HOTTINGER ET AL. -ROTALIID FORAMINIFERS 33 marginalonly on one or both sides of the test or complete. The septal Umbilicalbowl. -A deep, wide or narrowconical space in axial po- flap may extend into a umbilical plate, a foramenalplate, a bipartitor, sition formed between inner umbilical chamber walls, wherever the a cover plate, or a toothplate. [Pariesproximus, pars auct.] latterare separatedfrom the outer umbilicalwalls by a distinct edge or Supplementaryapperture(s). -Primarily formed openings either in shoulder; may communicate with a true open umbilicus or may be aperturalface ("aperturalpores") or (slit-like)in suturalposition, always closed by an umbilicalplug. [Pseudoumbilicus,pars auct.] in additionto a main cameralaperture. Sutural supplementary apertures Umbilicalplate. -A more or less contorted plate-like test element, are not convertedinto intercameralforamina because of their position extending between distal and proximal chamber walls and joined to and thus do not serve for passage of functional endoplasm between both, attachedto the intercameralforamen and to the main aperture, chambers.The same seems to be true of multiple supplementaryap- but not protrudinginto the latter. Separatesthe main chamberlumen ertures,which, althoughsituated in the septum,may be absentin earlier from a primary spiral umbilical canal, produced between plate and chambersand may be pluggedin partat a subsequentinstar. [Secondary adjacentcoil or by foliar chamberlets,or from stellarchamberlets. Be- aperture(s),pars. auct.] tween plate and adjacentcoil or within the plate itself an openingpro- Toothplate.-A contorted plate runningfrom an intercameralfora- vides connection between chamber and foliar or stellar chamberlet, men to an aperture,attached to both. Folded in a single, double, or whereverpresent. This openingmay remainopen in all chambersor it spiral fold. Folds (or tongues)with free, often serrateddistal ends and may be "plugged"in all but the ultimate chamberby a sealing plate. distally protrudinginto the aperture.A toothplate separatespartly or An umbilical plate may be single or composed of two symmetrical entirelythe main chamberlumen from an axial space (adaperturalde- branches in some planispiralgenera, thereby producing one or two pression)in post-embryonicstages. Interconnected toothplates produce spiralumbilical canals between plate and adjacentcoil. [Foramenalplate; a primarycanal. In contrastto an umbilicalplate, the toothplateis never umbilicalflap; murus reflectus; toothplate; paries proximus; pars auct.] associatedwith a foliar or stellarchamberlet and it does protrudewith a free edge distally and adaxiallyto the aperture.[Sipho-central pillar, pars auct.]

J. Paleont., 65(1), 1991, pp. 33-37 Copyright? 1991, The PaleontologicalSociety 0022-3360/91/0065-0033$03.00 DISTRIBUTION AND POSSIBLEPALEOECOLOGICAL SIGNIFICANCEOF ANNECTINA VIRIOSA,A NEW SPECIESOF AGGLUTINATED FORAMINIFERA FROM NONMARINE SALT PONDS IN R. TIMOTHY PATTERSON ANDW. BRIAN McKILLOP Ottawa-CarletonGeoscience Center and Departmentof EarthSciences, Carleton University, Ottawa,Ontario KIS 5B6, Canadaand ManitobaMuseum of Man and Nature, 190 RupertAvenue, ,Manitoba R3B ON2,

AnsTRAcr-Annectinaviriosa, a new species of Ammodiscidae(Foraminiferida), is describedfrom Recent brackishponds on a salt flat adjacentto LakeWinnipegosis, Manitoba. Colonization of the ponds was probablyby avian transport.The distinct morphotype is either the result of an allopatricspeciation event within the last 5,000 years or it is a previouslyundescribed species of shallow- waterAnnectina from nearbyHudson Bay. It is also possible that recoveredspecimens are previouslyunrecognized ecophenotypic variantsof some knownspecies produced by unknownhostile environmental factors within the pondecosystem. The latterhypothesis is difficultto test without extensive biologicalculturing. The largenumber of phenotypicallystable specimensliving in these ponds warrantsrecognition of these populationsas a distinct taxon.

