Part II.--STRUCTURE AND CHEMICAL NATURE OF THE SEED COAT AND ITS_Sweet-Clover Seed

RELATION TO IMPERMEABLE SEEDS OF SWEET CLOVER.[3]

[3] The writers wish to acknowledge the service rendered by Mr. H.

S. Doty, Instructor in Botany, Iowa State College, Ames, Iowa, in

assisting in the preparation of this article.

HISTORICAL SUMMARY.

When agriculturists first began to cultivate wild legumes they observed

that many seeds would not germinate within a comparatively short time

after planting. Thus the problem of impermeable seeds began to demand

attention many years ago. However, impermeable seeds are not confined

to the Leguminosæ, as they occur also in the Malvaceæ, Chenopodiaceæ,

Convolvulaceæ, Cannaceæ, and other families.

Since the first account of the structure of legume seed coats by

Malpighi (23 v. 1) in 1687, many investigators have contributed to our

knowledge of the structure of the coats of seeds belonging to this

family.

Pammel (31) made an extensive study of legume seeds, including all the

genera in the sixth edition of Gray's Manual, as well as genera not

included in that publication. He found that the seed coat uniformly

consisted of three layers, namely, the outer layer of Malpighian cells,

the osteosclerid layer, and the inner layer of nutrient cells. Pammel's

work included a study of the seed coats of _Melilotus alba_ and _M.

officinalis_, and the descriptions and illustrations in his publication

agree for the most part with the results obtained in the investigations

reported in this article. However, more variation was noticed in the

different layers of the seed coat than he describes.

The cause of impermeability in seeds has been investigated by many.

It has been found to be due to the embryos in some seeds, such as the

hawthorns, but in most cases to the structure of the seed coat, and

especially so in the Leguminosæ. Crocker (3) states that, exactly

opposite to the common view, the cause of delayed germination generally

lies in the seed coats rather than in the embryos. Nobbe (29) thought

that the hardness of leguminous seeds was due to the Malpighian layer,

and in a later publication Nobbe and Haenlein (30, p. 81) state that

the absorbent power of many seeds is inhibited or entirely arrested

by the cones of the Malpighian cells and the shields built up between

them, which consist principally of cutin. Huss (15) agrees with Nobbe

and Haenlein. Verschaffelt (39) found that the impermeability of the

seeds of Cæsalpiniaceæ and Mimosaceæ investigated was due to, the

inability of water to pass through the canals of the seed coat. By

soaking the seeds in alcohol or other substances which change the

capillarity of the pores, the seed coats were made readily permeable

to water. Gola (6) states that the cause of the impermeability of seeds

is the peculiar character of the Malpighian cells, which prevents their

infiltration and consequent increase in volume, while Bergtheil and Day

(2) found that the hardness of the seeds of _Indigofera arrecta_ was

due to their possession of a very thin outer covering of a substance

resistant to water. Ewart (5, p. 185) believes that in most impermeable

seeds the cuticle prohibits the absorption of water, but gives as an

exception _Adansonia digitata_, in which the whole integument seems to

be permeable to water with difficulty. The following is quoted from

White (42, p. 205):

As a general rule in small and medium-sized seeds the cuticle is well

developed and represents the impermeable part of the seed coat, while

in the cases of large seeds, such as those of _Adansonia gregorii_,

_Mucuna gigantea_, _Wistaria maideniana_, and _Guilandina bonducella_,

the cuticle is relatively unimportant and inconspicuous. In these

seeds the extreme resistance which they exhibit appears to be located

in the palisade cells.

In discussing the seed coat of _Melilotus alba_, Rees (33, p. 404)

states that the outer layer consists of palisade cells covered,

externally by a structureless membrane, which, however, did not

appear to be cuticle but hemicellulose, as it stained magenta with

chloriodid of zinc. The greater part of the walls of the palisade

cells also appears to be composed of hemicellulose and the outer ends

only were cuticularized. In order to find whether the outer membrane

was in itself impermeable to water, this author treated seeds for

short intervals in sulphuric acid to dissolve the outside covering

without directly affecting the palisade cells. Seeds treated in this

manner swelled in water and microscopic examination showed that the

ends of the palisade cells were quite intact, but had separated from

each other. From this it was concluded that the outer membrane is

instrumental in conferring impermeability on the seed, although not

directly responsible for it, as is the case with a true cuticle. It

is further believed that it probably served as a cement substance

by means of which the cuticularized ends of the cells were held

together closely, thus forming a barrier through which water could not

penetrate, but that as soon as this barrier was removed the ends of the

palisade cells separated and water passed in between them.

