N NOliniULSNI

z

NVINOSH1IINS S3ldVd8n

NVINOSH1II/MS S 3 I d Vd 8 I1_

^ z

v-V\ W 00

3 I

c <r If'

i § 23

*LI B RAR I ES*ZSMITHSONIAN~J INSTITUTION

N0I1I

1 LIBRARIES SMITHSONIAN

Z s ^

< w ? x5^>

Z H

! CO V CO

i Wl'Ss

INSTITUTION

^osv^

NOIlfUllSNI

NVINOSHJLIIAIS

^osv^

saidvdan

LI BF

cn — cn \ 5 10 ~

nvinoshiiws saiavaan libraries Smithsonian institution noijuuiisni nvinosh.

(/) Z W 2 "•

X

CO

O __

2 > X£2ms£X 2 ^ 2

0 2 (O V Z g * Z ^

SMITHSONIAN INSTITUTION NOIlDlllSNI NVINOSH1IWS S3l9Vdan LIBRARIES SMITHSO!

o ^ — co _ = co =: co

O «» x^uus^ q O

JNVINOSHilHS S3lbvaan UBRARIESZSMITHS0NIAN“'|NSTITUTI0N NOlinillSNI “'nVINOSHJ

' z r~ 2: r- z r*

o ? ro ° ° on v.„

& 5 i3t%i * 3 33

> is) t. dm W/o tefe jz °p

n '\.\SF’ z m x^osv^x ^ m x^osv^x ^ x^vas^x m

SMITHSONIAN "INSTITUTION NOlinillSNI "NVIN0SH1IWS S3 I 9 V9 9 H “ L I B R AR I ES SMITHSOI

* z * co z co z

„ 2 ' VX >' 2 X''ygs*gX > 2 vf ^ _

- CO Z c/3 Z co A Z

NVIN0SH1IIAJS S3 I 9 V9 9 11 LI BRAR I ES SMITHSONIAN INSTITUTION NOlinillSNI NVINOSH.

V* = C/3 “ C/3 “

UJ ‘ _ yXSa *3>v UJ

co x

t\ *

<

“H XatCaLM rjr>/

q v^osv^x ™ XjVAS^X o

SMITHSONIAN INSTITUTION NOlifliUSNI^NVINOSHlIWS2 S3 I ava SlA I B RAR I ES SMITHSO

d - o xsssx - 2

> ii. :id ? (ir"ii > «k^ § im >

- . "-mm

o> ± — „ \ z ^2^ £ — s ^

NVIN0SH1IIAIS^S3 I 9 V9 9 II L I B R AR I ES SMITHSONIAN “INSTITUTION^ NOimillSNI^NVINOSH

z _

< y^soy

5 /$*-

O

x vlS

> > -^pr 2 ^ ^ W ^ ^ XO/VDC

O Z CO V Z - - ^ -. H

SMITHSONIAN INSTITUTION NOlinillSNI NVIN0SH1IWS S3ldVa9IT

o x ^ _ ^ co . c o 5

Li X^vTsOv^X UJ 2 ^Xj- . yj ><tlAvXrx (/> W

% I

ZLIBRARIES°°SMITHSO

a)

cq -

O XsdcX ^ ymisy q

JNVINOSHilWS S3 I ava a n^LI B RAR I ESZ SMITHSONIAN-* INSTITUTION NOlifUllSNI^NVINOSH.

2 r- z r- 2 ■-

O yfCiSTIT//t\ r y^n O _ :k^A "*9 O

n s ^ ^VASV^ m _

SMITHSONIAN INSTITUTION^NOlinillSNI ^NVINOSHlIlAIS^Sa I 9V9 9 IT LI BRAR I ES SMITHSO

Z CO z CO . 2 > ^

the Wisconsin

Academy of

Sciences,

Arts &

Letters

Volume 67

1979

TRANSACTIONS OF THE

WISCONSIN ACADEMY

OF SCIENCES, ARTS

AND LETTERS

Volume 67, 1979

Editor

FOREST STEARNS

Copyright ©1979

Wisconsin Academy of Sciences, Arts, and Letters.

Manufactured in United States of America.

All Rights Reserved.

TRANSACTIONS OF THE

WISCONSIN ACADEMY

Established 1870

Volume 67, 1979

STATE OF THE ACADEMY 1

Robert A. McCabe

PALEOLATITUDE AND PALEOCLIMATE 4

R. H. Dott, Jr.

SOILS AND SURFICIAL GEOLOGY OF FOUR APOSTLE ISLANDS 14

S. J. Cary, P. F. McDowell, L. J. Graumlich

THE CADDISFLIES (TRICHOPTERA) OF PARFREY’S

GLEN CREEK, WISCONSIN 31

Thomas S. Karl and William L. Hilsenhoff

GARRISON LIFE IN THE NOVELS OF CHARLES KING 43

Kathryn Whitford

AGENTS OF THREE NATIONS IN THE

FOX RIVER VALLEY 1634 TO 1840 51

Charles D. Goff

PERCEPTION OF THE PRAIRIE IN A LETTER

FROM PRAIRIEVILLE 63

Charles F. Calkins

THE PHYCOPERIPHYTON COMMUNITY OF THE

LOWER BLACK RIVER, WISCONSIN 69

Michael R. Strenski

PRELIMINARY REPORTS ON THE FLORA OF WISCONSIN

NO. 67 VERBENACEAE— THE VERVAIN FAMILY 78

William E. Tans and Hugh H. litis

A MASS BALANCE OF NITROGEN IN WISCONSIN 95

Dennis R. Keeney

PRELIMINARY REPORTS ON THE FLORA OF WISCONSIN

NO. 68 CAPRIFOLIACEAE— HONEYSUCKLE FAMILY 103

Peter J. Salamun

DROUGHT IN WISCONSIN 1 30

Val L. Mitchell

T. C. CHAMBERLIN: THE KETTLE MORAINE AND

MULTIPLE GLACIATION 1 35

Susan Schultz

100 YEARS OF WISCONSIN PUBLIC WATER SUPPLIES 149

Alexander Zaporozec

CHANGING RAIL PATTERNS IN WISCONSIN 154

Curtis W. Richards

FIELD IDENTIFICATION OF PEROMYSCUS LEUCOPUS AND

P. MANICULATUS WITH DISCRIMINANT ANALYSIS 159

Mark R. Stromberg

THE NORTHERNMOST STATION FOR

ASPLENIUM PINNATIFIDUM 165

Martha G. Hanson and Robert P. Hanson

SENECELLA CALANOIDES JUDAY (CALANOIDA, COPEPODA),

MESOCYCLOPS LEUKARTI CLAUS (CYCLOPOIDA, COPEPODA),

AND DAPHNIA LAEVIS BIRGE (CLADOCERA) IN

INLAND WISCONSIN LAKES 171

Steven E. Mace

WISCONSIN ACADEMY OF SCIENCES, ARTS AND LETTERS

The Wisconsin Academy of Sciences, Arts and Letters was chartered by the State

Legislature on March 16, 1870 as an incorporated society serving the people of

the State of Wisconsin by encouraging investigation and dissemination of knowl¬

edge in the sciences, arts and letters.

OFFICERS

PRESIDENT VICE PRESIDENT — SCIENCES

Robert A. McCabe Kenneth W. Dowling

Madison Cross Plains

IMMEDIATE PAST PRESIDENT

Dale O’Brien

Spring Green

PRESIDENT ELECT

Thompson Webb

Waunakee

VICE PRESIDENT — ARTS

Tom Bliffert

Milwaukee

VICE PRESIDENT — LETTERS

Ruth Shaw Worthing

Fond du Lac

SECRETARY-TREASURER

C. W. Threinen

Middleton

ACADEMY COUNCIL consists of the above officers plus

COUNCILORS-AT-LARGE

TERM EXPIRES 1983

Emily H. Early, Madison

Hugh Highsmith, Fort Atkinson

TERM EXPIRES 1982

H. Clifton Hutchins, Madison

T. N. Savides, Merrimac

TERM EXPIRES 1981

Malcolm McLean, Ashland

Hannah Swart, Fort Atkinson

TERM EXPIRES 1980

David A. Baerreis, Madison

F. Chandler Young, Madison

PAST PRESIDENTS

(Presently serving on Council)

Katherine G. Nelson, Milwaukee

John W. Thomson, Madison

Adolph A. Suppan, Milwaukee

Norman C. Olson, Milwaukee

Louis W. Busse, Madison

APPOINTED OFFICIALS

EXECUTIVE DIRECTOR

AND PUBLISHER

WISCONSIN ACADEMY REVIEW

James R. Batt

Steenbock Center

1922 University Ave.

Madison, WI 53705

ASSOCIATE DIRECTOR

WISCONSIN JUNIOR ACADEMY

LeRoy Lee

Steenbock Center

Madison, WI 53705

MANAGING EDITOR WISCONSIN

ACADEMY REVIEW

Elizabeth Durbin

Steenbock Center

Madison, WI 53705

EDITOR TRANSACTIONS

Forest Stearns

Botany Department

UW Milwaukee, WI 53201

LIBRARIAN

Jack A. Clarke

4232 Helen White Hall

UW Madison

Madison, WI 53706

EDITORIAL POLICY

The TRANSACTIONS of the Wisconsin Academy of Sciences, Arts and

Letters is an annual publication devoted to original papers, preference being given

to the works of Academy members. Sound manuscripts dealing with features of

the State of Wisconsin and its people are especially welcome; papers on more

general topics are occasionally published. Subject matter experts review each

manuscript submitted.

Contributors are asked to submit two copies of their manuscripts. Manuscripts

should be typed double-spaced on 8V2 X 11 inch bond paper. The title of the pa¬

per should be centered at the top of the first page. The author’s name and brief

address should appear below the title. Each page of the manuscript beyond the

first should bear the page number and author’s name for identification, e.g.

Brown-2, Brown-3, etc. Identify, on a separate page, the author with his institu¬

tion, if appropriate, or with his personal address to be used in Authors’ Addresses

at the end of the printed volume.

The style of the text may be that of scholarly writing in the field of the author.

To expedite editing and minimize printing costs, the Editor suggests that the gen¬

eral form of the current volume of TRANSACTIONS be examined and followed

whenever possible. For Science papers, an abstract is requested. Documentary

notations may be useful, especially for the Arts and Letters papers, and should

be numbered for identification in the text. Such notations as a group, should be

separate from the text pages and may occupy one or more pages as needed. Liter¬

ature Cited should be listed alphabetically at the end of the manuscript unless

included in notations. The style of the references will be standardized as in the

current volume to promote accuracy and reduce printing costs.

Figures should be prepared to permit reduction. Lettering should be large

enough to form characters at least 1 mm high after reduction. High contrast glossy

prints, not larger than 8V2 X 11 inches, are most suitable.

Printing is expensive. Each paper will be subject to a per page charge to the

author, the rate being determined yearly. Galley proofs and edited manuscript

copy will be forwarded to the authors for proof reading prior to publication, both

must be returned to the editor within one week. Reprints should be ordered when

proof is returned; forms and instructions will be sent to the authors. Each author

will receive one complimentary copy of the volume of Transactions in which his

work appears.

Papers received on or before November 30 will be considered for the next

annual volume. Manuscripts should be sent to:

Forest Stearns

Editor: TRANSACTIONS

Dept, of Botany

University of Wisconsin-Milwaukee

P.O. Box 413

Milwaukee, WI 53201

HI!

ROBERT A. McCABE

57th President 1979

WISCONSIN ACADEMY OF

SCIENCES, ARTS AND LETTERS

STATE OF THE ACADEMY

Robert A. McCabe

Presidential Address

April 1979

Seven score and 19 years ago our fore¬

fathers brought forth for this State a new

association . . . our Academy. It was dedi¬

cated to the proposition that there was a

need to encourage investigation and dis¬

seminate views of the various departments

of Science, Literature and the Arts. Now we

are engaged in a grave financial crisis testing

whether this academy or any academy so

conceived and so dedicated can long endure

without direct public support — in whose

interest it directs its efforts and its programs.

5|C

I acknowledge paraphrasing our most il¬

lustrious U. S. president for that introduc¬

tion. It would have been a great relief for

me at the outset to recite for you our ac¬

complishments over the past year, compli¬

ment those who have performed well, tell

a light story or perhaps a few jokes and to

have thus amused you without being very

enlightening.

What I am about to relate to you is not

meant to be nor should it be construed as

criticism or complaint, neither is it a herald¬

ing of doom. But . . . our Academy is in

serious financial difficulty. The full impact

of this difficulty has not yet been felt in our

programing or by the membership. Infla¬

tion and a virtually static source of income,

coupled with litigation of the McCoy trust

suspending income from that source are

parties to the problem. In short we have had

to borrow from the capital of our endow¬

ment fund to maintain our program of ser¬

vice. This has been an anathema to the

Academy Council. Belt tightening has al¬

ready begun and more will doubtless follow.

Although some money saving adjustments

have been made, the major effort has been

voluntarily undertaken by our Executive

Director. He has launched a fund raising

campaign with the backing of the Academy

Council. Three quarters of the Director’s

time has been and will continue to be de¬

voted to this effort. Results thus far have

been encouraging but not overwhelming.

Laudable as this effort is both as a per¬

sonal undertaking and as an Academy pro¬

gram for self maintenance, the following

facts must be understood and acceptable to

the membership.

1. That the funds raised and to be raised

are not to increase the endowment, or

repay funds taken from the endow¬

ment; instead, they are to be used to

meet current expenses.

2. That until we adjust our life style or

obtain outside financial aid, this fund

raising effort will be an annual affair.

How long we can expect donors to re¬

peat on an annual basis is yet to be

determined.

3. If the Executive Director does indeed

spend most of his time fund raising,

who then minds the shop? We have a

dedicated and competent office staff,

but it would be unrealistic and unfair

to saddle them with responsibilities not

in their job descriptions.

4. Can we afford, financially and admin¬

istratively, to use our Executive Di¬

rector nine months of each year as a

fund raiser? My personal reply is that

we cannot!

5. Lastly, unless the Academy is the sole

winner in the McCoy court case our

financial woes will become chronic.

1

2

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

If this untenable situation distresses you,

as it does me, then we must delineate and

examine our options. When any company

or organization is losing money, its first

remedial action is to cut back on goods and

services. Our goods are our publications

and our first line responsibility. The Acad¬

emy Transactions by common accord must

not be diminished or diluted, but since most

scientific journals today require a page

charge for published articles, this possibility

is being exploited but is also under careful

study. The Academy Review is our most

popular publication and is perhaps the ma¬

jor offering the Academy makes to the lay

membership. Recent discussions on the Re¬

view have called for expansion, not retrench¬

ment. Our newsletter, the Triforium, is the

Academy tom-tom that keeps the member¬

ship informed of what is happening in the

organization, what events are to take place,

and what the status of the Academy pro¬

grams are at that moment. Who will say

what shall fall to the budget ax? Let him

also say, how, and then produce a consen¬

sus. The savings by reducing publication to

Transactions alone would not save enough

to keep us solvent, and our losses in mem¬

bership support would very likely outweigh

any hoped-for gains.

Raising dues is the standard “easy-out” of

a financial bind and it is also an “out” of

another kind, namely membership drop-out.

The gains of increased levy balanced against

lost members and hence lost participation

and revenue must be carefully considered.

This course should be taken only if it solves

the problem and only as a last resort.

The sum total of cost savings by discon¬

tinuing nonessential service programs now

conducted by the Academy is roughly $200

to $300. The sum is small because the

bulk of the cost is hidden in administration

and not obvious as cash outlay. This brings

us to staff salaries. These salaries by all odds

are not even competitive in today’s job mar¬

ket for the limited staff now conducting

Academy business. Indeed we have not been

able to increase salaries commensurate with

the rise in cost of living. Whatever services

we provide are those generated by the office

staff.

The last drain on our budget is the Steen-

bock Center. We are for all intents and pur¬

poses locked into this facility by today’s

housing costs and appreciating real estate

values. Maintenance costs of the Center

would be about equal to rent charges for less

desirable and certainly less functional quar¬

ters, if we were to sell the Center property.

* * *

So much for background. The basic op¬

tion in my opinion and one under which we

should have been operating from the outset

is: That the Academy be financially sup¬

ported by the State of Wisconsin. Histori¬

cally the Academy was meant to be state

supported and indeed it was in part until

1960. Each year prior to that time the

Academy presented its annual program re¬

quest (a pittance in relation to other re¬

quests) to the legislature and defended the

request. The Academy request was not but¬

tressed by great political clout and so was

undoubtedly on the list of easy-to-cut items.

Eventually by gentle but painful attrition

the small budget was cut . . . completely.

With the advent of the Steenbock bequest

to the Academy and later by the McCoy

trust the need to seek state support through

an annual hassle was diminished and all

effort abandoned. Ironically this occurred at

a time when Academy programs for the

people of Wisconsin were increasing.

On the surface this could be regarded as

a mistake, but if the Academy could sup¬

port itself and eliminate the time consum¬

ing and arduous task of budget preparation

and defense, then the move to become self

supporting was correct. Although the mon¬

etary saving to the state was insignificant,

it served notice to the legislature that we

were trying to help ourselves.

The fact that we assumed a financial

burden that had heretofore been in part a

1979]

McCabe — State of the Academy

3

state responsibility does not abrogate legis¬

lative involvement in the welfare of the

Academy as an agency serving the people

of the state.

If it were at all possible, we would not

now ask for renewal of state support and

conceivably there may come a time when

we will again become self sustaining.

We do not come with hat in hand to re¬

quest help in difficult times but we respect¬

fully ask for financial support because our

cause is just.

We ask for no brick or mortar financing.

We ask for no support of our publications.

We ask for no funding of our programs.

We ask for no remodeling or physical ex¬

pansion monies.

We ask for no land or real estate pur¬

chase.

What we request is a sum sufficient to

maintain our physical plant and an adequate

Academy staff. All other aspects of Acad¬

emy functioning will be covered by mem¬

bership dues and endowment proceeds. It

is rare when any Academy or professional

society can provide services or engage in

programs for public benefit on membership

dues alone.

The Wisconsin Academy of Sciences, Arts

and Letters is 4th in size among 45 state

academies and in prestige we rank near the

top. We are not elitist in membership; any¬

one interested in our three major disciplines

is welcome to join and participate.

We offer a forum for all who speak or

write through our meetings and publica¬

tions.

We have offered and hold open the offer

to provide an unbiased sounding board on

legislative or public policy discussions under

consideration by our elected representatives

at all governmental levels.

We will continue to provide programs

of education and enlightenment to the peo¬

ple of Wisconsin (and elsewhere).

We will exchange our Transactions of the

Academy with over 600 institutions in 60

countries promoting the progress, and the

scholarly stance of our State in the fields of

science, arts and letters.

We will strive to be self-supporting and

frugal with public funding.

We will promote Wisconsin’s people, its

programs and its products by all means at

our disposal.

* * *

The National Academy of Science is sup¬

ported by the U. S. Government and per¬

forms a service to the nation by its response

to problem solving and to policy solutions

for the national welfare.

We ask only that the Wisconsin Acad¬

emy be supported by the State of Wisconsin

so we may likewise respond for the welfare

of our State.

* * *

Thus my presidential address is ended.

The epilogue that I pass on to fellow mem¬

bers of the Academy is this: I will appoint

a committee to prepare an official position

on this basic premise of state support and

to develop a strategy on which to proceed.

There are key people I have in mind and a

tentative plan as a starting point for action,

but the team is not complete. If we are to

strive in this direction every last member will

be asked to help. At the proper time when

all is in readiness the program will be made

a matter of Academy record and member¬

ship participation will be solicited for sup¬

port in an effort to enlighten our legislature

on the necessity and wisdom for providing

state support.

So important is this effort that it must not

rest solely with a small committee of dedi¬

cated people. The Academy rank and file

must become involved. The need is urgent

and our cause justified!

PALEOLATITUDE AND PALEOCLIMATE

R. H. Dott, Jr.

Department of Geology and Geophysics

University of Wisconsin, Madison

A bstract

In light of large-scale displacements of continents through time, restoration

of paleolatitudes is critical to any paleoclimatic reconstruction. Paleomagnetic

studies over the past two decades have provided estimates of paleolatitude that are

more and more widely accepted. It was important, however, to test those estimates

with independent geologic evidence — especially for pre-Permian time for which

paleomagnetic data is less satisfactory. On the other hand, geologic coupled with

paleomagnetic evidence should provide powerful constraints for paleoclimatic

interpretations.

Paleomagnetic data suggest a Paleozoic equatorial position for most of North

America with a progressive relative counterclockwise rotation of the continent of

about 60°. Long-recognized reef and species-diversity patterns support a Paleozoic

low latitude, and Ordovician and Devonian volcanic ash fallout by trade wind

dispersal is consistent with the paleomagnetically-indicated equator positions. Mid¬

continent early Paleozoic paleocurrent and conglomerate size-rounding data also

agree with a tropical location characterized by episodic violent storms. And Permo-

Triassic plant and evaporite evidence is consistent with both latitude and terrain

restorations. As North America moved northward (and more counterclockwise)

since the Triassic, its central part came into the westerly wind belt as evidenced by

fallout of volcanic ash from the Cordillera. Most post-Triassic geologic indicators

of humid-arid conditions, paleocurrents and oxygen-isotope data, are consistent

with latitudinal and terrain restorations.

Introduction

III 1958, when I began teaching, enough

paleomagnetic evidence of ancient latitudes

had accumulated that the student of earth

history could hardly ignore it. Yet, there

were large gaps in the data and some un¬

tidy anomalies. Many, if not most, geolo¬

gists doubted the results, perhaps because

shifting poles or continents seemed non-

uniformitarian. But by the mid-sixties, the

data had begun to show enough of a sys¬

tematic pattern of latitudinal change that

one could begin to consider a number of

geologic implications of the paleomagnetic

results in addition to continental drift. In¬

deed, it became intriguing to compare — and

even test — the paleomagnetic data against

sedimentologic and paleontologic evidence

of latitude. While paleoclimatology still

seems more art than science, it is nonethe¬

less fruitful to examine the consistency of

phenomena.

The distributions of ancient reefs, evapo-

rites, tillites, coal, certain fossil plants, red

beds, and presumed eolian desert deposits

have long been used to try to interpret pa¬

leoclimatic zones and even paleolatitude

(e.g. Koppen and Wegener, 1924). Clearly,

all of these geologic phenomena should pro¬

vide tests of the paleomagnetic data. In 1956

Shotton and in 1960 Opdyke and Runcorn

presented the first comparisons of wind-

influenced cross bedding orientations with

paletomagnetically-indicated latitudes (see

also Opdyke in Nairn, 1961, and in Run¬

corn, 1962). Then Eaton in 1963, Powers

4

1979]

Dott — Paleolatitude and Paleoclimate

5

and Wilcox in 1964 and Wilcox in 1965

showed that fallout distributions of volcanic

ash could reveal prevailing paleowind direc¬

tions, adding still another dimension to

paleoclimatic and paleolatitudinal analysis.

About 1960 I began cataloguing these di¬

verse kinds of data on paleogeographic maps

(Dott and Batten, 1971).

To many the present large mass of paleo-

magnetic evidence of drastically different

past latitudinal positions of continents may

make further comparison of paleomagnetic

and paleoclimatic data seem redundant.

However enough skepticism exists (e.g.

Meyerhoff and Meyerhoff, 1972; Stehli,

1973) to warrant a concise, up-dated assess¬

ment. Although this paper deals almost ex¬

clusively with North America, the approach

is universally applicable. My objective is not

to prove or disprove continental drift, but

only to assess consistency of unlike data.

Distributions of reefs, evaporites, red beds,

coal and the like have been catalogued be¬

fore (e.g. Nairn, 1961; Runcorn, 1962,

Schwarzbach, 1963; Dott and Batten, 1971

and 1976). Therefore, more attention is

given here to paleowind indicators and to a

few stratigraphic intervals that have not pre¬

viously received adequate treatment.

Working Assumptions

Certain climatic assumptions provide a

background. We must first assume that a

significant temperature difference has always

existed between the poles and equator sim¬

ply as a consequence of the difference in

angle of incidence of solar radiation. The

intensity of the pole-equator gradient has

almost certainly varied through time, and

such variations would have greatly affected

world climate. This thermal gradient, cou¬

pled with the rotation of the earth and the

positions of land masses, controls the first-

order patterns of atmospheric circulation.

One consequence of variations of the pole-

equator gradient is the positioning of sub¬

tropical high pressure cells. With a stronger

gradient, these highs are pushed equator-

ward, whereas with a weaker gradient they

move somewhat poleward.

The general climatic importance of the

ratio of land to sea (or continentality) was

appreciated as early as Lyell’s time. It is

well known that the relative area of sea and

land greatly affects the overall heat budget

of the earth because of the 10-15 percent

difference in albedo of land versus water.

Moreover, being fluid, the seas can trans¬

port heat from low to high latitudes. It

follows that the location of a pole in a con¬

tinent or a restricted sea (such as the pres¬

ent Arctic Ocean) will be much colder than

a pole located in an open ocean area.

The spin of the earth and the resulting

Coriolis effect, allow steady trade winds on

either side of the equator to be assumed for

all times. The Westerly wind belts probably

also have always existed, but seemingly they

were more variable than the trades because

of variations of the pole-equator tempera¬

ture gradient and of land masses. Land-

masses greatly influence second-order atmo¬

spheric circulation patterns by localizing or

blocking the more or less permanent high

and low pressure cells. Mountainous barriers

obviously may produce strong rain shadows.

Paleomagnetic Latitudinal

Restorations for North America

Latitudinal shifts of North America are

indicated by paleomagnetic data for the past

700 million years (Fig. 1). The paleo-

equators have been up-dated from those of

Dott and Batten (1971 and 1976). Loca¬

tions are better established for some ages

than for others, nonetheless, the counter¬

clockwise rotation and northward motion of

the continent presents an impressively syste¬

matic picture. Only the details of exact po¬

sition for the more poorly controlled periods

remain to be documented.

Paleoclimatic Phenomena Compared

With Paleomagnetic Evidence

Working backward through time, let us

compare some climatically sensitive geo-

6

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

sf PALEOFLOW- PROBABLY WIND CONTROLLED

ASH FALLOUT

Fig. 1. Ancient latitudinal positions of North America based upon paleomagnetic data com¬

pared with sedimentary data. Note systematic counterclockwise rotation and northward shift

of the continent. (Paleomagnetic results from: Irving, 1964; McElhinny, 1973; Hicken, and

others, 1972; Van der Voo and Williams, 1975; French, and others, 1975; Proke and Har-

■ graves, 1973; McElhinny and Opdyke 1973; Van der Voo and French, 1974. Sedimentary data

after Dott and Batten, 1971 and 1976, who include detailed maps and references.)

EQUATOR AND TRADE WINDS

R-REEFS E-EVAPORITES T-TILLITES

logic phenomena with the paleomagnetic

latitudinal data to see whether or not the

two independent types of evidence agree

paleogeographically. The fallout patterns of

late Cenozoic (especially Pleistocene) vol¬

canic ash erupted in the Cordillera (Wilcox,

1965) and paleontologic evidence of a rain

shadow effect east of the mountains agree

with the present location of that region in

the mid-latitude westerly wind belt (Dott

and Batten, 1971, chapter 14). The same is

true also for Cretaceous and Jurassic ash

fallout (Fig. 1; Slaughter and Earley,

1965). Cretaceous humid, subtropical plant

types (Arnold, 1947; Andrew, 1961) and

oxygen isotope evidence for warm sea tem¬

peratures of 20°-25°C (Lowenstam and

Epstein, 1959) are consistent with a lower

latitude (Fig. 2). Cretaceous plants appar¬

ently adapted for humid conditions in the

present Rocky Mountain region, however,

seem anomalous as they grew leeward of the

rising Cordilleran mountainous mass. Ac¬

cording to Millison (1964), the lack of a

clear rain shadow here is the result of humid

air flowing from the southeast over the ad¬

jacent Cretaceous epeiric seaway and along

the eastern front of the Cordillera. In light

1979]

Dott — Paleolatitude and Paleoclimate

7

of the probable paleolatitude, however, it

appears more probable that moisture was

acquired from that seaway by winds of the

sub-tropical high pressure system flowing

south along the mountain front (Fig. 2).

However, fine ash blown higher into the

atmosphere would be blown straight east by

upper atmosphere winds.

The latest Jurassic Morrison Formation

of the western United States was deposited

to the lee of the embryonic Cordilleran

mountains (Fig. 3). The Morrison accumu¬

lated at about 30°-40° N latitude; it was

within a westerly wind belt as indicated by

volcanic ash that could only have blown

Fig. 2. Generalized Cretaceous paleogeography

and paleoclimatology showing paleomagnetic lati¬

tudinal results, volcanic ash dispersal (wide ar¬

rows), oxygen isotope results, and paleobotanical

evidence. Superimposed are the expected atmo¬

spheric circulation patterns; subtropical highs ap¬

pear at higher latitudes than today because of a

lesser pole-equator temperature gradient. Small

arrows indicate dispersal of warm- and cool-water

faunas. Sea water temperatures are obtained from

oxygen isotope studies. (Modified after Dott and

Batten, 1976). Compare with Figure 1.

from the west. Oxygen isotope studies of

shells from underlying Upper Jurassic (as

well as overlying Cretaceous) marine rocks

indicate warm ocean temperatures around

20°-25°C (Bowen, 1961), at least 10°

warmer than is typical of that latitude today.

Moreover, Bowen (1961) argues that the

maximum latitudinal surface Jurassic sea

water temperature gradient then was only

about 20°C as opposed to 60°C today.

Local evaporites in the same underlying

strata also indicate high evaporation poten¬

tial.

Sedimentary facies, petrography, and pa-

leocurrent data prove that the Morrison was

deposited on an eastward-sloping, low

coastal plain; Dawson (1970) estimated the

Fig. 3. Paleogeography during deposition of the

nonmarine Morrison Formation clastic wedge in

the western United States. The formation was de¬

posited on a coastal plain sloping eastward from

the rising Cordilleran volcanic arc whence ash

was carried by westerly winds. (Adapted from

Dawson, 1970 and Dott and Batten, 1971; see

latter, Fig. 13.19, for ash fallout.)

8

Wisconsin Academy of Sciences , Arts and Letters

[Vol. 67

gradient to have been of the order of 0.5-

1.0 m per km. Widespread fresh-water lime¬

stones attest to a rather wet landscape, and

the large herbivorous dinosaur fossils for

which the formation is famous suggest at

least moderate humidity. Yet, coal is almost

totally lacking and fossil plant material

(chiefly cycad wood and reeds) is rare.

Why? Dead plant material must have been

thoroughly oxidized at the soil surface,

which seems consistent with the abundance

of red- and yellow-colored sediments. Either

the Cordilleran mountains were not yet high

enough to form a strong rain shadow, or the

Morrison coastal plain was relatively dry

except along broad river valleys. Dawson

(1970) suggested a strong seasonal distribu¬

tion. This was based upon the apparent

thorough oxidation of vegetation, well-de¬

veloped growth rings in fossil wood, and the

texture of fluvial conglomerates interstrati-

fied with the dominantly fine-grained Mor¬

rison sediments. To move the largest cobbles

(13 cm), which are now at least 200 km

from their western upland source, maximum

shear velocity must have been at least 50 cm

per second, while the maximum river ve¬

locity near the water surface could have

been roughly 200-600 cm per second.1 Such

large velocities seem surprising for the ap¬

parently low gradient of the Morrison

coastal plain; the most probable explana¬

tion is that the coarse gravel was moved

only incrementally in wide, braided chan¬

nels during seasonal flooding.