INTRODUCTION of (Figures 1, 2; Patterson et al., 1990). URING A distributional analysis of nonmarine foraminifera Annectina is a comparatively rare genus-it has previously only D from salt ponds of the Lake Winnipegosis region of Man- been observed in Paleocene sediments from central Asiatic USSR itoba, a species of Annectina Suleymanov, 1963, was observed and from Holocene deposits in Australia--and has never been that was not referable to any previously described species (Pat- reported from marsh environments, although the closely related terson et al., in press). The foraminiferal fauna at most sampled genus Glomospira Rzehak, 1885, has occasionally been reported Lake Winnipegosis localities was dominated by Jadammina from estuarine environments (Scott et al., 1980). The tolerance macrescens with lesser numbers of Polysaccammina ipohalina. of this latter genus to low-salinity estuarine environments may This fauna is very similar to those found in marshes that border explain the tolerance of Annectina specimens for the brackish and James Bay (Patterson et al., 1990; Scott and water (17.00-20.72%o) characterizing the ponds at the Lake Win- Martini, 1982). Numerous previous reports attribute the estab- nipegosis locality. lishment of foraminiferal faunas in the Lake Winnipegosis re- gion to avian transport (Resig, 1974; Cann and De Deckker, METHODSAND MATERIALS 1981; Patterson, 1987; Patterson et al., 1990). Foraminifera were examined from a brackish pond on a salt Difficult to explain, however, is the enigmatic occurrence of flat 4 km north of the intersection of Manitoba Provincial High- an Annectina sp. dominated fauna in brackish ponds just west ways No. 20 and No. 271 (Figure 1). This locality consists of a

This content downloaded on Sun, 10 Mar 2013 22:08:30 PM All use subject to JSTOR Terms and Conditions 34 JOURNAL OF PALEONTOLOGY, V. 65, NO. 1, 1991

101000.0' 100000.0' TABLE1--Percent abundance of foraminiferaand arcellaceansfrom a brackishpond near Lake Winnipegosis, Manitoba; L = living,T = totalpopulation; SF = saltflat environment. Sample May88 Sep 88 Jul 89 53000.0' Lake53000.0 Salinity 17.00 20.72 - pH 8.2 8.0 - Watertemperature (0C) 23.0 16.0 - Environment SF SF SF No. of species 4 3 1 Numberof individuals/10cc L 0 11 0 T 27 120 24 Arcellaceans D. protaeiformis L T 4.0 Foraminifera A. viriosa L 8.3 Hudson T 74.0 47.5 100.0 Bay P. ipohalina L Manitoba T 1.7 J. macrescens L T 18.0 41.7 stu y J. macrescens L T 4.0 A. innipeg (polystomaphenotype) 52000.0' 52000.0'

Annectina viriosa (Table 1). Physicochemicalparameters, such as water temper- Sample Locality ature and pH, were measuredat the time of collection. Water salinity was calculatedfrom chlorinitymeasurements using the km equation (Stricklandand Parsons, 1968): 0 10 20 30 40 50 Winnipegosis salinity (%o)= 0.030 + 1.8050 x chlorinity (%o). I I I I I Bufferedformalin (7.5 percentwith borax added to bufferthe 101000.0' 100000.0' solution)was initiallyadded to the biologicalsamples to prevent decay. In the laboratory,rose Bengal stain (a few grains) was FIouRE1--Location map of LakeWinnipegosis, Manitoba, showing Annectinaviriosa type locality. added to the samplesto detect specimensof foraminiferaliving at the time of collection. After several hours the samples were rinsed to remove excess stain. Samples were then sieved using seriesof interconnectedshallow ponds (3-7.5 cm deep) partially a 35-mesh Tyler (500 Mm)screen to retain coarse organic ma- separatedby a numberof low hummocks(Figure 2). The entire terial and a 230-mesh Tyler (63 Mm)screen to retain the arcel- area covered by ponds is about 20,000 m2. One of the larger laceans and foraminifera.Scanning electron micrographswere ponds at this locality, whereall sampleswere collected,is about taken using the CambridgeStero-Scan 90 scanningelectron mi- 20 m long by 6 m wide (120 m2). This locality coincides with croscope located at the Geological Survey of Canada,Vancou- Sample Station 6 of Patterson et al. (1990). Samples analyzed ver, British Columbia, using Polaroid NP 55 film. for this report were collected from this sample station three times: 27 May 1988; 27 September 1988; and 11 July 1989 DISCUSSION Three hypothesespossibly explain the presenceof this distinct phenotype of Annectinain the Lake Winnipegosisregion: 1) An allopatric speciation event occurred in these ponds following the establishment of a small founder population. A good argumentcan be made for recognitionof this population as a separatespecies. This phenotypeis so distinct that it is not obviously referableto any known species,and a largepopulation (101 specimens)was recovered.This populationis characterized by a very stable phenotype (displayinglittle morphologicvari- ation) and many living specimens, indicating that the species has probably been at this locality for many generations.This also suggeststhat the species is very hardyand capable of with- standing salinity shifts and the desiccation events that occa- sionally occurin most salt pond environments.The ponds from which Annectina specimens were recovered, however, shrink slightlyduring drought conditions but have never been observed to dry completely. In winter these ponds do not freeze com- pletely but become slushy. Cann and De Deckker(1981) iden- tified species of trochamminidand elphidid foraminiferafrom FiGURE2-Annectina viriosa n. sp. typelocality, with numerous inter- ephemeral in southwest Australia that could withstand connectedsmall ponds. both salinity shifts and seasonal desiccation events.

