D0 
flowered variety with a pink-flowered variety, and illustrating 
Mendel’s law of segregation. The two parents and, between 
them, the first hybrid generation plants were shown in the first 
row. The other two rows contained plants of the second hybrid 
generation, in the kinds and proportions of each theoretically ex- 
pected when two pairs of Mendelian factors are involved. Thus 
in this cross, the two parents are yellow (YYrr) and rose pink 
(yyR?R?). The first hybrid generation plants have rose red 
flowers. The plants of the second hybrid generation are repre- 
sented in the bed in the ratio in which they occur, by 1 deep crim- 
son: 2 rhodamine purple: 2 scarlet red: 4 rose red: 1 vellow (like 
the parent flower color): 2 light yellow: 1 rose pink (like the 
male parent flower color): 2 light pink: 1 white. The colors, of 
course, refer to the flowers. 
Four-o'clocks are especially fine material for illustrating Men- 
delian laws of heredity for a number of reasons. They are com- 
paratively free from disease, easily grown, attractive in appear- 
ance, continuous bloomers, and they already have interesting as- 
sociations for the general public, since they are flowers of the 
old-fashioned flower gardens and since their flowers open and 
close according to certain environmental conditions. 
Furthermore, the tuberous roots live over winter easily when 
placed in a little sand in a dry, cool cellar, and farther south they 
live over out-of-doors. There are records of the roots living for 
over twenty-five years. The dome or mound-like habit of growth 
of the plants lends itself to orderly arrangement without distract- 
ing supports. Hence, the exhibit plot, appropriately labeled, can 
be arranged almost diagrammatically—the parents, F, (first filial), 
I, (second filial) generations, etc. In some regions such an 
exhibit bed could be enclosed with a low hedge of dwarf box or 
of other suitable materials. The plants are crossed easily. The 
flower colors are distinct and striking, and the heterozygotes 
(hybrid plants that do not breed true) of every genetic type 
are distinguishable by the beginner and the layman. Last of all, 
the great range of flower colors has a simple genetic basis—five 
factors in all, excluding the striping pattern—these being Y, R, 
R°, y, andr. Our exhibit proved quite effective; though it Shon 
ee ited more sun. The plants for this exhibit came from Mr. 
— 
