Chapter XXI: Echinodermata (continued): Development and Phylogeny 601 (3)

We have defined this group mainly by negative characters, as such are the only means for their differentiation from the remaining Sarcodina; and indeed from Flagellata, since in this group zoospores are sometimes formed which possess flagella. Moreover, indeed, in a few of this group (_Podostoma_, _Arcuothrix_), as in some Heliozoa, the flagellum or flagella may persist or be reproduced side by side with the pseudopodia. The subdivision of the Rhizopoda is again a matter of great difficulty, the characters presented being so mixed up that it is hard to choose: however, the character of the outer layer of the cytoplasm is perhaps the most obvious to select. In LOBOSA there is a clear layer of ectosarc, which appears to be of a greasy nature at its surface film, so that it is not wetted. In the FILOSA, as in most other Sarcodina, this film is absent, and the ectoplasm is not marked off from the endoplasm, and may have a granular surface. Corresponding to this, the pseudopodia of the Lobosa are usually blunt, never branching and fraying out, as it were, at the tip, as in the Filosa; nay, in the normal movements of _Amoeba limax_ (Fig. 1, p. 5) the front of the cell forms one gigantic pseudopodium, which constantly glides forward. Apart from this distinction the two groups are parallel in almost every respect.

There may be a single contractile vacuole, or a plurality; or none, especially in marine and endoparasitic species. The nucleus may remain single or multiply without inducing fission, thus leading to apocytial forms. It often gives off "chromidial" fragments, which may play an important part in reproduction.[67] In _Amoeba binucleata_ there are constantly two nuclei, both of which divide as an antecedent to fission, each giving a separate nucleus to either daughter-cell. _Pelomyxa palustris_, the giant of the group, attaining a diameter of 1''' (2 mm.), has very blunt pseudopodia, an enormous number of nuclei, and no contractile vacuole, though {53}it is a fresh-water dweller, living in the bottom ooze of ponds, etc., richly charged with organic débris. It is remarkable also for containing symbiotic bacteria, and brilliant vesicles with a distinct membranous wall, containing a solution of glycogen.[68] Few, if any, of the Filosa are recorded as plurinuclear.

The simplest Lobosa have no investment, nor indeed any distinction of front or back. In some forms of _Amoeba_, however, the hinder part is more adhesive, and may assume the form of a sucker-like disc, or be drawn into a tuft of short filaments or villi, to which particles adhere. Other species of Lobosa and all Filosa have a "test," or "theca," _i.e._ an investment distinct from the outermost layer of the cell-body. The simplest cases are those of _Amphizonella_, _Dinamoeba_, and _Trichosphaerium_, where this is gelatinous, and in the two former allows the passage of food particles through it into the body by mere sinking in, like the protoplasm itself, closing again without a trace of perforation over the rupture. In _Trichosphaerium_ (Fig. 9) the test is perforated by numerous pores of constant position for the passage of the pseudopodia, closing when these are retracted; and in the "A" form of the species (see below) it is studded with radial spicules of magnesium carbonate. Elsewhere the test is more consistent and possesses at least one aperture for the emission of pseudopodia and the reception of food; to avoid confusion we call this opening not the _mouth_ but the "pylome": some Filosa have two symmetrically placed pylomes. When the test is a mere pellicle, it may be recognised by the limitation of the pseudopodia to the one pylomic area. But the shell is often hard. In _Arcella_ (Fig. 10, C), a form common among Bog-mosses and Confervas, it is chitinous and shagreened, circular, with a shelf running in like that of a diving-bell around the pylome: there are two or more contractile vacuoles, and at least two nuclei. Like some other genera, it has the power of secreting carbonic acid gas in the form of minute bubbles in its cytoplasm, so as to enable it to float up to the surface of the water. The chitinous test shows minute hexagonal sculpturing, the expression of vertical partitions reaching from the inner to the outer layer.

{54}

Several genera have tests of siliceous or chitinous plates, formed in the cytoplasm in the neighbourhood of the nucleus, and connected by chitinous cement. Among these _Quadrula_ (Fig. 10, A) is Lobose, with square plates, _Euglypha_ (Fig. 8, p. 29), and _Paulinella_[69] are Filose, with hexagonal plates. In the latter they are in five longitudinal rows, with a pentagonal oral plate, perforated by the oval pylome. In other genera again, such as _Cyphoderia_ (Filosa), the plates are merely {55}chitinous. Again, the shell may be encrusted with sand-grains derived directly from without, or from ingested particles, as shown in _Centropyxis_, _Difflugia_ (Fig. 10, D), _Heleopera_, and _Campascus_ when supplied with powdered glass instead of sand. The cement in _Difflugia_ is a sort of organic mortar, infiltrated with ferric oxide (more probably ferric hydrate). In _Lecqueureusia spiralis_ (formerly united with _Difflugia_) the test is formed of minute sausage-shaped granules, in which may be identified the partly dissolved valves of Diatoms taken as food; it is spirally twisted at the apex, as if it had enlarged after its first formation, a very rare occurrence in this group. The most frequent mode of fission in the testaceous Rhizopods (Figs. 8, 10) is what Schaudinn aptly terms "bud-fission," where half the protoplasm protrudes and accumulates at the mouth of the shell, and remains till a test has formed for it, while the other half retains the test of the original animal. The materials for the shell, whether sand-granules or plates, pass from the depths of the original shell outwards into the naked cell, and through its cytoplasm to the surface, where they become connected by cementing matter into a continuous test. The nucleus now divides into two, one of which passes into the external animal; after this the two daughter-cells separate, the one with the old shell, the other, larger, with the new one.

If two individuals of the shelled species undergo bud-fission in close proximity, the offspring may partially coalesce, so that a monstrous shell is produced having two pylomes.

{56}Reproduction by fission has been clearly made out in most members of the group; some of the multinucleate species often abstrict a portion, sometimes at several points simultaneously, so that fission here passes into budding[70] (Fig. 9, 2, 6).

