BIOGEOGRAPHIC AFFINITIES OF THE NORTH ATLANTIC DEEP-WATER SCLERACTINIA Stephen D. Cairns and Ralph E. Chapman National Museum of Natural History, Smithsonian Institution Washington, D.C., USA SUMMARY We investigated thebiogeographic properties and relationships of Scleractinia in the North Atlantic. The North Atlantic was divided into 3 8 geographic regions and each of the 134 North Atlantic species known from deeper than 200 m was scored for each region, resulting in a data matrix of 5092 cells. The species composition of each region was then analyzed by calculating a matrix of Bray- Curtis distance coefficients and performing a cluster analysis (UPGMA) and ordination {Non-Metric Multi-Dimensional Scaling) to determine biogeo- graphic afEnities and relationships of these localities. Three major clusters (superclusters) were distinguished: western, eastern and North Atlantic. The westem Atlantic supercluster consists of the tropical and warm temperate west- em Atlantic regions and is characterized by 58% endemic species and a rel- atively low percentage of deep-water species. It consists of three subordinate clusters, which represent insular, continental and warm temperate areas. The eastern Atlantic supercluster extends from the Faroe Islands to the Gulf of Guinea, and is characterized by 38% endemic species and a disproportion- ately high number of amphi-Atlantic and cosmopolitan species, as well as a higher than average number of deep-living species. It also consists of three subordinate clusters: a Lusitanian warm temperate group, tropical localities, and a third cluster consisting of the Cape Verde and Canary Islands. The North Atlantic supercluster is characterized not by endemic species, but rather by a high number of cosmopolitan and amphi-Atlantic species, as well as a dis- proportionately high number of deep-living, unattached, and flabellid species. Two subordinate clusters correspond to the cold temperate northeast and northwest Atlantic. INTRODUCTION There is a deep-rooted desire among people to organize data into systems of classification (e.g., clusters). This probably originates in the need to organ- ize multi-dimensional data into smaller packets that can be visualized and manip- ulated by researchers, but also probably reflects a degree of real clustering in natural systems. Within natural history, if species are the objects of classifi- cation, then cladistics or phenetics may provide the algorithms used, but if zoogeography is the subject under consideration, then geographic regions are usually the units of classification and various clustering