mt. J. Sci. 158(1):51—56. 1997. © 1997 by The University of Chicago. All rights reserved. 1058-5893/97/5801 -0005$02.00

ORIGINS AND NATURE OF VESSELS IN . I. ACORUS

SHERWIN CARLQUIST’ AND EDWARD L. SCHNEIDER Santa Barbara Botanic Garden, 1212 Mission Canyon Road, Santa Barbara, California 93105

Vessels are demonstrated in metaxylem of both roots and rhizomes of both species of Acorus (Acoraceae) by scanning electron microscopy (SEM). The end walls of vessel elements are characterized by perforations that retain porose pit membranes; the pores sometimes coalesce into larger openings. Lateral walls lack pores in primary walls as detectable by our SEM. Sthations in pit membranes, as in Nymphaeaceae, were observed. Although vessel elements, as defined by porose pit membranes in end walls, occur in both roots and rhizomes, the long scalariform nature of end walls of vessel elements and the porose nature of the membranes are interpreted as primitive. Thus, the nature of vessels in Acorus is compatible with the recent idea that Acorus (Acoraceae) is the sister group to the remainder of the mono cotyledons. Cheadle’s (1942) view that presence of vessels only in roots is primitive in monocotyledons was based on light microscopy. The addition of SEM data has revealed new vessel distributions. The pores in pit membranes in end walls of vessels of Acorus are larger than pores in pit membranes in end walls of tracheids in vessel-less dicotyledons. The difference between vessel elements and tracheids in groups such as Acorus may be not the absence of pit mem branes in perforation plates but the size of pores in the membranes of the perforation plates. The conductive capabilities and air bubble transmission capacities of pores of various sizes in pit membranes need physiological analysis in view of SEM findings on porosity in pit membranes of vessel perforation plates.

Introduction Asia (Siberia and Manchuria), and A. gramineus Ait., from Japan and Southeast Asia (Ohwi 1965). There In recent years, DNA-based phylogenies have given are numerous synonyms for both species. Acorus cal enhanced significance to the relationships of Acorus, amus has become naturalized in Europe (Ohwi 1965). often included in but now more often placed Cheadle (1942) hypothesized that vessels in mono- in its own family, Acoraceae. The results of Duvall et cotyledons appeared first in roots, then phylogeneti al. (1993) on rbcL sequence data, Nadot et al. (1995) cally spread into stems, leaves, and inflorescence axes. on rps4 sequence data, and Davis (1995) on chioro He further hypothesized that phylogenetic specializa plast DNA restriction site variation place Acorus (and, tion of vessels, as judged from perforation plate mor in the case of Davis [1995], a second , Gymnos phology, has followed the same sequence. The data tachys) in a basal position in monocotyledons. Davis suggesting these conclusions have been updated by (1995) opts for “the placement of either Acorus or Wagner (1977). Until now, our knowledge of vessels Acorus + Gymnostachys as the earliest-diverging lin and their perforation plates in monocotyledons has eage within the monocots” (p. 521). French and Kes been based on light microscopy studies. Light micros sler (1989) find that Gymnostachys, which is less well copy is undoubtedly useful in defining the nature of known than Acorus, is also isolated; Duvall et al. perforation plates in monocotyledons that have marked (1993) offer strong support that Gymnostachys is an differences in morphology between end walls and lat aroid, and the data of Davis (1995), who studied few eral walls (e.g., simple perforation plates). However, genera, offered only weak support for a dade consist in tracheary elements of monocotyledons with long ing of Acorus and Gymnostachys. We did not have scalariform end walls, light microscopy cannot reliably material of Gymnostachys. Regardless of the ultimate give information about the presence or absence of pit placement of Gymnostachys, there is unanimous agree membranes in the end walls. In some monocotyledons ment that Acorus is the outgroup to the remainder of with long end walls, the morphology of the perforation the monocotyledons. Thus, the nature of conductive plate differs from that of the lateral walls (perforations tissue in Acorus is of prime importance in our under wider than lateral wall pits), leading to the assumption standing of evolution of xylem in monocotyledons. that pit membranes are lacking in what appears to be Acoraceae was recognized family as a as long ago a perforation plate. This assumption may be based on as 1822 by C. A. Agardh (see Takhtajan 1987). Until the fact that this is frequently true in dicotyledons. the molecular studies cited above added reasons for However, in many “primitive” woody dicotyledons recognition of Acoraceae, the two genera Acorus and with a wide range of plesiomorphic features, scanning Gymnostachys had often been treated as a tribe, Aco electron microscopy (SEM) reveals remnants of pit reae, in the subfamily Pothoideae of Araceae (see, e.g., membranes in the perforations, ranging from extensive Dahigren et al. 1985). Grayum (1987, 1990) favored to vestigial (Carlquist 1992). Such SEM data, based removal of Acorus as Acoraceae. According to most on secondary xylem, are not applicable to monocoty authors, Acorus includes two species, A. calamus L., ledons, where one is concerned with primary xylem. from eastern North America and eastern temperate However, our studies of two dicotyledon families that lack secondary xylem, Nymphaeaceae (Schneider and ‘Author for correspondence and reprints. Carlquist 1995a, 1995b; Schneider et al. 1995) and Manuscript received May 1996; revised manuscript received Sep Cabombaceae (Schneider and Carlquist 1996a, tember 1996. 1996b), have shown relatively little difference between

