The Bogs of Central Minnesota Author(s): Verona M. Conway Reviewed work(s): Source: Ecological Monographs, Vol. 19, No. 2 (Apr., 1949), pp. 173-206 Published by: Ecological Society of America Stable URL: http://www.jstor.org/stable/1948637 . Accessed: 30/01/2012 11:23

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Ecological Society of America is collaborating with JSTOR to digitize, preserve and extend access to Ecological Monographs.

http://www.jstor.org THE BOGS OF CENTRAL MINNESOTA

VERONA M. CONWAY The Universityof Shefflield,England CONTENTS

PAGE

INTRODUCTION ...... 175

SCOPE AND METHODS OF WORK .7...... 176

OUTLINE OF THE BOG SUCCESSION; THE DEVELOPMENT OF ACIDITY .177

THE BOG COMMUNITIES .185

AUTECOLOGICAL NOTES .193

CIIARACTERISTICS OF THE PEAT .199

GENERAL DISCUSSION .202

LITERATURE CITED .204

APPENDIX I. THE QUINHYDRONE METHOD OF PH DETERMINATION .205

APPENDIX II. DETAILS OF BOG LOCALITIES .206

174] THE BOGS OF CENTRAL AINNESOTA

INTRODUCTION peat chemistry or microbiology or the agricultural Ever since the earliestecological studies, ecologists uses of peat, are working together not only to in- seem to have been fascinatedby the study of bogs crease our basic understanding of peats and peat- and bog-dwellingplants. Perhaps thisis because the formiiingvegetation, but to solve the problems involved bog habitat is of very limited extent compared to in deciding the best practical policies for dealing the total area occupied by the vegetationof mineral with peat-bearing areas. soils. Hence bog plants are unusual, and therefore In the British Isles, the lowland bogs have suffered interesting,because they appear able to flourishin from human attention for many hundreds, if not environmentalconditions which are hostile to the thousands of years, and almost all of them have been great majorityof vascular plants. Not merelyare thoroughly degraded, to view them with an ecologist's theyunusual, but to earlier ecologistsmany of them eye, or converted to useful purposes, to take the view- appeared paradoxical,because theyare xeromorphic. 1)oint of a farmer, a peat-merchant or a shooter of This paradox is seen today in more perspective,since gamne-birds. The upland bogs are also for the most it is realizedthat the xeromorphyis exhibitedmainly part in a degraded state, with characteristics very by the Ericaceae; we have progressedbeyond the different from those which they must have shown teleologicalinterpretation implicit in the term"phys- during most of the period of their existence, and an iological drought"to a betterunderstanding of the understanding of the reasons for their degradation metabolicbehavior of the bog Ericads and of other is valuable not only for its own sake, but for its use of the vascular plants with the same habitat prefer- in forecasting the probable fate of the peat areas in ences. As examples of this approach one may take the future. We may, for example, need to know the the work of Priestleyand Hinchliff(1922, 1924) on answers to questions such as these: how far is bog the fat metabolismof the Ericaceae, or of Pearsall degradation a response to climatic change; is it de- and Wray (1927) on the ecology of Eriophorum sirable or possible to bring the bogs back to their angustifolium. natural condition; if peat formation has ceased, what Meanwhile,since the days of Schimperand Warm- new vegetation will appear, or what new plant types ing and "physiologicaldrought" other aspects of bog can we try to grow, and so on. For such an under- ecologyhave becomeprominent. First and foremost, standing, it is clearly desirable to have in the first bog communitiesprovide by far the clearestexamples place the widest possible knowledge of the relation of plant succession,since xeroseresof a truly pri- of bog types to climatic and topographic conditions, mary characterare hard to findand extremelyslow and it was therefore natural for an English investi- moving. Hence, ever since the workof Clements,the gator of bog ecology to seek the widening of experi- successional aspect of bog vegetation has rightly ence which could be given by a first-hand acquaint- claimed a large share of attention. Allied to this, ance with the bogs of at least one locality in the but later in time, has come the historicalstudy of wide area of the North American continent. For bogs by research on the underlyingpeats or lake this reason, and with the general background of deposits. Probably more bogs are today receiving thought which has been briefly sketched, the writer attentionfrom workersin this field than are being undertook a year's study of bogs in the state of studiedby ecologistswith other lines of interest. This Minnesota, from September 1946 to September 1947. is indeed as it should be, since, by a study of the The choice of this area, out of all the possible areas actual historyof development,we may confirm,or, in the U. S. A., was due to the encouragement of if necessary,modify the conclusionsalready reached Dr. W. S. Cooper of the University of Minnesota, from our knowledge of the present-dayecology. and his generous offer of hospitality within his de- Moreoverit is obvious that an understandingof bog partment. stratigraphyis of the highestimportance in the in- This work was carried out during the tenure of a terpretationof the pollen diagramson whichis based Virginia Gildersleeve International Fellowship given knowledgeof the post-glacial historyof vegetation by the American Association of University Women. typesin general. Grants were also received from Girton College, Cam- In recentyears a very large tributaryhas entered bridge, and the Leverhuline Fellowships fund. Grate- the river of knowledgewhich has flowedfrom the ful acknowledgment is made to all these institutions. work on bogs carried out by the ecologists,that is The writer has also to thank the University of Shef- to say, the "pure scientists." This tributary,derived field for granting leave of absence for a year, and from the territoryof "applied" science, might in the University of Minnesota for according the privi- fact be considered,from the point of view of volume leges of an Honorary Fellow. of work,as the main river! Be that as it may,there The help received from taxonomic experts is ac- is no doubtthat at presentecologists and expertson knowledged later. The writer is much indebted to 176 VERONA M. CONWAY Ecological Vo.19,M9oNograph No. 2 Mr. D. A. Christiansonof Hinckleyfor helpingwith Wisconsinglaciation, though as Figure 1 shows,some informationand transportin Pine County,and a great of them are not withinthe limitsof the Late Wis- deal is owed to the generosityof Mr. and Mrs. Borton consin maximum. of Minneapolis,who lent a car for fieldwork during Again, therewas need to use simple techniquesof the summerof 1947. Valuable assistance was re- study,rather than to spend time on the elaboration ceived from Dr. D. B. Lawrence and many other of new or lengthymethods, and thereforethe work membersof the Departmentof Botany in the Uni- (loes not claim to have gone beyond the traditional versity of Minnesota. Above all, the writer must lines of descriptiveecology. It has, however,seemed express her gratitude to Dr. W. S. Cooper, Pro- to the writerthat the data were worth putting on fessor of Botany in the Universityof Minnesota, record, since this would allow comparisonsto be who encouraged her in the firstplace to come to made betweenbogs of the area here describedand Minnesota,and who helped in every possible way those dealt with by Transeau (1903), Gates (1942), throughouther stay. Cooper (1913), and others. SCOPE AND METHODS OF WORK The problemof nomenclaturepresents some diffi- culty to ecologists who are not directlyconcerned Practical considerations,such as the limitedperiod with taxonomicprocedure. They are bound to rely of the research, and the difficultiesof transport, on thehelp of taxonomicexperts, and the writergrate- made it necessaryto choose an area as near as pos- fully recordsthe assistancereceived from Dr. J. W. sible to the laboratoryfacilities of Minneapolis and Moore,Dr. C. 0. Rosendahland Dr. MurrayF. Buell to restrictits size. The boundarymarked on Figure for the vascular plants; fromProfessor W. C. Steere 1 enclosesall the localitiesat whichecological studies for Sphagna; fromDr. C. Christensenfor the larger were made, and any generalizationsabout bog ecol- Basidiomycetes.Apart fromthe commonlyoccurring ogy will refer to this area only. Moreover,studies Angiosperms,of whichmis-identification is unlikely, were strictlylimited to wet habitatswhich were free specificnames which have not been checkedby ex- The abun- fromnany influenceby silt depositions. perts are markedwith an asterisk. The names and in ascribed dance of such habitats the area may be authoritiesgiven in the text are taken from the nature of the of course to the topographical region, followingliterature; exceptions to this are stated in All which has been stronglyaffected by glaciation. the text wherenecessary: the bog localities lie withinthe general area of the Floweringplants and ferns;Moore and Tryon (1946) Sphagna; Le Roy Andrews(1913) Othermosses; Grout (1928) Canada Basiodiomycetes;Graham (1944) Filamentousfungi; Gilman (1945) Soil acidities were almost all estimatedby the quinhydronemethod. This is open to criticismas being less accurate than more modernelectrometric methods,but thereis such a variationof pH reading given by several estimationsfrom one sample of peat of, say, 20 cc. in total volume,that the lack of great accuracy is out-weighedby the advantages Vu Ke of using simple, sturdy and portable apparatus. H.s Further, most of the more important and wide- rangingmeasurements of bog pH values in the Brit- - Wisconsin ish Isles have been made by the quinhydronemethod, moreespecially by Pearsall and his co-workers,so that the use of this methodin Minnesotahas the advan- tage of allowing comparisonwith the British data. Appendix I gives details of the use of the method and the type of resultsobtained. Occasionally,for water samples, a colorimetricmethod was used, the indicatorsbeing held by paper strips. Values so obtainedare qualifiedin the textby the word "colori- metric." The sketch-mapsof bog localitieswhich are shown Scale ? 50 '0miles in Figures 2, 4, and 5, are derivedby enlargement FIG. 1. Map of Minnesota, showing area of study from the relevantaerial photographs,commercially (dot-dash line). Dotted line, boundary of glacial Lake obtainable from the U. S. Departmentof Agricul- Agassiz. Hatched line, boundary of the Young Grey ture. Exact map locations and referencesto aerial drift (after Thiel (1930)). I, Itasca Park; G, Grand will be found in which Rapids; D, Duluth; Cu, Cuyuna; K. Kerrick; H. Hinck- photographs Appendix II, ley.; Ce, Cedar Creek; M, Minneapolis; S, Saint Paul. includes not only all the places mentionedin the April, 1949 THE BOGS OF CENTRAL MINNESOTA 177 seriesfrom one extremeto the other. This conclusion will be illustratedby an outline descriptionof six localities in which the bog vegetationwas studied. In order to simplifythe description,and to empha- size the commonfeatures of thesevaried successions, the vegetationis classifiednot under the headingsof particularassociations, but in terms of four major communitytypes, namely, pioneer-mat,moss-heath, bog-forestand marginal fen. These names are largely sdf-explanatory.The name "marginalfen," however,requires brief commentat this stage. This is the narrow zone which normallyseparates the communitiesof the mineralsoils aroundthe bog from the maincommunities which occupy the bog surfaceas a whole. It is characterizedby abundanceof Alnus incana and of basiphilous sedges, most commonly Carex riparia. Such a zone has been recognizedand describedin the literatureof northAmerican bogs, as for example by Cooper (1913) for Isle Royale, hut has receiveda differentdescriptive name from ScaleB nearly everyauthor. It is clear that it is the exact homologueof the "lagg" of Scandinavian writers; 10 100 20 300 400 50metes an account of the termand its meaningin relation to Britishbogs may be found in Godwin and Con- Open water way (1939). It is essentiallythat part of a bog mat which is permanentlyunder the influenceof the P10/oneer relativelybase-rich waters which drain down the higherground. The zone, . .. slopes of the surrounding / ar9/> cv,01/lanad withits neutralor nearlyneutral peat and its sedge- alder-willowvegetation, presents a physiognomyex- H~01/Ways a r.0",ro0a01s actly similarto that whichin Britainis called "fen," and hence the name "marginal fen" is used here instead of "lagg," which is a term apparentlynot FIG. 2. Sketch-map of Twin Lakes locality, Itasca widely recognizedin the U. S. A., if recognizedat Park. xy, position of transectshown in Fig. 13. all. text, but others which were examined during the Cedar Creek bog coiurseof the research. This bog, which lies withineasy reach of Minne- apolis, has receivedclose studyfrom Lindemann, but OUTLINE OF THE BOG SUCCESSION; THE mainlyfrom the limnologicalaspect. He has, how- DEVELOPMENT OF ACIDITY ever,given a description,most valuable to botanists, Withinthe area studied,there are bogs whichshow of the developmentof the bog as deducedfrom bor- associationsdominated by acidiphilousspecies, more ings along a transect(Lindemann 1941). The pres- particularlyspecies belongingto the Ericaceae, but ent area of open water is only a small fractionof thereare otherbogs in whichthis kind of association the lake which must originallyhave existed. The does not occur. This situationappears to be common greaterpart of this larger area is now occupied by in the northcentral parts of the U. S. A., and there bog-forest,but betweenthis and the lake water lies has been some discussionconcerning both the nomen- a narrow pioneer-matof varying vegetation. Not clature of bog types, and the underlyingfactors only does it vary fromplace to place at present,but which cause the difference.Waterman (1926) for it has varied from time to time in the past. Thus of example,deals withthese questions,and so also does Lindemann has described how the dominance Kurz (1928), who broadly separated his bog types Decodon verticillatus has alternatedwith dominance into "acid" bogs, and "circum-neutral"bogs.1 In of Typha latifolia. During the summerof 1947, centralMinnesota it seemsimpossible to draw a line Typha occupied by far the greaterarea of the two, whichshall even roughlyseparate two types in this but Decodon was growingvery luxuriantlyand cov- way, since there seems to be a continuouslygraded eringthe wholeof the mat!over a sectorat the north 1 An ecologistis apt to refer to soil types as "acid" or "non- end. Typha is not for the most part formingthe acid," in a way which must offend those who are used to a thinkingaccurately in chemical terms. Yet the practice makes youngermargin of the mat, but stands inland to for ease in writing, and ecologists usually understand each dominated Carices, especially C. lasiocarpa other well enough. To make the matter quite clear, however, zone by the followingequivalences will be used throughoutthis paper: and C. prairea. Bryophytes,though locally abun- Acid: pH below 5.0 Moderatelyacid: pH from 5.0 to 5.5 dant, appear as a minor componentof the mat Subneutral: pH from 5.5 to 6.5 Circumneutral: pH from 6.5 to 7.5 vegetation. VERONA Mf. CONWAY Ecological Monograph 178 Vo.19, No. 2 A few isolated saplings and trees of Larix laricina choosing samples from beneath Sphagnum patches, (tamarack) grow on the pioneer-mnat,but it appears and the peat of the bog as a wholeis not acid. It is that colonizationhas not been active in recentyears, thereforenot surprisingthat the marginalfen is not findthe bog-forestmargin is abrupt. The forestcon- conspicuouslydifferentiated in this bog, at any rate sists of mixed Larix and Thu-ja occidentali; in the in those parts of the margin which were examined. older parts the latterforms almost pure stands,which Itasca Park Twin Lakes area are denseand cast an intenseshade. The groundflora is basiphilousin the main,but Sphagnum magellani- South of the end of the S.E. arm of Lake Itasca, CUm,Ledum groenlandicum,and VacciniumOxycoc- beyond Lake Mary, lies a group of small hollows, bog develop- cus are of frequentoccurrence. Table 1 gives the each of whichshows stages of a type of resultsof pH estimations. The values for peat sam- mentwhich is commonto themall. Figure 2 shows ples taken in the bog-forestshow a wide range, but the dispositionof the hollows,which have for con- the mostacid values are obtainedonly by deliberately veniencebeen named with letters. A and B are the "Twin Lakes"; B shows signs of interference,for TABLE 1. Cedar CreekBog. pH values. example by the building of fences and the cutting out of treesfrom the communitywhich now resembles a pioneer-mat. A, on the other hand, gives the im- Place or type Mean pH of sample Comments value pressionof a more natural zonation,while D, which is not visible nor easily accessible from the road, Lake water... Two samples,7.0, 6.88 .6.94 presentsan apparently perfectexample of the suc- cessional stages characteristicof this locality. Fig- Pioneer-mat.. Six samples fromone quadrat. Range 6.30-6.85...... 6.53 ure 3 shows the general appearance of D. C was not examined,but appeared at a distanceto resemble Thuja forest.. Six samples fromone quadrat. D very closely. E and F no longer showed any Range 5.03-7.12...... 6.39 open water, but were covered in the center by the Ditto. Five samples each chosenfrom completed pioneer-mat,which was dominated here just below the surface where by Carex lasiocarpa. With this species were asso- Sphagnum magellanicumwas ciated a fair variety of other Angiospernis,and a growing. Range 4.10-4.48.;.... 4.26 small numberof widelydistributed but inconspicuous Marginal Fen. Two samples,5.62, 5.58 ...... 5.60 Bryophytes. Carex comosa was typical of the youngerpioneer-mats at A and D which were still Mineralsoil... Fromsloping ground juet above invading open water. Just as at Cedar Creek bog, thesite of thepreceding samples. 5.40 there is here no developmentof a moss-heathcom- munity,but Larix laricina directlyinvades the pio-

~~- -

Flu. 3. General view of Twin Lakes D. Foreground,pioneer-mat of Carex lasiocarpa with Typha; beyond the water, pioneer-mat,Larix and Picea zones, w"ithmixed Pine-hardwoodforest on upland in the background. April, 1949 THE BOGS OF CENTRAL MINNESOTA 179 TABLE 2. pH values along a transect across the Lake Alice bog communities,wtih the data for presence of selected species in quadrats at the sites where samples were taken for pH estimation.

