UNIVERSITY OF CALIFORNIA PUBLICATIONS

IN

AGRICULTURAL SCIENCES

Vol. 2, No. 5, pp. 191-204, plates 36-28 October 19, 1920

INTERSPECIFIC HYBRIDS IN CREPIS

I. CREPIS CAPILLARIS (L.) WALLR. X C. TECTORUM L.

BY

ERNEST B. BABCOCK and JULIUS L. COLLINS

Although the problem of the mechanism of heredity may be said
to have been solved by Morgan and others1 by means of the genetic
analysis of a species of flies, Drosopltila melanogaster, there yet
remains the highly important question regarding the generality of
the conclusions based on the Drosophila data. As has been pointed
out by Morgan,2 no method of determining the specific relation of
individual chromosomes to particular somatic characters appears more
promising than the study of hybrids between species.- Especially is
this the case when the species possess low chromosome numbers. If
such species can be subjected to extensive genetic analysis so that the
factorial composition of each pair of chromosomes may be described
with some degree of exactness, and if fertile hybrids between such
species can be obtained so as to permit of breeding as well as of cyto-
logical investigations of the hybrid progeny, we should find here the
most promising material with which to test the generality of the
chromosome theory of heredity.

Inasmuch as several species of Crepis were known to possess low
chromosome numbers, considerable attention has been given by the
writers3 to the genetic investigation of two of these species, viz.,
Crepis capillaris (vircns)'1 and C. tectorum L., the results of which

i Morgan, T. H., Sturtevant, A. II., Muller, H. J., and Bridges, C. B., The
Mechanism of Mendelian Heredity. New York, Holt, xiii + 262 pp., frontispiece,
and unnumbered diagrams. 1915.

2 Morgan, T. H., The Physical Basis of Heredity. Philadelphia, Lippincott,
pp. 1-305, 117 illustrations. 1919.

3 Babcock, E. B., "Crepis — a promising genus for genetic investigations."
Amer. Nat., vol. 54 (1920), pp. 270-276.

* Britten, James, and Rendle, A. B., "Notes on the 'List of British Seed
Plants.' " Jour. Bot., vol. 45 (1907), p. 102.

192 University of California Publications in Agricultural Sciences [Vol. 2

will be reported later. But before confining our attention too largely
to one or two species only, it was deemed advisable to look into the
possibility of obtaining fertile hybrids between these species.

Although the species in question show differences in many morpho-
logical characters and in at least one physiological character, only
those having to do with the seedling stage will be considered here,
due to the fact that the hybrids in all cases died during this stage.

The achenes of Crepis capillaris range in length between 2.0 and
2.5 mm. ; they have no beak, and the pappus sheds rather easily. The
cotyledons vary from broadly ovate to the condition where the breadth
of the widest part is greater than the length, in which case the tip of
the cotyledon is distinctly dentate. The first plumule leaf makes its
appearance very quickly after the cotyledons have expanded to their
normal size. The cotyledons are approximately 5 mm. wide and
4 to 6 mm. long (pi. 36, fig. 2).

The achenes of Crepis tectorum range from 3.5 to 4 mm. in length
and are correspondingly thicker than the capillaris achenes. The
tectorum achenes are also beakless but they retain their pappus more
persistently than do those of the other species. The cotyledons are
distinctly different in both shape and size. The general shape is nar-
rowly linear with bluntly pointed ends. Length of cotyledons varies
around 6 mm., the width around 3 mm. As in capillaris, the plumule
leaves appear very promptly, usually one at a time, but occasionally
in both species two plumule leaves appear simultaneously. Thus there
is evident a distinct difference in size and shape of the first or cotyledon
leaves of the two species corresponding with the difference in size of
achenes, and a resemblance in the prompt appearance of the plumule
leaves (pi. 36, fig. 1).

Crepis tectorum possesses one more pair of chromosomes than C.
capillaris, the former having four pairs, the latter, three pairs (pi. 38,
fig. 2).

Two methods of pollination were employed, which will be described
in detail in another paper: (1) Emasculation of female parent flowers;
(2) female parent not emasculated but washed free of its own pollen
by use of a fine jet of water. The second method was used when the
capillaris plant had given indications from selfing tests of being self-
sterile. The results of cultures thus secured are indicated below. The
female parent is mentioned first in each cross.

nii'o]

