Peristome Structure in the Mitteniales (ord. nov.: Musci), a Neglected Novelty Author(s): Jonathan Shaw Source: Systematic Botany, Vol. 10, No. 2 (Apr. - Jun., 1985), pp. 224-233 Published by: American Society of Taxonomists Stable URL: http://www.jstor.org/stable/2418348 Accessed: 17/11/2010 11:30

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=aspt.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

American Society of Plant Taxonomists is collaborating with JSTOR to digitize, preserve and extend access to Systematic Botany.

http://www.jstor.org SystematicBotany (1985), 10(2): pp. 224-233 C Copyright1985 by the American Society of Plant Taxonomists

Peristome Structurein the Mitteniales (ord. nov.: Musci), a Neglected Novelty

JONATHAN SHAW Department of Botany, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560

ABSTRACT. Cellular patternson the peristomeof Mitteniaplumula (Mitt.) Lindb., as revealed by scanning electron microscopy,support the interpretationmade by Ilma Stone in 1961 that the peristomeof this species is neither haplolepideous nor diplolepideous. The outer peristometeeth of Mitteniaare homologous to the endostome of diplolepideous ,and to the single peristome of haplolepideous mosses. There is no other peristomialstructure known in mosses that is homol- ogous to the inner peristome of Mittenia.The Mitteniaceae are removed fromthe Bryales,and a new order,the Mitteniales,is recognized. The Sorapillaceae and Eustichiaceae,traditionally placed near the Mitteniaceae, have peristomesthat are probably derived fromthe haplolepideous type.

The Mitteniaceae is a monotypicfamily rep- ily in the Bryales (Eubryales), suborder Rhizo- resented by Mittenia plumula (Mitt.) Lindb. goniineae. Almost all modern authors (Broth- of Mitteniaare verysmall (2-4 mm high), erus 1924; Robinson 1971; Crosby 1980; Vitt acrocarpous, and have somewhat complanate 1982) have included Mitteniaceaein the Bryales. foliation.The oblong-linguateto obovate leaves Ilma Stone (1961b) presentedobservations on are obliquely inserted, have well-developed the development of sporophytes and gameto- decurrencies, a single, rather weakly devel- phytes of Mitteniaplumula in exemplarydetail. oped costa, and isodiametric cells. Terminal Perhaps most significantlyfrom a taxonomic- sporophytesare borne on erectsetae (frequent- evolutionarystandpoint, Stone reporteda pat- ly several per inflorescence),and the capsules tern of peristomialdevelopment and structure, are erect,shortly cylindrical, and have rostrate which, although clearly arthrodontous,differs opercula. The peristome is double. Mitteniais fundamentallyfrom both the haplolepideous known fromAustralia, New Zealand, and Tas- and diplolepideous peristometypes (Evans and mania, where it grows on moist to rather dry Hooker 1913; Blomquist and Robertson 1941; soil, in the mouths of caves, on upturned roots, Mueller 1973). Haplolepideous and diplolepi- and on soil mounds in foresthabitats (Scott and deous peristomes develop fromthe innermost Stone 1976; Sainsbury 1955). It is most common threeconcentric layers of amphithecialcells (the at low elevations (Martin 1958). Mitteniahas outer, primary,and inner peristomial layers, frequently been compared to Schistostega respectively,from the outside inward; Blom- (Schistostegaceae)because the two genera share quist and Robertson 1941). Haplolepideous obliquely oriented, decurrent leaves and lu- peristome teeth and endostomes of diplolepi- minescent protonema (Goebel 1906; Stone deous peristomes are formed fromwall mate- 1961a). rial between cells in the primary and inner The generic name Mitteniawas published by peristomial layers, whereas diplolepideous Lindberg (1863) to replace MniopsisMitt. (1859), exostome teeth are formed fromwall material which is a homonym.of MniopsisMart. (1822, between cells in the outer and primaryperisto- Podostemaceae) and Mniopsis Dumort. (1822, mial layers. In Mittenia,Stone (1961b) reported Calobryaceae). When Mitten founded the ge- that the outer peristome teeth develop from nus, he included it in the Bryaceae, and this cells between the primaryand inner peristo- placement was supported by Jaeger (1875). mial layers, and the inner peristome develops Fleischer (1902) recognized the Mitteniaceae frominner periclinal walls of cells in the inner (bixtdid not provide a description) and allied peristomiallayer and adjacent walls of cells in the family with the Bryaceae in the Bryales. the outermostlayer of the endothecium. This Brotherus(1924), who was apparentlythe first is apparently the only legitimate report of to validate the Mitteniaceae, classifiedthe fam- endothecial cells involved in peristome for- 224 1985] SHAW: MITTENIALES 225

