Metaxya
Distribution
Africa present, America present, Australasia, Continental Africa present, Japan present, Neotropical present, Pacific: Samoa (Samoa present), Pantropic present, SE. Asia present, South Africa present, eastern South America present, pantropic-subtropic present
Pantropic, extending considerably beyond the tropics in Japan, Australia, South Africa, and eastern South America; comparatively weakly represented in continental Africa. Six genera: Odontosoria (10 American spp., 2 African spp.), Ormoloma (2 spp., neotropical), Tapeinidium (17 spp., SE. Asia to Samoa), Sphenomeris (11 spp., pantropic-subtropic), Xyropteris (monotypic, Malesian), and Lindsaea (c. 150 spp., pantropic-subtropic).
The group is much more diversified in the Old than in the New World, but species here regarded as primitive occur in both hemispheres, and the origin and early history of the group cannot be traced; there are no fossils that can be positively attributed to one of its genera.
The group is much more diversified in the Old than in the New World, but species here regarded as primitive occur in both hemispheres, and the origin and early history of the group cannot be traced; there are no fossils that can be positively attributed to one of its genera.
Morphology
For an account of the morphology and anatomy of the group see Perez Arbelaez (), Wagner (), and Kramer (). To this may be added the following notes on the sporangium. The annulus has a bow of c. 8-23 thickened cells. In Lindsaea very many species have 10, 11, or 12 bow cells, 11 being a particularly common number. Some species have, however, consistently larger numbers, especially in the sections Isoloma, Osmolindsaea, and Tropidolindsaea, where numbers between 14 and 20 are common. Here the sporangial head is also somewhat larger. This holds for Tapeinidium, Xyropteris, and Sphenomeris, too, where the annulus has 15-23, usually between 16 and 18 indurated cells.
In most species the stomium is morphologically not or scarcely differentiated from the non-indurated part of the annulus. Two to four well-marked lip cells occur, however, in the sections Isoloma, Tropidolindsaea, and nearly all species of sect. Schizoloma. The character appears to fluetuate, with many transitional cases, in Tapeinidium and Sphenomeris.
The bow of the annulus reaches up to or slightly beyond the insertion of the stalk in Tapeinidium and most species of Lindsaea; in the latter genus the taxonomic value of the character is slight. In Sphenomeris it is also variable but apparently well-marked and constant for each species ().
In most species the stomium is morphologically not or scarcely differentiated from the non-indurated part of the annulus. Two to four well-marked lip cells occur, however, in the sections Isoloma, Tropidolindsaea, and nearly all species of sect. Schizoloma. The character appears to fluetuate, with many transitional cases, in Tapeinidium and Sphenomeris.
The bow of the annulus reaches up to or slightly beyond the insertion of the stalk in Tapeinidium and most species of Lindsaea; in the latter genus the taxonomic value of the character is slight. In Sphenomeris it is also variable but apparently well-marked and constant for each species ().
Taxonomy
In the older literature the Lindsaea-group is usually associated with, or included in, the Davallioid ferns, even as late as 1928 by Perez Arbeläez (l.c.), although he noted the great differences in scale and rhizome structure, spore morphology, etc., between the two groups. More recent authors have tended to emphasize the differences between them and have placed the Lindsaeoids in a separate family (), associated them with the very broadly defined Pteris-group (), or placed them in Dennstaedtiaceae. In the present Flora, the question of formal delimitation of families has been left open (); therefore no formal status is here proposed for the Lindsaea group of genera. I would, however, express the opinion that, though the group is a very natural one, its Separation as a distinct family does not seem warranted, in view of its many similarities to Dennstaedtia and allied genera.
As expounded in the revision of the American Lindsaeoids (Kramer, 1957, l.c.), and in the chapter on the Classification of the Malesian representatives (), the leaf pattern, greatly and excessively used in the past, is by itself insufficient as a basis for generic Classification. Such genera as have been based entirely on characters of leaf architecture and venation: Schizoloma, Iso- loma, Synaphlebium, are here merged with Lindsaea. Comments on the circumscription of Sphenomeris and Tapeinidium can be found in the above-cited papers ().