This content downloaded on Sun, 10 Mar 2013 22:08:30 PM All use subject to JSTOR Terms and Conditions PATTERSONAND McKILLOP- NEW SPECIES OF AGGL UTINA TED FORAMINIFERA 35

It is not known whetherthe Annectinaspecimens from Lake obscured by misidentificationof these specimens with respect Winnipegosishave been at this location long enoughto speciate. to species. Rates of speciation vary greatly depending on the type of or- ganism and its population structure(Futuyma, 1986). For ex- SYSTEMATICPALEONTOLOGY ample, some plant populations that have been isolated for at least 20 million years, such as the Americanand Mediterranean Family AMMODISCIDAEReuss, 1862 plaintains, form viable hybirds (Stebbins and Day, 1967). In Subfamily AMMOVERTELLININAE contrast, Fryer and Iles (1972) reported that several distinct Saidova, 1981 species of cichlid fishes are endemic to Lake Nabugabo,which Genus ANNECTINASuleymanov, 1963 is separatedfrom LakeVictoria in Africaby a strip of low-lying ANNECTINA VIRIOSAn. sp. land that has been dated by radiocarbonanalysis to have de- Figure 3.1-3.6 veloped only 4,000 years before present(B.P.). Under the right Diagnosis.-Ovate species of Annectina with axis of coiling conditions speciation can thereforebe an almost instantaneous more or less in same plane. event (Eldredgeand Gould, 1972). Description.-Test free,compressed in plane of coilingof out- Following a warm phase, approximately 5,000 years B.P., er whorl, proloculus followed by undivided second chamber climatic conditions rapidly deteriorated in northern North tightly appressedagainst outer walls of earlierwhorls that serve America. Conditions similar to those presently found in the as floor for chamber; enrollment about elongate axis for 2-4 Lake Winnipegosisregion have existed for at least 4,000 years whorls; specimens with two whorls elongate, specimens with (Scott et al., 1987). Patterson et al. (1990) observed that the more than two whorls more circular;width of second chamber present distribution of foraminiferain the Lake Winnipegosis rapidly increasingas added, second chambercoiled in various areais significantlydifferent from that of the foraminiferalfauna planes, althoughby third or fourth whorl coiling more or less inhabitingthe region 5,000 years before present (Nielsen et al., planispiral,apparent planispiral coiling arrangementsomewhat 1987). Unfortunately, no cores were taken to determine the illusory due to large size of third and fourth whorls relative to duration of Annectina at this locality. However, based on the earlier whorls, occasional sigmoidally coiled specimens ob- above caveat on speciation rates, it is entirely feasible that this served;wall agglutinated, imperforate, of medium-to fine-grained populationof Annectinamay have existed in this area for over detrital sand, with pennate diatoms incorporatedinto walls of 4,000 years and has evolved to representa distinct species. some specimens, approximately8.5 Am thick in outer whorl; 2) The Annectina specimens found on the salt flat are eco- simple aperture at end of tube, of same width as chamber, phenotypic variants of populations living in marine environ- surroundedby slightly producedlip. ments. The distinct morphologydisplayed by individualsfound Etymology.-From the Latin viriosus, robust, strong, with at this locality is the result of a phenotypic response to envi- referenceto the survival of this species in this brackish non- ronmental factors. marine environment. Ecophenotypicvariation is well documentedin many species Types and Figured holotype (GSC of foraminifera,such as the shallow-waterspecies Cribroelphid- 96786), figuredoccurrence.