Westerly winds are also indicated for lat¬

est Triassic to earliest Jurassic time in the

western United States by the orientations

of large-scale cross bedding in the Navajo

and related sandstones. Whether this famous

sandstone represents entirely eolian or partly

shallow marine dunes — both of which have

been claimed — they must have been de¬

posited under the influence of the prevail¬

ing westerly winds of that time as first sug¬

gested by Poole (1957). 2 If the Navajo is;

in fact, largely eolian, as I now believe, it

would seem to represent a large dune field

Fig. 4. Alternate paleogeographic interpretations of the Navajo Sandstone. A. Classic hy¬

pothesis of a vast eolian dune field, for which the evidence seems strongest; B. Hypothesis of

a combination of shallow marine and coastal eolian dunes (Stanley, Jordan and Dott, 1971).

In either case westerly winds apparently controlled sand dispersal.

1979]

Dott — Paleolatitude and Paleoclimate

9

formed by onshore winds from the north¬

west (Fig. 4A). In its western extent, some

of it appears marine and one can infer that

nearshore dispersing currents were induced

by westerly winds, with a strong southerly

longshore drift component (Fig. 4B), (see

Stanley, Jordan and Dott, 1971; Dott,

1979).

For Permo-Triassic times, both the paleo-

magnetic and geologic data bearing upon

climate are among the most complete.

Tropical to sub-tropical latitudes for North

America are indicated by paleomagnetism

with the paleoequator (Fig. 1) almost ex¬

actly coincident with that inferred by Wege¬

ner a quarter of a century before any paleo-

magnetic data existed. The consequences of

expected atmospheric circulation patterns

for Permo-Triassic North America can be

tested geologically (Fig. 5). Paleowinds

over the western United States as indicated

by widespread Permian cross bedded sand¬

stones (whether eolian or very shallow ma¬

rine) conform well with expected trade wind

orientations. Extensive evaporites in the

central United States conform well with a

down-wind tropical continental margin to

the leeward of high mountains. Permian

plants with features adapted to aridity oc¬

cur in Arizona, while the inferred humid-

tropical lowland plants associated with

Pennsylvanian coals had given way to

drought-tolerant gymnosperms in eastern

United States and humid-temperate forms in

East Greenland (Frederickson, 1972). But,

with Africa and Europe adjacent to eastern

North America how could eastern America

have been humid? One would expect a dry

interior of the Gondwana supercontinent

like central Asia today? Apparently the

Tethys seaway and epeiric seas over central

Europe during part of Permian and Triassic

times supplied moisture to westerly-blowing

trade winds (Fig. 5).

The Carboniferous latitude of North

America seemingly was similar to that of the

Permian. Widespread Mississippian organic

phy and paleoclimatology showing comparison of

equator positions from paleomagnetism with sedi¬

mentary and paleobotanical evidence (see also

Fig. 1). Superimposed are the expected atmo¬

spheric circulation patterns with sub-tropical highs

shifted poleward for the same reason as in Fig. 2.

(After Dott and Batten, 1971 and 1976).

reefs are consistent with a low latitude loca¬

tion, and trade winds apparently influenced

the deposition of cross bedded Pennsylva¬

nian quartz sandstones over the western

United States as it did later in the Permian

(Fig. 1). Unlike Permian time, however,

no strong rain shadow is evident across the

continent, for humid-tropical to sub-tropical

lowland swamp forests cloaked the eastern

interior. This suggests that the Appalachian

mountains were not yet very high. Nonethe¬

less Mississippian evaporites of the Williston

Basin and Pennsylvanian ones in Utah indi¬

cate an evaporation potential that was sig¬

nificant where marine circulation became

restricted.

The abundance and widespread distribu-

10

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

tion of organic reefs in Devonian and Si¬

lurian times long ago provided arguments

for a low-latitude location of North Amer¬

ica, although a good case also could be

made for a weaker pole-equator temperature

gradient and a generally warmer, more uni¬

form climate during a time of small, low

continents (Dott and Batten, 1971, Fig.

11.46 or 1976, Fig. 13.24). Major evapo-

rites attest to high evaporation potential and

areas of restricted circulation. The inferred

tentative windward side of some of the Ca¬

nadian reefs (Andrichuk, 1958) would face

into the trade winds expected for the re¬

stored paleoequator. Finally, Late Devonian

volcanic ash distribution over the eastern

United States conforms to fallout for winds

expected for the paleoequator (Fig. 1).

For the Ordovician, ash fallout from vol¬

canic events in the Appalachian region con¬

forms well with trade winds expected for the

apparent paleoequator (Fig. 1). What little

is known of the dispersal patterns for the

St. Peter Sandstone also conforms to such

Fig. 6. Paieogeographic restoration of the middle Late Cambrian Baraboo (Wisconsin)

islands according to paleomagnetic evidence for a southern tropical setting. Trade winds in¬

ferred from regional cross bedding orientations in shallow marine sandstones. Episodic storm

waves must have approached from all sides judging from the uniform distribution of largest

rounded quartzite clasts (dots). Fine pebbles were dispersed to the left (lee) by normal cur¬

rents following storms (shaded area). (From Dott, 1974.)

1979]

Dott- — Paleolatitude and Paleo climate

11

winds (Dapples, 1955; Dott and Roshardt,

1972). Finally, the great species diversity

and sheer abundance of Ordovician marine

faunal elements is consistent with a tropical

to subtropical setting.

Cambrian strata of the central United

States provide some additional evidence for a

low latitude for early Paleozoic North Amer¬

ica. The paleocurrent pattern, for which

there is abundant data, conforms to trade-

wind-driven shallow marine currents expect¬

ed for the paleomagnetically indicated equa¬

tor. In Wisconsin, the presence of Cambrian

islands that shed coarse gravel has provided

an unusual opportunity to estimate the mag¬

nitude of ancient storm waves and also to

contrast effects of day-to-day processes with

those of so-called rare events. A broad range

of gravel clast diameters from 1 cm granules

up to 8 m boulders occurs, but there is no

rounding of boulders larger than 1.5 m.

Clearly the waves that pounded Cambrian

sea cliffs around Baraboo, Wisconsin were

not capable of moving larger boulders fre¬

quently enough to abrade them. Two slightly

different approaches are available for mak¬

ing quantitative estimates of the Cambrian

storm wave heights.3 Both results indicate

that breakers at least 8-10 m high crashed

upon the islands and tumbled boulders up

to 1.5 m in diameter with moderate fre¬

quency; almost certainly still higher waves

developed on rare occasions. Rounded

boulders occur all around the Baraboo is¬

lands, thus rare-event storm waves must

have approached from all directions, not

just from the apparent normal trade winds

direction suggested by paleocurrent data

(Fig. 6). Trade winds themselves, though

very steady, are not sufficiently strong to

generate the large waves required to move

boulders. Therefore, large storms are indi¬

cated, and, while other types are possible

(e.g. mid-latitude Nor’easters on our present

Atlantic coast), tropical storms seem most

probable in light of the conformance of the

regional paleocurrent pattern with the paleo¬

magnetically indicated latitudinal position

of Wisconsin (Fig. 1).

Conclusions

Comparison of climatically-sensitive ge¬

ologic phenomena with paleomagnetic evi¬

dence for ancient latitude presents generally

consistent restorations. North America ap¬

parently was dominated by Trade Winds

throughout Paleozoic time, but moved into

the Westerly wind zone during Mesozoic

and Cenozoic times. For some eras in North

American (e.g. late Mesozoic, Permo-Trias-

sic and Late Cambrian times), a satisfying

degree of completeness of paleoclimatic res¬

toration is possible. At best, however, such

restorations are largely qualitative. Not all

available data provide a clear picture; the

hardly disputable Eocambrian glaciation is a

case in point. Tillites or tilloids are known

today in Eocambrian strata from all conti¬

nents except Antarctica. Indeed, their wide¬

spread distribution is the problem, for they

seem to occur at all paleolatitudes. Was the

climate so rigorous and completely unzoned

that glaciation occurred from pole to equa¬

tor (Fig. 1)? At present the Eocambrian

paleoclimatic data cannot be satisfactorily

reconciled with paleomagnetic evidence,

which points up the need for additional

comparative studies of both phenomena.

Acknowledgements

My work and that of several former stu¬

dents that bears upon this paper has been

supported by grants from the Graduate

School of the University of Wisconsin,

Madison, from funds donated by the Wis¬

consin Alumni Research Foundation and the

University-Industry Research Program. Rob

Van der Voo of the University of Michigan

was extremely generous in helping me up¬

date the paleomagnetically indicated paleo¬

latitudes, and colleagues in the Department

of Meteorology, at the University of Wis¬

consin, notably R. A. Bryson, E. Wahl, and

J. E. Kutzbach, have been of great help over

12

Wisconsin Academy of Sciences , Arts and Letters

[Vol. 67

the years in trying to educate me on cli¬

matology. I have drawn upon the thesis re¬

search of James C. Dawson, William M.

Jordan and Kenneth J. Stanley, some of

which has not been published. Their work

and many stimulating discussions with them

are gratefully acknowledged.

Literature Cited

Andrichuk, J. M. 1958. Stratigraphy and

facies analysis of Upper Devonian reefs in

Leduc, Stettler, and Redwater areas, Al¬

berta. Amer. Assoc. Petroleum Geologists

Bull., 42:1-93.

Andrews, H. N. 1961. Studies in paleo¬

botany. Wiley, New York, 487 p.

Arnold, C. 1947. An introduction to paleo¬

botany. McGraw-Hill, New York. 433 p.

Blatt, H., Middleton, G. V., and Murray,

R. C. 1972. Origin of sedimentary rocks.

Prentice-Hall, Englewood Cliffs, New Jer¬

sey. 634 p.

Bowen, R. 1961. Paleotemperature analy¬

ses of belemnoida and Jurassic paleoclima-

tology. Jour, of Geol. 69:309-320.

Dapples, E. C. 1955. General lithofacies

relationship of St. Peter Sandstone and

Simpson Group. Amer. Assoc. Petroleum

Geologists Bull. 39:444-467.

Dawson, J. C. 1970. The sedimentology

and stratigraphy of the Morrison Formation

(Upper Jurassic) in northwestern Colorado

and northeastern Utah. Ph.D. thesis. Uni¬

versity of Wisconsin. 125 p.

Dott, R. H., Jr. 1974. Cambrian tropical

storm waves in Wisconsin. Geology 2:243-

246.

- . 1979. The Nugget-Navajo envi¬

ronmental war — can trace fossils help?

(Abstr.) Bull. Amer. Assoc. Petroleum Ge¬

ologists 63:441.

- and Batten, R. L. 1971. Evolution

of the earth. 1st ed., McGraw-Hill, New York

649 p. (2nd ed., 1976, 504 p.).

- and Roshardt, M. A. 1972. Analy¬

sis of cross-stratification orientation in the

St. Peter Sandstone in southwestern Wis¬

consin. Geol. Soc. America Bull. 83:2589-

2596.

Eaton, G. P. 1963. Volcanic ash as a guide

to atmospheric circulation in the geologic

past. Jour. Geophys. Res. 68:521-528.

Frederickson, N. O. 1972. The rise of the

mesophytic flora: Geoscience and Man, IV:

17-28.

French, R. B., Grubbs, K. L., and Van der

Voo, R. 1975. Paleomagnetism of Cam¬

brian dikes from Colorado (abs.). EOS,

Trans. Amer. Geophys. Union, 56:353-354.

Hicken, A., Irving, E., Law, L. K., and Hastie,

J. 1972. Catalogue of paleomagnetic di¬

rections and poles. Publ. Earth Physics

Branch, Dept. Energy Mines and Resources,

Ottawa, v. 45, 135 p.

Irving, E. 1964. Paleomagnetism and its

application to geological and geophysical

problems. Wiley, New York. 339 p.

Koppen, W. and Wegener, A. 1924. Die

klimate der Geologischen Vorzeit. Berlin,

Borntraeger. 255 p.

Lowenstam, H. A. and Epstein, S. 1959.

Cretaceous paleo-temperatures as deter¬

mined by the oxygen isotope method, their

relations to and the nature of rudistic reefs.

In El Sistema Cretacico (primer tomo), 20th

Internat. Geol. Cong., Mexico, p. 65-76.

McElhinny, M. W. 1973. Palaeomagnetism

and plate tectonics. Cambridge Univ., Cam¬

bridge. 357 p.

- and Opdyke, N. D. 1973. Remag¬

netization hypothesis discounted: a paleo¬

magnetic study of the Trenton Limestone,

New York state. Geol. Soc. America Bull.

84:3697-3708.

Meyerhoff, A. A. and Meyerhoff, H. A.

1972. The new global tectonics: major in¬

consistencies: Amer. Assoc. Petroleum Ge¬

ologists Bull. 56:269-336.

Millison, C. 1964. Paleoclimatology during

Mesozoic time in the Rocky Mountain area.

Mountain Geologist, 1:79-88.

Nairn, A. E. M. (editor). 1961. Descriptive

paleoclimatology. Interscience, New York.

380 p.

Opdyke, N. D. and Runcorn, S. K. 1960.

Wind direction in the western United States

in the late Paleozoic. Geol. Soc. America

Bull. 71:959-972.

Poole, F. G. 1957. Paleo-wind directions in

late Paleozoic and early Mesozoic time on

the Colorado Plateau as determined by

1979]

Dott — Paleolatitude and Paleoclimate

13

cross-strata (abs.). Geol. Soc. America Bull.

68:1870.

Powers, H. A., and Wilcox, R. E. 1964.

Volcanic ash from Mt. Mazama (Crater

Lake) and Glacier Peak. Science 144:1334-

1336.

Proko, M. S. and Hargraves, R. B. 1973.

Paleomagnetism of the Beemerville (New

Jersey) alkaline complex. Geology 1: 185-

186.

Runcorn, S. K. (editor). 1962. Continental

drift. Academic Press, New York (Internat.

Geophys. Ser. v. 3) 338 p.

Schwarzbach, M. 1963. Climates of the

past — an introduction to paleoclimatology.

Van Nostrand, London. 328 p.

Shotton, F. W. 1956. Some aspects of the

New Red desert in Great Britain. Liverpool

and Manchester Geol. Jour. 1:450-465.

Slaughter, M. and Earley, J. W. 1965. Min¬

eralogy and geological significance of the

Mowry bentonites, Wyoming. Geol. Soc. of

America Spec. Paper 83, 116 p.

Stanley, K. O., Jordan, W. M., and Dott, R.

H., Jr. 1971. New hypothesis of Early

Jurassic paleogeography and sediment dis¬

persal for western United States. Amer.

Assoc. Petroleum Geol. Bull. 55:10-19.

Stehli, F. G. 1973. Review of paleoclimate

and continental drift. Earth Sci. Rev. 9:1-18.

Wilcox, R. E. 1965. Volcanic-ash chron¬

ology. In Wright, H. E., Jr., and Frey, D.

G. (editors), The quaternary of North

America. Princeton Univ. Press, Princeton,

N.J., p. 807-816.

Van der Voo, R. and French, R. B. 1974.

Apparent polar wandering for the Atlantic-

bordering continents: Late Carboniferous to

Eocene. Earth Sci. Rev. 10:99-119.

- , and Williams, D. W. 1975. Paleo¬

magnetism of the Cambrian Wilberns For¬

mation (Texas) and the Cambrian and

Hadrynian poles for the North American

Continent (abs.). EOS, Trans. Amer. Geo¬

phys. Union, 56:596.

Notes

'Bed shear velocity (U*) was estimated from

the relation U* = V r/p where r is shear stress

(derived from the Shield’s Diagram) and is fluid

density. The lower limit for maximum channel

velocity above the bed was determined from the

Hjulstrom-Sundborg graph, while the upper limit

was estimated from the relation

assuming a value of about 0.05 for f for a grav¬

elly bed. (See Blatt and others, 1972). This equa¬

tion is suspect for hindcasting channel flow, how¬

ever, because the friction factor (f) is based upon

flow in pipes; it is very sensitive to bedforms and

hydraulic radius as well as to grain size and ve¬

locity in channels.

2 Some prominent cross bedded sandstones long

interpreted as eolian have been reinterpreted as

shallow submarine sand wave deposits like those

of the present Georges Bank, North Sea, Irish Sea

and English Channel (Stanley, Jordan and Dott,

1971). When they were first interpreted to be

eolian, the existence of underwater dune forms

was unknown. Because in shallow epeiric seas

currents are so strongly influenced by prevailing

winds, the orientations of cross strata formed in

submarine dunes also should reflect such winds

unless tidal currents (or the Ekman effect) com¬

pletely masked the wind’s influence on sand dis¬

persal. The thorny issue of environment of these

sandstones is sidestepped in this paper because

none of the cross bedded sandstones in question

show any definitive tidal features.

3 Breaker height can be estimated from experi¬

mental data for breakwater design derived in large

wave tanks, and from empirical equations relating

threshold velocity to boulder diameter and ve¬

locity to breaker height:

Ut = 9d12

and Cb = 2Ut and Hb

Where Ut is threshold velocity to move boulders

of diameter d; cb is solitary wave velocity at the

point of breaking, and Hb is height of breaker.

(Dott, 1974).

SOILS AND SURFICIAL GEOLOGY OF

FOUR APOSTLE ISLANDS

S. J. Cary, P. F. McDowell and L. J. Graumlich

Department of Geography

University of W isconsin-Madison

Abstract

The soils and surficial geology of Rocky, Oak, York, and Raspberry Islands,

members of the Apostle Islands group, were surveyed in 1976 and 1977. 1 The con¬

siderable diversity of environmental conditions affecting soil development has given

rise to complex soilscapes. Most common on Rocky, York, and Raspberry Islands

are fertile deciduous forest soils (Eutroboralfs of the Alfisol order) formed in de¬

posits of fine-grained glacial and glacio-lacustrine material. On Oak Island the

soils are primarily sandy Podzols (Harplorthods of the Spodosol order) formed

in coarse beach deposits more than one meter thick. Since moderate slopes of 5

to 20% provided the best opportunity for thick sand and gravel accumulations

in the ancient beach zones, the pedologic uniqueness of Oak Island is explained

by the sloping surfaces it presented to subsiding post-glacial lake levels.

Introduction

The Apostle Islands are located in the

northernmost part of Wisconsin, clustered

at the tip of the Bayfield Peninsula. Except

for Madeline Island, the Apostle Islands

are accessible only by boat. Isolation of the

islands has protected some of the best ex¬

amples of wilderness remaining in the east¬

ern United States. Although logged as re¬

cently as 50 years ago, the islands still har¬

bor some stands of mature northern mesic

forest. Many species of native animals also

inhabit the islands and are relatively undis¬

turbed.

In recognition of the special quality of the

recreational resources present in the area,

the U. S. National Park Service has been

assigned jurisdiction over most of the islands

(see Fig. 1). To provide a basis for deter¬

mining the recreational carrying capacity

1 Work was done under the general direction of

Prof. R. T. Brown, Michigan Technological Uni¬

versity, and funded through that institution by

the U. S. National Park Service. Work was guided

by Prof. F. D. Hole, University of Wisconsin-

Madison. Terminology is from Soil Taxonomy

(Soil Survey Staff, 1975).

the Park Service launched a program of re¬

source evaluation studies. Results of study

of soils and surficial geology of Bear Island

were reported by Kowalski (1976). This

paper presents observations made during

two field seasons on York, Raspberry,

Rocky, and Oak Islands. Descriptions and

maps of the major soils are supplemented

with a discussion of the environmental fac¬

tors which influenced soil development. Fea¬

tures of the soil landscapes and an impor¬

tant relationship between soils and topog¬

raphy are also discussed.

Geologic History of the

Apostle Islands Area

Although tectonically stable at present,

the Lake Superior region in early Precam-

brian time (about 3 billion years ago) was

a region of active mountain building. The

crystalline rocks which constitute the bed¬

rock in northern Wisconsin and most of

Canada are the fluvially and glacially eroded

remnants of the once broad mountain belt.

Further tectonic activity opened a huge

trough, or graben, which eventually filled

with thick deposits of sand and gravel

14

1979] Cary, McDowell and Graumlich — Soils and Geology of Apostle Islands

15

eroded from the nearby highlands and de¬

posited in the Precambrian sea (Martin,

1965). Excavation of some of the sandstone

over many millions of years by rivers and

glaciers created the Lake Superior basin.

The Bayfield Peninsula and all of the Apos¬

tle Islands are underlain by portions of the

remaining sandstones belonging to the Bay-

field group.

Much of the Lake Superior basin was

hollowed out during the Pleistocene Epoch

—the most recent 3 million years of geo¬

logic history. Continental glaciers repeatedly

advanced southward from Canada, follow¬

ing and deepening the lowlands in which

the Great Lakes now lie. Evidence from

the most recent glacial advance indicates

that after ice filled the Lake Superior basin

the Bayfield Peninsula upland forced the

growing ice mass to split. The Superior lobe

16

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

CA

CO

c

o

CO

>

Q)

<D

Oak Is.

Rocky Is.,

Raspberry

York Is.

Is.

years before present (xIOOO)

Fig. 2. Postglacial changes in the level of Lake Superior; names of some abandoned shore¬

lines are given. The topmost elevation of the four islands studied is given at right (after

Farrand, 1969).

flowed westward while the Chippewa lobe

filled Chcquamegon Bay and spilled into

what is now the Chippewa River basin. The

Apostle Islands were probably separated

from the mainland by glacial enlargement

of preglacial river valleys (Collie, 1901;

Irving, 1880). Its interlobate position caused

the Bayfield highland to receive great vol¬

umes of glacial debris. In general, the cur¬

rent land surface of kettles, kames, and mor¬

aines is more than 120 m (400 ft) above

the bedrock. Some of the till deposited by

the ice is very sandy and acid, but other till

bodies arc very clayey and somewhat cal¬

careous. The contrasting source materials,

Bayfield sandstone and Pleistocene lake

sediments, arc responsible for the differ¬

ences.

Retreat of the last continental glacier be¬

gan about 12,000 years before the present

(BP). As the ice mass wasted back into the

Lake Superior lowland, meltwater accumu¬

lated between the ice front and the southern

rim of the basin. Glacial Lake Duluth sta¬

bilized at about 329 m (1,085 ft) above sea

level and drained south at Solon Springs,

Wisconsin into the St. Croix and Mississippi

Rivers. These conditions lasted until about

11,000 years BP (Farrand, 1969, p. 195),

at which time all of the Apostle Islands

were submerged. The ice mass then wasted

rapidly, opening lower eastern outlets into

the Lake Michigan basin. Subsequent rapid

lake level decline was marked by only

minor halts until about 8,000 years BP,

when the low-water Houghton stage, 114 m

(375 ft) above sea level, was reached. In

about 3000 years the water level fell about

210 m (700 ft), for an average rate of 70

cm (2.33 ft) per decade, or almost 8 cm

(3 in) per year (Fig. 2).

During the first part of this period of lake

level recession all of the Apostle Islands

emerged. The abandoned strandlines visible

on the islands, including the Highbridge,

Moquah, Washburn, Manitou, and Beaver

1979] Cary, McDowell and Graumlich

Bay shorelines named by Farrand (1960),

were formed between 11,000 and 8,000

years BP. At the time of the Houghton

stage water levels were so low that the Apos¬

tle Islands were part of the mainland. Buried

peat found in places beneath 12 m (40 ft)

of water and 4 m ( 13 ft) of sand marks this

lake level minimum among the Apostle Is¬

lands (Taylor, 1931).

Lake level gradually rose for the several

thousand years which followed as isostatic

rebound lifted the outlet at Sault Ste. Marie,

Michigan. The Nipissing stage, only slightly

higher than the present water level, was

reached at 5,500 years BP (Saarnisto, 1975,

p. 312). Beaches formed at that time are

difficult to distinguish from current Apos¬

tle Islands beaches, as they are only 1 m

(3 ft) higher than the present lake eleva¬

tion of 182 m (602 ft) above sea level

(Farrand, 1969).

As soon as dry land became available,

colonization by tundra and boreal forest

plants and animals began. Spruce was the

dominant tree in these early woodlands, but

as climate ameliorated, spruce was grad¬

ually superceded by pine. Eventually the

transition was made to the present northern

mesic forest, which contains both hardwoods

and conifers (Saarnisto, 1974; Maher,

1977). Thus Quaternary geologic history

provided diverse initial conditions for de¬

velopment of the soils presently found on

the Apostle Islands.

Environmental Factors Affecting

Soil Formation

Initial Material

The recent geologic history of the Apos¬

tle Islands is one of continental glaciation,

inundation by postglacial lake waters, and

reexposure. In most places, deposits of gla¬

cial till are covered by lake sediments of

varying thickness. High water levels in the

Lake Superior basin and copious amounts

of sediment from melting glacial ice pro¬

moted deposition of clay and silt in deep

-Soils and Geology of Apostle Islands 17

water and coarser-textured material in shal¬

lows and on beaches. Typically one would

expect to find lacustrine sediments deposited

in a vertical sequence which coarsens up¬

ward because lake levels were falling, but

in many places there are only small amounts

of sand or gravel overlying clayey material.

Some soils formed in thin coarse deposits

over fine-grained lake sediments or till

(which are often difficult to tell apart),

while other soils formed in thick sand and

gravel deposits.

Surface modification of the islands, since

their emergence from receding glacial lake

waters, has been chiefly by stream activity.

On Oak Island, deep ravines are prominent.

In the ravine bottoms one finds small allu¬

vial terraces and narrow incipient flood-

plains.

Time

All soil formation presently evident on

the Apostle Islands has taken place in the

last 11,000 years. As soon as the topmost

points on the islands were exposed, weather¬

ing, plant colonization, and soil formation

(pedogenesis) began. The only major varia¬

tion results from recent increasing emer¬

gence of the lower parts of the islands. Soil

formation has proceeded for a longer time

on the top of Oak Island than on low-lying

York Island, for example. However, this

potential time difference is not very great

(Fig. 2). Rapid lake level recession exposed

all land elevations in about 2.000 years.

The 9,000 years that have since elapsed

have been sufficient to obscure any time-

related differences between the soils at high

elevations and those on the lower flanks of

the islands. This is particularly true for

Spodosols, which may be formed in only a

few hundred years (Buol, Hole, and Mc¬

Cracken. 1973. p. 254). On Oak Island,

spodosols with well-cemented ortstein hori¬

zons, indicators of advanced pedogenesis,

were identified throughout the full range of

elevations.

18

Wisconsin Academy of Sciences , Arts and Letters

[Vol. 67

Formation of soil horizons depends on

downward water movement. Soil develop¬

ment has not proceeded as far on steep

slopes where much water runs off as on

more nearly level areas where most water

percolates into the soil. Thus, rates of pedo¬

genesis are different in different landscape

positions.

Topography

There are considerable differences be¬

tween islands with respect to topography.

Most of the 22 islands in the group are rela¬

tively level and low-lying; York, Raspberry,

and Rocky Islands are good examples. On

these islands runoff is limited and drainage

networks are poorly developed. Most rain¬

fall and melted snow must move downward.

Such infiltration is often slow, as lacustrine

clays are frequently near the soil surface

where they impede drainage. The subdued

relief provides relatively uniform micro¬

climatic and hydrologic environments for

plant growth and soil formation.

Oak Island, on the other hand, is com¬

paratively rugged. Although it is less than

19 km2 (8 mi2) in area, it has almost 150

m (500 ft) of relief. The sloping landscape

is cut by numerous deep ravines. A diversity

of microclimates, microhabitats, and soils

matches the complexity of landform.

Climate

Although the Apostle Islands, at nearly

47° latitude, constitute the northernmost

land in Wisconsin, the climate is more tem¬

perate than in many areas to the south. The

moderating influence of Lake Superior

causes cool summer temperatures, relatively

mild winter temperatures, and growing sea¬

sons of more than 150 days — much longer

than in most of northern Wisconsin (Finley,

1975). Like much of Wisconsin the area of

the Apostle Islands receives about 0.75 m

(30 in) of precipitation each year. Most of

this precipitation falls as rain in the summer

when convectional storms are most frequent.

By comparison, winters are relatively dry,

although snowfall usually exceeds 1.75 m

(70 in) (Finley, 1975). Northwesterly winds

bring cold, dry Canadian or arctic air which

gains moisture from Lake Superior and oc¬

casionally drops snow. Springs and autumns

are moderately wet. In summary, the Apos¬

tle Islands experience moderately long grow¬

ing seasons and receive enough moisture to

provide significant downward movement of

water for leaching.

Organisms

The dominant plant communities on the

Apostle Islands are northern mesic forests

(Curtis, 1959). Although heavily logged in

the early 20th Century, the second and third

growth forests appear similar to presettle¬

ment composition. Important tree species

are sugar maple ( Acer saccharum ), hem¬

lock ( Tsuga canadensis) , yellow birch ( Be -

tula lutea), paper birch ( Be tula papyrifera) ,

and basswood (Tilia americana ). Locally

common are red oak ( Quercus borealis ),

white pine ( Pinus strobus), white cedar

( Thuja occidentals) , and balsam fir ( Abies

balsamea). Red oak and hemlock were se¬

lectively logged during the 1930’s (Beals

and Cottam, 1960). While conifers and

hardwoods make up the canopy, the forest

floor is usually heavily populated with sugar

maple seedlings. The Canadian yew ( Taxus

canadensis) was once a major ground cover

species, but over-browsing by deer has

made it rare on some islands (Beals, 1958).

Scarcity or absence of deer on some of the

smaller islands has allowed yew to remain

dense. The importance of coniferous and

hardwood vegetation and litter to soil de¬

velopment is discussed by Buol, Hole, and

McCracken (1973). Also important on the

Apostle Islands is the high frequency of

wind-throw. When trees are uprooted by

strong winds, tip-up mounds are formed.

This process creates depressions and mounds

and churns the soil locally. The impact of

the process on soil horizon development has

1979] Cary, McDowell and Graumlich — Soils and Geology of Apostle Islands

19

been studied by Gaikawad and Hole ( 1961 )

and Graumlich (1978).

Soils

Since the landscape of the Apostle Islands

is geologically young, soil forming processes

have had relatively little time to work. Con¬

sequently, soil development has usually af¬

fected no more than the upper 1.3 m (4 ft)

of surficial material. Soil profiles are shal¬

low. As stated earlier, the climate pro¬

vides sufficient water to ensure soil leaching,

a regime conducive to downward transloca¬

tion of nutrients and organic colloids as part

of the process of podzolization. Similar

downward movement of colloidal clay is

termed lessivage (Buol, Hole, and Mc¬

Cracken, 1973). Leaching occurs in most

humid climates. If nutrient-rich litter and

organic matter is not continually added to

the soil surface, podzolization will create a

Spodosol (Podzol), an infertile soil. Spo-

dosols are commonly found associated with

coniferous vegetation and loamy or sandy

parent materials.

Most evergreen trees produce nutrient-

poor, acid litter. Coarse-grained parent ma¬

terial is easily leached and is incapable of

retaining many nutrients necessary for plant

growth. The profile of a Spodosol usually

has an ashy grey, well leached zone at the

top over a dark brown horizon of illuvial

organic matter, iron and aluminum oxides,

and some colloidal clay. Beneath that spodic

horizon the concentration of illuviated ma¬

terials decreases and the brown color fades.

At depths of 1 m or more the geologic ma¬

terial has not been significantly affected by

pedogenesis.

Spodosol formation is inhibited if base-

rich litter is available to the soil, permitting

organic matter to accumulate. Most decidu¬

ous hardwood trees, particularly sugar ma¬

ple, supply abundant fertile litter to the

forest floor. Substantial amounts of silt and

clay, if present in the soil profile, allow re¬

tention of nutrients within reach of plant

roots and soil organisms. The soil which

forms under these conditions is called an

Alfisol (Gray-Brown Podzolic). The vertical

profile differs from that of a Spodosol in

that the B2 or illuvial horizon (called an

argillic horizon) is largely a zone of clay

accumulation rather than of iron and organic

material. Also characteristic of Alfisols is

a dark surface horizon rich in organic mat¬

ter, 7-16 cm (3-6 in) thick, overlying the

grey leached horizon.

