On the identity of Lottia strigatella (Carpenter, 1864) (Patellogastropoda : Lottiidae)

19 Pages·2003·10.2 MB·English

by  W B Simison

—
THE VELIGER
© CMS, Inc.. 2003
The Veliger46(1);1-19 (January 7. 2003)
On
the Identity of Lottia strigatella (Carpenter, 1864)
(Patellogastropoda: Lottiidae)
W. BRIAN SIMISON' and DAVID R. LINDBERG
Department of Integrative Biology and Museum of Paleontology, University of California,
Berkeley, California 94720-4780, USA
Abstract. Nomenclatural confusion has surrounded the northeastern Pacific lottiid currently referred to by the specific
names strigatella orparadigitalis for 135 years. Much of this confusion has resulted because of the supposed range of
this nominal taxon (GulfofCalifornia to the GulfofAlaska), its morphological variation within this range, and its overt
similarity to several earlier named taxa. Here we examine the relatedness and distribution of these taxa from localities
between Guaymas, Mexico, and the Aleutian Islands, Alaska. Relatedness is established by a maximum parsimony
analysis of partial 16S mtDNA genes and distance analyses of cytochrome c oxidase I and 16S. The results of these
analyses provide unequivocal evidence of the distinctness of Lottia strigatella (Carpenter, 1864), Lottia paradigitalis
(Fritchman, 1960), and the presence of a third previously unrecognized taxon, Lottia argrantesta Simison & Lindberg,
sp. nov. The taxa L. strigatella, L. paradigitalis, and L. argrantesta are not members of a species complex, but rather
members of three distinct subclades within the northeastern Pacific Lottiidae. Additionally, molecular data from Lottia
borealis (Lindberg, 1982) revealedthat this Alaskan taxon shouldbe synonymized with L. paradigitalis. Lottiastrigatella
and Lottia paradigitalis show characteristic Californian distributions with apparent range end points in the vicinity of
Point Conception, California. These data and the evolutionary history they reveal provide a compelling demonstration
of the levels of morphological variation present in the Patellogastropoda.
INTRODUCTION Sur, Mexico. In a second paper (Carpenter, 1864b) this
specific name was erroneously spelled strigillata. Car-
Nomenclatural confusion has surrounded the northeastern penter (1866:334) proposed Acmaea patina Van b. stri-
Pacific lottiid currently referred to by the specific name gillata for a second nominal taxon from the Vancouver-
strigatella or paradigitalis for 135 years. Much of this Californian provinces. He compared itto small specimens
confusion has resulted because of the apparent extensive
oiLottia pelta (Rathke, 1833), and remarked on the dif-
range of this nominal taxon (Gulf of California to the
Gulf of Alaska), its morphological variation within this ficulty in distinguishing it from "the A. strigatella of
Cape St. Lucas." Burch (1946) erroneously referred to
rkannogew,n atanxda.itUsndoevresrttansdimiinlgartihtye etxotesnetveorfalitsotdhiesrtriebaurtliioenr the northern species asAcmaeapersona strigillata, noting
in the northern portion of this range was further compli- ptheerssoinmail(arRiattyhkbee,tw18e3e3n).itSmaintdhs&malGlorsdpoenci(m1e9n4s8)oafnLdotAtbi-a
cated by the presence in Alaska of the morphologically
bott (1974) followed Burch. Grant (1933) placed A. per-
similar Lottia borealis Lindberg, 1982.
The tortured nomenclatural history began with the pro- sona strigillata in synonymy with Lottia digitalis (Rath-
posal of two similar specific names for a single nominal ke, 1833), butillustrated specimens ofBurch'sA.persona
taxon strigillata for the California population and stri- strigillata as "Acmaea persona." Four years later. Grant
gatella for the GulfofCalifornia population by P. P. Car- (1937) illustrated the same shells as supposed hybrids be-
penter in the 1860s. Palmer (1958) and McLean (1966) tween L. digitalis and L. pelta, but the name A. persona
gave detailed discussions ofthe subsequent nomenclatur- strigillata remained in synonymy with L. digitalis. It is
al confusion. interesting to note that Grant, who originally suggested
In summary, Carpenter (1864a) proposedAc/naea stri- thatthis taxon was ahybrid, neverdiscussedthis decision
gatella foralimpet from Cabo San Lucas, BajaCalifornia in any ofher texts. The hybrid designation only appeared
in figure captions without further comment (see also
Light, 1941; Smith et al., 1954).