Vegetationtype BOG-FOREST ______'WIoss-heath 0 o Old Picea Picea-Larix Distance fromzero of transectin paces (yardsapprox.) ...... 14 27 33 39 45 52 64 87 117 147 177 209 217 223 229 235 241 247 pH values at about 10 cm depth. . 6.8 6.4 7.3 5.5 6.8 4.3 7.1 6.3 4.4 5.0 6.3 6.3 6.0 6.5 5.5 4.8 6.7 6.6 Carexlasiocarpa Ehrh. var. americanaFernald ...... + + + + + + + Calliergonellacuspidata (Hedw.) L oeske.* ...... + + - Aulacomniumpalustre Schwaegr ...... + + + + + + + + + + + + + Salix pedicellarisPursh. var. hypoglaucaFernald ...... + + + + Potentillapalustris (L.) Scop ...... + + +. +...... + Andromedaglaucophylla Link ...... + + + + + + + Betulapumitla L. var. glanduliferaRegel ...... + + + + + + VacciniumOxycoccus L ...... ++ + + + . ... + Sphagnummagellanicum Bird.....+...... +...... + S. recurvumBeauv ...... + .... + + + + + + Ledumgroenlandicum Oeder ...... + + + + + + + + + + + + Chamaedaphnecalyculata (L.) Moench.var. angustifolia (Ait) Rehder...... + + Linnaea borealisL. var. amers- cana (Forbes) Rehder...... + + + + Maianthemumcanadensis Desf + + + + Aralia nudicaulisL ...... + neer-mat. Larix is followed,however, not by Thuja daphne calyculatais presentin the area, it is not this but by Picea mariana, and thereare indicationsthat species, but Andromeda glaucophylla which domi- this in turn may eventuallybe replaced by Abies nates the moss-heath,and that the chief moss of the balsamea. moss-heathis not a species of Sphagnum but Aula- comnium palustre. Lake Alice bog This bog also occursin the Itasca Park region,but Cuyuna bog is possiblyshallower than the Twin Lakes group just About a mile to the east of the re-awakenediron- described,since it carries a large population of mining village of Cuyuna there lies a depression floating-leavedhydrophytes. The pioneer-mat is filledby a lake and the bog communitieswhich are again dominatedby Carex lasiocarpa, but it passes colonizingit. The whole area is of much interest, overinto a well-developedmoss-heath in whichBetula because of the varietyof plant associationswhich is pumila var. glandulifera (the bog birch) is promi- exhibitedthere, but unfortunatelyit is threatened nent. Larix is activelycolonizing the moss-heathout- with destructionas a resultof miningactivities. side the margin of a wide bog-forestzone, and Where the trail fromthe road leads down to the dominatesthe youngerzone of the bog-forest,with bog, as indicatedin Figure 4, the slope is steep, and some admixtureof Picea. The lattertree dominates if the underwaterslope is correspondinglysteep on the older stage, but is itself being invaded by Abies, this radius,it may explain whythe pioneer-matis so as in the Twin Lakes area. narrow at this point. It is dominatedby Carex Table 2 gives the results of pH estimationsfor lasiocarpa,but quicklypasses into an alder-richzone peat samples taken along a line passing across the and inside this occurs a typicallytreacherous mar- vegetationzones from west to east, on the western ginal fenwith much coarse sedge. Roughcolorimetric side of the bog. It also gives the presenceor absence, estimatesof pH along this radius gave nothingmore in quadratslisted at the samplingsites, of a selection acid than pH 5.7. of the more characteristicand importantspecies of Fartheraround, on the west side of the bog, at E, the shrub and herb layers. The pH values suggest the bog zones are very much widenedand much of that sub-surfacepeat is not stronglyacid, but that the area is dominatedby bog birch. An older zone a skin of acidifyingmosses, more especially Aula- of tall birch,farther from the open water,is rather comnium palustre, Sphagnum recurvum and S. clearly distinct from a younger zone of shorter mnagellanicum,has developedextensively, so that peat bushes. Both among the bushes and in the pioneer- samples with pH values below 5.0 are readily en- inat between them and the water, the mixtureof countered. It is worth noting that while Chamae- species is surprising,since it includedseveral char- 180 VERONA M. CONWAY Ecological Mono1raph Vo.1,N.2 be consideredas a "surface skin" such as was de- scribedfor the bog-forestsat Cedar Creek and Lake Alice. This conclusionis based on samples taken with a peat borer at a spot in the Larix zone. The samples indicateda profilewhose main featuresmay be summarizedas follows: Depth from Surface in Feet Nature of Material 0- 6 . Fresh Sphagnum peat 6- 9. More humifiedSphagnum peat 9-11. Peat mainly of Calliergon cordifolium* 11-14.. Mainly Monocotyledonousremains 14-20.. Marl, with organic matterpresent above, but lessening downwards \~~~~~~~~~~0 C Beyond the Larix zone lies a well-definedmarginal fen, dominatedby Alnus incana, with much Carex rostrata and a great variety of basiphilous species. pH values were taken from quadrats extending along the line AB, the distance between the two ex- tremequadrats being about 130m. Six samples were taken from each quadrat, and the mean is here re- corded. Some colorimetricestimations for the lake Sc ale ,,metres water were also made, and the mean is again for six samples. The results were as follows: 100 200 300 400 500 Pu Lake water (colorimetric) ...... 6.7 FIG. 4. Sketch-map of Cuyuna bog. Symbols as in Carex lasiocarpa mat without Sphagna ...... 5.63 Fig. 2. Carex lasiocarpa mat with Sphagnum subsecundum5.45 Moss-heath (Vaccinium macrocarpon dominant) .. 4.72 acteristic membersof the moss-heath,although the Moss-heath (Chamaedaphne calyculata dominant) 4.42 pH values in the region appeared to be high; thus Bog-forest (Larix laricina dominant) ...... 4.39 two samples gave values of 6.07 and 6.15. After Marginal fen ...... 6.03 careful examination,however, it seemed clear that therehad been considerableinterference on this side In the region labelled D, there is a forest type of the bog by the removal of large trees, whose mature enough to show Picea as the dominant, but the stumps were still evident. This would be sufficient zone is narrow and the ground flora is very mixed, to explain the complexityof the area. probably owing to the influenceof the base-rich which lies On the other hand, along the transectAB, there ground waters of the wide marginal fen, behind the Picea zone. This influence is shown occurs a series of vegetationzones which give the by impressionof being almost,if not completely,pri- three pH readings, which gave values of 5.10, 5.40, mary in character. The pioneer-mathas formeda and 5.75. Finally, at C there is a vegetation type which is bridge across the center of the original lake; it is almost certainly secondary in nature, the modification dominated by Carex lasiocarpa, though Typha occurs induced the removal of trees from the fairly frequentlyin the mat, and is especially abun- being by orig- dant close to the water margin around the smaller inal bog-forest zone at that point. The effect has allow the moss-heath to to an extreme daughter-lake. The pioneer-matis succeeded by a been to develop density, with deep cushions of Sphagnum, especially moss-heath, and Sphagnum subsecundum, which ap- with S. as a pears here as the earliestmoss-heath invader, can be S. magellanicum, capillaceum frequent and a abun- found widely distributedat the base of the Carex dominant of the cushion summit, great dance of strictum. is iasiocarpa shoots in the older parts of the pioneer- Polytrichum Chamaedaphne are mat. At the timeof study,it was veryinconspicuous, dense and the other Ericads inconspicuous or since it was growingcompletely submerged. The most absent. advanced Ericaceous memberof the moss-heathis Pug Hole bog Vaccinium macrocarpon, which forms a conspicuous Pug Hole Lake lies alongside Highway 38, north zone in association with Sphagnum magellanicum, S. of Grand Rapids, and its southern end has been recurvum and other Sphagna. Chamaedaphne be- overgrown by an extensive development of bog. Only comes dominanta little way behind the Vaccinium, the southernmost end of the open water of the lake and this zone is itself being actively invaded by is indicated in Figure 5. By far the larger area of Larix. The bog-forestzone is wide here,with Larix the bog is moss-heath; bog-forest has not yet devel- still the sole dominant. The ground flora is domi- oped to a great extent, and the dotted line MN nated by Sphagna, and it is clear that the Sphag- roughlyindicates the northwardlimit of the scattered num dominationhas been long continued,and cannot tamaracks which are the bog-forest pioneers. 181 April, 1949 THE BOGS OF CENTRAL MINNESOTA to consider the moss-heathas secondary in nature, and in fact the general succession exhibited is ex- actly that which was seen along the transect AB at Cuyuna. Figure 6 shows some data from a transeet across part of Pug Hole bog, the line AB in Figure 5 showingits position. In collectingsamples for pH estimations,the peat was taken, as far as possible, from the sub-surface layers, say at about 10 cm. depth; in Sphagnum-richareas they were taken in the dead moss layer below the living surface. It was much harder to penetrate the tough surface of the sedge-mat peat, and therefore these samples were more superficial,though since the peat surface was submerged,it is unlikely that the readings would differgreatly' from those of samples 10 cm. deeper. The' nature of the plant communitiesis illustratedby the presence or absence data for selected species, taken from the records of listed quadrats extending along the transect. The quadrats were worked on a differentoccasion, and hence do not coincide with the sites of pH samples. The gradual developmentof acid conditions may clearly be seen in Figure 6, and is associated with the rise in importance of Sphagnum magellanicum and other mosses in lesser quantities,until the field- layer is formed entirelyof mosses. Chamaedaphne does not reach dominanceuntil the Sphagnum cover is firmlyestablished. The higher pH values of the marginal fen are shown also, but the abrupt transitionis betterillus- I 1 trated by the results shown in Figure 7, which gives Scal e pH 'values taken along the transect PQ, at the 100 200 300 400 500metre southernmostend of the bog, where the alder-fenis clearly definedand characteristic. FIG. 5. Sketch-mapof Pug Hole bog. Symbolsas in Kerrick bogs in Fig. 2. AB and PQ, positionsof transectsshown is a secluded spot, but easily acces- Figs. 6 and 7, respectively. MN, approximateouter Lake Margaret boundaryof area of tree colonization. sible from the village of Kerrick, and the surround- ing bog is typical of many which may be found in The pioneer-matis formed by Carex lasiocarpa, Minnesota at the presenttime. It is characterizedby and as at Cuyuna, Sphagnum subsecundumis abun- the fact that the youngest bog stage consists of a dant in the older parts of the mat. In June, 1947, floatingmoss-heath. Along parts of the margin of the younger parts of the mat were submergedto a Lake Margaret, this mat is very narrow, as can be depthof about 3 feet. The moss-heathis everywhere seen in Figure 8. Elsewhere it is wider, but it has dominatedby Chamaedaphne,which is thick and lux- everywhere the same major components, namely uriant, especially in the area north of the upland Chamaedaphneand Sphagnum recurvum. Vaccinium "islands" which stand out of the center of the bog. macrocarpon and V. Oxycoccus are also frequent. The appearance somewhat suggests that there may The most frequently occurring sedges are Carex have been some burning over the bog surface, but paupercula and C. oligosperma. C. lasiocarpa if thereis also a strongsuggestion that the bog surface present at all is not conspicuous, and no inflores- has been much drier at some time in the recent past cences of this species were found. than it is at present. For instance, right at the In a number of places at the very water edge, watermargin of the pioneer-matis a fringe of bog there were occasional shoots of Typha and of Carex birchand other shrubs,which are not likely to have rostrata. The interestof this lies in the comparison begun their development there if the mat was as with another bog about a mile to the south, which deeply submerged as it is at present. Again, the for convenienceis here called-Kerrick B. Here, in moss-heathis scatteredwith a numberof dead young walking out across the bog from the upland margin treesof Populus tremuloidesand a few also of Pinus toward the open water, one crosses first of all a strobus. moss-heathof dense Chamaedaphnewith much Poly- Nevertheless,in spite of the obvious effectsof trichumstrictum--an indicator of the drying out of recent drying (doubtless the work of the "dry a bog surface-then across a sharp transitioninto a thirties"),and possibly of burning,there is no reason zone of Carex rostrata, twenty meters or more in 182 VERONA M. CONWAY Ecological Monographs Vo.19, No. 2

Carer $ctiof. Ca/,erarostls car9erxs-s. Nsernf/uvv/ot//e ......

Sp/rnyw~mJc/Asecuz~dum ...... S rP7&,9e//a/~,curn ...... Io/M,o,&//e/7 . S5oly, l-9num recLrvum ......

Chamlaedaphne ca/ycu/aea , . . Andonooedag/suc/paz//al.q S2ahc//m ca/prceun ... 4r0.era2 rotuodbzfo/,X'. 9o/Zeryzchumcomnloe Calrex r moaria

Pponeer- mat Immature , moss -heath CommnnLn/ty1"a ufe mosse-sheeatht

0- , Tralvltlbn 1

, /values I

DIstance In paceS from zero of transeCt FIG. 6. pH values and occurrences of species in quadrats along the transect AB on Pug Hole bog.

