protoxylems corresponding to “ half-laterals."' 
HIPPOCREPIS UNISILIQUOSUS, Linn. 
The cotyledons are linear and long. 
The root is triarch and no fourth protoxylem appears. But just above 
the level where the third protoxylem divides in half (2 mm. below the node) 
two other small groups of vessels appear on the opposite side of the ellipse ; 
their appearance is exactly similar to that of the halves of the third xylem, 
and they behave in the same way ; there is no doubt that they represent the 
LINN. JOURN.—BOTANY, VOL. XLI. E 