mation in mosses, since early suggestions of tooth also consists of a single column of cell endothecial involvement were incorrectinter- wall plates with horizontal lines separatingthe pretations(Blomquist and Robertson 1941). plates (figs. 4-6). At the extreme base of the It is surprisingthat the systematicand evo- teeth,the ventral lamellae are not stronglyde- lutionary implications of Stone's (1961b) care- veloped (fig. 4), but somewhat higher up they ful observations seem not to have been evalu- are conspicuous (fig. 5). In addition to the la- ated previously.After all, in other species mellae, there are conspicuous thickeningsbe- with a double peristomeconsisting of outer ex- tween the lamellae (fig. 5). These thickenings ostome teeth and a delicate endostome of seg- are less developed near the apices and bases of ments arising from a basal membrane, the the teeth than at their middles (figs.4, 6). The peristome has a clearly demonstrable diplole- ventralplates tend to be somewhatbroader than pideous structure. The patterns of lines on the dorsal plates, as is the case in many exo- peristome teeth, reflectingthe pattern of cell stome teeth,which develop fromdifferent cell divisions in the peristomiallayers that gave rise layers. The ventral plates are smooth on the to the teeth, allow inferences about develop- surfacelike the dorsal plates. mental differencesbetween peristome types The inner peristome of Mitteniaconsists of even when direct developmental data are not (24-)28-32(-36) delicate, filiform,pale yellow- available (Edwards 1979). If Stone's (1961b) ob- ish teeththat arise froma low basal membrane. servationsare correctand the peristomeof Mit- The latterdoes not reach above the capsule rim, tenia is developmentally neither diplolepi- and can only be observed by examining the deous nor haplolepideous, the cellular patterns capsule mouth from the inside. The distance of the dorsal and ventral peristomial surfaces between adjacent inner peristomialteeth is not should reflectthese differences.Stone, in fact, constant in the circumferenceof the capsule, observed unusual cellular patterns on the in- some pairs lying closer together than others ner and outer peristomial surfaces in Mittenia. (see Stone 1961b, plate 2, fig.2). Adjacent teeth She did not, however, discuss the systematic often anastomose in an irregular manner (fig. implicationsof her observations. 8), and the inner peristomialteeth do not clear- As the cellular patterns characterizing the ly alternatewith the outer teeth. Stone (1961b) peristome of Mitteniaare difficultto see using reported that, in general, there are segments light microscopy,the peristomewas examined both opposing each outer toothand alternating with a scanning electron microscope. Because with them.Each inner tooth consistsof a single of the systematicsignificance of the peristomial column of cell wall plates on the dorsal surface cell-patterns,and because Stone (1961b) em- (fig.7). There are horizontal lines representing phasized developmental characteristicsof the anticlinal walls of inner peristomiallayer cells, Mitteniaperistome ratherthan the patternson but there are no vertical lines on the dorsal mature teeth, observations made possible by surfaces of the teeth. On the ventral surface, scanning electron microscopy will be de- each has a rather prominent vertical line, as scribed here in some detail. well as the horizontal lines thatseparate plates The sixteen outer peristometeeth of Mittenia in the columns (fig. 10). Occasionally a tooth are extremelylong and slender, up to 800 ,Am lacks a vertical line on the ventral surface,but in length. They are deep red, and when moist that is usually true only of a tooth that lies or dry, the teeth tend to coil backward upon especially close to another tooth (fig. 8, arrow themselves. (Processes of both the inner and at far right). The inner peristome teeth are outer peristome are referredto here as teeth; smooth throughouttheir lengths on both the problems of terminologywill be discussed be- dorsal and ventralsurfaces, including the basal low.) On the dorsal surface,each tooth consists membrane.The cellular patternon the ventral of a single column of cell wall plates, each plate surface of the basal membrane is rather vari- separated by a horizontal line representing able. There are prominenthorizontal lines, but remnantsof anticlinal cell walls in the primary there may or may not be one or more vertical peristomiallayer (figs.1-3). There is no median lines between teeth (figs. 8-9). The outer sur- vertical line as in diplolepideous exostome face of the basal membrane was not observed teeth. The plates are smooth throughout the as it does not extend above the capsule mouth length of the teeth.On the ventralsurface, each and is fused to the mouth. 226 SYSTEMATIC BOTANY [Volume 10