' Corresponding author: W. Brian Simison, Museum of Pale- The name Acmaea paradigitalis was proposed by
ontology, University of California, Berkeley, California 94720;
Ph: 510-642-3926; Fax: [phone]; e-mail: simey@socrates. Fritchman (1960) after a study of the radular basal plate
Berkeley.Edu morphology of L. digitalis, L. pelta, and the supposed
Page 2 The Veliger, Vol. 46, No. 1
Lottia paradigltalls
Lottia strigatella
Lottia argrantesta
LFoitgtuiraea1r.graSnkteetsctha mSiampisoofna&secLtiinodnbeorfgt.emspp.ernaotve.NroesrptehctAimveertiycpaeslhocoawliitnigeseoxfpencotmeidnadilsttraixbaut(isoynmsbooflsLotinticaoplauramdnig5i.taTlaibsl.eLo1t).tiaandstrliogcaatleiltliae.s
of molecular samples (reference numbers in column 5, Table 1). Expected distributions of haplotype groups are based on associated
shell morphologies.
hybrid. McLean (1966) synonymized L. paradigitalis character analysis of morphological characters as well as
with the Panamic species L. strigatella based on the sim- ecological studies have provided conflicting answers to
ilar shell characters of the two taxa. The similarities had the distinctness and distributions of these nominal spe-
been noticed first by Carpenter (1866) but were subse- cies. Clearly, a new data set is needed to address these
quently ignored by most workers. McLean's treatment questions.
was followed by later workers including Seapy & Hoppe Here we examine the phylogeny and distribution of the
(1973), Carlton & Roth (1975), Christiaens (1975), and lottiid taxa formerly known as strigatella. paradigitalis, and
Morris et al. (1980). This nomenclature remained rela- borealis from localities between Guaymas, Mexico, Bodega
tively stable until Lindberg (1981:75) revived the use of Bay, California, andthe AleutianIslands, Alaska. Phylogeny
the specific name paradigitalis for northern California wasestablishedby maximum parsimony analysisofapartial
specimens of L. strigatella based on radular differences sequence ofthe mitochondrial largeribosomal subunit(16S)
that distinguished the northern and southern California (Simison, 2000). After delimiting these taxa with molecular
taxa from one another. characters, shell and radular characters were examined to
The advent of molecular techniques provides new data determine the range ofmorphological variation within each
to examine levels ofrelatedness and to determine the dis- taxon. These morphological characters were then used to
tributions of populations and species-rank taxa. The stri- identify and delimit the regional occurrences ofthe taxaand
gatellalparadigitalis question is an ideal problem forsuch associate existing type specimens with specimens from
study. The debate has been ongoing for 135 years, and known haplotype groups.
W. B. Simison & D. R. Lindberg, 2003 Page 3
Table 1
Specimens and localities examined in the course ofthis study. Symbols and numbers referto type and additional sampling
localities, respectively. Shell and radula numbers refer to illustrated specimens and checkmarks to recovered molecular
sequences. Genbank accession numbers: L. argrantesta COI = AF295537, 16S = AF295540. L. paradif>italis COI =
AF295538. 168 = AF295541. L. strigatella COI = AF295539, 16S = AF295542.
Tables Figure Rad-
Specimen no. Taxon Locality 2&3 1 Shell ula COI 16S
