Molecular phylogeny of the subfamily Bostrychioideae ...

Molecular phylogeny of the subfamily Bostrychioideae (Ceramiales, Rhodophyta)... Giuseppe C Zuccarello; John A West Phycologia; Jan 2006; 45, 1; Academic Research Library pg. 24

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

ZI/ccarel/o & Wesl: Molecular phylogeny of Boslrychia IITchia pil/I/aw Ji. Tanaka & Chihara was synonymized with B. calliptera (Montagne) Montagne (Zuccarello & West 1002) because the absence of cortication in B. /iil/I/aw was shown not to be confined to a monophyletic lineage distinct from corticated B. calliplera. There has been extensive nomenclatural confusion around the species B. radical/s (Montagne) Montagne with morphologically similar species (B. ril"ll!aris Harvey) and forms (f. Iypica. f. depal/perala E. Post. f. haplemll/al/ica E. Post. f. II/ol/i/i/c}nl/is E. Post) recognized. but most names synonymi/.ed by King & Puttock ( 19R9) and the recognition of only two species. B. radinms and B. 1I/0ril~i(/}/(/ (Sonder ex Klit/.ing) J. Agardh for 13. radicalls f. I/IOI/i/!(orll/is. Later molecular work (Zuccarello & West 1995. 1003: Zuccarello el 01. 1999a) again raised doubts as to the utility of the morphological character used by King & Putlock (19/;9) to differentiate between B. radical/s and B. II/oril~ial/a (i.e. the presence or absence of monosiphonous laterals). revealing large amounts of genetic variation. as well as several monophyletic lineages. within the B. radical/s/B. II/oril~ial/a complex that did not conform to any clear-cut morphological character. The absence of monosiphonous laterals has also been used as the main character. along with more developed peripherohaptera. to separate B. lelll/issill/a R.J. King & Puttock from B. sill/p/icil/scl/la Harvey ex J. Agardh (King & Putlock 19R9). The species B. leI/ella is morphologically highly variahle (King el al. 19RR). but was reduced to a single species with two subspecies (ssp. leI/ella. ssp. flagellijc'ra (E. Post) RJ. King & PUtlock) hy King & Puttock (19R9). This proposal was not endorsed hy Kumano (1002) who still maintained B. flagellijc'ra E. Post for B. leI/ella ssp. f!agelli/c'ra. In this paper we further address the taxonomy of the subfamily Bostrychioideae from a molecular perspective. using the plastid-encoded rhcL gene and partial sequences of the nuclear-encoded large subunit of the rihosomal RNA gene (26S rRNA). The aims of this paper are: (I) to determine whether Boslr\"chia and 51iclOSiphollia are distinct and independent monophyletic lineages with the Bostrychioideae: (1) to understand the phylogenetic relationships of the species within the subfamily: (3) to check for monophyly of the described species: and (4) to uncover any biogeographic patterns of clades.

MATERIAL AND METHODS Methods for collection. isolation and maintenance of cultures are presented in West & Zuccarello ( 1999). Cultures for molecular analysis have been maintained for up to 20 years. Culture experiments on B. sill/plicil/scl/la and B. lel/l/issiJlla to investigate morphological variation were done on two clones of B. sil/lplicil/scl/la (1963. 35(2) and three clones of B. 1l'J1l/issill/a (2749. 330R. 3546). including hoth sexes and diploids. each grown at two temperatures (23 2: 2°C. 14 2: 2°C): two Iighl levels (33 2: 4 I-lmol III ~ S I: 17 2: I I-lmol m-~ S"I) and either stationary or on a rotary shaker 00 rpm). Samples sequenced are listed in Table I: we have included all presently recognized Bo.l"!Jychia species and all but two 51iClOsipilOI/ia species [5. gracilis RJ. King & Puttock: 5. mga (J.D. Hooker & Harvey) J.D. Hooker & Harvey. both of which are uncommon and restricted to New Zealand]. We

25

also include BOSIITchiocolax all.l"lralis Zuecarello & West ( 1(94). a parasite of Boslrychia radiCllIIS that is nested within the subfami Iy (Zuccarello el al. 2(04). DNA extraction followed a modified Chelex extraction method (Zuccarello el al. 1999b). Amplification of an approximately 900-1000 bp region of the nuclear-encoded 26S rRNA gene. corresponding to the middle third of the molecule (Y-fragment: Harper & Saunders 200 I) followed the procedure in Zuccarello & West (2002). Amplification and sequencing of the plastid-encoded large subunit of the ribulose bisphosphate carboxylase/oxygenase gene (rhcL) used amplification primers presented by Nam el al. (2000) and additional sequencing primers listed in Freshwater & Rueness (1994). The polymerase chain reaction (PCR) procedure followed Zuccarello cl al. (2001). All PCR products were electrophoresed in 1-2'1r agarose to check product size and sequenced following procedures in Zuccarello el al. (1999b). Sequences were assembled using the computer software supplied with the ABI sequencer. and aligned with Clustal X (Thompson el al. 1997). All sequences were compiled in SeAl version2a I I (Rambaut 1996). Phylogenetic relationships were inferred with PAUP"'4.0bI0 (Swofford 20(2). Outgroups used were selected from available GenBank deposits or sequenced for this sllldy. The rbcL sequences for the following species were downloaded from GenBank (Table I): RllOdomela cOII!('}"I'oidcs: Neorhodomela larix: Lallrel/cia ohlllsa; Pol"siphollia delllldala; CeJllroceras clani/alllm: Gelidiwn pl/sil/lIIl1. For the 26S rRNA gene. sequences from GenBank were: Caloglos.w lepriellrii: Celllroccras clal"li/alllm; G. PIlSiI/lim (see Table I for authorities). Maximum-parsimony trees (MP) were constructed in PAUP"'. using the heuristic search option. 500 random sequence additions. tree bisection reconnect ion (TBR) branch swapping. unordered and unweighted characters. gaps treated as missing data. For rhcL analysis rearrangements were run to completion. whereas for the 26S rRNA data only 100 tree were saved per replicate. The program Modeltest version 3.06 (Posada & Crandall 199R) was used to find the model of sequence evolution that best fits each data set by a hierarchical likelihood ratio test (a = 0.0 I) (Posada & Crandall 200 I). When the best sequence evolution model had been determined. maximum-likelihood (ML) was performed in PAUP* using the estimated parameters (substitution model. gamma distrihution. proportion of invariable sites. transition-transversion ratio) (one to five random additions) (Table 2). Distance trees were constructed using neighbour-joining reconstruction (NJ) either using the Modeltest parameters (rhcL data set) or the LogDet distance calculations (Lockhart el al. 1994) (26S rRNA data set). LogDet deals better with unequal hase frequencies and branch lengths as was seen in the 26S rRNA sequences of 5. kelallellsis (Grunow ex E. Post) R.J. King & Pultock (see Results). Support for individual internal branches was determined by hootstrap analysis (Felsenstein 1985). as implemented in PAUP*. For bootstrap analysis. 1000 bootstrap data sets were generated from resampled data (five random sequence additions). for both the MP and NJ analysis. Bayesian analyses were performed using MrBayes v3.0b4 (Huelsenbeck & Ronquist 200 I). Three separate runs were performed with a random starting tree and run for 1 x 10" generations. keeping 1 tree every 100 generations. The first I


