Is “3 AVITIib

   

NOON

STUDIES OF INHERITANCE
OF SHAPE IN BEANS

THESIS FOR DEGREE OF M. S.
WILLIAM KIA-SHEN SIE.

RW s
 
«-
s
N\.-

F Ch 743 Va.
STUDIES ON INHERITANCE OF SHAPE IN BEANS,

 

Thesis for Degree of M. 8S.

William Kia-Shen Sie.

1917
THESIS
ACKNOWLEDGMENT.

Upon the completion of his graduate work in Plant Breed-
ing, the writer wishes to take this opportunity to express
his sincere thanks to Professor F. A. Spragg, Expert in
Plant Breeding, at the Michigan Agricultural College, for his
patient teaching and for his criticism of this thesis.

10856
Ji 2
CONTENTS.

Introduction

Literature

The Problem

Sources of Material

Method of Measurements

Statistical Method

Planting

Harvesting

Summary

Bibliography

Page

10

10

14

26

27
INTRODUCTION.

As the knowledge of heredity is one of the principal
developments of the agricultural plant breeding it is very
important to investigate the influence of characteristics
that are transmitted from generation to generation upon
plant breeding practice. At the present time prominent
investigators have and are doing considerable work upon the
foundations of modern plant and animal breedings. Their
investigations include quality, quantity, color, size, and
last, but not least, shape. Ever since the breeders have
given their attention to agricultural problems the inherite
ance of size and shape has been much neglected because the
investigator had many easier problems at hand, and because >
his interests have only lately been called to these fields.

In the fall of 1914, Mr. P. K. Fu began an investiga-
tion of the inheritance of size and shape in beans. He
found the problem too large for the season of 1915 and he
was not able to cover the scope of my investigation. I am
using his seed to lay the foundation for a study of the
inheritance of shape in beans. Perhaps another may extend

this investigation.
ade

LITERATURE.

The references to Drs. Emerson, Belling, and Johannsen
are copied from Mr. Futs thesis. To this I have added a ree
view of the literature on methods of studying shape, though
none of it is upon beans,

Professor E. A. Emerson did much work on the inheritance
of color in the seed coat of beans when he was connected with
the Nebraska Experiment Station. In his work with bean cross-
es, he found that all the racial crosses of beans produced,
show little variation in the first generation, but pronounced
variation in the second and third generations. Under selec-
tion, they appeared fairly well fixed in the fourth and fifth
generations. The characters of the two parents (atavistic
tendencies) were usually reproduced among the offspring of
the second or third generation though often the new tendencies
were noticeable. Characters different from the parent forms
were usually blends in the crosses or united unchanged in
mosaics of small or large pattern.

In the study of size and shape in beans, he made num-
erous crosses between Fillbasket Wax having long flat seeds,
Longfellow having long slender seeds, and Snowflake Navy hav-
ing small round seeds. He then determined the mean, the
coefficient of variability for each of their lengths, weights,
breadths and thicknesses. He observed that in the first gen-

eration, the mean and the coefficient of variability were not
eo 3-

materially greater than for parents, but in the second gen-
eration, individuals exhibited marked segregation of size
and shape. From this, he concluded that "Shape may be defin-
itely inherited. Observations of the second generation bean
seeds where the parents differ in size but not in shape indi-
cate that length and breadth are probably not inherited
independently of each other. Large round beans crossed with
small round ones do not give any long slender beans in the
second generation, but only large medium and small round ones.
On the other hand, when the parents differ in shape as well
as in size, intermediate and parental shape as well as inter-
mediate and parental dimensions occur in the second generation. "*
Mr. J. Belling, Assistant Botanist of the Florida Experi-
ment Station in an attempt to secure a hybrid that would com-
bine the thin unopening hull of the Velvet bean with the
Lyon Beans! smooth pods which do not have the objectional
irritating bristles, has also studied the standard deviation
and coefficient of variability of the length, breadth, thicke
ness and weight, also the correlation of length and breadth
and thickness of the F., crosses between the Lyon and Velvet
beans. He measured from 50 to 200 seeds of each of his 118
plants and found that they varied between 10.5 and 20.05 mm.
in length, and from 8.3 to 13.55 mm. in breadth. In his
study of the weights of these seeds, he found that they
varied from .5 to 1.9 gm. He then concluded that, "The

Close agreement of the length and breadth of the hybrid seeds
with those of the Lyon beans and of tne thickness that of the
Velvet may possibly be genetic or may be due to special con-
ditions of growth." However, he did not investigate the size
and shape of beans in general.