Brood-division, either by resolution in the multinucleate species, or preceded by multiple nuclear division in the habitually 1-nucleate, though presumably a necessary incident in the life-history of every species, has only been seen, or at least thoroughly worked out, in a few cases, where it is usually preceded by encystment, and mostly by the extrusion into the cyst of any undigested matter.[71]

In _Trichosphaerium_ (Fig. 9) the cycle described by Schaudinn is very complex, and may be divided into two phases, which we may term the A and the B subcycles. The members of the A cycle are distinguished by the gelatinous investment being armed with radial spicules, which are absent from the B form. The close of the A cycle is marked by the large multinucleate body resolving itself into amoeboid zoospores (3), which escape from the gelatinous test, and develop into the large multinucleate adults of the B form. These, like the A form, may reproduce by fission or budding. At the term of growth, however, they retract their pseudopodia, expel the excreta, and multiply their nuclei by mitosis (7). Then the body is resolved into minute 2-flagellate microzoospores (8), which are _exogamous_ gametes, _i.e._ they will only pair with similar zoospores from another cyst. The zygote (9-11) resulting from this conjugation is a minute amoeboid; its nucleus divides repeatedly, a gelatinous test is formed within which the spicules appear, and so the A form is reconstituted. In many of the test-bearing forms, whether Lobose or Filose, plastogamic unions occur, and the two nuclei may remain distinct, leading to plurinucleate monsters in their offspring by fission, or they may fuse and form a giant nucleus, a process which has here no relation to normal syngamy, as it is not associated with any marked change in the alternation of feeding and fission, etc. In _Trichosphaerium_ also plastogamic unions between small individuals have for their only result the increase of size, enabling the produce to deal with {57}larger prey. Temporary encystment in a "hypnocyst" is not infrequent in both naked and shelled species, and enables them to tide over drought and other unfavourable conditions.

Schaudinn has discovered and worked out true syngamic processes, some bisexual, some exogamous, in several other Rhizopods. In _Chlamydophrys stercorea_ the pairing-cells are equal, and are formed by the aggregation of the chromidia into minute nuclei around which the greater part of the cytoplasm aggregates, while the old nucleus (with a little cytoplasm) is lost. These brood-cells are 2-flagellate pairing-cells, which are exogamous: the zygote is a brown cyst; if this be swallowed by a mammal, the original _Chlamydophrys_ appears in its faeces.[72]

_Centropyxis aculeata_, a species very common in mud or moss, allied to _Difflugia_, also forms a brood by aggregation around nuclei derived from chromidia. The brood-cells are amoeboid, and secrete hemispherical shells like those of _Arcella_; some first divide into four smaller ones, before secreting the shell. Pairing takes place between the large and the small forms; and the zygote encysts. Weeks or months afterwards the cyst opens and its contents creep out as a minute _Centropyxis_. Finally, _Amoeba coli_ produces its zygote in a way recalling that of _Actinosphaerium_ (pp. 73-75, Fig. 21): the cell encysts; its nucleus divides, and each daughter divides again into two, which fuse reciprocally. Thus the cyst contains two zygote nuclei. After a time each of these divides twice, so that the mature cyst contains eight nuclei. Probably when swallowed by another animal they liberate a brood of eight young amoebae. Thus in different members of this group we have exogamy, both equal and bisexual, and endogamy.

Most of the Rhizopoda live among filamentous Algae in pools, ponds, and in shallow seas, etc.; some are "sapropelic" or mud-dwellers (many species of _Amoeba_, _Pelomyxa_, _Difflugia_, etc.), others frequent the roots of mosses. _Amoeba coli_ is often found as a harmless denizen of the large intestine of man. _Amoeba histolytica_, lately distinguished therefrom by Schaudinn, is the cause of tropical dysentery. It multiplies enormously in the gut, and is found extending into the tissues, and making its way into the abscesses that so frequently supervene in the liver and other organs. _Chlamydophrys stercorea_ is found in the {58}faeces of several mammals. The best monograph of this group is that of Penard.[73]

2. FORAMINIFERA[74]

_Sarcodina with no central capsule or distinction of ectosarc; the pseudopodia fine, branching freely, and fusing where they meet to form protoplasmic networks, or the outermost in the pelagic forms radiating, but without a central or axial filament: sometimes dimorphic, reproducing by fission and by rhizopod or flagellate germs in the few cases thoroughly investigated: all marine (with the exception of some of the Allogromidiaceae), and usually provided with a test of carbonate of lime ("vitreous" calcite, or "porcellanous" aragonite?), or of cemented particles of sand ("arenaceous"); test-wall continuous, or with the walls perforated by minute pores or interstices for the protrusion of pseudopodia._

The classification of Carpenter (into _Vitreous_ or _Perforate_, _Porcellanous_ or _Imperforate_, and _Arenaceous_), according to the structure of the shell, had proved too artificial to be used by Brady in the great Monograph of the Foraminifera collected by the "Challenger" Expedition,[75] and has been modified by him and others since then. We reproduce Lister's account of Brady's classification.[76] We must, however, warn the tyro that its characterisations are not definitions (a feature of all other recent systems), for rigid definitions are impossible: here as in the case, for instance, of many Natural Orders of Plants, transitional forms making the establishment of absolute boundaries out of the question. In the following classification we do not think it, therefore, necessary to complete the characterisations by noting the extremes of variation within the orders:—

1. Allogromidiaceae: simple forms, often fresh-water and similar to
Rhizopoda; test 0, or chitinous, gelatinous, or formed of cemented
particles, whether secreted platelets or ingested granules. _Biomyxa_,
Leidy = _Gymnophrys_, Cienk.; {59}_Diaphorodon_, Archer; _Allogromia_,
Rhumbl. (= _Gromia_, auctt.[77] nec Duj.) (Fig. 14, 1); _Lieberkühnia_,
Cl. and Lachm. (Fig. 12); _Microgromia_, R. Hertw. (Fig. 11);
_Pamphagus_, Bailey.

2. Astrorhizidaceae: test arenaceous, often large, never truly chambered,
or if so, asymmetrical. _Astrorhiza_, Sandahl; _Haliphysema_, Bowerb.;
_Saccammina_, M. Sars (Fig. 13, 1); _Loftusia_, Brady.

3. Lituolidaceae: test arenaceous, often symmetrical or regularly spiral,
isomorphous with calcareous forms: the chambers when old often
"labyrinthine" by the ingrowth of wall-material. _Lituola_, Lam.;
_Reophax_, Montf.; _Ammodiscus_, Reuss; _Trochammina_, Parker and
Jeffreys.

4. Miliolidaceae: test porcellanous, imperforate, spirally coiled or
cyclic, often chambered except in _Cornuspira_: simple in _Squamulina_.
_Cornuspira_, Max Sch.; _Peneroplis_, Montf.; _Miliolina_, Lam. (incl.
_Biloculina_ (Fig. 15), _Triloculina_, _Quinqueloculina_ (Figs. 14, 4;
15, B), _Spiroloculina_ (Fig. 13, 5) of d'Orb.); _Alveolina_, d'Orb.;
_Hauerina_, d'Orb.; _Calcituba_, Roboz; _Orbitolites_, Lam.;
_Orbiculina_, Lam.; _Alveolina_, Park. and Jeffr.; _Nubecularia_, Def.;
Squamulina, Max Sch. (Fig. 14, 3).