26

Phycologia. Vol. 45 (I), 2006

Table 1. Sample u cd in this sllldy. COlleclion information included: location. dale collected ('olleeror if not authors). Culture GenBank accessions numbers _ pecies Bo tr chia 8. t:alliplut/ (Montagne) Montagne B. ealliplera

8. ealliprera B. eofliplera

8. eafliplera B..f/agelli/era E. POSt B. jlagell!fera B..f/ogelli/era B. IWI1'eyi Harvey

8. harveyi B. mOlllGglld Harvey

8. III(I/llagnei B. /IIorirzitllUl ( 'onder ex KUlzing) J. Agardh

8. morilziano 8. lIlorilzitJlltl B. lIIol'irzitJlla B. lIIorilziontl B. lIIurirziona B. lJIorirziana B. II/orir"ian(/ B. II/orirziano

B. piluli/era Montagne

Rio Silio Grande. Hha clo Cardoso. SP. Braz.il, 18 Jan. 1990 Some omapan. Vera Cruz. Mexi o. 10 Feb. 1994 Lim hu Kang. Singapore, 16 Jun. 1989 Ri Pereque. IIha do Cardoso. SP Brazil. 5 Apr. 1990 hund;1 Bay. QLD, Australia. 28 Scpo 1991 ( hri Pultock) Botany Bay. Sydney. Sw. Australia. 4 Jan. 1991 R;!inbQw. QLD. Australia. 29 OCI. 1995 PIage de Ouano. Nm Caledonia. I Feb. 1995 Havelock, TAS. Australia. 7 Jul. 1998 Doug McBridc) ew otfolk. DerWent River, TAS, Australia. 14 ct. 1998 (Doug McBridc) Twin Cay. Belize. J 5 Jul. 1989 (Laurie Su.1I ivan) Parr.! A"u. Maranhao. Brazil. 20 I ov. 1996 Buenaventura, lombia. 10 Sep. 1991 (Enrique Peria) Rio Guire. Edo Sucre. enezuela. I j Apr. 1991 Millers Landing. Wilson Promontory. VI . AUSlralia. 17 Dec. 1988 Beachwood. atal. South Africa, 0 I. 199/ (Alan Critchley) usa Dua. Bali. Indonesia. 10 May 1999 akama Ri er. lriomote I. land, Okinawa, Japan. 2 Mar. 1995 (M. Kamiya W. Sa wang. Sulawesi. Lndl'lnesia. 16 Nov. 1994 Pohnpei. Microne ia. 29 Sep. 1989 (Bill Rainey) Farasan I.. Saudi Arabia, 8 Jul. 2000 (M. Hussain) West Branch. Demerara River. Gu ana. 30 Jan. 2002 (Michael KUnzen)

8. mdicalls (Mol1lagne) Momagne B. radicans B. rodicalls

8. radicQns B. radiu/fls

B. radicans B. mdic'QIIS

8. radicalls B. radiCC/fls B. s orpioides (Hudson) Montagne B. simplicillscLlla Harvey ex J. Agardh

B. simplici/lscl1la B. simplici/lsc;ula

8. renella (l...amouroux) J. Agardh B. rellella B. lenella

8. lencl/a B. lenel/a 8. lcnetla

number'

rbcl~

26S rR A

3042

AY920805

AF382926

3400

AY920S06

AF382925

2961

3065

AF382932 AF382937

3216

AF382938

3113

AY920854

3553 3821

AY920856 AY920855

0618

AY92007

AY920857

0670

AY920R08

AY920858

2980

AY920859

3647

AY920860

3189

AY920812

AY92086I

3149

AY920SI1

Y920 62

2934

AY92009

AY920863

3204 3955

AY920864 AY920815 AY920814

AY920866

637 3453

AY920 13

AY920865

3001

AY920810

4069

Ayno81

4200

AY920817

Y92086

AY920818

AF_59407 AY920869 AY920870

A

Siio Sebastiiio. SP. Bra.zil. 2 Jul. 1982 Florence Bay. Magnetic 1.. QLD, Au tralia. 4 Jun. 1987 Rio Silio Grande. IIha do ardosu. Sr.. Brazil. 24 ov. 1989 Teluk Awang. Lumbok. Indonesia. -7 May 1999 Buena ViSta, Guatemala. 22 Mar. J 993 stero COYOle. Bahia an Ignacio. BCS. MeXico, 26 Mar. 1991 Cape Fear swary, SA 27 Jan. 1991 (Don Kapraun) Stamford. CT, SA. 13 Aug. 1991 (Charlie Yarish) Roscoff. Prance. 1999 (Frithjof Kuepper) Mandai, Singapore, 13 Jun. 1989 Forster, NSW. Ausu';llia. 23 Oct. 1995 Ballina. SW. Australia. 12 Feb. 1997 Tawi-Tawi, Mindanao. Philippines 15 Jan. 1988 Arembepe, Ballia. Brazil, 20 Jul. 1982 Bouncing Stones Beach. QLD. Au ·lralia. 13 Jun. 1987 La Parguera, Puerto Rico. 2 C)v. 19116 Darwin, NT. Australia. 4 Jun. 1989 Talabong. Negros O."ienlal. Philippines. 19 JuJ. 1990

2649 2837

AY920867

017

AY9208IQ

3980

AY920824

AY920874

3367 3124

AY920823 AY92082I

Y920 73 AY920872

3116

AY920820

AY92087I

3175

AY920822

130 2963 3562 3694

2871

AY920825 AY9208_6 AY920827 AY920828 AY920832

AY920875 AY920876 AY920877 AY920S78 AY920882

2641 2815

AY920829 AY92083I

AY920 79 AY92088I

2756 2986 ORO

AY920830

AY920880 AY920883 AY920884

Ayn0833 AY920834


Zuccarel/o & West: Molecular phylogeny of Bostrychia

27

Table l. Continued.

pecie

Laguna de la Reslinga. I. Margariw. newela. I May I 9 J Tutuila. meri an Samoa. _6 Jul. 1991 (Bill Rainey) v. IIha do Ilapari a. Bahia. Brazil. II 1996 Mangro e Tnlil. Broome. uSlfalia. 1 Jun. I 97 We tern Port Ba . VI u'lralia.2 a I. 19 6 William town, 1996 Brunswick Head. . Au lralia. 23 OCI. 1995 Go. ford. NSW, U tralia. 4 Jan. 1991 Gosford. SW. ustralia. 4 Jan. 199J Broughlon reek. uSlralia. II Jan. 199

