GENETICS: PEARL AND SURFACE 
223 
of possibilities. By means of Mendelian modes of research considerable 
progress has been made towards answering this question for many kinds 
of quaUtative variations. For fluctuating or quantitative variations 
also, progress has been made through the researches especially of East 
and his associates, following those of Nilsson-Ehle on colors determined 
by multiple factors. 
What we have tried to do in the present investigation is, by studying 
the growth of the individual, to analyze the adult variation curve into 
its component elements. We have attempted, in other words, to make 
a beginning at an understanding of the developmental physiology of the 
genes concerned in the production of the characters studied. 
A given individual, at an early stage of growth, may exhibit an ex- 
ceptional condition of a character, as compared with other individuals 
of the same age or stage of growth. It may be, for example, very short, 
the shortest plant in the whole population at three weeks of age. Will 
this same individual be the shortest plant in the adult population, after 
all growth is completed? If not, where is it in the adult variation curve, 
and how did it get there? This example will give a concrete idea of the 
general mode of approach to the problem of variation followed by this 
paper. 
Results. Measurements were made at twice-a-week intervals of the 
height of each individual of three series of plants of a variety of sweet 
corn used in other experimental work in this laboratory and described in 
earlier papers. The heights were measured to the tip of the tallest leaf. 
In addition to these, separate sets of measurements were also made of 
the tassel height as soon as the tassels appeared. The growth curves 
obtained by plotting the mean height at each measurement are rela- 
tively smooth. After July 3, the time of tasseling, the plants grow in 
height much faster than before. Growth in height ceases entirely as 
soon as the tassel blooms. 
The relative variability considered for the whole season shows a 
marked progressive diminution. It thus follows the general growth 
law of diminishing variability. Considered in detail, however, the rela- 
tive variability first shows an increase. After June 19 there is a rapid 
decrease until the time of tasseling. During the period of tasseling 
there is a very rapid increase in the relative variability. This is fol- 
lowed by an equally rapid decrease. A stable condition is finally reached 
which is some 10 per cent lower than the variability at the beginning of 
the season. 
The quantitative relations of tasseHng to the growth and variability 
of the height of com plants are very similar to the quantitative relations 
