Part 22

2. =The Macrospores.= The prothallium in _Salvinia_ and _Marsilia_ is still rather large, green, and capable of the independent assimilation of carbon. It projects more or less from the macrospore and bears (in _Marsilia_ only one, in _Salvinia_ several) archegonia, which however are embedded to a greater degree in the prothallium, and are more reduced than the archegonia of the true Ferns and Horsetails (Figs. 215, 216). The prothallium is still more reduced in _Isoëtes_ and _Selaginella_; _partly_ because it is smaller and is in a higher degree enclosed in the spore, it also contains less chlorophyll, or is entirely without chlorophyll, and in consequence incapable of independent existence, whilst the number of archegonia is less; and _partly_ because the archegonia are themselves reduced, the cells of the neck are fewer and embedded to the level of the surface of the prothallium without any, or with only a very slight projection (Figs. 235, 236).--Finally, the prothallium with its archegonia begins to develope in _Selaginella_ while the macrospore is still within its sporangium, and before it is set free from the mother-plant. After the spores are set free and germination has commenced, the spore-wall ruptures and the prothallium is exposed.

The GYMNOSPERMS go still further. The macrospore (embryo-sac) germinates and forms internally a cellular tissue, designated in former times by the name of _albumen_ (endosperm), which is _homologous with the prothallium_. It always _remains enclosed in the embryo-sac_, and is a parenchymatous mass containing a large supply of nourishment. In the upper part of the endosperm a number of archegonia are developed which are in the main constructed in the same manner as those in the Cryptogams, but are still more reduced, the neck consisting only of 4, 2, or 1 cell (Figs. 251, 252). The ventral canal-cell is also formed, in the majority, as a small portion cut off from the large central cell just beneath the neck; the larger remaining portion becomes the oosphere. When the pollen-tube has passed down to the oosphere (Fig. 253) and fertilisation has been effected, the oospore commences a cell-formation, the final result of which is the formation of _an embryo_ (_the asexual generation_) which is provided with a thinner, lower end, termed the suspensor. The embryo is forced more or less into the endosperm in which it may rest for a longer or shorter time, and generally is developed to such an extent that it has a distinct primary-root (radicle) and stem (plumule) with one or more embryo-leaves (cotyledons).

When the oosphere has been fertilised its nucleus sinks down to its lower end, and by repeated division into two, forms four cells lying in one plane (Fig. 253, see base of the left archegonium). Three tiers of cells are now formed by transverse division of these four. It is the intermediate one of these which elongates and forms the suspensor, or four suspensors, if they separate from each other, which push the lowermost four cells deep down into the endosperm. It is from these four lower cells that the embryo (or four embryos when the suspensors separate) is developed, but never more than one embryo attains full development. As several archegonia are contained in one and the same ovule, all of which are capable of forming embryos, there is the possibility that several embryos may be developed in a seed (polyembryony), but usually only one embryo attains perfect development.

At the same time that the embryo is being developed, other changes are taking place in the ovule, especially in the integument which becomes the shell of the seed (_testa_). The endosperm grows, and the embryo-sac supplants the cells of the nucellus. The _seed_ is now formed, and it consists in its most complete development, as in this instance, of three parts:

(1) The _testa of the seed_, formed by the enveloping integuments, with the remainder of the tissue of the nucellus lying outside the embryo-sac (the macrosporangium).

(2) The _endosperm_ or prothallium.

(3) The _embryo_.