width, and finally to the youngest bog stage domi- nated by Typha latifolia. It therefore seems quite possible that the Lake Margaret bog may develop in Moss- i MAlargina/the same way as that at Kerrick B. The bog-forests at Lake Margaret are dominated by Picea mariana, and in the area lying westward of Trarjto ;ln the S.W. corner of the lake, there is a mature stand of pure spruce which does not seem to have suffered from any removal of timber. The light intensity inside this stand is low, and the ground flora cor- respondingly poor. Larix is a relatively unimportant component, though it occurs throughout the fringe of bog-forest nearest the lake; there is no Larix zone, and in general, there are no early stages of the bog- forest to be found here. The marginal fen is well shown, especially by the stretch of bog at the S.E. corner of the lake; on the west side it is being invaded and in places obliterated by the more acidiphilous communities which are pushing outwards to the surrounding upland. In a patch of bog which stretches south from the lake towards Kerrick, there has been removal of timber, leaving a secondary moss-heath in which 40 I I Ledgm is the most prominent species. Determinations of pH were made as at Cuyuna, by taking six peat samples from each of a number of I 20 40 quadrats, and two colorimetric determinations of the Distance in braces lake-water were made. The results were as follows: L pH FIG. 7. values the transect on Hole pH along PQ Pug 1. Lake water (colorimnetric).Mean of 2 samples .. 5.0 bog. 2. Moss-heath. Mean of 6 samples ...... 4.30 3. Picea forest. Mean of 6 samples ...... 4.14 April, 1949 THE Boos OF CENTRAL MINNESOTA 183 4. Marginal fen. Mean of 6 samples ...... 5.03 DISCU8so0N 5. Mineral soil. One sample ...... 5.25 There is clearly a great differencebetween Cedar Mean of 6 samples ...... 4.22 6. Old Picea forest. Creek bog and Lake Margaret bog, yet the other 7. Secondary moss-heath. Mean of 6 samples .... 4.02 examples form a bridging series from one to the Sites 1 to 5 lay on a line running roughly S.E. from other. It appears also that a successionfrom open the lake margin to the upland at the S.E. corner of waterto bog-forestmnay occur, without involving any the lake. successionalstage in whichacid peat is formed,al- a "tend- These figures make it reasonable to call this an thoughone may recognize,to speak loosely, so thateven "acid bog," and, as in the Cuyuna Larix zone, the ency" towardthe developmentof acidity, to find Sphagnum Sphagnum of the spruce forestis not a thin skin, at Cedar Creek it is possible but extendsdeeply. Samples were taken witha peat inagellanicumin the Thuja forest. borer near the site of quadrat 3 in the series de- Nevertheless,the bogs may be consideredas all scribedabove, and the peat profileso revealed may belongingessentially to the same type,if one recog- be summarizedthus: nizes the factors which encourage the development of acidity,and realizes that these factors,singly or Depth from in combination,may vary in intensity. It matters Surface in Feet Nature of Material little whetheror not this idea is accepted, since it 0- 5..... Fresh peat Sphagnum is a way of lookingat the subject,but on the 5- 9..... Sphagnum present, but Monocotyledon- merely ous remains frequent other hand there is some ecological importancein 9-17.... Highly humifiedpeat passing down into correctlyrecognizing the factors just mentioned. organic mud with an increasing silt Thus, there are at least four conditionswhich en- content courage the establishmentof plants whichform acid Below 18 .... Sandy silt peat, and the subsequentspread of acidiphilouscom- munities. These conditionsare as follows: has played an importantpart Evidently Sphagnum (1) Whenithe Waters of the lake are relativelypoor at this locality. in the succession id bases. If a bog successionbegins in a lake whose waters AIL~~~~ are already fairly acid, the entry of acidiphilous phases of the successionwill clearly be more prob- able. At Kerrick the lake-waterhad a pH of 5.0, in contrastwith the value of almost 7.0 at Cedar __~~~~~~~~~~~~~~~~~~0 Creek, and correspondinglythe Kerrick succession has been imarkedlyacidiphilous in character. Another bog, Lesser Tamarack,was of the same generaltype as Kerrick,and gave a pH value of 3.5 for its open water. The base-contentand pH of any small lake will be determinedby a complexof factors. It will de- pend in part on the chemicalcharacters of the soil ~ ^ and rock fromtwhich the drainage wateris derived; in part on the rate of inflowand outflowof water, an(l the relation of this to the total volumeof the lake wnter; in part on the ratio betweenthe water receivedlfrom drainage inflowand that derived di- fi rectlyfrom precipitation. Thus forexample. one may contrast the state of Cedar Creek lake, which is swall in volume mndhas both inflowand outflows, with Pug Hole lake, wvhichis large in area, and has no outflow. It would he of Imuchinterest to collect statistical data concerningon the one hand these topographicalfeatures of the lakes and the resulting pH values of the water,and on the other,the rela- tion of the pH values to the associatedltypes of bog succession. Some veryinteresting results given by Thiel (1930) p)rovidlesonei inforimmationconcernimmg the variability of lake waters in relation to their calciumimcontent and to otherfactors whivlcldetermine wihetimer or not the calciumshall be depositedas marl. Thmielrives Fi. 8. Part of the water margin at Kerriek (Lake a map of Minnesotashowing all the localitiesfromm showing narrow zole of moss-heatlh,with Margaret), beds. The niap shows two Chanaedaphne at the water edge. which he recorded umarl 184184 VERONA AI.M. CONWAY Ecological~~~~~~~~~~~~~~~~~~~Vol. Monographs19, No. 2 egions of maximumdensity of such localities,the Creek bog gives a quantitativeillustration of an irst lying across the proximalend of the Grantsberg actual fluctuationin water level in thisregion. They sublobe of the young Grey drift (calcareous), the traced the recoveryfrom a minimumlevel in the fall second lying northand north-westof Mille-Lacs on of 1934, the level in the fall of 1939 being over two the youngRed drift (relativelynon-calcareous). The feethigher. second area of concentrationseems anomalous and It is worthconsidering the figuresfor total annual Thiel brings forwardother data and offersexplana- precipitationfor Minnesota since 1886; these are tory hypotheseswhich need not be discussed here. graphedin Figure 9. From themit is clear that the The main point which is relevantis that the lime- "dry thirties"were only the culminationof a long contentof the lake-deposits,and hence,probably, the downward trend of average precipitationlevel, a pH of the lake-water,is not simply related to the trendwhich started about 1905. This long period of nature of the glacial depositsimmediately surround- falling average must have been widely reflectedin ing the lake basin. Such a conclusionis of course graduallysinking water tables in manyof the smaller what would be expectedon the basis of the premises lakes of the region,and it is easy to see how the stated in the precedingparagraph. moss-heathspecies could have increasedtheir rate of (2) When a pioneer-mat is large in area. spread into the dryingand hence increasinglyacid pioneer-mats,until the latter were completelyoblit- and This point again would require quantitative erated. The normal types of pioneer-mathave in statisticaltreatment to make it more than a sugges- most cases not yet re-establishedthemselves in spite to tion, but it certainlyseems the easiest hypothesis of the rise in precipitationrates since 1935, and this explain,for instance,why at Cuyuna thereshould be may be taken as an example of the inevitabletime- acidiphilous communitieson some radii and not on lag between change of conditionsand vegetational others. A study of this bog stronglysuggests that response. It looks very much as thoughKerrick B and where the pioneer mat has spread rapidly is is an exceptionto the generalizationjust stated,since wide, the centralparts of it have been shieldedfrom it shows such an abrupt transitionfrom old moss- the directinfluence of the lake water,but that where heath to Carex rostrata and Typha zones,but unfor- the mat is narrow,the acid-formingSphagna have tunatelyno data are at hand to confirmthis idea. been excludedby the free access of water witha pH above 6.0. It seems likely that the extensivemoss- Precip? heath of Pug Hole bog has been encouragedin the same way. When the mat has (3) partial drying of pioneer 2S5 occurred. The surface of a pioneer-mattends to rise higher 2030 than the watertable level by the naturalprocesses of _ I peat accumulation,but it may suffera relativerise I I I 890 190 1920 930 1940 in much less time if the water level falls for any 1900 reason. To an ecologist, these bogs all bear the FIG. 9. Average annual precipitationvalues for Min- strongestimpression of fluctuatingwater levels im- nesota from 1886 to 1946 (from the U.S.D.A. Weather posed on the generalsuccessional development of the Bureau reports). vegetation. This will be discussed in its broader aspects later in this paper, but the relevant point (4) When trees have been removed in the early here is the application of this idea to explain the bog-forest stage. frequentoccurrence of the Kerrick (Lake Margaret) Sphagnum and Ericaceous species (especially bog type,where the Chamaedaphne associationforms Ledum), are normalcomponents of the groundflora the vegetationzone nearestthe water. In thesebogs of the bog-forest,and the removal of the shade of thereis no evidencethat Chamaedaphne can act as a the trees encouragesboth these groups to a marked pioneer-matformer; in fact, the youngestshoots of degree. Thus the most extensiveand dense areas of the plant are always found away from the water Chamaedaphne and its associates are usually found margin. Gates (1942) states that "the association alongsideroads and highwayswhere burning or tim- does not readily extendout into open water because ber removalhave obviouslyoccurred. It is in areas the plants of Chamaedaphne are slightlyheavier than such as these that peat samples with the lowest pH water,consequently must be held up." values are found,as for example at Kerrick (mean Pearsall (1938) has described for British peats value 4.02 in the secondarymoss-heath) and Cuyuna the acidifyingeffects of dryingand consequentoxida- (values of 4.20 and 4.08 in the Sphagnuum peat of tion,and the phenomenonis of course generallyrec- the regionlabelled C in Figure 4). ognized. It is, therefore,only reasonableto seek an Presumablythis factoris effectiveonly in increas- explanationof these marginal Chamaedaphne zones ing acidity and not in initiatingthe change from by consideringthem as relies fromthe dry years of circum-or sub-neutralconditions to acid conditions. the nineteen-thirties,when water table levels must In other words, acidiphilous species must already have fallen over the wholeregion. have been establishedbefore they can be encouraged The work of Buell and Buell (1941) at Cedar by increased light intensity. Thus, for instance,in CENTRAL MINNESOTA 185 April, 1949 THE BOGS OF many parts of the Cedar Creek Thuja forestwhere case in many of the bog areas of the northeastern the groundflora is entirelybasiphilous, it is doubtful, UT.S.A., and hence the name "'sedge-mat" is frequently to say the least, whether Chamaedaphne,or even used in the literature as a general name for the Ledum,would becomedominant on the cuttingout of earliest bog-forming stage. Since, however, what the Thuja. This type of ground flora in the bog- may be called the "functional importance" of the forestsis, however.probably less commonin central pioneer-mat in the hydrosere is the same whatever Minnesota,so that,broadly speaking, it is legitimate the dominant species, it seems useful to use a more to considertree removalas a factorwhich promotes elastic name in general discussions of the successions. the accumulationof acid peat. The phenomenonis One of the less comimon pioneer-mat formers in of someinterest in comparisonwith the parallel cases our area is Decodon verticillatus. Where it flourishes, in Europe whereremoval of forestcover has led to it forms a dense tall growth and other species form the dominance of Calluna vulgaris. a very small fraction of the total mass of vegetation. Having now describedthe factorswhich appear to The best examples of a Decodon mat were seen at be mainly operative in the central Minnesota area, Stacey A and B, where there were complete zones thework of Kurz (1928) on certainbogs in Michigan surrounding the lakes. At Cedar Creek bog the must be considered. His data are concernedmainly Decodon is only dominant in one part of the bog, withthe part played by Sphagnumin bringingabout and appears to be recovering from a phase during acid conditions,but his discussiondeals with rather which it has not been important. wideraspects of bog ecology. Later on in this paper Typha latifolia is the dominant of an extensive an accountwill be given of the resultson pH ranges mat at Kerrick B, and is probably colonizing open of Sphagnum in the Minnesota bogs; these results water in places at other localities, for example Cedar fall in exactly with those of Kurz. Moreover,the Creek, Floating Bog (Lake Itasca) and Cuyuna. writer concurs in the general discussion given by Carex prairea is abundant in the mat at Cedar Creek, Kurz, as far as he goes. He says "the relationbe- and appears to be the dominant of the mat at Grand tween the initial reactionof the free water and of Rapids C. the mat to the establishmentof Sphagnum is still No details were studied for the minor types of moot and needs more data." The presentdiscussion pioneer-mat just mentioned, but the Carex lasiocarpa has aimed at pushing analysis a little fartherthan mnatswere more closely studied, and Table 3 gives has been attemptedby Kurz. frequency estimates obtained from quadrat lists. Al- The fourthfactor described above may be omitted, together 50 species were recorded from the 48 quad- as being in a differentcategory from the otherthree, rats, but only 35 are given in the table, which since it only encourages and does not initiate the excludes any species with an average percentage of developmentof acid peat. The firstthree factors, less than 10, unless it had a percentage of over 20 however,may really be consideredas exhibitingdif- in some one locality. The species listed are divided ferentaspects of one general principle. This prin- into the three groups A, B, and C, on the basis of ciple is that while Sphagnum may begin to grow in general experience of Carex lasiocarpao mats in this a mediumwhich has a pH higherthan the optimum region as a whole. Within each group they are for the species concerned,it cannotbecome dominant placed in the order of their total frequency. unless protectedby some means from the continual It is worth noting that the total number of species access of water with a pH value much above the is much larger than that recorded by Woollett, Dean optimum. In otherwords, either the lake watermust and Coburn (1925) for the sedge-mat of Smith's bog, have a relativelylow pH value to start with,or it Michigan, where from over 1,000 quadrats they listed must be kept out of contactwith the incipientcol- only 29 species. It seems likely, since their data onies of acidifyingSphagnum, whether by the barrier showed Dulichium arunidinaceum with a frequency of of a large extentof pioneer-mat,or by the sinking 37 percent, that their sedge-miat was much younger of the watertable to a sufficientdepth below the peat than most of those which are here described. The surface. only young stages in the Minnesota area-as judged THE BOG COMMUNITIES by water-table level-were those at Cuyuna and Pug Hole. Even these were not widely different from THE PIONEER-MAT the rest in floristic composition, and it is the general As thename itself implies, it coversthe associations impression of homogeneity of these sedge-mats formingthe floatingmat which is the active agent throughout the area, which justifies the presentation hydro- in the process of "verlandung"in these lake of an average percentage value in the last column which seres. There are of course,plant communities of Table 3. lake water before any floatingmat is colonize the In addition to this floristic uniformity, there is formed,for instance, associations of Potamogeton nearly always a fairly abrupt transition between the and Nymnphaeaspecies, but it may well be doubted and the older seral stage which adjoins whethersuch associations are essential membersof pioneer-miiat support the idea that the hydrosere. it. These two observations Of the various types of pioneer-matobserved in favorable conditions for mat growth are not always these studies,the one dominatedby Carex lasiocarpa present, but that when they do occur, growth may is by far the most common. This appears to be the be rapid for a time, and then cease altogether for VERONA CONWAY Ecological Monographs 186 M. Vol. 19, No. 2 TABLE 3. Frequency data for the species of Carex lasiocarpa mats.

Twin Float- Twin Locality ...... Pug Lakes ing Lakes Total Hole A Bog F

Number of quadrats ...... 8 10 20 10 48

Carex lasiocarpa Ehrh. var. americana Fernald ...... 100 100 100 100 100 Lycopus uniflorusMi chx ..70 90 70 57.5 Menyanthes trijoliata L. var. minor Raf .. 80 100 40 55 Aster junciformis Rydb ...... 40 40 20 60 40 Potentilla palustris (L.) Scop ...... 12 90 10 28 . Equisetumfluviatile L ...... 100 .. .. 10 27.5 bD Calamagrostis canadensis (Michx.) Beauv ...... 40 70 .. 27.5 < Salix pedicellaris Pursh. var. hypoglauca Fernald ...... 10 90 .. 25 Ca>, Scutellaria epilobifolia A. Hamilton ...... 50 .. 20 17.5 0 g Calliergon cordifolium*(Hedw.) Kindb ...... 40 .. 30 17.5 ? Mniumrostratum* Schrad ...... 60 .. .. 15 Hypericum virginicumL. var. Fraseri (Spach) Fernald ...... 50 5 .. 14 Typha latifolia L ...... 30 .. 10 10 Carexcomosa Boott ...... 30 .. .. 7.5

X; Sagittarialatifolia Willd ...... 50 .. .. 12.5 $MUtricularia intermediaHayne ...... 40 ...... 10 D a3h0 Dulichium arundinaceum (L.) Britton...... 25 10 .. .. 9