and ordination algo- rithms are the tools typically used to synthesize these data. According to Clarke and Warwick (1994) there are literally hundreds of clustering methods, indeed 30 DEEP-SEA CORALS whole volumes (e.g., Hartigan, 1975) dedicated just to clustering algorithms. Cormack (1971) warned that the "availability of... classification techniques has led to the waste of more valuable scientific time than any other 'statisti- cal' innovation." Nonetheless, a judicious use of clustering and ordination tech- niques can lead to thought-provoking and valuable Insights about the distribution of organisms and it is becoming increasingly common for authors of bio- geographic analyses to include a clustering dendrogram and/or an ordination to explore their data set. Several examples using cnidarian taxa are presented below, in increasing order of sophistication, each with a slightly different method and purpose for the analyses, but all using the Bray-Curtis or Jaccard simi- larity coefficients and various clustering (typically UPGMA; Sneath and Sokal, 1973) and ordination algorithms (usually non-metric multi-dimen- sional scaling, MDS; Clarke and Warwick, 1994), For instance, Veron (1995) used clustering techniques in a simple way to document patterns of geographic affinity of shallow-water Scleractinia from adjacent geographic regions. Similarly, Wallace (1999) used clustering meth- ods to demonstrate geographic afifmities among acroporid Scleractinia, but also analyzed the composition of her clusters based on the overall distribu- tion of the component species. Ketchum and Reyes-Bonilla ( 1997) and Reyes- Bonilla and L?pez-P?rez (1998), based on shallow-water Scleractinia, clustered the regions of the eastern Pacific followed by an analysis of these patterns for evidence of the presumed direction of colonization, and to test for the valid- ity of Zoogeographie barriers. In an elegant pair of papers based on shallow- water corals, Sheppard (1997, 1999) used clustering and ordination (MDS) to find the geographic affinities of various localities in the Indian Ocean and then used these results to assess the role of the Chagos Archipelago as a pos- sible stepping stone for dispersal. Clustering and ordination can also be used for studying finer-scale ecological processes, such as assessing the relation- ship between a circumscribed ecological zone of occurrence versus various physical factors, such as light and substrate, as was done by S?nchez et al. (1998) and S?nchez (1999) for shallow-water Caribbean Octocorallia on selected reefs. The current study seeks to use