Cyathea
Content
Morphology
Dermal appendages. Scales on the stipes of Cyathea are of two kinds, and these appear to provide the best subdivision of the genus, at least in Malaysia. The two types of scale are called flabelloid and setiferous (see ).
Flabelloid scales () have a broad median portion consisting of longitudinally elongated cells with all walls thickened, and edges, of varying width, consisting of thin-walled cells diverging fan-wise outwards, with irregularly projecting marginal cells, some of them sometimes thick-walled and dark, often flexuous (). The scales develop at the apex of more or less massive (multicellular) outgrowths from the surface of the stipe and at right angles to these outgrowths (thus parallel to the surface of the stipe); the base of a scale is peltate, with a narrow part encircling the supporting outgrowth on the basiscopic side. The outgrowths become very large and spine-like in some species, in others they are quite small. As one proceeds from the base of the stipe to the ultimate axes, the scales become progressi- vely smaller, and their character changes in ways characteristic of individual species. These scales provide some of the most important diagnostic characters in Cyathea.
Setiferous scales (, c0 also develop at the apex of outgrowths from the surface of the stipe, but not at right angles to the outgrowths; the base of a scale widens more or less abruptly from the apex of the outgrowth, and in C. sangirensis may be seen transitions from stout erect hairs (like those of Dicksonia) to setiferous scales. All cells in setiferous scales are longitudinally elongate and all have walls of equal thickness (they are rarely so thick as in flabelloid scales); some marginal cells grow obliquely outwards at their distal ends to form straight or outcurved rigid, usually dark, setae on the edge of the scale ().
Flabelloid scales () have a broad median portion consisting of longitudinally elongated cells with all walls thickened, and edges, of varying width, consisting of thin-walled cells diverging fan-wise outwards, with irregularly projecting marginal cells, some of them sometimes thick-walled and dark, often flexuous (). The scales develop at the apex of more or less massive (multicellular) outgrowths from the surface of the stipe and at right angles to these outgrowths (thus parallel to the surface of the stipe); the base of a scale is peltate, with a narrow part encircling the supporting outgrowth on the basiscopic side. The outgrowths become very large and spine-like in some species, in others they are quite small. As one proceeds from the base of the stipe to the ultimate axes, the scales become progressi- vely smaller, and their character changes in ways characteristic of individual species. These scales provide some of the most important diagnostic characters in Cyathea.
Setiferous scales (, c0 also develop at the apex of outgrowths from the surface of the stipe, but not at right angles to the outgrowths; the base of a scale widens more or less abruptly from the apex of the outgrowth, and in C. sangirensis may be seen transitions from stout erect hairs (like those of Dicksonia) to setiferous scales. All cells in setiferous scales are longitudinally elongate and all have walls of equal thickness (they are rarely so thick as in flabelloid scales); some marginal cells grow obliquely outwards at their distal ends to form straight or outcurved rigid, usually dark, setae on the edge of the scale ().
Taxonomy
Early authors attempted to distinguish genera (within the genus Cyathea as here recognized) by characters of the indusium, whether cup-shaped, attached to one side of the receptacle, or absent. These three conditions do not cover all cases, and indusia have often been inadequately described. Species lacking indusia often resemble indusiate species more closely than other exindusiate ones. It thus appears that the exindusiate condition has arisen on more than one evolutionary line, and it does not give a natural subdivision of the genus. Presl attempted also to distinguish species in which the receptacle splits into two halves (Disphenia, Dichorexia), but this is not a significant characters.
Copeland at first united all Malaysian species in the genus Cyathea () but later attempted to distinguish the three genera Cyathea, Gymnosphaera and Schizocaena (), a division which I criticized (). Most of the species of Copeland's Gymnosphaera sect. 3 appear to be closely related to the type species of Schizocaena. Elim- inating these, I have found it extremely difficult to make a clear-cut Separation between Gymnosphaera and some species included by Copeland in Cyathea. Copeland's Cyathea is divisible, on the basis of scale-characters (see above) into two groups which I call subg. Cyathea (flabelloid scales) and subg. Sphaeropteris (setiferous scales). Gymnosphaera (excluded Copeland's sect. 3) is then a section of subg. Cyathea, and Schizocaena a section of subg. Sphaeropteris.