More than 20 years ago machines were devised by Kuntze, Michalowski

(27, p. 86), Huss (15), and later by Hughes (14), to scarify

impermeable seeds. Other methods have been recommended and employed to

some extent for hastening the germination of seeds. Hiltner (13, p.

44) treated seeds of red clover, white clover, and alfalfa 10, 30, and

60 minutes with concentrated sulphuric acid and found that the best

germination resulted from the 60-minute treatment. Love and Leighty

(21) also treated the seeds of various legumes with concentrated

sulphuric acid and obtained a better germination in all cases. In

their investigations with _Melilotus alba_ it was found that a 2-hour

treatment resulted in some injury to the seed, but that a treatment

varying from 25 minutes to 1 hour gave good results. In most cases in

our investigations the seed coats of sweet clover became permeable

to water after a treatment of 15 minutes in concentrated sulphuric

acid, and within 5 minutes all of the Malpighian cells were destroyed

down to the light line. Harrington (10) found that the soil, season,

climate, color, or size of red-clover seeds had no influence upon

the percentage of impermeable seeds and that the good germination

ordinarily obtained with red clover was due to the scarifying of the

seed coats by the rasps of hulling machines. Harrington (11) also

studied the agricultural value of impermeable seeds and found that

alternations of temperature cause the softening and germinating of

many impermeable clover seeds when a temperature of 10° C. or cooler

is used in alternation with a temperature of 20° C. or warmer and that

the effect of such an alternation of temperature is greatly increased

by previously exposing the seeds to germinating conditions at a

temperature of 10° C. or cooler and is decreased by previously exposing

the seeds to germinating conditions at a temperature of 30° C. It is a

well-known fact that impermeable seeds which remain in the field over

winter germinate readily the following spring.

The light line is the most important and interesting feature of

the Malpighian cell, at least so far as _Melilotus alba_ and _M.

officinalis_ are concerned. But one light line occurs in the

Malpighian cells in most Leguminosæ, although Pammel (32) reports two

well-developed light lines in _Gymnocladus canadensis_, Junowicz (16)

found three in _Lupinus varius_, and Sempolowski (36) two in _Lupinus

angustifolius_.

Many investigators have studied the light line, and different theories

have been advanced as to its function, physical properties, and

chemical nature. Schleiden and Vogel (35, p. 26) in describing the

mature testa of _Schizolobium excelsum_ in 1838 undoubtedly referred to

the light line when they stated that the walls of the Malpighian cells

were not equally thickened. Mettenius (26), in 1846, was probably the

first definitely to describe the light line. This author believed it

was composed of pore canals, all appearing at the same height in the

cells, but he was unable to prove this by cross sections. Lohde (20)

studied the light line in seeds of _Hibiscus trionum_ and found it

cutinized. Hanstein (8) states that the Malpighian cells are composed

of two cell layers and the light line is produced by the adjoining

walls of the ends of the cells. Later, this same author (9), according

to Harz (12), refers to the light line as a perforated disk composed of

tissue of strong refracting power.

Russow (34) concludes that the light line is produced by neither

chemical nor mechanical changes but is caused by a modified molecular

structure containing less water than the remainder of the cell wall.

Hiltner (13) agrees with Russow's explanation. Harz (12, p. 561) also

agrees with Russow and adds that he has observed that the light line

disappeared in a number of cases after applications of nitric acid.

Wigand and Dennert (43) suggested that the light line is due to a

series of erect fissures, while Tietz (37, p. 32) believes it is due

to a chemical modification and that the phenomenon results from the

exceptionally extreme density of parts of the cellulose membrane.

Junowicz (16) found evidence of cellulose material. The cell wall

at this point was strongly refractive and had a different molecular

structure. After studying _Phaseolus vulgaris_, Haberlandt (7, p.

38) agrees with the Russow explanation. In the seed of this plant

the light line colored blue after being treated with chloriodid of

zinc. Sempolowski (36), who investigated the light line in _Lupinus

angustifolius_, states that there is not only a difference in the

molecular structure but also a chemical modification of the cell wall

at this point, since with iodin and sulphuric acid the cell wall

colored blue, whereas the light line colored yellow. Wettstein (41),

who studied seeds of Nelumbo, agrees with Russow (34) and Sempolowski

(36) that chemical and physical modifications occur. He found that

iodin and sulphuric acid colored the Malpighian cells intensely blue,

the light line at first yellowish, and then later it gradually became

blue. This reaction may be accelerated by heat. Iodin produced the same

effect, and the light line colored blue more rapidly. When treated with

a water-withdrawing medium the light line was not altered for some

time, but finally disappeared with continued application. Cooking for a

long time in caustic potash or standing in cold caustic potash caused

the cells to swell, while the light line remained uninjured at first

but finally disappeared. He also believed that the absence of pore

canals in the region of the light line caused it to be more dense.