Babcoclc-Collins : Crepis capillaris X C. tectorum

193

Culture

Num-
ber

Parents

Cotyledon

Characters of

Seedlings

Method of
Pollin-
ation

Behavior

of
Hybrids

Number

of
Seedlings

Z 3

capillaris X tectorum

6 tectorum
5 capillaris
1 intermediate

2

All failed to
pass cotyledon
stage

12

Z 5

capillaris X tectorum

All tectorum,
showed hybrid
vigor

2

All failed to
pass cotyledon
stage

6

Z 7

capillarisy.tectorum

All tectorum,
showed hybrid
vigor

2

All failed to
pass cotyledon
stage

3

Z 8

capillaris X tectorum

Intermediate

2

All failed to
pass cotyledon

stage

1

Z10

tectorum X capillaris

All tectorum,
showed hybrid
vigor

1

All failed to
pass cotyledon
stage

12

Z 12

capillaris X tectorum

Small and dis-
torted, abnormal

1

Failed to pass

cotyledon

stage

1

Z13

capillaris X tectorum

Small and dis-
torted, abnormal

2

All failed to
pass cotyledon
stage

5

Z 3 Fx (18.42 P21 X p2 P51) capillaris X tectorum

Three heads from which pollen had been washed with a jet of water
were then pollinated with tectorum pollen and covered with a bag.
These heads produced forty-two achenes which were smaller even than
the average capillaris achenes. Of twenty-four placed in the germi-
nator, thirteen sprouted and the twelve surviving divided themselves
in the cotyledon stage into three groups. Six appeared like tectorum
seedlings, five like capillaris, and one was intermediate. These all died
at the end of the cotyledon stage.

Z5Fj (18.58 P4 X p2 P10) capillaris X tectorum

Achenes from this cross were planted at two different times. Of
the four placed first in the germinator, only two sprouted, both having
large tectorum-like cotyledons. The plumule leaves started on one

194 University of California Publications in Agricultural Sciences [Vol. 2

plant but failed to appear on the other. After remaining alive, but
not growing, for eighty-one days the plant with the rudimentary
plumule died. The one failing to produce even the rudimentary
plumule leaves lived for a shorter period.

After this unsuccessful attempt, twelve remaining achenes were
placed in the germinator. At the end of five days five had sprouted,
four of which showed robust, healthy cotyledons which resembled those
of tectorum seedlings but on an enlarged scale. They were essentially
as the first two plants secured from the same lot of seed. After having
produced abnormally large cotyledon leaves, and in some cases rudi-
ments of plumule leaves, all the seedlings began to turn yellowish, and,
despite efforts to revive or stimulate them, continued to decline until
finally they died.

One of the plants which was beginning to show signs of distress
was carefully removed from the soil by washing in a pan of water.
The root was one inch long and had a blunt rounded tip covered by
the rootcap. All along the root from the tip to the ground surface
line were small knots or wartlike protuberances as if lateral roots
might have been attempting to push out. Later cytological examina-
tion showed this to be the case. The root and the cap were turning
brownish as if growth had ceased and decomposition had begun,
although the above-ground parts had only begun to show signs of
unhealthiness. The fifth plant of this culture was smaller than the
others but otherwise like them.

When the plants began to show symptoms of declining health some
of them were treated with ether in an effort to stimulate them to new
growth, but this appeared to have no effect and all perished.

% 7 Pj (18.204 PB X p2 P51) capillaris X tectorum

Eight Fx achenes were produced. Three out of five achenes pro-
duced seedlings which showed their hybrid nature by developing large
tectorum-like cotyledons, by their failure to produce plumule leaves,
and by their inability to pass beyond the cotyledon stage.

Z8F; (18.201 Px X p2 P51) CAPILLARIS X TECTORUM

Only one achene was produced on one head washed free of its own
pollen and pollinated with tectorum pollen. This achene produced a
seedling which appeared to be intermediate between its parents, and
like other hybrid seedlings died at the end of the cotyledon stage.

1920] Babcock-Collins : Crepis capillaris X C. tectorum 195

Z 10 Ft (p2 P51 X e4 P2) tectorum X capillaris

This represents the reciprocal of the above mentioned crosses in
which capillaris was used as the female parent. All tectorum flowers
used in hybridization work were emasculated in the bud stage.

Six Pj seedlings were secured, all exhibiting tectorum cotyledons
on an enlarged scale. All died at the end of the cotyledon stage, some
having started to produce plumule leaves which resulted only in
rudimentary and abnormally shaped structures too small to be
described in detail. Ether treatment of two slowly dying seedlings
failed to stimulate them to renewed growth. These plants remained
alive in the cotyledon stage thirty days.

Z 12 Ft (18.42 Px X p2 PB1) capillaris X tectorum

In this culture capillaris flowers were emasculated in the bud stage
before the stigma was receptive. Three heads produced four achenes,
only one of which sprouted. It produced a small plant with under-
sized distorted cotyledons and no plumule. This weak seedling died
in the cotyledon stage.