51

p.m. 5 figures all in = Bars

membranes. pit in

absent

are

pores

element; vessel of

wall Lateral

4, Fig.

membranes. pit

the of remnants threadlike by other mostly each from separated large,

are

pores in

which

plate

a perforation of

Portion

3.

Fig.

pit membranes. in

the holes, larger into coalesced have which of some

pores,

circular

small

numerous

showing

plate,

a perforation of

2, Portion

Fig.

perforations. of the most in seen be may large to minute

from ranging

pores

plate;

perforation of

a

two-thirds 1, About

Fig.

photographs.

SEM from gramineus

Acorus of roots from of vessels Portions 1—4 Figs.

-4

the perforation In

holes.

larger into groups in lesced Cheadle by reported families other with

coa

have of

which some

remnants, membrane pit in along Acoraceae, expressions. xylary primitive retained

pores

circular

small numerous

illustrates 2) (fig. tion have to likely family monocotyledon a are Acoraceae

por

plate

perforation

The

found. be may pores a few the xylem, monocotyledon of evolution the in common

which in

and those

membranes in pit pores lack that quite are autapomorphies unless Thus, siomorphies.

pits

between

demarcation sharp a

not is

there although pie numerous of retention the for candidate a likely it

plate,

perforation of

part the

considered be not should makes (1995) Davis and al. (1993) et of Duvall grams

pores pits lacking

The

minimal.

or absent pores are 1), clado the in position its because Acoraceae with dons

(fig.

of the photograph

top

the at pit membranes the in monocotyle xylem of studies SEM initiated We

six of or five

ca.

In holes. circular

minute to area tion ecology. and morphology

perfora a of

half up to

occupying holes large atively ly vessel between synthesis better a achieve to able be

rel from

ranging

be seen, may

porosity of degrees all like will we increases, monocotyledons in distribution

1), (fig. plate

perforation

scalariform the of bars the and morphology of vessel knowledge our As 1965).

1).

Between (fig. shown

is end wall

of an two-thirds (Ohwi areas marshy wet in perpetually occurs Acorus

About elements.

vessel

of

end walls the on membranes 1975). (Carlquist habitats mesic more from chiefly

pit

of

nature

porose the

clearly very show 1—4) are (figs. that families monocotyledon also are These (1995).

gramineus Acorus

of of

roots longisections Radial Davis and (1993) al. et Duvall of cladograms the

with

(1977) Wagner and (1942) Cheadle of data the Results

compares one if primitive, more considered families

I.

Cheadle. to Vernon B. Tomlinson P via sent monocotyledon and the are These be reinvestigated. should

Birdsey by

Monroe

in Florida

cultivated 1252) was

(Cheadle pronounced not are and tracheids elements

vessel

A. gramineus of

specimen

The

Island. Rhode

Kingston,

between

differences

the

where

monocotyledons of ber

near collected was M-77)

(Cheadle A. calamus

of

Material

a num in

distribution

Vessel

described. been has erto

1940).