UCMP No. 57003 Lottia argrantesta Califin, La Paz, BCS, Mexico 1 1 17 23 y y
UCMP No. 57008 Lottia argrantesta Bahi'a de San Francisquito, BCS, Mexico 2 A 18 24 y y
UCMP No. 57005 Lottia argrantesta Tecolate, La Paz, BCS, Mexico 3 1 y y
UCMP No. 57006 Lottia argrantesta Tecolate, La Paz, BCS, Mexico 4 1 19 y y
UCMP No. 57007 Lottia argrantesta Bahi'a de San Francisquito, BCS, Mexico 5 A 16 25 y y
UCMP No. 57036 Lottiaparadigitalis Attn, Aleutian Is., Alaska, U.S.A. 6 y
UCMP No. 57037 Lottia paradigitalis Attu, Aleutian Is., Alaska, U.S.A. 7 y
UCMP No. 57038 Lottiaparadigitalis Attn, Aleutian Is., Alaska, U.S.A. 8 y
UCMP No. 57039 Lottiaparadigitalis Attu, Aleutian Is., Alaska, U.S.A. 9 y
UCMP No. 57040 Lottia paradigitalis Attu, Aleutian Is., Alaska, U.S.A. 10 y
UCMP No. 57041 Lottia paradigitalis Attu, Aleutian Is., Alaska, U.S.A. 11 y
UCMP No. 57042 Lottiaparadigitalis Attu. Aleutian Is., Alaska, U.S.A. 12 y
UCMP No. 57043 Lottiaparadigitalis Amchitka, Aleutian Is., Alaska, U.S.A. 13 5 y
UCMP No. 57044 Lottiaparadigitalis Adak, Aleutian Is., Alaska, U.S.A. 14 6 y
UCMP No. 57045 Lottiaparadigitalis Adak, Aleutian Is., Alaska, U.S.A. 15 6 y
UCMP No. 57046 Lottiaparadigitalis Adak, Aleutian Is., Alaska, U.S.A. 16 6 y
UCMP No. 57047 Lottiaparadigitalis San Francisco Bay, CA, U.S.A. 17 D y
UCMP No. 57020 Lottiaparadigitalis San Francisco Bay, CA, U.S.A. 18 D 6 26 y y
UCMP No. 57023 Lottia paradigitalis Bodega Bay. CA. U.S.A. 19 4 y
UCMP No. 57019 Lottiaparadigitalis San Francisco Bay, CA, U.S.A. 20 D 8 28 y y
UCMP No. 57018 Lottiaparadigitalis San Francisco Bay, CA, U.S.A. 21 D 7 27 y y
UCMP No. 57021 Lottiaparadigitalis Bodega Bay, CA, U.S.A. 22 4 5 y
UCMP No. 57022 Lottiaparadigitalis Bodega Bay, CA, U.S.A. 23 4 3 y
UCMP No. 57001 Lottia strigatella Guaymas. Sonora, Mexico 24 O 12 y y
UCMP No. 57009 Lottia strigatella Sta Maria, Cabo San Lucas, BCS, Mexico 25 2 22 y
UCMP No. 57002 Lottia strigatella Guaymas, Sonora, Mexico 26 O 15 y y
UCMP No. 57010 Lottia strigatella Cabo San Lucas, BCS, Mexico 27 2 21 y
UCMP No. 57014 Lottia strigatella Chileno, Cabo San Lucas, BCS, Mexico 28 2 20 y y
UCMP No. 57004 Lottia strigatella Califin, La Paz, BCS, Mexico 29 1 y y
UCMP No. 57017 Lottia strigatella Bahfa Tortugas, BCN, Mexico 30 3 y y
UCMP No. 57015 Lottia strigatella Bahia Tortugas. BCN, Mexico 31 3 14 y y
UCMP No. 157016 Lottia strigatella Bahia Tortugas, BCN, Mexico 32 3 y y
UCMP No. 57011 Lottia strigatella Cabo San Lucas. BCS, Mexico Na 2 9
UCMP No. 157012 Lottia strigatella Cabo San Lucas, BCS, Mexico Na 2 11
UCMP No. 57013 Lottia strigatella Cabo San Lucas, BCS. Mexico Na 2 13
MATERIALS and METHODS In the laboratory the coded specimens were sorted into
morphologically similar groupings inespective of locality,
In the course of this study we examined the morphology
and several specimens were then randomly chosen from
ofover 1500 putative specimens o^Lottia strigatella, Lot-
each group for sequencing. This approach increases thepos-
tia paradigitalis, and Lottia borealis from the Gulf of
Alaska to the Gulf of California, Mexico. In addition, sibility that all phenotypes present in a taxon will be sam-
nearly 100 specimens from 10 arbitraiy localities between pled as well as providing inultiple sequences for similar