As expounded in the revision of the American Lindsaeoids (Kramer, 1957, l.c.), and in the chapter on the Classification of the Malesian representatives (), the leaf pattern, greatly and excessively used in the past, is by itself insufficient as a basis for generic Classification. Such genera as have been based entirely on characters of leaf architecture and venation: Schizoloma, Iso- loma, Synaphlebium, are here merged with Lindsaea. Comments on the circumscription of Sphenomeris and Tapeinidium can be found in the above-cited papers ().
Cytology
Relatively very few chromosome numbers of Lindsaeoid ferns are known at present. Moreover, several of them are only approximately known, and in some cases the identity of the plants is uncertain, as can be concluded from the names under which they have been reported.
In sect. Schizoloma there are counts of n = 88 for Lindsaea ensifolia subsp. ensifolia (Manton & Sledge) and for L. 'tenera' (prob. L. orbiculata var. commixta) (Manton), c. 88 for L. viridis (Brownlie), 44 or 45 for L. ensifolia subsp. coriacea (Manton ex Holttum; see Kramer, 1968), c. 42 for L. trichomanoides (as 'cuneata') (Brownlie), c. 40 for L. prolongata (Brownlie), and c. 47 for L. vieillardii (Brownlie) and L. chienii (Kurita). It seems likely that most, if not all, of these numbers are 44 or twice as many. This was also found in other sections; in sect. Odontoloma there is a count of n = c. 44 for L. repens var. sessilis (Walker, in litt.) and 44 or 45 for L. pulchella var. blanda (Walker, in litt.), in sect. Lindsaea n = 88 in L. portoricensis (neotropical) (Walker) and c. 84 in L. arcuata (neotropical) (Mickel, Wagner & Lim Chen). Other numbers may or may not be related, e.g. L. 'nitida' (= integral) (Manton in Kramer) and L. 'scandens var. terrestris' (= L. doryphora) (Manton in Kramer), both c. 47. The number n = 47 has, however, been found unequivocally in certain species, e.g. L. imacraeana'' (= L. repens var.) (Wagner), L. ' concinna' (Australia; L. brachypoda?) (Manton in Kramer) and L. parallelo- gramma (Manton in Kramer). The numbers c. 50 and c. 100 reported for L. decomposita (= L. obtusa?) (Manton in Holttum), from the same section as L. parallelogramma, may have been 47 and 94, re- spectively. L. 'pectinata' (prob. L. oblanceolata) was also found to have c. 50 chromosomes (Manton in Holttum), but this species is closely related to L. repens (c. 44, see above). In L. odorata, placed in sect. Osmolindsaea, divergent in its monolete spores but not otherwise very distinct, a polyploid series was found, ranging from n = 150 (Mehra & Khanna) or 150-152 (Walker, in litt.), to c. 220 (Walker, in litt.); the report of 82 (Manton in Kramer) from Ceylon may be due to misidentification of the plant. A basic number of 50 or 51 for this species seems possible. L. linearis, a member of the distinct sect. Paralindsaea, was found to have n = 34 (Brownlie).
The picture is equally confusing in the related but much smaller genus Sphenomeris. In S. chinensis, by far the most widespread species, there are reports of n = 94 (Mehra & Khanna; Kurita & Nishida), c. 100 (Bir; Manton & Sledge), and 145, 146, 147 (Manton & Sledge). Its close relative S. biflora was counted as n — 48 (Kurita & Nishida). S. retusa had n = 88 and c. 88 (Walker, in litt.); another specimen, apparently of hybrid origin, with abortive spores, had 162-164 chromosomes, with Univalents (Walker, in litt.). Two counts of n = 38 and 39, respectively, for the New World S. clavata (Walker; Wagner) are again divergent. It has been suggested that Sphenomeris is not a natural genus (), but morphological data do not seem to support this.
Two species of Odontosoria from Jamaica have been counted as c. 96 (Walker). No counts for Tapeinidium or Xyropteris have been found in the literature.