The reduction in the ANGIOSPERMS is carried to the extreme limit. In the embryo-sac (the macrospore) the nucleus by continued division produces a prothallium consisting of primordial cells (Fig. 254). In the upper end of the embryo-sac (which is nearest the micropyle) are three cells, two of which are termed the “co-operating cells” (_synergidæ_) and the third is the _oosphere_. Three others are placed at the opposite end of the embryo-sac and are therefore termed the “antipodal cells.” Finally, a large cell is also formed, which occupies the space between the two groups and whose cell-nucleus, the central definitive nucleus, lies in the centre of the embryo-sac. These primordial cells are the slight remnant of the prothallium. The entire structure of the archegonium, with its neck and canal-cells, has disappeared, and nothing is left but the indispensable _oosphere_. When the oosphere has been fertilised, and has commenced the cellular divisions which lead to the formation of the embryo (Fig. 255), the synergidæ and antipodal cells are absorbed, and a cell-formation begins by a new process which emanates from the definitive nucleus and by which a parenchymatous cell-tissue, the nutritive-tissue, arises which may perhaps be considered as homologous with the endosperm of the Gymnosperms. The difference is that the nutritive-tissue of the Angiosperms is formed in two parts with an intervening interruption; the primary nutritive-tissue is first formed, and after fertilisation is absorbed, with the exception of one cell, which continues the development and gives rise to the nutritive-tissue proper, which is formed in the first instance of primordial cells, and later on of a cellular tissue; this nutritive-tissue formed in the embryo-sac is termed “endosperm”; in a few instances[23] a tissue which is derived from the nucellus functions as nutritive-tissue, and is termed “perisperm.” In many plants the seeds, when ripe, contain a very rich nutritive-tissue, in addition to the embryo, for the purpose of its nourishment during germination. These are termed albuminous (endospermous) seeds, in distinction to the ex-albuminous, or those in which the nutritive-tissue is stored in the embryo itself, before it is completely developed, and used for its sustenance.

In addition to the changes which fertilisation produces in the ovule itself, it also gives the impetus to a series of changes in the entire shoot which bears the ovule. The perianth, stamens, and style, generally wither, because the part they play is at an end; the wall of the ovary grows and becomes the wall of the fruit (pericarp). The entire gynœcium of a flower, transformed as a consequence of fertilisation, is termed a _fruit_. It consists of two parts, the _pericarp_ and the _seeds_, and according to the nature of the pericarp, the fruit is termed a capsule, nut, berry, or drupe.

The chief characteristic of the Phanerogams does not lie in the formation of the flower (although they may quite properly be termed “Flowering-plants”), because Equisetums and Lycopods have reproductive shoots as highly differentiated as those of certain Gymnosperms and other Phanerogams. As regards the SEXUAL GENERATION the characteristics are found:--(1) in its great reduction; (2) in the transmission of the microspore (pollen-grain) to the macrosporangium, and its germination, with the formation of a _pollen-tube_ (antheridium), the protoplasm of which is not differentiated into spermatozoids; (3) in the fact that the macrospore (embryo-sac) never leaves its sporangium (nucellus); and further in the Angiosperms, (4) in the peculiar development of the nutritive-tissue in two parts; and (5) in the great reduction of the archegonium.

As regards the ASEXUAL GENERATION the characteristic feature is that this generation is formed whilst the sporangium is still attached to the mother-plant, and for a long time is nourished by it; and that after the sporangium has become detached from the mother-plant, it spends a longer or shorter resting period as the embryo in the seed (enveloped by the testa), and does not make its appearance until the “germination” of the seed. In addition the shoot which bears sporangia undergoes greater modification than in the case of the Flowerless-plants.

The Phanerogams are separated into two Divisions as follows:--

Division 4. =Gymnospermæ.= The ovules, as well as the seeds, are borne _naked_ on the surface of _open carpels_, or on the apex of a stem (ovary wanting). The pollen-grains are conveyed by the wind to the ovules, and caught by drops of mucilage, secreted by the micropyle. A “stigma” is _wanting_. The entire _female prothallium_ (_the endosperm_), which serves for the nourishment of the embryo, is _formed before fertilisation_. The archegonia are _embedded in the upper part of the prothallium. The pollen-grains are “multicellular,” i.e._ there is always in their interior a distinct prothallium, formed by 1–3 cells, and a larger cell which gives rise to the pollen-tube.