W. Johannsen worked with the weights of beans and found
his pure line theory. He weighed the seeds of a single vari-
ety of beans and planted them separately. They arranged them-
selves in a normal curve round the weight of greatest frequency
where the seeds from the individual plants were harvested
separately. The crop from each individual again could be
grouped according to their weights in normal curve round the
most frequent weight characteristic of each individual. Thus,
there was a rough correspondence between the modes for the
individual plants and the weights of the individual seeds
from which they sprang. The heavier strains on the whole
come from the heavier seeds and the lighter from the lighter
seeds. But when he selected the heavier and lighter seeds
from a single strain and planted them separately, he found
that the modal weights were approximately the same for the
produce of both the heavier and the lighter seeds. This
indicates that selection inside the strain raised from a
single seed eos not alter the modal weight, i. e., the pro-
duct of the two selections are the same genetically.

To sum up, it may be said that none of these investiga-
tors have told us what sizes are separately inherited, nor

the number of inherited factors involved.
=5e

Johannsen has shown that there are such factors, because
the progeny of homozygous beans belonging to slightly different
sizes maintain separate means, and do not regress to the mean
of all sizes of beans.

Emerson nas shown that length and width are not inher-
ited separately, but together as inheritance of sizes of the
sane shape. Variations in the inheritance of shape, he finds,
occur only when the parents differ in shape.

Groth (N. J. Report 1911) has made a large number of to-
mato crosses, for example, “Currant upon Ponderosa” and "Plum
upon Peach® and followed the offspring for several generations.
His measurements of shape are the quotient of the measurements
for length and width and are illustrated by the symbol L/W=C.
In his summary we find the following: "In the Fo the variation
in size and shape are caused by the interaction of size and
shape factors." "The dependence of the Fy frequency distribu-
tion on that of Fy is much greater than its dependence upon
the frequency distributions of the parents." A close study of
the pictures (in bulletin) of the parents and their Fo, somatic-
ally speaking, shows the segregation of the Fo into the grand-
parental types, and their combinations of characteristics.

Professor Halsted (NW. J. Report 1910) and his associates
made crosses in peppers and secured one of considerable inter-
est, namely: Black Nubian with Coral Gem. From this cross 42
plants were grown in F 5. Measurements of length and width
of leaves and fruits were given in an accompanying table, but

the authors did not enter into any statistical dessication of
-6-

this data. For this reason I have calculated the quotients
between the length and width of each leaf and fruit, and obe-
tained the average (mean) in each case. These determinations
for shape are compared with the parents. The means for shape
of the leaves and the fruits of FP, were 2.143 and 3.250 re-
spectively. In F, the leaves varied from 1.7 to 4.3. <A study
of the results shows that the F, segregation (and therefore
variation) is considerably greater than the F,, but an F3 is
needed to separate the segregating types from one another.

In the report of 1913 the same authors secured similar re-
sults from many varieties of F, peppers.

Fast (Genetics, 1916, pp. 164-176) has recently presented
some data on the length of Corolla in a cross between two vari-
eties of tobacco (Nicotiana Longiflora Cav.). Upon a close
study of the data it was found that the Fo population was twice
as variable as the Fy and reached the extremes maintained by
the grandparents.

Drs. Emerson and East (Nebr. Research Bulletin 42) made

three crosses of distinct varieties of maize, and in their
summary stated that "in every case the Fo fraternities were
more variable than the F, lots. In most cases the F5's complete-
ly bridged the gap between the parents and in one case the F,
range of variation was from practically the shortest ears of
the short-eared parent to beyond the longest ears of the long-
eared parent."

The results of the above experiments show that the ¥5 is
more variable than Fi and usually includes the extreme of the

grand-parental shape.
-Je
THE PROBLEM.