Guaymas in the Gulf of California, Mexico and Alaska, individuals in each "lot." Afterreconstituting the groupings
California were collected for molecular sequencing (Fig- by locality it was discovered that 32 specimens from 13
ure 1). Specimens collected for sequencing were biased localities had been selected for sequencing (Table 1).
to represent as much morphological variation as possible Instit—utional abbreviations used herein are as follows:
from each locality. All specimens were labeled with a LACM Malacology Section, Natural History Museum
locality-based code and preserved in 70% ethanol of Los—Angeles County, Los Angeles, California;
(ETOH). UCMP Museum ofPaleontology, University ofCalifor-
'
Page 4 The Veliger, Vol. 46, No. 1
^ ^ ^ # ^ ^ #
1) M O O O — —
V—D —VD VdiOn 0d0 Od Od Od Or»", rm^,; ON -v^o iyfn m",O —r- rm<-,, in,
di d> rj ri ri ri ri ri ri ri ri
-oca odvdO'd-idt^drd^mdindo Oi—Nn"Or*i •r—+i o0—\orrNiioprri-O-N-^^OriOr—Ni
oo^ovr^O>nino VO^Ov^ONmONON^
odod-^dcNdrd-OdNd—d—do ^tr-^oinr^^-^int^
dvOd00d00d(Nd—-r^o —dd\OdvOdvO ri—n-roloooooooONN—-O^iO^r,ir—^-^i
' ri ri ri ri ri ri ri
(aU ^0'dt0d^OOOd^OOrOd^'(d-^NidV—^<0dm^^d0^-0d^^0OdOmOOdOO^ ^ONoroontd-c~nnI-Oim—TtO—N'rt<-^iimn
00 —
< ri ri ri ri ri ri ri ri
a doO-^d^odOoOdOOd'c-dn^d—md^^doPo-doOdoOdoO i'n^NOD^i^o^oionionOrN*^r,-O^N^O^
E
o
OdOOOd-od^odOoOdOOOdrfdoN-d^^d^^dO—OdoOdoOdoO 'ON^r<o-)or^--rr^jr<o-i'—* O—N(r^^imn
—
ri ri rj ri ri ri ri ri
T3
03
a, vd—od—idn-v^dD'dtv—Dd<i—dnid0n-do*d^-^r*cO-NNi^OnrdOoOdmid—n' '^o^qOrNj'N^DOONN'r-^^OTOfOinnOrN-^;
M
XfIlj di—ndooodooOd-'^dOoOdoOodO-^d^OtdvNrdr^-dOo\doidn—>drn-doo OmNNOOOo-m*oio^mitn^^miinn\ooo
c/1 ,H1^
rcrl tj
sOC ooO@i Ooo^Ooo^Oooi^Ooo^—^om'oomOooO^oooooor^o-vovoooionooooooooo i—n—O;NnOOm"r—i'Trij-—omono—o
ri ri ri ri ri ri ' ri ri
OdS^Od^OdOOdOOOdOOd-O^doOodOOdOOdOOdO-^d^i^dnOdcOJdOOdOOdOO O-*in—oqON"-*;Tttr;ni'n*—i'nr^oO-\^
ri ri ri ri ri ri ri ri ri
a O^ON^rOJ—'O^OriO-nO^isn^ruOJ-NOi"Nirn*innO—-\^r'i<*n-rvi~D—oiln—-in^(vONDno'-o^i\iDon>ionnOOiONTOi—N- Or—-;N-oo-o*oO—ONorndioOmdjNooidnoop-d-*o>-d£*o)romj
O—NiOrr-o^OnN—r-'-N;o—nvco-i;n—iorN-^—rJ—ONriinoro-rri)rr^O)NoO-N*r\-o; O—r»'^-ri~0O0Or<O-^ir-<*-rin^-ri<—-i0O0
ini^^'-^inininin^ininin'^'ND^'t '*'*' ONodododododododod
oEi'—in^ Co—EJ^ O^—NnOr;jr^-—'O;NOtr+N-lO-n——lnrvl-;no—l^no^'o-^nolon\—lrn^ilonnoT^rtl—^v-d~r'oJ*^O^OT'Ni^r-^~i'•n O0oNNdrro^<di-onidN0oD0dOo—'d*o'rdi*or—d^oiod'ro^~dO
OOtr^o of—<^n ro^Ei^ o^—;NrrDJj^OoiNonOoiOo\n'ri^-nOiiNOnTNitrn-ni'iO^nNiOrnNo^O—i•rni'ii^n^O^o^^r'o-iionnom—j—T;Omtn iONnord-oOdNoOdNodOo^dmoid-o*do—d
^ ^
Ta3. 5 o!i^n Oin OcOE-^> o—in "-tp^o; iiOnNnNONDrDT-ti-OnTN-tr^'-iOn-Ni—in'nir-iO~nNi—ono'otrotJriJ-nOi^Nnirinnio'O^oNNrodii^rn^-'r--jONNotOr-N'-rNO-tN~- Oo00Ndo(r-<di-iovrd~o;r-^ON^or—<d-,orcd-jiord—-jo'dorir-Or~N^
nw
k«Ph/"1 -^ rl r»-i + 1/1 Nor-oooNO — nr<-i"*inNOr- rC-: r-i*ir4nNriDrrl~-ro)orO)NrO<-, r—^, mrj
^
--i
W. B. Simison & D. R. Lindberg, 2003 Page 5
^
o
l/^
o•* NoO
ON Tt 00
o00 ot^ 0o0
—
NO vo
o o NoO ot^
^
o ^
o 5o o oOITn) o!^
e^ e* e* g« g^ g«
ON CnI r^ rmn 0—0 rn
0o0 Oon ro~ _: _: r_':l
mg^ mg« g« g; g=; g; g<
O — in On Ov "* On
-^ d dr- IdT) 0d0 dr^ 0d0
o
r4 ^ror^t^OvONtNCN
3 0dOrd<^dNOdNDd-*dONd00d00
<U c
C
O
u
^r-;Oinr-;Nq^>or-;
^ g=! g= g^ Og^ e^i ^* e^
W-1 -N*O 0O0N (-N* NOON 00 N'O^ I^ O0N0
^^ONinNO^ — m'^
812 -Hinoor-i^t^r-i'^tin
o o