5. Textulariaceae: test calcareous, hyaline, perforated; chambers
increasing in size in two alternating rows, or three, or passing into a
spiral. _Textularia_, Def.; _Bulimina_, d'Orb.; _Cassidulina_, d'Orb.

6. Cheilostomellaceae: test vitreous, delicate, finely perforated,
chambered, isomorphic with the spiral forms of the Miliolidaceae.
_Cheilostomella_, Reuss.

7. Lagenaceae: Test vitreous, very finely perforate, chambers with a
distinct pylome projecting (ectosolenial), or turned in (entosolenial),
often succeeding to form a necklace-like shell. _Lagena_, Walker and Boys
(Fig. 13, 2); _Nodosaria_, Lam. (Fig. 13, 3); _Cristellaria_, Lam.;
_Frondicularia_, Def. (Fig. 13, 4); _Polymorphina_, Lam.; _Ramulina_,
Wright.

8. Globigerinidae: test vitreous, perforate; chambers few, dilated, and
arranged in a flat or conical spiral, usually with a crescentic pylome to
the last. _Globigerina_, d'Orb. (Figs. 13, 6; 16, 2); _Hastigerina_, Wyv.
Thoms.; _Orbulina_, d'Orb. (Fig. 16, 1).

9. Rotaliaceae; test vitreous, perforate, usually a conical spiral (like
a snail), chambers often subdivided into chamberlets, and with a proper
wall, and intermediate skeleton traversed by canals. _Rotalia_, Lam.
(Fig. 14, 2); _Planorbulina_, d'Orb. (Fig. 13, 9); _Polytrema_, Risso;
_Spirillina_, Ehr. (non-septate); _Patellina_, Will.; _Discorbina_, P.
and J. (Fig. 13, 7).

10. Nummulitaceae: test usually a complex spiral, the turns completely
investing their predecessors: wall finely tubular, often with a proper
wall and intermediate skeleton. _Fusulina_, Fisch.; _Polystomella_, Lam.;
_Nummulites_, d'Orb. (Fig. 13, 11); _Orbitoides_, d'Orb.

The Allogromidiaceae are a well-marked and distinct order, on the whole resembling the Rhizopoda Filosa, and are often found with them in fresh water, while all other Foraminifera are marine. The type genus, _Allogromia_ (Fig. 14, 1), has an oval chitinous shell. _Microgromia socialis_ (Fig. 11) is often found in aggregates, the pseudopodia of neighbours fusing where they meet into a {60}common network. This is due to the fact that one of the two daughter-cells at each fission, that does not retain the parent shell, remains in connexion with its sister that does: sometimes, however, it retracts its pseudopodia, except two which become flagella, wherewith it can swim off. The test of _Pamphagus_ is a mere pellicle. In _Lieberkühnia_ (Fig. 12) it is hardly that; though the body does not give off the fine pseudopodia directly, but emits a thick process or "stylopodium"[78] comparable to the protoplasm protruded through the pylome of its better protected allies; and from this, which often stretches back parallel to the elongated body, the reticulum of pseudopodia is emitted. _Diaphorodon_ has a shell recalling that of _Difflugia_ (Fig. 10, D, p. 55), formed of sandy fragments, but with interstices between them through which as well as through the two pylomes the pseudopodia pass. In all of these the shell is formed as in the Rhizopods once for all, and does not grow afterwards; and the fresh-water forms, which are the majority, have one or more contractile vacuoles; in _Allogromia_ they are very numerous, scattered on the expanded protoplasmic network.

{61}

The remaining marine families may all be treated of generally, before noting their special characters. Their marine habitat is variable, but in most cases restricted. A few extend up the brackish water of estuaries: a large number are found between tide-marks, or on the so-called littoral shelf extending to deep water; they are for the most part adherent to seaweeds, or lie among sand or on the mud. Other forms, again, are pelagic, such as _Globigerina_ (Figs. 13, 6, 16, 17) and its allies, and float as part of the plankton, having the surface of their shells extended by delicate spines, their pseudopodia long and radiating, and the outer part of their cytoplasm richly vacuolated ("alveolate"), and probably containing a liquid lighter than sea water, as in the Radiolaria. Even these, after their death and the decay of the protoplasm, must sink to the bottom (losing the fine spines by solution as they fall); and they accumulate there, to form a light oozy mud, the "Globigerina-ooze" of geographers, at depths where the carbonic acid under pressure is not adequate to dissolve the more solid calcareous matter. Grey Chalk is such an ooze, consolidated by {62}the lapse of time and the pressure of superincumbent layers. Some Foraminifera live on the sea bottom even at the greatest depths, and of course their shell is not composed of calcareous matter. Foraminifera may be obtained for examination by carefully washing sand or mud, collected on the beach at different levels between tide-marks, or from dredgings, or by carefully searching the surface of seaweeds, or by washing their roots, or, again, by the surface or deep-sea tow-net. The sand used to weight sponges for sale is the ready source of a large number of forms, and may be obtained for the asking from the sponge-dealers to whom it is a useless waste product. If this sand is dried in an oven, and then poured into water, the empty shells, filled with air, will float to the surface, and may be sorted by fine silk or wire gauze.

From the resemblance of the shells of many of them to the Nautilus they were at first described as minute Cephalopods, or Cuttlefish, by d'Orbigny,[79] and their true nature was only elucidated in the last century by the labours of Williamson, Carpenter, Dujardin, and Max Schultze. At first they possess only one nucleus, but in the adult stage may become plurinucleate without dividing, and this is especially the case in the "microsphaeric" states exhibited by many of those with a complex shell; the nucleus is apt to give off fragments (chromidia) which lie scattered in the cytoplasm. At first, too, in all cases, the shell has but a single chamber, a state that persists through life in some. When the number of chambers increases, their number has no relation to that of the nuclei, which remains much smaller till brood-formation sets in.

The shell-substance, if calcareous, has one of the two types, porcellanous or vitreous, that we have already mentioned, but _Polytrema_, a form of very irregular shape, though freely perforated, is of a lovely pink colour. In the calcareous shells sandy particles may be intercalated, forming a transition to the Arenacea. In these the cement has an organic base associated with calcareous or ferruginous matter; in some, however, the cement is a phosphate of iron. The porcellanous shells are often deep brown by transmitted light.