As can be seen from Mr. Fu's thesis, he concluded (page
28 of his thesis) that his data on the inheritance of shape
is defective. In the original crosses, the kidneys were dom-
inant over navies, and the extracted navies gave only navies,
yet several exceptions were observed in later segregations, in-
dicating no dominance. His material is being further investi-
gated to get greater light along this line, and also to discover
the factors involved in this inheritance,

SOURCES OF MATERIAL.

From among the bean accessions that have been gathered by
the Michigan Experiment Station, Mr. Fu has listed eighteen
original numbers and four natural crosses from which his material
was drawn. The story of the origin of these crosses and their
care up to fall of 1914 when he made the plant-selections enter-
ing into his work has already been fully told by him. He
gathered all of the plants ripening satisfactorily in 1915, gave
them plant numbers, threshed them individually and stored them
in envelops in tin boxes away from the mice.

As each envelop contained the weds from an individual plant,
there were as many envelops for each plat or progeny as there
were plants represented. In the searoh for heterzygous material
in regard to shape, i. e., for plats of 1915 containing plants,
Some having navy seed, some intermediate and some kidney seed,
the envelops were sorted into three piles, the one navy, another

intermediate, and a third kidney. It is true that this is a
=8=

rough classification but served a purpose of observation and
comparison. Those lots that fell only into one of these groups
were not usable.

When the 1915 plats had thus been examined it was discoverea
that all of the plats showing satisfactory segregation had come

from two of Mr. Fu's groups, namely, accession 61 (one of the

Nav
e ey

natural crosses.) This means that all of the other lines failed

18 accessions) and the cross (one of the four
to furnish data in this investigation.

From among these lots, the plats showing the largest number
of productive plants were chosen, and the highest producers of
each lot were taken, because if we are to get reliable statis-
tical constants for each lot, we must have a reasonably large
number. However, certain envelops were discarded because of
the presence of too much disease, and in the chosen lots the
diseased individual beans were also discarded. Thus the most
productive and the healthiest beans were chosen from among

those lots that gave the most promising segregations.

METHOD OF MEASURELENTS.

Taking a piece of cross-section paper divided into half
inch squares and each square into tenths, the distances be-
tween the lines (1/20 inch) were taken as units of mf asurement.
The length and width and the quotient or ratio between these
measurements were obtained for each bean. This ratio repre-
sents the shape of the beans and is often indicated by the
symbol L/W =8 (or Length/Width=Shape.) The smaller the quotient,
= 9

the rounder the beans, and the greater the whotient, the
longer the beans. Because the berns were small it was
necessary to take a small unit of measure, in order to get
@ reasonably correct ratio between length and width. Thus
after some consideration the one-twentieth inch was chosen
as a convenient measure.

A question of how to measure naturally arose. The
measurements were taken by means of the aboe mentioned cross-
section paper, which was fixed upon a table. Besides this,
there were two thin straight narrow sticks with smooth faces
nailed (perpendicularly to one another) along the lines of
the cross-section paper from which the measurements were
counted; that is, one stick was placed close to a line of the
cross-section paper longitudinally and another was placed in
contact with the longitudinal stick close to a line of the
cross-section paper transversely. Then a bean was measured
by being placed in the wooden square in contact with the
sticks, and uvon the paper, the number of longitudinal lines be-
ing length, and a number of transversal lines being width. This
method for measurement was used in all work.

It was usually found that the beans did not measure exact
units. Fractions of more than half a division were considered
a whole division, and if less than half a division were dis-
carded. Judging the fractions and counting the lines needed
careful observation.

The measurements of the beans were tabulated in three
columns, i e., length, width and shape; however, the lengths and

widths were recorded before the ratios between them were obtair<«2?
-10-
STATISTICAL METHOD.

The beans were next classified according to their shapes,
placed on sheets for the study of variation and calculations
for statistical constants were executed for each envelop (that
is, the seeds obtained from one plant). The calculations were
for mean, standard deviation and coefficient of variability of

shape e
PLANTING.

The seeds were planted during the first week of June, when
the weather became quite warm. It was planted on a piece of
sandy soil south of the orchard near the east line of the College
farm and not far from the river. The lay of the land was as
flat as could be found. The land was well worked. Under these
conditions the plants had nearly an equal opportunity to develop.