It seems that the numbers 44 and 47 are widespread in the group, and that some counts of approximately one of these numbers are equal to them or have been derived from them. It is also certain that one or more divergent basic numbers occur besides. Differences in basic number are, however, not necessarily connected with considerable morphological ones. It may be hoped that a clearer picture emerges when more data are available and that then some more light may be shed on the affinities in the group.
;
In sect. Schizoloma there are counts of n = 88 for Lindsaea ensifolia subsp. ensifolia (Manton & Sledge) and for L. 'tenera' (prob. L. orbiculata var. commixta) (Manton), c. 88 for L. viridis (Brownlie), 44 or 45 for L. ensifolia subsp. coriacea (Manton ex Holttum; see Kramer, 1968), c. 42 for L. trichomanoides (as 'cuneata') (Brownlie), c. 40 for L. prolongata (Brownlie), and c. 47 for L. vieillardii (Brownlie) and L. chienii (Kurita). It seems likely that most, if not all, of these numbers are 44 or twice as many. This was also found in other sections; in sect. Odontoloma there is a count of n = c. 44 for L. repens var. sessilis (Walker, in litt.) and 44 or 45 for L. pulchella var. blanda (Walker, in litt.), in sect. Lindsaea n = 88 in L. portoricensis (neotropical) (Walker) and c. 84 in L. arcuata (neotropical) (Mickel, Wagner & Lim Chen). Other numbers may or may not be related, e.g. L. 'nitida' (= integral) (Manton in Kramer) and L. 'scandens var. terrestris' (= L. doryphora) (Manton in Kramer), both c. 47. The number n = 47 has, however, been found unequivocally in certain species, e.g. L. imacraeana'' (= L. repens var.) (Wagner), L. ' concinna' (Australia; L. brachypoda?) (Manton in Kramer) and L. parallelo- gramma (Manton in Kramer). The numbers c. 50 and c. 100 reported for L. decomposita (= L. obtusa?) (Manton in Holttum), from the same section as L. parallelogramma, may have been 47 and 94, re- spectively. L. 'pectinata' (prob. L. oblanceolata) was also found to have c. 50 chromosomes (Manton in Holttum), but this species is closely related to L. repens (c. 44, see above). In L. odorata, placed in sect. Osmolindsaea, divergent in its monolete spores but not otherwise very distinct, a polyploid series was found, ranging from n = 150 (Mehra & Khanna) or 150-152 (Walker, in litt.), to c. 220 (Walker, in litt.); the report of 82 (Manton in Kramer) from Ceylon may be due to misidentification of the plant. A basic number of 50 or 51 for this species seems possible. L. linearis, a member of the distinct sect. Paralindsaea, was found to have n = 34 (Brownlie).
The picture is equally confusing in the related but much smaller genus Sphenomeris. In S. chinensis, by far the most widespread species, there are reports of n = 94 (Mehra & Khanna; Kurita & Nishida), c. 100 (Bir; Manton & Sledge), and 145, 146, 147 (Manton & Sledge). Its close relative S. biflora was counted as n — 48 (Kurita & Nishida). S. retusa had n = 88 and c. 88 (Walker, in litt.); another specimen, apparently of hybrid origin, with abortive spores, had 162-164 chromosomes, with Univalents (Walker, in litt.). Two counts of n = 38 and 39, respectively, for the New World S. clavata (Walker; Wagner) are again divergent. It has been suggested that Sphenomeris is not a natural genus (), but morphological data do not seem to support this.
Two species of Odontosoria from Jamaica have been counted as c. 96 (Walker). No counts for Tapeinidium or Xyropteris have been found in the literature.
It seems that the numbers 44 and 47 are widespread in the group, and that some counts of approximately one of these numbers are equal to them or have been derived from them. It is also certain that one or more divergent basic numbers occur besides. Differences in basic number are, however, not necessarily connected with considerable morphological ones. It may be hoped that a clearer picture emerges when more data are available and that then some more light may be shed on the affinities in the group.
;