^ ^ # g*e=:^e^gig=;g=;g=ig^
1ro/-^ our--1 OoON Om—NirO-r--'^-o—;oOr—O4rO^Niryj-)ON—ODNoOo
' ri (N ' CN (N < ri
^ ^ ^ gig<g^g<g=;g;g^g^g;
o — _ n m
^ oVD oVD Iro—- inrr4~r0-0NMrO~-o(oNr-NOr<~rj-(;N0r0-~
ri fN rj (N (N (N r-j rj
g^*d* gOd0=N0; dg1^=/;1 si«o«n gOr-iNi rgOiongOO«NgOr«oOgNn^DgOo=oO;rgO-*rg—-*og-os^ogrN*f
di
gi g=; ©; Og* *g* moOTtOv^NooOONro
rNo-O ol—O'OtoNN "d1drt NO—'TlON — OOOOOO
O —
—r- «00 O—N rj ro
. (N (N (N <N
. ( 1 <.^ '
Pase 6 The Veliger, Vol. 46, No. 1
X3
c
oa
aj
o
o
^
^g< g=i
(N VD
ON >0 ro
^ ^ ^
e^ O
0O0 M3 lO *
On lO
^- ri r^ r*"]
C/5 ^ ^
g=; s^ Si
0——0 d."r(o- —r-.~ Or-N- I--T•)
< (^1 r^i (~<-j
OSi Og* ©; Si g« g*
Od r—) ^dNO <0N0 NOON v0o0
fi rn ro
M
TrcPunrjlh ON^"On N^O^O^N lt^^~O-' r^K^<-l Nri^rOoi xOr^l--' Nn^^DO
03
(IJ
rn cMrt ^~* ^ ^ ^m ^ ^ ^ # #
ic-/Jl HXm1i) ON0nO0 N'^O r—r'~-l rr—J^ CNiDOn CoKN NirOn-
0C1, c
Boc O(dSNNi oOd^ KS(OiN SrN-^O NS—Oi;—^";* S^liO SN^iO 0S(0N*
NO r-^ r-^ NO r-^ NO
oOOe^d(08=N0;dOOg=; O—SNi ^NS=O! —lSOi -Sr^io ImS/-;) OISTi) N^SOi
NO NO p-^ r-^ NO r-~' NO
^ ^ ^ # # # ^ # ^ ^ ^
(N .— o in r- •^ CM Ol NO O—N NO
in 00 '^ in 04 00 ro NO in
O (NNO CNNO (NNO (NNO (NNO oir)i Cr-N-l' 0r-4^ ONlO ONlO Ou-li
U
^ ^ ^ ^ ^ ^ ^ # ^ ^ ^
^.— lr-OJ rr-~^ *t-~- irno m0£^0 OO) rp4- OiNn ^m^ ^in >ool
r^i niin rir)i ri)n OiJn NrO) riin Cr-J^ Or-l^ 0N4O 0N4O Oiln
^ # # ^ # ^ ^ ^ ^ ^ # #
dWN-O1 N0O0 NNOO O(NN N1—O Np~O- iinn o^ ^m £riR-n 0040 "i*n ^o
ri mri! (iNn CiNrIi rirji NINO N(NO 0r-4^ 0r-4' NOOl ONlO 0i4n
^ ^ ^ # ^ ^ ^ ^ ^ ^ ^ ^ ^
5^=<! dpV-) doo ((NN "..5g^) riNNrjOOi MiO0n0 Mirin-n riON-nOnl "^QS NOifOln- ONo^)O^ 0O0r-44N^ 0Nr.S—4O^^1 N(iONn 0Nmm4O 0imrr4-i
c
5
ouc 1-- r<"i -^ in ^a. 1"^ o"o orj CMl .^0/} ^r4 Nr4O 0o0) 0O4N Oro r—o 0r4o
:
W. B. Simison & D. R. Lindberg, 2003 Page 7
—
nia, Berkeley, California; and USNM Division of Mol- mtDNA region, a 680+ bp fragment was amplified using
lusks, U.S. National Museum of Natural History, Wash- the 16Sar and 16Sbr primers described by (Palumbi, 1996;
ington, D.C. Kocher et al., 1989). In a 0.5 mL gene amp tube, on ice,
36.45 |xL double-distilled water, 5 |jl1 lOX PCR buffer (Per-
Molecular Sequence Data kin Elmer), 2.5 (jlI lOfxM dNTP's (Pharmacia), 2.5 |jlL 25
(xM MgC12 (Perkin Elmer), 1 |jlL each ofthe 10 |jlM prim-
Cytochrome c oxidase I (COI) and 16S mtDNA genes ers, 1 |xL of template, and 0.25 \yL of taq (Perkin Elmer)
were partially sequenced and compared among 15 and 32 were combined. A negative control containing all reagents
individuals, respectively, from 13 localities (Table 1). except the template was run in parallel. The tube was then
COI and 16S were chosen for this study based on their transferred to a Perkin Elmer 9600 geneamp. The cycling
interspecific and intraspecific levels of variation found parameters began with an initial denaturation at 95°C for 2
among sequences of eastern Pacific patellogastropods minutes followed by 36 cycles with three temperature pla-
(Simison, 2000). teaus of95°Cfor50 seconds, 45°C for50 seconds, and 72°C
Extraction. Two equally successful DNA isolation pro- for 90 seconds, ending with a 7 minute extension at 72°C.
tocols were used: (1) saturated salt/chloroform extraction, PCRproducts were purified usingWizard® PCRpreps DNA
and (2) CTAB/phenolchloroform extraction. For each ex- Purification System.