About 200 plants had been carefully chosen from the 1915
materia}. Each plant had a selection number as well as a plat
number recorded on a stake which was set up at the beginning
of each row. The system was that in regular use by the Michigan
Experiment Station and consisted of the following items: the
register number, the selection number and the accession number.
These were written in the above order on a card which was
paraffined and tacked to a stake.

We feared to use a drill because the lots of beans were
small and varied so greatly in size as not to insure equally

distant planting. The desire was that the seed may be so planted
that each plant may have the same amount of space. The land
was marked off, that the rows may be straight and equal dis-
tance, and that the recently mellowed soil may be grooved.
The seeds were dropped in these grooves, setting up the stake
first at the beginning of the lot of beans and leaving gap
of two or three feet before planting another lot of beans.
The envelops had the same number as the stakes. The
stakes were always set up before the seeds were planted, by
taking care to check up these numbers before each planting.
I made sure that no lots of beans were misplaced. The in-
dividual plats were planted in the same order as the selection
number given. The plats from 60100 te 614200 were progenies
belonging to work planned by Prof. F. A. Spragg. The plats
from 614300 to 630000 were those belonging to my thesis.
TABLE A.
Data on Shape of Seed Planted.
(Only productive plats are listed.)

 

No. of plants

 

 

Obv- Hare
Plat Plant cone Statistical Data tained vested
1916 1915 sion “N. M. GC. summer fall
614700 54021 B 131 1.4324.008 9.49%.39 7% ~—«-20
615300 53205 * 40 1.508+.012 7.86+.59 32 8
615700 «53211 +" 162 1.676+.005 6.14+.23 119 22
617300 514214 61 29 -:1.1194.009 6.66+.59 23 16
619400 514516 61 24 1.106+.009 6.05+.59 22 13
619900 514618 61 18 1.6014.019 7.714.86 13 7
620000 52630 $F 85 1.4364.009 9.044+.47 72 12
620400 52605 * 65 1.3954.001 9.74+.57 51 15
621800 53817 * 51 =: 1.3514.001 8.98+.59 41 15
624300 54322 * 151 =:1.1514+.003 4.25+.16 95 71
627800 513002 * 67 1.428+.006 5.81+.34 41 14
629200 513109 * 58 = =.1.2154+.007 7.20+.45 26 16
629300 513114 * 45 1.241+.009 7.98+.56 30 17
629400 513102 * 169 = 1. 390+.004 6.30+.23 111 31
629500 513105 * 67 1.3514.008 7.97+.46 32 13
629600 513120 " 71 = =1.2714.008 8.31+.47 37 28
-13-

A week or so after the seeds were planted the weather be-
came suddenly cool and then it rained most of the remainder of
the month. The germination was checked and also the growth.
Approzimately 25 percent of the seeds thus failed to produce
thrifty plants. This includes some sickly plants that were
pulled out to prevent the spread of diseases.

During the growing season the plants were given personal
care, being hoed and cultivated a number of times. When the
plants were about six inches high the number of thrifty plants
belonging to each plat were counted and recorded. The differ-
ence between this number and the number of the seeds that were
planted showed how many failed in germination or became sickly.
The remainder of the plants grew vigorously and were attractive
looking, although they had passed through an unfavorable period.

In the first part of the growing season the field was soaked
with water all the time and the sun seldom paid them a visit.

This favored disease, but the sicklyones were pulled out, as men-
tioned above, in order to check the spread of disease.

Following the wet and cool June came a hot and dry July. The
temperature climbed to 102.2 degrees daily during the blosseming
period. Besides the heat, it was very dry, not giving two inches
rainfall in the whole latter part of the season. Notes were taken
in regard to the color of the flowers during the blossoming period,
and recorded in the bean register. During the very hot days the
flowers dropped off; the leaves turned yellow; a few undersized
pods formed without beans; and the growth of some plants was stunt-
ed. However, this was the experience in nearly all the fields in
Michigan and some other states, producing scarcity and the highest
price in history.
-14.
HARVESTING.

Harvesting began September 20th. The plants were collect-
ed as soon as they were ripe. The foregoing description dis-
courages hove for a good harvest. The harvesting of a few
beans here and there was more trouble than harvesting plants
full of beans. In some plats the plants yielded no beans at
all. In other plats some plants gave one or two pods, thus
veing easily missed in collection.