{63}

Despite the apparent uniformity of the protoplasmic body in this group, the shell is infinitely varied in form. As Carpenter writes, in reference to the Arenacea, "There is nothing more wonderful in nature than the building up of these elaborate and symmetrical structures by mere jelly-specks, presenting no traces {64}whatever of that definite organisation which we are accustomed to regard as necessary to the manifestations of conscious life.... The tests (shells) they construct when highly magnified bear comparison with the most skilful masonry of man. From the same sandy bottom one species picks up the coarsest quartz grains, unites them together with a ferruginous cement, and thus constructs a flask-shaped test, having a short neck and a single large orifice; another picks up the finer grains and puts them together with the same cement into perfectly spherical tests of the most extraordinary finish, perforated with numerous small pores disposed at pretty regular intervals. Another species selects the minutest sand grains and the terminal portions of sponge-spicules, and works them up together—apparently with no cement at all, but by the mere laying of the spicules—into perfect white spheres like homoeopathic globules, each showing a single-fissured orifice. And another, which makes a straight, many-chambered test, the conical mouth of each chamber projecting into the cavity of the next, while forming the walls of its chambers of ordinary sand grains rather loosely held together, shapes the conical mouths of the chambers by firmly cementing together the quartz grains which border it." The structure of the shell is indeed variable. The pylome may be single or represented by a row of holes (_Peneroplis_, _Orbitolites_), or, again, there may be several pylomes (_Calcituba_); and, again, there are in addition numerous scattered pores for the protrusion of pseudopodia elsewhere than from the stylopodium, in the whole of the "Vitrea" and in many "Arenacea"; and, as we shall see, this may exercise a marked influence on the structure of the shell.

In some cases the shell is simple, and in _Cornuspira_ and _Spirillina_ increases so as to have the form of a flat coiled tube. In _Calcituba_ the shell branches irregularly in a dichotomous way, and the older parts break away as the seaweed on which they grow is eaten away, and fall to the bottom, while the younger branches go on growing and branching. The fallen pieces, if they light on living weed, attach themselves thereto and repeat the original growth; if not, the protoplasm crawls out and finds a fresh weed and forms a new tube. In the "Polythalamia" new chambers are formed by the excess of the protoplasm emerging and surrounding itself with a shell, organically united with the existing chamber or chambers, and in a space-relation which follows definite laws characteristic of the species or of its stage of growth, so as to give rise to circular, spiral, or irregular complexes (see Fig. 13).

{65}

In most {66}cases the part of the previously existing chamber next the pylome serves as the hinder part of the new chamber, and the old pylome becomes the pore of communication. But in some of the "Perforata" each new chamber forms a complete wall of its own ("proper wall," Fig. 13, 8_b_), and the space between the two adjacent walls is filled with an intermediate layer traversed by canals communicating with the cavities of the chambers ("intermediate skeleton"), while an external layer of the same character may form a continuous covering. The shell of the Perforata may be adorned with pittings or fine spines, which serve to increase the surface of support in such floating forms as _Globigerina_, _Hastigerina_, and the like (Fig. 17). In the "Imperforata" the outer layer is often ornamented with regular patterns of pits, prominences, etc., which are probably formed by a thin reflected external layer of protoplasm. In some of the "Arenacea" a "labyrinthine" complex of laminae is formed.

A very remarkable point which has led to great confusion in the study of the Foraminifera, is the fact that the shell on which we base our characters of classification, may vary very much, even within the same individual. Thus in the genus _Orbitolites_ the first few chambers of the shell have the character of a Milioline, in _Orbiculina_ of a _Peneroplis_. The arrangements of the Milioline shell, known as Triloculine, Quinqueloculine, and Biloculine respectively, may succeed one another in the same shell (Figs. 14 4, 15). A shell may begin as a spiral and end by a straight continuation: again, the spherical _Orbulina_ (Fig. 16 1) is formed as an investment to a shell indistinguishable from _Globigerina_, which is ultimately absorbed. In some cases, as Rhumbler has pointed out, the more recent and higher development shows itself in the first formed chambers, while the later, younger chambers remain at a lowlier stage, as in the case of the spiral passing into a straight succession; but the other cases we have cited show that this is not always the case. In _Lagena_ (Fig. 13 2) the pylome is produced into a short tube, which may protrude from the shell or be turned into it, so that for the latter form the genus _Entosolenia_ was founded. Shells identical in minute sculpture are, however, found with either form of neck, and, moreover, the polythalamial shells (_Nodosaria_, Fig. 13 3), formed of a nearly straight succession of _Lagena_-like chambers, may have these chambers with their {67}communications on either type. Rhumbler goes so far as to suggest that all so-called _Lagena_ shells are either the first formed chamber of a _Nodosaria_ which has not yet become polythalamian by the formation of younger ones, or are produced by the separation of an adult _Nodosaria_ into separate chambers.

Many of the chambered species show a remarkable dimorphism, first noted by Schlumberger, and finally elucidated by J. J. Lister and Schaudinn. It reveals itself in the size of the initial chamber; accordingly, the two forms may be distinguished as "microspheric" and "megalospheric" respectively (Fig. 15), the latter being much the commoner. The microspheric form has always a plurality of nuclei, the megalospheric a single one, except at the approach of reproduction. Chromidial masses are, however, present in both forms. The life-history has been fully worked out in _Polystomella_ by Schaudinn, and in great part in _Polystomella_, _Orbitolites_, etc., by Lister; and the same scheme appears to be general in the class, at least where the dimorphism noted occurs. The microspheric form gives birth only to the megalospheric, but the latter may reproduce megalospheric broods, or give rise to swarmers, which by their (exogamous) {68}conjugation produce the microspheric young. The microspheric forms early become multinucleate, and have also numerous chromidia detached from the nuclei, which they ultimately replace. These collect in the outer part of the shell and aggregate into new nuclei, around which the cytoplasm concentrates, to separate into as many amoeboid young "pseudopodiospores" as there are nuclei. These escape from the shell or are liberated by its disintegration, and invest themselves with a shell to form the initial large central chamber or megalosphere.

In the ordinary life of the megalospheric form the greater part of the chromatic matter is aggregated into a nucleus, some still remaining diffused. At the end of growth the nucleus itself disintegrates, and the chromidia concentrate into a number of small vesicular nuclei, each of which appropriates to itself a small surrounding zone of thick plasm and then divides by mitosis twice; and the 4-nucleate cells so formed are resolved into as many 1-nucleate, 2-flagellate swarmers, which conjugate {69}only _exogamously_.[80] The fusion of their nuclei takes place after some delay: ultimately the zygote nucleus divides into two, a shell is formed, and we have the microsphere, which is thus pluri-nucleate _ab initio_. As we have seen, the nuclei of the microsphere are ultimately replaced by chromidia, and the whole plasmic body divides into pseudopodiospores, which grow into the megalospheric form.