--Recent.paratypes (GSC 96787-GSC 96790) and unfig- ium excavatum(Miller et al., 1982), the marsh species Jadam- ured paratypes(GSC 96791) from a brackishpond on a salt flat mina macrescens(Scott and Medioli, 1980; Patterson, 1990), 4 km north of the intersection of Highways No. 20 and No. and the planktonicforaminifer Neogloboquadrina pachyderma 271, near Lake Winnipegosis,Manitoba. Deposited in the col- (Reynolds and Thunell, 1986; Yeh et al., 1989). Miller et al. lections of the Geological Survey of Canada,Ottawa, Ontario. (1982) observed that the greatest morphologicalvariability in Dimensions.-Maximum length, 350 Am; maximum width, Cribroelphidiumexcavatum is found in nearshore temperate 325 Am; maximum thickness, 125 tm. estuaries that are subject to extremes in climate and environ- Remarks.-Comparison oftype illustrationsshows Annectina mental conditions. Similarly,Jadammina macrescenshas a ten- viriosato be very similar in appearanceto Trochamminamili- dency towardsubareal supplementary apertures in highersalin- oloides Jones, Parker,and Kirkby (1869) from the Permian of ity environments, whereas phenotypes from lower salinity England. Due to the tremendoustemporal differenceit is un- environments typically have none (Scott and Medioli,~1980; likely that these forms are conspecific,although it is difficultto Patterson, 1990). Finally, northeasternPacific populations of assess similarities and differencesdue to a lack of detail in the Neogloboquadrinapachyderma tend to reversecoiling directions type descriptionof Trochamminamilioloides. It is also impos- depending on the water temperature.In addition, the amount sible to compare type material as the respositoryof this latter of nutrientsin the watercolumn determinesthe shape and level species is unknown(Ellis and Messina, 1940).Annectina viriosa of encrustationof the test (Reynolds and Thunell, 1986). The is also similarin outlineto Ammodiscusperuvianus Berry (1928). environmentalconditions existing in the ponds may be stressing However, Berry'sspecies has 4-5 whorls and is entirely plani- the metabolismofAnnectina. Although the observedphenotype spiral, as opposed to the 2-4 whorls and more variablecoiling is dissimilar to any phenotype displayed by Annectina else- characterizingthe Annectinaviriosa test. It was initially thought where,it is possible that the recoveredspecimens are ecopheno- that Annectina viriosa might be an ecophenotypic variant of typic variants of an existing species. Trochamminasquamata gordialis Jones and Parker(1860), the 3) These specimens are phenotypicallysimilar to an unde- type species of GlomospiraRzehak, 1885 (now known as Glo- scribed species living in the Hudson-James Bay area, since the mospira gordialis). This was based on the probable original foraminiferaof this region are not well known. colonization of the Lake Winnipegosis area by foraminifera The presenceof a large,morphologically stable populationof transportedfrom shallow-marineenvironments elsewhere (Pat- living and dead specimens at this locality supportsthe hypoth- terson et al., in press). Glomospiragordialis has been reported esis that this speciesis the resultof an allopatricspeciation event, fromrelatively shallow, brackish estuarine environments in Nova or is an unknown species from Hudson or James Bay. If later Scotia (Scott et al., 1980) with salinities similar to the brackish studies should prove this species to be a radicallydifferent eco- values recordedat the Lake Winnipegosisarea locality. How- phenotypicvariant of some well-knownspecies, it can easily be ever, Glomospiragordialis has numerousstreptospirally coiled placed in synonymy. The taxonomic relationshipsand ecolog- whorls and very distinct sutures between whorls. In contrast, ical significanceof this and other Ammodiscidaecould only be Annectina viriosahas a maximum of only four whorls, despite

This content downloaded on Sun, 10 Mar 2013 22:08:30 PM All use subject to JSTOR Terms and Conditions 36 JOURNALOF PALEONTOLOGY,V. 65, NO. 1, 1991

1P I*~40 r Iir r

i40 *-

Ilk.' )