Z 13 F1 (el4 P. X 212 P21) capillaris X tectorum (Kew)

One washed head produced eight achenes. Three sprouted, and the
plants had enlarged cotyledons which persisted for some time. One
seedling produced several abortive plumule leaves but they all stopped
growing when about 4 or 5 mm. long. It appeared unable to produce
typical plumule leaves. Those formed were tiny threadlike structures
and not at all like plumule leaves of normal seedlings. The diameter
of this plant at sixty days was three fourths of an inch. (Some of the
normal tectorum plants produce seed in ninety days.) Another seedling
went through essentially the same process and died when four months
old. The third plant had twisted, deformed cotyledons, and each
appeared to have a separate root. They were separated, each cotyledon
placed in a pot of soil, where one died after four days, the other con-
tinuing the struggle for thirty-six days before it too perished.

It will be noted that cultures Z 10 and Z 12 are reciprocal crosses
in which each female parent was emasculated, thus insuring hybridity,
and that the behavior of the resulting seedlings was similar. The
plants of both cultures failed to develop past the stage in which the

196 University of California Publications in Agricultural Sciences [Vol. 2

young seedling is nourished from the food material stored in the seed.
Apparently in the combination of capillaris and tectorum the germinal
elements are incapable of interacting in such a way as to cause the
seedling to develop normally (pi. 36, fig. 3).

In a number of cases (not listed) where the capillaris female parent
was washed and pollinated with tectorum pollen, a number of achenes
were secured which germinated well, producing seedlings which
appeared and behaved in every way like typical capillaris plants.
These did not stop growth at the end of the cotyledon stage but con-
tinued normal development. They were maternal in all respects. Thus
we get two kinds of results when depollination by water is substi-
tuted for emasculation, and tectorum pollen applied: (1) plants which
show the tectorum type of cotyledon on an enlarged scale, and which
die at the end of the cotyledon stage of development ; (2) plants which
show maternal inheritance and are able to develop past the cotyledon
stage, the limit of development in class one. Class two occurs only
when capillaris is the female parent and the water method of depollina-
tion is used. Of eleven crosses where the female was depollinated by
means of the water jet, six produced F1 seedlings having tectorum
cotyledons on an enlarged scale, and all six failed to develop beyond
the cotyledon stage ; five produced Fx seedlings typically capillaris
which developed normally into capillaris plants.

From the evidence furnished by the equivalent results of reciprocal
crosses when the female plant was emasculated (Z 10 and Z 12), we are
led to conclude that seedlings of the second class described above,
exhibiting maternal inheritance, were the result of self-fertilization
of the capillaris plant, which may have occurred before washing or
because of incomplete removal of the pollen by the water method, and
that those of the first class, showing dominance of tectorum in F1 and
the failure to continue development, were true hybrids. As a check
a number of heads were depollinated with water and bagged without
pollination. In one case selfed seeds were produced in a bag covering
heads so treated. This indicates that the method is responsible for
the appearance of the capillaris plants where crossing was attempted.
In no case were achenes produced on heads which had been emasculated
and bagged without pollination as checks.

The above conclusions were confirmed by cytological examination.
Cells from the mature plants (Z 9 P„) were found to contain six
chromosomes, the typical number for capillaris. Cells of the root tips
from young seedlings of the hybrid class (Z 5) were found to contain

1920] Babcock-Collins: Crepis capillaris X C. tectorum 197

seven chromosomes, the sum of the haploid numbers of capillaris and
of tectorum. Nothing can be learned of the reduction division
because the plants never reached maturity, but there seems to be no
difficulty in somatic division, all seven of the chromosomes dividing
in an apparently normal fashion.

Examination of a young ¥t seedling (Z 5) which had reached
the limit of development, revealed a most unusual teratological cell
condition (pis. 37 and 38). The tissue systems of the plant were in
a chaotic condition. Patches of embryonic tissue were distributed
here and there among the larger vegetative cells, patches or sections
of tracheary cells were likewise distributed here and there throughout
the mass. Groups of vegetative cells were separated by streaks of
disorganized and disintegrated tissue. It apeared as if the force that
directs the organization and coordination of cell systems, whatever it
is, was lacking. This lack of order in the cell systems prevented the
functioning of these systems and caused cessation of development.

The principal features of the interspecific hybrids here recorded
are :

1. Reciprocal crosses are equivalent.

2. Fj shows dominance of tectorum cotyledon characters and hybrid
vigor, as expressed by the increased size of the seedling parts.

3. Absence of complete organization and coordination of the func-
tioning systems, which absence causes the death of the plant at the
end of the cotyledon stage.

The possible origin of species having a larger chromosome number
from species having a smaller number by fragmentation or segmenta-
tion of some of the latter has been suggested a number of times. Metz5
shows a diagrammatic gradation of chromosome numbers for different
species of Drosophila. Hance6 applies the same idea to the origin of
chromosome number variations in Oenothera species. Rosenberg7
recently concluded that the origin of Crepis species with three, four,
and five pairs of chromosomes could best be explained by non-dis-
junction occurring during the reduction division. Bridges8 actually

5 Metz, C. W., "Chromosome Studies in the Diptera, I. A preliminary study
of five different types of chromosome groups in the genus Drosophila." Jour.
Exp. Zool., vol. 17 (1914), pp. 45-59.