B. fell ella

B. fell ella

B. felle/lo B. fellella B. fellU;s ima R.J. King

'PUll

k

B. felluissima

8. fem';.!' ima 8. temli ;ma B. felll/;ss;ma 8. fell"; ;II/{/

31 I

Y92036

Y920

6

6 5

920 37

Y920

7

374

A 920

2747

Y920 39

Y920

5 I

Y9_0 44

V9209

920 43

Y920

9

9

3110 III 30

Y9_0 42 Y9_0 40

Y92041

Y920 90

D 5

Y92045

Y920 94

2 76

Y _046

Y90891

ti losiphonia · tlrbl/ cIlia Har ey) R.J. King & Putt 'k · ;llIr;caUl

B r de aim- incent) P. . Silva

· ;llIr;cata

472

S. ;lIIricafa

S. keloll Ilsi

e Gruno\ ex E. Po I R.J. King & Putlock

T.

· kelallell ;

· tallgatell; (Po I) R.J. King

u tra-

u lIaIia. 2

S. kelollell i

fri a. 20 Jan.

k

21-t

'J.. We t

AY 20 49

Y9205

6

Y920

7

Y920 Y920 99 Y920900

3241

A 920 50

Y920901

266

Y920 51

Y920902

fri a. 22 De .

· fallgmen'; 80 Ir)' hio olax ausfral; Zuc arell

29

920 4

961

pro

PUll

67

Zea-

Y920

u tra·

Y920903

Rhodomelaceae Louren ia obmsa (Hudson) Lamouroux L rigida J. gardh

Ireland Be !any Ba

W.

U

Iratia, I I MOl

G21

_818 I 920 52

2000 eorhodomela lari . (Turner) M:l uda Rhodomela (lJIfervoides Hudson) P. . il a Polysiphollia delll/dara Dillwyn) Grcville alogia' a leprieurii (Montagne G. arten elllrocera' lavomm (C. gardh) M ntagne a/llacalllh/ls I/smlat/l Turner) Kiitzing

59499 FOg33 1

Kiel Bight. German Plymoulh. Devon. nilcd Kingdom Picnic Point, Georgc' River. S Iralia. 15 cb. I 99

F 42914

F522 17

D

F25949 Ima Bay. agneli I land. QLD. Iralia 6 Jun. 1987

X 105 generations (burn-in) were discarded and the remaining 9000 trees (representing 9 X 105 generations) were used to calculate a 50% majority-rule tree. The three separate runs were compared (to determine if a similar topology was reached), and the saved trees of all three runs were combined and used to produce a majority-rule tree to determine the posterior probabilities for the individual branches.

us-

2

F259414 Y920 04

RESULTS Partial 26S rRNA phylogeny Fifty-six samples of the Bostrychioideae were used for phylogenetic reconstruction. including one of the parasite Bostrychiocolax australis. Only three species were represented by


28

PhycoLogia, Vol. 45 (I), 2006

only one accession [Bostrychia piluL((era Montagne, B. scurpioides (Hudson) Montagne and S. arbuscuLa (J.D. Hooker & Harvey) RJ. King & Puttock]. Complete sequences range from 875 to 1081 bp. After removal of regions of ambiguous alignment a data set of 798 bp was produced with 266 potentially phylogenetically informative sites. This produced over 15,000 MP trees of 838 steps (CI = 0.5835). ML analysis produced a single tree with a -In L score of 5349.24024. LogDet distances were also used to produce a NJ tree topology shown in Fig. I. The difference between the different reconstruction methods was the position of S. keLanensis (three samples). This species had a highly divergent 26S rRNA sequence as can be seen from the long branch in Fig. I. The Mp, NJ and Bayesian analysis placed S. keLanensis as a poorly resolved sister group (72% MP bootstrap percentage (BP), 0.90 posterior probabilities (PP» to a clade containing the cladohapteron-possessing species B. radicans, B. moritziana, B. piLuli(era and the parasite BostrychiocoLax austraLis. The ML analysis placed S. keLanensis in a derived position with certain species of Bostrychia tenelLa (not shown), although this position did not receive BP support. The sister group relationship of S. kelanensis to species of Bostrychia possessing cladohaptera as seen in the NJ tree is also similar to the results of the rhcL (Fig. 2). Two consistently produced major clades with all reconstruction methods, except for ML (see above), are (1) a group containing the species of Bostl)lchia and Srictosiphonia with cladohaptera (B. radicans, B. moritziana, B. piLulifera and S. keLanensis), plus the parasite BostrychiocoLax austraLis and (2) a clade containing all species of Bostrychia and Stictosiphonia that produce peripherohaptera. These will be referred to as the 'cladohapteron-clade' and 'peripherohapteron-clade', respectively. The relationships within the cladohapteron-clade show several moderately to well-supported lineages. These lineages are consistent with other studies on this group (Zuccarello & West 2003). Two samples are found in this dataset that are not found in previous studies. Bostrychia pilulifera Montagne possesses cladohaptera but is corticated, in contrast to the other species, and is polysiphonous throughout. It is associated with uncorticated polysiphonous B. radicans samples. BostrychiocoLax austraLis. a parasite of Bostrychia (Zuccarello & West 1994), is well embedded within the remaining B. radicans/B. moritziana/B. piLuL(fera samples. Most of the basal relationships within this clade are not supported, as has also been shown in other studies (Zuccarello & West 2003). The relationships within the peripherohapteron-clade are more complex. Again, many of the basal relationships are unsupported but several well-supported groups are evident. This clade presently contains 10 species names [B. calliptera (Montagne) Montagne (including the previously synonymized species B. pinnata), B. harveyi Montagne, B. montagnei Harvey, B. tenella (Lamouroux) J. Agardh, B. tenuissima R.J. King & Puttock, B. scorpioides, B. simpliciuscuLa, S. arbuscuLa, S. intricata (Bory de Saint-Vincent) P.c. Silva, and S. tangatensis (E. Post) R.l. King & PuttockJ. Several of these species are clearly not monophyletic. Contrary to the results presented in Zuccarello & West (2002) B. calliptera does not appear to be monophyletic. Although the three lineages of B. calliptera presented in Zuccarello & West (2002) are clearly seen in Fig. I, the clade