FIouRE3-1-7, Annectinaviriosa n. sp., brackishpond nearLake Winnipegosis, Manitoba, Recent. 1, side view of holotype(GSC 96786) showing somewhatelongated test, x 302; 2, aperturalview of damagedparatype (GSC 96787) showing largeaperture, which is just an opening at the end of the enrolled second chamber;damage to test (while reorienting)in lower part of very delicate test shows relative thickness of wall to test size, x 489; 3, enlargementof apertureof same specimen showing largenumbers of pennatediatoms on inner surfaceof test, x 1,720; 4, side view of same paratypeshowing slightly flared apertural opening, x 379; 5, side view of paratype(GSC 96788) showingchange in orientation of coiling of final whorl, x 200; 6, edge view of paratype(GSC 96790) showing compressedtest and aperturalregion where wall of previous chambercan be seen servingas the floor of final whorl;opening at bottom of illustrationis damageto the test, x 261. being about the same size as Glomospira gordialis, and is elon- dialis. Based on such radical morphologic differences, it is not gate and compressed with the axis of coiling of later whorls possible to refer the present species to Glomospira gordialis. more or less in a single plane. Annectina viriosa also has much Numerous unicorporated pennate diatom frustules were found less distinct sutural contacts between whorls and the relative on the interior of many tests ofA. viriosa (Figure 3.3), indicating breadth of the whorls is much greater than in Glomospira gor- a probable food source.

This content downloaded on Sun, 10 Mar 2013 22:08:30 PM All use subject to JSTOR Terms and Conditions PATTERSON AND McKILLOP-- NEW SPECIES OF AGGL UTINA TED FORAMINIFERA 37

