Physiological Studies in Plant Anatomy 123 
the fact that if root tips are soaked in an alcoholic solution of carotin, 
the yellow pigment is held by the wall behind the growing apex in 
the region of the future endodermis. Carotin is a hydrocarbon 
offering no other possibility of chemical reactiveness than the un¬ 
saturated CH = CH linkage. 
(h) Basal Substances of the Casparian Strip. 
No conclusive evidence exists as to the nature of these sub¬ 
stances, but we agree with Kroemer that cellulose does not enter 
into their composition. If the impregnating substances are removed 
by oxidising agents such as eau de javelle, or by boiling alkali, then 
the subsequent behaviour of the section in iodine reagents shows 
clearly the absence of any cellulose matrix from the Casparian strip. 
The basal substance present is far more resistant to concentrated 
sulphuric acid and less resistant to oxidising agents like eau de javelle 
than the cellulose membranes surrounding it. It is quite possible to 
obtain isolated networks of Casparian strip by submitting the section 
or tissue, first to prolonged boiling in strong potash and then, after 
washing, transferring to concentrated sulphuric acid. Unfortunately 
endodermal tissue is usually entangled with xylem and cuticle or 
exodermis, and the isolation of the Casparian strip for more thorough 
investigation is difficult. It can be completely isolated in the way 
described above in the case of Potamogeton and therefore in time 
macro-chemical evidence of its nature may be forthcoming. 
Preliminary observations have supplied the facts that nitrogen 
is present (about 4 per cent.) and also sulphur. (We have to thank 
Mr H. D. Kay of the Department of Physiology, University of 
Leeds, for his help with the macro-chemical work.) These results, to¬ 
gether with the great resistance of this substance to sulphuric acid, 
show clearly that we are dealing with something which is quite 
definitely not cellulose nor a pure carbohydrate, but its actual 
nature is as yet only conjectural. We would, however, emphasise the 
facts that the Casparian strip is formed in the root at a very early 
stage, and that its formation probably involves the deposition of 
the impregnating substances in a wall which has not yet lost the 
characters of the walls of merismatic tissue, which do not at this 
stage readily give cellulose reactions. The nature of these walls is 
described somewhat more fully in the next paper of this series, where 
it is shown that the cell wall retains the embryonic character for a 
longer time at the root apex than in the stem apex of the plant. 
Provisionally we would emphasise the significance of the fact in 
explaining the constant occurrence of a Casparian strip and a primary 