'I j,' s f w.' ", w

FIGS. 1-6. SEM photographsof the outerperistome of Mittenia plumula (Archer s.n., holotype: NY). 1. Dorsal surface near the base; capsule rim visible near the bottom. 2. Dorsal surface about midway up a tooth. 3. Dorsal surface near the top. 4. Ventral surface near the base. 5. Ventral surface about midway up the tooth. 6. Ventral surface near the top. Scale on 1-2, 4-6 = 10 Asm;scale on 3 = 25 jAm.

Cellular patternscharacterizing haplolepi- on the inner and outer peristome of Mittenia deous peristometeeth were describedby Ed- do not conformto eitherthe haplolepideous or wards (1979). Shaw and Rohrer (1984) de- diplolepideous type. Stone's (1961b) develop- scribed patterns on the endostomes of mental study demonstratedthat the outer peri- diplolepideousperistomes. The patternsof lines stome in Mitteniais developmentally equiva- 1985] SHAW: MITTENIALES 227

FIGS.7-10. SEM photographs of the inner peristome of Mitteniaplumula (Archer s.n., holotype: NY). 7. Dorsal surface. 8. Ventral surface of the basal membrane and three teeth; the tooth on the far right lacks the vertical line found on others. 9. Ventral surface of the basal membrane. 10. Ventral surface of one tooth.Scale = 10 Am.