Division 5. =Angiospermæ.= The carpels surround the ovules and form an entirely closed chamber (_ovary_), in which the ovules mature and ripen into seeds. The surface of a portion of the apex of the carpel is transformed into the “stigma,” which, by a sticky fluid and also by hair-structures, is capable of retaining the pollen-grains conveyed to it by the wind, or more frequently by insects. The pollen-tube grows from the stigma, through the “conducting cellular tissue” (_style_), to the ovules. The pollen-grains contain two cells, a vegetative and a free generative cell. The latter passes into the pollen-tube and there divides into two, one of which is the sperm-nucleus. The female prothallium, which is intended to serve as nutritive-tissue, is formed _after fertilisation_. Archegonia are wanting.

DIVISION IV.

GYMNOSPERMÆ.

The following characters should be added to those already given on page 2:--

The Gymnosperms comprise only trees or shrubs. The flowers are always _unisexual_ and destitute of perianth (except _Gnetaceæ_); the female plant of _Cycas_ is the only one which has no flower. The MALE FLOWERS are constructed on the same type as the cones of the Horsetails and Club-Mosses, and are _most frequently long shoots_ (Figs. 243, 258, 260 _A_, 267 _J_) bearing a number of spiral or verticillate stamens. The FEMALE FLOWERS are of a more varied structure (see the orders). The OVULE _is orthotropous_ (except _Podocarpus_ which is anatropous) and projects from the carpel uprightly, inverted, or horizontally; it has usually _only one integument_ (compare however Taxaceæ) which proceeds from the upper part of the nucellus, so that the embryo-sac in part is placed below the integuments (Figs. 251, 264). The drop of mucilage which catches the pollen-grains dries up and draws the pollen-grain through the micropyle to a space just above the nucellus--_the pollen-chamber_--in which the germination of the pollen-grain commences.

In each seed, only one of the many embryos which are formed proceeds to its full development. The seed is always _endospermous_, and the embryo has one, two, or a whorl of several cotyledons. A vigorous primary root is developed on germination. THE VASCULAR BUNDLES in the stem are arranged in a ring, and _increase in thickness_ takes place by a closed cambium-ring which forms bast (_phlœem_) externally, and wood (_xylem_) internally with distinct annual rings, _as in the Dicotyledons_. Only certain of the Cycadeæ deviate from this arrangement. The _secondary wood_ is very uniform, as it is almost exclusively _formed of tracheides_ with bordered pits, but _true vessels are wanting_; this also indicates a relationship with the Pteridophyta (see page 202).

The Gymnosperms are biologically lower than the Angiosperms; they are wind-fertilised, and without extra floral-nectaries.

This Division embraces three classes: CYCADEÆ, CONIFERÆ, And GNETEÆ. It is no doubt monophyletic, and has taken its origin from heterosporous Ferns, now extinct, most nearly related to the Ophioglossaceæ and Marattiaceæ. The Cycadeæ appear to be the oldest class. The Coniferæ are related to these through Ginkgo. The Gnetaceæ are more isolated. The Division is not continued into the higher Flowering-plants; it has evidently attained its highest development, and is now in a retrograde condition. The similarity which has often been pointed out between certain Coniferæ and Lycopodinæ is only in analogous resemblances, and does not entitle one to suppose that there is a nearer relationship, or that the former take their origin from the latter.

Class 1. =Cycadeæ.=

The stem is very _rarely ramified_. The leaves are _large_, _pinnate_, and arranged spirally. The flowers are _diœcious, without perianth_.

There is only one order, the =Cycadaceæ=.--In habit they resemble the Ferns, especially the Tree-Ferns (compare Figs. 207 and 256). The stem is tubercular (Fig. 258), or cylindrical (Fig. 256), but not very tall (as much as about 12 metres), and very rarely ramified. [In Ceylon, unbranched specimens of _Cycas_ are rarely met with in the wild state. The stems of _C. circinalis_ occasionally branch in greenhouses.]