Beginning with the first row the plats were examined in
order. Plants with ripe pods were pulled, leaving those with
green pods, and discarding sterile plants to save trouble with
them later.

After pulling, the ripe plants belonging to each plat were
tied in a bundle and labeled, indicating the plat number. The
bundles were hung upon wires in the seed room in the order of
the plat numbers. When dry, more notes were taken. Most of
the green plants were frozen, and shriveled in the field, as it
was too late to mature.

In the laboratory each plant was given a selection number
beginning with one for each plat in accordance with the sta-
tion's system, and the following notes were taken: the height
of plant, the condition of disease, the number of pods, the
length of pods, and the number of seeds. These were registered
in the bean selection book. The seeds produced on each plant
were threshed by hands, counted, and put in an envelop, writing
selection number on the envelop to indicate the year planted,
the olat number and the selection within the plat. The en-

velopes were arranged in the order of their numbers in tin boxes
-15-

to protect them from the mice. Thus when the work was done, all
could be easily turned to.

In threshing it was found that some of the seeds were not
naturally ripened. Such deformed or undersized seeds were dis-
earded, and not counted. The seeds having mature shape were
the only ones that were saved. It was necessary to eliminate
the shriveled seeds in order to get accurate results, as de-
formed seeds would only deceive us in the study of inheritance
of shape.

The plats were now studied systematically and grouped ac-
cording to whether they had 10 to 24 seeds; 25 to 49 seeds; or
50 seeds and more. The plants producing fewer seeds were dis-
carded. Most of the groups were at a disadvantage because of
the limited number in their populations, but a few plants in
each of the 15 plats produced quite well, and two plats were

nearly normal in production.
Distribution of Individual Plants.

-16.

TABLE B.

(Based on mean of shape)

 

No. of

plants

vested

seed _ 1.1 #1.2 $21.3 +#+21644 1.5 #+44.6 #+41.67 1.8

614700 20 4 12 4
615300 8 1 3
615700 22 2 8 3
617300 16 3 2 6 2 3
619400 13 3 4 4 2
61.9900 7 1 2 3 1
620000 12 8 1
620400 15 2 6 3 2
621800 15 1 1 6 5 1 1
624300 71 5 9 11 15 18 6 2 5
627800 14 1 2 4 4 2 1
629200 16 3 5 5
629300 17 2 3 5 2
629400 31 1 2 5 12 3 5 2 1
629500 17 2 2 2 4 2
629600 28 4 14 4 2 4
-17-

There were 158 plats at planting time. Only sixteen of these
produced enough seeds to give reliable results. These seeds were
measured just as their mothers were, taking the length and width
and then finding the ratio ¥ or shape. This determination is
the measure of shape.

Table B gives the distribution of shape for the plants of
each of the sixteen productive plats. The plants are classified
according to the mean of shape, the shape being determined for
each seed. Considerable variation may be noted among the plats,

some having a much greater variability than others.

TABLE C.

1.1 1.2 1.3 #4164 #.+11.5 1.6 1.7 1.8

 

619400 3 +4 4 2
629200
629300 2 3

TABLE D.
617300 3 6 2 3
620000 3 1
629500 2 2 2 4 1
629600 4 14 4 2 4
-18-

 

TABLE E.
1.2 #1462 1.3 1.64 15 21.66 1.67 1.48
614700 4 12 4
615300 1 3 3 1
615700 2 9 8 3
619900 1 2 3 1
620400 2 2 6 3 2
621800 1 1 6 5 1 1
627800 1 2 4 4 2 1
TABLE F.
624300 5 9 11 15 18 6 2 5
629400 5 12 3 5 2 1
-19-

Tables C, D, E, and F give different groups showing
similar variabilities and indicate that each group has nearly
a constant center of variation. A comparison of these groups
shows that there is fairly definite segregation. Some groups
have a limited range of variation, while others are extremely
variable, probably due to a heterozygous condition.

As the two plats 624300 and 629400 (shown in Table F)
are the most variable and also the most productive of the
whole series, they are studied much more in detail The seeds
that planted plat 624300 (as can be seen in Table A) was the
Least variable and almost the roundest beans that were listed;
and the seeds that planted plat 629400 was about average in
size and variability. Yet the offspring of these plats vary
from very round (rounder than the ordinary navy) to excellent
kidney types. It appears that the individual plant has a con-
trolling influence on the shape of its seeds, the seed on a

plant being reasonably constant.
-20-
TABLE G.