designated 'clade I' does not group with the other B. calliptera samples from clades 2 and 3 (Zuccarello & West 2002), although again this is not well supported. The next polyphyletic species is the complex containing B. tenuissima and B. simpliciuscuLa that is in three separate wellsupported lineages. These two species are separated by a single character: the presence (B. simpLiciuscuLa) or absence (B. tenuissima) of monosiphonous laterals. These three lineages correspond to the three RuBisCo spacer haplotypes of B. tenuissima reported in Zuccarello et at. (l999c) (H I, H2, H3). The present study shows that samples identified as B. simpLiciuscuLa are in two of the lineages along with B. tenuissima samples (H2, H3). RuBisCo spacer analysis shows that the B. simpLiciuscuLa samples have identical sequences to B. tenuissima samples of the same lineage (not shown). A third polyphyletic species is B. teneiLa. Samples identified as B. tenella ssp. flagellifera are in a clade distinct from other B. tenelLa and B. montagnei samples. Putting B. tene/la ssp. fiagellifera aside for now, we are still left with the core B. tenella clade, containing three well-supported clades. The first corresponds to samples of B. tene/La from both the Pacific and Atlantic Oceans (2871, 3080, 3181, 2641, 2986, 3753). The second clade containing samples identified as B. montagnei and a third clade contains samples of B. tenelLa, again from both the Pacific and Atlantic Oceans (2756, 3 141, 2815, 3655). The relationship between these clades is not resolved. Other relationships of note are (I) the sister group relationship (although not supported with partial 26S rRNA, but supported with rhcL; see Fig. 2) of heavily corticated S. arbuscuLa to uncorticated S. intricata; (2) the clear phylogenetic distinction between B. scorpioides and B. harveyi; and (3), the sister group relationship of B. .I'corpioides with B. tenella ssp. fiagellifera and some B. tenuissima samples (2747, 3581) of lineage H I (see above).

RbcL phylogeny Forty-eight samples of the Bostrychioideae were used for phylogenetic reconstruction. Only three species were represented by only one accession (B. piLuLifera. B. scorpioides and S. arbuscuLa). The data set consisted of 1163 bp with 440 potentially phylogenetically informative sites. This produced eight MP trees of 2177 steps (Cl = 0.3550). ML analysis produced two trees with a -In L score of 11,397.489. The ML topology of one of these trees is shown in Fig. 2. All reconstruction methods gave very similar results. Plotting of absolute distances vs ML corrected distance. including third codon positions, did not show significant saturation of this gene (data not shown). The rbcL phylogeny differs from the 26S rR A phylogeny mainly due to taxon sampling. With the rbcL primers employed we were not able to amplify samples of B. tenelLa ssp. fiagellifera, B. montagnei, or samples of B. calliptera from clade I and 2 (see above). We also sequenced a rhcL gene from BostrychiocoLax austraLis. which gave a sequence identical to its host Bostrychia radicans (2837) (data not included), which is not surprising because it is known that red algal parasites contain the plastids of their hosts (Goff & Coleman 1995). Other topology differences between the two data sets revolve around unsupported branches. The rbcL phylogeny again shows the unsupported grouping


ZlIccarelio & West: Molecular phylogeny of Bostrn'hia

29

99/ 2/0.

6MP/0. PP

5 Au traJia 21 ew Cal d nia Au tralia

H

L3 L2

H2 *

*

*

* *

. kelalletl i lind n ia . kelamm i 214 Au tralia . kelalll!lIs' Au tralia

' - - - - - - - Calogl SlI lepriel/Jii Celltr I!r clflvntlllll CIIIIIII 'III/tilll II tlllfltu Gelidiulll pll illlllll

L..-

L..L..-

- 0.00

ub titution / ite

Fig. I. J topology LogOct di lance' of panial 26 rR equen e data 0 pecic within the ubfamily Bo_tr chioidca ... = boot'trap . uppon > 9-'i! for both MP and J anal i plu PP > .95. ther branch support Ii. ted in the order (BP-MP/BP- JIPP). r a . pecified. Celidium pu ilium u. cd a outgroup. ample' with culture code de' ribcd in Table I or fr m en Bank in Mal rial and Method. HI. H . H indi ate Bos/rychia /elluissima haplOl pc lineages from Zuccarell e/ 01. (199 ). L I. L_. L3 indicatc B. callip/era lincage from Zuccarello We t (2002 (ee text).


30

Phycologia, Vol. 45 (1),2006

B. tellcl/a "743 Ausb:aha B. te/lel/a 2986 AustrLlliu

B. tenelln 2871 Philippine 13. fenel/a31 1 American Samoa .---------1 B. tene/la 3080 Philippines 8. tenel/a 2641 Brazil 71/70/1.0 8. tenel/a3141 Venezuela B. trmel/a 2756 Puerto Rico * B. ten ella 2815 Australia 1.B. tenel/a 3655 BrazjJ 75NJ/l.OP? ,....-_----:*_ _-1 D. lenlr/issillla 2747 Au tralia HI .--....:.:.-----l B. te/'luissillla 35 1 Au tralia l..... B. scorpioides France 1.OPP S. t(l/lgnlensis 3241 South Africa S. tangatensi 3266 South Africa 1."'_._ _--I B. telil/i /1ua3546 Australia H3 B. silllplieiuscula 2963 Singapore ,....---- . iutricata 3472 Australia . illtricala 3867 New Zealand . intn'catn 2876 South Atrica 1.S. arbllscula D33 New Zealand L~--.r---:: B. calliptem 3042 Brazil L3 B. cal/iplern 3400 Mexico B. telluissima 3308 Australia B. simplieil/scula 3562 Australia L---*-----l B. simplieil/scllin 3694 Australia H2 B. feuuissil1la 3111 Austral ia B. tenui sima 3110 Au tralia L---*--------t B.lmrtleyiD61 Au tralia B. hnrveyiD670 Australia 52/6 /1.0 .-----1 B. radicrlt/ 2649 BraziJ B. rndimlls 3017 Brazil L..-_ _ B. rndit;au$ 39 0 Indonesia 1.8. pilulifem Guyana '----- B. moritziarw3189 Colombia l..... B. moriLilllla 3149 Venezuela 8. radicans 3367 Guatemala B. radicaus 3124 Mexico * B. fa tieans 116 USA B. radicallS 175 USA * B. murit;::ulI1U 2934 Australia .-----f B. moribnnl1 3955 Ind nesia B. moritziana 3637 Japan -{===~B~,::"((fori/?illntl345 fnd nesia L_..:*~ B. mori/::itll1f1 001 Micronesia B. moritzillTlfl 4069 Saudi Arabia . kelnllen i 29 Australia * S. kelrmensis 214 Au tralia ...-----...;*;;....-----1.-------- Rhodomeln conferooides 1 eur}wdumelu Irll1~Y L-----*-----r--- LtlU1'erlcia ob/llsa wl/.rencia rigida LPolysiphonill delludnfa Centrocern clllvtltllm Geliditwi pu ilium

*

I

I

I

1..-------

L-

L-

- - - - 0.05 substitutions/site Fig. 2. ML lopology of rbcL DNA equ nce data of species within the subfamily Bostrychioideae. * = bOOtstrap support> 95'il for both MP and J 3Jlaly. is plu PP > 0.95. Other branch supports listed in the order (BP-MPIBP- J/PP). or as pecified. GelidiulI! pilsil/unt used as OIHgroup. Samples with culture coded de cribed in Table I or from GenBank in Material and Methods. 1-11. H2. H3 indicates 8. rellllissima haplotype lineages from Zuccarello e/ al. (1999c). L3 indicates B. callip/era lineage from Zuccarello & West (2002) (see text).