In the Apostle Islands, conditions pro¬

mote development of Spodosols in some

areas and Alfisols in others. Where glacial or

lacustrine fine-grained material is at or near

the surface, Alfisols have formed. The initial

material was somewhat calcareous and the

clays have retarded leaching of the calcium

and other cations. Where surface drainage

is good these soils are termed Typic Eutro-

boralfs (Hibbing and Ontonagon series)

(Fig. 3a). Where drainage is somewhat

poor these soils are termed Aquic Eutro-

boralfs (Rudyard and Selkirk series). Eu-

troboralfs are fertile northern Alfisols and

are extensive on four of the five mapped

Apostle Islands.

Sand mixed with gravel occurs in deposits

which may exceed 1 m in thickness. These

materials have experienced long-continued

podzolization and typically have formed

Spodosols. On steep slopes, where much

water runs off, the development of soil hori¬

zons has been retarded, giving the soils a

young appearance. In other areas, the spodic

horizon and podzolic profile are well formed.

These two kinds of soils are classified as

Entic (young) and Typic Haplorthods (Bi-

bon, Vilas, Rubicon, Rousseau, and Karlin

series). Moderately well drained soils in the

catena are included in the Typic Haplor¬

thods (Croswell and Orienta series) (Fig.

3b), but somewhat poorly drained mem¬

bers are Typic Haplaquods (AuGres series).

Extremely gravelly deposits associated with

ancient beaches have also given rise to Typic

Haplorthods (Pence, Waiska, and Allouez

20

Wisconsin Academy oj Sciences, Arts and Letters

[Vol. 67

Fig. 3a. Schematic diagram of a Typic Eutro-

boralf on the Apostle Islands. The soil is well-

drained, fertile, and characteristic of northern de¬

ciduous forests.

Fig. 3b. Schematic diagram of a Typic Hap-

lorthod on the Apostle Islands. The soil is sandy,

well-drained, infertile, and characteristic of conif¬

erous forests.

series) (Fig. 3c). Typic Haplorthods rep¬

resent the central concept of the Spodosol

soil order.

In many areas, coarse materials overlying

the glacial and lacustrine clays are of only

moderate thickness (.3-1 m). There soil

formation can be complex. The usual se¬

quence of events includes leaching and for¬

mation of a spodic horizon in the coarse

upper layer followed by leaching and clay

concentration at and below the clay contact.

The appearance is that of a Spodosol formed

on top of an Alfisol (Fig. 3d). Such soils are

termed “bisequal” (Hole, 1976). Bisequal

soils are one form of intergradation between

Spodosols and Alfisols; appropriately they

are called Alfic Haplorthods (Superior, Dry-

burg, and Manistee series) when well

drained, and Aqualfic Haplorthods (Allen¬

dale and Dafter series) when somewhat

poorly drained.

In scattered depressions drainage is so

poor that pedogenic processes have done

little in 10,000 years but add organic ma¬

terial. These youthful soils are placed in the

Inceptisol order and are called Mollic Hap-

laquepts (Munuscong series) because of

the thick organic-rich surface horizon simi¬

lar to that of Mollisols formed under prairie

vegetation.

1979] Cary , McDowell and Graumlich — Soils and Geology of Apostle Islands

21

Description

forest litter

pale, leached loamy sand

iron-rich and humus-

rich sandy loam

pale, leached sandy loam

reddish, blocky clay loam

clayey sediments,

slightly altered

Fig. 3c. Schematic diagram of a Typic Hap-

lorthod formed in gravelly beach deposits on the

Apostle Islands.

Fig. 3d. Schematic diagram of an Alfic Hap-

lorthod formed in surficial beach sediments and

underlying clayey sediments on the Apostle Is¬

lands. The soil is well-drained and has character¬

istics of both Alfisols and Spodosols.

Since soils are three-dimensional bodies

(Buol, Hole, and McCracken, 1973) study

of horizontal soil variability is important.

The soil maps of Raspberry, York, Rocky,

and Oak Islands (Figs. 4, 5, 6, 7) show

major soil bodies. The soil pattern is more

complex than can be mapped at the scales

indicated. Not shown are inclusions of

smaller soil bodies and the gradual and in¬

terfingering nature of lateral boundaries for

which a drawn line is a gross simplification.

Also designated for each soil body is the

approximate land surface slope. Level land

is placed in slope category a (0-2% ), while

land inclined at 20-30% is placed in slope

category e. The steepest land, with slopes

greater than 30%, usually occurs in associa¬

tion with bluffs or ravine systems, shown as

land types.

The physiographic uniqueness of Oak

Island is evident in the slope designations.

Oak Island has a large part of its land sur¬

face in moderate slopes in contrast to York,

Rocky, Raspberry, and most other Apostle

Islands, which are chiefly level (0-2%) and

gently sloping (2-6%) land (Fig. 8).

The soil maps reveal that Oak Island is

also pedologically distinct. The soilscapes of

the three smaller islands are dominated by

Eutroboralfs and Alfic Haplorthods — soils

22

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

Table 1. Legend for figures 4, 5, 6 and 7.

Eutroboralfs

A Typic Eutroboralf

(Ontonagon, Hibbing)

B Aquic Eutroboralf

(Rudyard, Selkirk)

Haplorthods

C Alfic Haplorthod

(Superior)

D Alfic Haplorthod

(Manistee, Dryburg)

E Aqualfic Haplorthod

(Allendale, Dafter)

Haplaquepts

F Mollic Haplaquept

(Munuscong)

Haplorthods

G Typic Haplorthod

(Bibon)

H Aquic Haplorthod

(Orienta)

I Typic (Entic) Haplorthod

(Vilas, Rubicon)

J Typic (Entic) Haplorthod

(Croswell)

L Typic Haplorthod

(Rousseau, Karlin)

M Typic Haplorthod

(Waiska, Pence, Allouez)

Haplaquods

K Typic Haplaquod

(AuGres)

Miscellaneous Land Types

V Shallow to bedrock

W lake sand

X organic material

Y ravine complex

Z bluff

Slope Classes

a 0-2%

b 2-6%

c 6-12%

d 12-20%

e 20-30%

SOIL MAP OF

Fig. 4. Soil map of Raspberry Island.

1979] Cary, McDowell and Graumlich — Soils and Geology of Apostle Islands

23

Fig. 5. Soil map of York Island.

24

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

Hg. 6. Soil map of Rocky Island.

SOIL MAP OF NORTHERN

1979] Cary , McDowell and Graumlich — Soils and Geology of Apostle Islands

25

Fig. 7a. Soil map of northern Oak Island.

SOIL MAP C

SOUTHERN

OAK ISLAND

26

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

Fig. 7b. Soil map of southern Oak Island.

Cary , McDowell and Graumlich — Soils and Geology of Apostle Islands

27

1979]

slope class (%)

Fig. 8. Proportion of land area on Raspberry, Rocky, York and Oak islands

in different slope classes.

formed in less than 1 m of coarse sandy ma¬

terial overlying glacial or glaciolacustrine

clayey material. The coarse surficial layer,

deposited by beach processes at times of

higher water levels, is generally thin (less

than 1 m) on these islands. On the other

hand, more than 75% of the surface of Oak

Island is covered with Typic Haplorthods

or less well drained Spodosols, which typi¬

cally form in thick (more than 1 m) de¬

posits of coarse-grained materials. In only

a few locations on Oak Island does the basal

clayey sediment lie close enough to the sur¬

face for soil formation. However, the clayey

substrate can significantly affect pedogenesis

even if it is buried by more than 1 m of

coarse beach deposits. Indirect evidence of

subsurface clayey material, in the form of

mottling was often present in profiles which

were of loamy or sandy texture throughout

their upper 1.5 m (5 ft). Oak Island soils

are sandy and sloping, while the soils of

Raspberry, Rocky, and York Islands are

generally fine-textured and fairly level.

More detailed analysis suggests that sandy

soils are associated with sloping land while

clayey soils are associated with level land

(Table 2). Observation points were selected

by randomly overlaying grids (adjusted for

differences of scale) on the soil maps; ob¬

servations were recorded at each grid point.

If no relationship existed between thick¬

ness of coarse surface deposits and land sur¬

face slope, the observation frequencies

should resemble those of Table 3. A con¬

tingency table analysis was performed to

test for a possible relationship. The Chi

Square statistic (x2) was calculated accord¬

ing to the formula

X2 - X fo2/fe - N

where f0 is the observed frequency at a given

position in the table, fe is the expected fre¬

quency assuming no relationship between

the variables, and N is the total number of

observations (252) (Blalock, 1972). The

calculated chi square value was 62.6, highly

28

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

Table 2. Frequencies (%) for given combina¬

tions of land surface slope and surficial sand

thickness, includes all points on Rocky, Oak, York

and Raspberry Islands.

Thickness of coarse

surface layer (cm.)

Table 3. Frequencies (%) at which points on all

four islands would be expected to possess the

specified combinations of land surface slope and

surficial sand thickness, if the two properties are

not related.

Thickness of coarse

surface layer (cm.)

significant at a probability level of 0.001.

There is less than one chance in a thousand

that the positive association between land

surface slope and coarse surficial material

thickness results from chance.

The most probable explanation for the

observed correlation between land surface

slope and thickness of the coarse surficial

material relates to the beach processes which

deposited the sandy, gravelly sediment 8,000

to 11,000 years ago. Thick layers of coarse

sediments can accumulate in the littoral

(beach) zone only if three conditions are

more or less satisfied:

1) a sediment surplus must exist; incom¬

ing material from erosion or longshore

transport must exceed outgoing ma¬

terial;

2) a suitable depositional zone must be

present; in excessively steep areas the

suitable zone is very narrow; on

broad, level beaches wave action

causes the sand deposits to be exten¬

sive but thin;

3) sufficient time must be available; for

sedimentation to build thick deposits,

i.e. lake level must remain constant

for a period.

The degree to which these three conditions

were met differs between the islands, thus

there are differences in the thickness of

coarse beach sediments.

Although longshore drift has created a

few small spits and forelands at former and

present lake levels (Engstrom, 1972), the

major source of sediment for beaches on

all of the islands has been bluff erosion.

Bluff erosion is favored by the presence of

steep slopes above the beach level. When

a relatively steep face is present, sediment

is eroded from the face through under¬

cutting by wave and ice action, followed by

slumping, through sheet and rill erosion on

the bluff face, and through gravity-induced

slope processes. Sorting by wave action at

the beach may remove the fine fraction of

the bluff material, leaving coarser beach

deposits, including boulders.

If conditions are favorable for bluff ero¬

sion, sediment surplus at the beach will lead

to the accumulation of thick beach deposits.

During emergence of the Apostle Islands

these conditions were met on Oak Island, an

island with great vertical relief and relatively

steeply sloping flanks. There bluff erosion

has contributed large amounts of sediment

to beaches at all lake levels throughout the

post-glacial period. On York, Rocky, and

Raspberry Islands a large proportion of the

land surface is in gentle and flat slopes, with

no steep land above these surfaces to con-

1979] Cary , McDowell and Graumlich-

tribute sediment. Wave action on the rela¬

tively flat tops of these islands probably

sorted and reworked the original surfaces,

but the supply of new beach material was

small.

Slope of the island surface at beach level

determines the zone available for deposition

of beach sediment. Thick accumulations of

material are favored by moderate slopes.

When the littoral zone is fairly level expen¬

diture of wave energy over a wide area dis¬

courages thick sand accumulation by spread¬

ing the material laterally. On flat-topped

islands like Raspberry, York, and Rocky,

the wide littoral zone has a thin layer of

beach deposits spread over the fine-textured

material from the original surface. On an

island with moderately steep slopes, such as

Oak Island, the width of the littoral zone is

limited, and sand accumulation is thicker.

On a very steep original slope, a shelf-like

littoral zone, with relatively thick sediment

accumulation, may be constructed by wave

action if ample time and surplus sediment

are available. Some of the flatter slopes on

the west side of Oak Island may be beaches

built onto or cut into the original steep face.

Although the lake level fell rapidly, the

unconsolidated glacial debris mantling the

islands and the relatively high wave energy

of Lake Superior allowed beaches to build

rapidly. An island with moderately sloping

sides always had sand accumulating as long

as island emergence continued. The flat-

topped, steep-sided islands were less likely

to receive similar accumulations. Regard¬

less of the elevation at which lake level tem¬

porarily stabilized, thick sand accumulations

were unlikely; either there was a paucity of

coarse sediment or there was an unfavorable

depositional zone. Moderate slopes (ap¬

proximately 5-20%) thus seem to be impor¬

tant in the accumulation of thick sand de¬

posits. Differences in initial soil material be¬

tween islands can be partially explained by

this phenomenon.

The above discussion is important from

a practical, as well as a theoretical, view.

■Soils and Geology of Apostle Islands 29

Recreational development of the Apostle

Islands is proceeding under the direction of

the U. S. National Park Service. Assessment

of soil resources and evaluation of soil limi¬

tations for various recreational activities is

essential for planning. Drainage is an im¬

portant soil property and poor drainage pre¬

sents serious problems for septic tank dis¬

posal fields or for more primitive waste dis¬

posal methods, as well as for paths, trails,

and campgrounds. Poor drainage is most

often encountered on level, fine-textured

soils. Steep, sandy soils may present a seri¬

ous problem because of the great erosion

potential. The apparent correlation between

surface slope and thickness of coarse sur-

ficial material suggests that most areas will

have one problem or the other. Level land

with loamy or sandy soil has few limitations

but is not common on the mapped islands.

It does not seem likely that other islands

in the archipelago will have soils more suited

to recreational uses, because all the islands

were subject to the same beach processes.

The biologic and pedologic environments of

the Apostle Islands are fragile and develop¬

ment will demand great care.

Conclusion

This paper presents the findings of two

field seasons work in the Apostle Islands.

The prevalent soils on the four islands stud¬

ied are of two basic kinds: fertile Alfisols,

formed in deposits of fine-grained glacio-

lacustrine material (Eutroboralfs), and

sandy Spodosols developed in coarse beach

deposits (Haplorthods). The Apostle Is¬

lands possess a landscape with widely vary¬

ing characteristics of initial material and

topography, and which, in conjunction with

the cool, humid continental climate, allow

a variety of biotic communities to thrive.

The coexistence in this region of Alfisols,

usually found in more southerly locations

under deciduous forest, and Spodosols, typi¬

cally found in more northerly climatic zones

with evergreen vegetation, attests to the di-

30

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

versity and complexity of the pedogenic en¬

vironment.

Development of soil maps for the four

islands surveyed allows analysis of the spa¬

tial variability of soils. Reconnaissance soil

surveys completed 10 or more years ago

(Ableiter, and Hole, 1961; Hole, et al,

1968) show virtually all the Apostle Islands

as having soils formed in thin coarse de¬

posits overlying basal clayey material (Eu-

troboralfs, Alfic Haplorthods). This por¬

trayal is accurate for York, Rocky, and

Raspberry Islands, but not for Oak Island.

Oak Island is dominated by thick surficial

sand and gravel deposits and the Typic and

less well drained Haplorthods formed in

them. The preponderance of sandy soils on

Oak Island is explained by the correlation

between thick coarse deposits and steep land

surface slope.

Literature Cited

Ableiter, J. K., and F. D. Hole. 1961. Soil

Survey of Bayfield County, Wisconsin.

U.S.D.A. Soil Cons. Service, Series 1939,

No. 30, 77 p., 12 maps.

Beals, E. W. 1958. The phytosociology of

the Apostle Islands and the influence of deer

on the vegetation. M.S. thesis, Univ. of Wis¬

consin, Madison.

Beals, E. W., and G. Cottam. 1960. The

forest vegetation of the Apostle Islands,

Wisconsin. Ecol. 41:743-751.

Blalock, H. M., Jr. 1972. Social Statistics.

McGraw-Hill, New York. 583 p.

Buol, S. W., F. D. Hole, and R. J. McCracken.

1973. Soil Genesis and Classification. Iowa

State Univ. Press, Ames, 360 p.

Collie, G. 1901. Wisconsin shore of Lake

Superior. Geol. Soc. Amer. Bull. 12:197-

216.

Curtis, J. C. 1959. The Vegetation of Wis¬

consin. Univ. of Wisconsin Press, Madison,

657 p.

Engstrom, W. N. 1972. Spatial patterns in

beach morphology and sedimentology in the

Apostle Islands of northern Wisconsin.

Ph.D. dissertation. Univ. of Wisconsin,

Madison, 236 p.

Farrand, W. R. 1960. Former shorelines

in western and northern Lake Superior ba¬

sin. Ph.D. dissertation. Univ. of Michigan,

Ann Arbor, 226 p.

Farrand, W. R. 1969. The Quaternary his¬

tory of Lake Superior. Proc. 12th Conf.

Great Lakes Res.: 181-197.

Finley, R. W. 1975. Geography of Wiscon¬

sin. Univ. of Wisconsin Press, Madison. 472

P-

Gaikawad, S. T., and F. D. Hole. 1961.

Characteristics and genesis of a Podzol soil

in Florence County, Wisconsin. Trans. Wis.

Acad. Sci., Arts, Letters 50:183-190.

Graumlich, L. J. 1978. An analysis of the

spatial variation of soil microtopography

and surface horizon thicknesses. M.S. thesis.

Univ. of Wisconsin, Madison.

Hole, F. D. 1976. Soils of Wisconsin. Univ.

of Wisconsin Press, Madison. 223 p.

Hole, F. D., M. T. Beatty, C. J. Milfred, G. B.

Lee, and A. J. Klingelhoets. 1968. Soils

of Wisconsin. Univ. Wisconsin Ext., Madi¬

son, map (1:710,000).

Irving, R. D. 1880. Coastal features of the

eastern Lake Superior district. Geol. of Wis.

1873-1879, v. 3, pp. 70-76.

Kowalski, W. L. 1976. Geology, soils, and

recreational limitations of Bear Island, Wis¬

consin, Apostle Islands National Lakeshore.

M.S. thesis. Michigan Tech. Univ., Hough¬

ton, 207 p.

Maher, L. J., Jr. 1977. Palynological

studies in the western arm of Lake Superior.

Quaternary Res. 7:12-44.

Martin, L. 1965. The Physical Geography

of Wisconsin. Univ. of Wisconsin Press,

Madison, 3rd ed., 608 p.

Saarnisto, M. 1974. The deglaciation his¬

tory of the Lake Superior region and its cli¬

matic implications. Quaternary Res. 4:316-

339.

Saarnisto, M. 1975. Stratigraphical studies

on the shoreline displacement of Lake Su¬

perior. Can. J. Earth Sci. 12:300-319.

Soil Survey Staff. 1975. Soil Taxonomy.

U.S.D.A., S.C.S., Agric. Handbk. 436,

Washington, D.C., 754 p.

Taylor, F. B. 1931. Submerged peat beds

among the Apostle Islands. Science 74:265-

267.

THE CADDISFLIES (TRICHOPTERA) OF

PARFREY’S GLEN CREEK, WISCONSIN1

Thomas S. Karl and William L. Hilsenhoff

Department of Entomology

University of Wisconsin, Madison

A bstract

Life cycles, biology and ecology of 17 species of Trichoptera in Parfrey’s

Glen Creek are reported. Wormaldia moestus and Oligostomis ocelligera, rare in

Wisconsin, were found. Other species such as Diplectrona modesta, Parapsyche

apicalis, Rhyacophila vibox, and Dolophilodes distinctus, which seldom occur

commonly in any stream, were a significant part of the fauna. Closely related

species were often segregated temporally or spatially. Limnephilus rossi is reported

from Wisconsin for the first time and its larva is described.

Parfrey’s Glen Scientific Area in Sauk

County, Wisconsin has a unique flora and

fauna that has received much attention

(Vorhies, 1909; Hilsenhoff, 1974; Webb,

1974; Wynn and Loucks, 1975). Flowing

southward through the area is Parfrey’s

Glen Creek, a small, permanent, spring-

fed stream. Its insect fauna differs signifi¬

cantly from most other streams in southern

Wisconsin, and is dominated by an unusual

assortment of caddisfly larvae (Trichop¬

tera). It also has significant populations of

the stoneflies Isoperla clio (Newman), Al-

locapnia nivicola (Fitch), A. rickeri (Fri-

son), Amphinemura delosa (Ricker), and

A. linda (Ricker), the mayfly Baetis vagans

(McDunnough), the beetles Optioservus

fastiditus (LeConte), Helichus striatus (Le-

Conte), Agabus seriatus (Say), A. con-

fusus (Blatchley), and Hydrobius melaenus

(Germar), and several unidentified Diptera.

Because of increased public use of this area

(Wynn and Loucks, 1975), a study was

begun in 1975 to document the unique cad¬

disfly fauna and to learn more about their

biology and ecology.

1 Research supported by the College of Agricul¬

tural and Life Sciences, University of Wisconsin-

Madison and carried out in cooperation with the

Wisconsin Department of Natural Resources.

Parfrey’s Glen Creek is characterized by

series of riffles, rapids, small waterfalls, and

occasional pools. In winter it is frozen, and

in summer most of it is densely shaded by

vegetation, and in the gorge by cliffs. Only

a few patches of aquatic mosses are pres¬

ent, mostly outside of the gorge. The aver¬

age flow is 21 1/sec and varies little season¬

ally, with currents mostly 15 to 45 cm/sec

and depths of 1 to 40 cm. Temperatures

vary from 0 to 18°C, the maximum occur¬

ring in early spring before the trees are in

leaf. After the stream is shaded, tempera¬

tures remain fairly constant at 16.5°C

throughout the summer. Except for some

siltation due to visitor use, the stream is

free of pollution with oxygen at or near

saturation level. The stream has a gradient

of 1.92 hm/km, a pH of 7.8, total alkalinity

of 216 mg/1, and a specific conductance of

528 ^mhos/cm.

Methods

To collect caddisflies for determination

of life cycles, a D-frame aquatic net with a

1.0 mm mesh was held against the stream

bottom and the substrate immediately up¬

stream was thoroughly disturbed (using

one’s feet) for approximately one minute to

dislodge insects. Insects and debris col¬

lected in the net were placed in a white

31

32

Wisconsin Academy of Sciences , Arts and Letters

[Vol. 67

plastic pan with a little water. Caddisflies

were removed from large pieces of debris,

water, and the net and preserved in 70%

ethanol. The remaining debris was pre¬

served for more thorough examination in

the laboratory. Large rocks or logs that

could not be sampled by this method were

lifted from the water and caddisflies were

scraped into the net or directly into collect¬

ing jars. Prior to disturbing the substrate to

dislodge larvae, and with the D-frame net

in position, substrate materials were re¬

moved from the water and examined for

pupae. All possible microhabitats were sam¬

pled semimonthly from March 19, 1975 to

November 6, 1975; six or seven sites were

sampled on each date.

To determine microdistribution, sampling

sites were selected to be as homogeneous as

possible with respect to current, substrate

type, and water depth. Current was meas¬

ured with a pigmy current meter 2.5 cm

above the substrate and water depth was

calculated as the average of measurements

taken at the corners and center of a sampl¬

ing site. Stones were measured after sampl¬

ing to determine the greatest diameter. Sub¬

strate units sampled were silt, detritus,

leaves, moss, sand (1 mm to 2 mm), gravel

(3mm to 2 cm) and stones 2-5 cm, 5-10

cm, or >10 cm). An average of 45 sam¬

ples were collected each spring ( April -

June), summer (July-September), autumn

(October-December), and winter (January-

March).

Flight periods were determined from

adult collections and laboratory reared in¬

sects. Daytime collections were made by

sweeping riparian vegetation and by picking

adults from tree trunks near the stream or

exposed rocks in and along the stream; at

night, adults were collected in light traps.

In the laboratory, adults were reared from

pupae, prepupae or mature larvae in aerated

water maintained within 2°C of the stream

temperature.

Life cycles were determined by analyzing

larval growth rates and the presence of pre¬

pupae, pupae, and adults. Each larval instar

was determined by head capsule width

measured with an ocular micrometer.

Larvae were identified using keys by Ross

(1944), Flint (1960, 1961, 1962), and

Hilsenhoff (1975). Lepidostoma larvae

were identified using the metamorphotype

method of Milne (1938). Adults were iden¬

tified using keys and illustration by Betten

(1934, 1950), Ross (1944, 1946), and

Leonard and Leonard (1949a). Specimens

were also compared with identified adults

in the University of Wisconsin Insect Col¬

lection, and voucher specimens were de¬

posited in this collection. The reported dis¬

tribution in Wisconsin is based on Long-

ridge and Hilsenhoff (1973), and speci¬

mens in the University of Wisconsin Insect

Collection.

Seventeen species of caddisflies were col¬

lected from Parfrey’s Glen Creek; these are

the only species known to occur in this

stream. Life cycles and numbers collected

of the fourteen most common species are

summarized (Table 1 ) and notes on the

biology and ecology of all species follow.

Caddisfly Biology and Ecology

Dolophilodes distinctus (Walker) 1852

Dolophilodes distinctus has been found

in rapid, cool streams in northern Wiscon¬

sin and a few streams in Sauk County. It

was relatively common in Parfrey’s Glen

Creek. Adults were captured every month

except September and October, and other

workers have noted adults present over

much or all of the year (Longridge and Hil¬

senhoff, 1973; Ellis, 1962; and Leonard

and Leonard, 1949b). The population in

Parfrey’s Glen Creek is bivoltine with peaks

of emergence in March-April and July-

August. Tebo and Hassler (1961) also

noted two peaks of emergence in this spe¬

cies, in April and in September.

Brachypterous females were collected

from late February to the first week of

April. They were found crawling on the

stream bank, under rocks and logs near

1979]

Karl and Hilsenhoff — Caddisflies of Parfrey’s Glen

33

Table 1. Development of 14 species of Trichoptera in Parfrey’s Glen Creek indicating life stages1

1 Life stage noted only if it constituted 10% or greater of the months sample of that species. 1 ,2, 3, 4, 5 = Larval Instar;

PP = prepupae; P = Pupae; A = Adult.

the stream, and on exposed rocks in the

stream. Ross (1944) also found only wing¬

less females in winter and noted that winter

males were winged and tended to be larger

than in other seasons. In warmer months,

adults were found on tree trunks close to

the stream and under rocks and logs along

the stream.

Larvae that hatched from eggs laid in

March and April grew very rapidly and by

the end of June nearly 90% were in the

fifth instar. Pupae were first noticed in early

July. Early instar larvae from the July-

August emergents were first noticed in mid-

August. They grew rapidly and by Novem¬

ber all were in the fifth instar. The winter

was spent mainly as pupae, but prepupae

and fifth instar larvae were also present. By

late February more than 80% were in the

pupal stage. Throughout the winter a small

number of adults emerged to produce the

brachypterous phenotype. The presence of

some adults quite removed from periods

of peak emergence may be attributable to

a long adult life span.

Larvae inhabited a wide range of sub-

34

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

strates, currents and water depths. They

were found in currents ranging from 5 cm/

sec to 97 cm/sec, mostly on the under sur¬

face of rocks 12 cm or larger. Depths

greater than 15 cm were preferred, but

larvae were not uncommon in water as shal¬

low as 5 cm.

Pupal chambers were fastened to the un¬

der surface of stones, usually 10 cm or

larger. Pupae were found in currents of 19

to 70 cm/sec at depths of 4 cm to 20 cm

and preferred currents faster than 30 cm/

sec.

Wormaldia moestus (Banks) 1914a

Wormaldia moestus is rare in Wisconsin.

Besides Parfrey’s Glen Creek and one other

Sauk County stream, it has been collected

only in the northeastern counties. Eight lar¬

vae were found, all of them in the densely

shaded stream gorge, mostly on the under

sides of rocks 10 cm or larger, in currents

of 23 to 70 cm/sec and water 10 to 16 cm

deep. The two smallest larvae were col¬

lected in early and late April (head cap¬

sule widths 0.35 and 0.50 mm respec¬

tively). Those collected in June, November,

and February had head capsule widths of

1.70-1.80 mm. Larvae apparently grow

rapidly in spring and spend summer, autumn

and winter in the terminal instar. Pupation

probably occurs in late winter with emer¬

gence in early spring. Ross (1944) col¬

lected adults in Illinois during the first week

of March.

Diplectrona modesta (Banks) 1908b

Diplectrona modesta frequents rapid

clear brooks and streams throughout Wis¬

consin. In Michigan Leonard and Leonard

(1949b) found it only in tiny, spring-fed

brooks. Parfrey’s Glen Creek is one of the

few streams in Wisconsin where it com¬

prises a significant part of the caddisfly

fauna, and here it was the dominant hydro-

psychid. Diplectrona modesta was univol-

tine with adults present from early May to

July, and peak emergence the last week of

May and first week of June. Adults were

found resting on tree trunks.

A wide range of instars was often pres¬

ent, indicating an extended hatching period.

Larvae grew steadily over the summer and

by mid- August more than 80% were in the

fourth and fifth instar. The larvae developed

little during the winter, with growth resum¬

ing in early spring. Prepupae were found

from April 29 to July 9, pupae from June

12 to July 24.

First and second instar larvae preferred

2- 5 cm stones in currents of 13 to 29 cm/

sec; depth was not important. They were

mostly on the undersurface, but were found

on other faces if vegetation was present.

Later instars were less selective and in¬

habited all sizes of stones and occasionally

vegetation in currents of 12 to 68 cm/sec,

but preferred currents of 15 to 36 cm/sec.

Mostly on the undersurfaces of stones, they

inhabited other faces if fissures or protuber¬

ances protected them from the current. In

late spring, fifth instar larvae preferred

deeper water (>12 cm), larger stones

(>10 cm) and slower currents (<46 cm/

sec) than in other seasons, probably as a re¬

sult of movement to pupal habitat in spring.

Pupae attached chiefly to the undersurface

of rocks greater than 10 cm in water deeper

than 11 cm with currents of 15 to 44 cm/

sec.

Parapsyche apicalis (Banks) 1908b

In Wisconsin Parapsyche apicalis occurs

in small, cold, rapid streams, mostly in the

north. It was not common in Parfrey’s

Glen Creek. Adults of this univoltine spe¬

cies were present from April 30 to June 12

with the emergence peak the last two weeks

of May. Longridge and Hilsenhoff (1973)

and Leonard and Leonard (1949b) col¬

lected adults into summer and early autumn.

First instar larvae were collected first on

July 9 indicating an incubation period of

3- 4 weeks. The larvae grew slowly in July

1979]

Karl and Hilsenhoff — Caddis flies of Parfrey’s Glen

35

and August and more rapidly in September

so that by the end of September almost

80% were in the fourth instar. Winter was

spent mainly as fifth instar larvae, but sec¬

ond, third and fourth instars were also pres¬

ent. Prepupae and pupae were present from

April 29 to May 12.

Developmental stage and season influ¬

enced larval distribution. First and second

instar larvae clearly preferred vegetation in

currents of 30 to 78 cm/sec and depths

less than 12 cm. In autumn, third and

fourth instar larvae were most common on

stones 5 cm or larger in currents of 23 to

53 cm/sec and depths greater than 6 cm.

In winter, they also colonized vegetation

growing on rocks. Fifth instar larvae were

found at all depths and in autumn and win¬

ter were most common on rocks greater

than 10 cm and in currents faster than 30

cm/sec. They occurred on the bottom of

bare rocks, and on other faces if vegetation

was present.

Approximately 60% of the pupal cham¬

bers were constructed of vegetation, the rest

of sand grains. Flint (1961) noted that

pupal cases were almost exclusively organic

material and only rarely constructed of

sand grains. Pupal cases of vegetation were

found in vegetation and completely en¬

closed the pupa. Sand cases were found on

the undersides of stones larger than 5 cm in

diameter in currents less than 15 cm/sec

and at depths greater than 6 cm.