traction, pedamlmtissue was cut from the foot mmamrgin approx- Cycle Sequencing. Direct double-stranded cycle se-
imately 3-5 along the margin and 3-5 toward the quencing of 20 to 30 ng of PCR product was performed
center ofthe foot. The tissue was soaked in deionized water in both directions using the aforementioned primers and
to remove any residual ETOH and finely diced to bits. For the ABI® cycle sequencing kit following a half reaction
the saturated salt technique, the diced tissue was digestemdMin ABI® cycle sequencing protocol. Cycle sequencing was
TaR1I.S5,ml10tumbeMcoEntDaTinAingan2d5040|0xLmisMolatNiAonCLb)uf,fe6r0(1\0iX010% pcleirnfgorpmaeradmeutseirngs awePreerk2i5n cEylcmleers 9at60906°gCenfeoram1p0.sTehceondcsy,-
SDS, and 10 |jlL proteinase K. The mixture was then vor- 50°C for 5 seconds, and 60°C for 4 minutes. Cycle se-
texed and stored on a shaker at 37°C overnight. Following quencing product was purified using Princeton Separa-
tissue digestion, 175 |jlLofsaturated NaCl solution was add- tions Centrisep spin columns, then dried in a speed vac.
ed. The samples wereinvertedfor5 minutes andcentrifuged The dried, purified cycle sequencing product was resus-
at 13 k rpm for 30 minutes. The supernatant was washed
pended in 2.5 |xl loading solution of 5:1 deionized form-
EmwwiiaTtxsOheHdp,crheblccyoieprniiotntfrvaioetfrerumdsgieouudnssiiafnntoggr1322twkmotiinrmvueptosmelsu.sfmuoerpTsehr1en5oaftsmauinipncteeurtnveacostolalundatmned1D0dNa0insA%d- 4aa|jlnm%1ioAdfeaB:csIra2ymP5lpralimmesimmna®dnMed3E7lg7DeolaT.DdiATNnhgAweistosghleeulq5tu0iweonanmcsgwe/arrmsulnlBoalanuddeedDaenoxantlryaaz3ne.6dc1o.mn5
cvaorlduemde,sthoef r7e0m%ainEiTnOgHp.ellTehtewa7s0%waEsTheOdHtwwiacsehwwiatsh tdiwso- werAleiaglnimgennetd b&y Ahnaanldysuissi.ngThtehe P16ASUPan4d.0CbO3Ia tseextqueednitcoers.
carded and the pellet dried for five minutes in a speed vac.
TsthoereDdNatA-w2a0s°Ce.luted in 50 |xL ofdouble-distilled waterand AannduCnOcoIrrpeacrtteidtioPnsvawlaues,cpaalicruwliasteedco(Tmapbalreisso2naonfdt3h)e. 1W6eS
For the CTAB technique, diced tissue was digested in a used the 16S northeast Pacific lottiid dataset generatedby
1.5 mL tube containing 600 |jlL 2XCTAB and 9 |xL of Simison (2000) to compare the phylogenetic relationship
proteinase k then incubated at 37°C overnight. 600 |xl of of the members within the nominal strigatella-paradigi-
phenol chloroform isoamyl alcohol (25:24:1) was added to talis "complex." This dataset was chosen because it in-
the tiss:ue mixture a:nd mixed via inversion for 5 minutes. cludes a representative ofeach ofthe nominal taxa under
The solution was then centrifuged at 13 k rpm for 15 min- study here. COI data was not included in Simison's
utes. The supernatant was added to 600 (xL ofchloroform (2000) phylogenetic analysis because ofpoor sample size.
isoamyl alcohol (24:1), mixedfor5 minutesandcentrifuged
at 13 k rpm for 15 minutes. DNA was precipitated using Morphology
600 |xL isopropanol and stored at -20°C for 2 hours. The
precipitate was centrifuged at 13 k rpm for 30 minutes at Digital images ofthe ventral, dorsal, and profile views
4°C. The supernatant was discarded and the pellet washed of 18 shells were captured with a digital camera con-
twice with two volumes of 70% ETOH and centrifuged at nected to a Scion LG-3 Scientific Frame Grabber system.
13 k rpm for 20 minutes. The ETOH was discarded and the In addition, an anterior portion ofthe radular ribbon from
pellet dried by speed vac for 5 minutes and eluted in 100 nine specimens (Table 1) was dissected from the head
|xL ofdeionized water region posterior to the odontophore and placed in a 0.5%
Amplification. Amplification ofa 700+ bp coding region sodium hypochlorite solution for 5 minutes or less to dis-
of COI was achieved with the HCO-2193 and LCO-1490 solve associated organic material and rinsed in distilled
primers described by Folmer et al. (1994). For the 16S water The radular ribbon was examined using an
Page 8 The Veliger, Vol. 46, No. 1
Table 4
Summary of range of distances for interspecific and intraspecific pairwise comparisons.
16S COI
Intraspecific Intraspecific
Taxon distances Taxon distances
strigatella 0.15%-1.33% strigatella 0.0%-3.79%
paradigitalis 0.0%-1.4% paradigitalis 0.0%-0.29%
argrantesta 0.0%-1.2% argrantesta 0.0%-2.68%
Interspecific Interspecific
Comparison distances Comparison distances
strigatella v. paradigitalis 11.11%-12.91% strigatella v. paradigitalis 16.43%-17.16%
strigatella v. argrantesta 17.96%-19.17% strigatella v. argrantesta 25.56%-27.75%
paradigitalis v. argrantesta 14.71%-16.4% paradigitalis v. argrantesta 25.47%-26.55%
ElectroScan Model E3 Environmental Scanning Electron strongly demarcated shell patterns such as found in L.