lent to the endostomeof diplolepideous mosses, cal distinctiveness,the peristomeof Mitteniais and to the single row of teeth in haplolepi- at least as differentfrom both the haplolepi- deous mosses. The inner peristomeof Mittenia, deous and diplolepideous typesas the latterare since it is formedfrom inner peristomiallayer fromeach other.There are generally only eight cells and those of the outermost endothecial cells in the inner peristomiallayer of Mittenia, layer,has no homologue among other mosses. whereas there are 24 cells in this layer in hap- In termsof developmental and morphologi- lolepideous peristomes,and anywhere from16 228 SYSTEMATIC BOTANY [Volume 10 to 112 cells in diplolepideous peristomes (Ed- diplolepideous peristomes and alternate en- wards 1979; Shaw and Rohrer 1984). The peri- dostome segments. Sixteen families were in- stomial formula (Edwards 1979) of Mitteniais cluded in the order. Two families, Eustichi- usually 4:2:1, and seems to be unique among aceae and Sorapillaceae, classifiedin proximity arthrodontousmosses. (Derivation of peristo- to the Mitteniaceae contain species whose peri- mial formulae is described in the legend for stome structurehas been the subject of varying figs.21-26.) Cell divisions in the outermosten- interpretations.The Calomniaceae, also classi- dothecial layer occur independentlyof those in fiedclose to the Mitteniaceae,are eperistomate. the amphitheciallayers, and the numberof cells The Eustichiaceae contain only the genus subtending the eight inner peristomial layer Eustichia(Brid.) Brid. and about eight species. cells is variable, leading to the variable number The peristomeof Eustichiais single, and has tra- of inner teeth(Stone 1961b). Both amphithecial ditionallybeen interpretedas endostomial. Ed- and endothecial cells contributewall material wards (1979) recently suggested that it is ac- to the inner peristome of Mittenia.Wall depo- tuallyhaplolepideous, and presenteda drawing sition coincides with points where anticlinal that clearly showed a 2:3 cellular pattern.Vitt walls of endothecial cells meet periclinal walls (1982) subsequently transferredthe family to of inner peristomiallayer cells, resultingin the the . On the dorsal surface,the teeth verticallines thatare visible on the ventralsur- are verticallystriate, and with no verticallines face of the inner peristome.According to Stone they appear to be haplolepideous (fig. 11). Ad- (1961b), wall deposition by endothecial cells is ditional thickeningsare present externalto the less than depositionfrom the amphithecialcells, teeth (fig. 11). On the ventral surface,the cel- and may be absent in places, especially near the lular pattern characteristicof haplolepideous tips of the teeth. peristomesis lacking on most teeth of most of With its highly unusual peristome,the Mit- the (ten) capsules examined (figs. 12-14). Each teniaceae are obviously misplaced in the tooth consists of a single column of cell wall Bryales. The family does not, in fact, fit into plates, ratherthan one and one half plates per any of the orders presentlyrecognized for the tooth as in haplolepideous peristomes (Ed- mosses, and a new order, the Mitteniales, is wards 1979). At the base of the peristomewhere warranted. the teeth are fused, the presence of only one line between each tooth indicates thatthe peri- Mitteniales Shaw, ord. nov.-TYPE GENUS: Mit- stomial formula is 4:2:2 ratherthan 4:2:3. The tenia Lindb. Oefv. K. Vet. Ak. Foerh. 19: 4:2:2 formula seems to be the most common 606. 1863. condition in Eustichialongirostris (figs. 12-14). In Plantae parvae acrocarpae in caulibus foliosis only a few places on two of the capsules stud- complanatae, cellulis foliorum isodiametricis. ied was the haplolepideous 2:3 patternevident Capitula erecta; peristomiumduplex, dentibus (figs. 15-16). exterioribusin parietibus inter series cellula- The Sorapillaceae, a small family of two rum peristomialem primariam et interioribe species in the genus SorapillaSpruce & Mitt., procreantibus,dentibus interioribusin parie- have had an unsettledtaxonomic history, being tibus inter series cellularum peristomialemin- shifted between haplolepideous groups (Mit- teriorem et endothecialem exteriorem pro- ten 1869; Brotherus 1909) and diplolepideous creantibus. groups (Brotherus1924; Dixon 1932). The peri- Small acrocarpous plants with ? complanate stome was interpretedby Brotherus(1924) as foliationand isodiametricupper leaf cells. Cap- single and endostomial.Allen (1981), however, sules erect; peristome double, the outer teeth recently presented observations using scan- derived from periclinal walls separating pri- ning electron microscopy,and concluded that mary and inner peristomiallayer cells, the in- the peristome is diplolepideous, consisting of ner teeth derived fromwalls separating inner long endostome segments and short or rudi- peristomial layer cells and cells of the outer- mentaryexostome teeth.On the dorsal surface, most endothecial layer. the long inner teeth consist of a single column The concept of the Bryales presented by of plates (fig. 17). This patternis similar to pat- Brotherusin the second edition of Die natiir- terns on the outer teeth of Mitteniaand Eusti- lichen Pflanzenfamilien(1924) included those chia, on haplolepideous teeth, and on the en- families containing acrocarpous mosses with dostome segments of the Funariaceae. On the 1985] SHAW: MITTENIALES 229

jA

X s- { | X * A - I I

IL

I ~ ~ ~ I W.411

i s_.~;.-~ A .

FIGS.11-16. SEM photographs of the peristome of Eustichia longirostris.11. Dorsal surface; note the ad- ditional thickeningsin frontof the long teeth (Dusen 284, NY). 12. Ventral surface (Dusen 284, NY). 13. Ventral surface (Dusen s.n. 1896, NY). 14. Ventral surface (Hatcher& Engel 145, NY). 15. Ventral surface (Dusen s.n. 1896, NY). 16. Ventral surface (Hatcher& Engel 145, NY). Scale = 25 Am. ventral surface,there appears to be only one The last familytraditionally placed in prox- verticalline between each tooth,as in Eustichia imityto the Mitteniaceae is the Calomniaceae. (figs.18-20). On one capsule, however, there is The familyconsists of a single genus, Calomnion some evidence of a reduced 2:3 pattern(fig. 20). Hook.f & Wils., and three eperistomatespecies. 230 SYSTEMATICBOTANY [Volume 10

r~~~~~~~~~~~o tj

'~~~~~~' A.A~~~~~~~~~~~~~~AS 4.

.~~~~~~~~M .;-C f

FiGs.17-20. SEM photographsof theperistome of Sorapilla sprucei (Spruce 559, holotype: NY). 17. Dorsal surface. 18. Ventralsurface. 19. Ventralsurface. 20. Ventralsurface. Scale = 25 ,um.