Drepanocladus aduncus (Hedw.) Warnst. var. Kneiffii(Bry. Eur.) Monkem. *. . 80 85 90 64 Campanula uliginosa Rydb .. 80 75 50 51 Bryum bimum* Schreb .. 30 75 90 49 Lysimachiathyrsiflora L .. 60 55 70 46 Galium triftdumL ...... 80 .. 100 45 ' Drepanocladus aduncus (Hedw.) Warnst. var. typicus (Ren.) Wynne.* ...... 40 45 80 41 < a3 Viola pallens(Banks) Brainerd ...... 70 .. 70 35 X Epilobium leptophyllumRaf ...... 10 45 70 31 I. Impatiensbiflora Walt ...... 70 30 20 30 _ Cicuta bulbiferaL ...... 70 25 20 29 _ Triglochinmaritima L ...... 40 75 .. 29 0 Muehlenbergia racemosa (Michx.) B.S.P ...... 10 25 50 21 0 Agrostis hyemalis (Walt.) B.S.P ...... 30 .. 40 17.5 o Scheuchzeria palustris L. var. americana Fernald ...... 10 50 .. 15 Z Drepanocladus vernicosus*(Lindb.) Warnst ...... 10 40 .. 12.5 V Bidenscernua L ...... 40 .. .. 10 Dryopteristhelypteris (L.) A. Gray var. pubescens(Lawson) ...... 30 .. .. 7.5 another period of years, until favorable conditions which have been discussed already. Another very occur again. good example may be seen at Grand Rapids B, just east of Pug Hole. The moss-heath at the latter place THE MOSS-HEATH may have sufferedslight burning, and is of a thicker A Chamaedaphne association is the most common and drier type with much Polytrichum strictumr.This type of moss-heath in the area, and the chief mosses species is a normal component of the moss-heath, but accompanying the Chamaedaphne are always, as far remains in small quantity, unless the bog surface as observation went, Sphagnum recurvum and S. suffersburning or considerable drying, in which case magellanicumn. However, as has been described al- Polytrichum strictum becomes very conspicuous. Les- ready, Andromeda and Aulacomium palustre were the ser Tamarack bog is of the same general type as main components of the Lake Alice mnoss-heath,and Lake Margaret, but there is some evidence that trees Vacciniurn macrocarpon with Sphagnutm subsecundurm have been removed and that the rather extensive formed the younger zone of the moss-heath at moss-heath is secondary in nature. Cuyuna. Table 4 shows a clear justification for the name Table 4 gives frequency data from quadrat lists "moss-heath," since, with only one exception, every taken in Chamaedaphne-Sphagnum associations at species with an average frequency over 20 percent is three differentlocalities, and though the three were either a heath or a moss. One qualification must be not precisely similar, it is convenient to show the made, however, which is that Carex average percentages, since these do express the gen- lasiocarpa may eral floristic character of the Chamaedaphne asso- occur, as a relic of the pioneer-mat stage; it occurs ciation of this region. The Kerrick moss-heath does in fact with an average frequency of 32 percent. not appear to have suffered froni burning, and is a This and other relics of the pioneer-mat are not type of common occurrence at present, for reasons included in Table 3. Apart from Carex lasiocarpa April, 1949 THE BOGS OF CENTRAL MINNESOTA 187 TABLE 4. Frequencydata for the species of moss- THE BOG-FOREST heaths. A superficialsurvey of a large numberof bog- forestsmight lead to the conclusionthat they show Kerrick LAN" Llcality...... (Lake Pug Tama- Total markedvariations in type. Thus, for example, the Margaret) Hole rack maturestage may be dominatedeither by Thuja or Numberof quatrat.. 50 38 20 108 Picea; the ground flora may presenta considerable varietyof aspect and the indicationsof finalclimax ChamaedapAnecalyculda (L.) Moench. var.anguatifolia (Ait) Rehder...... 92 95 100 94 stages may differfrom each other. Sphagnumrecuruum Beauv ...... 98 68 100 88 These variationsdo indeed exist, but just as with 8. magcUanicumBrid ...... 30 95 80 62 the successionsas a whole,so with the bog-forests, Polytrichum#tridtum Banks...... 14 97 100 59 VacciniumOzycoccus L...... 65t 39 10 54 it is impossibleto classifythem into separate types, Aulacomniumpalusre Sahwe u ...... 68 55 34 for they grade into one anotherin such a way as Andromedaglaucophula Link...... 12 45 50 31 to defy classification.It is, therefore,easier to dis- Cares oligoepermaMich...... 56 .. .. 26 cuss the characteristicsof the bog-forestconsidered Polptrichumcommune L...... 36 18 10 25 Kalmiapolifolia Wang ...... 28 16 5 19 as a unityrather than as a numberof discretetypes, Cares pauperculaMichx ...... 36 .. .. 17 though for conveniencethe differentstages of ma- Spraeatomenlosa L. var. romea (Raf.) turitymay be consideredunder separate headings. Fernald...... 2 .. 20 17 Sphagnumcapiliaceum (Weis) Schrank . 2 23 35 16 (1) Invasion stoges EriophorumPirginicum L ...... 26 5 13 Good examples of forestinvasion may be seen at Bduk pumilaL. var.glandulifera Regel ... 26 .. 9 Droeerarotundifolia L...... 50 9 Pug Hole, Cuyuna, Lake Alice and Foating bog Ledumgroenlandicum Oeder ...... 12 5 .. 7 (Lake Itasca). The successful invader is always CaUa palustri L.16 .7 Larix laricina. At first,only a few of the invading Pleurosiumschreberi (Brid.) Mitt. ... . 5 30 7 seedlingssurvive to maturity,and theseisolated indi- CaUiergoneiramineum (Dickl.) Kindb. 12 .. .. 6 Vacciniummacrocoarpon Ait 81...... t 5 viduals long remain conspicuous. The Lake Alice Drwopterispinuloaa (0. F. Mueller) bog illustratesthis point very well. Later, however, Watt...... 2 8 5 5 thereseems to be a "mass invasion,"and the southern 2 SarraceniapurpureaL2 ...... 2 3 . end of the Pug Hole moss-heathseems to be at just this stage at present. Nevertheless,even an invasion tOutof 20 quadrats of this kind does not guarantee that bog-forestis finallyestablished, for it is possible to find places none of themhas a frequencyabove 15 percent. where a whole zone of younigtainaracks has died. The low frequencyof Vacciniummacrocarpon may Such a case is illustratedin Figure 10, whichis taken seem surprising,since it is well known as being in the marginalzone of the bog-forestof Floating highlycharacteristic of acid bogs. The explanation bog. The trees,which are here showndead, forma probablylies in the fact that it is a pioneer species small part of a zone of invading tamuaracks,which of the advancing moss-heath,and becomes crowded in 1939 were alive and flourishing,and fringedthe out in the maturestages, so that random sampling of a large moss-heatharea is not likelyto give high frequenciesfor this species. Betula pumila var. glanduliferais included as a memberof the moss-heathcommunity, although its ecological status does not seem clearly definableat present. It may in some cases forma zone in which it is itselfdominant; such a case was seen at Cuyuna, but therewas every appearance here that the domi- nance of the Betula followedsecondarily after the removalof bog-foresttrees. Gates (1942) recognizes it as a dominantof a "High Bog Shrub" association in Michiganbogs, but in the Minnesotaarea it accords betterwith general observations to retainit as a minor associate of both the moss-heathand the bog-forest stages. Lastly, we may note the small numberof species with frequenciesof 25 percentor over-9 are given in Table 4, and Carex lasiocarpa would make the tenth. This figure,compared with that of 19 from the sedge-matdata of Table 3, must be a reflection of the relativelyhigh aciditiesof the surfacepeat in these communities,which exclude all but the small Fw;. 10. Zoneof dead taniar:acksbetween the bog for- numberof acid tolerantspecies. est and the lioIneer-mat at Floating bog, Itasea Park. 188 VERONAVERONA Al.M. CONWAY Ecological Monographs ~~~~~~~~Vol.19, No.2 bog-forestnearly all the way round. In 1947 all the trees of this zone were dead. This can almost cer- tainly be attributedto risingwater levels in the bog, |ti3la C f thoughthe data needed for proof are not available. var 2 7: Very much the same can be seen at Cuyuna in the wt - - moss-heathoutside the Larix forest. Here a new -4 -tT I}f invasion has already begun, as shown by the quite young saplings (up to 8 feet high) which are fre- quent among the taller dead trees. 4 ~~~~~7' In spite of such set-backshowever, the establish- mnentof bog-forestappears to depend solely on the /a ricn i I t ultimatesuccess of Larix in the invasionzone. Thus, althoughhuge numbersof Thuja seedlingsare found inside a Thuja forest,during the whole of a year's fieldwork only one seedlingof Thuja was ever seen outside the margin of a bog-forest(and this one, Picea~~~~~7 oddly enough,was at Pug Hole, where there is no ,778r~ana matureThuja forestin the immediatesurroundings). Again, whilePicea seedlingsand saplings may some- times be found on moss heathsor pioneer-matsout- side the bog-forest,no instance was observed in * 60 which this species appeared to have established,or to be establishing,a primarybog-forest without the Is/samrea interpositionof a Larix phase. Apparent exceptions to this may be seen where Picea is regeneratingin __i~t*fttt~i an area of secondarymoss-heath. This differencebetween Picea and Larix in their FIG. 11. Diagram to show the data collected at Pug capacityfor invasionmay be tentativelyexplained in Hole bog for ten seedlings of each of four species. Each termsof two physiologicaldifferences, shown by the drawing shows the total length of root and shoot and two species in the early stages of the life-history. the relationto the soil surface (horizontalare) and water One is that the root systemsof Larix table (w.t.). Where the root ends in an interrupted seedlingsare line, this indicates that only part of the root system tolerant of a greater degree of submergencethan was dug out of the ground. The horizontalhatching of those of Picea seedlings. This differenceis indicated the root axis represents the depth to which the roots by the data in Figure 11, thoughobviously the latter were still alive. The numbersgive the age of the seed- are statisticallyinadequate. General ling as derived from ring-counts. Vertical scale 1/40 observationsug- natural size. gests that such a differencecould be experimentally demonstrated. The other feature is a differencein growth-ratewhich may easily be deducedby examin- (2) The mature bog-forest ing the annual rings in stems of young Picea and While it seems that Larix never remains perma- Larix of the same height,or by comparingthe ring- nently as the bog-forest dominant, it may be a long widthsin stemsof the same diameter. Thus to take time before it is finally ousted by either Picea or an example, a Larix tree of 14 feet in heightwas Thuja. It is generally stated that Thuja will not 17 years old; a Picea tree of 17 feet in heightwas grow on acid peat, and this certainly holds good for 137 years old. central Minnesota, where the species was never found on a The results expressed graphically in Figure 11 bog where there was evidence that a moss-heath stage might have occurred. On the other the were obtained in the moss-heathzone at Pug Hole, hand, behavior of Picea means the in the region of active forest invasion. One may is not by any converse of that of Thuja, since Picea forests were common conclude from these data that the roots of the bog in bogs without any trace of previous acid birch may remain alive at a much greaterdepth of stages, as, for example, at and at in Itasca submergence,and that its shoots show a more rapid Hinckley places Park. Moreover, Picea and growth in height in the first4 years, when com- Thuja sometimes occur as pared with the main bog-forestspecies; it also codominants. appears that Abies seedlings make slower growth On the whole, Picea forests seemed to be of much than those of Picea and Larix, and that their tap- more frequent occurrence than Thuja forests, and for roots are more deeply penetrating. The older ones this reason it was easier to find examples which among theseAbies saplings probablygerminated dur- showed the successional history of the forest. Two ing the period of lowerwater table already discussed, studies were made in the Itasca Park region, to pro- and are now sufferingfrom the killingof large parts vide illustrations of this "historical structure." of theirroot systemsby risingwater levels. Signif- The first was carried out at the Lake Alice bog, icantly also, their shoots in many cases are being and the results are shown in Figure 12. The pro- smotheredby a luxuriantgrowth of Sphagnum. cedure was to mark out a line from the pioneer-mat April, 1949 THE BOGS OF CENTRAL IINNESOTA 189

0 20 4080 0 0 00 20 140 60 80 200 220 240 DIscance Inpaces from zero of transect Bog forest margin WI Iter margin

G~rths Inchescnas Abiestrees 201

0 20

Chs PRceatrees | nGcer

I 20] I 540 I~ ~~~~~~~~~~~~~~~~~inchesl~lly ~~ Irimsr~dll, ~~~~ ~~~~~~~~~~~l q

f'icea.. sapl~ings IIIIIIII1III 11111 11111 I111 I 11 I 1111111111 1 11111111 I 1IiI11 III 11 11111 I I I . ~I &irthis I Larx trees incheS CnV 540 I

-20

1n111E~~~~~~~~~~~~~11 1 11 ~l11 b 1 1 iI1 I I J11 I Larixsaplings I ii FIG. 12. Results of a transect made across the bog communitiesat the Lake Alice bog. The girth at breast height is recordedfor all the trees within each strip of five paces in length and two paces in width, along the length of the transect. Any youag tree between 6 inches and 6 feet high is recorded as a sapling. throughthe forestzones to the bog marginand then Figure 13 gives the second set of data, taken this to work along the line measuringgirths of trees (at time fromTwin Lakes F. The successionalstory is breast height) and noting saplings, for consecutive the same, but caught,so to speak, at a slightlydif- strips of fivepaces (approximatelyyards) along the ferent phase. Here again there is Piceca invasion line, and two paces wide. Thus a crudebelt transect into the Larix zone, but no Picea saplings have of the forest was obtained,and since the diameter yet reached maturity. Correspondingly,the Piceca (and hence girth) of trunkwill be roughlypropor- zone includedin it a few still living Larix trees,tall tional to age for any one species, the resultingdia- and old, of whichonly one occurredon the transect. gram gives a picture both of tree densityand of Abies again dominatesthe oldest stage, thoughthere age-class for the differentspecies along the transect. Fraxinus nigrais also colonizingactively. The Pinus Reading the diagram fromright to left, one may strobuswhich formed the startingpoint of the tran- distinguish: sect is growingat the junction of bog and upland; (1) the moss-heathzone, with saplings of both therewas no marginalfen here. Picea and Larix. As already described,this The sedge-matof this bog was complete,leaving zone containeda fewisolated mature tamaracks, no open water,and in it were a few Larix saplings. but none occurredon the transect. These may be seen in Figure 14, just projectingfrom the sedge-matvegetation, and one occurredon the (2) a wide belt of Larix as the dominant,but transect. The extremelyabrupt margin between with some mature spruce trees in it, and an and sedge-matis clearly shown in this abundanceof spruce saplings. bog-forest photograph,and as at Lake Alice, the forest zones (3) a zone of spruce dominance,without any live themselvesare clearlydemarcated. tamaracks,and The characteristicsof the ground flora may be (4) a zone of invasionby Abies balsamea. judged from Table 5, of which the firsttwo lists The successionalstory speaks for itself, but the give the species most commonlyoccurring in the ecological differencebetween Larix and Picea in re- mature bog-forest. In each list the species are spect to shade toleranceof the seedlings is worth placed in order accordingto the sum of their per- emphasizing; the completeabsence of Larix repro- centages,and no species is given which has a sum ductioninside the forestis reallystriking. The other lower than 17.5. The separation into these lists is point of importanceis the relativelyabrupt transition based on general fieldexperience of the species con- betweenthe zones. cerned,and althoughit is arbitrary,since a species 190 VBRONAVERONA M. CONWAY Ecological Monographs 190 ~~~~~~~~~~~~~~~~~~~Vol.19, No. 2 0 20 40 60 8 0 100 120 I Distance 809 forest mnarg9n-*

Abies balsamea trees /tCheiS,Cr/7) Pinusstrobus 35 Fraxinus nigra 50 80 25 60 I 20- -40

I ~~~~~~~~~~~~~~~~10-20 I 5-

Preesie7cha p Picea c

I0~~~~~~~~~~~~~020~~~~~~~~~2

trees 20t40 L a r x ba p2 401.Taetaostbo-rsaTwnaeF(tePa)maasdeiefrF.1 Larix -60 TLarx Splin FIG.13Transet aross thebo-frsatTiLaeF(tacPakmdasecibdorig12

placed underone headingmight also be placed under found in any otherbog-forest and are characteristic another,it puts the frequencydata in the most con- membersof the neighboringforests on mineralsoils. venientway. The species concernedwere Carex pedunculataMuhl., Perhaps the most noteworthyfeature is the broad Dryopteris phegopteris (L.) C. Chr., Ribes triste distributionand high frequencyvalues of the more Pall., and Hylocomium triquetrium* (L.) B. & S.; acidiphilousspecies; such data formthe basis for the theyhave not been includedin Table 5. previously expressed idea that there is a strong (3) The latest stages "tendency"toward the developmentof acidity even in apparentlynon-acid successions. From the data given in Figures 12 and 13 it List F gives data on tree seedlingfrequencies and appears that Abies balsameea will probably become bears out the conclusionswhich may be derivedfrom the forestdominant in successionto Picea at the two the frequenciesof saplings in the transectdata. In localities concerned,and the same could be said of particular,Larix seedlings were not once recorded manyother Picea forests. Abies is also invadingthe in the bog-forestquadrats. A furtherpoint of note mixed Picea-Thuja forest which occurs at Grand is that Thuja seedlingswere absent in the Boutwell Rapids C. However,at KerrickB therewas a good Creek bog, althoughAbies seedlingsabounded. The growth of Pinus strobus and Betula papyrif era on tree cover was thin in this area, due to earlier re- peat, and at Floating bog, and other localities also, moval of timber,and it may be that the light inten- Fraxinus nigra is an importantcomponent of the sity was too high for the reproductionof Thuja. oldestbog-forest. The questionof theultimate climax Certain other species of the ground flora here were of the bog successionis a large one, and could not also anomalous, in the sense that they were not be studiedin the time available. It must sufficeto April, 1949 THE BoGS OF CENTRAL MINNESOTA 191

-~~~

FIG. 14. View across the completed sedge-nat at Twin Lakes F, to illustrate the abrupt margin of the bog- forest. Carex roqtrata is abundant in the foreground,but most of the mat is formedby C. lasiocarpa. say from superficialobservation that possiblymany burning,and also whenFraxiwets ?sigra becomes domi- bogs in central Minnesotamay end by carryingan nant. A very good example of marginal fen with Abies balsameaclimax similar to thatso widelyoccur- dominantFraxinus may be foundalong the west side ring in the northernconiferous forest region. There of Hinckleybog. is, on the otherhand, little if any evidencethat the The ground floravaries considerably,according to other main climax of centralMinnesota-the Maple- the densityof tree or bush growth,but very com- Basswood type-will ever establishitself on bog sites. monlyhas Carex riparia as the mostimportant spe- cies. Carex rostratais sometimesabundant, and a THE MARGINAL FEN greatvariety of othersedges could be listed. Calarna- Althoughthis communityis characteristicof bogs grostis canadensis and (lIceria canadensis are in general,and clearly recognizablewhen it is pres- frequent,together with a long list of basiphilous ent, it is very variable, and hence cannot easily be hygrophytes,such as Impatiens biflora and Viola described in general terms, nor illustratedby the pallens. Brvophyta are varied in species, but not descriptionof typical examples. This variabilityis importantin volume. partly due to the marginalposition which makes it Early stages of the marginalfen region would be liable to disturbancesand alterationsthrough human interestingto see, but are naturallynot easy to find, activity. Thus for example,in long establishedmar- since it is the oldest region of any bog. However, ginal fens,fine trees of Fraxinu.snigra may be found, at Pug Hole, where the bog is extendingup the it is not uncom- but theseare oftencut out. Again, lengthof a narrowlake, instead of growingcentrip- findthat ditcheshave been dug betweenthe mon to etally, as more usually occurs, there are young upland and the rest of the bog. Anotherpractical stages of marginalfen betweenthe Carex lasiocarpa difficultyfor a researchworker comes fromthe ex- mat and the upland boundaryof the bog. The tran- tremely awkward and treacherous nature of the the zone, which has ground, and the frequent dense growth of alder sect shown in Figure 6 crossed thickets. For these reasons it was not possible to been labelled "marginal caricetum,"since Alnus is makeadequate frequencystudies in the timeavailable. not yet establishedthere. The zone is narrow in Alnutsincana is the mostcharacteristic species, but mnostplaces, commonlyabout 8 to 10 feet wide, and may sometimesbe absentthrough cutting, or perhaps is dominatedby mixturesof sedges; the most corn- M. CONWAY Ecological Monographs 192911), ~~~~~~~~~~~~VERONA ~~~~~~~~~~~~~~~~~~~~~~~~~~Vol.19, No. 2 TABLE 5. Frequency data for the species of bog-forests.