In Cyathea subg. Cyathea all possible conditions of the indusium occur, and it appears that species with cup-shaped indusia can be closely related to others with hemitelioid indusia; in some species, e.g. C. javaniea Bl. () and C. hymenodes Mett., intermediate conditions may occur on the same leaflet as typical indusia. Some species have very large hemitelioid indusia which cover the sorus almost to maturity, often breaking later (C. loheri Christ, C. oinops Hassk., ); these have usually not been distinguished from species in which the indusium is at first quite complete (e.g. C. crenulata Bl., ). Other species show various stages of reduction of the hemitelioid type of indusium; in many cases this is quite hidden by the mature sorus () and has been reported as lacking (the species thus being placed in Alsophila). In subg. Sphaeropteris the hemitelioid condition has not been found. Most species of this subgenus have a complete indusium, breaking at maturity (never truly cup-shaped) or none; two cases where partial indusia occur (C. alternans (Wall.) Pr. and C. diseophora Holttum) appear to be intermediate between fully exindusiate and exindusiate species and are probably hybrids.
Descriptions of species of Cyathea have rarely been satisfactory, and misidentifications have been frequent; hence names in collectors' lists, and distribution data based on them, are often unreliable. For clear distinction between species, detailed descriptions of scales, hairs and indusia are essential, and often these cannot be seen satisfactorily with a X 10 lens. The only authors who described such details adequately were Mettenius and Christensen. Because of inadequate early descriptions, many species have been named more than once, and the only way to know this is to examine type material. I have seen such material of almost all the 350 species described from Malaysia, and have examined also types of species from the mainland of Asia and from the Pacific.
Subdivision of the genus Cyathea: recent proposals:
R. M. TRYON () has recognized Sphaeropteris BERNH. as a distinct genus, in which he includes the tropical American species mentioned on p. 124 of the present work, and also several others which disagree in scales and sori from the specification on p. 65; in my opinion the latter should be excluded. TRYON limits the genus Cyathea to some tropical American species, distinguishing them from all Malesian species here included in Cyathea subg. Cyathea solely on the lack of a seta at the apex of stipe-scales. He has transferred all Malesian species of Cyathea subg. Cyathea sensu HOLTTUM 1963 to the genus Alsophila (type species A. australis R. BR.).
I agree that subg. Sphaeropteris as defined on p. 76 is clearly distinct from all other members of the family, but I cannot agree that the remaining species, including those of tropical America, are divisible into natural groups on such clearly defined characters. CONANT has shown that hybrids exist between species of different genera as recognized by TRYON, and in one case such hybrids have good spores ().
All species so far examined, of several different genera as recognized by TRYON (including Sphaeropteris) have the chromosome number 69. This is a very strong indication that Cyathea in the broad sense adopted in Flora Malesiana is a phy- letic unity, and I adhere to my recognition of it as a single genus. As Sphaeropteris is the only sub-group separable on well- defined characters, I still place all the rest in a subgenus Cyathea, the further subdivision of which seems to me still un- certain. The species C. australis (R. BR.) DOMIN (type of Alsophila) has spores which differ considerably from those of the majority of Malesian species transferred to Alsophila by TRYON (see . 751).
Taxonomy of Malesian species of Cyathea: Since 1963 many new collections have been made, especially in New Guinea. I have not been able to examine all of these. The following data are based on specimens which have come to my attention; probably more new species remain to be recognized, and new information about many species remains unrecorded.
Copeland at first united all Malaysian species in the genus Cyathea () but later attempted to distinguish the three genera Cyathea, Gymnosphaera and Schizocaena (), a division which I criticized (). Most of the species of Copeland's Gymnosphaera sect. 3 appear to be closely related to the type species of Schizocaena. Elim- inating these, I have found it extremely difficult to make a clear-cut Separation between Gymnosphaera and some species included by Copeland in Cyathea. Copeland's Cyathea is divisible, on the basis of scale-characters (see above) into two groups which I call subg. Cyathea (flabelloid scales) and subg. Sphaeropteris (setiferous scales). Gymnosphaera (excluded Copeland's sect. 3) is then a section of subg. Cyathea, and Schizocaena a section of subg. Sphaeropteris.