Nobbe and Haenlein (30) treated sections of seed coats of _Trifolium

pratense_ with iodin and sulphuric acid and found that the light line

colored blue as readily as the thickened ridges that radiate inward

from it, but that the outer processes of the palisade cells projecting

from the light line toward the cuticle stained dark brown. They also

state that various causes work to produce such unusual lusters in

the light line, the principle one of which is the thickened ridges

which radiate inward, reach their greatest development at this point,

and coalesce in the lumen of the cell. The result is that the light

line falls upon a continuously homogeneous medium, while in the inner

portions of the ridges the light passes through media of varying

opacity, such as cellulose, water, and protoplasm, whereby it is

progressively subdued in varying degrees by partial reflection. Pammel

(31, p. 147) studied the light line in _Melilotus alba_ and found that

it consisted of a narrow but distinct refractive zone below the conical

layer. The refractive zone colored blue with chloriodid of zinc. The

whole upper part was, however, more or less refractive, while the

remainder of the cell wall contained pigment and colored blue with

chloriodid of zinc. Small canals project into the walls, in some cases

extending beyond the light line.

Beck (1) found that the light-refracting power of the light line was

much greater than that of the undifferentiated membrane and stated that

there may be in addition to this a chemical difference which can not be

detected with the present microchemical methods. He does not believe

that it is cuticularized or that it contains less water than the rest

of the cell.

Marlière (24, p. 11) gives a physical explanation and states that

the true cause of the light line lies in the peculiar structure

of the secondary membrane of the Malpighian cell. Tunmann (38, p.

559) observed that it did not hydrolize in weak acids and therefore

decided that it was not hemicellulose. He found that it dissolved in

concentrated sulphuric acid more readily than the regions surrounding

it and that it was composed of pectin or callose. In our investigations

the main portion of the light line of _Melilotus alba_ and _M.

officinalis_ was very resistant to concentrated sulphuric acid, only

the narrow outer portion being attacked. It showed evidence of callose.

MATERIAL AND METHODS.

Permeable and impermeable seeds[4] of _Melilotus alba_ and _M.

officinalis_ were obtained from commercial samples and also from

samples collected in the field. Those selected for sectioning were

allowed to dry after being removed from the germinator and then

embedded on the ends of pine blocks in glycerin gum, which was made by

dissolving 10 grams of powdered gum arabic in 10 c. c. of water and

adding 40 drops of glycerin. After the glycerin gum had dried for 24

hours, the seeds were easily sectioned. This method of embedding causes

no change in the seed coat. It is more satisfactory than the paraffin

method for holding the seeds firmly. The glycerin gum dissolved readily

when the sections were mounted in water.

[4] The term "permeable" is used in this paper to designate seeds whose

coats are permeable to water in two weeks or less at temperatures

favorable for germination, while the term "impermeable" is used to

designate seeds whose seed coats are impermeable to water for this

length of time when temperatures are favorable for germination.

Impermeable seeds are commonly referred to as "hard seeds," and they

may become permeable in time.

In the microchemical studies Sudan III, alcanin, chlorophyll solution,

and phosphoric acid iodin were used to test for cutin or suberin;

sulphuric acid and iodin, chloriodid of zinc, and chloriodid of

calcium for cellulose; phloroglucin and hydrochloric acid for lignin;

ruthenium red for pectic substances; and sulphuric acid, Congo red, and

aniline blue for callose.

Where very thin sections were necessary for detailed study of the

structure of the seed coat, pods in various stages of development were

collected, and after the usual preliminary treatment they were embedded

in paraffin and sectioned with the microtome. Microchemical tests were

made with these sections by using various specific stains. Safranin was

used to test for cutin, suberin, and lignin; haematoxylin and methyl

blue for cellulose ; methylene blue, methyl violet B, mauvein, and

ruthenium red for pectic substances; and aniline blue and Congo red for

callose. In studying some points with reference to the pore system of

the seed coat, it was necessary to use free-hand sections of fresh pods.

In studying the seed coat in relation to the absorption of water,

both permeable and impermeable seeds were soaked in water solutions

of safranin, gentian violet, eosin, and haematoxylin, then dried and

embedded in glycerin gum for sectioning. Seeds were soaked in stains

dissolved in 95 per cent alcohol to test the penetration of alcohol. It

was evident that the seed coats did not act as a filter, as the stains

passed through them with the water or alcohol.