Microscope (ESEM). strigatella (e.g.. Figures 11-13).
Lottia paradigitalis and L. strigatella are substantially
RESULTS
more similar to one another than either is to L. argrantesta.
Molecular Sequence Data Both taxa have a wide range of overlapping shell pattern
variation (compare Figures 7, 8 with 9, 13), strongly de-
Uncorrected pairwise comparisons of 32 specimens of marcated shell patterns (compare Figures 3, 5 with 11, 15),
the nominal strigatella-paradigitalis "complex" and the and dark tessellate forms (compare Figure 5 [central area]
Simison (2000) 16S phylogeny of northeast Pacific lot- and 12). A solid, yellow-tan form has been found only in
tiids reveal three distinct lineages among the specimens. L. strigatella (Figure 14). Both taxa lack ribbing, and pri-
Sequence divergence within lineages was low while se- marily concentric growth lines texture the exterior shell sur-
quence divergence among lineages was greater (Table 4). face although microscopic radial treads are sometimes pre-
The three lineages are each nested in different clades of sent; shell profiles are virtually identical in both taxa. One
the Simison 16S phylogeny. The geographic distributions discernible difference between L. paradigitalis and L. stri-
of these lineages are sympatric over portions of their gatella shell color patterns is the stronger bifurcating pat-
ranges; the San Francisco Bay group (L. paradigitalis) terns ofthe whitemarkings presentinL.paradigitalis(com-
overlaps with the Baja California group (L. strigatella) in pare Figures 3, 5, 7 with 9, 11, 13).
southern California, while the Baja California group co- The radula of L. argrantesta is readily distinguishable
occurs with the Gulf group (L. argrantesta n. sp.) in the from those of both L. paradigitalis and L. strigatella. In
southern Gulf of California (Figure 1). L. paradigitalis (Figures 26—28) and L. strigatella (Fig-
ures 20-22) the inner margins ofthe second lateral teeth
Morphology
appear convex, while in L. argrantesta the edges appear
Examination ofradular and shell morphologies oftaxa concave (Figures 23-25). This places the cusps of the
sorted by genotype revealed previously unsuspected mor- second lateral teeth of L. paradigitalis and L. strigatella
phological differences, especially between Gulf speci- closer to the cusps of the first lateral teeth than they are
mens of L. strigatella and L. argrantesta. Although both in L. argrantesta. Lottia paradigitalis and L. strigatella
taxa have a wide range of shell pattern variation, speci- radulae are very similar in overall morphology. One pos-
mens of L. argrantesta (Figures 16 and 19) tend to be sible difference we noted was that radular segments in L.
lower in profile than specimens of L. strigatella (Figures paradigitalis appear slightly shorter than in L. strigatella.
9 and 15). Lottia strigatella specimens also tend to have There is minor radular variation in L. paradigitalis (com-
more convex posterior shell profiles. Both taxa have var- pare Figures 27 and 28), but it is not as marked as that
iegated forms that are similar in shell color and pattern reported in the Panamic taxon Lottiafascicularis (Simi-
&
(compare Figures 9 and 16) as well as dark tessellate son Lindberg, 1999).
forms with random white markings (compare Figures 12
DISCUSSION
and 19). In many cases L. argrantesta can be distin-
guished from L. strigatella by the presence oflow coarse After 135 years of conjecture, the results of this study pro-
ribs on its shell, but relatively smooth specimens also vide unequivocal evidence ofthe distinctness ofLottia stri-
occur (Figure 17). Lottia argrantesta appears to lack gatella, Lottia paradigitalis, and a third previously unrec-
W. B. Simison & D. R. Lindberg, 2003 Page 9
ognized taxon, Lottia argrantesta, sp. nov. Moreover, these studies. Moreover, the pattern has implications for clade
taxa are not members ofa "species complex" oreven sister selection (Jablonski, 1987; Lloyd & Gould, 1993; Ver-
taxa, but rather members of three distinct subclades within meij, 1996).
the northeastern Pacific Lottiidae (see below). These data
SYSTEMATICS
and the evolutionary history they reveal provide a compel-
ling demonstration of the levels of morphological conver- Patellogastropoda Lindberg, 1986
gence present in the Patellogastropoda.