ZlIccarello & WesT: Molecular phylogeny of Bostrychia

of S. kelanensis with the other cladohapteron-containing species. Long branches seem to be less of a problem for the rbcL vs the 26S rRNA of S. kelanensis (i.e. all reconstruction methods gave similar topologies). The other relationships within the cladohapteron-clade are similar to the 26S rRNA phylogeny, and previous studies which contained larger sample sizes were used (Zuccarello & West 2003). although taxon sampling is not completely identical. The relationships within the peripherohapteron-clade are also similar to those indicated by the 26S rRNA data set. BosTn'chia simplicillscllla/B. Tellllissima is in three separate clades corresponding to the RuBisCo spacer haplotypes H 13 (Zuccarello eT al. 1999c). BO.I·Tn·chia scorpioides is a sister group to H I lineage B. Tenllissima samples, a difference between the two data sets being that S. TaIl/:aTellsis is the sister to these two clades. whereas in the 26S rRNA data set its position was unresolved. STicfOsiphonia arbllsclIla is a sister group to S. inTricma. BosTrychia harpeyi is again distinct from B. scOlpioides. The level of genetic variation within B. Tenelhl is quite high. and two distinct clades are seen.

Morphological observations Tier-cell numbers used in the separation of BosTrychia (2 tier cells/axial cell; Fig. 3) and STicTosiplulIlia (3-5 tier cells/axial cell: Fig. 4) by King & Puttock (1989) are consistent within species but are not consistent in our phylogenetic results with monophyletic genus-level discrimination. Cladohaptera within the species B. radical/s. B. lI1oriT:.:ial/a (Fig. 5) and B. pillllifera (Fig. 6) are formed on specialized branches that arise from the first (or second) axial cell at the base of primary determinate laterals. The hapteron apex is composed of elongated axial and pericentral cells. A second cladohapteron-type is seen in S. kelal/el/sis. in which the hapteran is formed at the apex of an unspecialized branch or at the apex of an indeterminate axis (Fig. 7). Peripherohapteron rhizoids are formed from pericentral or cortical cells on the ventral sides of the axis Ie.g. S. imricaTa (Fig. 8). S. wngaTensis (Fig. 9) and B. sill1p/icillscllla (Fig. 10)]. The two species B. sil/lplicil/scllla and B. Tel/llissill1a are distinguished from each other by the presence or absence. respectively, of monosiphonous laterals (Fig. I I). [n several culture isolates that initially had monosiphonous laterals. the plants of some produced only polysiphonous laterals for many years, whereas others retained monosiphonous laterals. We also looked for morphological variation between the three separate lineages of B. Tenllissima/B. simplicillscllla (H 1-3) identified by our molecular work, which was done by submitting the clones to a variety of culture conditions (see Material and Methods). Although morphology did change slightly under different conditions, no consistent expressions that consistently corresponded to the three molecular lineages were seen. We investigated the branching pattern of B. Tenella and B. Tenelhl spp. fiaRellifera and confirmed that successive branching occurred at almost every axial cell in B. Tel/ella (Fig. l2), whereas it occurred at intervals of two to five axial cells in B. fiagellifera (Fig. 13). There was very little deviation from this pattern.

31

DISCUSSION It is clear from our data that the continued recognition of two genera within the subfamily Bostrychioideae, which is presently based on differences in tier-cell numbers. is not supported by molecular phylogenetic analysis. Samples with pericentral cells that divide more than once (i.e. resulting in three or more tier cells) are not monophyletic. These formerly STicTosiphonia species are found in three clades, one containing S. TWIKatellsis. one containing S. kelanensis that possibly forms a sister group to Bostr",chia species with cladohaptera. and the third containing S. arbllscllla and S. inTricara. We propose that all species placed in SticfOsipllOnia by King & Puttock (l989) be returned to Bostrychia (Table 3). The only subfamilial division that is even moderately supported by our data is the informal 'section' first proposed by Post (1936, 1939) that consists of two groups: the Flagelliflllcrarae (for species with peripherohaptera) and the Ramifllleratae (for species with cladohaptera). There are several reasons why we think raising these sections to generic status is not useful. Firstly. the grouping of all species with cladohaptera into a single clade is not well supported (i.e. the position of S. kelanensis is still ambiguous). Moreover. the cladohapteron form in S. kelanensis is different from that of other cladohaptera-producing species (being on a nonspecialized branch vs a specialized determinate lateral). These two types of c1adohaptera are therefore possibly not homologous. Secondly. because the type species of the genus Bostlychia is B. scorpioide.l' (which has peripherohaptera), it would require that the c1adohaptera species. including B. radicans and B. l/1oriT:.:iana. be given a new generic name. Because both these species have been used extensively in molecular. biogeographie. physiological and life-cycle studies (e.g. Karsten et al. 1993. 1994: Zuccarello & West 1995. 1997.2003: West & Zuccarello 1999: Zuccarello eT al. 1999a). a generic name change that could be particularly vexing. if not actually confusing. to researchers. Thirdly. a generic name change to a monophyletic genus for species with cladohaptera will have to include the parasite BO.l'trrchiocolax allsTralis, a species completely lacking cladohaptera. The taxonomic conundrum of a red algal parasite being nested among several species with its host genus but lacking virtually all of the diagnostic characters of the genus (Goff eT al. 1996. 1997: Zuccarello eT 01. 2004) has not been specifically addressed by taxonomists to date. Because of their grossly different morphologies and modes of nutrition, the practice up to now has been to make parasites exemptions to the general goal of achieving monophyly for taxa at all levels of classification. Our sampling has also highlighted the nonmonophyly of several other species that have previously been studied genetically. BosTrYchia callipTera and B. pinnaw are a distinctive pair of species due to the uniquely shared character of regular alternate-distichous branching. This synapomorphy led to a biogeographic study of these two species by Zuccarello & West (2002) which demonstrated that actually three lineages existed within the species complex. these not being separable on the basis of any consistent presence or absence of cortication. They thus concluded that ecorticate B. pinnuw should be considered as a synonym of corticated B. callipTera. This work also suggested that the new re-circumscribed species B.