The LEAVES are arranged spirally, and so closely together that no free stem-surface is left between them, and have only a slight sheath (which is not amplexicaul, as in the Palms). They are compound (most frequently pinnate; in _Bowenia_, bipinnate); in some genera the leaves are rolled up in various ways, resembling the vernation in Ferns (Fig. 257); they are leathery and perennial. In some, stipules are present, as in the Marattiaceæ. Groups of scale-leaves alternate in the majority with groups of foliage-leaves.

The FLOWERS are without perianth. The MALE FLOWER is sometimes an enormous collection of stamens (Fig. 258), which are flat in some (_Cycas_, Fig. 241), shield-like in others (_Zamia_, _Ceratozamia_) like the sporophylls in Horsetail (Fig. 259); but in all, the pollen-sacs are situated in large and varying numbers on the back of the stamens, and arranged in groups of 2–5, like the sporangia in the sori of the Ferns (Fig. 241 _b_, _c_). FEMALE FLOWERS _are wanting_ in _Cycas_, because the carpels do not terminate the apical growth of the stem. After a group of foliage-and of scale-leaves, a group of carpels is developed, which are pinnate and resemble the foliage-leaves, bearing on their edges a number of ovules (most frequently 5–6) (Figs. 245, 256); the same stem produces successively scale-leaves, foliage-leaves, and carpels. The differentiation is not much more advanced than in certain Ferns (_Struthiopteris_, _Blechnum_), where barren and fertile leaves of different form regularly alternate. _The other genera have female flowers_; the carpels are shield-like in _Zamia_ and _Ceratozamia_ (Fig. 246), and collected into cone-like flowers, which terminate the growth of the stem (Fig. 259). The number of ovules in these instances is two to each carpel.

The SEEDS are large (most frequently 2–6 centimetres long) and plum-like; the external layer of the testa is fleshy, while the internal one is hard and horny. There are two systems of vascular bundles in the testa, one outside, the other inside the stone. The embryo is straight, attached to the end of the suspensor, which is often long, filamentous, and rolled up; it has one or two cotyledons.

The embryo in _Ceratozamia_ and others is very slightly developed, at the time when the ripe seed is detached from the carpel; and it is not until after sowing that its further development and germination proceed. This calls to mind the Cryptogams, especially _Selaginella_, whose macrospores are thrown off filled with endosperm; but the oosphere is not fertilised till after the separation of the macrospore from the parent-plant, while in the Cycadeæ fertilisation is effected before the separation. In _Cycas_ the testa may rupture, and the endosperm grow and become green in the light, even though no embryo has been formed. This also is an indication of its prothalloid nature.

Gum-passages are present in all organs. Collateral vascular bundles, with spiral and scalariform tracheides, are found; and normal thickening takes place by means of a cambium. An exceptional mode of growth is found in _Cycas_ and _Encephalartos_, the cambium ceases to divide after a time and is replaced by a new cambium which arises in the cortical parenchyma just outside the bast, and which forms a new ring of xylem and phlœem. This may be repeated so that a number of concentric rings are produced. In _Ceratozamia_, structures resembling corals extend from the roots in a vertical direction and appear on the surface of the soil; these are peculiar roots, in which a symbiotic Alga (_Anabæna_) is found.

The Cycadeæ were formerly (from the Coal period to the Later Cretaceous) far more numerous than at the present day. They appear to have been most numerous in the Trias and Jurassic. The remnant (75 species) which have persisted to the present time are found in all tropical countries. _Cycas_ (Trop. and Sub-trop., Eastern Hemisphere); _Dioon_ (Mexico); _Macrozamia_ (Australia); _Encephalartos_ (Trop. and S. Africa); _Stangeria_ (Fig. 258, Sub-trop. South and East Africa); _Bowenia_ (Trop. Australia); _Ceratozamia_ (Mexico, New Granada, Western Brazil); _Microcycas_ (Cuba); _Zamia_ (Trop. and Sub-trop. N. America.)