STRUCTURE OF THE SEED COAT.

There is very little endosperm present in mature seeds of _Melilotus

alba_ or _M. officinalis_. That which is present is quite permeable to

water and therefore bears no relation to the impermeable seeds of these

plants.

The outer layer of the seed coat, which is the modified epidermal layer

of the ovule, is known as the Malpighian layer. (Pl. V, figs. 1 and

2.) The cells constituting this layer, commonly called palisade cells,

are the most highly modified cells of the seed coat. They are very

much elongated, their length varying in the different regions of the

coat, and their outer tangential walls and the outer portions of their

radial walls are so much thickened that their lumina are confined to

the inner portion of the cells, sometimes occupying less than half the

length of the cells. The inner tangential walls and inner portions of

the radial walls are thickened just previous to the death of the cells,

the thickening sometimes being only slight and sometimes so much as to

leave only very narrow lumina.

There is a very thin layer on the outer surface of the Malpighian

cells which has been called cuticle by previous investigators, but

the chemical composition of this layer and its perviousness to water

indicate that there is very little cutin present. This layer is

probably the primary epidermal cell wall rather than a deposit on the

outer surface of the wall. To determine this a study of the development

of the Malpighian cells is necessary.

Beneath the so-called cuticle there is the much thickened outer portion

of the Malpighian cells in which there are two rather distinct regions,

one constituting the conelike structures and the other forming a

continuous layer over the conelike structures, separating them from

the cuticle and filling in between them. These two regions separate

easily, and in cutting sections the outer region, called by some the

cuticularized portion, often breaks away, leaving the entire surface of

the cones exposed.

The term "cuticularized layer" will be used to designate all of the

thickening covering the cones, including that around the cones as well

as the portion between the cones and the cuticle. This term is not

entirely appropriate, for the region is practically free from cutin,

but for the want of a better term it will be used. There are canals

in the cuticularized layer and cones, which are easily seen when the

sections are treated with chloriodid of zinc or sulphuric acid. A

surface view of a section showing the cones and cuticularized layer

when mounted in glycerin shows the canals as dark lines due to the

air inclosed. The canals are most abundant along the lines where the

lateral walls of the cells join, but many are within the cones and in

the cuticularized substance between the cones. (Pl. V, fig. 5.)

The well-developed light line in _Melilotus alba_ and _M. officinalis_

is found just below the bases of the cones. In some seed coats only

a few and in others none of the canals which are common in the cones

and cuticularized region cross the light line. A very distinct line of

small canals filled with air and thus forming a dark band is present

just above the fight line, thus making the light line more conspicuous.

(Pl. V, fig. 3.) When the lumina of the cells extend across the light

line, they are exceedingly small. The light line is the most compact

region of the Malpighian layer and is conspicuous because it refracts

the light much more than the regions above and below it.

Just below the Malpighian is a layer of cells variously modified and

known as the osteosclerid. The cells of this layer are often referred

to as the hourglass cells on account of their shape. In some regions

of the seed coat they are expanded at both ends and their walls are

much thickened, the thickenings forming ridges on the radial walls,

while in other regions only the upper tangential wall and a portion of

the radial walls are thickened and the cells are expanded only at the

inner end, thus having the shape of the frustum of a cone. Beneath the

osteosclerid layer is the nutrient layer.

The nutrient layer contains chloroplasts. It varies not only in the

number of layers of cells composing it, but also in the modifications

of these cells. This layer ranges from four to seven cells in thickness

in the different parts of the seed coat.

PLATE V.

[Illustration]

Structure of the Seed Coat of Sweet Clover.

Fig. 1.--Section of the seed coat of _Melilotus officinalis_. × 450.

Fig. 2.--Another section of the seed coat of _Melilotus officinalis_,

showing the variation in size and modifications that occur in the

three layers. × 450. Fig. 3.--Section of the Malpighian layer of a

_Melilotus alba_ seed, showing a line of canals just above the light

zone. × 450. Fig. 4.--Section of the Malpighian layer of a permeable

_Melilotus alba_ seed. × 450. Fig. 5.--Tangential section of the

Malpighian cells cut between the cuticle and tops of the cones, showing

pores. × 530. Fig. 6.--Section through the Malpighian layer of an

impermeable _Melilotus alba_ seed. × 450. Fig. 7.--Section through

the Malpighian layer of an impermeable _Melilotus alba_ seed, showing

the region through which water and stains readily passed. × 450. Fig.