Locality...... Lake Cedar Lake Twin Boutwell Cuyuna Alice Hinckley Kerrick Creek Alice Lakes F Creek Bog-foreststage ...... Early to Mature Early Mature Mature Mature to Late Late Late Late Dominant....;...... Larix Larix Picea Picea Thuja Picea Picea Thuja Numberof quadrats...... 10 7 40 40 40 12 10 12 A. Speciesfound throughout the bog-forest, but tendingto diminishor disappear in the laterstages. Sphagnumrecurvum Beauv ...... 80 86 100 95 12.5 25 90 33 S. magellanicumBrid...... 100 57 37 100 40 33 30 25 Ledumgroenlandicum Oeder ...... 10 100 77 57.5 20 75 10 25 Aulacomniumpalustre Schwaegr...... 70 70 60 7.5 30 33 60 25 VacciniumOxycoccus L...... 80 70 67 27.5 5 16.5 10 8 Pleuroziumschreberi (Brid.) Mitt.* ...... 3 22.5 32.5 17.5 33 50 25 Mniumrostratum* Schrad...... 12.5 . 27.5 8 80 75 Smilacinatrifolia (L.) Deaf...... 14 67 47.5 2.5 .. 40 16.5 DryopterisThelypteris (L.) A. Gray var. pubescens (Lawson)A. R. Prince...... 50 2.5 62.5 . 40 Carexleptalea Wahl...... 14 25 .. 60 .. 10 25 C. trispermaDewey ...... 10 30 55 10 5 .. 60 8 Brachytheciumrutabulum* (L.) B. and S.....20 14 5 .. 25 8 30 16.5 Hypnumpatientiae* Lindb ...... 14 2.5 .. 5 16.5 50 Lonicera.villosa(Michx.) R. and S...... 14 32 .. 10 32 Climaciumamericanum* Brid ...... 14 . .. 20 8 20 25 Camptotheciumnitens* Schimp...... 14 27.5 .. 2.5 .. 40 Dicranumundulatum* Ehrh...... 20 43 12.5 7.5 Carexbrunnescen~s (Pers.) Poir...... S ..60 Trientalisborealis Raf ...... 20 5 42.5 Calliergoncordifolium* (Hedw.) Kindb...... 22.5 .. 17.5 .. 20 Sarraceniapurpurea L ...... 57.5 .. . Sphagnumplumulosum* Roll 29...... 298.. 20 S. sq'uarrosumCrome ...... 37 2.5 .. . 16.5 Polytrichumstrictum* Banks...... 38 5 20 . Sphagnumcapillaceum (Weiss) Schrank ...... 1. 1 30 7.5 Pohlia nutans*(Schreb.) Lindb ...... 15 5 8 Droserarotundifolia L ...... 27 . Polytrichumcommune* L ...... 10 . .. 17 . B. Relicsof pioneer-matand moss-heath Chamaedaphnecalyculata (L.) Moench.var. angustifolia(Ait) Rehder...... 100 29 2.5 2.5 . Equisetumfluviatile L...... 14 72 .. . . 10 Betulapumila L. var. glanduliferaRegel .....40 43 .. . . 8 Potentillapalustris (L.) Scop...... 28 15 .. . . 30 Eriophor'umvirginicum L...... 60 . .. 5 . Carexlasiocarpa Ehrh. var. americanaFernald. 50 14 . Menyanthestrifoliata L. var. minorRaf...... 14 27.5 .. 2.5 .. 10 Kalmia polifoliaWang...... 50 . .. 2.5 . A ndro Li...... Link.10 nk 29 .. 2.5 .. 8 Carex ...... 35 . . .. Drepanocladusvernicosus* (Lindb.) Warnst.28...... 28 Sphagnumsubsecundum Nees ...... 20 .. Drepanocladusaduncus (Hedw.) Warnst.var...... 10 .. 10 1 Kneiffii*(Bry. Eur..) Mbnke-n. C. Speciescharacteristic of the laterstages. Mitellanuda L...... 7.5 25 100 100 Thuidiumdelicatulum* (L.) Mitt...... 47.5 8 90 58 Rubuspubscen~s Raf ...... 16.5 90 75 Cornu~scanadensis L...... 30 . 7.5 33 . 92 Fragariavesca L. var. americanaPorter ...... 5 16.5 50 67 Galiumtri~florum Michx...... 8 70 50 Circaeaalpina L ...... 0 5 Linnaea borealisL. var. americana(Forbes) ...... 0 5 Rehder...... 14 5 . .. 50 10 33 Aralia nudicauli~sL...... 12.5 16.5 20 58 Cornusstolonifera Michx...... 15 25 50 16.5 Thuidiumvirginianum* (Brid.) Lindb...... 17.5 60 16.5 Chiogeneshispidula (L.) T. and G...... 32.5 10 8 20 16.5 Rubusidaeus L. (agg.)...... 15 52.5 8 . 8 Vacciniumangustifolium Ait...... 12.5 20 2.5 8 . 16.5 April, 1949 THE BOGS OF CENTRAL MINNESOTA 193 TABLE 5 (Continued)

Locality ...... Lake Cedar Lake Twin Boutwell Cuyuna Alice Hinckley Kerrick Creek Alice Lakes F Creek Bog-foreststage ...... Early to Mature Early Mature Mature Mature to Late Late Late Late Dominant...... Larix Larix Picea Picea Thuja Picea Picea Thuja Numberof quadrats.10 7 40 40 40 12 10 12

Coptis groenlandica (Oeder) Fernald ...... 8 70 50 Hypnum crista-castrensis*L ...... 2.5 .. .. 40 Pyrola secunda L ...... 42 Hylocomium proliferum*(L.) Lindb ...... 8 .. 16.5 Ribes lacustre (Pers.) Poir...... 10 10 2.5 Bromus ciliatus L ...... 10 10 D. Speciesintolerant of soil pH values below5.5 Impatiens bifloraWalt ...... 70 .. 40 8 Viola pallens(Banks) Brainerd...... 20 45 .. 30 16.5 Maianthemum canadense Desf . . 5 . 52.5 33 .. 16.5 Bidenscoronata (L). Britton...... 37.5 Carexriparia Curtis var. lacustris(Willd.) Kukenth...... 14 10 .. .. 10 Lysimachiathyrsiflora L ...... 32.5 Polygonumsagittatum L ...... 27.5 Bidens cernua L ...... 15 .. 10 Calthapalustris L ...... 14 10 Saxifragapennsylvanica L ...... 2.5 15 Leersiaoryzoides (L.) Swartz...... 17.5 .. E. Tree seedlings. Picea mariana (Mill.) B.S.P ...... 14 15 15 .. .. 10 8 Thuja occidentalisL ...... 65 Fraxinus nigra Marsh ...... 2.5 .. 20 33 Abies balsamea (L.) Mill ...... 17 30 67