In the Perforate genera, _Patellina_ and _Discorbina_, plastogamy precedes brood formation, the cytoplasms of the 2-5 pairing individuals contracting a close union; and then the nuclei proceed to break up _without fusion_, while the cytoplasm aggregates around the young nuclei to form amoebulae, which acquire a shell and separate. In both cases it is the forms with a single nucleus, corresponding to _megalospheric_ forms that so pair, and the brood-formation is, _mutatis mutandis_, the same as in these forms. Similar individuals may reproduce in the same way, in both genera, without this plastogamic pairing, which is therefore, though probably advantageous, not essential. If pseudopodiospores form their shells while near one another, they may coalesce to form monsters, as often happens in _Orbitolites_.[81]

The direct economic uses of the Foraminifera are perhaps greater than those of any other group of Protozoa. The Chalk is {70}composed largely of _Textularia_ and allied forms, mixed with the skeletons of Coccolithophoridae (pp. 113-114), known as Coccoliths, etc. The Calcaire Grossier of Paris, used as a building stone, is mainly composed of the shells of Miliolines of Eocene age; the Nummulites of the same age of the Mediterranean basin are the chief constituent of the stone of which the Pyramids of Egypt are built. Our own Oolitic limestones are composed of concretions around a central nucleus, which is often found to be a minute Foraminiferous shell.

The palaeontology of the individual genera is treated of in Chapman's and Lister's recent works. They range from the Lower Cambrian characterised by perforated hyaline genera, such as _Lagena_, to the present day. Gigantic arenaceous forms, such as _Loftusia_, are among the Tertiary representatives; but the limestones formed _principally_ of their shells commence at the Carboniferous. The so-called Greensands contain greenish granules of "glauconite," containing a ferrous silicate, deposited as a cast in the chambers of Foraminifera, and often left exposed by the solution of the calcareous shell itself. Such granules occur in deep-sea deposits of the present day.[82]

3. HELIOZOA

_Sarcodina with radiate non-anastomosing pseudopodia of granular protoplasm, each with a stiff axial rod passing into the body plasma; no central capsule, nor clear ectoplasm; skeleton when present siliceous; nucleus single or multiple; contractile vacuole (or vacuoles) in fresh-water species, superficial and prominent at the surface in diastole; reproduction by fission or budding in the active condition, or by brood-formation in a cyst, giving rise to resting spores; conjugation isogamous in the only two species fully studied; habitat floating or among weeds, mostly fresh water._

1. Naked or with an investment only when encysted.

APHROTHORACA.—_Actinolophus_ F.E. Sch.; _Myxastrum_ Haeck.;
_Gymnosphaera_ Sassaki; _Dimorpha_ (Fig. 37, 5, p. 112) Gruber;
_Actinomonas_ Kent; _Actinophrys_ Ehrb.; _Actinosphaerium_ St.;
_Camptonema_ Schaud; _Nuclearia_ Cienk.

{71}2. Invested with a gelatinous layer, sometimes traversed by a firmer
elastic network.

CHLAMYDOPHORA.—_Heterophrys_ Arch.; _Mastigophrys_ Frenzel;
_Acanthocystis_, Carter.

3. Ectoplasm with distinct siliceous spicules.

CHALAROTHORACA.—_Raphidiophrys_ Arch.

4. Skeleton a continuous, fenestrated shell, sometimes stalked.

DESMOTHORACA.—_Myriophrys_ Penard; _Clathrulina_ Cienk.; _Orbulinella_
Entz.

This class were at first regarded and described as fresh-water Radiolaria, but the differences were too great to escape the greatest living specialist in this latter group, Ernst Haeckel, who in 1866 created the Heliozoa for their reception. We owe our knowledge of it mainly to the labours of Cienkowsky, the late William Archer, F. E. Schulze, R. Hertwig, Lesser, and latterly to Schaudinn, who has monographed it for the "Tierreich" (1896); and Penard has published a more recent account.

_Actinophrys sol_ Ehrb. (Fig. 18) is a good and common type. It owes its name to its resemblance to a conventional drawing of the sun, with a spherical body and numerous close-set diverging rays. The cytoplasm shows a more coarsely vacuolated outer layer, sometimes called the ectosarc, and a denser internal layer the endosarc. In the centre of the figure is the large nucleus, to which the continuations of the rays may be seen to converge; the pseudopodia contain each a stiffish axial filament,[83] which is covered by the fine granular plasm, showing currents of the granules. The axial filament disappears when the pseudopodia are retracted or bent, and is regenerated afterwards. This bending occurs when a living prey touches and adheres to a ray, all its neighbours bending in like the tentacles of a Sundew. The prey is carried down to the surface of the ectoplasm, and {72}sinks into it with a little water, to form a nutritive vacuole. Fission is the commonest mode of reproduction, and temporary plastogamic unions are not uncommon. Arising from these true conjugations occur, two and two, as described by Schaudinn. A gelatinous cyst wall forms about the two which are scarcely more than in contact with their rays withdrawn. Then in each the nucleus divides into two, one of which passes to the surface, and is lost (as a "polar body"), while the other approaches the corresponding nucleus of the mate, and unites with it, while at the same time the cytoplasms fuse. Within the gelatinous cyst the zygote so formed divides to produce two sister resting spores, from each of which, after a few days, a young _Actinophrys_ escapes, as may take place indeed after encystment of an ordinary form without conjugation.

The axial rods of the pseudopodia may pass either to the circumference of the nucleus or to a central granule, corresponding, it would appear, to a centrosome or blepharoplast; or again, {73}in the plurinucleate marine genus _Camptonema_, each rod abuts on a separate cap on the outer side of each nucleus. The nucleus is single in all but the genera _Actinosphaerium_, _Myxastrum_, _Camptonema_, and _Gymnosphaera_. The movements of this group are very slow, and are not well understood. A slow rolling over on the points of the rays has been noted, and in _Camptonema_ they move very decidedly to effect locomotion, the whole body also moving Amoeba-fashion; but of the distinct movements of the species when floating no explanation can be given. The richly vacuolate ectoplasm undoubtedly helps to sustain the cell, and the extended rays must subserve the same purpose by so widely extending the surface. _Dimorpha_ (Fig. 37, 5, p. 112) has the power of swimming by protruding a pair of long flagella from the neighbourhood of the eccentric nucleus; and _Myriophrys_ has an investment of long flagelliform cilia. _Actinomonas_ has a stalk and a single flagellum in addition to the pseudopodia; these genera form a transition to the Flagellata.

Several species habitually contain green bodies, which multiply by bipartition, and are probably Zoochlorellae, Chlamydomonadidae of the same nature as we shall find in certain Ciliata (pp. 154, 158) in fresh-water Sponges (see p. 175), in _Hydra viridis_ (p. 256), and the marine Turbellarian _Convoluta_ (Vol. II. p. 43).