32

Phyco!ogia. Vol. 45 (I), 2006

Fig. 3. Boslrychia radical/s. I:bemOLOx,ylin- tained squash preparation showing the paired lier cells d rived from each of the six periccmral cells of one ;Lxi,; segment lframed within Ihe parallel horizontal lines). The distal of the lier-cell pair i1\ binucleate. the proximate i. uninu JelltC. Fig. 4. St/etosipllOnia k(l!clllell'.~is. DAPI-l1uorescencc of a squash prep,tnlli n. howing lhe Ihree lier cells derived frol11 each of the four pericenlral cells of one axis segment (framed wilhin the p,lrallel horizontal lines). The dislal of each series of three tier cells is binucleate. the proximalal twO cells are uninucleate.


ZlIeeare!lo & West: Molecular phylogeny of Bostryehia

33

peripherohapteron derived from vel1lml lier cell at a node opposite an indetemlinate lateral branch (arrowl. Fig. 10. 8o.nryc!lia simplicillsclI/o. Fig. II. 80strychia simptici/lsclI/o. An indeterminate axis showing partially monosiphonous lateral branche (arrowheads) and a ing.le peripherohapteron (arro' . Fig. U. 80 lI:vchia tenet/a. Apex of an indeterminale axis with altemat -di tkhou$ determin,lIe lateral branchc borne on successive axial cell (a teri k ). Ethanol-pl'lSmolysed sample. Fig. 13. 80 tryc!lia jlagetti/era. Apices of indeterminale axes with successi e altcrnale-dislichou laleral branche ..eparated b four unbranched axial. egments (asterisks). Elhanol-ptasmoly. ed sample.

ca!liptera was monophyletic. The outgroup species used were B. radicalls and B. morit~i(((l((, the particular choice of which lead to an apparent monophyletic ingroup. Our present data using further species sampling and partial 265 rRNA data suggest that there is no support for the monophyly of B. ca!liptera and that it could even be polyphyletic. with lineage-I samples appearing as a sister group to several species. RbeL seyuencing of the lineage-I and -2 samples was unsuccessful after multiple tries. but alternate rbcL primer design and seyuencing may resolve the issue of monophyly vs nonmonophyly of B. ca!liptera, A more convincing instance of nonmonophyly within what constitutes a single morpho-species is the B. simp/icillscll/a/ B. tellllissima complex. Firstly. we have demonstrated molec-

ularly that B. tellllissima should not be recognized as distinct from B. simp/ieillsell/a. in which case the latter name has priority. The presence or absence of monosiphonous laterals is not useful for designating species in this subfamily. as has already been demonstrated in the B. radicalls/B. morit~ialla complex (Zuccarello & West 1995. 2003; Zuccarello et at. 1999'1). The three molecularly indicated lineages correspond to the three haplotype lineages presented in Zuccarello et al. (1999c) and also have a consistently different biogeographic distribution. with the H I haplotype found in the southern part of southeast Australia and the H2 and H3 haplotypes only encountered further north. H3 being the northern-most. This biogeographic pattern is similar to that of B. simp/ieiuscu/a, in which samples in lineage H2 are found in southeastern

Fig. 5. SO.I·tn'chia lIIorit~i(///(/, Cladohaptera (arrows) arising from the basal axial cell of lateral branches in a male gametophyte releasing spermatia (arrowhead). Fig. 6. BO,l'trYchia I'i/Iltifera. C1adohaptera arising from the basal axial cell of a lateral branch of a heavily cortieated determinate axis. Fig. 7. Sticto,lil'//(lIIia "e/allellsis, The apical cell (arrow) and surrounding tier cells (asterisks) of a eladohapteron on an indeterminate axis. Fig. 8. Stictosil'hrlllia intricata, A peripherohapteron arising from tier cells and fanning out into an expanded multicellular attachment pad. Fig. 9. Stictosil'h(lIIia tallgatellsis. An early stage in the development of a peripherohapteron arising from ventral tier cells.


34

Phycologia, Vol. 45 (I), 2006

Table 2. ummary of molecular data and likcliho terlllined from a hierar hi 31 likelihood ration tc>!

J 16~

798 266 38 > 15.000

ligncd charactcrs Informative har:1Ctcrs teps in MP trees umber of M P trees Modcl •eject d amma . hapc par:1meter Proportion of in ariable ,itc,

440 2177 8 Tr

TR

0.51\. ~ 0._858

0.6569 0.4415

0.2

O. 5 I 0.0856 0.1962 0.3651

Base frcqucncies

C G T

0.1 41

0._671 0.2608

ubstitulion ralc matrix

1.0000 2.6578 1.0000 1.0000 4.6381 1.0000

R(a) lA- I R(b [A-Gl R( ) [A-T] R(d [C- 1 R(e) [ -TJ R(f) l -TJ

6.2371 5.724. 2.6188 4.5697 5 . 061 1.0000

Australia and those of lineage H3 so-far only known from Singapore. Although these three B. simplicillsculalB. lenuissima lineages (H I, H2, H3) are genetically quite distinct, there are no obvious morphological characters by which they can be distinguished. Even in controlled culture using several temperatures, irradiance regimes and water motion. no consistent morphological distinction was apparent. This may be due to the seeming lack of distinguishing characteristics for this species in general. lts morphological characters. apart from the presence of peripherohaptera, are mostly negative (an absence of synapomorphies, such as noncircinate apices and lack of regular branching and cortication). It is now generally accepted among taxonomists that a lack of morphological synapomorphies is strong prima~facie grounds for questioning the validity of any supposedly monophyletic groupings. It is likely for many taxa in the Ceramiales. and other orders, that detailed molecular analyses and increased geographic sampling will show that morphological innovation has not kept pace with genetic divergence leading to species formation. An interesting outcome of our analysis concerns the clade of B. scorpioides from Europe and the B. le/l.uissima haplotype

Table 3. (1989).