mon are Carex strictior, C. rostrata, and C. riparia. AUTECOLOGICAL NOTES Calamagrostis canadensis and Iris sp. (not seen in The scatteredcomments which are collectedin this flower)were also common,while Carex lasiocarpa and sectionare not intendedto give an all-roundauteco- Equisetum fluviatile are found as the transitionto logical accountof any of the species mentioned. In the true floatingsedge-mat begins. the firstplace, it is only the behaviorof the species Marginalfen is not foundon everyradius of every whilegrowing on peat bogs thatis considered. Thus, bog, and there are types of situationin which its for example, although Thuja occidentalisgrows on absence can readilybe explained. Thus,for example, limestonebluffs and other types of mineralsoil, its there are localitiesin which the peat formedin the behaviorin such habitatsis not consideredat all here. bog remainsso close to neutralpoint throughoutthe Secondly,various aspects have been dealt with by bog successionthat there is no contrastin pH between other authors,and no attemptwas made to collect the main bog peat and that of the marginalregion. parallel data in Minnesota. An example of such an At Cedar Creek, the marginal region even has a aspect is the questionof mycorrhiza-itspresence or lower pH than the main mass of the peat below the absence in the various bog species. Then again, a bog-forest,and Thuja, the bog-forestdominant, ex- numberof observationswhich have been made in the tends to the mineralsoil margin,though a few alders Minnesotaarea are not new but only confirmsimilar can be found to indicate the marginalnature of the observationsmade in bogs of other regions, more region. especially by Cooper, Gates and Kurz. They are Then again, as at Kerrick,the drainagewater from describedhere in order to give some measureof the the surroundingupland may be insufficientlyrich in ecologicalsimilarity of the bogs of Minnesotato those bases, or too small in quantitywhen the reliefis very in otherparts of the U.S.A. low, to maintain high pH values in the marginal region against the acidifyingactivity of Sphagnum PICEA MARIANAAND THUJA OCCIDENTALIS fromthe morecentral bog areas. In such cases,moss- In dealing with the mature bog-forest community, heath may be found invading the marginal fen, the pH ranges of these two species were mentioned followed by bog-forestspecies, and sometimesthe briefly. It seems worth while to enquire more closely invasionmay be carriedright out to the upland mar- into the fact that either species may be dominant in gin and the marginalfen is thus entirelysuperseded. the less acid peat, and that they may be codominant Monographs 194 VRONAVERONA M. M. CONWAY CONWAY Ecological~~~~~~~Vol. 19, No.2 occasionally. One may well ask, in relation to any fairlylow, than in the Picea forestat Kerrick,where particularbog whichlacks stronglyacid stages,what the peat was everywhereacid. ecologicalconditions determine the dominanceof one It would seem then, that if the two species are or other species or whetherchance plays the major not complementaryto each other,with respect to pH part in the determination. preferences,they must competewith each other for Some small experimentswere carried out in the dominanceon sub-neutralpeat areas. Let us examine laboratoryto confirmor extendthe fieldobservations their distributionareas to see if they give any clue on pH tolerance. First,there were trials with Thuja to the probable outcome of the competition. The seedlings. Seeds could not be found,but large num- distributionmaps are given by Munns (1938). For bers of first-yearseedlings could be collectedfrom both species the distributionarea reachesa boundary Cedar Creek bog, and these were placed with their in Minnesotawhich runs diagonallyfrom northwest roots in buffersolutions, one batch in a solutionof to southeastacross the state,and mustcoincide fairly pH 6.2, the other in a solution of pH 4.2. After closelywith the boundaryof the area of studyshown fourdays the firstbatch was muchmore healthy than in Figure 1. The distributionarea withinthe U.S.A. the second. The buffersolution of pH 4.2 was a is roughlythe same for the two species; the differ- phthallatemixture, and as it could have been that ence betweenthem lies in the huge westwardand the phthallate was more toxic than the phosphate northwardextent of the Canadian area for Picea, mixtureused for the solutionof pH 6.2, a phthallate which is quite absent in the case of Thuja. How- mixtureof pH 6.2 was given to the seedlingspre- ever,the centralMinnesota bogs lie at some distance viouslyin the phosphatemixture. They did deteri- fromthe northernlimit for Thuja, and thereis no orate gradually,but not nearlyso rapidlyas the ones reason to think,just fromlooking at the distribution in the solution of pH 4.2. The experimentwas areas given on this continentalscale, that the climatic repeated,this time using four batchesof eight seed- conditionsof central Minnesota are likely to favor lings each, in the four followingsolutions: acetate one species ratherthan the other. mixturesof pH 4.2 and 6.2 and phthallatemixtures of pH 4.2 and 6.2. All the seedlingsin the more NOOf Nlot Testa acid solutionswere discoloredwithin three days. In seeds Root Expt pH seed,5 rmln slitd lengtn (mm! 0 2 ~4 6 the less acid solutionsabout half the seedlingsap- atedl peared unhealthyafter 4 or 5 days; there was no apparent differencebetween the effectsof acetateand phthallate. The controls in tap-water remained healthy. These experimentsconfirm the conclusion c-1 that low pH is an adverse conditionfor Thuja. The correspondingexperiments with Picea were carried out as germinationtests with seeds. A pre- liminarytest was made withsets of 10 seeds in each of a set of 8 solutionswith pH values rangingfrom 3.0 to 6.5. It was clear fromthis firsttrial that the percentagegermination was higherand the growth- rate greater,the higher the pH value. Two more experimentswere carriedout, using contrastingsolu- tions of pH 4.1 (phthallate mixture) and pH 6.5 (phosphatemixture). The resultsare shownby the FIG. 15. Results of germinationexperiments with frequencygraphs given in Figure 15. In Experiment Picea mariana. 1 an attemptwas made to eliminatefungal attacks on the seeds by dustingthem with copper sulphate In fact,the firstimpression received from a study powderbeforehand. A large numberof seeds showed of these bog-forestsis that the dominanceof Picea a split testabut no furthergrowth, and thissuggested or Thuja is decided mainly by chance. There is, that the copper sulphate was toxic to the seeds, and however,one line of observationfrom which one the treatmentwas omittedin Experiment2. The muaytentatively suggest the limitationsunder which, to speak figuratively,chance has to work. This is, results could be taken to mean merelythat sodium namely, the distributionof Thuja seedlings as it is more to Picea seedlings than phthallate toxic appeared during the fall of 1946 and the summer sodiumphosphate, and furtherexperimentation would of 1947. They were,with one single exception,never be needed to meet this criticism. Assumingthat it seen outside the mlargin of a maturebog-forest, not could be met, the conclusionis clear that Picea is even in the stage of mass-invasionby Larix. On the not acidiphilousat all, but that it can tolerateacidity otherhand, they were extremely abundant on the floor of a degreewhich cannot be toleratedby Thuja. The of forestswith Thuja as dominantor co-dominant. conclusionis supported by the observationin the Picea seedlingsand saplings, however,occurred fre- field that Picea seedlings and saplings were much quentlyamong the more rapid growingLarix trees more abundantin the Larix dominatedbog-forest in of invading bog-forests,yet they were not of such the Lake Alice bog where acidities were in general frequentoccurrence in maturePicea stands as T~huja April, 1949 THE BOGS OF CENTRAL MINNESOTA 195 seedlingsin matureThuja stands. The seeds of par- closely behind Aulacomniuim palustre. This bog, as ent plants at some distanceare morelikely to arrive discussedearlier, shows only a moderatedevelopment in numberson a pioneer-mator on an open Larix- of acidity,and in general,it seemedthat Andromeda invasionzone, than on the floorof a maturebog-forest; has a pH optimumrather higher than the optimum hence, if the germinationdifferences just discussed for Chamaedaphne. This is probably only another are general,and not a featureof one particularsea- way of expressingthe fact thatAndromeda character- son, Picea will have a greater chance of being the izes the youngerstages of moss-heathformation, since successor to Larix and will thereforebecome the the older a moss-heath,the greaterits acidity,up to dominantin a larger numberof areas. On the other a limitingvalue of about 4.0. This is well illustrated hand, in those places where a few Thuja trees do by the Pug Hole transectin Figure 6, whichshows become established.they will be likely to reproduce the absence of Andromeda fromthe inner and older more freelythan Picea and so, in those few places, part of the moss-heath,dominated here by Chamae- Thuja will end as the dominantof the maturebog- daphne. forest. Chamaedaphne is the characteristicdominant of This suggestionneeds to be testedby experiments the moss-heathin its matureand late stages and per- on the light and shade toleranceof the seedlings,as sists until a completebog-forest cover is established. well as on other conditionsfor germination,and by It is the most acidiphilousof all the species in the long-termobservations on the areas in which Picea list given above (except perhaps for Chiogenes his- and Thuja at presentgrow together. pidula, for whichno observationson pH range were made), and was neverfound growing where the sub- ERICACEAE surface peat has a pH value above 5.0. The plant It would be possible,by takinga broad view-point, appears not to be shade tolerant,however, for it thins to group togethera number of Ericaceous species out as soon as the Larix cover becomesat all dense, and to considerthem as formingan ecologicaltype, and its place is taken by Ledum, which seems, in the "Bog Ericads." Such a group,in centralMinne- contrast,to be intolerantof exposureand character- sota, would include Chamaedaphne calyculata, Vac- izes the bog-forestrather than the mioss-heath,though cinium macrocarpon, V. Oxycoccus, Andromeda glau- found occasionallyin the latter. There is, however, cophyila. Kalmia polifolia, Ledum groenlandicum and a lower limit of light intensityfor the growthof Chiogenes hispidula. It is even possible to imagine Ledgum,a limitwhich is passed in areas of verydense that one mightfind them all growingtogether within bog-forest,such as some parts of the Thuja forestat the limits of a single meter quadrat. To look at Cedar Creek bog, and the old Picea forestat Lake themin this way, however,is to obscurethe interest- Margaret,though it remainsflourishing in thebrighter ing and fairlyobvious ecological differencesbetween openings whichoccur here and there. The plant is them. nearly always associated with Sphagnum, and it is, Take first,for instance,the two species of Vac- therefore,possible that it requires an acid medium cinium. During this study,V. macrocarpon was only for its functionalroots, though this point requires found in any quantity at Cuyuna and at Kerrick study. It can certainlygrow in bog-forestswhich (Lake Margaret). In boththese places V. Oxycoccus have not been preceded by a moss-heath(as far as was also common,but it was dlearlygrowing further may be judged by generalobservations) and in which away fromthe water marginthan V. macrocarpon, the peat at the level of the water-tablemay have a and was also frequentand characteristicinside the pH of well over 6.0. bog-forest,where V. macrocarpon was never found. Kalmia polifolia is highly characteristicof these The zone of overlap betweenthe two species was not bogs, but not presentin any quantity. It appears to wide, usually less than a yard across, and Vaccinium have muchthe same habitatpreferences as Ledum. Oxycoccus was neverseen to play the part of a pio- These ideas may be summarizedcrudely by the neer of the moss-heathcommunity. At Cuyuna,long diagram given in Figure 16; the diagram must not runnersof V. macrocarpon could be found pushing be taken to suggestthat the occurrenceof Ericaceous out ahead of the Sphagnum magellanicum; theserun- species in the bog-forestis dependenton the previous ners were infrequentlybranched in the horizontal existenceof a moss-heathphase in the bog succession plane, thoughbearing vertical side-shoots. A num- of which the bog-forestforms a part. As Table 5 ber of runnerswere pulled out and examined. One shows,species such as LDedumand Vaccinium Oxycoc- of them,for example, was 13.5 feet long from the cus are characteristicof all typesof bog-forest. apex to the point of breakage, carried 7 vertical Vaccinium Oxycoccus occurs in Britain,and there lateral branches,had grown a lengthof over 2 feet too it is highlyacidiphilous. The BritishAndromeda the currentseason, and appeared to have all its live polifolia closely resemblesA. glaucophylla in habit, rootsconfined to the proximal2 feet,where they were thoughit is smaller and less gregarious; it also in- growingamong Sphagnum plants. habitspeat bogs,and interestinglyenough it has been Andromeda, like Vaccinium macrocarpon, is char- dyingout fromBritain in the last hundredyears, a acteristicof the youngestmioss-heath stages, though it fact whichmay well be causally connectedwith the surviveslonger. At Lake Alice bog, it was not only extremeacidity which has developedin the surface the dominantof the moss-heathas a whole,but also peats of manyBritish bogs as theyhave degenerated appeared to be invading the pioneer-mat,following throughdraining, burning, and erosion. Perhaps the VERONA lxi. CONWAY EcologicalMonographs 196 Vo.19, No. 2 most striking parallel, to a British ecologist, is that niental approach to the phenomenon of the produc- which may be found between Cha-naedaphne and tioIn of acidity by Sphagnum plants, a phenomenon Calluna vulgaris. Both are strongly acidiphilous, which has been dealt with at greater length by other both vary in growth form from a sparse low growth authors, for instance by Baas-Becking and Nicolai to a dense bushy growth, and both are strongly en- (1934), and much more fully by Ziegenspeck (1936). couraged by burning and by removal of tree cover Though the exact mechanism of ion-exchange may not froni areas of suitable soil character. The outstand- vet be known, it is perfectly clear from the work ing ecological difference between them is that Cal- of these and other research workers that Sphagnum, luna will grow on mineral soils, but they appear in and other mosses also, will alter the pH of the medium other respects to be completely vicarious. in which they are growing until the pH is within the A last but important point concerning these bog range which is optimal for their growth. The initial Ericads is that in the area under consideration, no pHi must not be so far from the optimum as to be seedling of any of the species was ever found, in toxic to the plant, but within fairly wide limits the spite of careful searching. Obviously, germination species can "create" its own optimum, whether the Ro1usthave occurred from time to time, to give the initial pH is higher or lower than the optimum. plants their present distribution, but it seems quite This holds good not only for those with an acid evident that vegetative reproduction is far and away optimum, but for basiphilous species also. more important at the present tine. This serves to The field observations made during the present explain the strongly gregarious growth which is study were entirely in accord with these findings, shown by all the species, and also the apparently Ibotlhas to the general principle, and as to the be- haphazard absence of species from bogs which would havior of those individual species which were among appear to afford all the right habitat conditions. those dealt with in the experiments of the authors Thus, for example, there is no obvious reason for mentioned above. It is for this reason that the word the absence of Vacciniumn macrocarpon from the "acidifying" has been used for certain of the Sphag- Lake Alice and Pug Hole bogs. The phenomenon nitrn species; acidifiers are of course acidiphilous, but clearly calls for intensive study of the conditions for the converse is not necessarily true. germination, which might well throw some interesting The species are dealt with roughly in the order of light on the equilibrium or lack of equilibrium be- increasing acidity of their optimum. The majority tween present climiatic conditions and the bog vege- of the pH data came from a set of observations taken tation of this region. in the bog-forest at Hinckley bog. (1) S. squarrosumn appears to flourish in sub- lPIONEER-' neutral situations and is moderately shade-tolerant. MAT MOSS-HEATH BOG-FOREST The greatest abundance of it that was seen was on the Cedar Creek sedge-miiat. It also occurs in mar- Vcaccn17um macroc,-olooo g-inal-fens,as at Pug Hole, where the pH was over 84dro'rec~i ,oo/,fo/,s: 6.0 (transect PQ). Three samples at Hinckley from just below tufts of S. sqiarrosum gave pH values of Chamaedaphre cjycL'/at2 ' _ 5.40, 5.48 and 6.33. The lowest value found for ,VcC/n/csr OXYCOCCUS such a sample was 4.55, but in general the moss is not at all characteristic of acid situations. It is use- ' ~~~A'&/fl749,oo/,fo//, ful to ecologists as an indicator, especially as it is easily recognizable in the field, but it does not occur in sufficientbulk to be an important member of a 0C9/eloenes h ',dL'1/a plant community. (2) S. subsecutndumn.This species has been much lv I'cc/lqn a1f79a5t/fo//41n split by some taxonomists, but Le Roy Andrews FIG. 16. Diagramto representthe ecological ranges (1913) retains the name to represent a species, and of the bog-livingEricaceous species in the Minnesotathere seems every justification for following his no- bogs. menclature since there was a close similarity between all the specimens except for a few plants found at SPHAGNUMSPP. Pug Hole, which were larger, and may have been a Thespecies of Sphagnumshow an ecologicaldiver- growth-forimor may need a differentnomenclature. sitywhich makes them such valuable field indicators These were confined to a small area, and are not of habitatconditions that their importance to ecol- considered in the remarks which follow. ogistscan hardlybe over-estimated.Moreover, since This species again is easily recognizable in the theyare an old groupwith their species very widely field,and is characterized by a dirty olive green color distributed,they form a veryvaluable link in the when submerged, but a bright orange yellow color comparisonof partlysimilar vegetation types in when exposed, at any rate in the fall season. It is widelyseparated continents. found abundantly in the older parts of the Carex The data here collectedshow a close agreementlasiocarpa miats at Cuyuna and Pug Hole. At both withthe findings of Kurz (1928) in his fieldstudies places it persisted in the moss-heath in hollows which of bogs in Michigan. Kurz also made an experi- reached close to or below the water table, where the April, 1949 THE BOGS OF CENTRAL MINNESOTA 197 pH was probably higher than in the surrounding Immediate 6 ins. tussocks. It occurredalso, thoughless abundantly, Sub-Surface Lower in the pioneer-matof the Lake Alice bog. On the 4.08 4.22 15 ins. lower; 6.72 (below other hand, it was never found in the pioneer-mats water-table) of those bogs whichcould be called "circum-neutral" 4.49 in the nomenclatureof Kurz, that is to say in bogs 4.50 5.75 4..55 4.90 such as thoseof Twin Lakes and Cedar where 16 ins. lower; 6.80 (below Creek, water-table) the factorsleading to moss-heathformation are not 4.58 operative. 4.63 (3) S. cuspidatum was not often found in any 4.82 4.25 15 ins. lower; 6.90 (below abundance,though it was frequent,for example,on water-table) the sedge-matat Pug Hole, mixed with S. subsecun- 4.87 dum. 5.07 4.43 18 ins. lower; 6.55 (below water-table) (4) S. recurvum ranges from the earliest moss- 5.47 heath stages into the oldest bog-forestswhere it finallybecomes shaded out. It also occurs in fair S. mnagellaniciunivery quickly builds up deep dense quantityin the pioneer-matat Cedar Creek, where cushions of moss, and the figures show that the old pH values are not below 6.0. Nevertheless,it is dead remains of the moss may be more acid than the definitelyan acidifyingspecies as shownby the fol- the still living stemnbases. lowing list of pH values obtained from below S. When it grows in the open, this species has a recurvum surfacesat Hinckley: purple-red color which is quite unmistakable, but in Immediate the shade it may often have a pure pale green color. Many samples from Sub-Surface 6 ins. Lower the bog-forests were examined in the laboratory to make sure that they really be- 4.38 6.73 longed to this species; they always did, and no other 4.42 6.75 member of the Sphagna cymnbifoliawas ever found in 4.52 5.55 the bog-forests. 