J.D. Hooker and W. Wilson(1854) remarked groupof familieshaving complanate foliation. aboutsimilarities between Calomnion and Tetra- Brotherus(1924), following Fleischer's system, phis,and otherbryologists who have related placed the Calomniaceaein the Bryales,close Calomnionto theTetraphidaceae include Dixon to the Mitteniaceaeand the Rhizogoniaceae. (1932)and Vitt(1982). Jaeger (1875) recognized Mostrecent authors have followedBrotherus's the Calomniaceaefor Calomnion, and placed it placementof the Calomniaceaein the Bryales withthe Schistostegaceaeand the Epipterygi- (Robinson1971; Schultze-Motel1971; Crosby aceae (thelatter now partof the Bryaceae)in a 1980). 1985] SHAW: MITTENIALES 231

The ovate to elliptical leaves and isodiamet- habit and the presence of pseudoparaphyllia ric leaf cells of Calomnionare indeed much like would appear to be inconsistentwith a place- those of Tetraphis,and in habit there is a strik- ment of this genus among the haplolepideous ing similaritybetween the genera. The cap- mosses (Allen 1981). Whether or not Sorapilla sules of both Calomnionand Tetraphisare erect and (or) Eustichiarepresent an intermediatelevel and cylindrical,and have rostrateopercula. Vitt in a particularevolutionary line relating hap- (1982) suggested thatCalomnion may have thal- lolepideous mosses and Mitteniais difficultto lose protonema like that of Tetraphis,but he determine.There is a need for developmental presented no detailed observations and stated studies of peristomesto ascertainwhether those that the matterneeds to be investigated fur- of Mittenia,Sorapilla, and Eustichiahave similar ther. If Calomniondoes have thallose protone- peristomial formulae by virtue of convergent ma, this would provide strongevidence, in my developmental pathways leading to similar opinion, for a close phylogenetic relationship mature structures,or whether the similarities between Calomniaceae and Tetraphidaceae,in- reflectuniform developmental patterns. stead of the formerwith Mitteniaceae. In the The peristome of Mitteniaposes some prob- absence of such evidence, J. D. Hooker (1867) lems of terminology.Should the double peri- accurately observed that Calomnionis "A very stome of Mitteniabe referredto as exostome curious moss of doubtfulaffinity." and endostome, even though the teeth are not Because of a lack of developmental infor- comparable in position to the exostomeand en- mation about peristomes in the Sorapillaceae dostome of diplolepideous mosses? Although and Eustichiaceae, these families cannot be as- it is well known that haplolepideous peri- signed definitelyto the Mitteniales.The essen- stomes are developmentally equivalent to dip- tial featuresof the order are the few number lolepideous endostomes,the termendostome is of cells in the inner peristomial layer and the rarely applied to them. The term pre- or pro- contributionof endothecial cells to peristome peristome also is applied rather loosely in formation.Neither Sorapillanor Eustichiahas an bryology. Uses that have appeared in the lit- inner peristomecomparable to that of Mittenia, erature include thickeningsexternal to haplo- and there is no reason to suspect that endothe- lepideous peristome teeth (Edwards 1979), cial cells contributewall material to the peri- thickenings external to diplolepideous exo- stome. The peristomial formula of Mittenia(4: stomes (Brotherus 1924-1925), and exostomial 2:1) is approached by Sorapillaand Eustichia(both thickenings external to diplolepideous endo- most commonly 4:2:2) more closely than by stome segments(Horton 1982). Should the out- either the haplolepideous or diplolepideous er teeth of Sorapilla(fig. 17) be called preperi- types (4:2:3-4:2:12). The relationships of moss stomes or exostome teeth?It seems most logical species having 4:2:2 formulae are difficultto to apply the term "exostome" to any thicken- interpretsince this pattern could be derived ings thatdevelop fromouter and primaryperi- from either diplolepideous or haplolepideous stomial layer cells, and "endostome" to any patterns(figs. 22, 26). In Pseudoditrichummirabile thickeningsthat develop from cells one layer Steere & Iwats. (Pseudoditrichaceae) the 4:2:2 inward. Preperistomewould then be applica- patternis probablyderived froma more typical ble only to peristomialthickenings external to diplolepideous type (Shaw 1984; figs. 25-26). the outer peristomial layer. The outer peri- However, in such taxa as Venturiellasinensis stome of Mitteniawould then be an endostome. (Vent.) C. Muell. (Erpodiaceae), the homology The inner peristomeis unique, although it does of the peristomeis unclear, as is the systematic not seem necessary to provide a new term for placement of the species possessing it (Edwards this structure.It is remarkable that the inner 1979). The presence of a 2:3 patternon at least peristome of Mitteniaconverges in form and some of the peristome teeth of Eustichiaindi- delicate appearance to such a degree with dip- cates that this species may be best interpreted lolepideous endostomes, and that the same is as a haplolepideous type. Vertical striationson true of the outer teeth and diplolepideous ex- the dorsal surfaceof the Eustichiaperistome are ostomes. The parallel evolution of such similar consistentwith Vitt's (1982) placement of this inner peristomesin diplolepideous mosses and genus in the Dicranales. The same also may be in Mitteniamay indicate a functional signifi- true of Sorapilla; however, the pleurocarpous cance to these structures.On the other hand, it 232 SYSTEMATIC BOTANY [Volume 10