Reproduction by fission is not rare, and in some cases (_Acanthocystis_) the cell becomes multinuclear, and buds off 1-nucleate cells. In such cases the buds at first lack a centrosome, and a new one is formed first in the nucleus, and passes out into the cytoplasm. These buds become 2-flagellate before settling down. In _Clathrulina_ the formation of 2-flagellate zoospores has long been known (Fig. 20, 3). In _Actinosphaerium_ (Figs. 19, 21), a large species, differing from _Actinophrys_ only in the presence of numerous nuclei in its endoplasm, a peculiar process, which we have characterised as _endogamy_, results in the formation of resting spores. The animal retracts its rays and encysts; and the number of nuclei is much reduced by their mutual fusion, or by the solution of many of them, or by a combination of the two processes. The body then breaks up into cells with a single nucleus, and each of these surrounds itself with a wall to form a cyst of the second order.

{74}

Each of these divides, and the two sister cells then conjugate after the same fashion as in _Actinophrys_, but the nuclear divisions to form the coupling nucleus are two in number, _i.e._ the nucleus divides into two, one of which goes to the surface as the first polar body, and the sister of this again divides to form a second polar body (which also passes to {75}the surface) and a pairing nucleus.[84] The two cells then fuse completely, and surround themselves with a second gelatinous cyst wall, separated from the outer one by a layer of siliceous spicules. The nucleus appears to divide at least twice before the young creep out, to divide immediately into as many _Actinophrys_-like cells as there were nuclei; then each of these multiplies its nuclei, to become apocytial like the adult form.

Schaudinn admits 24 genera (and 7 doubtful) and 41 species (and 18 doubtful). None are known fossil. Their geographical distribution is cosmopolitan, as is the case with most of the minute fresh-water Protista; 8 genera are exclusively marine, and _Orbulinella_ has only been found in a salt-pond; _Actinophrys sol_ is both fresh-water and marine, and _Actinolophus_ has 1 species fresh-water, the other marine. One of the 14 species of _Acanthocystis_ is marine; the remaining genera and species are all inhabitants of fresh water.[85]

4. RADIOLARIA

_Sarcodina with the protoplasm divided by a perforated chitinous central capsule into a central mass surrounding the nucleus, and an outer layer; the pseudopodia radiate, never anastomosing enough to form a marked network; skeleton either siliceous, of spicules, or perforated; or of definitely arranged spicules of proteid matter (acanthin), sometimes also coalescing into a latticed shell; reproduction by fission and by zoospores formed in the central capsule. Habitat marine, suspended at the surface (plankton), at varying depths (zonarial), or near the bottom (abyssal)._

{76}

The following is Haeckel's classification of the Radiolaria:—

I. PORULOSA (HOLOTRYPASTA).—Homaxonic, or nearly so. Central capsule
spherical in the first instance; pores numerous, minute, scattered;
mostly pelagic.

A. SPUMELLARIA (PERIPYLAEA).—Pores evenly scattered; skeleton of solid
siliceous spicules, or continuous, and reticulate or latticed, rarely
absent; nucleus dividing late, as an antecedent to reproduction.

B. ACANTHARIA (ACTIPYLAEA).—Pores aggregated into distinct areas;
skeleton of usually 20 centrogenous, regularly radiating spines of
acanthin, whose branches may coalesce into a latticed shell; nucleus
dividing early.

II. OSCULOSA (MONOTRYPASTA).—Monaxonic; pores of central capsule limited
to the basal area (osculum), sometimes accompanied by two (or more)
smaller oscula at apical pole, mostly zonarial or abyssal.

C. NASSELLARIA (MONOPYLAEA).—Central capsule ovoid, of a single layer;
pores numerous on the operculum or basal field; skeleton siliceous,
usually with a principal tripod or calthrop-shaped spicule passing, by
branching, into a complex ring or a latticed bell-shaped shell; nucleus
eccentric, near apical pole.

D. PHAEODARIA (CANNOPYLAEA, Haeck.; TRIPYLAEA, Hertw.).—Central capsule
spheroidal, of two layers, in its outer layer an operculum, with
radiate ribs and a single aperture, beyond which protrudes the outer
layer; osculum basal, a dependent tube (proboscis); accessory oscula,
when present, simpler, usually two placed symmetrically about the
apical pole; skeleton siliceous, with a combination of organic matter,
often of hollow spicules; nucleus sphaeroidal, eccentric; extracapsular
protoplasm containing an accumulation of dusky pigment granules
("phaeodium").

{77}

A. SPUMELLARIA.

Sublegion (1). COLLODARIA.[86]—Skeleton absent or of detached spicules;
colonial or simple.

Order i. COLLOIDEA.—Skeleton absent. (Families 1, 2.) _Thalassicolla_
Huxl.; _Thalassophysa_ Haeck.; _Collozoum_ Haeck.; _Collosphaera_ J.
Müll.; _Actissa_ Haeck.

Order ii. BELOIDEA.—Skeleton spicular. (Families 3, 4.)

Sublegion (2). SPHAERELLARIA.—Skeleton continuous, latticed or spongy,
reticulate.

Order iii. SPHAEROIDEA.—Skeleton of one or several concentric spherical
shells; sometimes colonial. (Families 5-10.) _Haliomma_ Ehrb.;
_Actinomma_ Haeck. (Fig. 23).

Order iv. PRUNOIDEA.—Skeleton a prolate sphaeroid or cylinder,
sometimes constricted towards the middle, single or concentric.
(Families 11-17.)

Order v. DISCOIDEA.—Shell flattened, of circular plan, simple or
concentric, rarely spiral. (Families 18-23.)

Order vi. LARCOIDEA.—Shell ellipsoidal, with all three axes unequal or
irregular, sometimes becoming spiral. (Families 24-32.)[87]

{78}

B. ACANTHARIA.

Order vii. ACTINELIDA.—Radial spines numerous, more than 20, usually
grouped irregularly. (Families 33-35.) _Xiphacantha_ Haeck.

Order viii. ACANTHONIDA.—Radial spines equal. (Families 36-38.)

Order ix. SPHAEROPHRACTA.—Radial spines 20, with a latticed spherical
shell, independent of, or formed from the reticulations of the spines.
(Families 39-41.) _Dorataspis_ Haeck. (Fig. 25, A).

Order x. PRUNOPHRACTA.—Radial spines 20, unequal; latticed shell,
ellipsoidal, lenticular, or doubly conical. (Families 42-44.)

C. NASSELLARIA.

Order xi. NASSOIDEA.—Skeleton absent. (Family 45.)