Hydropsyche slossonae (Banks) 1905

Hydrosyche slossonae is widely distrib¬

uted in clear, cool, swiftly flowing streams

throughout Wisconsin, and was the second

most common hydropsychid in Parfrey’s

Glen Creek. It is univoltine with a long

flight period (Longridge and Hilsenhoff,

1973; Leonard and Leonard, 1949b; Ellis,

1962; and Ross, 1944). Adults were col¬

lected from May 19 to September 18. On

August 28, 1975, 161 females and 5 males

were caught in a light trap. While males

may not be as strongly phototropic as fe¬

males (Leonard and Leonard, 1949a),

Schuhmacher (1970) found females to be

generally more prevalent than males in the

genus. Eggs were found in late August and

September on the sides and beneath rocks

and logs. They were roundish, off-white,

and laid in one-layered patches of about

260 eggs.

Early instar larvae were first evident in

early July. Most grew very slowly and

overwintered in the second instar. Growth

resumed in late March and by the end of

April, 80% were in the third instar. Some

larvae overwintered in the fifth instar and

pupated in late spring, accounting for

adults in May and June.

Segregation of instars was evident. First

and second instar larvae were most com¬

mon on vegetation while later instar larvae

preferred bare stones 5-15 cm in diameter

at depths less than 1 1 cm. Nets were spun

in areas protected from the full force of

the current. Fifth instar larvae were also

found in vegetation on the upper surface

of rocks in currents greater than 65 cm/sec.

All larvae were most common in currents

greater than 18 cm/sec.

Pupae were attached to vegetation or to

the bottom of stones 5 cm or larger in di¬

ameter, usually in currents of 12 to 36 cm/

sec. Depth did not significantly limit distri¬

bution, although pupae were not found at

depths less than 4 cm unless the current was

greater than 30 cm/sec.

The three hydropsychids, H. slossonae,

D. modesta and P. apicalis were segregated

temporally and spatially. Hydropsyche slos¬

sonae was temporarily isolated by growing

in late spring while D. modesta and P. api¬

calis grew most actively in autumn. Early

instars of D. modesta and P. apicalis were

segregated spatially, with P. apicalis being

found in vegetation on upper surfaces of

rocks in fast moving shallow water and

D. modesta mostly under bare stones over

a wide range of depths and currents. Later

36

Wisconsin Academy of Sciences , Arts and Letters

[Vol. 67

instars were not segregated spatially, but

mature larvae eat different foods (Shapas

and Hilsenhoff, 1976).

Rhyacophila vibox (Milne) 1936

Rhyacophila vibox is found throughout

Wisconsin in small, cold, spring-fed streams.

Flint (1962) reported larvae only in springs

and spring brooks less than two feet wide.

It was a common univoltine species in Par¬

ley’s Glen Creek. Adults were collected

mostly from cracks or fissures of tree bark.

They were found May 22 to June 20 cor¬

roborating published flight periods (Long-

ridge and Hilsenhoff, 1973; Ross, 1944;

Ellis, 1962). First instar larvae appeared

in early July and grew slowly so that by

mid-August more than 70% were still in

the first and second instar. Growth acceler¬

ated in September, and by early November

more than 70% were in the fifth instar.

Winter was spent in the fourth or fifth in¬

star and by March most were in the fifth

instar. Prepupae were found April 20 to

May 5 and pupae April 29 to June 12.

Life stages showed some segregation by

microhabitat. First instar larvae were found

only in currents less than 45 cm/sec, while

later instars had no current limitations, ex¬

cept in spring and summer when they were

most abundant in currents faster than 30

cm/sec. All larvae avoided pools. Vorhies

(1909) reported R. vibox larvae in the

moderately swift portions of Parfrey’s Glen

Creek. First instar larvae were limited to

the lower surfaces of stones. Later instars

also inhabited vegetation on the upper sur¬

faces and sides of stones, but were most

common under stones of 5-10 cm. During

autumn, winter, and spring, larvae were also

found in leaf mats, under the bark of sub¬

merged logs, and on large stones isolated in

sandy stretches. Depth never restricted lar¬

val distribution.

Pupae were most common in currents

greater than 25 cm/sec and depths greater

than 10 cm. Stones greater than 3 cm in

diameter were preferred as pupation sites,

although some pupal chambers were at¬

tached to stones barely larger than the

chamber.

Glossosoma intermedium (Klapalek) 1892

Glossosoma intermedium occurs in cool,

woodland streams throughout Wisconsin,

and was the most abundant caddisfly in

Parfrey’s Glen Creek. At some sites their

cases almost covered the stream bottom.

Glossosoma intermedium was univoltine

with overlapping generations, the adults

having two periods of emergence. From

April 19 to June 1, less than 10% emer¬

ged, with peak emergence during the first

half of May and a male: female ratio of

9:1. More than 90% emerged from July 7

to October 23, mostly during the last two

weeks of July and first week of August.

During this second emergence period the

male: female ratio was 1:2. Adults were

found resting on trees and riparian vegeta¬

tion.

First instar larvae of the spring genera¬

tion were noticed first in mid-June. They

grew steadily and by the first week of No¬

vember were in the fifth instar or prepupal

stage in which they overwintered. By early

April most individuals were prepupae, and

pupae were found from the first week of

April to the end of May. First instar larvae

from the summer emergence were first col¬

lected July 24 and grew slowly over the

autumn, overwintering as third and fourth

instars. Growth accelerated in mid-April

and by mid-June more than 70% were pre¬

pupae.

Larvae were found over a wide range of

currents, substrate types, and water depths.

There was some segregation of life stages,

and seasonal differences in distribution were

also evident. First and second instar larvae

preferred currents less than 30 cm/sec and

stones 3-7 cm. A few were found in gravel.

Instars three to five were found in currents

ranging from pools to 96 cm/sec, but most

were in currents of 15 to 30 cm/sec. They

inhabited a wide range of substrates, but

1979]

Karl and Hilsenhoff — Caddisflies of Parfrey’s Glen

37

were most common on stones 5 cm or

larger. Larvae occurred only on the down¬

stream side of rocks or other protected

areas in currents greater than 30 cm/sec,

but on all sides in slow currents. Larvae in

pools were found only on stones or logs free

of silt or detritus. In summer, fifth instar

larvae even occurred on rocks and logs that

protruded above the water’s surface pro¬

vided the surface was continually wetted.

Just prior to pupation the bottom strap

of the case is cut away and the dome¬

shaped upper portion is cemented to a rock

or other support (Ross, 1944). Most pupae

were found in currents greater than 22 cm/

sec, at depths greater than 12 cm, and on

rocks larger than 10 cm in diameter. Pupae

were also found on logs and smaller stones,

though much less frequently. A few were

found in pools. In currents greater than 30

cm/sec pupae occurred on surfaces pro¬

tected from the full force of the current,

while in slower currents pupae were present

on all surfaces. Pupae occurred in aggrega¬

tions, with some clusters containing more

than one hundred individuals.

Oligostomis ocelligera (Walker) 1852

Oligostomis ocelligera is uncommon in

Wisconsin and rare in Parfrey’s Glen Creek;

only two larvae were found during this

study. Both had head capsule widths of 2.25

mm and were collected February 29, 1976,

from detritus on the stream bottom at a

depth of 10 cm in a current of 24 cm/sec.

Lloyd (1921) stated that the larvae are

bottom dwellers that crawl over the stream

floor, or rest among the trash and litter. He

reported pupation in mid- or late April in

the stream bed or in dead wood; in New

York emergence occurs during the last two

weeks in May. Longridge and Hilsenhoff

(1973) collected adults May 28 and June 4

in Wisconsin.

Brachycentrus occidental is (Banks) 1911

Brachycentrus occidentalis occurs in cold

streams in all but the southeastern corner of

Wisconsin. Only 58 larvae were collected,

and the species was regarded as uncommon

in Parfrey’s Glen Creek. Extensive notes on

various aspects of its biology have been

published recently by Gallepp (1974,

1976) and Gallepp and Hasler (1975).

Adults were found in Parfrey’s Glen Creek

April 29.

Early instar larvae were noticed first in

late July. They grew steadily and by the end

of September almost all were in the fifth in¬

star, the stage in which they overwintered.

Gallepp and Hasler (1975) observed the

same pattern of growth in Lawrence Creek,

Wisconsin.

In summer, larvae were found on the up¬

per surface of rocks in currents of 10 to

31 cm/sec and water less than 6 cm deep.

Throughout the year larvae were found in

open areas or lightly shaded sections of the

stream. Gallepp (1976) also noted the pref¬

erence of this species for areas exposed to

bright sunlight. No pupae were collected.

Pycnopsyche guttifer (Walker) 1852

Pychnopsyche guttifer has been collected

throughout Wisconsin. Adults of this un¬

common, univoltine species were collected

September 18 to November 6 from riparian

vegetation. On November 6, when snow

was present, they were found crawling on

the stream banks.

Early instar larvae were noticed first in

mid-October. They grew actively over the

autumn and winter months so that by early

March all were in the fifth instar. This was

the only caddisfly in Parfrey’s Glen Creek

to grow rapidly during the winter, thus

avoiding competition with most other her¬

bivores and detritivores. Only fifth instar

larvae were found in spring and early sum¬

mer; no larvae were found in late summer.

Cummins (1964) reports that during the

summer larvae fasten their cases to sticks,

aquatic vegetation, or cobbles, and remain

inactive until pupating in early autumn.

Neither pupae nor prepupae were found in

this study.

38

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

Larvae were most abundant in pools or

along the banks where they crawled over

debris and leaves. Depth was not signifi¬

cant in determining distribution. A few in¬

dividuals collected in the autumn were

found on the bottom of loosely packed

stones (5-10 cm in diameter) in currents

up to 20 cm/sec. Cummins (1964) found

that during low flow larvae moved into the

middle of the stream if periphyton was

abundant. Feldmeth (1970) found that

P. guttifer larvae were swept off the sub¬

strate by currents greater than 15 cm/sec.

Early instar larvae constructed cases

from leaf and stick matter, the leaf material

predominating. As the larvae grew, the ratio

was reversed and stick material dominated

as described by Flint ( 1960) .

Hesperphylax designatus (Walker) 1852

Hesperophylax designatus occurs through¬

out Wisconsin in small, cold streams. Only

27 larvae were found in Parfrey’s Glen

Creek. Adults of this univoltine species were

collected May 22 to July 24 from riparian

vegetation, with peak emergence in late

May. In other Wisconsin studies Longridge

and Hilsenhoff (1973) noted a flight period

from April 27 to August 24, and Vorhies

(1905) found adults March 15 with peak

emergence in mid-April.

Early instar larvae were noticed first the

last week of July. They grew steadily over

the summer and autumn, and overwintered

as fifth instar larvae. Prepupae were found

the first week of April and pupae the last

week of April and first week of May. Vor¬

hies (1905) noticed that in mid-February

most larvae had begun to prepare for pupa¬

tion.

A seasonal difference in habitat selection

was evident. In summer, larvae inhabited

mostly logs and occasionally gravel and

small stones (less than 5 cm) . They were

found only in currents slower than 20 cm/

sec. In autumn and winter, larvae preferred

the undersides of stones 5-10 cm in diam¬

eter or vegetation on these stones, and were

found mostly in currents faster than 40

cm/sec. Water depth did not restrict distri¬

bution.

Detailed descriptions of cases are in Vor¬

hies (1905, 1909) and Lloyd (1921). Pu¬

pae were found clustered on undersides of

rocks greater than 10 cm in pools or in cur¬

rents less than 15 cm/sec and depths

greater than 20 cm. Clustering of pupae

has been reported previously for this spe¬

cies (Vorhies, 1909; Denning, 1937).

Limnephilus rhombicus (Linnaeus) 1758

Limnephilus rhombicus has previously

been collected in Wisconsin only from the

northeast and southwest counties. One larva

with a head capsule width of 1.55 mm was

collected July 9, 1975, from a pool with a

detritus substrate and a depth of 8 cm. Re¬

ported flight periods are June 10 to June 29

for Wisconsin (Longridge and Hilsenhoff,

1973), the end of June to the first half of

July for Michigan (Ellis, 1962), and early

June to late August for Illinois (Ross,

1944).

Limnephilus rossi (Leonard and Leonard)

1949a

Leonard and Leonard (1949a) described

this species from adults collected September

18 to October 6 from dry grass overhanging

the banks of spring-fed streams in Michigan.

The larvae were not described, and the

identity of those collected from Parfrey’s

Glen Creek remained unknown until 1977

when two males were reared September 21

and 27 from pupae or prepupae collected

September 5. This is the first record of this

species in Wisconsin.

Early instar larvae were collected first at

the end of February. They grew slowly un¬

til the end of April, and then rapidly, reach¬

ing the fifth instar by early summer. The

summer was spent in the fifth instar, with

pupation in late summer. Early instar lar¬

vae were collected easily, but fifth instar

1979]

Karl and Hilsenhoff — Caddisflies of Parfrey’s Glen

39

larvae and pupae were difficult to find. Lar¬

vae and pupae occurred on the bottoms of

rocks of all sizes and on logs over a wide

range of currents. A distinct preference was

exhibited for depths greater than 8 cm.

Larval description: (fifth instar): Length

16.2-18.4 mm, width 2. 7-3.0 mm. Head

mostly light brown with dark muscle scars

most numerous on the vertex and dorso-

posterior and ventral sections of the genae.

Ventral half of genae medium brown, gula

light brown. Frons with an anteriorly di¬

rected triangle composed of dark muscle

scars. Legs light brown with infrequent

muscle scars. Thoracic nota light brown

with numerous dark muscle scars. Anterior

edge of pronotum medium brown; posterior

corners black. Mesonotum with central third

of posterior margin and posterior third of

sides black. Metanotum with four ovoid

sclerites having 11-12 setae each. Abdominal

segments 1 and 8 with 15-19 setae dorsally

on each side. Gills mainly in clusters of two

or three, with occasional single gills. Chlor¬

ide epithelia present on ventor of segments

2-7. Anal claw with one large and one small

accessory tooth.

Larval case : Constructed mainly of min¬

eral particles averaging 0.8 mm anteriorly

and 0.4 mm posteriorly with a few bits of

organic material distributed randomly. Ta¬

pered posteriorly and slightly curved with

a dorsal hood projecting about 2 mm an¬

teriorly. Total length 13 to 16 mm for ma¬

ture larvae.

Neophylax concinnus (McLachlan) 1871

Previous collections from Wisconsin indi¬

cate that N. concinnus is restricted chiefly

to the northern half of the state, but it was

common in Parfrey’s Glen Creek. Adults

were present September 6 to November 30,

with peak emergence the last half of Sep¬

tember. Sedell (1972) found emergence

in this species was highly synchronized.

Longridge and Hilsenhoff (1973) collected

adults July 7 and September 13 in Wiscon¬

sin, while Leonard and Leonard (1949b),

and Vorhies (1909) found adults flying as

late as November.

Early instar larvae were first noticed in

early November. They grew slowly in late

autumn and winter so that by early April

more than 60% were still in the third instar.

Growth accelerated in late April and by the

end of June nearly all were prepupae. Pre¬

pupae were found April 29 to August 15,

but first occurred in significant numbers the

end of June. Other workers have noted this

long prepupal period for Neophylax (Vor¬

hies, 1909; Lloyd, 1921). The pupal pe¬

riod appeared to be short, as pupae were

collected only on August 15.

Early instar larvae were found mostly on

stones 2 cm or larger. In currents less than

15 cm/sec larvae occurred on all faces of

stones, but in faster currents they sought the

downstream side or other sheltered areas.

They were frequently found on large stones

isolated in sandy stretches, and in spring

preferred larger rocks. A few individuals

were found on logs. Terminal instar larvae

were often found out of the water on rock

surfaces that were continually wetted.

Throughout the year, larvae were found in

a wide range of currents and depths.

Larvae pupated in the larval case, which

they attached chiefly to rocks 8 cm or

larger in currents ranging from pools to 30

cm/sec. In currents of 15-30 cm/sec pupae

were found mainly in sheltered areas on the

lateral and downstream sides of rocks; in

pools they occurred on all sides except the

bottom. Water depth did not limit distribu¬

tion, and some larvae pupated so close to

the water line that when the water level

dropped slightly during the summer they

became exposed and dried out.

Lepidostoma try anti (Banks) 1908a

Lepidostoma bryanti has been found in

all except the southeast corner of Wiscon¬

sin. In Parfrey’s Glen Creek it was less

abundant than the other two species of

40

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

Lepidostoma. Adults were collected April

30 to June 27, with most emerging during

mid-June. They were found generally rest¬

ing on tree bark and in riparian vegetation.

First instar larvae were noticed first the

last week of June. They grew slowly

throughout the summer and 90% were still

in the third instar in late September. Growth

increased greatly in late autumn and by the

end of November more than half were in

the fifth instar. The winter was spent as

fourth or fifth instar larvae; by mid-March

almost all were in the fifth instar. Pupae

were found May 5 to May 28. Vorhies

(1909) noted that in Parfrey’s Glen Creek

pupae were abundant on June 1.

Except in autumn, larvae were found

exclusively in pools, the preferred sub¬

strate being leaf material and detritus. In

autumn, perhaps in response to an increased

need for food during their period of active

growth, larvae were found on the exterior

and beneath the bark of submerged logs,

and were common beneath leaves. A few

individuals were found in crevices or holes

in rocks isolated in sandy stretches. Water

depth did not limit distribution in any season.

Lepidostoma costalis (Banks) 1914b

Lepidostoma costalis was common in

Parfrey’s Glen Creek, but the only previ¬

ous Wisconsin record is from farther north

in Waushara County. Adults were collected

August 2 to September 18, mostly in the

last half of August. They were found most

commonly resting in bark fissures.

Larvae first appeared in mid-August and

grew slowly. Winter was spent as second

and third instar larvae. Active growth re¬

sumed in mid-April and by late June most

larvae were in the fifth instar. Prepupae

and pupae were found July 24 to August

15.

In late fall and winter, larvae were found

mostly in pools with a substrate of leaves

or detritus, but they were also present on

logs and in gravel. Larvae were found only

in areas where the current did not exceed

14 cm/sec; water depth did not limit dis¬

tribution. In spring, larvae were found in

currents of 15 to 30 cm/sec, clinging to

leaf jams in mid-stream or to rocks greater

than 6 cm in diameter. Those on stones were

either in crevices or on the bottom. As the

larvae grew they moved back to the pool

habitat and areas of very slow currents;

fourth and fifth instar larvae were found

only in these sites.

Pupae attached directly to vegetation in

currents up to 78 cm/sec. They also com¬

monly attached to the exterior of logs and

cracks and holes beneath the bark. Pupae

in pools were attached to pieces of bark,

twigs, or bits of leaves. Some attached to

lateral and downstream faces of rocks in

the middle of sandy stretches.

Lepidostoma griseum (Banks) 1911

Lepidostoma griseum has previously been

collected only from northern Wisconsin.

Although uncommon in Wisconsin, it was

the most common lepidostomatid in Par¬

frey’s Glen Creek. Adults were collected

August 12 to September 28, with most

emerging the last half of August. This flight

period is similar to that reported by Long-

ridge and Hilsenhoff (1973), Ellis (1962),

and Leonard and Leonard (1949b).

Early instar larvae were collected first on

October 16. Growth was slow throughout

the autumn, and winter was spent in the

second and third instar. Active growth re¬

sumed in late April and at the end of June

70% were in the fifth instar. Prepupae were

found on August 15; pupae occurred from

August 15 to November 6.

During autumn and winter, larvae oc¬

curred in pools and currents ranging up to

30 cm/sec. The preferred substrate in pools

was detritus; in riffles it was stones greater

than 8 cm in diameter. In spring, when

growth was most rapid, larvae concentrated

in pools that had leaves and detritus as a

substrate. Depth did not limit distribution

in any season.

Larval cases were constructed of organic

1979]

Karl and Hilsenhoff — Caddisflies of Parfrey’s Glen

41

material and were similar to those of L.

bryanti, but the cases of mature larvae dif¬

fered in being shorter and stockier. Early

instar larvae constructed cases entirely of

small sand grains. Some fourth and fifth

instar larval cases were mainly organic ma¬

terial, but even these always had at least a

few sand grains present.

Pupae were found in currents of 4 to 31

cm/sec. They were common in cracks and

holes in the bark of submerged logs, but

also occurred on the underside of stones 5

to 10 cm in diameter and in sheltered areas

on the lateral and downstream sides.

The three species of Lepidostoina co¬

existing in Parfrey’s Glen Creek were sep¬

arated temporally or spatially during pe¬

riods of active growth. L. bryanti larvae

grew most rapidly in summer and early au¬

tumn and by mid-autumn were in the fourth

or fifth instar. L. costalis and L. griseum

larvae grew most rapidly in late spring and

early summer, and were segregated during

this period, utilizing different habitats. L.

griseum occurred almost exclusively in

pools while L. costalis was most common

in areas of moderate current and stony sub¬

strates. During periods of slow growth or

pupation, no significant segregation of these

two species was evident.

Summary

The 17 species of caddisflies that inhabit

Parfrey’s Glen Creek represent a fauna that

is unusual in southern Wisconsin and more

representative of faunas in northern Wis¬

consin streams. The small size of the stream,

its low summer temperature (16.5°C), and

dense woodland setting are probably re¬

sponsible for maintenance of this fauna.

Species competing for the same resources

were found to be segregated either tempor¬

ally or spatially, permitting maximum use of

these resources and allowing many species

to coexist in the stream. Distribution of spe¬

cies within the stream results from a com¬

plex interaction of environmental factors,

of which current, substrate and food are

probably most important.

Literature Cited

Banks, N. 1905. Descriptions of new Nearctic

neuropteroid insects. Trans. Am. Entomol.

Soc. 32:1-20.

— - . 1908a. Some Trichoptera and allied

insects from Newfoundland. Psyche 15(4):

61-67.

- . 1908b. Neuropteroid insects-notes and

descriptions. Trans. Am. Entomol. Soc. 34:

255-267.

- . 1911. Descriptions of new species of

North American neuropteroid insects. Trans.

Am. Entomol. Soc. 37(4): 335-360.

- . 1914a. American Trichoptera-notes

and descriptions. Can. Entomol. 46:201-

205.

- . 1914b. American Trichoptera-notes

and descriptions. Can. Entomol. 46:261-

268.

Betten, C. 1934. The caddisflies or Trichoptera

of New York State. N.Y.S. Mus. Bull. 292:

1-576.

- . 1950. The genus Pycnopsyche (Tri¬

choptera). Ann. Entomol. Soc. Am. 43(4):

508-522.

Cummins, K. W. 1964. Factors limiting the

microdistribution of larvae of the caddis¬

flies Pycnopsyche lepida (Hagen) and Pyc¬

nopsyche guttifer (Walker) in a Michigan

stream. Ecol. Monogr. 34:271-295.

Denning, D. G. 1937. The biology of some

Minnesota Trichoptera. Trans. Am. Ento¬

mol. Soc. 63:17-44.

Ellis, R. J. 1962. Adult caddisflies (Trichop-

era) from Houghton Creek, Ogemaw

County, Michigan. Occ. Pap. Mus. Zool.

Univ. Mich. 624:1-16.

Feldmeth, C. R. 1970. The respiratory ener¬

getics of two species of stream caddisflies in

relation to water flow. Comp. Biochem.

Physiol. 32:193-202.

Flint, O. S. 1960. Taxonomy and biology of

nearctic Limnephilid larvae (Trichoptera)

with special reference to species in eastern

United States. Entomol. Am. 40:1-117.

- . 1961. The immature stages of the

Arctopsychinae occurring in Eastern North

America (Trichoptera: Hydropsychidae).

Ann. Entomol. Soc. Am. 54(1): 5-11.

42

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

- . 1962. Larvae of the caddisfly genus

Rhyacophila in eastern North America

(Trichoptera: Limnephilidae). Proc. of U.S.

Nat. Mus. Smithsonian Institute 113:465-

493.

Gallepp, G. W. 1974. Behavioural ecology of

Brachycentrus occidentalis Banks during the

pupation period. Ecology 55(6): 1283-1294.

- . 1976. Temperature as a cue for the

periodicity in feeding of Brachycentrus oc¬

cidentalis ( Trichoptera ). Animal Behavior

24:7-10.

- and A. Hasler. 1975. Behavior of lar¬

val caddisflies (Brachycentrus spp.) as in¬

fluenced by marking. Am. Midi. Nat. 93:

247-254.

Hilsenhoff, W. L. 1974. The unusual larva

and habitat of Agabus confusus (Dytisci-

dae). Ann. Entomol. Soc. Am. 67:703-705.

- . 1975. Aquatic insects of Wisconsin.

Generic keys and notes on biology, ecology

and distribution. Tech. Bull. 89 Wisconsin

Dept, of Natural Resources.

Klapalek, F. 1892. Trichopterologicky Vyz-

kum Chech v.r. 1891. Ceska akademie

cisare Frantiska Josefa provedy, slovesnost

a umeni v Praze. Rozpravy 5:1-22.

Leonard, J. W. and F. A. Leonard. 1949a.

Noteworthy records of caddisflies from

Michigan, with descriptions of new species.

Occ. Papers. Mus. Zool. Univ. Mich. 520:

1-8, pis. 1-5.

- . 1949b. An annotated list of Michigan

Trichoptera. Occ. Papers Mus. Zool. Univ.

Mich. 522:1-35.

Linnaeus, C. 1758. Systema Naturae. Tenth

Edition. Stockholm.

Lloyd, J. T. 1921. The biology of North

American caddisfly larvae. Lloyd Lib. Bull.,

Entomol. Ser. 21: 1-124.

Longridge, J. L. and W. L. Hilsenhoff. 1973.

Annotated list of Trichoptera (Caddisflies)

in Wisconsin. Trans. Wise. Acad. Sci. Arts

and Letters 61:173-183.

McLachlan, R. 1871. On new forms etc., of

extra-European Trichopterous insects. J. of

the Linnean Soc. of London, Zoology 1 1 :

98-141.

Milne, L. J. 1936. Studies in North American

Trichoptera. Pt. 3, 56-128, with 2 pis. Cam¬

bridge, Mass.

Milne, M. J. 1938. The metamorphotype

method in Trichoptera. J. N. Y. Entomol.

Soc. 46: 435-437.

Ross, H. H. 1944. The caddisflies or Trichop¬

tera of Illinois. Ill. Nat. Hist. Surv. Bull.

23:1-326.

- . 1946. A review of the nearctic Lepi-

dostomatidae (Trichoptera). Ann. Entomol.

Soc. Am. 39: 265-290.

Schuhmacher, H. 1970. Untersuchungen zur

Taxonomie, Biologie und okologie einiger

Kocherliegenarten der Gattung Hydro¬

psyche Pictet (Insecta, Trichoptera). Int.

Revue ges. Hydrobiol. 55: 511-557.

Sedell, J. R. 1972. Trophic ecology and nat¬

ural history of Neophylax concinnus and

N. oligius. Diss. Abs. 5747-B.

Shapas, T. J. and W. L. Hilsenhoff. 1976.

Feeding habits of Wisconsin’s predominant

lotic Plecoptera, Ephemeroptera, and Tri¬

choptera. Great Lakes Entomol. 9:175-188.

Tebo, L. B., Jr. and W. W. Hassler. 1961.

Seasonal abundance of aquatic insects in

western North Carolina trout streams. J.

Elisha Mitchell Scient. Soc. 77:249-259.

Vorhies, C. T. 1905. Habits and anatomy of

the larvae of the caddisfly. Platyphylax des¬

ignate Walker. Trans. Wise. Acad. Sci.

Arts Letters 15:108-123.

- . 1909. Studies on the Trichoptera of

Wisconsin. Trans. Wise. Acad. Sci. Arts

Letters 16:647-738.

Walker, F. 1852. Catalogue of the specimens

of neuropterous insects in the collection of

the British Museum. Pt. I. 192 pp. London.

Webb, D. W. 1974. New Species of Panorpa

(Mecoptera: Panorpidae). Entomol. News.

85:171-173.

Wynn, S. and O. Loucks. 1975. A social and

environmental history of human impact on

Parfrey’s Glen. Trans. Wise. Acad. Sci. Arts

Letters 43:26-54.

GARRISON LIFE IN THE NOVELS OF CHARLES KING

Kathryn Whitford

Department of English

University of Wisconsin — Milwaukee

Life in the “Old Army” of the Indian

Wars has been admirably documented in

such books as Forty Miles a Day on Beans

and Hay and Frontier Regulars, both of

which treat the enlisted soldier primarily as

a fighting man, despite the inclusion of brief

episodes of garrison life.1 More recently

Glittering Misery has made a conscientious

effort to document the garrison life of of¬

ficers’ dependents in their sometimes heroic

efforts to maintain islands of civilization on

the western frontier.2 But Glittering Misery

by the sheer painstaking research that went

into its composition, chiefly proves how lit¬

tle evidence remains for a reconstruction of

garrison life among the officer class. The

reasons for that state of affairs are not far to

seek. First, of the innumerable letters that

were written to mothers and friends back

east, only a small portion has survived and

few of those contain the kind of information

historians desire. Second, officers’ wives,

after all, were no more likely to be gifted

correspondents than the general proportion

of educated women. And third, much of

what historians now wish to know were the

commonplaces of garrison social life and

were omitted because of their very famili¬

arity.

On the other hand, when officers or their

wives wrote for publication they designed

their works for an Eastern audience which

wanted to read of Indians and picturesque

frontiersmen rather than the problems of

educating children in Wyoming or Arizona.

Probably neither men nor women habitu¬

ated to garrison life foresaw a day when

army social rituals enacted in the wilderness

would seem as exotic as the Indian rituals

they supplanted. Moreover, publishing

memoirs was complicated by conventions

decreeing that few people could be named

in such memoirs and circumscribing what

could be said even of those. Nineteenth

century decorum threw a fierce privacy

about a “good” woman. Even in conversa¬

tion a gentleman spoke only of “my wife”

or “Mrs. _ He never employed her

given name except among her intimates.

He could not describe his wife’s activities

or those of other officers’ wives without

giving offense. In consequence even married

officers so completely omitted accounts of

social life from their recollections that on

a first reading their books appear to have

been written by bachelors.3

In the absence of detailed and even gos¬

sipy diaries, perhaps the best pictures of

garrison life available to historians today

are those created by the Milwaukee novel¬

ist, General Charles King (1844-1933)

who had served in the Indian fighting army

before turning author. General King was

the son of the editor of the Milwaukee

Sentinel He grew up in Milwaukee and

began his military career by serving as an

aide to his father, General Rufus King, dur¬

ing the Civil War. He subsequently grad¬

uated from West Point, served briefly as an

instructor there, and after being stationed

at New Orleans requested assignment to

Arizona where he took part in the Apache

campaign of 1874. He was severely wound¬

ed at Sunset Pass and awarded the Silver

Star for “gallantry in action” at Diamond

Butte. His regiment, the 5th cavalry, moved

to Wyoming for the Sioux campaign of

1876-77 during which King was promoted

to Captain and shortly thereafter retired

because of disability due to his earlier

wounds. He returned to Milwaukee where,

because his account Campaigning with

43

44

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

Crook had been published by the Sentinel,4

he already enjoyed a literary as well as

military reputation, and turned to novel

writing as a means of supplementing his re¬

tirement pay. He was recalled to service

during the Spanish American War, serving

in San Francisco and Manila and participat¬

ing in the defense of Manila during the

Philippine Insurrection which followed.

Once more he returned to Milwaukee and

novel writing. Although he was active in

the Wisconsin National Guard and trained

troops for World War I he was not per¬

mitted to go overseas in that war. He re¬

mained a familiar figure in military and

social circles in Milwaukee until his death

in 1933.