clustering and ordination analyses to address many of the parameters described above. First, we seek to establish how many species of Scleractinia occur in the North Atlantic below 200 m, where they occiu", and at what depths. Using these data, we then try to docu- ment the relationships among the faunas of various regions, in both gross and fine detail, based on the shared presence of these taxa. We attempt to char- acterize these groupings based on physical characteristics of the environment and the morphological characteristics of the species, in an effort to understand why these various localities cluster in the way they do. In so doing we com- pare otir clusters to the traditional shallow-water Zoogeographie provinces. We also compare the results of traditional clustering with the results of the ordination (MDS) in an effort to better understand the clusters. Finally, we ask if those transitional localities that do not fit nicely within any single clus- STEPHEN D. CAIRNS AND RALPH E. CHAPMAN 3T to -S P o 3 B-JB' o -Go -fa '5 " TJ T3 ?B .3 I ^ o B 13 I O .a S? ^ I s 3 - ^ y 'S o M 2? S o B 3 -4-? ??-? RI i 5 Z < 5 S. a CA G i 1 1^ S o g.2 sa ?-a s.2 .3 3 s.s 13 ^ ^ 3 u o ^ S s- .boss 'S -6 -2 T3 .g g >, ? a^ .S u O .3 i 2 S JB .2 N tg ?^ 'S o o ? u 00 5> Si B JO <: 4-1 ?3 ? 8=5 se ???S i s 8 5> -r; (U ? ? >; 00 g ?K ^ -S a 8 Q la- s s: c .8 ?1 "3 e ji ? cj u u u u Q u U CJ u u u u O ? O il u i U \ %.. \ f.^ ?* r4 o r- p?. 1^ t^ i ? i i o M ^ OS fN ^ ri 4 ^ os 1^ o ? ? ? ? ? ?o rt -; IN ? .. " '. 1-^ 1^ ??* r--? tr\ u^ f^\ t^ t^ T?- ^ l? o\ ?N S O S = ri ?? ?? s -r 7 " m M ?" o s tn ": ?, -^ ?rt ^ .-j i & ? i S P ^ t ? ? i i i 2 -S g :? - I I ?fl " I ? '3 U 3 " m so 00 * I^ oo t^ ^ =o ? " 1 i I ? u I I 1^ I -I is" Oft ti, I -3 E i I- I ? E i I ?5 5 (3\ O ?t fS 2 i ? ?i Q Z I .s I % M I o -a o o ?S '? ^ -Ti 0) u .2 u C ^ -^ c o ?? I 1^ 2 II ?3 Ig i ? 3 -a s ?? u ,A I 'S B a ?o ?3 ?2 'G " -o o ? 2 ? ? 5 ?3 I*' < " u ??s 13 ?? ? s ?a 5 2 " ff T3 S < -9 9 .a "2 -a g ^ E -S .a 3 -s -s ?f2 I " -S VI -S g ? " s < .2 ? 3 a .D ^ 'S .2 g E -2, 3 :? u J 00 2 ?= en 60 ? i- S g- o G S < ^ Ou ^ J3 ?^ &3 Q ^. u (J ? ' ' u Q a Q Q Q Q Q D Q Q Q U U fj U U (J ?J U ? U O U U u U tb d. b. b. < < < < < -< <; <; < < < < < < < Ik MH < -< < < -< < < < <: < < o M 1/1 m U w CA m tiO 1? m m m M m 00 w M M m M w m W M M m ? m Cfl M M ?T3 u 1 _S_ s 4> i ? _&_ _S^ 1 _&^ ? U _i_ ? ? u u o i i q> i i U S u u i i 1 CO v^ ?t Os Os '2 - - i i r- ro ^ r*> *f t^ o t^ w "ij rt\ ^^ t-3 f-3 i~S -?"? ?N ?N (S o ^ W? ?O '.I ? Os T vs 5 oo g v> f^ ^ o (N o ^ u u > ? ^ ? .3 ,a ,a a a a a 8 a a a n 4- 8 5 Uj Q ?? ? o s 3^ a -?. t<^ O O ?S ?o c5\ m fs \0 o g ?! -* J ^4 ?