Order xii. PLECTOIDEA.—Skeleton of a single branching spicule, the
branches sometimes reticulate, but never forming a latticed shell or a
sagittal ring. (Families 46-47.)

Order xiii. STEPHOIDEA.—Skeleton with a sagittal ring continuous with
the branched spicule, and sometimes other rings or branches. (Families
48-51.) _Lithocercus_ Théel (Fig. 26, A).

Order xiv. SPYROIDEA.—Skeleton with a latticed shell developed around
the sagittal ring (cephalis), and constricted in the sagittal plane,
with a lower chamber (thorax) sometimes added. (Families 52-55.)

{79}Order xv. BOTRYOIDEA.—As in Spyroidea, but with the cephalis 3-4
lobed; lower chambers, one or several successively formed. (Families
56-58.)

Order xvi. CYRTOIDEA.—Shell as in the preceding orders, but without
lobing or constrictions. (Families 59-70.) _Theoconus_ Haeck. (Fig.
25, B).

D. PHAEODARIA.

Order xvii. PHAEOCYSTINA.—Skeleton 0 or of distinct spicules; capsule
centric. (Families 71-73.) _Aulactinium_ Haeck. (Fig. 26, B).

Order xviii. PHAEOSPHAERIA.—Skeleton a simple or latticed sphere, with
no oral opening (pylome); capsule central. (Families 74-77.)

Order xix. PHAEOGROMIA.—Skeleton a simple latticed shell with a pylome
at one end of the principal axis; capsule excentric, sub-apical.
(Families 78-82.) _Pharyngella_ Haeck.; _Tuscarora_ Murr.;
_Haeckeliana_ Murr. (Fig. 28).

Order xx. PHAEOCONCHIA.—Shell of two valves, opening in the plane
("frontal") of the three openings of the capsule. (Families 83-85.)

We exclude Haeckel's Dictyochida, with a skeleton recalling that of the Stephoidea, but of the impure hollow substance of the Phaeodaria (p. 84). They rank now as Silicoflagellates (p. 114).

The Radiolarian is distinguished from all other Protozoa by the chitinous central capsule, so that its cytoplasm is separated into an outer layer, the _extracapsular_ protoplasm (ectoplasm), and a central mass, the _intracapsular_, containing the nucleus.[88]

The _extracapsular_ layer forms in its substance a gelatinous mass, of variable reaction, through which the plasma itself ramifies as a network of threads ("sarcodictyum"), uniting at the surface to constitute the foundation for the pseudopodia. This gelatinous matter constitutes the "calymma." It is largely vacuolated, the vacuoles ("alveoli"), of exceptional size, lying in the nodes of the plasmic network, and containing a liquid probably of lower specific gravity than seawater; and they are especially abundant towards the surface, where they touch and become polygonal. On mechanical irritation they disappear, to be formed anew after an interval, a fact that may explain the sinking from the surface in disturbed water. This layer may contain minute pigment granules, but the droplets of oil and of albuminous matter frequent in the central layer are rare here. {80}The "yellow cells" of a symbiotic Flagellate or Alga, _Zooxanthella_, are embedded in the jelly of all except Phaeodaria, and the whole ectosarc has the average consistency of a firm jelly.

The _pseudopodia_ are long and radiating, with a granular external layer, whose streaming movements are continuous with those of the inner network. In the Acantharia they contain a firm axial filament, like that of the Heliozoa, which is traceable to the central capsule; and occasionally a bundle of pseudopodia may coalesce to form a stout process like a flagellum ("sarcoflagellum"). Here, too, each spine, at its exit from the jelly, is surrounded by a little cone of contractile filaments, the _myophrisks_, whose action seems to be to pull up the jelly and increase the volume of the spherical body so as to diminish its density.

The _intracapsular protoplasm_ is free from _Zooxanthella_ except in the Acantharia. It is less abundantly vacuolated, and is finely granular. In the Porulosa it shows a radial arrangement, with pyramidal stretches of hyaline plasma separated by intervals rich in granules. Besides the alveoli with watery contents, others are present with albuminoid matter in solution. Oil-drops, often brilliantly coloured, occur either in the plasma or floating in either kind of vacuole; and they are often luminous at night. Added to these, the intracapsular plasm contains pigment-granules, most frequently red or orange, {81}passing into yellow or brown, though violet, blue, and green also occur. The "phaeodium,"[89] however, that gives its name to the Phaeodaria, is an aggregate of dark grey, green, or brown granules which are probably formed in the endoplasm, but accumulate in the extracapsular plasm of the oral side of the central capsule. Inorganic concretions and crystals are also found in the contents of the central capsule, as well as aggregates of unknown composition, resembling starch-grains in structure.

In the Monopylaea, or Nassellaria (Figs. 25, B, 26, A), the endoplasm is differentiated above the perforated area of the central capsule into a cone of radiating filaments termed the "porocone," which may be channels for the communication between the exoplasm and the endoplasm, or perhaps serve, as Haeckel suggests, to raise, by their contraction, the perforated area: he compares them to the myophane striae of Infusoria. In the Phaeodaria (Fig. 26, B), a radiating laminated cone is seen in the outermost layer of the endoplasm above the principal opening ("astropyle"), and a fibrillar one around the two accessory ones ("parapyles"); and in some cases, continuous with these, the whole outer layer of the endoplasm shows a meridional striation.

The _nucleus_ is contained in the endoplasm, and is always at first single, though it may divide again and again. The nuclear wall is a firm membrane, sometimes finely porous. If there are concentric shells it at first occupies the innermost, which it may actually come to enclose, protruding lobes which grow through the several perforations of the lattice-work, finally coalescing outside completely, so as to show no signs of the joins. In the Nassellaria a similar process usually results in the formation of a lobed nucleus, contained in an equally lobed central capsule. The chromatin of the nucleus may be concentrated into a central mass, or distributed into several "nucleoli," or it may assume the form of a twisted, gut-like filament, or, again, the nuclear plasm may be reticulated, with the chromatin deposited at the nodes of the network.

{82}

The skeleton of this group varies, as shown in our conspectus, in the several divisions.[90] The Acantharia (Figs. 24, 25, A) have a skeleton of radiating spines meeting in the centre of figure of the endoplasm, and forcing the nucleus to one side. The spines are typically 20 in number, and emerge from the surface of the regular spherical forms (from which the others may be readily derived) radially, in five sets of four in the regions corresponding to the equator and the tropics and polar circles of our world. {83}The four rays of adjacent circles alternate, so that the "polar" and "equatorial" rays are on one set of meridians 90° apart, and the "tropical" spines are on the intermediate meridians, as shown in the figures. By tangential branching, and the meeting or coalescence of the branches, reticulate (Figs. 23, 24, 25) and latticed shells are formed in some families, with circles of openings or pylomes round the bases of the spines. In the Sphaerocapsidae the spines are absent, but their original sites are inferred from the 20 circles of pylomes.