He employed all his military experience

in the approximately sixty volumes of nov¬

els, stories, and histories which he published

during his lifetime, but his best and most

interesting novels are those laid on the

Great Plains or in Arizona during the years

1871-78, particularly the group of novels in

which he follows the fortunes of a cavalry

regiment, rather like the 5th Cavalry, from

Arizona to Wyoming.5 In these novels he

drew heavily upon his own experiences and,

in fact, probably appears in them in the

secondary role of Billings, the adjutant.6

Because he unabashedly wrote to make

money, King developed a fool-proof plot

formula which combined a garrison love

story with one or more episodes in which

the cavalry is hotly engaged against hostile

Indians, thus assuring his readership among

both men and women. When his garrison

plots are stripped of such obvious plot de¬

vices as concealed identities, mistaken iden¬

tities and service in the field, a reader dis¬

covers that a fourth reason so few officers’

wives kept diaries or wrote their recollec¬

tions was the same reason that drove

trooper Charles E. Springer to discontinue

his daily journal, “to repeat day for day the

life of a frontier garrison in winter is no

charm.”7 Winter or summer, garrison life

was more notable for its sameness than its

variety. The degree of hardship undergone

in following one’s officer husband might

vary with the distance to the nearest rail¬

head; the chief discomfort might be heat or

cold, sandstorms or blizzards, depending

upon the geographical location of the fort;

there might be many women or few, de¬

pending upon the size and remoteness of

the fort; but army routine and army proto¬

col were the same. Everywhere their hus¬

bands were posted, the women awakened

to the same bugle notes and made the same

duty calls among virtually the same people.

In King’s novels, whether laid in Arizona

or on the great plains, it is the routine and

predicability of garrison life which forms

the background for the action of the plots

and, in fact, sometimes sets the plots in mo¬

tion. Ennui drives garrison inhabitants to

seek new amusements and focuses their at¬

tention upon minor infractions of social or

military convention.

Even the passage of years did not alter

army routine; although it added new forts

to the frontiers and slowly and subtly

changed the composition of the officer

corps.

In King’s early novels, the majors and

colonels who commanded posts were usu¬

ally veterans of the Civil War who had

risen to command by service in the Union

Army. Among them were men who held

brevet rank considerably higher than their

permanent army rank. Men such as these,

without formal military training, might be

admired by their troopers, but in King’s

novels were eyed askance by their subordi¬

nate West Point officers unless they were

truly exceptional commanders.8 The young

first and second lieutenants and some of the

captains were usually West Point graduates,

often the sons of army officers who, like

King himself, were carrying on a family

military tradition. The social uneasiness

generated by a situation in which the West

Point men were often outranked by men

they considered their social inferiors was

not dissipated until, with the passage of

1979]

Whitford — Garrison Life in the Novels of Charles King

45

time, King’s West Point heroes rose to com¬

mand. Under those conditions an officer’s

wife needed the diplomatic skills of an am¬

bassador to prevent being drawn into one

camp or another, but such social manuever-

ing is a subject better fitted for a novel than

for memoirs.

King wrote about the officer class not

only because it was the class he knew best

but probably because as Fred Dustin has

commented, “In 1876 the average enlisted

man counted for little: in the army he was

only a number on the rolls. By the com¬

fortable civilian he was despised.’’9 King

created a few vivid and sympathetic

sketches of cavalry troopers but he almost

completely ignored their wives. Although

there was an area of each post known as

“sudsville” it constituted an entirely sep¬

arate social entity. Even when, as in the

novel Captain Blake , troopers and their

wives attended a play produced by the

officers, they were seated in a separate sec¬

tion from that reserved for officers and their

wives.10

Although in King’s garrison novels,

money was never an acceptable substitute

for courage or breeding, officers were ex¬

pected to have private means by which to

supplement their pay. It was axiomatic that

a lieutenant could hardly maintain himself,

much less a wife and family, upon his of¬

ficer’s pay. Precisely because garrison life

was remote, it was also costly. Household

goods had to be transported long distances

and were expensive upon the frontier. In

addition there were visits to be made to

aging parents, and son and daughters to be

sent away to appropriate Eastern schools.

The sons in these novels often followed in

their father’s footsteps at West Point. The

women vacationed near West Point in or¬

der to visit their sons and brothers and to

meet eligible young officers. As a result of

army marriages, garrison society was knit

not only by bonds of comradeship but by

ties of blood.

The education of officer’s daughters was

obtained in good finishing schools rather

than colleges. None of the women in King’s

novels was bookish, although a few of the

young officers had scholarly interests. In

fact, although army wives supervised the

children’s lessons, in King's novels they

were rarely seen with books in their hands

except when reading to invalids. Neither

did they do much sewing, usually buying or

ordering their clothes on trips to Chicago

or New York. Paris couture was a matter

for comment, but dresses from the East

were commonplace.

Officers’ daughters became officers’ wives

at 18 or 19 in many cases, so that in these

novels they were often ten years younger

than their husbands. Families, however,

were small and that fact combined with the

presence of servants freed the women from

the demands of domesticity. Many of the

women rode well and several of the wives

who had grown up in frontier garrisons

could handle a Smith and Wesson revolver

competently. These were women whom

King and the garrison unanimously ap¬

proved.

Once married, King’s heroines conducted

their domestic lives in a succession of adobe

or drab frame houses constructed by army

labor. Most of them were built as double

houses with a common wall.11 They varied

somewhat in overall size and in the number

of bedrooms, but all were shaded by ver¬

andas facing the parade ground. They usu¬

ally ended in lean-to kitchens giving upon

small yards bounded by a low fence. They

had at best only partial basements, or cel¬

lars. In summer the wooden walls cracked

and peeled under the heat of the plains

winds; in winter the foundations were

banked with straw and the snowdrifts piled

to the second story. Gales from the moun¬

tains swayed the Navajo blankets (brought

from Arizona) hanging across the interior

doorways, despite storm sashes and the

layer of cotton stuffed around the windows.

The carpets, stretched over several thick¬

nesses of newspaper, rose and bellied under

46

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

the force of winter winds. Parlor and

kitchen stoves provided heat and kerosene

lamps, light, during the long winter eve¬

nings. The cry of “fire” echoes through

General King’s novels somewhat monoto¬

nously, but undoubtedly reflects an ever¬

present danger. Whether because of the pos¬

sibility of fire or because there was no

suitable storage place provided, in at least

two of King’s novels the family silver seems

to have been kept in a box or basket under

the bed when it was not used for entertain¬

ing.12

Such were the houses available to the

officers and their wives. And even these

were assigned by rank. One house, the

largest and best, was designated for the

Colonel commanding. Two houses, on

larger posts, might be generally considered

to belong to the majors. Thereafter the

captains chose their quarters in order based

on the date of their promotions. The 1st

Lieutenants followed, also in order of pro¬

motion and finally the 2nd Lieutenants, al¬

most invariably bachelors, doubled up in

the houses no superior had claimed. An of¬

ficer newly transferred to a post had “bump¬

ing rights” over all his subordinates so that

he had the power to set off an elaborate

series of housing shifts each giving rise to

politely concealed ill-will. King illustrates

the operation of this ritual when an entire

regiment was assigned to a new post.

The bachelor officers pitched tents on the

parade and placidly waited their turn to

choose quarters, a ceremony which im¬

pressed Miss Leroy as something incompre¬

hensible. It was not easy to make her realize

just why Captain Ray couldn’t move Mrs.

Ray and the baby boys up from the hotel

until Captain Freeman had chosen, and why

Mrs. Blake should remain at Cheyenne near

her old home until the Trusc'otts and Rays

had settled on what houses they would take.

(They wanted the big double brick next but

one to the Colonel’s, but were afraid to

move in, lest the new surgeon ordered out

from Omaha should take a fancy to that

very set.)13

It is no wonder that seasoned army wives

like Mrs. Stannard who appears in many of

King’s novels learned to live with as few

chattels as possible. But beyond the incon¬

venience of housing shifts there was another

cruder consequence of the limited post

housing. An officer killed in action was

promptly replaced on post. That meant that

his widow was forced to vacate her house,

leaving behind not only the army life that

she knew but the friends who could best

understand her grief and bewilderment.14

In addition to the officers and their fami¬

lies, garrison houses often accommodated

live-in servants. Bachelor officers had army

strikers to serve them, but married men em¬

ployed either a combination cook and

housemaid or both. In families with chil¬

dren a nursemaid was not uncommon. Such

servants were drawn from diverse sources,

but the majority of the servants in King’s

novels were Irish. Sometimes a veteran

trooper no longer fit for military service

took civilian service with an officer in order

to stay with the life he knew. The wives

and daughters of troopers were also em¬

ployed by officers’ households, but since

many of the troopers were Irish they and

their families merely swelled the Irish ser¬

vant brigade. These servants were impor¬

tant in King’s novels, as they probably were

in life, because they provided a channel of

unofficial communication between the

troops and the officers’ households. They

formed an excitable, gregarious society

from which the Blacks, Chinese and the oc¬

casional Indian girl were largely excluded.

However an immeasurable gulf separated a

pretty Irish servant girl from the daughter

of the house. The servant girl could be

courted by troopers; the daughter of the

house only by officers. Despite the scarcity

of women on post, in King’s novels no

young officer ever falls in love with an

1979]

Whitford — Garrison Life in the Novels of Charles King

47

Irish servant girl. On the other hand, Mrs.

Snaffle and Mrs. Wilkins, captain’s wives,

lapse into brogue when excited, thus sug¬

gesting that officers of an earlier generation

had been less fastidious.15

The lives of all these people, officers,

troopers, wives and servants, moved to the

measure of trumpet calls and focused, in

good weather at least, upon the parade

ground. Each day began with the boom of

the morning gun and ended with the sunset

gun. Between those events the trumpet

called Reveille, Stables (this was a cavalry

regiment), Sick call and Fatigue, Boots and

Saddles for morning parade, Adjutant’s call,

Guard mounting, Drill call, Recall, Dinner,

Squad drill, Company drill, Stables, Retreat

and Evening Dress Parade.16 After the Sun¬

set Gun it called Tattoo, and Lights Out,

although those calls spoke only to the bar¬

racks. No village clock ever regulated lives

as rigidly as did those trumpet calls. They

also had power to disrupt lives — to sound

“officers call” or more importunately “the

General” (general assembly) at any hour;

thus sending troops and their officers to

intercept Indians off reservations, to relieve

some beleaguered stage station, or to quell

riots in town. In King’s novels men wel¬

comed “the General” almost as fervidly as

women dreaded it.

As the day was ordered by trumpets so

the night was divided by sentry calls. The

sentries posted on the perimeter of the gar¬

rison did not walk beats which met. Instead,

at regular intervals, they called off and the

sound moved in order around the post. Like

Tattoo and Lights Out the call of the sen¬

tries did not concern most of the officers,

but only the Officer of the Guard.

For the other officers and their wives

evening parade signalled a transition from

business to pleasure, from army time to

time that might be called their own. Eve¬

ning parade was not only a military but a

social event. The women of Officers Row,

freshly dressed and at leisure before dinner,

sat on the verandas facing the parade or

strolled from house to house to exchange

the day’s news, to admire their husbands in

dress uniform to watch parade and listen

to the band. Even small posts maintained a

band and large forts such as Russell and

Cushing prided themselves on the excel¬

lence of their music. As parade broke up,

the officers, released from their day’s rou¬

tine, had time to join their wives and

friends. Courtesy required that all acquain¬

tances be acknowledged and since everyone

knew everyone else the way homeward was

more social than military.

Bachelor officers might go to the post or

trader’s store, there to get a drink and settle

down for an evening of poker. Married of¬

ficers sometimes joined them briefly. Al¬

though a room at the store was usually the

only approximation of an Officers’ Club,

the officers who spent too much time there

were suspect in King’s novels. If married,

they were assumed to frequent the club be¬

cause they did not get along with their

wives; if single, because they were unable

to command invitations to the better homes

on the post. The cardroom, King implied,

attracted grumblers, created ill will, and

fomented the male equivalent of female

gossip.

Generally the social life of the Garrison

was orderly, decorous, and predictable.

There were frequent invitations to dine, but

since rooms were small, dinner parties of

more than ten were practically unknown.

However there was a great deal of casual

calling and stated occasions when formal

calls were de rigeur. Thus dinner required

a proper dinner call the following day. All

guests in the garrison and all new families

posted to the fort were to be called upon

immediately. Engagements, birthdays and

promotions required calls of congratulation,

and bereavement calls of sympathy. All

these were the common courtesies of a

somewhat formal age, and would have been

conducted equally punctiliously by ladies

48

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

and gentlemen in any Eastern city. How¬

ever, garrison social life produced some dis¬

tinctive patterns. For one thing many of

the wives, all in fact of the most admirable

in King’s eyes, did not entertain while their

husbands were off on campaign. When

three or four troops were out of the fort

this convention necessarily created periods

of social doldrums even during the winter,

when the social season was at its height.

During the summer, the usual campaign

season, the effect was exaggerated by the

fact that many wives summered in the East.

When wives returned in the fall they often

brought guests to spend the winter. One

might suspect that the number of attractive

feminine guests thus introduced to fort so¬

ciety was merely a device employed to pro¬

vide the love interest General King required

for his plots. However, a story which might

have been written by General King himself

is contained in the memoirs of General

Cruse. His wife’s sixteen year old sister

spent the winter at Fort Lowell, was courted

by Lt. Hodgson over the family objections

that she was too young, and married her

cavalry officer the following year.17

The Custer’s, also, apparently made a

practice of inviting pretty girls for extended

visits.18 Thus Elizabeth Custer’s narratives

of life in the Plains forts serve to substan¬

tiate the pretty roommates, cousins and

nieces who enliven garrison life in King’s

novels.

For entertainment these girls were of¬

fered dinners with the middle aged, the

spectacle of parade, informal evening par¬

ties, escorted rides into the surrounding

country, skating on the Platte, sleigh rides

after sufficient snowfall, band concerts, an

occasional formal German and even ama¬

teur theatricals. In suitable fall weather

there might also be a jack rabbit hunt over

the hills, ending in a picnic to which the

ladies who neither rode nor hunted were

conveyed by ambulance. The chief enter¬

tainments, however, were the frequent in¬

formal dances, or hops. When young Mrs.

Turner complained bitterly that the hops,

held almost nightly in Arizona, were less

frequent in the Wyoming garrisons, Mrs.

Stannard explained that in Arizona the

band had not been mounted and therefore

did not have to attend morning and evening

stables.19 Therefore, evening duty for hops

was not an unreasonable demand upon

them. Since the bands in Wyoming were

mounted, nightly hops would be an unwar¬

rantable imposition even though hops broke

up early. In fact all the social life of the

garrison except the most formal dances

ended by eleven or twelve o’clock. The

army world believed in early to bed and

early to rise. Stable call, attended by the of¬

ficers, sounded at 6:00 a.m.

Although much of an officer’s life was

regulated by official duties and protocol

there was another, generally female, side of

garrison life not subject to official rules.

Secrets and news, passing through official

channels among the men, passed also from

veranda to veranda among the women, and

from kitchen to kitchen among the cooks.

Cut off from their families women sup¬

ported each other through illnesses and

bereavements. Their hospitality was bound¬

less, for new faces and new conversation

were both welcome means of enlivening

garrison social life. Not only were guests

invited for months, but the same generosity

was extended to the relatives of bachelor

officers, to wives until they were settled on

a post and to young officers convalescing

from wounds. Women lent silver, dishes

and recipes, sat with invalids, encouraged

courtships, wrote innumerable letters link¬

ing garrison to garrison and frontier to

families in the East, mentored young offi¬

cers, read weeks old newspapers, and spec¬

ulated on their next removal. In such en¬

terprises the women were restrained only

by their own discretion and their husbands’

authority. And therein lay the danger.

Duane N. Greene wrote in a sketch of

army life that “The most discordant garri¬

sons are those comprising the greatest num-

1979]

Whit ford — Garrison Life in the Novels of Charles King

49

ber of ladies.”20 Novelist King, despite his

frequent praise for Officers’ wives who left

comfortable Eastern homes to follow the

cavalry into frontier garrisons, tends to sub¬

stantiate Greene’s pronouncement. Women

were at the root of almost all garrison trou¬

ble in his novels. Mrs. Pelham interferred

in the promotions and leaves granted her

husband’s men.21 Mrs. Turner gossiped un¬

til her husband’s friends one by one de¬

serted him.22 Mrs. Wilkins was the terror

of social engagements despite her warm

heart.23 Nanette Flower proved a spy and a

traitor24 and Mrs. Granger enthralled

Tommy Hollis and Captain Blake for her

own ends.25 Younger women disrupted the

post by their very presence — -two of them

attracted undesirable kinsmen to the area

of the fort.26 Other girls evoked rivalries

between officers. It may seem unreason¬

able to include male rivalries in the list of

women’s derelictions, but King himself

raised the issue in Marion's Faith when he

wrote “Things were in almost as eruptive a

state at Russell ... as they had been at old

Sandy during the Pelham regime, only —

only who could this time say there was a

woman at the bottom of it? And yet was it

not Gleason’s unrequited attention to our

heroine that prompted much of the trou¬

ble?”27

Taken altogether the roles of women in

General King’s novels are fairly unambigu¬

ous. The women he admired, the women he

assigned to his heroic officers as rewards for

valor and patience, proved to be fine wives,

but they served the garrison by their discre¬

tion and their adoption of the officer’s code.

They became army. Other women, and even

those before their marriages, were sources

of disruption. Since the army could not drill

women into shape like a slack cavalry

troop, it was with a sense of real relief on

the part of both author and characters that

men took the field. Once on campaign the

desirability of order and discipline became

self-evident and the foe could be neatly

identified by his warpaint. When King’s

officers responded to “the General” they

moved from a world complicated by women

to the simpler, if more violent, world of

primal landscapes and single purposes. In

King’s novels they went gladly, for as Cap¬

tain Ray, one of King’s chief heroes an¬

nounced profoundly, “Garrison life and

girls spoil many a good cavalryman.”28

Notes

1 Don Rickey Jr., Forty Miles a Day on Beans

and Hay (Norman, Oklahoma, 1963), Robert

M. Utley, Frontier Regulars (New York, 1973).

2 Patricia Y. Stallard, Glittering Misery, De¬

pendents of the Indian Fighting Army (San

Rafael, California, 1978).

3 General Crook refers to “Mrs. Crook” only

in two consecutive paragraphs. General George

Crook , his Autobiography ed. Martin F. Schmitt

(Norman, Oklahoma, 1946) p. 155. General

Miles refers once to “my wife” and once, men¬

tioning the Custers at Fort Hays, says “Mrs. Miles

being with me, we frequently met them socially

and enjoyed many hunts and pleasure parties

together.” Nelson A. Miles, Personal Recollec¬

tions of General Nelson A. Miles (New York,

1969) pp. 151, 256. George A. Custer, My Life

on the Plains (Norman, Oklahoma, 1962) refers

to his wife once as “Mrs. Custer” and once as

“my wife,” pp. 68, 66.

4 Campaigning with Crook was first published

serially in the Sentinel and then in a paperbound

edition (Milwaukee, 1880 before being published

by Harper and Brothers, New York, 1890.

5 Charles King, The Colonel’s Daughter (Phila¬

delphia, 1882). Charles King, Marion’s Faith

(Philadelphia, 1886). Charles King, Captain

Blake (Philadelphia, 1892). Although the char¬

acters introduced in The Colonel’s Daughter ap¬

pear in, or are mentioned in, other novels by

King, the three listed here are the most important

to the history of King’s fictitious regiment.

0 For King’s employment of his own experiences

see Harry H. Anderson, “Home and Family as

sources of Charles King’s Fiction,” Milwaukee

County Historical Society Historical Messenger ”

31:68 (Summer, 1975).

1 Charles E. Springer, Soldiering in Sioux Coun¬

try, ed. Benjamin Franklin Cooling III (San

Diego, 1970), p. 73.

8 In King, Marion’s Faith, Colonel Whaling

and his wife are condescended to by almost all

the cavalry officers. Also in Charles King, Ray’s

Recruit (Philadelphia, 1898) Capt. Mainwaring

is a decent officer whose selfconsciousness about

50

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

his lack of formal education makes him the butt

of some of the younger officers’ humor.

9 Fred Dustin, in W. A. Graham, The Custer

Myth (New York, 1953), p. 369.

10 King, Captain Blake, p. 190.

11 The composite description derived from

King's novels can be compared with the following

descriptions from the annual inspection of public

buildings which appears in Richard Upton, Fort

Custer on the Big Horn (Glendale, California,

1973).

No. 6. Commanding Officer’s Quarters, frame,

single set, IVi stories, size 42 by 36 feet, four

rooms on each floor; “L” to same embracing

kitchen, pantry, closets, with attic chamber; size

20 by 16 feet. Porch in front.

No. 1 to 5, and 7 to 11. Officer’s quarters,

frame, ten double sets, IVi stories, size 48 feet

by 46 feet six inches, six rooms on first and

four rooms on second floor; “L” to same, size

34 by 14 feet embracing two kitchens, with pan-

tries, closets and attic chambers. Porch in front.”

p. 280.

12 Charles King, A Daughter of the Sioux (New

York, 1903), p. 215 and Charles King, A Gar¬

rison Tangle (New York, 1896), p. 163.

53 King, Ray’s Recruit, p. 102.

14 Mrs. Farrar is widowed in Charles King’s

Fort Frayne (New York, 1895), Mrs. Turner in

Captain Blake and Mrs. Winn in Charles King’s

A Trooper Galahad (Philadelphia, 1898).

15 Charles King, Lanier of the Cavalry (Phila¬

delphia, 1892) and The Colonel’s Daughter.

30 This list (which is not exhaustive) is based

on a list of calls in Charles King, Trials of a Staff

Officer (Philadelphia, 1895), p. 15.

17 Thomas Cruse, Apache Days and After

(Caldwell, Idaho, 1941), pp. 181, 191.

18 Elizabeth Bacon Custer, Tenting on the

Plains, 1887 (reprinted Williamstown, Massa¬

chusetts, 1973) p. 208.

19 King, Marion’s Faith, p. 35.

20 Duane N. Greene, Ladies and Officers of the

United States Army; or American Aristocracy, A

Sketch of the Social Life and Character of the

Army (Chicago, 1880).

21 King, The Colonel’s Daughter .

22 King, Captain Blake.

23 King, The Colonel’s Daughter.

24 King, A Daughter of the Sioux.

25 King, Captain Blake.

20 King, A Garrison Tangle, Lanier of the Cav¬

alry.

27 King, Marion’s Faith, p. 314.

28 King, Ray’s Recruit, p. 40.

AGENTS OF THREE NATIONS IN THE

FOX RIVER VALLEY, 1634 TO 1840

Charles D. Goff

Department of Political Science

University of Wisconsin-Oshkosh

While explorers, fur traders, Indians,

priests and a very few white settlers were

the major actors in the early history of the

Fox Valley, an impressive number of note¬

worthy events were set in motion by people

carrying out the policies of distant govern¬

ments.

French Era Travelers Through the

Fox Valley 1634-1763

The first government to wield any sov¬

ereign authority in the present-day Fox Val¬

ley of northeastern Wisconsin was the colo¬

nial government of French Canada. Perma¬

nent French settlements were established at

Quebec in 1608, Three Rivers in 1634 and

Montreal in 1642. Only twenty-six years

elapsed from the founding of Quebec to the

day when Jean Nicolet, an official emissary

of Governor Champlain, fired his pistols and

wore his finest damask robe to impress Win¬

nebago Indians at Red Banks northeast of

Green Bay. Nicolet’s arrival within the

boundaries of present-day Wisconsin was

the first expression of European authority in

the Fox Valley wilderness.

The first reference on any map to the Fox

River Valley was a map drawn by Jean

Boisseau at Paris in 1643 which shows the

Fox River (Riviere des Puans), Lake Win¬

nebago (Lac des Puans), and the tribal re¬

gion in which the Winnebagoes resided in

1643 (La Nahon des Puans). The map ref¬

erences to the Winnebagoes presumably

came from hearsay reports obtained by

Nicolet from Indians familiar with the top¬

ography of the Fox Valley.1

The first official agents of governmental

action in the Fox Valley appear to have

been the seven members of the Louis Joliet

expedition. Joliet and his interpreter-cartog¬

rapher-chaplain associate, Father Jacques

Marquette voyaged up the Fox River in late

May and early June of 1673 at the order

of Canadian Intendant Talon to investigate

reports that there was a great river in the

west of New France called the Mississippi.2

Father Marquette’s map accurately shows

the Fox-Wisconsin river system including

Lakes Winnebago and Butte des Morts, in¬

dicates the Wolf River and sketches cor¬

rectly the Illinois River from the Mississippi

(R. de la Conception) to Lake Michigan

(Lac des Illinois).

The first large scale governmental action

within the Fox Valley was the invasion in

1716 by a French military expedition during

the First Fox War. The French “army” was

organized at the order of Canadian Gover¬

nor Philippe Vaudreuil and commanded by

the King’s Lieutenant at Quebec, French

Army Major Louis de la Porte, Sieur de

Louvigny. The force consisted of some 225

French soldiers and fur traders, and 600

armed Indians, two brass cannon and a brass

grenade mortar. Louvigny’s orders were to

clear the Fox River of all hostile blockading

forces, including a fortification of the Fox

Indians located on the southwestern shore

of Big Lake Butte des Morts.

The Fox fort, some four miles west of

the present-day site of Oshkosh and a short

way east of the main Fox village on Big

Lake Butte des Morts was described by

Father Pierre F. X. Charlevoix, the expedi¬

tion’s chaplain, to his Jesuit superiors, as a

“sort of fort surrounded by three ranges of

oak palisades with a good ditch in the rear.”

Both Louvigny and Charlevoix said the Fox

fort was garrisoned by 500 warriors (300

51

52

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

additional warriors were absent on a war

party) and 3,000 women.

Louvigny bombarded the Fox fort with

his two cannon and grenade mortar, until

“On the third day . . . while I was preparing

to undermine their works by placing mine

boxes under their wall the Foxes proposed

terms of capitulation . . . and I . . . con¬

cluded peace with the Foxes.3

Records of Louvigny’s siege do not indi¬

cate the fort’s exact location, but in “The

Bell Site: an early Fox village” archaeolo¬

gist Warren Wittry reported finding iron

grenade fragments on the southwest shore

of Lake Butte des Morts, thus suggesting

that:

“. . . this site might be the one attacked by

Louvigny in 1716 ... So far as is known to

the writer, the grenades fired by Louvigny

are the only ones ever shot in central Wis¬

consin. . . .”4

The First Fox War ended in 1716 and

the Second Fox War began in 1728, largely

because the Foxes, according to Father

Charlevoix:

“infested with their robberies and filled with

murders not only the neighborhood of

Green Bay, their natural territory, but al¬

most all the routes communicating with the

remote (French) colonial posts.”5

Canadian Governor-General Beauharnois,

desiring to end the Fox Indians outrages,

believing French lives must not go un¬

avenged and being determined to overawe

both the Iroquois and the Foxes by a bril¬

liant stroke, sent a second military force

to present Winnebago County in the sum¬

mer of 1728. Commanded by Constant

Marchand, Sieur de Lignery this force, con¬

sisting of 400 French and 1,100-1,200 In¬

dians left Mackinac August 10, 1728 and

advanced along Green Bay in bateaux and

canoes (Fig. 1). Lignery provoked and then

defeated the Menominees on August 15th

and reached the west end of Big Lake Butte

des Morts by August 25th.6

Having had timely notice of the French

army’s approach, and lacking warriors to

match Lignery’s force, the Foxes retreated

up the Fox River and into the forests west

of Omro, where Lignery felt it imprudent

to pursue them.

Fig. 1. This “Map of the Land of the Savage Foxes,” was published in 1730 by Gaspard

Chaussegros de Lery, to illustrate his account of the Lignery expedition’s attack in 1728 on

the fort of the Fox Indians, then resident on Big Lake Butte des Morts and Lake Winnebago

in east central Wisconsin. De Lery's map appears to be the first to show details of the Fox

River valley including vicinities of present day Omro, Oshkosh, Appleton, Kaukauna, and

Green Bay. Source: L. P. Kellogg, French Regime in Wisconsin (1925), 314.

1979]

Goff — -Three Nations in the Fox River Valley

53

Lignery remained at Lake Butte des

Morts only briefly because his army had to

return to Montreal before the rivers froze

and the first snows fell. Consequently he

was able only to burn the Fox villages and

destroy their unharvested cornfields and

gardens. In retreating down to Green Bay,

Lignery withdrew the La Baye garrison

and burned the fort since his expedition

had chiefly succeeded in infuriating the

Foxes without inflicting significant damage.

Lignery concluded that the small fort at

Green Bay would not be able to defend it¬

self against the certainty of revenge attacks

in the following fall and winter.7

Weighed on a scale of military success or

failure, it would seem that it was the Foxes,

not the French who were victorious. Ac¬

tually, a delayed consequence of Lignery’s

invasion of 1728 was that the Fox Confed¬

eracy of anti-French tribes suddenly began

to crumble. News came to the Fox villages

on Lake Butte des Morts that their Kicka-

poo and Mascouten allies had made peace

with the Illinois Indians. At the same time

the Sioux refused to continue their agree¬

ment to grant the Foxes refuge in Sioux

country if they could not resist the onslaught

of the French. Perhaps most upsetting to the

Foxes was the fact that Montreal was

crowded in 1729 with delegates from west¬

ern tribes of the upper Great Lakes, all de¬

claring their love for the French and their

hatred for the Foxes.

Canadian Governor Beauharnois now

changed his commander. Paul Marin, former

commandant at Fort La Baye was selected

to lead a third expedition against the Foxes

in 1729. Marin’s party left Montreal in the

summer of 1729 but he came only as far

west and south as the vicinity of former Fort

La Baye, where he wintered. Early in the

spring of 1730 a great military opportunity

suddenly presented itself to Marin. A por¬

tion of the Winnebago who had deserted the

Fox confederacy returned to their village

site on Doty Island in Lake Winnebago and

were suddenly attacked by a large force of

Foxes. Marin’s party hurried to the assis¬

tance of the Winnebagoes with a force of

Menominees and a few fur traders. After

five days of grim fighting the Foxes broke

off their attack during the night.8

Disasters now began to fall on the Foxes

like hailstones. One was an incredible effort

in 1730 to move the entire tribe to present-

day New York state to seek refuge among

the Iroquois who had long solicited their

alliance and had offered them asylum. This

effort was thwarted by the cooperation of

the French settlements at St. Joseph and

Detroit, Michigan; in the Illinois Valley; at

Vincennes, Indiana and Miami, Ohio, which

raised a combined force of 1,400 French

and Indians to besiege the Foxes who had

halted and erected a fort on the Illinois

prairie on an eastern branch of the Kas-

kaskia River. The Foxes defended their

position for 23 days, abandoning it under

cover of darkness and a storm on September

9, 1730. Next day, the Foxes, burdened

with their families, were overtaken and two

or three hundred warriors with an equal

number of women and children were mas¬

sacred. Four or five hundred more were

captured and scattered as slaves among the

Indian allies of the French. Only several

hundred Foxes escaped and fled back to

Wisconsin.9

Another disaster befell the Foxes in the

winter of 1731-32 when Governor Beau¬

harnois connived at the departure from

Montreal of a war party of mission Indians

who came to Wisconsin to “eat up Rey¬

nards.” Proceeding part of the way over¬

land on snow shoes, the mission Indians

attacked a Fox village of fifty cabins on the

Mississippi River about five miles above

the mouth of the Wisconsin River, killing

or capturing 300, only “thirty ‘true’ Rey¬

nards” escaping.10

It was a poignant disaster for the Foxes

when war chief Kiala and three fellow chiefs

surrendered themselves to Commandant Vil-

liers at Fort La Baye as hostages to the

French, in the hope the remainder of the

54

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

Fox tribe would be spared. Instead of being

merciful, Governor Beauharnois condemned

the Fox chiefs to a slow death by chain

gang labor and tropical heat at a sugar cane

plantation on Martinique.11

The crowning Fox disaster came in Sep¬

tember, 1733. Captain Villiers returned to

Fort La Baye with orders from Governor

Beauharnois to exterminate all remaining

Foxes; men, women and children. When

rumors of Villiers’ orders reached the Sauk

village at La Baye, the Sauks were seriously

alarmed, as they had permitted the few re¬

maining Foxes to take refuge with them in

their village. If the French were to insist

on the surrender of the Fox refugees, the

Sauks would be forced to break the cus¬

tomary Indian obligation of hospitality and

protection to their persecuted relatives.