624300
Division l.
Sel. No. 1.9 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0

62 30 0 32
63 20 Oo 38
66 34 0 37

Division 2.

3 5 1
13 3 3 4
20 5

28 7 oO 1
46 11 60OoO 1
55 51 1 38

Division 3.

4 1 6 2 2 1
6 1 1 0 2
10 2 1 1
11 3 3 4 3
12 3 1 1 7
14 4 3 2 3
15 1 3 2 1
16 4 oO 3 1
17 6 6 3 5
21 2 23 2 4 3
23 7 13 #5 10
=2l-
Division 4.

Selec. No. 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0

 

24 1 24 Oo 17 3
32 14 21 2
61 1 14 #=0. 23

Division 5.
59 29 0 26
67 1 0 18 1 24 4

Division 6.
42 6 0 2
497 2 2 0 12

Division Te

37 3 o 30 0 5

43 6 0 44 0

44 3 0 20 0 2 1
49 5 0 64 0 12

51 3 0 22 0 1

53 4 0 40 0 12 1
57 27 0 32 0 1

69 21 0 41 0 1

70 39 0 40 oO 1
o22-

Division 8.

 

Selec. Now 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0
’ 2 14 8 1 5 1
8 5 16 28 7 8
9 6 1 2 0 1

18 2 1 10 2 5
22 2 3° «4215 4 6 2
33 1 18 2g 1 44
35 6 11 38 7 37
36 2 17 34 5 36
38 5 14 47 #22 23 4
40 1 8 53 0 35
45 1 3 2 19 3 4
52 3 29 16 6 6
54 1 27 56 6 3 1
56 1 20 58 1 3 1
64 26 9 21 0 1
65 1 10 69: 3 5 1. 1
71 2 26 54 5 4 1
2 1 3 7 3 3 2
60 1 16 45 2 0 4
-23-

Division 9.

1.0 lel 1.2 163 164 165 166 1.67 1.8 1.9 2.0

 

Selec. No.
5 1 7 2 7 0 1

26 0 2 0 2

58 2 1 22 2 22 2

09 29 0 26 1
Division 10.

19 3 1 18 15 #20

29 0 4 8 oOo 4

39 7 1 27 18 0 13

68 12 2 25 6 0 1
Division ll.

25 1 5 8 4 1 0 1

31 23 0 69 0 23 0 6

34 3 7 57 +O 12 +O 2

27 5 9 5 3 0 3

48 30 37 «7 22 0~CtOl

50 18 = =36 2 29 1 3
Division 12.

41 3 4 10 19 38 29 o 28
-24-
TABLE H.

629400
Division l.

Selec. Now. 1.0 1.1 1-2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.9

 

17 8 0 4
19 2 1 8 1
20 1 1 3 1

Division 2.

2 2 7 1

14 1 6 0 1

18 4 1 2
21 3

24 3 4 16 2 14
29 2 0 1

31 1 6 0
13 2 0 1

Division 3.

5 3 1
23 1 3
25 2 0 3 1
12 1 0 3
=25-
Division 4

Selec. Now 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0

 

1 1 5 0 1 1
6 0 3 1
7 2 0 4
8 6 5 5 2 2 1
9 2 2 0 0 1
10 2 0 4 0 1
11 2 0 0 1 1
16 3 #211 0 2 1 1
22 5 0 1 0 1
26 1 0 1 1
27 1 1 0 2 1
Division 5.
28 1 2 7 0 6 2 1 1
30 1 6 6 0 2 0 1
Division 6,
3 2 0 1 1 0 1 0 1
1 0 0 1 0

15 3 0 6 6 0
-2be

Tables G and H give the distribution of shape as determined
for the individual seeds for each plant. Table G gives the data
for plat 624300 and Table H for plat 629400. At the left of
these tables will be found the selection numbers within the plats.
The selection numbers are not in their numerical order, but are
grouped so as to bring tegether the data for the plants having
similar variability. These groups again indicate segregation of
shape and variability. Some of the groups have narrow ranges of
variation while others have a wide range of variation. The
mothers have dominated the shape in the past generation. This
should be studied another year to follow the segregation of the
next generation; see how powerful the mother plants are in dom-
inating shape, and find out how far it is possible to fix types.
In other words, is there a certain basal probably environmental

variability that is always present?

