Ranales , Rhoeadctles, and Rosales. 
697 
in the root ; while in Type 2 the number of collateral strands in the upper 
part of the hypocotyl is twice the number of phloem or xylem strands in 
the root. 
Thus the nature of the ‘active’ strand (xylem or phloem, or both) 
and the relation in number of stem to root strands are the distinguishing 
characters between the types. The determination of these characters, as 
well as the observation of ‘ rotation ’, is bound up with the identification of 
the first-formed protoxylem elements. Chauveaud 1 maintains that the 
phenomenon of ‘ rotation ’ is an illusion produced by early obliteration of 
protoxylem in the upper part of the seedling, or, as he terms it, ‘ acceleration 
basifuge ’. However this may be, much confusion has arisen through 
failure to identify the first-formed protoxylem elements with consequent 
variations in description of transition phenomena, due largely to difference 
in age of seedlings examined. 
Two changes due to age may be noted now : (a) obliteration of primary 
tissues, one effect of which is to make a single structure appear to be two 
(see Scott 2 ) ; ( b ) production of secondary tissues, one effect of which is to 
unify separate strands. As these changes go on together, the net result at 
any one age will depend upon the relative rates of change of these pheno- 
mena. Difference in the level at which the cotyledons were examined has 
probably also played a considerable part in the divergence of accounts. 
It is sufficient now to draw attention to one or two concrete instances of 
discrepancy. Thus Dr. Scott 2 describes the cotyledons of the Wallflower 
as having two bundles, while the precisely similar vascular system met with 
in the cotyledons of many Piperales, 3 Centrospermae, 4 Cactaceae, 5 and 
Tubiflorae, 6 is treated as single by the respective authors, who in conse- 
quence regard these as cases of Van Tieghem’s Type 3, while Dr. Scott’s 
description would make them examples of Type 2. 
The practically identical transitional phenomena seen in Dipsacus is 
instanced by Van Tieghem as an example of Type 1. Phaseolus and 
Medicago furnish Van Tieghem with examples of Type 2 and Type 3, 
respectively, though the course of events is very similar in both. 
My own observations on Dicotyledons, begun in 1902, very soon 
convinced me that there is a common ground plan, if not absolutely for all, 
for the majority of dicotyledonous species. 
(< a ) The feature of greatest constancy is the existence of two root poles 
1 Loc. cit., 1911. 
2 Scott, D. H. : Structural Botany, 7th ed., Part I, p. 78. 
3 Hill, T. G. : On the Seedling Structure of Certain Piperales. Ann. Bot., vol. xx, 1906. 
4 Hill, T. G., and de Fraine, E. : On the Seedling Structure of Certain Centrospermae. Ann. 
Bot., vol. xxvi, 1912. 
5 De Fraine, E. : Seedling Structure of Certain Cactaceae. Ann. Bot., vol. xxiv, 1910. 
6 Lee, E. : Seedling Anatomy of Certain Sympetalae. Pt. I, Ann. Bot., vol. xxvi, 1912. 
Pt. II, id., vol. xxviii, 1914. 