6 Hance, E. T., "Variations in the number of somatic chromosomes in
CEnothera scintillans de Vries. " Genetics, vol. 3 (1918), pp. 225-261.

7 Kosenberg, O., "Chromosomenzahlen und Chromosomendimensionen in der
Gattung, Crepis." Arkiv for BotaniJc, Bd. 15 (1918), p. 11.

s Bridges, C. B., "Non-disjunction as proof of the chromosome theory of
heredity." Genetics, vol. 1 (1916), pp. 1-52 and 107-163.

198 University of California Publications in Agricultural Sciences [Vol. 2

found a female Drosophila melanogaster with five pairs of chromosomes
which originated after secondary non-disjunction in both parents.

Assuming that Crepis tectorum, a species with four pairs of
chromosomes, originated by non-disjunction of one pair of the capil-
laris chromosomes, we would expect a cross between these two to be
compatible inasmuch as the chromosome complex should be identical,
tectorum merely having one of the capillaris chromosomes in duplicate.

The demonstrated inability of hybrids between the two species to
function normally leads to the conclusion that Crepis tectorum is not
related in such a direct way to Crepis capillaris.

The results reported here indicate the desirability of making pre-
liminary experiments in hybridizing all the species of Crepis that give
promise of being of value for genetic investigations. Experiments
with other species are now under way.

Grateful acknowledgment is made to Dr. Ruth F. Allen for her
assistance in preparing the cytological material.

Digitized by the Internet Archive

in 2012 with funding from

University of California, Davis Libraries

http://archive.org/details/interspecifichyb25babc

EXPLANATION OF PLATES

PLATE 36
Crepis seedlings. X 2.

Fig. 1. Crepis tectorum. Normal seedling showing elongated cotyledons
extending horizontally from the center. Note the two plumule leaves at right
angles to the cotyledons.

Fig. 2. Crepis capillaris. Normal seedling showing short rounded cotyledons
extending horizontally, one plumule leaf showing at right angles to the
cotyledons.

Fig. 3. F, hybrids, C. tectorum $ X C. capillaris <$ (above). Ft hybrids, C.
capillaris $ X C. tectorum $ (below). Seedlings show stage at which develop-
ment ceases.

Fig. 4. Selfed seedling resulting from crossing method No. 2. Note the
roundish C. capillaris type of cotyledons (marked c) not at all like those of the
Fj hybrids. Five plumule leaves are shown, the plant being the same age as
the hybrids.

Fig. 5. Fj hybrid seedling produced by crossing method No. 2. Notice the
two small abnormal plumule leaves between the cotyledons. This shows the
highest stage to which any of the hybrids developed.

[200]

UNIV.

CALIF. PUBL. AGR. SCI. VOL. 2 | BABCOCK-COLLINS] PLATE 36

> - ***v

t. <

PLATE 37

Teratologieal tissue of F, hybrid Crcpis capillaris X C. tcctorum. Z 5.

Fig. 1. A vertical, not quite median, longitudinal section of the abortive
plumule of a seedling. X 530.

a. An isolated patch of tissue surrounded by the slime (dark in reproduc-
tion) of disintegrating cells. Within the patch are the leaf tracheids and an
irregular mass of meristematic dividing cells.

b. Above the main patch of meristem is a second smaller layer, also lying
free in a layer of slime. It is about ten cells long and two or three cells thick.
The cells nearest it on all sides are fully matured parenchyma.

c. Side section of apex of plumule showing only mature cells.

[202]

UNIV. CALIF. PUBL. AGR. SCI. VOL. 2 [BABCOCK-COLLI NS J PLATE 37

PLATE 38

Fig. 1. Teratological tissue in Ft hybrid Crepis capiUaris X tectorum. Z 5.
Cross-section of root just below ground level. X 530.

a. A vegetative cell dividing (metaphase). This cell is completely sur-
rounded by fully differentiated cells.

b. Very much crumpled and distorted cells of outer wall of seedling.

c. Black areas showing decomposition of cells.

Fig. 2. Chromosomes of Crepis capiUaris. Polar view showing two J 's and
four more or less rodlike. X 1500.

Fig. 3. Chromosomes of Fj hybrid C. capiUaris X C. tectorum (Z 5) show-
ing two J's, one V, and rodlike ones. X 1500.

[204]

UNIV, CALIF. PUBL. AGR. SCI. VOL, 2

[BABCOCK-COLLINS] PLATE 38

v

■ i • •

r^**

^ VA

■k

Hh

Y

»

Fig. 1

; * -

Fig. 2

V

Fig. 3