SUMMARY.

All plats show variability and some also show segregation,
but variability does not definitely prove segregation.

There appears to be a dominance of shape held by theparent.

Certain shapes seem to be quite generally used as centers of
variation, indicating segregation in certain populations, These
are 1.2, 1.4, and 1.6. Perhaps there are other centers.

Plats 624300 and 629400 show a range from 1.1 to 1.8. These
plats are extremely heterozygous, and are studied more fully in

detail. Both of these plats come from the cross Wea Raney °
-27.
BIBLIOGRAPHY,

l. Belling, J., 1911.
a. Second Generation of the Cross Between Velvet and
Lyon Beans.
Florida Station Report, p. 82.
obo. The Mode of Inheritance of Semi-Sterility in the offe-
spring of Certain Hybrid Plants.
Florida Agricultural Experiment Station.
2. Davenport, C. B., 1912.
The Interpretation of Deminance and Some of Its
Consequences,
American Naturalist, p. 128.
3. Emerson, R. A.
a. 1902, Preliminary Account of Variation in Bean.
Annual Report of Nebraska, p. 30.
be 1904, Heredity in Bean Hybrids.
Annual Report of Nebraska, pe 33.
Ce 1909, Inheritance of Color in the Seeds of Common
Bean, Phaseolus Vulgaris.
Twenty-second Report of Nebraska.
d. 1909, Production of White Bean Lacking the Factor for
Total Pigmentation.
Procedure of American Breeder's Association, Vol.6,p. 396.
e. 1910, Inheritance of Size and Shape in Plants.
American Naturalist, p. 738.
f. The Inheritance of Recurring Somatic Variation in Vari-
egated Fars of Maize. .

Research Bulletin of Nebraska, No. 4.
4.

9.

LO.

=2§«

Bast, E..M., 1910.
a. A Mendelian Interpretation of Variation that is Ap-
parently Continuous.
American Naturalist, p. 65.
b. Genetics, January, 1916.
c. Genetics, July, 1916.
Bast, BE. M., and Emerson, R. A., 1913.
The Inheritance of Quantitative Characters in Maize.
Nebraska Research Bulletin No. 2.
Bast, B. M., and Hayes, H. Ke,
a. 1911, Inheritance in Maize.
Bulletin 167, Connecticut Station.
be 1915, Further Experiments of Inheritance in Maize.
Connecticut Bulletin 188.
Groth, B. H. Alfred, 1914.
The Fy Heredity of Size, Shape and Member in Tomato
Leaves.
New Jersey Bulletin No. 238, 239.
Halsted, D.,
Experiments with Peppers.
New Jersey Reports 1909, 1910, 1911, 1912, 1913, 1914.
Halstead, B. Ae,
Heredity and Correlation of Structures in Tomatoes.
New Jersey Reports 1911, 1912, 1913, 1914.
Hayes, H. K.,
Correlation and Inheritance in Nicotiana Tabacun.

Connecticut Expt. Station, Bulletin 171, 19l2.
ll.

12.

13.

14.

15.

16.

17.

Johannsen, W., 1909.
Elements of Exact Study of Heredity.
Review in Nature. (London.)
Jennings, H. S&.,
Genetics, Sept., 1916.
Little, C. C., Dec. 18, 1914.
A Possible Mendelian Explanation for a Type of In-
heritance Apparently Non-Mendelian in Nature.
Science Weekly.
Owen, Earle J., 1910.
Inheritance Studies with Beans.
New Jersey Report, pe 277.
Weldon, W. Fo. R., 1902.
Mendel's Law of Alternative Inheritance.
Biometrika, No. 2, pe 228.

The Fy Heredity of Size, Shape and Number in Totato
Leaves.

New Jersey Agri. Expt. Station; Bulletins 238, 239, 242.

Some Results in Size Inheritance.

New Jersey Agri. Exp. Station; Bulletin No. 278.

-Q=
 

 
 
iain

 

Fal tt