Their alarm would have been even greater

had they realized that the stage was being

set for the bloodiest battle ever to occur in

Winnebago County, the Battle of Little Lake

Butte des Morts.

When Commandant Villiers walked up

to the locked gate of the Sauks’ palisaded

village at La Baye on September 16, 1733

with less than a squad of the more than 300

garrison troops, fur traders and allied In¬

dians under his command and rashly tried

to break open the gate, a shot from within

the village killed one of Villiers’ sons. Vil¬

liers angrily returned the fire, at which a

twelve year old Sauk in one shot killed the

commandant. A brief fire fight at the Sauk

gate now ensued in which ten French at¬

tackers, including Repentigny, second in

command at Fort La Baye, were killed and

three others wounded.12

When the Sauk chiefs realized this much

French blood had been shed they also real¬

ized that French retribution would come

quickly and that they had no alternative but

to retreat from La Baye as quickly and

secretly as possible. The Sauk-Foxes left

their village under cover of darkness on the

third night following the fight at their vil¬

lage gate, proceeding up the Fox River.

The next day the dead commandant’s

oldest son, Ensign Louis Villiers, organized

a pursuit force of most of the La Baye

garrison plus whatever fur traders were in

town. Villiers’ party overtook the fleeing

Sauk-Foxes by late afternoon at a point

somewhere close to the present-day high

level bridge across Little Lake Butte des

Morts. The historic Battle of Butte des

Morts then began. According to Louise Kel-

logg:

“The French lost heavily as well as the

Sauk-Foxes. Among the French officers

alone, one or more members of the Villiers,

Ailleboust, Du Plessis and other well-known

families, mourned the death of their youth.

. . . Ensign Louis Villiers was wounded and

Augustin Grignon recalled at a later time

that two of his uncles were killed in this bat¬

tle. It is believed that the great mound at

(Little) Butte des Morts was erected to cover

the tribesmen slain in this battle.”13

Canadian Governor-General Beauharnois

reported to France on November 11, 1733,

that the losses of the French and their In¬

dian allies at the Battle of Butte des Morts

were 34 dead and about 20 wounded. These

casualties were in addition to the ten killed

at the Sauk-Fox village gate. The Sauk-Fox

left 20 Sauk and six Fox dead on the field

and had an unknown number killed and

wounded.14

The Sauk-Foxes who could travel now

abandoned the Fox River country, moved

west of the Mississippi, found refuge among

the Sioux and built a palisaded village on

the Wapsipinicon River in eastern Iowa.

Years later they established themselves for

several generations on the lower Rock River

in northwestern Illinois. After the Battle of

Butte des Morts, several other tribes, se¬

cretly pleased with the resistance of the

Foxes, returned their Fox captives, so that

Sauk-Fox numbers in Iowa and Illinois

again became substantial.

The final French effort to exterminate

1979]

Goff — Three Nations in the Fox River Valley

55

the Foxes in the Second Fox War was a

winter expedition in 1734-35 organized at

the order of Governor Beauharnois and

commanded by Captain Nicolas Joseph de

Noyelles. De Noyelles’ force consisted of

84 French soldiers, habitants and 200 mis¬

sion Iroquois Indians. When de Noyelles

moved up the Fox Valley, winter had al¬

ready begun. Abandoning their canoes and

proceeding on snow shoes, de Noyelles

finally found the tribesmen he sought in

early April, 1735, fully entrenched on an

island in the Des Moines River. After futile

attacks in which several French officers

were killed, de Noyelles abandoned his ob¬

jective, his party making their way to the

nearest French settlement south of present-

day St. Louis.15

When the chiefs from the upper country,

assembled at Montreal in 1737, asked for

mercy for the Sauks and Foxes, Governor

Beauharnois agreed and sent Pierre Paul

Marin to conciliate them. Marin succeeded,

first at an Iowa post and later at a trading

post among the Sioux on Lake Pepin.

When friction between French and British

colonies in North America developed into

armed conflict in 1755, French governmen¬

tal recruiters and their Indian enlistees

passed through the Fox Valley repeatedly.

Paul Marin and Charles de Langlade both

recruited and commanded increasingly large

contingents of Western Indians in support

of French Canada. Langlade and his Indians

were sent home in 1760 only a few days

before the surrender of the final French

Canadian stronghold, Montreal, to British

General Amherst. Langlade brought the

news to Mackinac and La Baye that all of

French Canada had surrendered to the En¬

glish on September 8, 1760.

The last French governmental force to

traverse the Fox-Wisconsin was the com¬

bined garrisons and officers of fort Mac¬

kinac and La Baye, commanded by Captain

Louis Beaujeau. They retreated up the Fox

River in October of 1760 and ultimately

reached New Orleans. The French era in

Wisconsin was at an end.10

British Era Travelers Through the

Fox Valley 1763-1816

Jonathan Carver in 1766 was the first

Englishman to travel up the Fox Valley,

and Peter Pond, in September, 1773, was

probably the first Yankee to travel from

Green Bay to Prairie du Chien.17 Neither

Carver nor Pond came to the Fox Valley as

a consequence of any governmental policy.

However, during the American Revolu¬

tionary War, it was British policy that

caused Charles Langlade of Green Bay to

lead Wisconsin Indians to Quebec in 1776,

1777, and 1778. Some of Langlade’s Wis¬

consin Indians supported British General

Burgoyne’s campaign in upstate New York

and became a scourge to American frontier

settlements.

In 1780 Menominee and Winnebago In¬

dians accompanied the unsuccessful British

attack of Emanuel Hess on the American

fort at Cahokia (East St. Louis) and the

Spanish fort at St. Louis. George McBeath,

a fur trader and interpreter for Langlade,

was the last British traveler along the Fox-

Wisconsin having a role in the American

Revolutionary War. He announced to a

council of tribesmen at Prairie du Chien on

May 24, 1783, the approach of peace and

advised the Indians to devote their future

energies to hunting.18

Once the United States had become an

independent nation, its sovereignty extended

west to the Mississippi and north to the Ca¬

nadian border. Wisconsin supposedly was

American. In reality British fur traders con¬

tinued to operate at will throughout the area.

Wisconsin Indians were British in sympathy

and the Wisconsin fur trade functionally

was still Canadian.

It was about May 1, 1812, that Francois

Reaume, a secret courier from Canadian

General Isaac Brock, passed along the

Fox-Wisconsin with an urgent message for

56

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

Robert Dickson, chief trader at Prairie du

Chien for the Northwest Company. Gen¬

eral Brock’s message to Dickson, who now

became a colonel in the Canadian army,

warned of the near approach of war be¬

tween Britain and the United States and in

guarded terms suggested Dickson begin re¬

cruiting Indians. In the ensuing months,

hundreds of Indians from the Sioux, Sauk,

Winnebago and Menominee tribes, Includ¬

ing Menominee Chief Tomah and his pro¬

tege Oshkosh and 200 Sauk-Fox warriors

under Sauk Chief Black Hawk reported

to Robert Dickson for British military

duty.

A British surprise attack on Fort Mac¬

kinac captured the American island fortress

in the War of 1812 before the American

garrison even knew war had been declared.

The British had not only prepared for war

by inflaming Indian tribes against the Amer¬

icans, but had secretly set their various

forces in motion before the actual outbreak

of war, thus enabling them to achieve tacti¬

cal surprise at Fort Mackinac.20

The only military action on Wisconsin

soil during the War of 1812 occurred in

1814 when Americans at St. Louis moved

up the Mississippi and built Fort Shelby

at Prairie du Chien. British Colonel Robert

Dickson recognized that an American strong

point at such a strategic location would un¬

dermine continued British control of the fur

trade in Wisconsin and the upper Missis¬

sippi valley. Accordingly, he quickly or¬

ganized a force of Indians, fur traders and

a few British soldiers under the command of

Lt. Col. Wm. McKay which compelled the

American construction party commanded by

Lt. Perkins to surrender Fort Shelby before

the regular American garrison arrived.21

While Col. McKay’s capture of Fort

Shelby temporarily restored the British flag

to all of Wisconsin, the Treaty of Ghent in

1815 required all British Canadians to leave

Wisconsin. Captain Andrew Bulger, the last

commandant of Fort McKay, lowered the

Union Jack, burned the fort and embarked

his garrison on the Fox-Wisconsin water¬

way for their voyage back to Canada.22

The Fox Valley 1816-1840

American defeats at Fort Mackinac, Fort

Dearborn, Fort Shelby and the surrender of

American General Hull’s army at Detroit in

the War of 1812 caused the U.S. War De¬

partment to establish and garrison a chain

of frontier forts from the Canadian border

to the Gulf of Mexico. It was hoped the

forts would protect American frontier com¬

munities against future Indian attacks, aid

in the establishment of American control of

Wisconsin and block Canadian fur traders

from using the Fox-Wisconsin and the up¬

per Mississippi Rivers.

An early step in implementing the War

Department’s plan for frontier forts was the

arrival of the American Third Infantry

Regiment at Green Bay on August 7, 1816,

and the subsequent construction of Fort

Howard. When Col. Chambers and Third

Infantrymen were ordered to Prairie du

Chien in the spring of 1817 to rebuild and

garrison a fort there, Fort Howard’s second

commanding officer was Major Zachary

Taylor, the hero of Buena Vista and later

the President of the United States. When

Col. Chambers’ Third Infantrymen passed

up the Fox River in the spring of 1817

they were the first American military force

to travel through the valley.23

Two years later in the summer of 1819,

the Fifth Infantry Regiment under the com¬

mand of Col. Henry Leavenworth traveled

up the Fox, through Lakes Winnebago and

Butte des Morts and up to the portage, in

the process of being redeployed from De¬

troit to the western frontier. The Fifth In¬

fantry garrisoned Forts Armstrong and

Crawford, and built Fort Snelling, the north¬

ern anchor of the frontier fort system, at

St. Paul, Minnesota. Captain Henry Whit¬

ing, an officer in Leavenworth’s command,

wrote a journal describing the Fifth Infan-

1979]

Goff — Three Nations in the Fox River Valley

57

try’s journey and sketched a scene at Neenah

and Big Lake Butte des Morts and two of

the “Grand Kakalin” portion of the Fox

River rapids which are probably the oldest

existing sketches by an American of scenes

in the Fox Valley. Captain Whiting’s report,

including these watercolored sketches

brought him a commendation from the

Secretary of War, John C. Calhoun.24

In the summer of 1820, the Territorial

Governor of Michigan, Lewis Cass, his secre¬

tary, James Duane Doty, later a Wisconsin

Territorial governor. Henry Schoolcraft, fa¬

mous Indian agent-author-anthropologist,

Dr. Alexander Wolcott, expedition physician

and a party of forty men passed down the

valley bound for Green Bay and Detroit on

the last leg of perhaps the most unusual ca¬

noe trip ever taken through this area. The

Cass party were completing a 4,200 mile

“Voyage of Inspection” through Michigan

Territory which then included most of the

present-day states of Minnesota and Wiscon¬

sin, as well as Michigan. The Cass expedi¬

tion rode in three birch bark canoes each of

which was 36 feet long, had a beam of seven

feet, carried a mast and sail and was pro¬

pelled by eight-man crews.25

By August of 1823 James Duane Doty

had become the newly appointed Circuit

Judge of western Michigan and had been

married. Doty and his bride traveled up the

valley on the Fox-Wisconsin to Prairie du

Chien in August of 1823 to spend the win¬

ter and to hold the first session in Wisconsin

of the Michigan Territorial Superior Court

at Prairie du Chien in May, 1824. 26 In the

fall of 1824, Doty convened his court at

Green Bay and there built his first home in

Wisconsin.

When a series of murders of Whites by

Winnebago Indians in the vicinity of Prairie

du Chien and La Crosse threatened to pre¬

cipitate an Indian war in 1826-27, a detach¬

ment of more than 500 men under the com¬

mand of Gen. Atkinson moved from Jeffer¬

son Barracks at St. Louis to Wisconsin by

keel boat. Two hundred soldiers from Col.

Snelling’s garrison at Fort Snelling came

down the Mississippi to Fort Crawford at

Prairie du Chien. An additional force of sol¬

diers, militia and Indians, under the com¬

mand of Major Whistler from Fort Howard,

marched up the Fox Valley to the Fox-Wis¬

consin portage and dug in on the hill east of

the Fox River where Fort Winnebago was

built in 1828.

The rapid concentration of American mil¬

itary power in the heart of the Winnebago

country so stunned the Winnebago chiefs

that they concluded fighting would be use¬

less,, and assigned Chief Red Bird and an

accomplice in the murders to sacrifice them¬

selves to the Americans to preserve the

Winnebago tribe. The Winnebagoes had

helped to demonstrate that interlocking sup¬

port from the frontier forts could not only

safeguard the white man’s frontier but had

made effective Indian warfare in Wisconsin

obsolete.27 Unfortunately, Sauk Chief Black

Hawk triggered a second demonstration of

the futility of frontier forays by Indians

when he precipitated the Black Hawk War

in 1832.

When Sauk Chief Black Hawk led his

“British Band” of nearly 2,000 Indians east

across the Mississippi River in 1832, it was

the final Sauk effort to reoccupy lands in the

Rock River Valley of northwestern Illinois

which the Sauks had owned for several gen¬

erations.

Black Hawk was aware that Americans

had built forts at Rock Island, Prairie du

Chien, St. Paul and Green Bay since the

War of 1812, but probably had no adequate

understanding of the improved effectiveness

of the American army despite the Winne¬

bago tribe’s humiliation by the Americans

in the Red Bird Disturbance of 1828. Black

Hawk’s ignorance or bull-headedness was to

cost his band the loss of nearly ninety per¬

cent of their number in about ninety days.

The Sauk’s stunning losses in 1832 made it

clear to tribes in the upper Mississippi Val-

58

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

ley that the advance of the white man’s

agricultural frontier could no longer be

halted by Indian murders or gunfire. Percep¬

tive tribal chiefs, i.e., Sauk Chief Keokuk

and Menominee Chief Oshkosh, also con¬

cluded that tribesmen had no alternative to

negotiating a final surrender of their tribal

lands to the accelerating flood of land-

hungry Americans.

During the Black Hawk War two com¬

panies of Menominee Indians were enlisted

at Green Bay and commanded by Col. Sam¬

uel Stambaugh, the new Indian agent at

Green Bay. Augustin Grignon of Butte des

Morts was the captain of one of the Me¬

nominee companies; his lieutenants were his

son, Charles, and his nephew, Robert.

George Johnston, former sheriff of Brown

County, was the captain of the second Me¬

nominee company; Johnston’s lieutenants

were William Powell whose farm was in the

present-day town of Algoma, and James

Boyd, son of the former Indian agent at

Green Bay.28

The last traveler up the Fox Valley from

the Black Hawk War was Chief Black Hawk

himself. After captivity as a prisoner of

war at Jefferson Barracks, Missouri and Fort

Monroe, Virginia, Black Hawk was escorted

back to Fort Armstrong and released to his

detested tribal rival, Sauk Chief Keokuk.

When Black Hawk traveled up the Fox Val¬

ley and west along the Wisconsin to the

Mississippi, the Black Hawk War was

finally over.29

This chronicle of early Fox Valley events

triggered by governmental action should

include the survey and construction of the

military road from Fort Howard to Prairie

du Chien. In 1829, frontier promoters

James Duane Doty of Menasha and Morgan

L. Martin of Green Bay urged General

Macomb in Washington, D.C. to use troops

to open a road from the head of Lake Win¬

nebago to Green Bay. Since it was politi¬

cally more justifiable to build a road linking

the frontier forts, in 1832 a party led by

Lt. Alexander Center completed surveys and

construction plans for a 234 mile military

road from Fort Howard at Green Bay to

Fort Winnebago at Portage and to Fort

Crawford at Prairie du Chien (Fig. 2). Ac¬

tual road construction began in the spring of

1835.

The section of the military road between

Fort Crawford and Fort Winnebago was

built by three companies of soldiers from

Fort Crawford under the direction of Col.

Zachary Taylor. Three companies from Fort

Winnebago built the middle section of the

road from Portage to Fond du Lac and the

third section from Fond du Lac to Green

Bay was built by three companies from Fort

Howard.

The Fort Howard-Fort Crawford military

road was a thirty foot wide lane through

the timber. It was unpaved and in wet

weather was virtually impassable for wagons.

Grading was slight and so many high

stumps were left standing that carriages

and sleighs found travel difficult. It was not

until the end of 1837 that the road was

even nominally complete. Nevertheless, for

many years the military road was the only

overland route between Green Bay, Fond

du Lac, Portage, Madison, the lead region

and Prairie du Chien. Military troops and

supplies passed over it frequently, and a

growing stream of settlers increased the flow

of traffic.30

In August, 1836, Henry Dodge, the newly

appointed Territorial Governor of Wiscon¬

sin, crossed the Fox River at Algoma in

present-day Oshkosh, paying a toll on a

ferry belonging to James Knaggs. Governor

Dodge was on his way down the Fox to ne¬

gotiate the Treaty of Cedar Point by which

the U.S. Government purchased 4,100,000

acres from the Menominee Indians on Sep¬

tember 3, 1836 for $700,000. By this treaty

the Menominees gave up title to all land in

Wisconsin north of the Fox River and east

of the Wolf.31

In the summer of 1837, troops com¬

manded by General Henry Atkinson jour¬

neyed up the Fox Valley by the new mili-

1979]

Goff-— Three Nations in the Fox River Valley

59

2nd Lt. Alexander J. Center commanded an Army

survey party in 1832 which mapped the Fox

River in the process of looking for an appro-

''' 3 priate location for a military wagon road

|I between Forts Howard, Winnebago and Crawford.

'n The Military Road was built in 1835-37, and

\ ran along the east shore of Lake Winnebago.

h Source: State Historical Society of Wis .

Fig. 2. Lieutenant Center’s map of Lake Winnebago in 1833.

tary road from Fort Howard to Fort Win¬

nebago and thence by keel boat to Fort

Snelling. In 1839 a survey party of the U.S.

Bureau of Topographical Engineers led by

Captain Thomas J. Cram surveyed the po¬

tential of the Fox-Wisconsin rivers for

steamboat navigation, concluding that the

route could be made navigable (Fig. 3). 32

A different kind of survey of the land area

of the Fox Valley was conducted by con¬

tract surveyors working under the direction

of the U.S. Surveyor General’s office. Since

enactment of the Northwest Ordinance of

1787, a rectangular system of mapping had

been developed in which the government

surveyors established boundary lines of In¬

dian treaty lands, exterior township lines

and then established subdivision lines in¬

cluding section, town, range and lot lines.

Mapping of the Fox Valley south of the

Fox River and Lake Butte des Morts was

completed by 1834 and the map published

by the Surveyor General’s office in Cincin¬

nati, October 23, 1835. The field work of

mapping valley areas north of the Fox River

and east of the Wolf seems to have still been

in progress in January, 1839, but the

completed map was published by the U.S.

Surveyor’s office at Dubuque, September 28,

1839. Once the maps were published, the

60

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

Fig. 3. U. S. Topographical Engineers Captain T. G. Cram led a party in 1839 which sur¬

veyed the potential for steamboat navigation of a canal from Lake Winnebago to Wrights-

town. While this canal was never built, the cartographic skill of the engineers clearly had

reached modern professional standards.

land offices could begin to throw the land

open for public sale.33

County government first appeared in the

Fox valley wilderness in 1818 when Gover¬

nor Cass of Michigan Territory proclaimed

a division into three counties of the terri¬

tory’s lands lying west of Lake Michigan

and north to the Straits of Mackinac. Two

of the counties, Brown and Crawford, com¬

prehended the Fox-Wisconsin waterway.

The county seat for Brown county was

Green Bay and for Crawford County Prairie

du Chien.

County affairs in Michigan Territory in

1818 and hence in the Fox River valley,

were in the hands of a board of three county

commissioners. Originally appointed by the

territorial governor, by 1825 the county

commissioners had become elective. The

boards of county commissioners in Michi¬

gan Territory were replaced in 1827 by

boards of county supervisors, one super¬

visor to be elected from each town.34

Direct participation by Fox valley citi¬

zens in the government of Michigan Ter¬

ritory began at Green Bay in 1821 when

42 votes were cast for Michigan territorial

delegate to the U.S. Congress. More to the

point, Fox valley and other Brown County

voters in 1823 elected Robert Irwin of Green

Bay representative to the Territorial Legis¬

lative Council from west of Lake Michigan.

The Fox valley was still a wilderness in

1823 but the tendency of significant histori¬

cal events in the valley to be initiated by

the agents of distant governments had now

been largely reversed by the significant fact

that the citizenry of the valley could express

their needs to county, territorial and Con¬

gressional representatives or to judicial of¬

ficers.

The Michigan Territorial Legislature

1979]

Goff — Three Nations in the Fox River Valley

61

created Iowa County in 1830 and Milwau¬

kee County in 1835 from the southern por¬

tions of Crawford and Brown counties, re¬

spectively. Thus when Wisconsin Territory

began to function on July 3, 1836 there

were four counties within the present day

boundaries of Wisconsin, i.e. Brown, Mil¬

waukee, Crawford and Iowa. The total

white population of these counties in 1836

according to the territorial census was

11,683. Once Wisconsin legally became a

territory in 1836 the Wisconsin Territorial

Legislature created fifteen additional coun¬

ties; those in or near the Fox valley were

Calumet, Fond du Lac, Manitowoc, Sheboy¬

gan, Marquette and Portage. Three addi¬

tional counties were created in 1840 includ¬

ing Winnebago in the Fox valley.35

The closing of the American Fur Com¬

pany at Green Bay in 1844 was an histori¬

cally symbolic requiem for Indians and fur

trading in the Fox valley. Governmentally,

the arrival of a new era was symbolized by

the creation in the valley of the new coun¬

ties of Fond du Lac, Winnebago and Calu¬

met. Federal troops left the valley with the

deactivation of Fort Howard in 1841. The

arrival in the valley of the American agri¬

cultural frontier was marked by the first

sizeable influx of settlers at Oshkosh in the

summer of 1846, and by the platting of the

villages of Neenah in 1847 and Menasha in

1849. Wisconsin Territory became the State

of Wisconsin in 1848. Clearly, an era of

more than two centuries of exploration, fur

trading and the initiation of most major

events in the Fox valley by governmental

agents came to an end shortly after 1840.

Notes

4Thwaites, Reuben Gold, ed., The Jesuit Re¬

lations and Allied Documents . . . Jesuit Mission¬

aries in New France 1610-1791 (Cleveland, 1896-

1901) 23: 1642-43. Jean Boisseau’s map faces

page 234.

2 Charlevoix, Pierre F. X., History of New

France, III (Paris, 1744), 178-179 (translation of

John Gilmary Shea).

Severin, Timothy, Explorers of the Mississippi

(New York, A. A. Knopf, 1968), 83-92.

3 Charlevoix, op. cit., V, Book 20, 257, 305-307.

S.H.S.W., Collections, 5:77-85.

Smith, Alice E., The History of Wisconsin:

From Exploration to Statehood, I (Madison, Wis.,

State Historical Society of Wisconsin, 1973), 41.

4 Wittry, Warren L., “The Bell Site: An Early

Fox Village,” Wisconsin Archaeologist, 44(1):

45-46.

5 Charlevoix, op. cit., 5:305.

G Hebberd, S. S., History of Wisconsin Under

the Dominion of France (Madison, Wis., Mid¬

land Publishing Co., 1890), 121.

7 Kellogg, Louise P., The French Regime in

Wisconsin and the Northwest (Madison, Wis.,

State Historical Society of Wisconsin, 1925), 320-

321; S.H.S.W., Collections, X, 47-53. The Chaus-

segros de Lery map in Kellogg, op. cit., 314 seems

to be the earliest existing map which shows sub¬

stantial detail of the general vicinity of present-

day Oshkosh, Lake Winnebago (Lac des Puants),

Big Lake Butte des Morts (Petit Lac des Rey¬

nards) and the lower Fox River (Riviere des

Reynards). The Fox fort east of the largest Fox

village had been in existence, according to de

Lery’s map, since at least 1723 (Fort ou les

Reynards sentient fortifie in 1723).

8 Alvord, Clarence W. (ed.), Centennial His¬

tory of Illinois: The Illinois Country 1673-1818

(Springfield, Ill.: Illinois Centennial Commission,

1920), 163; Kellogg, op. cit., 324.

Kellogg, L. P., “Fox Indians During the French

Regime,” S.H.S.W., Proceedings, 1907, 142-188.

9 Kellogg, L. P., “Fox Indians . . .” op. cit., 327;

Alvord, op. cit., 163-164.

30 Kellogg, op. cit., 329-330.

11 Kellogg, French Regime, op. cit., 329-331.

S.H.S.W., Collections, 17:210.

12 Letter from Governor Beauharnois and Hoc-

quart to the French Minister, Nov. 11, 1733 and

Oct. 5, 1734, in S.H.S.W., Collections, 17:189-

191, and 202-203.

13 Kellogg, op. cit., 331-332.

S.H.S.W., Collections, 17:188-191, 200-204; 3:

200; 8:207-208.

14 Letter from Governor Beauharnois ... to

French Minister, op. cit., 17:188-191.

15 Kellogg, op. cit., 334-335.

S.H.S.W., Collections, 17:221-229 and 216-221.

16 Kellogg, op. cit., 432, 436.

17 “Narrative of Peter Pond,” in Gates, C. M.,

Five Fur Traders of the Northwest (Minnesota

Historical Society, 1965), 30-39.

S.H.S.W., Collections, 18:314-354.

18 S.H.S.W., Collections, 2:165-170,174.

Kellogg, op. cit., 191-192.

62

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

19Thwaites, Reuben Gold, Wisconsin in Three

Centuries, II, 101, 140.

Kellogg, op. cit., 282.

20 Kellogg, op. cit., 284.

21 Kellogg, op. cit., 315-319.

22 Kellogg, op. cit., 325.

23 Prucha, Father Francis Paul, Broadax and

Bayonet (Madison, Wis., S.H.S.W., 1953), 18-21;

54.

24 Captain Whiting’s Journal, zmpzzblished mss.

in National Archives Microfilm Reel #1, No.

130, Rep. Book Pa397, in Fort Howard Consol.

File 1819-1873.

23 Schoolcraft, Henry R., Summary Narrative of

an Exploratory Expedition to the Sources of the

Mississippi River in 1820 . . . (Philadelphia: Lip-

pincott, Grambo & Co., 1855), 190.

26 Smith, op. cit., 168.

27 Prucha, op. cit., 23-24. For a superb descrip¬

tion of Red Bird at the moment of his surrender,

see S.H.S.W., Collections, 5:178-204 in Col.

Thomas McKenney, “The Winnebago War.”

28 Strong, Moses, “The Indian Wars of Wiscon¬

sin,” S.H.S.W., Collections, 8:276.

29 Hagen, William T., Sac and Fox Indians

(Norman, Okla., University of Oklahoma Press,

1958), 200-201.

Smith, op. cit., 133-150.

Nesbit, R. E., Wisconsin: A History (Madison,

Wis., Univ. of Wisconsin Press, 1973), 95-100.

30 Prucha, op. cit., 135-144.

31 Harney, Richard J., History of Winnebago

County (Oshkosh: Allen and Hicks, 1880), 101.

32 Source of Capt. Cram’s map: U.S. Senate,

26th Congress, 1st Session (1840), Document

318, 16.

33 Smith, op. cit., 457.

34 Wisconsin Historical Records Survey, County

Government in Wisconsin, Vol. 1. 1942, 3-4.

35 State of Wisconsin, Wisconsin Blue Book,

1977, (Madison, Wisconsin, 1977) 727.

PERCEPTION OF THE PRAIRIE IN A

LETTER FROM PRAIRIEVILLE

Charles F. Calkins

Carroll College, Waukesha, Wisconsin

The arrival of a few Scots in 1840

marked the beginning of a sizable rural

Scotch settlement that would develop in

what was then Milwaukee County, but to¬

day Waukesha County, with Vernon Town¬

ship as its center.1 Within a short period of

time, additional pioneers joined the van¬

guard to form a relatively dense settlement

of Scots. Whereas a substantial number

came directly from Scotland to the Wiscon¬

sin Territory, most of the initial settlers

came from the general area of Livingston

County, New York, where they or their

families had settled earlier in the nineteenth

century.

Included among the earliest arrivals in

the Vernon area was the Robert Weir

family,2 which had reached Milwaukee by

boat on October 26, 1841, and journeyed

westward to Prairieville3 the following day.

Like so many others who eventually set¬

tled in Vernon, Weir had emigrated to New

York from Scotland, where he was born in

1809. 4 Weir may have been attracted to

the Prairieville area by a letter much like

the one he sent back to New York encour¬

aging friends to move west and join the

growing settlement.

Weir’s letter was addressed to William

Fraser, postmaster of Fowlerville, New

York, a small village in Livingston County.5

The contents indicate, however, that it was

an open letter intended for old friends; be¬

cause of his position as postmaster, Fraser

was being asked to pass on the news. Weir,

in addressing the letter to Fraser, might

have been epitomizing the popular stereo¬

type of the frugal Scotsman. It has been

suggested that “In order to save the 25 cent

postage, Weir addressed the postmaster,

. . . postmasters then enjoying the frank.”6

On the other hand, sending the letter to

Fraser was necessitated, perhaps, by Weir’s

own strained financial resources resulting

from a costly trip and subsequent invest¬

ment in land. Although one cannot be sure

that Weir was describing his personal finan¬

cial situation, he wrote in the letter that “An

empty purse and a new country maketh a

hard beginning.”

On the surface, Weir’s letter might be

considered typical of its genre. Many of the

topics common in pioneer letters — the trip,

family health, finances, general economic

conditions and opportunities, prospects for

agriculture, and perception of the new en¬

vironment — are all covered in some depth.

It is this last item, Weir’s perception of the

prairie, oak opening, and forest biotic com¬

munities, that is of particular interest and

importance in this letter. The attitudes ex¬

pressed and preserved in the personal pa¬

pers of people like Weir have provided for

a continuing re-evaluation of the impor¬

tance of the prairie in the process of west¬

ward settlement.7 Weir’s pre-arrival percep¬

tion of a section of the Milwaukee County

prairie was somewhat at odds with his on-

the-scene perception. Whereas Weir left

New York intending to settle on a prairie

farm, upon arriving he chose land8 that,

according to his own description, most

likely would be classified as an oak open¬

ing. This change did not detract from Weir’s

enthusiasm for his new home. Weir’s opti¬

mism remained high, although it was tem¬

pered by the reality of the frontier situation

in which he found himself. His plan for

success was altered somewhat (as is re¬

vealed in the letter) to fit his changing per¬

ception of the new surroundings, which

were quite unlike those of his native Scot-

63

64

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

land or those of New York, his intermedi¬

ate residence.