(Johansen

method chloride

than

hith

plant the within extensive more is

Acorus

in

acid—ferric

tannic

Foster’s of

modification

to

Northen’s ing

occurrence vessel that

reports

study Our

microscopy.

correspond combination green

safranin—fast a

with

stained

light

of basis

the on

reported

distributions the than and slides on glass mounted were microscopy light for

extensive more is SEM, with revealed used as be to sections occurrence, Other SEM. Nanolab Lomb and Bausch

that vessel a with possibility examined the is and there sputter-coated, microscopy, of paraffin, light cleansed

stubs, by aluminum on mounted demonstrated were be cannot sections that Some pared. walls end in rosities

were

pre

(rhizomes) stems and roots po of sections reveals tudinal SEM Because microscopy. light on based

longi

radial and

transverse Both 1940). (Johansen

methods are 1977) Wagner 1942; (Cheadle

monocotyledons

employed commonly to the

according sectioned and

paraffin

in

morphology

and

distribution

vessel of

Reports

in

embedded were portions stem and

Root Ferren. Wayne

be investigated. to needs walls

and Thorsch, Jennifer

Mary Cheadle, of

courtesy

the through

end

such of significance physiological

The

friction.

provided were collection

this

from Specimens

species.

more

considerably

offer

must

remnants

membrane

each of segments root and stems several

we

studied lections;

pit

with

plates

perforation

scalariform on membranes col in these present were individual many how know

pit

porose the but plates, perforation do not simple do we plants, than entire than rather preserved were roots

flow water to and stems of resistance more segments offer Because Barbara. likely Santa membranes California,

of University at pit the of Collection devoid Cheadle I. plates Vernon the perforation from Scalariform discussed.

obtained

were alcohol, ethyl

aqueous dilute or stored widely in neus, appreciated been yet not has of which cance

grami Acorus and calamus

of Acorus

stems and

Roots signifi the patterns, morphological of a series form

walls end porose with or plates perforation in nants methods and Material

rem membrane pit with elements tracheary These

1992). (Carlquist states

character plesiomorphic Cabombaceae.

in rich

dicotyledons

in woody to

be they appear as and Nymphaeaceae in findings to similar are plates

vessels, of

plates

perforation in

common are remnants in perforation remnants membrane pit with cotyledons

membrane pit

whether to

find

likely we are families, di- woody primitive in seen patterns the Thus, ment.

such

of

study

By SEM.

with

of study worthy therefore our equip with resolvable any least at porosities, lack

are

and xylem

monocotyledon

in states character ized pits wall lateral the whereas sizes, varied of porosities

on unspecial

information

to

provide are likely plates, with membranes, pit porose bear walls end the that

perforation

scalariform with

vessels

have to and roots, shows SEM However, microscopy. light using urations

only in

vessels

to

have

(1977) Wagner and (1942)

config wall secondary in walls lateral and walls end

SCIENCES PLANT OF JOURNAL INTERNATIONAL 52 U

C,,

L) z w

(D z

- CD c-) 1-

.