* ^_. 1-J __ ?.1 ?1-1 S ? S S S S s ? w \>i ?*' ^; "? c: ^ s Pi 2 vi ? T^ *- s i> s ? o o o (M 8 S VI os CTi ^D ? ?* r- r^ ^ tn ii% o f*^ o o IN 00 <0 ? ? ^ ^ ^ oo ?^ t^ Q? 5^ ^ ? ?? '^ go , ^^ o i^ t^ ^ '?j- ?N n * ?? ? ? ? ? S ? ^ ?? S T? 2: Q 5 I \ ?F- a a- ?2 ? % ^ %^ \ UH tl. b u. ?H tJL, b< ? ? ? ? ? ? ? ' b. U. O u ? O Q Q u ? ? ? ' ? ? U U U ?L, b. (b M:< eu tu u. bL. 0. u. u. b <; < < < < < < IL. < < < < < < < < < < < < < U ? ? ? (A ? M M m w m m w ? m M M M w m M U] ? O O u u O O U u D? w" i i S 4> ? i s 1 u o ? ? u ? o O 1 S ? u u u u ? ? s y V ^ u S^ 4> 1 E= 8 " Os g r^ . es \o 1^ J^ ^ ?n S Tl .:j' o iC "^ 1-^ rr; 0\ ^^ ? S O ? *0 e^ 1?h ??p s ? 2 " s 2 2 = 8 2^a n ~ es 3 2 8^ ? M ^ ^ ^ ? ^ tl ^ oo oo ?-? tS ^ Os r^ o? vo ^ n f^ S S H " ? ^ 2 ;? -r ? T "f ? A ? s y? >" ^ ? ? ? ? ? m se M ? si ^ s - 7 s ~ s ? " (S 2 '*> -* fi, ?f f? ? s et r^ _, i^^ S S 2 ^ A s ?2^ ?/i tn ? ? ? o Ko Cj i-i "" S 'f T S *? S \o ;:; ?v ^ o IN ~- 4 "C? -Q -o a a a O t. It, b, ft. ? g. ?S -3 3 N a SI ? ? " 3 a. T? S ? I ^Js I 3 ? S ? J5 -??; y "Q ? ?? .^ ? ? ? la, o. ^ a ?? ??! & & g- ?J -^ --5 3 o 1^ *ri \o r^ ?^ r^ r-- m ? ^ m oo OCr OO 00 OO 00 ?O 8 5 ?padd?w jou uoiSsu jtfpoj -opuj its^ipuD 9if1 m pDjoos SU01S3J g? 3tji/0 tpjOf^/o dvui z I ? ?t Q Z 5 .5 fc E 'S o,* =" S s g M ?a ?. ? "AS ?? S l? i '^ 5 ? UM * '- * 'S ? S S S o (Q ^1 'S ^ S 5 f "^ "0 s? S ^ a. __^ .2 2 C -3 O s o. S in o " 1. ^ ^ ^ ,S o k. ?^ R b o ^ CQ O w m w 5 O < <: fc D 6 w J J X o ? -< 2; Pi p I-? j w pa H ; m E o PO u u H u Zii? to tn O 5 8 Q m^mms^mmi^^mm&m:-mm^m^^m^^0m^mm^mmmi^m^mtm 0 0 1 0 0 0 0 0 01 0 0 O?IO 0 0 ? 0 0 0 0 0 0 0 0 1 10 1A3 Lt m??m3m^m'i^Bmmm^^mm'^f^mm^s^mm^mM?mm'm:?mm^mmm?i^^9mm?mmms^mimmm^m^ HKmiSS oil 10100100 01 I lOOlOlOOOO 0 0 0 0 10 aVH S? mm^ffmf^mmsmmmmk&M-vmmm^mMm^rM^^^mmmmm^m?mmm^^mmmmmi^a^s^m 9^^snB^t$ 00010100000 OIOIOOIOIOOOO 0 0 0 0 0 0 VKS te a^i??K??i?A?M!|^:??ami??^:W?H??BKA??aK?l?^;-?^aft-W^^^ OOIIOIOQOOO 0010000000000 0 10 110 OO I? w^iira?)i^ii???raE?i^isi?05i3S?^ass^s^5?'^'mM?s???*MA^ssiiom?te^^ ?IOIOOOOI 10 001 lOOOOIOOOO 0 1 10 10 SAUV 61 mm^^m^&siimmm^sm^m.';m^Mm^Emm-^^?m^m^i!^i!m^m?mmEmsmmmm^m?mim^^m^sm^^ O?OdOOOOl fO OOOOOOOOOOOOO 0 110 11 oaw Li ai^[^??ii^i!?i^?^i?^??i^^?ai???f??imai'n B^w*r?HM<''^A^jca?rii'?j?iiijriu^r^ijiiu?jitjrjjffi'im"rtM7iri^jir.ipj?iiLLi\'"ii.'t]i)ini:Ti2r.iL"rn^ ?iTTi?jjj"T!'zanrniii'i'j^^r'^.!:^ir.':t;^"n''Ti^i'rf?1r1?!!?rt'SlKi?Cf?ilJfft^ 0 0 0 0 10 as ?.1 00 0 00100000 000O1OOO1OOO1OOOOOO A?ID SI mmrnm^^m^smmm^m^m^^mmm^^m^^Mm^mms^^mammmmfmsmmm?mmmm?sm?mmm 00010000000 0001 lOOOOOOl 1 0 0 0 0 10 DfAS El m^^?msm^em^^^i^Mm?S^^x^?mmmm^^^s^mwmmm^m^^^^m^?