Weir’s initial, positive perception of the

Milwaukee County prairie he expected to

encounter upon arrival was contrary to the

general attitudes many historians have at¬

tributed to the so-called eastern woodland

mentality. In his discussion of the settle¬

ment of the Lake Plains (which encompass

a major portion of Wisconsin), Ray Allen

Billington, for example, commented on the

impressive beauty of the prairie, but he was

quick to suggest that “Yet the prairie coun¬

try was shunned by the first settlers, whose

frontier technique was adjusted to a wooded

country.”10 Moreover, he accounted for the

early settlement of the wooded sections of

southern Indiana and Illinois specifically

by suggesting that “ adequate soil [italics

mine] and familiar vegetation” were avail¬

able.11 In a far more balanced treatment of

the prairie-woodland controversy in South¬

eastern Wisconsin, Schafer felt obliged to

point out that “The answer [to the question

of whether to take a prairie or woodland

farm] turned sometimes on men’s belief

in the superiority of forested land as

land. . . .”12 That Weir intended to take a

prairie farm while he was still in New York

indicates that he did not fit the supposed

popular mold. Whatever his sources of in¬

formation were about the physical funda¬

ment of a portion of old Milwaukee

County, they did not allow him to conclude

that prairie soil was “inadequate” or in¬

herently less fertile than that of the East.

A change in Weir’s on-the-scene percep¬

tion is evident when he wrote that “. . . I

do not think they [prairie farms] are as

good for wheat as the [oak] openings or

timber land. . . .” However, it does not

seem reasonable to conclude that the change

was prompted by any empirical evidence

indicating the prairie soils were inferior;

quite the opposite appears to have been the

case. Later in the same paragraph where

he compared the Genesee country with his

new home, Weir was very explicit: “. . . the

soil is better [in the Prairie ville area], there

is no mistake about it. . . .” Earlier in the

letter Weir expressed satisfaction with his

choice of land when he stated, “I think that

I have got a good wheat farm.”

One question remains: What motivated

the change in perception then, if it was not

the real or imagined low fertility of prairie

soils? Throughout the letter, Weir’s pre¬

occupation with the forest resource was

evident. He wrote of his desire to acquire

timber land to complement this prairie

farm; he bemoans the fact that “There is

but little timber on it [the farm], say enough

to fence it into 20 acre lots.” Another ref¬

erence to “trees [being] thinly scattered

over” the farm suggests a dissatisfaction with

the timber supply and is not a reflection

of poor soil quality. This writer contends

that Weir’s perception is summarized well

when he writes that “. . . it is hard to get

a prairie farm and timber enough for the

same.” In other words, he valued the

prairie soil but viewed the paucity of timber

as a definite drawback.

Weir’s actions support the contention. In

the spring of 1842, he sold his land in sec¬

tion 12 and purchased 160 acres in section

1, about one mile north.13 Apparently his

desire to obtain some “first rate timber

land” was realized in this transaction.14 It

is quite possible that included within his

purchase was some of the desirable timber

land to which he made reference in the

letter.

It seems clear that Weir’s perception and

subsequent actions were not unique among

settlers on those portions of the American

frontier where the prairie was encountered.

Mounting evidence suggests that early set¬

tlers did not shun the prairie because of as¬

sumed greater fertility of woodland soil.

Studies centered in Iowa,15 Illinois,16

Texas,17 and Michigan18 have revealed, as

John Fraser Hart has suggested, “that the

pioneers, wherever possible, preferred to

settle along the prairie — -woodland edge,

so that they might enjoy the advantages of

1979]

Calkins — Perception of the Prairie

65

both types of country.”19 The fertility of the

prairie soil combined with the wood from

forested areas used to build houses, barns,

fences, and used to make fires for warmth

and cooking offered an attractive induce¬

ment for settlement among pioneers like

Weir.

In his letter that follows, Weir wrote

forcefully in a style that ranges from folksy

to simple eloquence. For clarity, punctua¬

tion and capitalization have been provided

where they were absent in the original. The

spelling and grammar have not been

changed from the original, because they do

not detract from a complete understanding

of Weir’s thoughts.

Mr. Wm. Fraser

Dear Sir:

According to promise I set down to give

you a few lines. I hope you will excuse me

in being so long in writing, as I sent word

to you by Mr. Dn. [Duncan] Cameron20 [of]

Caledonia [New York] that we had got here

safely. And I am happy to inform you that

we are all in good health at present. And it

may aford you comfort to know that we

are in a healthy country blessed with a

good appetite and plenty to satisfy the same.

It may be that you have heard that we had

a rough passage on the Las [lakes]. We

were nine days on the water and part of

the time it was very stormy. When when

[sic] we came into Lake Huron we had a

high head wind and a heavy sea, so much

so that we had to turn round and go with

the wind for sometime, and when we came

into Lake Michigan we were again able to

put round again and take shelter at an is¬

land and remain for one day. Yet we en¬

joyed ourselves as well as could be ex¬

pected. We were on a first rate Boat, had

a good Captain and a respectable set of

hands, likewise a good Bar, and we were

some[what] like Tom O’Shanter — we did

not mind the storm a whistle. I did not get

sea sick at all. My family did a little, yet

stood it better than might be expected. We

landed at Milwaukie on the 26th [of] Oct.

and came to Prairieville [Waukesha] the

27th. James Wallace, James Begg [Jr.] and

their families arrived at Prairieville on the

26th. James Begg [Sr.] and his [family] who

came by land arrived on the 28th. Strange

to think that we all started from home at

different times to come such a distance,

some by water and some by land and to ar¬

rive here in such rotation on the 26th, 27th

and 28th. Prairieville is 18 miles west from

Milwaukie and I think in a short time will

be a great place for business. There is now

a first rate flowering mill, one saw mill, one

blacksmith’s shop and plenty of business

for three hands (James Wallace is em¬

ployed in the same). There is a good wag-

gonmaker’s shop built last faull. There is

two copper shops, three stores and one

tavern, one drug store, one saddler, a num¬

ber of shoemakers and tailors well em¬

ployed there, one meeting house finished

and two not finished, one stone Academy

not finished inside, preachers without num¬

ber, such as they are, and two or three

Doctors but little employment for the same.

Milwaukie will be a great place. She is in

much want of a Harbor and has likewise

been wanton for water power, but it is

thought money will be granted this winter

for a harbor and there is a damm across

the Milwaukie river nearly finished, which

is said will give her water power sufficient

for one hundred run of stores. If a good

harbor is got and good flowering mills built,

the market here will be much better. When

these things are done Milwaukie will be to

this country as Rochester is to western N.

[New] York. Wheat has been carried from

here to Buffalo for 15 cents per bus.

[bushel] but 18 pence [or] 2/ is the general

price.21 Wheat is worth at present from 70

to 75 cents, corn 2/9, oats 1/6, pork 2

dollars and 50 cents, beef about the same.

Goods are but little higher here than they

are in Fowlerville and plenty of everything

to be got. Money is rather scarce, but it is

known to be so in every new country. I

66

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

think this will be fine wheat growing coun¬

try although there is but little raised as yet

for export. Farmers that have been here 4

or 5 years appear to have but little done

by way of improvements. They say it is

hard beginning. The fact is many come here

poor and have nothing to begin with. An

empty purse and a new country maketh a

hard beginning. This country is settling very

fast. I was much disappointed in finding the

land so much taken up as it is. I looked

round thinking to find land at government

prices, [$1.25 per acre] but I could not find

it good and in a convenient place, although

I did not go more than 35 miles back from

the Lake [Michigan]. The people say that

more land has been taken since last spring

than was taken in two years previous.

You may tell Neil McDugal22 that I

have bought a farm in his neighborhood.

It is between his son-in-law23 and the Mil-

waukie road.24 My south line and his north

line is the same on the west side of the cross

road, or as it is called, the Rochester

road,25 and James Begg has bought on the

west side of the said Rochester road. His

south line and McDugal’s north line is the

same. James Wallace has bought John Mc¬

Intyre’s farm in the same neighborhood. I

am nearly two miles west from him.26 Wil¬

liam Begg is thinking of buying Mr.

Plumm’s27 farm in said neighborhood. It is

going to be quite a Scotch settlement, and

tell the Deacon28 if he doth not come on to

his land we will either sell it or burn it up,

and tell him that I think there is Deacons

enough where he is and we want him to be

the old Deacon amongst us here, and tell

him further that the best wheat that I have

seen in the Ter. [Territory] was raised in

that neighborhood. I think that I have got a

good wheat farm. I bought 160 acres, and

I intend to buy 40 acres more which joins

said 160, which will make 200 acres for

cultivation, and there is 80 acres of first rate

timber land IV2 mile to the north which I

was intending to take at gov’t, price to keep

as a timber lot, but it is thrown out of mar¬

ket at present, and I do not know if I can

get it. If not, there is 60 acres equally as

good timber and about the same distance

from me which I can get for 3 Dollars per

acre. My best respects to James Hamilton

and not forgetting his Lady. Tell him that

I think I have got as good a wheat farm as

there is in Sugarburry [New York].29 There

is but little timber on it, say enough to fence

it into 20 acre lots. Large trees [are] scat¬

tered thinly over it consisting of white oak,

black oak, black walnut and some maple,

and part of it is thickly grown over with

hazel brush. There is a small log house on

it and 8 acres broken up [and] 10 acres

fenced. It will require three yoke of oxen

to plough it the first time. Stone enough

but not too many apparently.

When I left [New] York state I was fully

in the belief that I would take a prairie

farm. They are pleasant to the eye, but I

do not think they are as good for wheat as

the [oak] openings or the timber land, and

it is hard to get a prairie farm and timber

enough for the same. Water is likewise hard

to be got on many of them. I think the land

in this country is good in general, but I do

not think the face of the country so hand¬

some as the Genesee [New York] country

is. It is more rolling or broken, but the soil

is better, there is no mistake about it, and

the healthiness of this country will speak

loudly her praise. I have been in through

Milwaukie, Racine and Walworth counties,

and I have not seen man, woman or child

sick.

Give my respects to all enquiring friends

and acqus. [acquaintances] : Edward Craig

the first time you see him, Js. Hamilton, Jn.

Hamiltons, Rt. Wallance, An. Frrasers, As.

McBean, Ts. Howie, all the old women

about Fowlerville and espec. in Sugarburry

and write to me as soon as this comes to

hand.

If any letters come from Scotland send

them to me. Direct [them] to me, Rt. Weir,

1979]

Calkins — Perception of the Prairie

67

Prairieville Post Office, Milwaukie County,

Wisconsin Territory.

Robt. Weir

Dec. 27, 1841

Notations

1 A brief history of the settlement is the focus

of Edward C. Wicklein, The Scots of Vernon and

Adjacent Townships, Waukesha County, Wiscon¬

sin (Waukesha, 1974). Wicklein reprints a copy

of the letter (pp. 28-29) that is taken from a

transcribed copy of the original, which contains

errors and omissions. For a description of the

contents and availability of this work see “Wis¬

consin History Checklist,” Wisconsin Magazine

of History, 57:326 (Summer, 1974).

2 At the time of arrival, the family consisted of

Robert and his wife Mary, both 30 years of age,

and young daughters Margaret and Mary Jane,

5 and 2 years of age, respectively.

3 The present City of Waukesha was formerly

known as Prairieville. When Waukesha County

was separated from Milwaukee County in 1846 —

to create a separate political unit — the name

change took place. Prior to being called Prairie¬

ville, the little settlement was known as Prairie

Village for a short period of time.

4 It is difficult to determine the date of Weir’s

arrival in the United States. A Declaration of

Intention to Become a Citizen of the United

States in the name of Robert Weir is not on file

in either Livingston County, New York (personal

communication from Miss Anna Patchett, County

Historian, Livingston County, May 29, 1975) nor

Waukesha County, Wisconsin. It is certain that

Weir was in New York by 1832 as his name is

included on a list of members of the United

Presbyterian Church of York for that date (Mary

R. Root, History of the Town of York, Living¬

ston County, New York (Caledonia, 1940), 99).

6 The original Weir letter is in an unclassified

collection of the Waukesha County Historical

Museum, Waukesha, Wisconsin.

6 John G. Gregory, ed., Southeastern Wiscon¬

sin: A History of Old Milwaukee County, Vol. II

(Chicago, 1932), 965. Excerpts of the Weir letter

are provided on pages 965 and 966. However, er¬

rors are apparent when the excerpts are com¬

pared with the original letter.

7 An effective use is made of letters and other

personal papers in this regard by Douglas R. Mc-

Manis, The Initial Evaluation and Utilization of

the Illinois Prairies, 1815-1840 (Chicago, 1964).

8 A deed dated December 11, 1841 records

Weir’s purchase of the NW1/4 of Sec. 12, T. 5

N., R. 19 E., from John C. Snover for a sum of

$450, or just over $2.80 per acre, which was more

than twice the price of government land.

9 A map entitled Presettlement Vegetation of

Waukesha County, 1836, prepared by Marlin

Johnson and Jerry Schwarzmeir (1973) from the

Public Land Survey Notes and Plats shows this

to be the case. In addition, a portion of the land

that became Weir’s farm was reported to have

been burned over shortly before the surveyor

reached the area and appears on Johnson and

Schwarzmeir’s map as “burned.”

10 Ray Allen Billington, Westward Expansion

(New York, 1949), 294.

11 Ibid.

12 Joseph Schafer, Four Wisconsin Counties:

Prairie and Forest (Madison, 1927), 107. In an

explanatory footnote for this point, Schafer adds:

“Many Yankees had this belief until the virtues

of the prairie lands had been thoroughly tested.”

13 Weir sold his farm for $650.00 and purchased

the SW1/4, Sec. 1, T. 5 N., R. 13 E. from Caleb

Nanscawen for the sum of $775.00.

14 In addition to wanting timber for use on the

farm, it seems reasonable to suggest that Weir’s

desire for acquiring good timberland was moti¬

vated by his intent to engage in the sawmill busi¬

ness in the very near future. “Mr. Robert Weir

. . . built a saw-mill at Big Bend, which ran by

steam. At this mill, most of the plank used in

constructing the Muk[wonago] and Milwaukee]

road was sawed” ( The History of Waukesha

County, Wisconsin (Chicago, 1880), 791). The

mill was built between December 27, 1841, the

date of Weir’s letter, and 1848, the date of in¬

corporation of the road {Ibid., 387).

15 Leslie Hewes, “Some Features of Early

Woodland and Prairie Settlement in a Central

Iowa County,” Annals, Association of American

Geographers, 40:40-57 (March, 1950).

16 McManis, The Initial Evaluation and Utiliza¬

tion of The Illinois Prairies, 1815-1840.

17 Terry G. Jordan, “Pioneer Evaluation of

Vegetation in Frontier Texas,” Southwestern His¬

torical Quarterly, 76:233-54 (January, 1973).

18 Bernard C. Peters, “Pioneer Evaluation of the

Kalamazoo County Landscape,” Michigan Acade¬

mician, 3:15-25 (Fall, 1970).

19 John Fraser Hart, The Look of the Land

(Englewood Cliffs, 1975), 7-8.

20 Like Weir, Cameron reached Vernon Town¬

ship in October, 1841, in the company of three

other bachelors. Apparently Cameron returned

to New York shortly thereafter, as did Donald

68

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

Stewert ( Waukesha Freeman, September 3,

1910), a fellow traveller, to report in person to

relatives on the opportunities in the Wisconsin

Territory. In subsequent years a number of Cam¬

eron’s relatives settled in Vernon Township.

21 The following monetary equivalents apply

to the prices quoted:

One pence = 1 cent or 10/125 shillings

One shilling = 12 Vi cents

22 The reference here is obviously to Neil Mc-

Dougal, who has been credited with being the

founder of the Scotch settlement in Vernon (see

the Waukesha Freeman, September 3, 1910 and

September 16, 1943). McDougal was a wealthy

New York farmer who came west to invest in

real estate. In 1840 he purchased a number of

tracts of land, and then returned to New York

where he was at the time Weir’s letter arrived

with the message for him.

23 McDougal’s daughter Ellen was married to

Findley Fraser, who owned the El/2, SW1/4, Sec.

12, which, in part, was adjacent on the south to

Weir’s property.

24 Reference here is to the Milwaukee-Janesville

(via Mukwonago) road, which passed through

section 1 about a half mile north of Weir’s prop¬

erty. Presently this route is designated CTH ES

(formerly STH 15) or National Avenue. For a

map of territorial roads in Waukesha County,

see: Southeastern Wisconsin Regional Planning

Commission, A Jurisdictional Highway System

Plan for Waukesha County, Planning Report

No. 18, (Waukesha, 1975), 19.

25 The road mentioned here, with a north-

south trend, passed through the middle of section

12 and in section 13 angled to the west and

south where it entered Big Bend, and then it

continued in a southerly direction, closely paral¬

leling present-day CTH F, to Rochester in Racine

County.

26 The record does not indicate such a purchase.

A Hugh (not John) McIntyre owned the Sl/2,

NW1/4 of Sec. 5, T. 5 N., R. 20 E., the Town¬

ship of New Berlin, which is immediately east of

Vernon. Weir’s property would have been “nearly

two miles west from” the McIntyre land.

27 Apparently Weir is referring to Joseph A.

Plumb here. There are no Plumms listed as be¬

ing land owners in the vicinity at this time.

28 The reference to “the Deacon” in the broader

context of the idea being developed coupled

with the fact that Neil McDougal was an elder

in the United Presbyterian Church of York, New

York, suggests that Weir is still referring to Mc¬

Dougal here, as he was at the outset of this para¬

graph.

29 The common spelling is Sugarbury. This small

village was located in the Town of York across

the Genesee River from Fowlerville, which was

in the Town of Avon.

THE PHYCOPERIPHYTON COMMUNITY OF THE

LOWER BLACK RIVER, WISCONSIN

Michael R. Strenski

University of W isconsin-La Crosse

A bstract

Phycoperiphyton taxonomic composition was determined from materials

gathered from the Black River, La Crosse County, Wisconsin. Phycoperiphyton

samples were collected from two stations at each of four sites during the open

water season, April through November 1976.

Fifty-six genera and 205 species were found. The phycoperiphyton com¬

munity was dominated by members of the class Bacillariophyceae (ca. 98% of

the total community). Achnanthes lanceolata, was the dominant species during

spring months. Melosira varians dominated by mid-June and was succeeded by

Cocconeis placentula euglypta during July through late September. Navicula cryp -

tocephala followed in the autumn, and during the final sampling period in Novem¬

ber Diatoma vulgare was the dominant species.

Introduction

The Black River is a tributary of the

Mississippi River with the confluence lo¬

cated 11.5 km (7 mi) north of La Crosse,

Wisconsin. From its headwaters, located ap¬

proximately 267 km (160 mi) NNE in Tay¬

lor County, to the Mississippi drains 5490

km2 (2129 m2) . This investigation, per¬

formed during April through November

1976, was undertaken to determine the

taxonomic composition of the phycoperiphy¬

ton community and to establish the density

and relative density of the several species.

Description of the Study Area

The southern portion of the Black River,

located between, lat. 43° 57" to 44° 03"

and long. 91° 12" to 91° 16", forms the

northern boundary of sections 1 and 2 and

the western boundary of sections 10, 16, 21,

and 27 of Holland Township, La Crosse

County, T. 18 N., R. 8 W. (Fig. 1). United

States Geological Station 05382000 is lo¬

cated about 50 m upstream from the first set

of sampling stations used in this study. The

study area covered 11 km (6.6 mi) of river

which in the study area meanders through

an area known as the Black River Bottoms.

Agricultural crops and a few pastures are

found along the Trempealeau County side of

the river. Swimming, canoeing, and fishing

are popular recreational activities on the

Black River.

Two stations were established at each of

four sites. At each site, one station was

placed where the current velocity was great¬

est, and a second station where current ve¬

locity was least. Direct sunlight was also

considered in station location because it

greatly affects phycoperiphyton distribution

(Table 1). United States Highway 53 spans

the Black River at a crossing known as

Hunter’s Bridge, 3.5 km (2 mi) southeast

of Galesville, Wisconsin. This bridge serves

as a reference point for site locations (Fig.

1 ) . Stations have the designation of N or S

which indicate that the locations were nearer

to the northern (N) or southern (S) bank

of the river.

Materials and Methods

Phycoperiphyton samplers were con¬

structed from plastic slide boxes (Strenski,

1977). Two samplers, each containing eight

evenly spaced glass slides, were set at each

69

70

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

Fig. 1. Black River between Trempealeau and

LaCrosse counties. Locations of paired sampling

stations are indicated as 1-N and 1-S, etc.

station. Slides were inserted vertically to

avoid sedimentation. Duration of slide ex¬

posure ranged from 14 days during the

spring and summer, to 18-21 days during

the autumn.

Material for identification was scraped

from the glass slides into a solution of “M3”

preservative (APHA 1975). Diatoms were

cleared and permanently mounted in Hyrax®

prior to analysis (APHA 1975). The phy-

coperiphyton was identified to the lowest

taxonomic level possible using bright-held

microscopy. Taxonomic keys used included

those of; Bourrelly (1968), Cleve-Euler

(1955), Hansmann (1973), Hohn and Hel-

lerman (1963), Huber-Pestalozzi (1942),

Hustedt (1930a, 1930b), Patrick and Rei-

mer (1966), Prescott (1962), Smith

(1950), Tiffany and Britton (1952), and

Weber (1971).

Table 1. Direct sunlight estimates, mean and

range values for current velocity (m/sec), and

depth (m) at each station in the Black River, 10

April through 20 November 1976.

a Direct Sunlight = 1 — Direct sunlight at least un¬

til early evening.

2 ■ — 6 to 8 hr of direct sunlight

daily.

3 — Less than 3 hr of direct

sunlight daily.

A standard Palmer-Maloney cell (Palmer

and Maloney 1954) was used to determine

density of all living cells (diatoms and non¬

diatoms) and for the identification of non¬

diatoms. At least 500 organisms per count¬

ing cell were tallied. The relative densities

of each taxa within the diatom community

were estimated by the species proportional

count method (APHA 1975). A minimum

of 500 cells were counted from each Hyrax®

mount.

Results and Discussion

The phycoperiphyton in the Black River

represented four classes of algae: Bacil-

lariophyceae (diatoms), Chlorophyceae

(green algae), Dinophyceae (dinoflagel-

lates), and Myxophyceae (blue-green al¬

gae). Fifty-six genera and 205 species were

identified; twenty-nine of the 205 species

1979]

Strenski — Phy coperiphyton Community of the Lower Black River

71

were further identified to include 75 varie¬

ties.

Members of the class Bacillariophyceae

were the major dominants. This class was

represented by 192 species in 35 genera and

accounted for approximately 98% of the

phycoperiphyton community. Diatoms from

28 species belonging primarily to the genera

Fragilaria, Synedra, Achanthes, Navicula,

and Pinnularia were further identified to

variety level. Seasonally dominant taxa in¬

cluded Achnanthes lanceolata, Melosira

varians, Cocconeis placentula euglypta, Na¬

vicula cryptocephala, and Diatoma vulgare.

These five taxa comprised about 44% of the

phycoperiphyton. Two species, Navicula

cryptocephala and Nitzschia palea, occurred

in each sample while 15 taxa appeared in

at least 93.9% of the samples. Navicula

cryptocephala ranked first in relative density

(16.72%). Cocconeis placentula euglypta

and Melosira varians ranked second and

third with 12.55% and 8.24%, respectively.

The Chlorophyceae were represented by

20 genera but the group was not a major

community component. This class contrib¬

uted only 2.24% of the algae encountered

but representatives of Chlorophyceae were

found in 85.7% of the samples. The most

frequently encountered green algae were

Scenedesmus quadricauda, Pediastrum du¬

plex, and Scenedesmus dimorphus, which

were found in 71.4%, 43.9%, and 42.9%

of the samples, respectively.

Single species of Merismopedia and of

Oscillatoria represented the class Myxo-

phyceae; each occurred in early autumn at

different stations.

Ceratium hirundinella was the sole rep¬

resentative of the class Dinophyceae and

was noted only once.

The rank of the common phycoperiphy¬

ton taxa according to mean relative density

(% of the total community) was determined

(Table 2). Common taxa are those taxa

with a relative density of > 1%. Eighteen

taxa were specified as common. Navicula

cryptocephala (16.72%), Cocconeis pla¬

centula euglypta (12.55%), and Melosira

varians (8.24%) ranked first through third,

respectively. These were followed by

Nitzschia palea (4.44% ) and Navicula

exigua capitata (3.91%). Six taxa had rela¬

tive densities between 3.72% and 2.74%,

while seven other taxa each comprised be-

Table 2. Common phycoperiphyton taxa in the

Black River ranked according to mean relative

density. Ranking is based on samples from all

stations, 10 April through 20 November 1976.

Relative Density (%)

Rank Taxon (mean and range)

1 Navicula cryptocephala 16.72

(0.36-60.78)

2 Cocconeis placentula euglypta 12.55

(0.00-56.02)

8.24

(0.00-54.43)

4.44

(0.37-12.38)

3.91

(0.00-14.26)

3.71

(0.00-18.55)

3.67

(0.00-34.55)

2.92

(0.00-12.35)

2.87

(0.00-66.60)

2.85

(0.00-14.38)

2.74

(0.00-15.30)

12 Achnanthes lanceolata rostrata 2.31

(0.00-12.05)

13 Gomphonema angustatum producta 2.07

3 Melosira varians

4 Nitzschia palea

5 Navicula exigua capitata

6 Cocconeis placentula

1 Achnanthes lanceolata

8 Melosira italica tennuissima

9 Diatoma vulgare

10 Cyclotella stelligera

11 Cyclotella atomus

72

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

Table 3. Common phycoperiphyton taxa in the Black River ranked according to frequency of occur¬

rence (%). Ranking is based on samples from all stations, 10 April through 20 November 1976.

tween 2.31% and 1.19% of the total com¬

munity.

Common taxa were ranked by frequency

of occurrence (Table 3). Frequency of oc¬

currence is defined as the percentage of sam¬

ples within which a taxon was found. Com¬

mon denotes those taxa whose frequency of

occurrence was at least 50%. This ranking

included only one non-diatom taxon, Scene-

desmus quadricauda. It ranked twenty-eighth

with three diatom taxa. Several taxa includ¬

ing Cyclotella meneghiniana (frequency

93.9%), Synedra ulna (93.9%), Amphora

ovalis (89.8%), Surirella tenera (84.7%),

Gyrosigma kutzingii (81.6%), and Ope-

phora martyii (80.6%) ranked in the top

twenty but were never abundant in any

sample. Only 52 of the 254 taxa occurring

in the Black River had a frequency of oc¬

currence >51%.

The algal density (cells/m2) was calcu¬

lated at each station. Stations which were

heavily shaded, e.g. 3-S and 1-S which aver¬

aged about 1.70 X 109 cells/m2, had phy¬

coperiphyton communities with the lowest

densities. This contrasted with 4-N (no

shade) and 3-N (shaded only in the eve¬

ning) where densities were about 3.00 X109

cells/m2 (Fig. 2). Direct sunlight affected

the density at each station and was catego¬

rized into three groups: those exposed to di¬

rect sunlight at least until evening (1-N,

3-N, 4-N), those exposed to 6 to 8 hr of di¬

rect sunlight (2-N, 2-S, 4-S), and those ex¬

posed to less than 3 hr of direct sunlight

(1-S and 3-S). The “Student’s” t-distribu-

1979]

Strenski — Phycoperiphyton Community of the Lower Black River

73

tion was used to examine the effect of direct

sunlight. Algal communities at stations with

direct sunlight until evening had significantly

greater densities than those at stations re¬

ceiving less than 3 hr of direct sunlight

(P <0.001). Similarly, densities at sta¬

tions with 6 to 8 hr of direct sunlight had

significantly greater densities than those re¬

ceiving less than 3 hr of direct sunlight

(P = 0.012).

Sampling began during April when the

high water resulting from the spring thaw

began to subside. Low densities were found

during April and May (8.8 X 108 cells/m2).

This may have resulted from the high spring

discharge scouring the glass substrate. A

spring maximum of 3.34 X 109 cells/m2 was

noted during early June; however, the great¬

est density occurred in early autumn when

3.47 X 109 cells/m2 were observed. The

lowest density (7.0 X 108 cells) was found

during the final sampling period in Novem¬

ber (Fig. 3).

Five species reached dominance during

the study. Early in the season, from April to

mid June, Achnanthes lanceolata was the

dominant species. Its highest relative density

was observed from 19 May through 2 June

Fig. 2. Mean density of phycoperiphyton with

respect to direct daily sunlight at each sampling

station in the Black River, 10 April through 20

November 1976.

when it accounted for 21.00% of the total

community. It remained common until 12

October when it comprised <1% of the

total community. During its dominance,

other common taxa included Gomphonema

angustatum producta, Achnanthes lanceolata

rostrata, Nitzschia palea, Navicula crypto-

cephala and Achnanthes lanceolata ventri-

cosa (in order of decreasing relative den¬

sity). Peak density of Achnanthes lanceolata

(4.3 X 108 cells/m2) occurred after Melo-

sira varians became dominant (Fig. 4). At

heavily shaded stations (1-S and 3-S)

Achnanthes lanceolata had a relative density

of 13%, while at stations with little or no

shade (1-N and 4-N) this species had a

relative density of about 18%.

Melosira varians dominated the commu¬

nity by mid-June. It attained a peak rela¬

tive density of 34.14% and an average of

27% during the period 16 through 30

June. This was the only species that also

exhibited a fall peak, reaching a value of

13.17% during the final sampling period.

It remained a common taxon during each

sampling period. Its relative density fell to

approximately 1.50% during August and

September. Peak density 16 through 30 June

Fig. 3. Seasonal mean density of phycoperiphyton

in the Black River, 10 April through 20 Novem¬

ber 1976.

74

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

was 8.2 X 108 cells/m'2. A fall peak of 4.0

X 10s cells was observed during 12 October

through 2 November (Fig. 4), Smith

(1950) documented this genus as the most

commonly encountered of all freshwater

Centrales. He further stated that Melosira

varians may dominate during the early

spring and late fall months. This species was

also one of the two principal species found

during October and November on the Ohio

River (Weber and Raschke 1970). In the

Black River the distribution of this species

appeared to be relatively unaffected by di¬

rect sunlight. Relative density of Melosira

varians at well-shaded stations differed from

Fig. 4. Seasonal mean density of the five dom¬

inant phycoperiphyton taxa in the Black River, 10

April through 20 November 1976.

unshaded stations by only 3%. For exam¬

ple, during the period 2 June through 14

July relative density at 1-N and 4-N (no

shade) was 24% while 1-S and 3-S (well-

shaded was 21% for this species. Other

common taxa occurring while Melosira vari¬

ans dominated were Navicula cryptocephala,

Cocconeis placentula euglypta, Achanthes

lanceolata, and Nitzschia palea and Navicula

exigua capitata.

During July, August, and much of Sep¬

tember, the dominant taxon was Cocconeis

placentula euglypta. During that time this

taxon ranged in relative density from

19.41% to 39.53% (14 thru 28 July). In

contrast, Achnanthes lanceolata and Melo¬

sira varians were found in rather uniform

relative densities while they were dominant.

Cocconeis placentula was also a common

taxon at this time. Massey (unpublished

data) found this species and several varie¬

ties dominant in Pool 8 of the Missis¬

sippi River during midsummer. Weber and

Raschke (1970) reported it as dominant in

the Kalmath River, Oregon, during August.

From 30 June through 22 September shad¬

ing appeared to inhibit growth of this spe¬

cies. A relative density of about 37% was

noted at stations 1-N and 4-N. During the

same period the relative density at stations

1-S and 3-S only about 18%. A peak den¬

sity of 9.2 X 10s cells/m2 was noted during

the sampling period 14 through 28 July (Fig.

4). Navicula cryptocephala, Cocconeis pla¬

centula, Navicula exigua capitata, Nitzschia

pales and Cyclotella stelligera were also

common throughout the summer months.