In the Spumellaria the simplest form of the (siliceous) skeleton is that of detached spicules, simple or complex, or passing into a latticed shell, often with one or more larger openings (pylomes). Radiating spines often traverse the whole of the cavity, becoming continuous with its latticed wall, and bind firmly the successive zones when present (Fig. 23).

_Calcaromma calcarea_ was described by Wyville Thomson as having a shell of apposed calcareous discs, and _Myxobrachia_, by Haeckel, as having collections of the calcareous Coccoliths and Coccospheres. In both cases we have to do with a Radiolarian not possessing a skeleton, but retaining the undigested shells of its food, in the former case (_Actissa_) in a continuous layer, in the latter (_Thalassicolla_) in accumulations that, by their weight, droop and pull out the lower hemisphere into distinct arms.

The (siliceous) skeleton of the Nassellaria is absent only in the Nassoidea, and is never represented by distinct spicules. Its simplest form is a "tripod" with the legs downward, and the central capsule resting on its apex. The addition of a fourth limb converts the tripod into a "calthrop," the central capsule in this case resting between the upturned leg and two of the lower three regarded as the "anterolateral"; the odd lower leg, like the upturned one, being "posterior." Again, the skeleton may present a "sagittal ring," often branched and spiny (Fig. 26, A), or combined with the tripod or calthrop, or complicated by the addition of one or more horizontal rings. Another type is presented by the "latticed chamber" surrounding the central capsule, with a wide mouth ("pylome") below. This is termed the "cephalis"; it may be combined in various ways with the sagittal ring and the tripod or calthrop; and, again, it may be prolonged by the addition of one, two, or three chambers below, {84}the last one opening by a pylome (Fig. 25, B). These are termed "thorax," "abdomen," and "post-abdomen" respectively.

In the Phaeodaria the skeleton may be absent, spicular (of loose or connected spicules) or latticed, continuous or bivalve. It is composed of silica combined with organic matter, so that it chars when heated, is more readily dissolved, and is not preserved in fossilisation. The spicules or lattice-work are hollow, often with a central filament running in the centre of the gelatinous contents. The latticed structure of the shell of the Challengeridae (Fig. 28) is so fine as to recall that of the Diatomaceae. In the Phaeoconchida the shell is in two halves, parted along the "frontal" plane of the three apertures of the capsule.

The central capsule (rarely inconspicuous and difficult, if not impossible to demonstrate) is of a substance which resembles chitin, though its chemical reactions have not been fully studied hitherto, and indeed vary from species to species. It is composed of a single layer, except in _Phaeodaria_, where it is double. The operculum in this group, _i.e._ the area around the aperture, is composed of an outer layer, which is radially thickened, and a thin inner layer; the former is produced into the projecting tube ("proboscis").

REPRODUCTION in the Radiolaria may be simple fission due to the binary fission of the nucleus, the capsule, and the ectoplasm in succession. If this last feature is omitted we have a colonial organism, composed of the common ectoplasm containing numerous central capsules; and the genera in which this occurs, all belonging to the Peripylaea, were formerly separated (as Polycyttaria) from {85}the remaining Radiolaria (Monocyttaria). They may either lack a skeleton (Collozoidae, Fig. 22), or have a skeleton of detached spicules (Sphaerozoidae), or possess latticed shells (Collosphaeridae) one for each capsule, and would seem therefore to belong, as only differentiated by their colonial habit, to the several groups having these respective characters. Fission has been well studied in _Aulacantha_ (a Phaeodarian) by Borgert.[91] He finds that in this case the skeleton is divided between the daughter-cells, and the missing part is regenerated. In cases where this is impossible one of the daughter-cells retains the old skeleton, and the other escapes as a bud to form a new skeleton.

Two modes of reproduction by flagellate zoospores have been described (Fig. 22). In the one mode all the zoospores are alike—isospores—and frequently contain a crystal of proteid nature as well as oil-globules. In the Polycyttaria alone has the second mode of spore-formation been seen, and that in the same species in which the formation of isospores occurs. Here "anisospores" are formed, namely, large "mega-," and small "micro-zoospores." They probably conjugate as male and female respectively; but neither has the process been observed, nor has any product of such conjugation (zygote) been recognised. In every case the formation of the zoospores only involves the {86}endoplasm: the nucleus first undergoes brood division, and the plasma within the capsule becomes concentrated about its offspring, and segregates into the spores; the extracapsular plasm disintegrates.[92]

The YELLOW CELLS (_Zooxanthella_), so frequently found in the Radiolaria were long thought to be constituents of their body. Cienkowsky found that when the host died from being kept in unchanged water, the yellow cells survived and multiplied freely, often escaping from the gelatinised cell-wall as biflagellate zoospores. The cell-wall is of cellulose. The cell contains two chloroplastids, or plates coloured with the vegetal pigment "diatomin." Besides ordinary transverse fission in the ordinary encysted state in the ectoplasm of the host, when free they may pass into what is known as a "_Palmella_-state," the cell-walls gelatinising; in this condition they multiply freely, and constitute a jelly in which the individual cells are seen as rounded bodies. They contain starch in two forms—large hollow granules, not doubly refractive, and small solid granules which polarise light. We may regard them as Chrysomonadaceae (p. 113). Similar organisms occur in many Anthozoa (see pp. 261, 339, 373 f., 396). Diatomaceae (yellow Algae with silicified cell-walls) sometimes live in the jelly of certain _Collosphaera_. Both these forms live in the state known as "symbiosis" with their host; _i.e._ they are in mutually helpful association, the Radiolarian absorbing salts from the water for the nutrition of both, and the Alga or Flagellate taking up the CO_{2} due to the respiration of the host, and building up organic material, the surplus of which is doubtless utilised, at least in part, for the nutrition of the host. A similar union between a Fungus and a coloured vegetal ("holophytic") organism is known as a Lichen.

The Suctorian Infusorian _Amoebophrya_ is parasitic in the ectoplasm of certain Acantharia, and in the peculiar genus _Sticholonche_ which appears to be intermediate between this group and Heliozoa.

The Silicoflagellate family Dictyochidae are found temporarily {87}embedded in the ectoplasm of some of the Phaeocystina, and have a skeleton of similar nature. Their true nature was shown by Borgert.

The Amphipod crustacean _Hyperia_[93] may enter the jelly of the colonial forms, and feed there at will on the host.[94]