ACKNOWLEDGMENTS -, D. B. Sco-rr, ANDW. B. MCKILLOP.1990. Recent marsh-type foraminiferafrom Lake 765- This research was partially supported by the Natural Sciences agglutinated Winnipegosis,Manitoba, p. 781. In C. Hemleben, M. A. Kaminski, W. Kuhnt, and D. B. Scott and Research Council of Canada (NSERC) Oper- Engineering (eds.), and Taxon- Grant OGPOO41665 to R.T.P. Field research were Paleoecology,Biostratigraphy, Paleoceanography ating grants omy of AgglutinatedForaminifera. NATO Advanced Science Insti- provided to W.B.M. from the Manitoba Museum of Man and tute Series, Series C, Volume 327. Nature Foundation and the Manitoba Heritage Foundation Inc. RESIG,J. M. 1974. Recent foraminiferafrom a landlockedHawaiian We also wish to thank J. Dubois for assistance in the field, as lake. Journalof ForaminiferalResearch, 4:69-76. well as F. Clark, S. T. Goldstein, E. MacLellan, and W. V. Sliter REUSs,A. E. 1862. Entwurfeiner systematischenZusammenstellung for critically reviewing the manuscript. We also wish to thank der Foraminiferen.Sitzungsberichte der KaiserlichenAkademie der J. Luternauer for providing use of the scanning electron micro- Wissenschaftenin Wien,Mathematisch-Naturwissenschaftliche Classe 44:355-396. scope at the Geological Survey of Canada, Vancouver, and to (1861), L. ANDR. C. THUNELL.1986. Seasonal and P. Krause for assistance in its operation. REYNOLDS,A., production morphologicalvariation of Neogloboquadrinapachyderma (Ehren- REFERENCES berg)in the northwestPacific. Micropaleontology, 32:1-18. RZEHAK,A. 1885. Bemerkungeniiber einige Foraminiferender Oli- BERRY,E. W. 1928. The smallerforaminifera of the middle Lobitos gocainFormation. Verhandlungendes NaturforschendenVereins in shales of northwesternPeru. EclogaeGeologicae Helvetiae, 21:390- Briinn (1884), 23:123-129. 405. SAIDOVA,KH. M. 1981. O sovremennom sostoyanii sistemy nadvi- CANN,J. H., ANDP. DEDECKKER. 1981. Fossil Quaternaryand living dovykhtaksonov Kaynozoyskiikh bentosnykh foraminifer (On an up- foraminiferafrom athalassic (non-marine) saline lakes, Southern Aus- to-date system of supraspecifictaxonomy of Cenozoic benthonicfo- tralia.Journal of Paleontology,55:660-670. raminifera).Institut Okeanologii P. P. Shirrshova,Akademiya Nauk ELDREDGE,N., ANDS. J. GOULD. 1972. Punctuatedequilibria: an SSSR, Moscow, 73 p. alternativeto phyleticgradualism, p. 82-115. In T. J. M. Schopf(ed.), ScoTT,D. B., ANDI. P. MARTINI.1982. Marshforaminifera zonations Models in Paleobiology.Freeman, Cooper and Company,San Fran- in westernJames and Hudson Bays. Le NaturalisteCanadien, 109: cisco. 399-414. ELLIS,B. F., ANDA. MESSINA.1940. Catalogueof Foraminifera.Amer- -, ANDF. S. MEDIOLI.1980. Quantitativestudies of marsh fora- ican Museumof NaturalHistory, New York, 1940 and supplements. miniferaldistributions in Nova Scotia.Implications for sea level stud- FRYER,G., ANDT. D. ILEs. 1972. The Cichlid Fishes of the Great ies. CushmanFoundation for ForaminiferalResearch Special Pub- Lakes of Africa. T.F.H. Publications, Neptune City, New Jersey, lication, No. 17, 58 p. 641 p. , ANDA. A. L. MILLER.1987. Holocene sea levels, paleo- FUTUYMA,D. J. 1986. EvolutionaryBiology, 2nd ed. SinauerAsso- ceanography,and late glacial ice configurationsnear the Northum- ciates, Inc., Sunderland,Massachusetts, 600 p. berlandStrait, MaritimeProvinces. CanadianJournal of EarthSci- JONES,T. R., ANDW. K. PARKER.1860. On the Rhizopoda fauna of ences, 24:668-675. the Mediterranean,compared with that of the Italianand some other - , C. T. SCHAFER,AND F. S. MEDIOLI.1980. Eastern Canadian Tertiarydeposits. Quarterly Journal of the GeologicalSociety of Lon- estuarineforaminifera: a frameworkfor comparison.Journal of Fo- don, 16:292-307. raminiferaResearch, 10:205-234. , ANDJ. W. KnmvB. 1869. On the nomenclatureof the STEBBINS,G. L., ANDA. DAY. 1967. Cytogenicevidence for long con- foraminifera;Part XIII--The Permian Trochamminapusilla and its tinued stabilityin the genus Plantago. Evolution, 21:409-428. allies. Annals and Magazineof Natural History, ser. 4, 4:386-392. STRICKLAND,J. D. H., ANDT. R. PARSONS.1968. A practicalhandbook MILLER,A. A. L., D. B. SCOTTrr,AND F. S. MEDIOLI.1982. Elphidium of seawateranalysis. Fisheries ResearchBoard of Canada,Bulletin, excavatum (Terquem);ecophenotypic versus subspecificvariation. 167, 311 p. Journalof Foraminiferal 12:116-144. Research, SuLEYMANOV,I. S. 1963. Novyy rod i dva novykh vida iz semeystva NIELSEN,E., D. H. McNEIL,AND W. B. McKILLOP.1987. Originand Ammodiscidae[A new genusand two new species of the family Am- paleoecologyof post-LakeAgassiz raised beaches in Manitoba.Ca- modiscidae]. Dokladari Uzbekiston SSR, Fanlar Akademiyasining, nadian Journalof EarthSciences, 24:1478-1485. Tashkent, 11:41-43. PATTERSON,R. T. 1987. Arcellaceansand foraminiferafrom Pleis- YEH,C. C., R. T. PATTERSON,AND R. H. OSBoRNE.1989. Fourier tocene Lake Tecopa, California.Journal of ForaminiferalResearch, analysisof the planktonicforaminifer Neogloboquadrina pachyderma 15:333-343. (Ehrenberg)from the PleistoceneSanta BarbaraFormation, Califor- 1990. Intertidalbenthic foraminiferalbiofacies on the Fraser nia. Journalof Micropalaeontology,8:77-85. River Delta, British Columbia:modern distributionand paleoeco- logical importance.Micropaleontology, 36:229-244. ACCEPTED11 JULY 1990

This content downloaded on Sun, 10 Mar 2013 22:08:30 PM All use subject to JSTOR Terms and Conditions