21 22

Funaria Sorapilla Eustichia

23 24

IPL Mittenia haplolepideous

25 26

:~~~~~~~~~~iI I e idT diplolepideousp FIGS. 21-26. Schematic transversesections of one-eighth of the circumferenceof moss capsules showing the cell layers involved in peristomeformation. The number of cells in each layer (fromtop to bottom)gives the peristomial formula;for example, the formula for fig. 22 is 4:2:2. The cellular patternon the surface of each peristometype can be inferredfrom the arrangementof anticlinal (vertical) and periclinal (horizontal) walls in the figures.Note the differencein cellular patternbetween peristomeswith 4:2:2 formulaederived fromhaplolepideous types (fig. 22), and diplolepideous types (fig. 26). OPL, PPL, and IPL referto the outer, primary,and inner peristomial layers, respectively.These figuresare based on observations presented by Edwards (1979), Shaw (1984), and Shaw and Rohrer (1984).

is interestingthat no other mosses with single these two groups of arthrodontousmosses is peristomes evolved a structurecomparable tc ancestralto the Mitteniales. If Sorapillaand (or) that found in Mittenia. Eustichiarepresent phylogenetic intermediates Although the phylogenetic position of the between Mittenia and other arthrodontous Mitteniales is uncertain,the order is pheneti- mosses, a relationship between the Mitteni- cally isolated fromother orders of arthrodon- aceae and haplolepideous groups is indicated. tous mosses. It is likely that Mitteniaevolved However, similarities in peristome structure from an ancestor having a haplolepideous or between Mittenia,Sorapilla, and Eustichiamay be diplolepideous peristomestructure but there is due to convergentevolution. Similarly,the ga- little evidence at present to evaluate which of metophyticsimilarities between Schistostegaand 1985] SHAW: MITTENIALES 233