USES. Sago is made from the starch-containing pith of _Cycas revoluta_ and _circinalis_. The leaves are often used at funerals and church festivals, under the name of “palm-branches.”

Class 2. =Coniferæ= (=Pine-trees=).

The stem _branches freely_. The leaves are _entire_, relatively small, linear or reduced to scales. The flowers are without perianth. The ovules naked. It is seldom that the female flower is reduced to only one carpel.

Whilst the Cycadeæ principally resemble the Ferns, the Conifers partly resemble the Lycopods, and partly the Equisetums--the former especially in the _needle- or scale-like_, leathery, simple, and often perennial leaves (“evergreen plants”), which _never possess stipules_ (Figs. 263, 270, 272). _Ginkgo_ deviates from this, being no doubt the oldest, and the Conifer which stands nearest to the Cycadeæ (Fig. 260). The resemblance to the Equisetums is especially owing to the fact that the stem ramifies abundantly, and often very regularly, forming a pyramid with verticillate branches. In addition to the foliage-leaves, scale-leaves (bud-scales) are present in the majority of species.

The FLOWERS are monœcious or more rarely diœcious. _Perianth is wanting._ The stamens of the _catkin-like male flowers_ (Fig. 267, _J_) are of different forms, but as a rule more or less shield-like. As in the Cycadeæ, the pollen-sacs are in all cases situated _on the underside_. There are, as a rule, two pollen-sacs (the Abietaceæ, Fig. 267), or 3–5, (the Cupressaceæ and Taxaceæ, Fig. 243); a few have more, _e.g. Araucaria_ (Fig. 242); they dehisce by clefts.

If, in commencing our consideration of the _female flower_, we begin with that of _Ginkgo_, we shall observe in the corner of a scale- or foliage-leaf a small flower, which consists of two carpels, each bearing one ovule, and reduced almost to the ovule itself (Fig. 260 _C_, _D_). The flower in _Podocarpus_ is still further reduced, viz. to a single carpel with one ovule, which is anatropous and has two integuments. This ovule is situated in the axil of a cover-scale (_c_, in Fig. 262 _D_), and several female flowers of this description are collected in a small cone, the stalk and bracts of which become fleshy (Fig. 262 _C_). The external integument also becomes fleshy (an aril). _Dacrydium_, which is clearly related to _Podocarpus_, has an external integument which developes more independently as a fleshy aril (Fig. 262 _B_, _B’_). _Microcachrys_ also is clearly allied to these: the bracts are more fleshy, and the ovule (_i.e._ the female flower) is protruded beyond the bract (Fig. 262 _A_, _A’_). _Taxus_ stands in a more isolated position: a flower which has been reduced to an ovule is situated, in this instance, on the apex of a secondary branch which is studded with floral-leaves (Figs. 263, 264); an external integument is developed on all sides and surrounds the seed as a scarlet aril. According to this conception _the aril corresponds to an external integument_, and the Taxoideæ thus possess a partly dichlamydeous ovule. Only _Ginkgo_ and _Cephalotaxus_ appear to deviate from this, as in these there is only one integument (unless the small outgrowth indicated by _ar_, in Fig. 260 _D_, really is a rudimentary, external integument); in CYCADEÆ, to which _Ginkgo_ is most closely related, there is likewise only one integument. But in these genera the testa is differentiated into two layers, and the seed resembles a drupe; like the Cycadeæ there is an external fleshy covering and an internal hard one, and these two layers may probably be considered homologous with the two integuments. This theory is also borne out by the arrangement of the vascular bundles in _Cephalotaxus_ and _Podocarpus_, which present the xylem in the fleshy external layer to the _outside_ of the testa, which is therefore the upper side of the integument (Celakovsky).

The coalescence of the integuments into one is only slight in _Torreya_, more pronounced in _Podocarpus_ and strongest in _Cephalotaxus_ and _Ginkgo_. Celakovsky terms these ovules “holochlamydeous.”