266 
Hill and de Frame . — A Consideration of the 
stored in a small seed, and of the seed itself, is the same as for a larger seed 
of the same species. In fact, the most satisfactory method of attacking the 
problem is to ascertain the amount of water required, not only for the 
ordinary activities of the living tissues, but also for the hydrolysis and 
sufficient dilution of the food materials present, and to consider this in 
relation to the tracheae and the sieve- tubes. Also, the time factor and the 
temperature at which germination takes place must be taken into account 
if a complete investigation is to be made, for it is obvious that if hydrolysis 
takes place slowly, the capacity for water-carriage need not be so great as 
when hydrolysis occurs quickly ; and, with regard to temperature, it need 
hardly be mentioned that the degree of heat considerably influences this 
process. In short, since germination is intimately bound up with physico- 
chemical reactions, it is obvious that any conditions influencing these must 
be severally investigated before their resultant can be ascertained. 
It may not at first appear obvious how questions relating to transloca- 
tion are connected with the number of bundles. Briefly put, the connexion 
appears to be this : Confining our attention to water-supply, it is obvious 
that in order that efficiency may be maintained a vascular strand must 
convey a certain amount of water to the part which it is its function to 
supply. The xylem of a bundle does not pour out water at its tip, like 
a fountain, but supplies the surrounding tissues throughout its whole course. 
If the xylem be so compact that there is little or no parenchyma between 
the tracheae, it is obvious that, no matter how numerous the tracheae may 
be, their capacity for supply is limited according to the extent of the outer 
surface of the xylem mass ; so that, in order to obtain an increased supply 
of water, it is necessary in the first instance not to increase the number of 
tracheae, but to increase the amount of surface in contact with the tissues 
to be supplied. This, obviously, may be done by the branching of the 
strand, and possibly this may in part explain the larger number of 
cotyledonary bundles in plants possessed of large seeds, e. g. Cycads, 
Quercus and Ricinus , and also the presence of two bundles where one might 
be expected. For example, in Ephedra, although the cotyledons are not at 
all massive they may be very long, and are characterized by the possession 
of two vascular strands which remain separate throughout their course 
within the seed leaves. The xylem of these bundles is, however, very 
compact. 
The same end may be attained by different means : it is obvious that 
if, instead of dividing, the bundle opens out tangentially, as it were, by the 
development of much parenchyma between the tracheae, the surface of the 
wood would be greatly increased ; this is seen in Cephalotaxus , Cupressus , 
Juniperus , and also in Persoonia , where the seed-leaf bundles at the 
cotyledonary node are much elongated tangentially, whereas in Pinus there 
may be two bundles at the corresponding level. 