Mitteniamay reflectconvergent evolution in GOEBEL,K. 1906. ArchegoniatenstudienX. Flora 96: similar habitats, or phylogenetic proximity. 1-202. There is a need for additional developmental HOOKER,J. D. 1864. Handbookof theNew Zealand studies of moss capsules and peristomesin or- flora:A systematicdescription of thenative plants of der to answer these and other questions con- New Zealand and the Chatham,Kermadec's, Lord Aukland's,Campbell's, and Macquarrie'sislands, part cerning phylogenetic relationships in mosses. 2. London: Reeve and Co. Although Schistostegais eperistomate,studies of and W. WILSON. 1854. The botanyof theAnt- capsule development with particularreference arcticVoyage of H.M. DiscoveryShips Erebusand to the peristomiallayers would provide essen- Terrorin the years 1839-1843, underthe command tial informationfor evaluating its relationship of Sir JamesClarke Ross, Kt., R.N., F.R.S. II. Florae to Mitteniaand other mosses. Novae Zealandeae,part 2. London: Reeve and Co. HORTON,D. G. 1982. A revision of the Encalypta- ACKNOWLEDGMENTS. Stimulatingconversations ceae (Musci), with particular reference to the with Harold Robinson, Bruce Allen, Joseph Rohrer, North American taxa. Part I. J. Hattori Bot. Lab. and Howard Crum contributedsignificantly to many 53:365-418. of the concepts presented in this paper (for which I JAEGER, A. 1875. Genera et species muscorum sys- alone, however, take full responsibility).Present ad- tematicedisposita seu adumbratio florae musco- dress: Departmentof Botany,Duke University,Dur- rum totis orbis terrarum.Ber. St. Gall. Naturw. ham, North Carolina 27706. Ges. 1873-74:53-278. LINDBERG, S. 0. 1863. Epipterygium,nytt moss-slagte. Oefv. K. Vet.-Akad. Foerh. 19:599-609. LITERATURE CITED MARTIN,W. 1958. Survey of moss distributionin ALLEN,B. 1981. A reevaluation of the Sorapillaceae. New Zealand. Bryologist61:105-115. Bryologist84:335-338. MITTEN, W. 1869. Musci Austo-americani.J. Linn. BLOMQUIST, H. L. and L. L. ROBERTSON. 1941. The Soc., Bot. 12:1-659. development of the peristome in Aulicomnium MUELLER, D. M. J. 1973. The peristome of Fissidens heterostichum.Bull. TorreyBot. Club 68:569-584. limbatusSullivant. Univ. Cal. Publ. Bot. 63:1-34. BROTHERUS,V. F. 1901-1909. Musci. Pp. 207-235 and ROBINSON, H. 1971. A revised classificationfor the 277-1246 in Die naturlichenPflanzenfamilien, I. Teil, ordersand familiesof mosses. Phytologia21:289- 3. Abteilung, Lieferung, eds. A. Engler and K. 293. Prantl. Leipzig: Wilhelm Engelmann. SAINSBURY, G. 0. K. 1955. A handbook of the New . 1924-1925. Musci. Pp. 1-478 in Die natiir- Zealand mosses. Bull. Roy. Soc. New Zealand 5: lichenPflanzenfamilien, ed. 2, vol. 10-11, eds. A. 1-490. Engler and K. Prantl. Leipzig: Wilhelm Engel- SCHULTZE-MOTEL, W. 1971. Vorlaufiges verzeichnis mann. der Laubmoose von Samoa. Bryologist 74:357- CROSBY,M. R. 1980. The diversityand relationships 358. of mosses. Pp. 115-129 in The mossesof North SCOTT, G. A. M. and I. STONE. 1976. The mossesof America,eds. R. J.Taylor and A. E. Leviton. San southernAustralia. London, New York, Toronto, Francisco: Pacific Division AAAS. Sydney, and San Francisco: Academic Press. DIXON,H. N. 1932. Classificationof mosses. Pp. 397- SHAW, J. 1984. A reinterpretationof peristome 412 in Manual of bryology,ed. F. Verdoorn. The structure in Pseudoditrichummirabile Steere & Hague: Martinus Nijhoff. Iwats. (Pseudoditrichaceae). Bryologist 87 (in EDWARDS,S. R. 1979. Taxonomic implicationsof cell press). wall patterns in haplolepideous moss peri- and J. R. ROHRER. 1984. Endostomial archi- stomes. Pp. 317-346 in Bryophytesystematics, eds. tecturein diplolepideous mosses. J. Hattori Bot. G. C. S. Clarke and J.G. Duckette. London, New Lab. 57:41-61. York, Toronto, Sydney, and San Francisco: Aca- STONE, I. 1961a. The highly refractiveprotonema of demic Press. Mittenia plumula (Mitt.) Lindb. (Mitteniaceae). EVANS,A. W. and H. D. HOOKER. 1913. Develop- Proc. R. Soc. Vict. 74:119-124. mentof the peristomein Ceratodonpurpureus. Bull. 1961b. The gametophoreand sporophyteof TorreyBot. Club 40:97-109. Mitteniaplumula (Mitt.) Lindb. Austral. J. Bot. 9: FLEISCHER, M. 1902. Die Musci der Flora von Buiten- 124-151. zorgzugleich Laubmoosflora von Java mit Berucksich- VITT, D. H. 1982. .Pp. 307-336 in Synopsis tigungaller Familienund Gattungender gesamten and classificationof livingorganisms, vol. 1, ed. S. Laubmooswelt,vol. 1. Leiden: E. J. Brill. P. Parker. New York: McGraw-Hill.