4.60 (7) Sphagnum capillaceuin. The figures given be- 4.62 5.95 low indicate that this species is the most strongly 4.75 6.75 acidiphilous of any of the bog-species. 4.82 6.76 Immediate 4.87 Sub-Surface Lower Samiples 5.10 6.33 4.08 6 ins. lower; 4.22 15 ins. lower 6.72 5.40 6.65 4.10 12 ins. lower 7.33 4.15 From these figuresit would appear to have its opti- 4.20 6 ins. lower; 4.65 mumrange from4.5 to 5.0, whichis muchthe same 4.30 6 ins. lower; 4.88 as that suggestedby the 4.60 8 ins. lower; 4.15 figuresgiven below for S. 4.62 magellanicum. From what has been said previously, however,and fromthe fact that the pH values were Yet it may be doubted whether this species has the so much higherin samples only 6 inches below the strongest capacity for acidification, since it only ap- living plants, one must concludethat the species can pears in places where S. recitriurnor S. inagellanicitm invade successfullyinto areas with a pH very much are already growing abundantly. It is especially higherthan its optimum. This is supportedby the characteristic of the secondary moss-heath with tall observationat Hinckleythat S. recurvum was invad- thick Chamaedaphne and forms the upper part, some- ing muchmore freelythan S. magellanicum into the times the whole, of tall deep moss tussocks. It is wetter parts of the bog-forestwhich were rich in characterized by its small, soft, bright red heads, basiphilouspioneer-mat relies such as Typha, Equi- packed together to form a mosaic pattern. Its ap- setum fluviatile, and Menyanthies. pearance and ecological behavior so closely resemble (5) S. teres was found occasionallymixed with S. that of the moss which is described by writers on recurvum and havingthe same ecologicalrange. British bogs as S. rubellrmnthat it is hard to believe that the two names are not (6) S. magellanicum is the acidifierpar excellence. really synonymous. (8) Lastly, for the sake of Its behavioris well illustratedin the Pug Hole tran- comparison with British bogs, we may record that, S. papillosuminwas found sect shown in Figure 6. It follows after S. sub- only once (at Cuyuna). This is at present probably secundum and is the dominantthroughout the zone the most active peat builder on the British Isles, of falling pH values, until a stable low pH value is though S. magellanicunmis also very important. Fur- establishedand the accompanyingmature moss-heath ther, S. imbricatumn. was never found. This species develops with Chamaedaphne as the dominant. pH has in the past been responsible for the formation data were obtained from Hinckleyin the same way of large volumes of peat in the British Isles, but is as for S. recurtlum,and were as follows: now very rarely found, and nowhere in large quan- 198 VERONAVERONA A. M. CONWAY Ecological Monographs CONWAY ~~~~~~~~Vol.19, No.2 tities. It would be of the greatestinterest to know therefore,it is found in quantityin a peat profile whetherthis species, easily recognizablefrom a single as, for example,at Cuyuna,it is likelyto indicatea leaf fragment,has at any timnebeen abundant in fairlyyoung bog-forest stage in whichnot muchacid AmnericanSphagnum regions.Such informationcould peat, if any, has been formed. readily be obtained by any one who is workingon (3) Polytrichum stricturn. This species is markedly pollen analysesof peat profiles. acidiphilous, but like Sphagnum capillaceum, may not necessarilybe strongly since it always OTHER BRYOPHYTA acidifying, occurs in assocation with the acidifyingSphagnum. (1) Aulacomniuzmpalustre. This species was found It occurs in fact miosttypically in moss-heathcom- by Ziegenspeckto be an acidifier,with its optimiumnmunities, and in primarymoss-heaths it is usually at a pH of about 4.4. It may be seen at work,so thinly scattered. In secondary types, however,or to speak, on the pioneer-miiatof the Lake Alice bog, underconditions of markeddryness, as at KerrickB, w-hereit is the most abundant miioss,and precedes it multipliesvery rapidly and may formnnumerous Andromeda. It remains as the main imiossof the conspicuousdark green cushions,and one may find moss-heathhere_, and in general it persists through patches in whichthere is very littleother vegetation all the stages of a bog successiononce it has become apart fromthis species and Cham.aedaphne. It occurs established,even into the latest bog-foreststages in small quantitiesin bog-forests,but not consistently, that were seen. Like Sphagnum recurrum, it can and it is perhaps intolerantof shade. toleratefairly high pH values; it is found for in- stance with fair frequencyon the pioneer-matat TYPHA LATIFOLIA Cedar Creek. In the paper by Ziegenspeck(1936) Figure 17 illustratesa point of interestconcerning it is suggestedthat there are two ecotypesincluded the ecologyof this species, namelythat it has a dual in the species,one acidiphilous,the otherbasiphilous. type of distribution. On the left of the photograph The mode of occurrenceof the species in Minnesota is a patch of Typha whichborders the water-margin suggests,however, that it nmaymore easily be con- of the pioneer-mnatat Floating bog (Lake Itasca). sideredas one species witha wide range of tolerance, In the backgroundis the bog-forest,with a clearly since it occurs in practicallyeverv bog in this area. definedzone of Typha betweenit and the pioneer- (2) Calliergon cordifolium. (giganteurn). All the mat. The two areas of Typha dominanceare sepa- specimensfound during this study agreed with the rated by a wide zone of compact Carex lasiocarpa descriptionsgiven for C. giganteum in the standard mat. Other examples of this behavior could be texts. However, the name C. cordifolium is here quoted. A reason for this may be suggestedby the used in accordancewith the recenttreatment of the fact that whereTypha is abundantand luxuriant,the genus Calliergon, given by Wynne (1945). The spe- peat surface is not held togetherinto a flrmnsolid cies is highly characteristicof small shallow pools mat,but is usuallysoft and treacherousto walk upon or wet depressionsin bog-forestswhere the peat is between the Typha tussocks. On the other hand, circunm-neutralat the level of the watertable. When, where a sedge-mat has a tough compact surface,

- ~~~~~~~~~~~a -

FIG. 17. Typha latifolia in the pioneer-matof Floating bog, Lake Itasca. In the foreground,plants growing close to the water margin; behind, the Typha zone at the margin of the bog-forest. April, 1949 TIHE BOGS OF CENTRAL MINNESOTA 199 Typha shoots rarely occur in it, and when they do, the Carex lasiocarpa mat, the large one on the left they are single and small. It would seem then that came fromthe marginalzone of the bog-forestwhere apart fromthe places whereTypha is itselfthe domi- it adjoins the sedge-matand whereTypha is tall and nant of a whole pioneer-mat,the species can only abundant. succeed as a componentof a pioneer-matin those areas where the mat is loose-texturedand does not CHARACTERISTICS OF THE PEAT offertoo much mechanicalresistance to penetration Anyone who is interestedin the study of plant by the Typha rhizomes. Such areas will naturally successionsmust be preparedto analyze the progres- be found firstlyat the young margin of the mat sive changes in edaphic conditionsas well as to de- whereit is invadingthe water surface,and secondly scribe the plant communitieswhich partly or wholly at the oldest part of the mat, where tree growthis cause the soil changes and respond to them. For- causing death of the main mat-dominantsand the tunatelyfor studentsof bog successions,there is a ground surface is again becomingloosened. Figure vast literaturedealing withthe physicaland chemical 18 illustratesthe effectsof this competitionbetween propertiesand the microbiologicalpopulation of peat, Typha and the sedge-matdominants. Two Typha and thoughmuch of it is writtenfrom an agricultural shootswere dug out fromthe bog at Twin Lakes F; point of view,the resultsare of basic importancefor the small one on the rightcame fromthe centerof ecologists. Thus, for example,much useful, general informationconcerning peats in Minnesotamay be obtainedfrom the surveyby Alway (1920) and from the separate countyreports of the U. S. Department of AgricultureSoil Survey. More searchinganalyses ~IIs may be foundin the workof Waksmanand his col- leagues, and of Dachnowsky-Stokes,2as well as a vast range of otherliterature dealing with peat chem- istry and with the distributionand metabolismof soil microorganisms.From all this workover nearly a century there has emerged a body of firmly- establishedand well-knownprinciples concerning the types of organismwhich are active under different conditionsof aeration and chemicalbackground, the substratesthey use and the substancesthey produce, the relativerates of degradationof differentchemical L- types such as carbohydrates,liquins and proteins, and numerousother related problems. Within the limits of a short-termstudy it was obviouslyimpossible to embarkupon elaboratechemn- ical and bacteriologicalanalyses, either to verifythe application of these general principlesto the peats of central Minnesota,or to extend and refinethem. However, there seemed to be various points which mightbe dealt with by simple and rapid methods, and two in particularwere chosen because theymight yield data for contrastor comparisonwith the bog types of the highlandregions of Britain. The first was concernedwith the correlationbetween the de- gree of humificationof the upper peat layers,and the type of plant communitygrowing upon it. The second concernedthe natureof the fungiwhich char- acteristicallygrow in the peat types,and moreespe- cially thoseof acid peats.

DEGREE OF HUMIFICATION OF THE PEAT European peat investigatorshave often made use of the scale of humnificationwhich was originallyde- signed in Scandinavia. This is a methodof classify- ing peat samples on the basis of theirtexture and of the amountof liquid whichcan be squeezed out of a damp sample in the hand, and so on. The classes are given numbersfrom 0 to 10, correspondingto FIG. 18. Two plantsof T?>pha latifolia dug out from increasingstages of humiflcation,but the methodis tile bog at Twin Lakes F. On the left, a plant from essentiallyqualitative, and thereforeunsatisfactory for Mie bog-forestmargin; on the right, a plant from the 'The book on Peat by Dachnowsky-Stokes,promised by middle of the sedge-mat. Chronica Botanica, was not available at the time of writing. Monographs 200Ecolo0gical VERONA ML CONWAY Vo.1,Io the comparisonof data obtained in differentplaces drip out, and sieve and sample were weighed. Then by differentauthors. Moreover,the surfacepeats of a strong jet of water was run on to the sample, the Minnesotabogs were so inhoimogeneousin texture which was spread out thinly over the sieve, and that the method was almost impossible to apply. turned over with a glass rod. Washing was con- Hence it was decided to make use of the ordinary tinued until only the coarsest fractionof the peat sieve methodfor estimatingthe coarser fractionsof remainedon the sieve. After allowing to drain for mineral soil, and to modifyit where necessary,to 5 minutes,sieve and residue were re-weighed. The make it applicable to peats. The main modification filtratewas passed througha sieve with holes of 1 arose fromthe fact that air-driedor oven-driedpeat mi. diameterand similarlywashed free of finerpar- bears no relationin its propertiesto the same peat ticles,and the processrepeated with a finesieve (60 whilestill in its naturallydamp or waterloggedstate. to the inch mesh). Thus coarse, medium,and fine Hence it was impossibleto work on a dry weight fractionswere obtained, and calculatedas percentages basis, and instead,both the original sample and the of the original wet weight. The differencebetween separated fractionswere dealt with in the saturated the sumnof these percentagesand 100 percentrepre- state. The procedurewas as follows: 40 gm. to 50 sents the "remainder,"that is, the finestfraction of gmn.of moist peat was placed on the coarsest sieve all. Estimateswere made in duplicate. The method (a meshof 8 wiresto the inch) and waterwas played is undoubtedlycrude, but the results,as given in on to it verygently from the finejet of a wash-bottle Figure 19, show some degree of consistency. The until the sample was uniformlywet, and water was greatest error arises where the peat is wholly or just beginningto drip out frombelow the peat mass. largelycomposed of Sphagnum,for it happened sev- The excess of "gravitationalwater" was allowed to eral timesthat negativevalues were obtainedfor the

EARLY MIDDLE LATE C M F R C M F R C M F R L | f I 4 ff 6 ffi f to 'f + f Cedar Creek Lesser Tamarack Kerrick Sedge - mat Moss heath Secondary Sedqe-mat ej~~~~~~~ ~ mj j t ath

2 2 fljllk Kerrick 2 ~~~~~7 ~ Kerrick Old Plirea Cuyuna Moss-heath Sedge-mat

_ _ 8 Pken

Cuyuna Sedge-mat with Creek I T_ _-Kerrick | ,51 swbsec. _t Cedar 7h~e Cuyuna ii TiL

14 Cuya _ KerrickMarginal fen Maturemoss- ~ ~~heath I Scale * I I Cuyuna Marginalfen Vertical so

Cuyuna L j Earlymos- 6 Kerrick&-iopZ4o,-vm1 Y heth C Coarse I Medium fractions F I Fine R Remainder _ FIG. 19. Results of mechanical analyses of peat samples into fractions as described in the text. The analyses were made in duplicate, the differencebetween the estimates being representedby the blank part of the column, and the mean shown by the horizontal line. April, 1949 THE Boos OF CENTRAL MINNESOTA 201 "remainder" fraction, since the sum of the three though not freshly collected, had been stored in a weighed fractions added up to more than 100 percent. refrigerator. This is seen in diagrams 2, 4, 7, 8, and 9 of Figure Agar plates were prepared at pH values of ap- 19. It seems that Sphagnum leaves are capable of proximately 4.5, 5.5 and 6.5, and the medium used retaining greater Xolumes of water when they have for any peat sample was the one having a pH value been washed out separately, than when they are still closest to that of the peat mass from which the sam- massed together in a compressed state in the original ple was taken. In a number of cases platings were peat sample. made in duplicate, and one of each pair was placed All the samples tested were from sub-surface peats, in a sealed chamber through which nitrogen was that is, from depths of between 10 cm. and 20 cm. passed for four hours or more. The lids were re- below the surface. moved from the dishes immediately before placing in the chamber, to avoid the retention of oxygen Turning to the results, a clear agreement may be inside the closed dish. The nitrogen was sterilized seen between samples 1 to 5, grouped together as be- oxygen by passing it longing to the earlier stages of pioneer-mat or pri- and freed from traces of before it reached the mary mioss-heath. All show a great preponderance through pyrogallol solution chamber. Even so, contaminants on the plates might of the coarse fractions. Samples 6 to 9 belong to well have been expected, but in fact none of the spe- the younger bog-forest types or to their equivalent control in secondary moss-heaths. These show greater variety cies which were apt to contaminate the plates plates, and hence it in the shape of the frequency graph. This is prob- ever appeared on the "nitrogen" seemed reasonable to assume that organisms which ably because they are all Sphagnum-rich, and only had been derived from it. a very small difference in state of preservation is appeared on the peat fell in with the general experience that needed to determine whether Sphagnum remains will The results are checked at pH values adhere together as whole plants and hence appear in bacterial growths usually than and that the variety of micro-organic the coarse fraction, or break up into separate leaves lower 5.0, at than at and so appear in the medium fraction. Samples 10 type is much greater higher lower pH Table 6 gives the data concerning all the to 13 are from mature bog-forest regions, or from values. on more than one an equivalent secondary moss-heath; 14 and 15 are filamentous fungi which appeared from marginal fens, and 16 is from an area of drained peat with a water table far below the sur- TABLE 6. face. These last seven samples are arranged accord- ing to the order in the successional series of the plant rH OF PEAT AND AGAR community growing on the peat. Under5.0 Over 5.0 Developed The results as a whole show a clear correlation (10 (12 in between the proportion of finer fractions in the peat samples) samples) Nitrogen and the successional position of the vegetation. That Trichodermaviride* is to say, in the later stages, the conditions are such Pers. ex Fries...... 6 5 Yes as to cause a greater degree of disintegration of the Mucorsilvaticus* dead plant tissues as they become converted into Hagen...... 9 7 Yes peat. The data provide illustrations of two well- Penicilliumsp...... 9 5 Yes Zygorhynchusmoelleri* known principles of soil microbiology. Firstly, the Vuillemi...... 2 Yes greater the activity of aerobic organisms, the more Gliocladiumfimbriatum* rapid, in general, is the breakdown of organic mat- Gilmanand Abbott.. . 2 No Fusarium .. No ter; the degree of this activity will depend on the sp...... 2 successional status of the growing plant community, because the raising of the surface level above the water table is the essential process in the succession. plate. The first three organisms on the list in the Secondly, the lower the acidity of the peat, the greater table are the only ones of frequent occurrence, and the variety of the micro-flora,and hence again, the the only acid-tolerant types. Moreover, all three greater will be the rate of breakdown, as nmaybe appear capable of developing in an atmosphere of judged by graphs 14 and 15 for marginal fens and nitrogen. Further tests were therefore made to in- graph 13 for the non-acid bog-forest type. vestigate their power to withstand submergence. Innoculations were made on to malt-agar plates It thus appears from the correlations shown by (pH 4.5) and the plates were submerged, without Figure 19, that the method might usefully be applied lids, in beakers of distilled water so that the agar of conditions under as one means determining the was about 3 cm. below the surface of the water. All which any particular sample of peat was formed. three species produced mycelium freely, the Mucor FUNGAL FLORA OF PEAT TYPES in greatest volume. The latter species produced many chlamydospores, but no sporangia. Tricho- Growths of filamentous fungi were obtained by dernmaviride3 and thePenicillium produced sporangio- planting small lumps of peat directly on to imialt-agar phores where the mycelium reached the air-surface. plates. The samples of peat were taken from the 3 This name is used in accordance with the findings of Bisby freshly opened centers of large jarfuls of peat. which, (1939) concerning this species and its nomenclature. 