16 
R. H. Compton. 
transition phenomena according to the earlier writers are sum¬ 
marised in column 1 of the figure which represents the Third Type. 
The xylem bundles remain in place, twisting through 180" as they 
ascend the axis; and the halves of the lateral phloems fuse radially 
outside the xylem. The chief merits of this view of the transition 
are that the vascular bundles are regarded as continuous from root 
to cotyledons, and that it furnishes a convenient method of 
description for comparative purposes. For this latter reason it 
has been almost universally adopted in England, despite many 
and obvious discrepancies with the facts of seedling structure. The 
most important fact ignored by this theory of Van Tieghem, Ge'rard, 
&c., is the common occurrence of a median protoxylem group between 
the main xylem bundles of the middle region of the transition. The 
error is not so serious in the simple case illustrated ; but in a 
hypocotyl such as that of Lupinus spp., in which the lateral xylem 
bundles are widely separated from one another, the median proto¬ 
xylem cannot be overlooked. In such a case we are justified in 
speaking of the xylem bundles of the hypocotyl as forming triads, 
each consisting of a median polar “ protoxylem ” and a pair of 
lateral “ metaxylems.” Miss Thomas, while recognising the presence 
of the median protoxylem, yet speaks of the “double-bundle” as an 
anatomical unit: and the median and lateral xylems are regarded 
as constituting a single complex group. The work of Chauveaud 
throws much light on the nature of the median protoxylem, without, 
however, surmounting all the difficulties : and Miss Thomas’ view 
can perhaps more easily be brought into accord with these than 
can the schemes of other writers. This subject must be considered 
later. 
The views of Bonnier shew certain analogies with those of 
Ge'rard and Van Tieghem, and may be considered next. The 
scheme of transition according to Van Tieghem is not merely one 
of rotation : the position of the pole of xylem-differentiation changes 
relatively to the centre of the axis, the endodermis, &c. In the 
root the protoxylem lies towards the exterior of the central 
cylinder—a position of physiological value in relation to water 
absorption, according to Bonnier. As we ascend the axis the pole 
of xylem differentiation passes inwards, and is eventually separated 
from the pericycle by the whole thickness of primary xylem and 
phloem. Taking account of this gradual displacement by the 
xylem pole, Bonnier considered that the “central cylinder presents 
the same general plan of construction in both root and stem.” He 