ullllllary of nomenclatural l:hangc,. Olher inforllllllion

011

H I (Fig. 2). Haplotype H I is found in southern Australia but not further north (Zuccarello el a/. 1999c), whereas our B. scorpioides sample is from northern Europe [although it is reported from temperate South Africa (Stegenga el al. 1997)J. The temperate distribution in opposite hemispheres of these two sister species raises the question of trans-tropical dispersal, a phenomenon also seen in several other seaweed species but not fully explained in any of them (van Oppen el al. 1994; Wieneke ef al. 1994). Our data show that B. len ella ssp . .ffagellij'era is genetically distinct from B. len ella and that the morphological distinction between them (branching frequency) is quite consistent. We propose that B. fen ella ssp. .f1agellij'era should be re-elevated to the species levels (as B ..ffagellij'era E. Post) as was originally conceived by Post (1936). One of our sequenced samples is from Botany Bay. New South Wales. an estuary just south from the type locality (Parramatta River, Jackson's Bay), where it is commonly found (personal observation). This species seems to be the sole corticate Boslrychia in southern and central New South Wales, although it co-occurs with B. lenella in more tropical environments. We have collected B. jiagellij'era from eastern and northern Australia, ew Caledonia (the sources of our molecular data). Thailand and Madagascar (data not shown), but have not examined it from other localities (South Africa, New Zealand and Japan) where it has been reported (King & Puttock 1989; Silva 1'1 al. 1996). Our data indicate that many of the morphological characters previously used in species descriptions are not useful in the subfamily Bostrychioideae. Cortication seems to have been a feature gained and lost in several lineages (e.g. within lineages of B. callipfera, the lineage grouping S. arbuscula and S. intricafa). Monosiphonous branches have also been gained and lost, and are variably present or absent in culture, in the B. radicanslB. morifziana complex and in B. simpliciusculalB. tenuissima. Although morphological stasis (maintenance of morphological characters over long evolutionary periods) has been evoked in the B. radicanslB. r110rilziana cladohapteronclade (Zuccarello & West 2(03), this is clearly not correct. Within this complex there arc two lineages that are clearly distinct, the heavily corticated B. pilulij'era and the nonphotosynthetic parasite Bosfrychiocolax ausfralis. Although many of the lineages appear to have retained their ancestral condition (ecorticate, variably branched axes) the,e other lineages clearly have morphologically diverged. In conclusion, our data do not support the maintenance of

typc.. typ

10

alitie. and oldcr s m\J1yms can b' found in King

Prcsent nal11es

Baslryehia rellella spp. jlagelli/era (E. Po.t) RJ. King & Pullock 8. rerll/i. sillla R.J. King & PUllock B. pimwra Ji. Tanaka 'Chihara lierosiphollia arbll.ellia (J.D. Ho ker & Harvey) RJ, King & Pullock S. il/frieafa Bor de aint- in cnt) P. . Sil a . kelallellsis (Grunow ex E. Post) RJ. King ,Pullock , IlIl1gafell.~is (E. Po t) R.J. King & PUllock . vaga (J.D. Ho ker & Harvcy) J.D. Ho ker Harvey , gracilis RJ. King & ?ullock

PUlLOCk

cw binol11i,lI

80,"II)'c!lia jlage/Ii/era E. PO'I Bos/ryehia silllplieil/scl/la Hal' C 80 /rycllia callip/em (Mol1lagn) lontagne Baslrycllia arbl/.,·clIla J.D. Hool-..cr 'Harve 80sll)'ellio ill/rica/{( (Bory de !lim- inccm Monwgne Bosrrycllia kelmlt!llsis Grunow ex E. Post 80 'rry Ilia ral/gntel/sis E. POM 80srn'l'l,;a \'{Ii/a (J.t . Hooker Harvey) Bo.,·rryellia gracili,,' (RJ. King & Punock) Zuccarello J. . clllllh. 1/01'.'

\"'C',

ricrosiplu)//oia gracilis. (RJ. King & Puuock. 1989. Ausrralitlll


Zuccarello & West: Molecular phylogeny of Bostrychia two genera within the subfamily Bostrychioideae based on tier-cell number. Our molecular data does support the separation of the subfamily into sections based on hapteron type as proposed by Post (1936), although strong support for this subgeneric separation needs continued research with other DNA markers, and developmental studies need to be undertaken to determine developmental homology in cladohaptera types. We therefore propose that all species of both Bostrychia and Stictosip!zonia now be referred to the genus Bostrychia. In addition, we advocate that B. tene/fa ssp. fiaf?ellifera should be re-elevated to independent species status as B. fiagell(fera E. Post. Although the Bostrychia complex of species has been well studied, many taxonomic problems still remain. The molecular indicated polyphyly of species such as B. simpliciuscufa and B. radicans/B. morit::.iana, although evolutionary interesting, is not an outcome easily encompassed by our present taxonomic/nomenclatural system. We recommend that in phylogenetic studies at the species level. multiple sampling within species, especially in the case of widely dispersed populations, should be incorporated in molecular analyses. Further research on this fascinating genus will probably increase its status as a model for showing the primacy of molecular data in the elucidation of cryptic and pseudo-cryptic red algal species.

ACKNOWLEDGEMENTS We would like to acknowledge all the collectors who have sent us samples through the years. Their assistance is greatly appreciated and has been indispensable in our research. This work was supported by the Australian Research Council and the Australian Biological Resources Survey. We thank Gerry Kraft for useful comments on this manuscript.

REFERENCES DE BERG R.F 1949. The New Zealand species of Bostn'chia related to Bostrl'chia scorpioides Mont. Farlowia 3: 499-502. FALKENBERG P 1901. Die Rhodomelaceen des Golfes von Neapel und der angrenzenden Meeres-Abschnitte. Fauna und Flora des Golfes von Neapel, Monographie (Berlin) 26: 754. FELSENSTEIN J. 1985. Confidence intervals on phylogenies: an approach using the bootstrap. El'Olution 39: 783-791. FRESHWATER D. & RL'ENESS J. 1994. Phylogenelic relalionships of some European Gelidiwn (Gelidiales, Rhodophyta) species, based on rbeL nucleotide sequence analysis. Phycologia 33: 187-194. GOFF L.J. & COLEMAN A.W. 1995. Fate of parasite and host organelle DNA during cellular-transformation of red algae by their parasites. Plant Cell 7: 1899-1911. GOFF L.J., MOON D.A., NYVALL P, STACHI' B .. MANGIN K. & ZI'CCARELLO G.c. 1996. The evolution of parasitism in the red algae molecular comparisons of adelphoparasites and their hosts. Journal of Ph."colog." 32: 297-312. GOFF L.J., ASHEN J. & MOON D. 1997. The evolution of parasites from their hosts - a case study in the parasitic red algae. El'Olution 5\: 1068-1078. HARPER J.T. & SAUNDERS G. W. 200 I. The appl ication of sequences of the ribosomal cistron to the systematics and classification of the florideophyte red algae (Florideophyceae, Rhodophyta). Cahiel'S de Biologie Marine 26: 25-38. Hml~IERsAND M.H. 1963. The morphology and classification of some

