8 
F. F. Blackman 
the condensed hexose units. The lowest molecular weight for starch 
is 15,000, which indicates the condensation together of at least 90 
hexose units and so suggests many more possible stereo-isomers than 
there are starch-bearing plants. 
Reichert’s first investigation, published in 1913, was directed to 
enquiring if the starch-grains of all species of plants are demonstrably 
different. For the 300 species examined the answer is in the affirma¬ 
tive. A few weeks ago there reached this country a second publica¬ 
tion containing a minute study of the starch-grains of parents and 
crosses. These two investigations together fill four huge quarto 
volumes for which the Carnegie Institution has subsidized the publica¬ 
tion of 800 photo-micrographs of starch-grains, 1100 graphic charts 
of their reactions, as well as some 1700 pages of text. 
For each plant, the isolated starch-grains are examined for some 
twelve characteristics; four points of visible form, two of appearance 
in polarized light, two of staining with aniline dyes, several of re¬ 
actions with iodine and other agents and finally the solution- 
temperature or rate at which the grains lose their solid form when 
heated with water, which temperature ranges from 49 0 to 90° Centi¬ 
grade. No two of these characteristics seem to be closely correlated 
and Reichert treats them all as ‘unit characters.’ 
Graphic charts are constructed by giving a numerical value to 
the degree of reaction of a given starch to each of these agents, so 
that the eye can take in, as a whole, all the different behaviours of 
each starch in the form of a curve. No great accuracy is claimed for 
the detailed values, as the whole work is entitled a preliminary 
exploration. The results certainly support the thesis that no two 
species or cultural forms have quite identical starches, and that in 
crosses the starches of the cross show a varying behaviour, but 
incline more towards the characters of the seed-parent. 
From his review of genera and species, the author concludes that 
nearly always the starches of species of one genus resemble each other 
more than they resemble species of other genera. Genera, too, often 
show similarity within one family. Indeed, it is interesting to find 
how closely, on the whole, the starch-characters accord with the 
subdivisions of classification accepted for flowering plants. Starch- 
evidence has definite contributions to make to the doubtful relation¬ 
ships of certain genera, and indicates that some families are homo¬ 
geneous and others heterogeneous. 
No individual details are given as to the part of the plant the 
starch-grains were taken from, but the general statement is made 