^mma sEffimnamsHm^ira fflH^Hmi OOOOOOIOOOO OOIOIOOIIOOI 1 0 0 0 0 0 0 AVI 11 OOOIOIOIOOI OOllIOOOOOOlO I 0 0 0 0 0 wvr 6 'lOOtlllllOI IllllIlOOOOil 1 0 0 1 0 0 ana L OOOOOIOOOOI OOIIIOIOOOOOI 10 0 10 0 H9AV s ^m^m^^iK^^mmi?mm?mmm?^mm^mm^^^^^m^^^^mmm?msm m?omm 00001010100 I1111011I1I00 0 0 0 1 10 VTi ? m^^m^s^?^m^?mm^^^m?m^smmmmms^^m^^^^m^mK:wmm^m HaMlRH?HwlwHeHffl 00010000100 OlIllOOOOOOlO 0 0 0 0 0 0 Baa t 09 es 8S iS 9S SS K CS IS IS OS 6* 8fr LP 9fr Sfr frfr Cfr Ifr Ifr 0fr 6e 8C it 9t St frt ft rt 1? ?nil/3[?jiFi Z ? ? l Q 2: 3 z UJ ^mmi^&im^osm^^^^mw&-m^Mi^mms:^^. Oe 61 ?l Ll 91 visX/?l?o^ 8 Q "1 OOOOOOOOOO 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 lAD L? f^m^^^^s^^^^s^m^mmmmfm^m^^^mm^&i^mmm^i^mmmmi^^&mss^pm? ^^^smm^ 1000000000 0 l 1 0 0 1 0 I 110 0 0 ! 0 0 0 0 0 avw se 7:mm^m^^mmm^^s^m^^mmmm^mmm9^^m^tm^?^^mmm^mem:sm'mmm?mm iW14]raS^ OOOOOOOOOO 0 ? 0 0 0 0 0 I 0 0 0 0 0 0 0 0 0 0 0 VMS ?t ?i<:iiPSiaJM??^i?i^^ffl?^ii^rawH^??io^i^i;wsi?j????MH^s;l*Ml (psnutiuoo) xu^BUi BjBp aqx '? 'II^^X V) iW OOtOOIOOOO 11O000OO OIIOOIOIIO 1 100001 I ?^1 0 wm 0 mm 0 0 0 Km? 0 ilH wm wm 1 HI I mm 0 iMI 1 mm 0 wm. 0 wm~ wn um WS???^BiHt 0 0 0 m 0 CaSMH I BHi 0 nrrnai 0 ma 0 0 ism 0 0 g? 1 IA3 ?? QVM 5E 0 0 ?S?B 0 1 I 0 I 0 I Hufs 0 0 0 MM 0 0 ms 0 Km 0 me 0 0 0 0 0 0 0 0 0 mi 0 0 ji|T|fl: 0 0 KM I lira VMS K 0 0 0 nfoa 0 0 ISHHl 0 BUEU 1 0 rifff?f 0 ggauH 0 0 ?KBiBi 0 Iffll 0 0 ? 0 SMI 0 0 0 0 IM! 0 0 1 BBfflB 0 0 0 0 0 0 HH? OD It 0 0 0 IHHBI I 0 EE1B3H 0' i 0 0 0 0 tfljjlUjin 0 m^?^ 0 0 0 0 1 0 W?MA 0 0 0 0 1 I 0 0 K?tttU 0 0 1 SMH 61 SB 0 ?Kl 0 0 KSK I BKH .0 0 1 0 0 1 0 0 . 0 uaaga 0 0 0 1 1 0 0 WBMKt 0 MHUn 0 ?all 0 aagm 0 0 mgem 0 0 0 0 qaH a. 0 0 GHSf 0 U?fBE 0 0 wem 1 iHra?i 0 0 NHH 0 ?nBHI 0 tmam 0 a^gB 0 0 0 0 1 0 i^ 0 ?fiW 0 0 0 ?um 0 p 1 0 0 3S!M ? r 0 0 VJS S? 0 0 0 0 0 0 I 0 0 0 0 0 0 0 0 0 0 0 0 ?> 0 w 0 0 1 0 0 ^^'air? l?EBS 0 0 0 0 0 0 S?sW 1 0 0 0 0 0 I^SB mem 0 0 wsaL 0 ? 0 0 0 0 ?SUB? 0 0 Ma?M 1 0 WSSk 0 1 MHR 0 0 VDS ir 0 0 0 0 0 0 0 0 0 0 0 0 lK??KKt Q 0 o' SjjSXl?K ? IHM 0 0 I 0 i? 1 I IMW 0 mA 1 ?a wm. 1 1 0 0 0 0 0 0 KftlHIlt! 0 0 0 W?etSA 0 0 DJO ?1 WKSk 0 0 OOOOOOIOOO r 0 I 0 I t 0 0 0 I I 0 I 0 I 0 1 0 I I 0 0 0 1 0 0 I 0 0 I 0 0 0 1 0 1 0 1 0 1 0 0 1 0 0 0 0 ?mm 0 0 0 0 0 0 0 ms? 0 0 wm 0 0 mm 0 0 0 0 0 0 0 0 mm 0 0 wm 0 0 tm 0 dJS ?I AIID SI Mi 0 0 mm 0 0 wm 1 0 ! 0 0 0 !^?^^?..?..T?~?.>? 0 0 0 0 0 0 ttsoEb H?ls 0 I Hill 0 0 0 DMS n 0 0 0 I 0 l 1 0 0 0 1 0 I 0 I 0 0 l 1 0 I 1 0 0 0 I 0 1 0 1 0 0 ! 