-z I —::;----

— I

- I 54 INTERNATIONAL JOURNAL OF PLANT SCIENCES plate (fig. 3), lysis of pit membranes is more extensive. walls of vessel elements of both roots and stems of Although some circular pores are evident (fig. 3), larg both species of Acorus, and we designate these cells er holes in the pit membranes, separated from each as vessels. The lateral walls of tracheary elements in other by threadlike pit membrane remnants, are pres both roots and stems of both species lack any percep ent. tible pores. The pores in the pit membranes of the The lateral wall pitting of A. gramineus (fig. 4) vessel elements are often small, and their small di shows narrower pit areas and wider wall bands be ameter doubtless likely confers higher friction for wa tween them as compared to the perforation plates (figs. ter flow than would be the case if no pit membrane 1—3). No pores could be detected in the pit membranes remnants were present. The pores in pit membranes of of lateral wall pits of vessels. A portion of a lateral perforation plates of Acorus are generally larger than wall of a root tracheary element of Acorus calamus the pores found in the margo regions of conifer tra (fig. 5) shows scalariform and transitional pitting. Stri cheid pit membranes (Frey-Wyssling 1976). The Aco ations are evident in the pit membrane of the lateral rus pores are also larger than pores illustrated in end wall pitting shown in figure 5 (arrow). Striations in pit wall pits of tracheids of the vessel-less dicotyledons membranes are evident in the perforation plates of A. Bubbia (Cariquist 1983) and Tetracentron (Carlquist gramineus (fig. 1, near bottom, arrow; fig. 3, arrow at 1988). While noting the larger size of pores in the end upper left). wall pit membranes of Acorus, we note that the defi The perforation plate portion of an A. calamus root nition of a vessel element is changing as a result of (fig. 6) illustrates perforations as seen from the lumen SEM studies: vessels can no longer be defined in terms side of the cell; absence of a pit membrane in the cen of absence of pit membranes on end walls, but we ter perforation may be an artifact, but the circular must take into account the size of pores in membranes pores in the lower perforation are undoubtedly natural. of end walls in those angiosperms in which extensive The perforation plate portion of an A. calamus root pit membrane remnants are present in perforations. (fig. 7) illustrates perforations as seen from the outside Coalescence of neighboring pores into larger holes can of a tracheary element; the primary wall is intact de be considered one possible criterion of a vessel. The spite separation of this vessel element from a neigh pit membrane remnants in perforations, and the pores boring vessel element. Striations are prominent in the of moderate size in these remnants can be considered pit membrane remnants, and circular pores are con primitive expressions of vessel formation, indicating spicuous in the membranes. incomplete lysis of the pit membrane at vessel element End walls are represented in the rhizome tracheary maturation. These primitive structural conditions in element portions shown (figs. 8 and 9). In both of these Acorus agree with the basal position accorded Acora photographs, end walls are represented. The narrow ceae by Duvall et a!. (1993), Davis (1995), and Nadot ness of the perforations, however; is more like the nar et al. (1995) and is in agreement with the designation row lateral wall pits of tracheary elements than is true of pit membrane remnants in perforation plates of di- of roots: there is less differentiation of end walls from cotyledons as a plesiomorphic character state (Carl lateral walls in rhizomes than in roots of Acorus. A quist 1992). The long scalariform perforation plates of perforation plate of A. calamus (fig. 8) viewed from the primitive woody dicotyledons in which pit rem the lumen side of the cell shows an irregular surface nants occur and the presence of such remnants are fea on the secondary wall. In most of the perforations tures that could provide friction for water flow, in shown in figure 8, several circular to oval pores can comparison to the minimal friction offered by vessels be observed. The rhizome vessel element of A. gra with simple perforation plates. The relatively greater mineus (fig. 9) is viewed from the outside of the wall; friction of the primitive perforation plates is likely not the adjacent vessel element has been sectioned away. of much negative selective value in plants of wet or Two perforations shown (fig. 9) lack any wall rem marshy places where water is perpetually available to nants; this absence may be an artifact due to the sec roots and where transpiration is moderate because leaf tioning process. The remainder of the perforations (fig. surfaces are relatively restricted. Acorus ecology and 9) contain intact pit membranes with one or more leaf morphology (Ohwi 1965) accord with these con pores. We have not figured lateral wall pits for the ditions. rhizomes of the two Acorus species; they are much The striate nature of pit membranes in end walls and like the perforations except that, as in the roots, the also in some lateral walls of vessels of Acorus is note lateral wall pits lack any perceptible pores. worthy, because similar striations occur in primary xy Discussion lem of dicotyledons. We reported this condition for Nymphaeaceae in Barclaya (Schneider and Carlquist Cheadle (1942) and Wagner (1977) reported that in 1995b) and Eurvale and Victoria (Schneider and Carl Araceae (in which they included Acorus) vessels are quist 1995a). Such sthations may be much more wide restricted to roots. Cheadle (1942) states that in Ara ly distributed in vascular plants—we found them in ceae, “the stems of several (including Acorus calamus the fern Woodsia (Carlquist and Schneider; in press)— L.) were injected with India ink to confirm the absence but primary xylem has not, as yet, been sampled ex of vessels” (p. 444). Using SEM, however, we con tensively with SEM. clude that porose pit membranes are present in end SEM offers an ideal tool with which to examine CARLQUIST & SCHNEIDER—VESSELS IN ACORUS 55 I t

, I

p4 V i%t1$J

I. ii aA1A7

-

1.