202 VERONA ALM. CONWAY Ecological. Monographs 202 ~~~~~~~~~~~~~~~~~~Vol.19, No. 2 Finally, the three species were tested for growth in such places be the climax type, or perhaps,more under both submergedand anaerobicconditions. In- accurately,the sub-climax,if one is to adhere to the noculationswere made into 1 percentmalt-agar solu- terminologyof Clements. tions in test-tubeswhich were so fittedthat nitrogen Turningnow to the descriptionof the bog succes- could be passed throughthe liquid, and the tubes, sions on Isle Royale, given by Cooper (1913) it with nitrogenabove the liquid, could be sealed off appears that these would fit broadly with the sum- from the atmosphere. The nitrogen was bubbled marystatement which has been made about the Min- vigorouslythrough the liquid for two hours. It was nesota bogs. However, under this broad similarity foundthat all threespecies producedmycelium under thereare a numberof differences,some only of detail, these conditions,though not as freelyas the controls but some of more weight,which must express some not subjected to the nitrogen treatment. Again, genuine differencebetween the successional factors spores were only produced at the liquid-gas inter- which operate in the two regions. One such point face, and thiswas trueof the controlsalso. concernsPicea mariana,which in Minnesotamay cer- It appears then, that these three fungi at least tainlyform a seral stage whichis not precededby a can producea vegetativegrowth when submerged and Sphagnum-richstage, whereas on Isle Royale it is lacking oxygen,and since they commonlyoccur in said that "Ledum and Picea mariana follow the the peats of this region,they may be consideredas sphagnum,being dependentupon its presence." An- importantagents in the decompositionof plant mate- otherpoint of differencemay be seen in the statement rials below the level of the water table. There is by Cooper that in those successionsin whichSphag- another set of fungal organismswhich are also of numis abundant,"the sphagnumis a superficiallayer ecological importance,but which cannot be studied supportedupon the sedge-mat,and contributeslittle by plate-culturemethods. These are the mycorrhizal towards peat formation." While this is sometimes fungi,and the Sphagnicoloushigher Basidiomycetes. true of bogs in Minnesota,there are bog types,such The mycorrhizalroots of the coniferousspecies are as thoseat Kerrickand Cuyuna,in whichSphagnunm found predominantlyin the moist Sphagnum layers has played an importantpart as a peat-former. of the bog-forests,above the water table, and the Here then are a few of the points of difference Basidiomiycetefructifications were associated with con- that emergein makingcomparisons with bog-bearing spicuous myceliagrowing in the loose upper mass of areas not too far removedfrom Minnesota; the state- Sphagnum stemsjust below the livingsurface. Only mentof themmust be left as a merestatement, since a smallnumber of theseBasidiomycetes were observed without a first-handexperience of American bogs to occur withany frequency;these were Cantharellus outside Minnesota,it would be idle to offerexplana- infundibuliformis,Inocybe scabella and Russula rose- toryhypotheses. tpes. From theirmode of occurrence,it is probable If we carry the comparisonacross the Atlanticto that both mycorrhizalfungi and Basiodiomycetesare the BritishIsles, thereis no difficultyin findinggen- aerobic, and hence may be consideredas important eral resemblancesin the lake hydroseres. These re- in the initial breakdownof plant materialsduring semblancesare greatestin the middle stages of the the formationof acid peats in the moss-heathand hydroseres,for in both regions we find shrub and bog-foreststages. open forest stages invading the surface which has been built up to the water-levelby the earlier com- GENERAL DISCUSSION munitiesof herbs or low shrubs. In Britain again The central Minnesotabogs may be describedin both "acid" and "circum-neutral"bogs, and inter- the shortestand most simple way by saying that mn'ediatetypes can also be found. In the Norfolk they show a succession from pioneer-mat,through Broads region for instance,there is no development bog-forest,to some climax stage whose natureis not of acidiphilous communitiesbefore the "zwischen- yet certainlyestablished, with the interpolationof an moorwald" stage, which is probably the equivalent acidiphilousmoss-heath stage if conditionsare suit- of the early bog-forestof Minnesota. On the other able. Even this bare statementis enoughto show a hand, the "acid" type may be illustratedby the bogs differencefrom the summarywhich is given by Gates of Cheshire and neighboringcounties. Of all the (1942) for the bogs of northernLower Michigan. types of topographyin England and Wales, that of In the diagram whichpresents the successionalrela- the Cheshiremeres region nmost nearly resembles what tionshipsfor his area, it is shownthat all the lines is found in central Minnesota. It is a countryof of primarysuccession pass throughthe stage of a low relief with fairlv numeroussmall shallow lakes, Chamaedaphne association.Moreover, all lines of suc- and it is here that a floatingpioneer-mat is most cession in Michigan are shownto end in the Thuja coammmmonlyfound. On thesethe "cranberry-bog"type associationwhich Gates statesto be "the climaxasso- of communityusually occurs; it is the Britishrepre- ciation in boggy areas in the region." In view of sentative of the American moss-heath. In British what has been observedin the Minnesotabogs con- hydroseresas a whole the floating-matis less coma- cerningthe prevalenceof Abies balsamea in the late mon as the main pioneer stage than the reed-swamp stages,it does not seem likelythat Thuja is the only dominated by Phragmiites comnmunis and its associates, possible climax association for the Minnesotabogs, whichhave theirroots in the solid substratumbelow thoughit seemsin places, as at Cedar Creek,to form the water. This is one of the major differencesbe- a self-reproducingcommunity, and may, therefore, tween the hydroseresof Minnesota and those of April, 1949 THE BOGS OF CENTRAL MINNESOTA 203 Britain. It is interestingthat both areas include in the temperatureconditions of a region,provided the their floras both Phragmites communis and Carex waterlevel is maintainedat a constantheight. Hence lasiocarpa, yet the relativeimportance of these two the climatic conditions which are unfavorable to species in one area is exactlythe reverseof that in hydroseralprogress will be those which lessen the the other. volume of water which drains into the depression, A secondmajor differencelies of coursein the ulti- that is to say, it will be primarilythe total precipita- mate stages to which the hydroseresare tendingin tion, or closely related variables,which will govern the two areas. For the rainfall in Britain is suffi- the rate of progressof a hydrosere. In the center cient to allow the formationof ombrogenouspeat of an optimumarea for bog development,one may above topogenous peat, sometimeswith, and some- postulatethat conditionsof precipitationwill almost timeswithout a stage of "zwischenmoorwald"and oak always be above the lower limitfor the maintenance forest intervening,as described for the Norfolk of the necessarysteady water levels. At the margins, Broads area by Godwin and Turner (1933). There however,climatic conditions must be less favorable, are of course ombrogenousbogs in the U.S.A., but that is, they are such that the water table levels are all the peat in central Minnesota is topogenously not constantlymaintained, but showfluctuations. This formed. is in fact the interpretationgiven to the observation One of the strikingfeatures in this comparison that the centralMinnesota hydroseres appear to have may be found in the numberof bog species which proceeded by intermittentrather than by steady are commonto both areas. To mentionsome that progress. comeat once to mind,there are Menyanthes trifoliata, From this interpretationat least three questions Scheuchzeria palustris, Carex riparia, C. rostrata and arise. Firstly, has this fluctuatingprogress been manyother Carex species,Vaccinium Oxycoccus, Equi- characteristicof thesebogs all throughtheir history? setum fluviatile, Dryopteris Thelypteris, most of the On the basis of the surface ecology, one cannot Sphagnum spp. and so on. Then again one may find whollyanswer this question,but it seems fair to say examplesof closelyrelated species, or membersof the that it has been characteristicof all the later stages. same family,playing preciselyparallel roles in the These would cover a period lasting for at any rate bog-ecology. The case of Calluna vulgaris and four generationsof spruce trees in the cases illus- Chamaedaphne calyculata has received commental- tratedin Figures 12 and 13, cases whichwere chosen ready. We can comparealso Alnus rotundifolia and as being typical examples of what was generally A. incana, Andromeda polifolia and A. glaucophylla, noticeablein the area as a whole. Betula pubescens, and B. papyrifera. Secondly,are thereany signs that the fluctuations Lastly, there are points of interestwhich arise have any kind of regularityand could,therefore, be fromexamining how the Minnesotabog area is situ- taken as expressionsof some nmoreor less regular ated in relationto a map given by Transeau (1903) climaticrhythm? which indicates the North Americandistribution of Thirdly,supposing that some such rhythmhad been bogs. The optimumarea for bog developmentis operating, is it in any way connected with the shownas the area in whichoccur the greatestnum- fluctuationof precipitationvalues whichis shownin ber of the characteristicbog species. The regionof Figure 9? This coversthe last 60 years, and makes studywhich is shownin Figure 1 lies partly within a very strong impressionof a wave-likevariation and partly outsidethe S.W. boundaryof Transeau's witha period of 40 to 50 years. Very likelythis is optimum area; the boundary crosses Minnesota entirelyfortuitous, but it would be of the greatest roughlyfrom the N.W. cornerof the state to Grants- interestto possess the data whichmight be given by berg (Wis.), near the eastern border of the state. historicalstudies of the vegetationrecord which were Just as with individual species, so with vegetation designedto answerthese threequestions. types one may expect to see marked sensitivityto The subject of climaticrhythms- has naturallybeen small fluctuationsof conditionsat the marginof the always in the mindsof climatologists,but plant ecol- distributionarea. This is exactlywhat may be seen ogists have only gradually come to realize the imi- in the Minnesotabog series. portance of the topic in consideringthe status and There is on the one hand the observationon the historyof existingvegetation. A most stimulating presentspread of the moss-heathto the water-margin presentationof the possibilitiesof the idea, and of in many bogs and the tentativecorrelation of this methodsof studyderived from it, has been presented withlow waterlevels in the reccntpast; on the other by Von Post (1944, 1946). He discusses both hand there are the many observationsof abrupt longerperiod fluctuationsand also the "fine-grained" transitionsbetween one vegetationzone and another, rhythmsobtainable by the study of closely placed and of apparentlyuniform age withina particular pollen samples. Results of this latterkind of study zone. This latterset of observationspoint to a very had been published at least as early as 1935, but unevenmethod of progressionof the hydrosere;that have not, to the writer'sknowledge, appeared in an is to say, to phases of rapid extensionin area of a Enolish text until 1946. These small fluctuationsin vegetationtype, separated by intervals of less or the pollen curves (Betula is the pollen type mainly no extension,but of consolidationand maturation. concerned)might be of the same order of magnitude Broadly speaking, some kind of hydroserewill de- as those hypotheticalones which ioightbe found to velop in any water-containingdepression, whatever have controlledthe Minnesota hog successions. Be 204 VERONAVERONA M.M. CONWAY CONWAY Ecological Monograps ~~~~~~~~Vol.19, No.2 that as it may,there is not muchdoubt that a num- Godwin, H., & V. M. Conway. 1939. The ecology of a ber of rhythmicoscillations of climate,of different raised bog near Tregaron, Cardiganshire. Jour. Ecol. periodicities,will in the end be recognized. We have 27: 313-363. evidence not only from the large-scale fluctuations Godwin, H., & J. S. Turner. 1933. Vegetational suc- associatedwith the several glaciationsof the ice-age, cession at Calthorpe Broad, Norfolk. Jour. Ecol. 21: but we have the repeated minorre-advances during 235-262. the retreatfrom the last glacial maximum,we have Graham, V. 0. 1944. Mushroomsof the Great Lakes Von Post's descriptionof the post-glacial climatic Region. Chicago Acad. Sci. & Chicago Nat. Hist. optimum and subsequent deterioration,and the Mus. Chicago. probablyequivalent xerothermnic theory of American Granlund,E. 1932. De Svenska Hogmossarnas Geologi. authors; we have the work of Cooper (1942) on the Sverig. Geol. Unders. 26C No. 273. advances and retreatsof glaciers in Alaska, and the Grout, A. J. 1928. The moss flora of North America, data of Granlund (1932) on recurrencehorizons in north of Mexico. New York. European bogs, to name outstandingexamples. Kurz, Herman. 1928. The influenceof Sphagnum and other bog mosses on bog reactions. 9: It would seen, therefore,that since the Minnesota Ecology 56-69. area possesses a large numberof bogs still in an Le Roy Andrews,A. 1913. Sphagnales. North Amer- ican Flora 15, part I. undisturbedor almost undisturbedcondition, which show especial sensitivityto climatic fluctuations,it Lindemann,Raymond L. 1941. The developmentalHis- toryof Cedar Creek Bog, Minnesota. Amer. Mid. Nat. would be the ideal centerfor a plant-historicalstudy 25: 101-112. designed to throw light on both ecological and Moore, John W., & R. M. Tryon, Jr. 1946. A prelim- climatichistory. The combinationof pollen analysis inary check-list of the floweringplants, ferns, and and stratigraphyin a really detailed study,even in fern allies of Minnesota. Dept. of Botany, Univ. of a singlebog, would be certainto vieldvaluable data, Minn. even if no climaticrhythm were demonstrable. If, Munns, E. N. 1938. The distribution of important however,such a rhythmcould be discoveredon the forest trees of the United States. U. S. Dept. Agr. basis of the recordof the bog species themselves,it Misc. Pub. 287. should be possible to get some idea of the order of Pearsall, W. H. 1938. The soil complex in relation to magnitude of the rhythmicperiod, provided that plant communities. I. Oxidation-reductionpotentials thereis also a recordof the regionalforest changes, in soils. III. Moorlands and bogs. Jour. Ecol. 26: whichmay serve,in a broad way, as a time-scale. 180-193,298-315. Studies of this sort, carried out on eitherside of Pearsall, W. H., & E. Marjorie Wray. 1927. The the Atlantic,may enableus to decidehow far climates physiology and ecology of the calcifuge habit in in general are characterizedby unceasing rhythms, Eriophorumangustifolium. Jour. Ecol. 15: 1-32. and how far such rhythmsare local in character,or Post, Lennart von. 1944. Pollenstatiska perspektivpa general over the whole earth's surface. They will Jordens klimathistoria. Ymer: 79-113. also constituteone of the means by whichour ideas 1946. The prospect for pollen analysis in the study of the relation betweenclimate and vegetation of the earth's climatic history. New Phytol. 45: 193- may 217. be based ever increasinglyon our knowledge of historicalfacts, rather than on theoreticalassumptions. Priestley,J. H., & Mildred Hinchliff. 1922. The phys- iological anatomy of vascular plants characteristicof peat. Naturalist, pp. 263-267. LITERATURE CITED 1924. Further notes on the vascular plants character- Alway, F. J. 1920. Agriculturalvalue and reclamation istic of peat. Naturalist, pp. 201-208. of Minnesota peat soils. Univ. Minn. Agri. Expt. Sta. Thiel, G. A. 1930. A correlation of marl beds and Bul. 188. types of glacial deposits. Jour. Geol. 38: 717-728. Baas-Becking, L. G. M., & E. Nicolai. 1934. Ecology of Transeau, E. N. 1903. On the geographic distribution a Sphagnum Bog. Blumea 1: 10-45. and ecological relations of the bog plant societies of Bisby, G. R. 1939. Trichodermaviride Pers. ex Fries. northernNorth America. Bot. Gaz. 36: 401-420. Brit. Mycol. Soc. Trans. 23: 149-168. Waterman, W. G. 1926. Ecological problems from the Buell, Murray F., & Helen Foot Buell. 1941. Surface Sphagnum bogs of Illinois. Ecology 7: 255-272. level fluctuations in Cedar Creek bog, Minnesota. Woollett, Edith, Doris Dean, & Helen Coburn. 1925. Ecology 22: 317-321. An ecological study of Smith's bog, CheboyganCounty, Cooper, W. S. 1913. The climax forest of Isle Royale, Michigan. Mich. Acad. Sci. Arts & Letters Papers Lake Superior, and its development. Bot. Gaz. 55: 5: 201-210. 1-44, 115-235. Wynne, Frances E. 1944. Studies in Drepanocladus. 1942. Vegetation of the Prince William Sound Region, IV. Taxonomy. Bryologist 47: 147-189. Alaska; with a brief excursion into Post-Pleistocene 1945. Studies in Calliergon and related genera. Climatic History. Ecol. Monog. 12: 1-22. Bryologist 48: 131-155. Gates, Frank C. 1942. The bogs of Northern Lower Ziegenspeck, H. 1936. In wieweit sind die Sphagnen Michigan. Ecol. Monog. 12: 213-254. und andere Moose befihight im Leben ihre Umgebung Gilman, J. C. 1945. A manual of soil fungi. Iowa auf ein bestimmtes pH-optimum einzustellen? Bot. State College Press, Ames, Iowa. Arch. 38: 1-15. April, 1949 THE BOGS OF CENTRAL MINNESOTA 205 APPENDIX I THE QUINHYDRONE METHOD OF PH CUYUNA BOG KERRICK (Lake Margaret) DETERMINATION Site pH values Mean Site pH values Mean The followingtable gives some data which will 5.50 4.20 show the degree of variabilityin pH value of sam- 5.58 5.58 4.25 4.23 ples of peat fromwithin a small area. At each site, Sedge- 5.65 Moss- 4.25 a meterquadrat was chosen,and fromit six samples mat (5.50 heath 4.10 were taken in the sub-surfacepeat, that is, at a T 5.62 5.66 T 4.35 4.36 5.85 4.62 depthof about 10 cm. below the actual surface. The samples were placed in glass vials, and to three of 5.18 3.90 them (labelled T in the table) toluol was added and Late 5.35 5.35 4.03 4.04 into the peat with a glass rod. This pro- Sedge- 5.53 Picea 4.18 stirred mat 5.45 forest 4.18 cedure was used because it was found by Pearsall T 5.50 5.54 T 4.25 4.24 (1938) that by the use of toluol, the pH and the _5.68 _4.30 redox potentialof the peat could be stabilized,so as to minimizethe errorinvolved in transportingpeat 4.65 4.00 Early 4.70 4.74 Old 4.30 4.22 samples and allowing some time to elapse before moss- 4.87 Picea 4.35 testingin the laboratory. In the presentstudy, the heath 4.52 forest r4.08 sampleswere taken to the laboratoryas soon as pos- T 4.60 4.69 T{4.18 4.21 sible after collection,and stored in a refrigerator; 4.95 J4.38 the pH determinationswere carried out withintwo 4.10 3.85 or threedays of collection. It will be seen fromthe Mature 4.35 4.27 Secondary 3.93 3.97 resultsthat the samplestreated with toluol gave mean moss- 4.35 moss- 4.03 heath 4.38 heath p3.98 X alues close to the means of the untreatedsamples, T 4.57 4.58 T 4.01 4.06 and hencein the later collectionstoluol was not used. 4.80 {4.20 4.25 4.85 4.43 4.47 5.08 5.05 Larix 4.72 Marginal 5.22 forest 4.25 fen (4.95 T 4.35 4.32 T 5.02 5.02 4.35 (5.10

6.00 HINCKLEY BOG 6.02 6.03 -_ _ _- Marginal 6.07 6.60 fen 5.90 6.60 6.64 T 6.02 86.03 Picea 6.71 6.16 forest 6.48 V- T_6.48 6.55 6.68 6.55 6.62 6.61 Marginal 6.67 fen (6.55 T16.55 6.58 16.63 Ecological Monographs 206206 VERONA Al.M. CONWAY ~~~~~~~~~~~~~~~~~~Vol.19, No. 2 APPENDIX II visited in the course of the study. Reference is given not only to the map square, but to the quarter DETAILS OF BOG LOCALITIES of the square, thatis NW, NE, SE, or SW, in which The accompanyingtable gives details from which the localityis situated. The air-photographnumbers it would be possible to find those areas which are referto the collectionin the Universityof Minnesota described in the text and others which were also library.

Air- Locality County Countymap square photograph Comments Cedar Creek Anoka T. 34N, R.23W, 27NW BIM-3-68 Describedin the text. Grandy Isanti T. 36N, R.23W, 1ONW BIX-6-8 MatureLarix forest. Stacey A Isanti T. 34N, R.22W, 12NE BIX-7-6 Narrowpioneer-mat and nioss-heath;drained Larix foreston acid peat. Stacey B and C Isanti T. 34N, R.22W, 11SW BIX-7-6 Pioneer-matsof Decodon. Hinckley Pine T. 41N, R.20W, 28SW BYD-2-137 Picea forest;good marginalfen (Fraxinus dominant). Lesser Pine T. 42N, R.17W, 28SE-SW BYD-4-95 Moss-heath(mainly secondary); Picea forest. Tamarack' Duxbury Pine T. 42N, R.17W, 10SE BYD-4-157 Larir-Piceaforest Kerrick(Lake Pine T. 45N, R.18W, 26SW BYD-6-125 Margaret) Describedin the text. KerrickB Pine T. 45N, R.18W, 34SE BYD-6-125 Cuyuna CrowWing T. 47N, R.28W, 32NE BXT-2-115 Describedin the text. Pug Hole Itasca T. 57N, R.26W, 24SW CIP-6-163 Describedin the text. Grand Rapids B Itasca T. 57N, R.25W, 18SW CIP-6-163 Extensivemoss-heath. Some indications of burning. Grand Rapids C Itasca T. 56N, R.26W, 14SE2 CIP-13-22 Pioneer-matof neutrophiloussedges; mixed Picea-Thujaforest with some Abies.

Lake Alice' Hubbard _SET. 143N, R.35W, BXU-3-41 Describedin the text. Twin Lakes Hubbard T. 143N, R.35W, 3ONW BXS-5-23 Describedin the text. FloatingBog Clearwater T. 143N, R.36W, lINE BXS-4-105 ExtensiveCarex lasimcarpa mat; narrowbog- (Lake Itasca) forestzone (Larir, Picea, Fraxinusetc.) BoutwellCreek Clearwater T. 143N, R.36W, 15NE BXS-5-21 Mlixedold Thuja-Picea-Abiesfore3t; evidl33ce Bog of muchinterference.

Taiss.nall lake had nu name oa themap, bab was the s nallestof three lviaig close togeth3r, the largest being called Tamarauk Lake. 2 Thequarter of the square indicated is theactual position as givenby the air-photo, not the position which would be deducedfrom the county road map, which incorrectlyindicates a bend in the road. 3 LakeAlice is thenearest named topographical feature to thebog, which is notitself related to that lake but occurs around a smalllake situated about a mileto the -eastof Lake Alice, just north of the road.