0 0 msmm 0 1 0 0 0 0 1 0 0 0 lOOOOOOOIOOO 11I10000I?O? 1 lOIOlODIOOO i 00000000010 OOIIIOOIIOIO 100100000000 ?V7 II wvr 6 RMmSK MOM s vu c Hafl I 06 6S ?g 98 ?8 M ?S tS I? 08 6i 8?. U 9?. Si. fi ?i ?i U a? 69 89 ?9 99 S9 H ?9 19 19 n?JL/?|??l 5 8 Q :2 S-. |3 73 B-^ SPS-^ - S u (U ?a g-Si 'S S s -s T3 s s ? s-g g> => -g. ^ -q .? ^ ^ p-f^ B 3' ? 5 S -3 5 I -? t? ^ o "5 ? ? ;- 03 S ^ S <^ ?^ ^ -^ -? ?? 'S ? -^ .> (U .a -g o ,ti ^ -g g ?" 2 'o i,^ t? "HS 3 aj s? ?? ?1 .o o n ?* 8 2 g '5 ? SP-? ia w! ^ a S 3 ?! m ?.s ^ fi 1 ^ ? O " -is ??0 ?i. S? ^J3 5 5'-' Ile,-s "3 3 o ^ !/? SWOI - [ 1 ?] UGSUBXiaiipS)^ SI/6 U/L es/?? Of/IE 9?/91 ZilZZ 6?/8! 91V 6t/8I ?eXfr)rz - [Z9]VinJ!5?sniD !I/E 9I/? ES/OI ZVI% 9I/C 8?/lI ?W6 9I/? 0/0 ??/(l)? - teiJpuBpailS 61/CI 0I/? e?/9e 9?/8? iVL\ ?WE? 9E/8I 9/f 6ZI0Z 8?X9e)9r ~ [89] opueuv 0/0 6/1 w/? iW? LZIL ?W? ??/9 81/Z 0/0 LZ/Z - flljail-isisno 6/? 6/e SWOl ?W01 LZI9 9E/S OOl/CC - - - - [JZ) otjtoBd-opui Zl/C 6Z/? ES/6 iWS Sll^ \UZ\ ?WS CZ/fr 9/1 - t-z/COt- [?l]VII"?sniD 6/? 91/P ?S/EI iWOI um ??/91 Sfr/OI Sl/f ?/I t'l/E 81/(1)1' {ZZ] apaopv ?/l ?/T 89/?I t9/CI zm 9VS IE/9 . ?/I 9C/? ?- ??/? [6l]Bpmmsa ?l/OI ?I/? 8fr/6i KWoZ ZE/El ZVL 0?/Z? i/fr ?E/?l - 8t'/Ct')6E [09] VI JaisniD oc/e i/t' ?s/ze ??/frE 0E/8I UV ?W?l ?/I ?i/?l ' 81-/(1)63 [09] (..?"09 {?)|]3pmrl,Y ?I/EE 6/z:i 95/S? fr9/98 re/o? ??/0? ?l/?C E/t- ?1/0Z 61/93 9^19 [KI]3|HreilVN a j D paq?vnv I?l"?i?3 m 0001 49A0 ?03 3BB POB 3 JUi sj?|sii|3/sj?|sn]3j>dns ? a ty ?c Si's Hi jij} 3ij} Jo dvyj f ajnSij 'A 8 Q z I 5 Q Z es 5 8 5 Lu n a c a D. I-H ??? u ni i5 S'a-S (S j= "^ o B) -3 O T3 m y a ^ ti 1? o s -s ? c ?I s a PS g .S o ? t <: -a ? s b- ^ 14 !;^-5 5 5 ^ ^ .a-S." ?B S ? es SJ T3 -p j O a f? e S 3 " a a i? ? 3 t^ '" Eo M u ? O ?S S u B 3 s= .a C '-S -" ? W ,nt s. -S a ) o H s &" ? 6i . 3 F? *-! ? ? g| ? S ) ?? *^ W ,. ? a<=> -S 3 ^ I > S ^ 1 ? ;TJ^^ 8 ? ^ EO -^ ^rf S s o ^ Ja'S < -s 1 f -? t ci o " o II ' K ^ t> s S o g.P w K ? o g ? o 2 e S g ? g ? ? S -c a S3 ? S < -0 T c ' S ?- <; R*.2 ? >> >^ S) 3 JS 1-? ? ?f-: fc ? Q s i^ -S .a I ?a !-?? ?S W '"' ^ 'C - o m,o I -S ? 11 ? ^ ? .2 '5 ^ ^ W g I p o 2 ?3 (u b o 2 o ? 'S tr " bT m FM W "^ J^ -Il JJ s? ?a "J Q '? ? E? ? U ? 8 tu Q I ? Q lu I "cS '-S ? ^.fa 0 8-1 o 5 ? S 1^ a s th cie s. Q J3 o Q, 3 ? ?1 tw u O U i^ U U O "fl CO C V P t, a (4 O W)5 ?3>l in z I Q Z O? 5 I w _J _ .00 6 'C ai '3) .3 s.? J3 " ?D-" =3 o 3 o? ? ??I 27? " O ils a K ? Il .2 0 a <^ b S fi '^ O?l "g S S ? ? s Si a o ?s ? C ?l o 1 o 5 U ^gggs; ?2?? ? o u o o tri U o u ?5 o o n t2 ^* ?-* 'ri u '^ ti ?" S es JW own 'S ? ? o ko ? (M O ts (S ?* ?g ??^^ 2 o U & u Si) ^ I ?^ ?? o ^ .. nt X a ^ S c es U g .g 'C CL,