8LI 9 Figs. 5—9 SEM photographs of vessel elements from roots and rhizomes of Acorns. Figs. 5—7, Vessel portions from roots calamus. Fig. 5, Portion of lateral wall; pits are scalariform of Acorus and transitional and have somewhat striate membranes. perforation plate, seen from inside of vessel; Fig. 6, Portion of perforation at bottom contains small pores; pits along membranes. Fig. 7, Portion left edge of photograph have striate of perforation plate, seen from outside of vessel; pit membranes and oval are markedly striate; relatively large circular pores in membranes are evident. Figs. 8, 9, Portions of perforation plates from rhizomes of Acorus. Fig. 8, Perforations calamus, seen from inside of vessel; perforations at left contain from A. several large pores each. Fig. 9. Perforations from from outside of vessel; although two wider Acorus gramineus seen perforations lack pit membranes (likely an artifact), the others to several circular pores are contain membranes in which one evident. Bars in all figures = 5 .tm. 56 INTERNATIONAL JOURNAL OF PLANT SCIENCES porose pit membranes in perforation plates and dem However, the morphology of vessels in Acorus is of a onstrate presence or absence of pit membrane rem markedly primitive type. Moreover, SEM studies may nants. Our results show that Acorus has vessels in both reveal vessels to be more extensive organographically roots and stems, rather than just in roots as claimed in other monocotyledons in which vessels are currently by Cheadle (1942). According to the Cheadle criteria reported in roots. In any case, our data are not in con of progressive organographic spread and specialization flict with the basal position hypothesized for Acorus of vessels throughout the plant body in monocotyle within monocotyledons by Duvall et al. (1993), Davis dons, this would represent a moderate advance in ves (1995), and Nadot et al. (1995). sel phylesis over presence of vessels in roots only. literature cited

Cariquist S 1975 Ecological strategies of xylem evolution. Univer Grayum MH 1987 A summary of evidence and arguments sup sity of California Press, Berkeley. porting the removal of Acorus from the Araceae. Taxon 36:723— 1983 Wood anatomy of Bubbia (Winteraceae), with com 729. ments on origin of vessels in dicotyledons. Am J Bot 70:578—590. 1990 Evolution and phylogeny of the Araceae. Ann Mo 1988 Comparative wood anatomy. Springer-Verlag, Berlin. Bot Gard 77:628—697. 1992 Pit membrane remnants in perforation plates of prim Johansen DA 1940 Plant microtechnique. McGraw-Hill, New York. itive dicotyledons and their significance. Am J Bot 79:660—672. Nadot 5, G Bittar, L Carter, R Lacroix, B Lejeune 1995 A phylo Cariquist S, EL Schneider In press SEM studies on vessels in genetic analysis of monocotyledons based on the chloroplast gene ferns. 1. Woodsia obtusa (Sprengel) Torrey (Dryopteridaceae). rps4, using parsimony and a new numerical phenetics method. Am Fern J. Mol Phylogenet & Evol 4:257—282. Cheadle VI 1942 The occurrence and types of vessels in the various Ohwi J 1965 Flora of Japan. Smithsonian Institution Press, Wash organs in the Monocotyledoneae. Am J Bot 29:441—450. ington, D.C. Dahlgren RM’L HT Clifford, PF Yeo 1985 The families of the Schneider EL, S Carlquist 1995a Vessel origins in Nymphaeaceae: monocotyledons. Springer-Verlag, Berlin. Euryale and Victoria. Bot J Linn Soc 119:185—193. 1995b Vessels in the roots Davis JI 1995 A phylogenetic structure for the monocotyledons, as of Barclaya rotund(folia. Am J Bot 82:1343—1349. inferred from chioroplast DNA restriction site variation, and a 1996a Vessel origin in Cabomba. Nord J Bat (in comparison of measures of dade support. Syst Bot 20:503—527. press). 1996b Vessels in Brasenia (Cabombaceae): new perspec Duvall MR, MT Clegg, MW Chase, WD Clark, WJ Kress, HG Hills, tives of vessel origin in primary xylem of angiosperms. Am J Bot LE Eguiarte, JF Smith, BS Gaut, EA Zimmer, GH Learn Jr 1993 83:1236—1240. Phylogenetic hypotheses for the monocotyledons constructed from Schneider EL, S Carlquist, K Beamer, A Kohn 1995 Vessels in rbcL sequence data. Ann Mo Bot Gard 80:607—619. Nymphaeaceae: Nuphar, Nytnphaea, and Ondinea. mt j Plant Sci French JC, CT Kessler 1989 Molecular systematics of Araceae: are 156:857—862. Acorus and Gyinnostachys aroids? Am J Bot 76:242 (abstract). Takhtajan A 1987 Systema magnoliophytorum. NAUKA, Lenin Frey-Wyssling A 1976 The plant cell wall. Pages 1—294 in Hand grad. buch der Pfianzenanatomie. Vol 3, Pt 4. Gebruder Borntraeger, Vv’agner P 1977 Vessel types of the monocotyledons: a survey. Bot Berlin and Stuttgart. Not 130:383—402.