In Cyathea subg. Cyathea all possible conditions of the indusium occur, and it appears that species with cup-shaped indusia can be closely related to others with hemitelioid indusia; in some species, e.g. C. javaniea Bl. () and C. hymenodes Mett., intermediate conditions may occur on the same leaflet as typical indusia. Some species have very large hemitelioid indusia which cover the sorus almost to maturity, often breaking later (C. loheri Christ, C. oinops Hassk., ); these have usually not been distinguished from species in which the indusium is at first quite complete (e.g. C. crenulata Bl., ). Other species show various stages of reduction of the hemitelioid type of indusium; in many cases this is quite hidden by the mature sorus () and has been reported as lacking (the species thus being placed in Alsophila). In subg. Sphaeropteris the hemitelioid condition has not been found. Most species of this subgenus have a complete indusium, breaking at maturity (never truly cup-shaped) or none; two cases where partial indusia occur (C. alternans (Wall.) Pr. and C. diseophora Holttum) appear to be intermediate between fully exindusiate and exindusiate species and are probably hybrids.
Descriptions of species of Cyathea have rarely been satisfactory, and misidentifications have been frequent; hence names in collectors' lists, and distribution data based on them, are often unreliable. For clear distinction between species, detailed descriptions of scales, hairs and indusia are essential, and often these cannot be seen satisfactorily with a X 10 lens. The only authors who described such details adequately were Mettenius and Christensen. Because of inadequate early descriptions, many species have been named more than once, and the only way to know this is to examine type material. I have seen such material of almost all the 350 species described from Malaysia, and have examined also types of species from the mainland of Asia and from the Pacific.
Subdivision of the genus Cyathea: recent proposals:
R. M. TRYON () has recognized Sphaeropteris BERNH. as a distinct genus, in which he includes the tropical American species mentioned on p. 124 of the present work, and also several others which disagree in scales and sori from the specification on p. 65; in my opinion the latter should be excluded. TRYON limits the genus Cyathea to some tropical American species, distinguishing them from all Malesian species here included in Cyathea subg. Cyathea solely on the lack of a seta at the apex of stipe-scales. He has transferred all Malesian species of Cyathea subg. Cyathea sensu HOLTTUM 1963 to the genus Alsophila (type species A. australis R. BR.).
I agree that subg. Sphaeropteris as defined on p. 76 is clearly distinct from all other members of the family, but I cannot agree that the remaining species, including those of tropical America, are divisible into natural groups on such clearly defined characters. CONANT has shown that hybrids exist between species of different genera as recognized by TRYON, and in one case such hybrids have good spores ().
All species so far examined, of several different genera as recognized by TRYON (including Sphaeropteris) have the chromosome number 69. This is a very strong indication that Cyathea in the broad sense adopted in Flora Malesiana is a phy- letic unity, and I adhere to my recognition of it as a single genus. As Sphaeropteris is the only sub-group separable on well- defined characters, I still place all the rest in a subgenus Cyathea, the further subdivision of which seems to me still un- certain. The species C. australis (R. BR.) DOMIN (type of Alsophila) has spores which differ considerably from those of the majority of Malesian species transferred to Alsophila by TRYON (see . 751).
Taxonomy of Malesian species of Cyathea: Since 1963 many new collections have been made, especially in New Guinea. I have not been able to examine all of these. The following data are based on specimens which have come to my attention; probably more new species remain to be recognized, and new information about many species remains unrecorded.
Citation
Smith 1865: Syn. Fil.: 16
Kaulf. 1824: En. Fil. Chamisso: 254
Copel. 1947: Gen. Fil.: 99
Diels 1899 – In: E. & P., Pfl. Fam. 1: 123
Christ 1897: Farnkr. Erde: 10, 317
Smith 1947: Gen. Fil.: 95
J. Smith 1842: pp. 659-668. – In: Lond. J. Bot.
Copel. 1909 – In: Philip. J. Sc.: Bot. 353
Hook. 1839: Gen. Fil.: t. 23
Smith 1844 – In: Sp. Fil.: 14
Smith 1875: Hist. Fil.: 244
Copel. 1947: Gen. Fil.: 98
Presl 1836: Tent. Pterid.: 54
Swartz 1806: Syn. Fil.: 139, 364