SOLUBLE, NITROGENOUS CONSTITUENTS OF PLANTS 2/7 
other plants and organs, e.g. pipecolic acid in the green bean and y-methyleneglutamine 
in tulip bulbs (cf. refs. 29, 30). 
From this beginning, large numbers of discrete compounds have been detected on 
paper chromatograms and their chromatographic characteristics have been recorded 
on “maps” (STEWARD, ZACHARIUS AND POLLARD), Moreover, as methods of con- 
verting the free keto acids to their amino acid analogues were adopted, the number 
of these keto acids to be recognized increased greatly, far beyond those that were 
known from their presence as Krebs-cycle intermediates®®. As more plants and families 
have been examined, the number of these unidentified substances has become very 
large indeed, but concomitantly an ever expanding number have been isolated and crit- 
ically identified. The surprise is that so many of the recently discovered substances 
are of low molecular weight and that their presence was totally unexpected in plants, 
even in common food plants. 
Systematic surveys of plant families by the newer methods are few. One such, on 
the the Liliaceae, was made with special reference to seven recently discovered nitro- 
genous compounds (pipecolic acid, hydroxypipecolic acid, y-methyleneglutamine 
and its acid, y-methylglutamic acid, y-methyl-y-hydroxyglutamic acid, y-amino- 
butyric acid and azetidinecarboxylic acid), all of which were known to be present in 
some liliaceous plants. Although the distribution of these compounds in the family 
was very variable, some like y-methyleneglutamine being confined to a few species, 
the family also yielded 45 unidentified substances (FOWDEN AND STEWARD"). Also, 
the members of the Leguminoseae, conspicuous because of their role in the discovery 
of the substituted piperidines and the y-substituted glutamyl compounds are also 
found to contain many other compounds which have yet to be identified as shown 
on the “map” of compounds recognized in this laboratory by ZACHARIUS** (1952) and 
later by BARRALES? (1959, see Fig. 2), or the maps published from VIRTANEN’s labora- 
tory”? (MIETTINEN, 1955). 
Precision in the mapping of the positions of compounds as they occur on chromatog- 
rams is achieved by rigorous control of the variables that apply during the chromatog- 
raphy (especially temperature); the use of pure solvents and precise mixtures; and 
at least one of the now somewhat standard combinations such as phenol saturated 
at pH 5.5, followed by collidine—lutidine (x : 3) saturated with water, or by butanol- 
acetic acid—water (9 : I : 2.9). Even so, it is better to refer the position of a substance 
to that of an internal standard like alanine (R. alanine) than to the solvent front 
(R;-). Position alone, even when recorded in several solvents, is only a guide to the 
identity of a component. The final identification requires supplementary techniques, 
and there is no general substitute for the isolation and characterization of a pure 
substance, followed by synthesis and critical matching with the unknown. Where 
two asymmetric centers exist in a given compound (for examples see GROBBELAAR, 
POLLARD AND STEWARD!*), the task is not completed until the ultimate stereoiso- 
meric configuration is known (cf. ref. 14 and the contribution of WrniTz to this Sym- 
posium). 
A few selected sequences of compounds recognized or identified in this way will 
be mentioned in order to emphasize certain general conclusions which can now be 
drawn. 
References p. 42 