35

Ceramiaceae and Rhodomelaceae. Unil'ersit." of California Publications in Botam' 35: 165-366. HOOKER J.D. & HARVEY W.H. 1845. Algae antarcticae. London Journal of Botany 4: 249-276. 521-551. HOOKER J.D. & HARVEY W.H. 1847. Algae tasmanicae. London Journal of BOlan." 6: 397-417. HUELSENBECK J.P & RONQUIST F 2001. MrBayes: Bayesian inference of phylogenetic trees. Bioinformatics 17: 754-755. KARSTEN U .. WEST J.A. & GANESAN E.K. \993. Comparative physiological ecology of Bostrychia morit;:,iana (Ceramiales, Rhodophyta) from freshwater and marine habitats. Phl'cologia 26: 401-409. KARSTEN U .. WEST J.A., ZI'CCARELLO G. & KIRST G.O. 1994. Physiological ecotypes in the marine alga Bostrychia radiulII.I' (Ceramiales, Rhodophyta) from the east coast of the USA. Joul'/lal of Ph."cology 30: 174-182. KING R.J. & PurrocK C. 1989. Morphology and taxonomy of Bostrychia and StictosipllOnia (Rhodomelaceae/Rhodophyta). Australian Systematic Botan." 2: 1-73. KING R.J .. PLTTOCK C.F & VICKERY R.S. 1988. A taxonomic study on the Bostrvchia tenellel complex (Rhodomelaceae, Rhodophyta). Phycologia 27: 10-19. KU~IANO S. 2002. Freshwater red algal' of the ,w}rld. Biopress limited, Bristol. UK, 375 pp. LOCKHART P, STEEL M .. HE"DY M. & PE""Y D. 1994, Recovering evolutionary trees under a more realistic model of sequence evolution. Molecular Biolog." and El'olution II: 605-612. LUCAS A.H.S. & PERRIN F 1947. The seaweeds of South Australia. Part II. The red seaweeds. Government Printer. Adelaide, Australia. 458 pp. NAM K.W.. MAGGS C.A .. McIVOR L. & STA"HOPE M.J, 2000. Taxonomy and phylogeny of Osmlllidea (Rhodomelaceae, Rhodophyta) in Atlantic Europe. Journal of Ph."cologl· 36: 759-772. POSADA D. & CRANDALL K.A. 1998. ModelTest: testing the model of DNA substitution. Bioinformatics 14: 817-818. POSADA D. & CRANDALL K.A. 2001. Selecting the best-fit model of nucleotide substitution, S".,·tematic Biolog." 50: 580-60 I. POST E. 1936, Systematische und pflanzengeographische Notizen zur Bostrychia-Caloglossa-Assoziation. Rel'ue Algologie 9: 1-84. POST E. 1939. Weitere Daten zur Verbreitung des Bostrychietum III. Archi,' fiir Protistenkunde 93: 6-37. PRUI/HOMME VA" REINE W.E & SLUMAN H.J. 1980. Red algae found on European salt-marshes. I. Bostrvchia seOlpioides (Rhodomelaceae). Aquatic Baton." 9: 323-342. RAMBAI.T A. 1996. Se-AI: sequence alignment editor. Available at: http://evolve.zoo.ox,ac.uk/ (15 August 20(5). SILVA pc., BASSON PW. & MOE R.L. 1996. Catalogue of the benthic marine algae of the Indian Ocean. Unil'ersity of California Publications in Botan." 79: 1-1259. STEGE"GA H .. BOLTON J. & ANDERSON R. 1997. Seaweeds of the SOllth African West Coast. Contributions Bolus Herbarium. Cape Town. 655 pp. SWOFFORD D.L. 2002. PAUP* Phvlogenetic analysis using parsimonv ("'and other methods), version 4. Sinauer Associates, Sunderland. MA. USA. THOMPSO:" J.D .. GIBSO:" T.J .. PLEWNIAK F. JEANMOUGI:" F & HIGGll\s D.G. 1997. The Clustal X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Research 24: 4876-4882. VAN OPPEN M.J.H .. DIEKMANN 0 .. WIENCKE c.. STAM W.T. & OLSEN J.L. 1994. Tracking dispersal routes: phylogeography of the ArcticAntarctic disjunct seaweed Acrosiphonia areta (Chlorophyta). .10111'nal of Pln'('(}logv 30: 67-80. WEST J.A. & ZLCCAREI.LO G,c. 1999, Biogeography of sexual and asexual populations in Bostrychia m()rit~i{///(/ (Rhodomelaceae, Rhodophyta). Ph."cological Research 47: 115-123. WIENCKE C .. BARTSCH I.. BISCHOFF B .. PETERS A.F & BREB1AN A.M. 1994. Temperature requirements and biogeography of Antarctic, arctic and ampli-equatorial seaweeds. Botanica Marina 37: 247259. ZUCCARELLO G,c. & WEST J.A. 1994. Comparative development of


36

Phyc%gia, Vol. 45 (1), 2006

the red algal parasites Bostrychiocolax australis gen. et sp. nov. and Dawsoniocolax bostrychiae (Choreocolacaceae, Rhodophyta). Journal of Phycology 30: 137-146. ZUCCARELLO G.c. & WEST J.A. 1995. Hybridi/.ation studies in Bostrychia: I. B. radicans (Rhodomelaceae, Rhodophyta) from Pacific and Atlantic orth America. Phyc%gical Research 43: 233-240. ZUCCARELLO G.c. & WEST J.A. 1997. Hybridization studies in Bostrychia: 2. Correlation of crossing data amI plastid DNA sequence data within B. radicans and B. moritziana (Ceramiales, Rhodophyta). Phycologia 36: 293-304. ZUCCARELLO G.c. & WEST J.A. 2002. Phytogeography of the Bostrychia ca/lipteralB. pinnata complex (Rhodomelaceae, Rhodophyta) and divergence rates based on nuclear, mitochondrial and plastid DNA markers. Phycologia 41: 49-60. ZUC:CARELLO G.c. & WEST J.A. 2003. Multiple cryptic species: molecular diversity and reproductive isolation in the Bostrychia radicanslB-moritziana complex (RhodomeJaceae, Rhodophyta) with focus on North American isolates. Journal of Phycology 39: 948-959. ZUCCAREl.LO G.c., WEST J.A. & KING R.J. 1999a. Evolutionary divergence in the Bostrychia moritzianalB. radicans complex (Rho-

domelaceae, Rhodophyta): molecular and hybridization data. Phycologia 38: 234-244. ZUCCARELLO G.c., WEST J.A., KAMIYA M. & KING RJ. 1999b. A rapid method to score plastid haplotypes in red seaweeds and its use in determining parental inheritance of plastids in the red alga Bostrychia (Ceramiales). Hydrobiologia 401: 207-214. ZUCCARELl.O G.c., WEST J.A., KARSTE U. & KI G R.J. 1999c. Molecular relationships within Bostrychia tenuissima (Rhodomelaceae, Rhodophyta). Phycological Research 47: 81-85. ZUCCARELLO G.c., SANDERCOCK B. & WEST J.A. 2002. Diversity within red algal species: variation in worldwide samples of Spyridia jilamentosa (Ceramiaceae) and Murraye/la peric/ado'\" (Rhodomelaceae) using DNA markers and breeding studies. European Journal of Phycology 37: 403-417. ZUCCARELLO G.c., MOON D. & GOFF LJ. 2004. A phylogenetic study of parasitic genera placed in the family Choreocolacaceae (Rhodophyta). Journal '1' Plzycology 40: 937-945.

Received 25 January 2005; accepted 27 June 2005 Communicating editor: H. Kawai
