Key points are not available for this paper at this time.
The percentage of pantropical and amphioceanic of ferns far exceeds similar percentages for angiosperms. Conversely, the number of fern is much lower than the comparable figure for flowering plants. Comparison of the two groups of plants on islands points out even more strikingly the lower percentages of generic and specific endemism of ferns. These distributional differences cannot be accounted for by the longer geologic history of ferns, but can be attributed to their greater dispersibility. The wider distribution of ferns, together with the absence of many of the speciation barriers operative in angiosperms, has contributed to a relatively slower rate of speciation in the ferns, and probably also to a lower extinction rate. Delimitation of generic and suprageneric taxa is therefore more difficult in many cases. T1HE DISTRIBUTION OF extant ferns has generally been neglected as a source of biogeographical data. Christ (1910) was the first to deal comprehensively with fern distribution. Unfortunately, Christ's work is outdated in some respects, primarily because circumscription of has been greatly altered in the past several decades. In his introduction, Christ acknowledged that many ferns do indeed have wider distributions than flowering plants do, but that in general the distribution of pteridophytes parallels that of flowering plants. Subsequent authors dealing with general plant geography have dismissed the ferns with statements such as phytogeographical conclusions concerning flowering plants can be extended to cryptogams (Cain 1944, p. 284). A comparison of the distribution of the of flowering plants and ferns is presented in table 1. Figures for the flowering plants are extracted from Good ( 1964), while the data for the ferns are largely the result of my own compilations. In general, I have adopted a relatively restricted delimitation of many fern genera, largely following Copeland's Filicum (1947); broader generic circumscriptions would point out more emphatically the differences in distribution of ferns and flowering plants. Copeland recognized a total of 309 fern genera. The slightly enlarged list used to compile the figures herein (351 genera) reflects, for the most part, a better understanding of certain neglected and taxonomically difficult groups. Where recent studies have shown Copeland's generic concepts to be too narrow, I have relied on the judgments of other pteridologists (e.g., Trismeria I include in Pityrogramma, Hemipteris in Pteris, Schizoloma and Isoloma in Lindsaea, and Egeinolfia in Bolbitis). For purposes of the comparisons below, I accept 19 fern described since the publication of Filicum. Copeland's classification within two groups of ferns, the Cyatheaceae and the thelypteroid ferns, is now so obsolete as to require treatment according to more recent monographic work: I have followed Tryon (1970b) in his circumscription of the Cyatheaceae and accepted most of the thelypteroid of Ching (1963). In compiling the figures for the Hymenophyllaceae, I have accepted most of Copeland's genera, but with the circumscription given them by Morton (1968). Not unexpectedly, difficulties arise in trying to place certain in one of the categories of table 1. For example, Adiantum is predominantly a genus of the tropics and subtropics, and I have placed it accordingly; nevertheless, there are a very few species native to temperate regions. On the other hand, Polystichum contains a larger proportion of species adapted to the exigencies of temperate climates, and I have placed it in the cosmopolitan-subcosmopolitan class. Following Good (1964, p. 132), I refer to as genera actually confined to one floristic region or having a distribution not greatly exceeding the average size of a floristic region. This definition, too, leaves one with problems, but they are largely unavoidable. CHOROLOGICAL CONSIDERATIONS Three values from table 1 encourage me to believe that fern systematists have circumscribed fern on the basis of criteria that are roughly similar in importance to the criteria used to define angiosperm 4 BIOTROPICA 4N1): 4-9 1972 This content downloaded from 207.46.13.85 on Tue, 23 Aug 2016 05:48:21 UTC All use subject to http://about.jstor.org/terms TABLE 1. Compairison of distribution of generia in flowering plants and ferns. Flowering plants Ferns Total ca. 12,500 ca. 351 Total species ca. 225,000 ca. 9,000 Average genus size ca. 18 species/genus ca. 26 species/genus Genera with 100 or more species each ca. 470 3.1% 21 6.0%I Monotypic ca. 4,200 [33% 104 30% Cosmopolitan or subcosmopolitan ca. 130 ca. 1% 9 2.6%) Circumboreal genera' 11 [3.1% Pantropical ca. 250+ 2% 60 17.1% Pantropical + cosmopolitan and subcosmopolitan ca. 380 3% ca. 69 19.7%) Neotropical-Asian and Pacific Basin ca. 40 [0.32% 11 3.1% Neotropical-African ca. 92 0.74% 9 2.6% African (Madagascar)-Indo-Australasian-wide ca. 445 3.6% 28 8.0% Asian-Pacific Basin widespread ca. 374 3.0%) 64 [18.2% Neotropical widespread ca. 350 2.8% 38 10.8% Narrow endemic genera-Asia, Pacific 73 20.8% Narrow endemic genera-Neotropics 42 11.9% Narrow endemic genera-Africa and adjacent islands 6 1.7%) Total ca. 10,000 [80% 121 34.5% a Extending into high elevation tropics in a few instances. genera. On a percentage basis, average genus size, the number of with 100 or more species, and the number of monotypic are comparable in the two groups of plants. Meaningful differences emerge when the percentages of pantropical in the flowering plants (2 percent) and in the ferns (17 percent) are compared. About 69 (20 percent) of ferns are either cosmopolitan-subcosmopolitan or pantropical. These 69 contain nearly two-thirds of the total number of fern species, a figure likely to be far in excess of the percentage for flowering plants, unknown at the present. Similar differences in distribution emerge when one looks at the percentages of tropical-subtropical in flowering plants and ferns which occur in the Neotropics and Paleotropics (but are not pantropical). Widely distributed Neotropical in the ferns constitute nearly 11 percent of the total number of fern genera, while less than 3 percent of the flowering plant have such a distribution. Figures for Asian-Pacific Basin widespread reflect similar distribution frequencies. A corollary of these features of fern-flowering plant distribution is that the proportion of endemic in the flowering plants (ca. 80 percent) is much higher than in the ferns (35 percent). The greatest number of these fern endemic is known from the floristically rich Indo-Australasian region, and many are monotypes (45 of 73). Twentyseven of 39 endemic in the Neotropics are monotypic, while two of six African are monotypic. A rather high proportion of these monotypic could, without stretching generic limits too severely, be accommodated within other genera, or their origin from a larger and more widespread genus is evident. Relatively few monotypes are obviously relict: Loxsoma (New Zealand) and Thyrsopteris (Juan Fernandez) are two of the notable
Alan Р. Smith (Sat,) studied this question.