8.--Cross section of a Malpighian cell of a permeable _Melilotus alba_

seed through the region of the light zone, showing the lumen not

entirely closed. × 530. Fig. 9.--Section through the Malpighian layer

of a _Melilotus alba_ seed shaded to show the portions which react to

the cellulose and pectose tests. × 450. Fig. 10.--Section through the

Malpighian layer of a _Melilotus alba_ seed which shows the condition

of the seed coat after 60 minutes' treatment of concentrated sulphuric

acid. That portion above the light zone was destroyed, and the lumina

as small pores through which much of the stain now passed were seen

extending across the light line. The lines between the cells were much

more distinct, appearing as small intercellular spaces through which

some stain passed. × 450. _a_, Cuticle; _b_, cuticularized layer; _c_,

conelike portion of the thickening of the Malpighian cells; _d_, light

line; _e_, region of a hard seed coat through which water and stains

readily passed; _l_, lumen; _M_, Malpighian cells; _N_, nutrient cells;

_O_, osteosclerid cells; _p_, canals just above light zone.

MICROCHEMISTRY OF THE SEED COAT.

Tests for cutin showed that there was very little present in the seed

coat. Slight reactions for cutin were observed in the cuticle, in the

outer margin of the cuticularized layer, and in the basal portion of

the cones. These reactions were so slight as to be almost negligible.

It is evident that the cuticle and cuticularized layer are not well

named in _Melilotus alba_ and _M. officinalis_. Tests for cellulose

showed that it was present in the cuticle, cuticularized layer, cones,

the walls of the Malpighian cells below the light line, and the walls

of the cells of the osteosclerid and nutrient layers. (Pl. V, fig. 9.)

The reaction for cellulose in the Malpighian cells was quite distinct

in the walls below the light line, less distinct in the cones and

cuticle, and least distinct in the cuticularized layer.

Tests for lignin occasionally showed slight traces in the Malpighian

cells below the light line. When treated with reagents for pectic

substances, the cuticle, cuticularized layer, cones, and all cell walls

below the light line gave a definite reaction. The reaction of the

cones and cuticle was more pronounced than the cuticularized layer.

Tests for callose gave no reaction except in the upper part of the

light line. This part of the light line stained slightly blue with

aniline blue and was easily dissolved with sulphuric acid. In cutting

free-hand sections of fresh material the Malpighian layer sometimes

broke along this line. The greater part of the light line reacted to

none of the tests, and its chemical nature was not determined.

When microtome sections of seeds in different stages of development

were treated with various stains, the results were in accord with those

obtained with free-hand sections. Thus with safranin the periphery and

cones of the Malpighian cells were slightly stained, while haematoxylin

and methyl blue stained all the seed coat except the light line. The

cones and cuticle stained more readily than the cuticularized layer,

but neither stained as deeply as the cell walls below the light line.

Methylene blue, methyl violet B, and mauvein stained all above the

light line, indicating the presence of pectic substances; however, the

staining was more prominent in the cones and cuticle.

The difference in reaction of the cones and cuticularized layer to

the cellulose and pectose tests probably indicates a difference in

density rather than a difference in chemical composition. Since the

cuticularized layer separates readily from the cones, there may be a

difference in physical properties.

With Congo red the upper part of the light line was only very slightly

stained, but aniline blue had a more pronounced effect.

The microchemical tests applied to the seed coat show that in the

region above the light line there is only a slight trace of cutin

or suberin, but a considerable amount of cellulose and pectose. All

cell walls below the light line are mainly cellulose but contain some

pectose. The upper portion of the light line contains callose, but the

remainder of the light line appears to be chemically different from all

other parts of the seed coat or else so dense as to resist the attack

of the reagents.

THE SEED COAT IN RELATION TO THE ABSORPTION OF WATER.

A study of permeable seeds soaked in water containing stains showed

that there were no local regions through which the water passed.

The stains passed through all regions of the seed coat. Coating the

micropylar region with vaseline retarded germination, but had no

effect upon the percentage of germination at the end of three days. In

seed coats through which the stain had passed, the light line was not

stained. Some stain was found in the canals which crossed the light

line, and much more in the cell cavities. There was no evidence that

the stain had permeated the substance of the light line. It was able to

cross the light line only when pores were present.

In impermeable seeds the stains passed readily to the light line.

(Pl. V, fig. 7.) It was evident that the absorption of water was

not prevented by either the cuticularized layer or the cone-shaped

structures of the Malpighian layer, but by the light line. The region

outside of the light line became stained in a few hours, but there

was no trace of the stain within the light line after the seeds had

remained a week in the stains. Alcohol did not penetrate the seed coat