Without the molecular data Lottia argrantesta would Lottiidae Gray, 1840
likely have gone unrecognized. And while Lindberg Although this taxon is the most diverse and abundant of
(1981:75) revived the use of the specific name paradi- all patellogastropod clades in the world, it is diagnosed
gitalis for northern California specimens ofL. strigatella by few characters, and most notably by an absence of
based on radular differences, it was thought at that time calcitic foliated shell microstructures and the presence of
that Lottia strigatella and Lottia paradigitalis likely rep- fibrillar ones. Foliated shell structures are present in the
resented a species pair which transitioned at Point Con- Patelloidea, Nacelloidea, and many Acmaeoidea, but are
ception, California. This scenario was consistent with the absent in the Lottiidae. The remaining anatomical and
range ofmorphological shell and radular variation shared shell characters of the Lottiidae are all found in different
by these two taxa, their similar habitats, and their contig- combinations in one or more of the outgroups.
uous ranges. Moreover, allopatric divergence during a Two major subclades, Lottiinae and Patelloidinae, have
glacial or interglacial period provided a plausible mech- been previously recognized on radular and shell micro-
anism. structure characters; they are also delimited by molecular
However, this scenario is falsified by the 16S phylog- characters (Simison, 2000). Both groups contain numer-
eny. Instead, the shared morphology and habitats ofthese ous subclades that have been named, as well as previ-
taxa appear to result from convergence, not common an- ously unrecognized ones. In North America, Australia,
cestry, and range size is characteristic ofthe larger, more Japan, and South America, members of the Lottiidae
inclusive clades to which each taxon belongs and not the compose the vast majority ofthe species in the nearshore
outcome ofa recent divergence from a common ancestor patellogastropod guilds. Unlike the Acmaeoidea, mem-
While disconcerting relative to the more familiar scenar- bers of the Lottiidae are not found in the deep sea. In-
io, this result suggests that deeper divergences are also stead, they are primarily intertidal in habitat and rarely
affected by modern day Point Conception. This barrier is occur deeper than 30 meters. They occupy a wide range
possibly thermal in nature and acts to limit the distribu- of intertidal heights and habitat types. Some species are
tions of either larvae or adults. For members of the Col- tolerant of brackish water and can be found in estuarine
lisella and "A" subclades (Figure 2) potential southern habitats. Several species are associated with algae and
limiting temperatures appear to occur near the southern marine angiosperms while others are found only on car-
California Bight; northern limiting temperatures do not bonate substrates.
appear to be reached until the northern GulfofAlaska or The Lottiidae are distributed worldwide with the ex-
Aleutian Islands. For members of the sister clade that ception of Antarctica. There are no strong biogeographi-
contains L. strigatella, Macclintockia, and Nomaeopelta cal trends within the global distribution of Lottiidae, and
(Figure 2), northern limiting temperatures are seldom different taxa in a single clade may range from cool tem-
found north of central California, and the majority are perate to subtropical environs. Members of the Lottiidae
south of the Bight. Possible southern limiting tempera- are identifiable in the Cretaceous based on shell micro-
tures in the L. strigatella + Macclintockia + Nomaeo- structure and radularcharacters (Akpan et al., 1982; Lind-
pelta clade occur at the mouth of the Gulf of California. berg, 1988). By the Eocene, circulatory characters that
Thus ranges in CoUisella -\- subclade A average about diagnose living taxa are visible as impressions preserved
7900 km, while ranges in the L. strigatella + Macclin- on the interioroffossil shells (Lindberg & Squires. 1990).
tockia -\- Nomaeopelta clade average only about 1600 km.
Moreover, these different thermal tolerances appear to be Lottia Gray, 1833
clade-level traits that first appeared in their respective
common ancestors in the Late Miocene orEarly Pliocene, Lottia Gray, 1833:800. Type species, by subsequent desig-
nation of Dall, 1871: Lottia gigantea Sowerby, 1834.
long before glacial and interglacial sequences provided a
Northeastern Pacific.
plausible mechanism for divergence. Subsequent diver- Tecturella Carpenter. 1860:3. Type species, by monotypy:
gences in both clades produced taxa with similar toler- TectitrellagrandisGray (= Lottiagigantea = Sowerby,
ances, suggesting that thermal tolerance was heritable in 1834) (not Stimpson, 1853:36).
these clades and this trait constrained descendents to sim- TectiirinaCarpenter, 1861:219.Type speciesbyoriginaldes-
ilar range sizes. This finding offers a deeper historical ignation: Tectiihna grandis "Gray" (= Lottiagigantea
view of the potential makeup of latitudinal barriers and Sowerby, 1834).
range size than is attainable through classical taxonomic Shell profile varies from high to low with the apex
Page 10 The Veliger, Vol. 46, No. 1
+6 A
Lottia argrantesta
Nomaeopelta Clade
+4
+7
Macclintockia Clade
+3
Lottia strigatella
+3
+4
undescribed SNI taxon
+3 Collisella Clade
Lottia Lottia paradigitalis
+4 Clade A
{"Notoacmea'TTectura" spp.)
1 of 1 trees 01 = .3149
tree length = 3309 steps HI = .6851
informative characters = 437 Rl = .5409
Figure 2. Simison's (2000) 16S maximum parsimony (PAUP 4.0b3a: Swofford, 2000) phylogenetic hypothesis of the relationships
among the major clades oftemperate northeastern Pacific patellogastropods showing the placement ofthe three taxa discussed herein (^^).
Numbers on branches = decay values. SNI is an unidentified San Nicholas Island, California taxon with a unique haplotype. CI =
consistency index, HI = homoplasy index, RI = retention index.