A shift in dominance to Navicula crypto¬

cephala occurred during the autumn. This

species was found only as a trace during the

extended first period (10 April through 19

May), but comprised at least 5.24% of the

algal community and was common through¬

out the remainder of the study. It attained

its peak relative density of 53.27%, from

12 October through 2 November. At the

same time it reached a peak density of 1.76

1979]

Strenski — Phy coperiphyton Community of the Lower Black River

75

X 109 cells/m- (Fig. 4). This was the only

taxon to yield a density greater than 109

cells/m2 during any one sampling period

and it was found in all samples. A variety of

Navicula cryptocephala was reported by

Bahls (1971) as a dominant species in the

Upper East Gallatin River, Montana and

was found in all samples collected from nat¬

ural substrates. In the Black River, direct

sunlight did not appear to significantly affect

its relative density. Heavily shaded stations

had 28% and essentially unshaded stations

29% from 8 September through 2 Novem¬

ber. Taxa also common at this time were

Melosira varians, Synedra ulna oxyrhynchus,

Nitzschia palea, Nitzschia linearis , and Dia-

toma vulgare .

Diatoma vulgare appeared only in small

amounts throughout most of the season;

however, it became the dominant alga dur¬

ing the cooler autumn months, comprising

32.3% of the community. This taxon was

reported by Blum (1957) in the Saline

River, Michigan, as characteristic of the fall

and winter diatom flora in unpolluted waters.

Drum (1964) also found it most abundant

during October and November in the Des

Moines River, Iowa. Low relative densities

were observed in the early sampling periods

of the study, thus suggesting that it had

been present throughout the winter months.

Apparently direct sunlight was not a sig¬

nificant factor affecting distribution. Its rela¬

tive density at station 3-S during the final

period was 66.60%. This was the greatest

dominance exhibited by any taxon during

this study. A density of 2.4 X 108 cells/m2

was recorded at this time. Other common

taxa during this time included Melosira

varians , Synedra ulna oxyrhynchus, Nitz¬

schia palea, and Navicula exigua capitata.

Diversity indices were calculated for each

sample both spatially and seasonally using

the Shannon-Wiener (1963) index which

ranged from 3.45 (4-S) to 4.19 (1-S) over

the eight stations. Shading appeared to have

a marked influence on diversity. Stations

with less than 3 hr of direct sunlight daily

had significantly greater diversities than sta¬

tions that received direct sunlight until the

evening (P = 0.014). Those stations which

received less than 3 hr of direct sunlight also

had significantly greater diversity than sta¬

tions with 6 to 8 hr (P = 0.034). High

mean diversity (>4.36) was observed dur¬

ing the early sampling periods (10 April

through 2 June). Seasonal changes were

most significant in the autumn when the

mean diversity was reduced from 3.93 (July

through mid-October) to 2.95 (for the re¬

mainder of the study). The overall seasonal

mean diversity was 3.89.

Lloyd and Ghelardi proposed the term

“equitability” to compare the observed mean

diversity with the maximum possible diver¬

sity (EPA 1973). The normal range is 0 to

1. In the Black River this index ranged

from 0.28 (4-N) to 0.46 (1-S) with respect

to spatial distribution, with an overall mean

of 0.36. Seasonal changes ranged from 0.19

(12 October through 2 November) to 0.55

(19 May through 2 June), with an overall

mean of 0.33.

Current velocity is also a critical factor

affecting the distribution of algae. Odum

(1956) implied that the depleted life re¬

quirements needed by the algal flora were

renewed and the accumulated by-products

were removed by the water flowing over the

organisms. Whitford (1960) stated that

swift currents rapidly remove the somewhat

impoverished water around a community

and replenish it with fresh nutrient-rich

water. Experiments conducted by Whitford

and Schumacher (1961) demonstrated high

rates of respiration and phosphorous uptake

in algal cultures when exposed to currents

of 0.15 m/sec as compared to still water.

Mclntire (1966) investigated the structure

of two laboratory stream periphyton com¬

munities in which the currents were very

slight in one stream and moderate (compar¬

able to the swifter currents at stations in

this study) in the other. He found that spe-

76

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

cies composition of the two laboratory

stream periphyton communities were ap¬

proximately the same but that the relative

densities differed significantly. This was also

observed in the Black River with Achnan-

thes lanceolata which had relative densities

of 11% (slow) and 17% (fast). Melosira

varians showed relative densities of 17%

(slow) and 22% (fast), while Cocconeis

placentula euglypta had 27% (slow) and

33% (fast). Navicula cryptocephala with

relative densities of 24% (slow) and 28%

(fast), showed the smallest difference.

Diatoma vulgare was markedly affected with

11% in slow currents in contrast to 24% in

fast currents.

In a review of 165 studies, Palmer

(1969) listed 80 of the most tolerant spe¬

cies of algae with respect to organic pollu¬

tion. Included in his list were greens, blue-

greens, diatoms, and flagellates. Twenty-

three species that appeared on Palmer’s list

were identified in the Black River; however,

only five were noteworthy. These species

were (Palmer’s rank in parenthesis) Nitz-

schia palea (2), Melosira varians (13),

Navicula cryptocephala (17), Diatoma vul¬

gare (40), and Cocconeis placentula eu-

glyypta (58). Nitzschia palea was a com¬

monly occurring species; however, in the

Black River it never attained a position of

dominance and always appeared in rather

uniform relative densities. The remaining

four species appeared in significant quanti¬

ties; however, their presence was related to

seasonal changes and apparently not to or¬

ganic pollution.

Acknowledgments

This study is based on a dissertation in

partial fulfillment of the requirements for

the degree of Master of Science-Biology,

University of Wisconsin-La Crosse. The au¬

thor wishes to express his appreciation to

Dr. Donald Rada who served as advisor

during the study. This study has been sup¬

ported in part by the River Studies Center

of the University and the Biology Depart¬

ment, the author extends thanks to these

organizations as well.

Literature Cited

American Public Health Association. 1975.

Standard methods for the examination of

water and wastewater. 14th ed. APHA, New

York.

Bahls, L. L. 1971. Ecology of the diatom

community of the Upper East Gallatin

River, Montana with in situ experiments on

the effects of current velocity on features of

the Aufwuchs. Ph.D. Disertation. Montana

State University, Bozeman, Montana.

Blum, J. L. 1957. An ecological study of

the algae of the Saline River, Michigan.

Hydrobiol. 9:361-405.

Bourelly, P. 1968. Les algues d’eau diuce.

Tome II les algues jaunes et brunes Chryso-

phycecees, Pheophycees, Xanthophycees et

Diatomees. Boubee and Cie, Paris.

Cleve-Euler, A. 1955. Die diatomeen von

Schweden und Finnland, Vol. I-V, Almquist

und Wiksells Boktryckeri, Stockholm.

Drum, R. W. 1964. Ecology of diatoms in

the Des Moines River, Ph.D. Disertation.

Iowa State University of Science and Tech¬

nology, Ames, Iowa.

Hansmann, E. W. 1973. Diatoms of the

streams of Eastern Connecticut. State Geo¬

logical and Natural History Survey of Con¬

necticut, Department of Environmental Pro¬

tection, Bulletin 106.

Hohn, M. H. and J. Hellerman. 1963. The

taxonomy and structure of diatom popula¬

tions from three eastern North American

rivers using three sampling methods. Trans.

Amer. Microscop. Soc. 82:250-329.

Huber-Pestalozzi, G. and F. Hustedt. 1942.

Die Kieselalgen. In A. Thienemann, ed.,

Das Phytoplankton des Susswassers, Die

Binnengewasser, Band XVI, Teil II, Halfs

II. E. Schweizerbant’sche Verlagsbuch-Han

dlung, Stuttgart.

Hustedt, F. 1930a. Bacillariophyta. In A.

Pascher, ed., Die Suswasser-Flora Mitteli-

uropas, Heft 10. Gustav Fischer, Jana.

Hustedt, F. 1930b. Die Kieselalgen. In L.

Rabenhorst, ed., Kryptogamen-Flora von

Deutschland, Osterreich, und der Verlagsell-

schaft m.b.h., Leipzig.

1979]

Strenski — Phy coperiphyton Community of the Lower Black River

77

Mclntire, C . D. 1966. Some effects of cur¬

rent velocity on periphyton communities in

laboratory streams. Hydrobiol. 27:559-570.

Odum, H. T. 1956. Primary production in

flowing waters. Limnol. and Oceanogr. 1:

102-117.

Palmer, C. M. 1969. A composite rating of

algae tolerating organic pollution. J. Phycoh

5:78-82.

Palmer, C. M. and T. E. Maloney. 1954. A

new counting slide for nanoplankton. Amer.

Soc. LimnoL and Oceanogr. Spec. Pub. No.

21: 1-6.

Patrick, R. and C. M. Reimer. 1966. The

diatoms of the United States. Vol. 1. Mono¬

graph 13, Philadelphia Acad. Natur. Sci.

Prescott, G. W. 1962. Algae of the Western

Great Lakes Area, 2nd ed. W. C. Brown,

Dubuque, Iowa.

Shannon, C. E. and W. Weaver. 1963. The

mathematical theory of communication.

University of Illinois Press, Urbana.

Smith, G. M. 1950. The freshwater algae

of the United States, 2nd ed. McGraw-Hill

Book Co., New York.

Strenski, M. R. 1977. A study of the phy-

coperiphyton community in the Black River,

La Crosse County, Wisconsin. M.S. thesis.

University of Wisconsin-La Crosse, La

Crosse, Wisconsin.

Tiffany, L. H. and M. E. Britton. 1952.

The algae of Illinois. University of Chicago

Press, Chicago.

United States Environmental Protection

Agency. 1973. Plankton, p. 13-20, and

Periphyton, p. 1-6 in Biological field and

laboratory methods for measuring the qual¬

ity of surface waters and effluents. EPA-

670/4-73-001. EPA, Cincinnati.

Weber, C. I. 1971. A guide to the common

diatoms at water pollution surveilance sys¬

tem stations. EPA, Cincinnati.

Weber, C. I. and R. L. Raschke. 1970. Use

of a floating periphyton sampler for water

pollution surveilance. USFWPCA, Cincin¬

nati.

Whitford, L. A. 1960. The current effects

and growth of fresh-water alga. Trans.

Amer. Microscop. Soc. 79:302-309.

Whitford, L. A. and G. J. Schumacher. 1961.

Effects of current on mineral uptake and

respiration by a freshwater alga. Limnol.

and Oceanogr. 6:423-425.

Whitford, L. A. and G. J. Schumacher. 1973.

A manual of fresh-water algae. Sparks Press,

Raleigh, North Carolina.

PRELIMINARY REPORTS ON THE FLORA OF WISCONSIN

NO. 67. VERBENACEAE— THE VERVAIN FAMILY

William E. Tans

Scientific Areas Section

Department of Natural Resources

Madison, Wisconsin

Hugh H. Iltis

Department of Botany

University of Wisconsin

Madison, Wisconsin

A bstract

The Verbenaceae, a huge, tropical plant family, is represented in Wisconsin

by the genera Verbena, with five native and three cultivated species, and six

hybrids, and Phyla, with one species. Wisconsin distribution maps plotted from

herbarium specimens show the ranges of the above taxa. Dichotomous keys to

genera and species, and descriptions, geographic range, ecology, phenology, and

chromosome number for each species are given.

Introduction

The Verbenaceae, a large, primarily tropi¬

cal family, is relatively poorly represented

in Wisconsin, with only two genera and 15

taxa, including six native species, their six

hybrids, as well as three garden ornamentals

which are rare escapes. Of these, the taller,

common, often weedy Verbena hastata, Blue

Vervain, V. urticifolia, White Vervain, and

V. stricta, Hoary Vervain are familiar plants.

The Verbenaceae present no special prob¬

lems to floristic workers despite the many

hybrids, largely because Harold N. Mol-

denke, formerly of the New York Botanical

Garden, has monographed the Verbenaceae

and published treatments in many regional

floras. Some 70 publications on Verbena

alone by this prolific author have appeared

in Phytologia since 1961; those pertinent to

Wisconsin are listed in the bibliography.

Much information regarding species ranges,

descriptions of hybrids, and taxa of infre¬

quent occurrence in Wisconsin has been

taken from the publications of Dr. Mol-

denke, who, in addition to verifying the iden¬

tification of most of our specimens, com¬

mented helpfully on this manuscript. Since

Dr. Moldenke, above all others, has made

an accurate treatment of the Wisconsin Ver¬

benaceae possible, this report is dedicated

to him.

This study is based primarily on speci¬

mens deposited in the herbaria of the Uni¬

versity of Wisconsin-Madison (WIS), UW-

Milwaukee (UWM), and the Milwaukee

Public Museum (MIL), as well as UW-

Oshkosh (OSH), UW-La Crosse (UWL),

UW-Stevens Point (UWSP), UW-Superior

(SUWS), UW-River Falls (RIVE), UW-

Rock County Campus, Janesville (UWJ),

the University of Iowa (IA), University of

Minnesota (MIN), University of Michigan

(MICH), Beloit College (BELC), the Mor¬

ton Arboretum (MOR), Field Museum of

Natural History (F), and the private her¬

barium of Mrs. K. Rill, Oshkosh. Grateful

acknowledgment is extended to the admin¬

istrators of these herbaria for loan speci¬

mens. Dr. Edward G. Voss, University of

Michigan, Dr. Gerald Ownbey, University

of Minnesota, and Vicki Funk, Ohio State

University, provided distribution records of

Verbena urticifolia to complete the range

map for that species.

Each solid or hollow circle, square, or

plus ( + ) symbol on the Wisconsin maps

represents the exact location where speci¬

mens were collected; triangles indicate coun¬

ty records without specific locations. Dots

on the United States map indicate specimen

occurrence by county. Numbers in the state

map insets record by month the number of

78

1979]

Tans and litis— Preliminary Reports — -Verbenaceae

79

specimens examined in flower or fruit, i.e.

the phenology of the species.

Since the Verbanaceae, world-wide, are

morphologically a most diverse group, the

descriptions that follow refer only to the

Wisconsin species.

VERBENACEAE Saint-Hilaire

Vervain Family

Herbs, shrubs, vines or trees in the trop¬

ics. Leaves (in ours) opposite, exstipulate,

simple; blades entire to dentate, incised or

cleft. Inflorescences axillary or terminal,

ours in heads {Phyla) or spikes {Verbena).

Flowers perfect, hypogynous, more or less

irregular. Calyx lobes fused, campanulate to

tubular or salverform, 2- or 5-lobed. Corolla

with fused petals, funnelform or salverform,

tube well developed, the limb 4- or 5-lobed,

more or less weakly 2-lipped. Stamens 4,

didynamous (2 long, 2 short), inserted on

the corolla tube. Ovary superior, sessile,

somewhat 4-lobed, 2-carpellate, 2- or 4-

celled with one ovule in each cell. Fruit a

dry schizocarp splitting into 2 {Phyla) or 4

{Verbena) nutlets at maturity.

A large, morphologically heterogeneous

family, widely distributed throughout the

world, most abundant in the tropics, with 76

genera and 3,400 species and subspecies

(Moldenke 1971), the largest genera, Vi-

tex, Clerodendrum, Verbena (including

Glandularia) and Lantana, containing both

widely used ornamentals and pernicious

weeds. The hard, heavy, durable wood of

teak {Tectona grandis) is economically im¬

portant in many tropical areas.

The Verbenaceae, together with Lamia-

ceae (Labiatae), Boraginaceae, Callitricha-

ceae and Phrymaceae, comprise the order

Lamiales, with 7800 species (Cronquist

1968). The Lamiales, Scrophulariales and

Polemoniales comprise the super-order Tu-

buliflorae. The monotypic, temperate, her¬

baceous Phrymaceae {Phryma leptostachya;

cl. litis 1957), sometimes included in Ver¬

benaceae (Thorne 1976) or segregated as

a separate family on the basis of its reduced

and highly specialized ovary, is clearly de¬

rived from Verbenaceae (Cronquist 1976).

Most closely related to Verbenaceae, but

more advanced, are the Lamiaceae, distin¬

guished by a gynobasic style, deeply 4-

cleft ovary, sharply 2-lipped flowers, and

aromatic foliage. In contrast, Verbenaceae

have a terminal style, only slightly or un-

lobed ovaries, at most weakly 2-lipped flow¬

ers, and often nonaromatic foliage.

Key To Genera

A. Flowers in elongated spikes, these terminating stems and branches; calyx

5-lobed; and cylindrical; corolla 5-lobed fruits separating into 4 nutlets .

. . . . . . 1. Verbena.

AA. Flowers in dense heads axillary on long, bare peduncles; calyx 2-lobed and

compressed; corolla 4-lobed; fruits separating into 2 nutlets . 2. Phyla.

1. Verbena L. Vervain

Herbs, erect, ascending or procumbent,

perennial. Leaves opposite, toothed, incised,

lobed or pinnatifid. Inflorescences terminal

or axillary, spicate, loose to dense, few to

often many and paniculately arranged.

Flowers small or rather large in Sect. Glan¬

dularia, solitary and sessile in the axil of a

bractlet. Calyx tubular, 5-ribbed, unequally

5-toothed. Corolla salverform or funnel-

form, its tube curved, villous within, a ring

of hairs closing its mouth, the limb spread¬

ing, weakly 2-lipped, 5-lobed, slightly zygo-

morphic. Stamens 4, inserted on the upper

portion of the tube, included. Style one, 2-

lobed, the anterior lob stigmatiferous. Ovary

superior, with one ovule in each of four

cells. Fruit enclosed by the mature calyx,

splitting at maturity into four small, linear

nutlets.

A highly complex genus of 206 species,

122 subspecific taxa and 49 hybrids, native

80

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

primarily to the New World (Moldenke

1971), most frequent in the southcentral

United States, Mexico, and South America,

with three species native to the Mediter¬

ranean region, several American species are

weeds in the Old World and many species

widely cultivated ornamentals. The showy

Section Glandularia is now often elevated to

generic status (Umber, 1979).

Several Wisconsin species of Verbena

(e.g. has tat a, stricta, urtici folia, brae teat a)

contain in their leaves and stems the bitter

glucoside Verbenalin. Verbenalin, with the

disagreeable tasting seed oils, may provide

a competitive advantage in Wisconsin pas¬

tures, the plants being selective ignored by

livestock. Singly or in various combinations,

these species (especially V. stricta ) are

abundant in grazed areas throughout south¬

ern and western Wisconsin.

Originally, the three most common Ver¬

vains were probably ecologically segregated,

with V. hastata in marshes, V. stricta in

prairies, and V. urticifolia in woodlands.

Today however, disturbance in most areas

has obliterated this separation, and our na¬

tive Vervains not only flourish but hybridize

in such disturbed habitats. Considering the

ubiquitous joint occurrences of many of these

species, the relative rarity of hybrids is re¬

markable.

Verbena officinalis has been attributed

erroneously to Wisconsin on the basis of two

specimens (WIS) collected at Baraboo,

Sauk County in 1861 (Hartley 1966). Both

are V. x perriana (V. urticifolia x V. brac-

teata ), their pinnatifid leaves resembling

those of V. officinalis.

Key to the Wisconsin Species of Verbena

(Hybrids among the first five species are not separable in this key; cf. pp. 89-93,

Figs. 1, 4, 5, and The New Britton and Brown Illustrated Flora 3: 126-127).

A. Sterile style-lobe not protruding beyond the stigmatic surface; flowers 3-9 mm

long, 3-9 mm broad; calyx 1-6 mm long; flowers in elongate, slender spikes;

native species (Section VERBENA).

B. Spikes usually many, paniculate; fruiting calyx 1.8-3 mm long; corolla

2. 5-4. 5 mm broad.

C. Spikes slender, the fruiting calyces more or less remote; calyx 1.8-

2.4 mm long; corolla dull white, its tube 1.8-2. 5 mm long. .....

. 1. V. URTICIFOLIA.

CC. Spikes thick, densely flowered, the fruiting calyces imbricate; co¬

rolla deep blue, lavendar or purple (white in occasional albinos);

calyx 2.5-3 mm long; corolla tube 3-4 mm long or more .

. . . . . . 2. V. HASTATA.

BB. Spikes solitary or in 3’s; fruiting calyx 2. 9-5-6 mm long; corolla 3-9

mm broad.

D. Stems prostrate-ascending, often forming extensive mats, 5-10

(20) cm tall; bracts divergent, foliaceous exceeding the calyx;

leaves lobed or incised usually near the base. . .3. V. bracteata.

DD. Stems erect, 3-11 dm tall; bracts appressed, subulate-lanceolate,

shorter than or equalling the calyx; leaves toothed, rarely lobed.

E. Plants sparsely strigose; leaves lanceolate to oblanceolate, 3-

11 mm wide; rare in southeastern Wisconsin. 4. V. simplex.

EE. Plants densely pubescent; leaves elliptic to orbicular, 13-50

mm wide; common throughout southern Wisconsin .

. 5. V. STRICTA.

1979]

Tans and litis — -Preliminary Reports — V erbenaceae

81

AA. Sterile style-lobe protruding beyond the stigmatic surface; flowers 10-30 mm

long, 8-30 mm broad, very showy; calyx 8-15 mm long; flowers in flat-

topped, foreshortened spikes; rare garden escapes (Section GLANDU-

LARIA).

F. Corolla 20-30 mm long, 10-25 mm broad, the tube twice as long as the

calyx.

G. Leaves appressed strigose or glabrate; flower rose to purple. ....

...................................... 6. V. CANADENSIS.

GG. Leaves densely pubescent; flowers violet to red and white, often

with a white central eye spot in the throat. ... .7. V. x hybrida.

FF. Corolla 10-15 mm long, 8-10 mm broad, the tube 1.3- 1.5 times as long

as calyx; flowers rose to purple. .............. 8. V. bipinnatifida.

Section Verbena (Native Species)

1. Verbena urticifolia L. Maps 1-3.

White Vervain, Nettle-leaved Vervain

Herb, perennial, 5-15 ( — 25) dm tall.

Stems erect, simple or with ascending

branches, quadrangular. Leaves ovate to

elliptic, occasionally lanceolate, 4-14 cm

long, 3-7 cm wide, with pustulate hairs

above, subglabrous to densely velutinous

beneath, acute, abruptly attenuated to the

1.5-3. 5 cm long petiole, coarsely serrate or

biserrate. Leaf hairs simple, sometimes

gland-tipped on the main veins beneath.

Spikes slender, becoming remotely fruited,

5-28 cm long, numerous in open panicles.

Bractlets ovate-acuminate, 0.7-1. 2 mm long.

Calyx 1.8-2. 4 mm long, minutely pubescent,

the subequal teeth not connivent. Corolla

dull white , slightly surpassing the calyx, 2.5-

3.5 mm long, the tube 1.8-2. 5 mm long.

Nutlets oblong, 1. 5-2.0 mm long, raised re-

82

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

ticulate-striate. 2n = 14 (Lewis & Oliver

1961).

Widespread and common in eastern North

America (Map 1), from Quebec to Sas¬

katchewan south to Florida and Texas, in

Wisconsin primarily south of the “Tension

Zone” in a great variety of habitats, from

dry oak to moist maple-basswood and flood-

plain forests, and sedge meadows (there

modal, fide Curtis 1959), an aggressive

weed tolerant of disturbance, common in

grazed or cut-over woods, pastures, grazed

or drained sedge meadows, roadsides, and

railroad rights-of-way and other sunny or

shaded, mesic and moist degraded sites.

Verbena urticifolia is one of the few na¬

tive weeds frequently seen co-dominant

with grasses in heavily grazed pastures.

Here an obvious increaser (as is V. stricta,

which may occur with it but in drier spots),

it is avoided by cattle.

Flowering from the end of June through

the end of September, fruiting from July

through October.

Verbena urticifolia tends to be autoga¬

mous, its minute flowers setting seed whether

excluded from or exposed to insect pollina¬

tors. Self-pollination occurs when the wilting

of the persistent corolla allows the anthers

to contact the stigma. Pollinating insects are

chiefly flies (Perkins et ai 1975).

In Wisconsin, Verbena urticifolia hybri¬

dizes most frequently with V. hastata ( V . x

engelmannii ) and rarely with V. bracteata

(V.x perriana ) and V. stricta (V. x illicita) .

V HASTATA

V BRACTEATA V. SIMPLEX

Fig. 1. Species of Verbena native to Wisconsin

and their hybrids. Those connected by dashed

lines have not been collected in the state.

var. leocarpa — after Fernald (1936) and from Wisconsin material identified by Moldenke.

1979]

Tans and litis — Preliminary Reports — Verbenaceae

83

A hybrid with V. simplex (V x stuprosa )

has not been found here (Fig. 1).

Variability in Verbena urticifolia

Verbena urticifolia exhibits much varia¬

bility, as expected of such a wide-ranging

species occurring in diverse habitats. One

of the variants has been segregated as var.

leiocarpa Perry & Fernald (Fernald 1936).

The characters originally used to separate

the variety, as well as an additional one cor¬

related with it, are listed in Fig. 2. In the

original description, var. leiocarpa was said

to occur from “Eastern Virginia to South

Carolina” (and rarely northward to Con¬

necticut), in contrast to the widespread dis¬

tribution of the typical variety. While this

variety was originally claimed to be a

Coastal Plain endemic, Moldenke subse¬

quently identified many of the Wisconsin

specimens as var. leiocarpa, apparently on

the basis of the velutinous pubescence on

the underside of the leaves.

On the basis of Perry and Fernald’s origi¬

nal taxonomic characters (length of calyx,

corolla and bractlet, inflorescence angle,

nutlet ribbing) the Wisconsin specimens of

Verbena urticifolia can not be separated

into two varieties. There is no correlation

between calyx length, bractlet length and

nutlet length for specimens separated into

two varieties on the basis of pubescence

(Fig. 3). Even Fernald’s best distinguishing

character, the soft pubescence on the under¬

side of the leaves, is variable and intergrad¬

ing. Nevertheless, it does seem to correlate

slightly with geography, for var. leiocarpa

is found primarily south of the Tension

Zone in Wisconsin (Map 3), while the typi¬

cal variety (Map 2) does range north of the

Tension Zone, although infrequently. A

similar, and hence significant, restriction of

var. leiocarpa to more southerly areas has

been reported for Western Pennsylvania

(Jennings 1953). Correlated with distribu¬

tion, again in a poorly defined way, are the

additional characters of presence and rela-

O VERBENA URTICIFOLIA VAR. URTICIFOLIA

• VERBENA URTICIFOLIA VAR. LEIOCARPA

CALYX LENGTH MM

Fig. 3. Scatter diagram of Wisconsin specimens

of Verbena urticifolia var. urticifolia and var.

leiocarpa (determined by Moldenke) illustrating

the lack of correlation between three of the taxo¬

nomic characters Fernald utilized to separate the

variety.

tive abundance of gland-tipped hairs, these

primarily on the rachis, petioles, upper

nodes and midveins on the ventral leaf sur¬

face.

In summary, while these distinctions seem

hardly worthwhile, the map and scatter dia¬

gram suggest that originally there may have

been a Coastal Plain population differenti¬

ated from an inland one. With the waning

of the Pleistocene ice, both varieties mi¬

grated into the formerly glaciated lands to

become sympatric. With differential selection

pressure relaxed, morphological distinctness

was reduced by exchange of genetic material

in many populations, and continued evolu¬

tionary divergence of the varieties was

swamped through recombination. Perhaps

the descendants of the Coastal Plain popu¬

lation are the velutinous ones which now

seem to be restricted to the warmer parts

of Wisconsin, a region harboring many

Coastal Plain taxa.

While most floras disregard Fernald’s va¬

riety (e.g. Radford, et al. 1968), calling at¬

tention to such variability does serve a use¬

ful biogeographic purpose. Fernald, a genius

at recognizing subtle plant variation in the

1979]

Tans and litis — Preliminary Reports — V erbenaceae

85

86

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

field, may not have been totally wrong! Only throughout its range can resolve this ques-

detailed population studies of this species tion.

Key to Varieties

A. Leaves nearly glabrous to hirtellous beneath; hairs 0.5-1. 1 mm long; gland-

tipped hairs rarely present except in more densely pubescent specimens;

nearly throughout Wisconsin. . . la. Verbena urticifolia

var. URTICIFOLIA.

AA. Leaves more or less densely velutinous below; hairs mostly less than 0.4 mm

long; scattered gland-tipped hairs usually present on upper nodes, rachis,

petolies and midrib of ventral leaf surface; mostly south of the Tension Zone.

. lb. Verbena urticifolia

var. LEIOCARPA.

la. Verbena urticifolia L. var.

urticifolia Map 2.

lb. Verbena urticifolia L. var.

leiocarpa Perry & Fernald Map 3.

Morphological intermediates between

these ill-defined variants possess either a

dense pubescence of mixed long and short

hairs with gland-tipped hairs, or a sparse

but very short pubescence without or with

few gland-tipped hairs, in each example il¬

lustrating continuous variation. These are

indicated on Map 2 with plus symbols.

2. Verbena hastata L. Blue Vervain,

Simpler’s Joy Map 4.

Verbena hastata L. var. scabra Moldenke

Herb, perennial, 2-13 dm tall. Stems

erect, 1 -several, simple or ascending

branched, scabrous with short antrorse hairs

on the angles. Leaves to 15 cm long, 5.5 cm

wide, narrowly ovate to elliptic, on narrow

winged petioles, reduced upwards, gradually

accuminate and rounded at base, coarsely

serrate or biserrate, the larger sometimes

hastately lobed, scabrous to rough hirsute

on both surfaces but softer beneath. Spikes

5-12 cm long, rarely solitary, usually nu¬

merous in a dense terminal panicle, erect

and densely many-flowered. Bractlets 1.8-

2.8 mm long, lanceolate-subulate, shorter

than the calyx. Calyx 2. 5-3.0 mm long, its

acute lobes with subulate, connivent tips.

Corolla deep lavender or purple (white in

forma albiflora Moldenke, rose in forma

rosea Cheney), 3-4.5 mm broad, the tube

3-4 mm long surpassing the calyx. Nutlets

linear, 1.8-2. 1 mm long, faintly striate. 2n

- 14 (Mulligan 1961).

Widespread from Nova Scotia to British

Columbia, south to Florida, Texas, and Ari¬

zona, but only sporadic in the Western

States, rare in the Southern and Southeast¬

ern States (cf. Moldenke 1963b for map),

in Wisconsin the most common Vervain,

occurring in a variety of usually moist,

sunny habitats, as in marshes, stream edges,

lake shores, shrub-carrs, low prairies, sedge

meadows (where modal: Curtis 1959), and

rarely in moist forests, tolerating much habi¬

tat disturbance, hence common in heavily

grazed pastures, roadsides and railroad

rights-of-way, and occasionally abandoned

sandy fields.

Flowering from late June through Sep¬

tember, fruiting from early July to mid-

October.

In describing var. scabra, Moldenke

(1963b) states that “This variety differs

from the typical form of the species in hav¬

ing its leaf blades more rigid, conspicuously

scabrous above, and often more or less con¬

spicuously pubescent beneath,” and later

(1971), “Probably most, if not all, of the

material now passing as Verbena hastata

from west of the Mississippi, or at least,

west of the Great Plains, is actually

var. scabra .” While Moldenke has identified

1979]

Tans and litis — Preliminary Reports — Verbenaeeae

87

specimens from several counties in the state

as van scabra, all Wisconsin specimens of

V. hastata are rough hispid to scabrous

above and pubescent beneath, and seem to

be part of the western, more scabrous end

of a subcontinental, taxonomically indivisi¬

ble cline.

In Wisconsin, Verbena hastata hybridizes

with V. stricta ( V. x rydbergii) and V.

urticifolia ( V . x engelmannii) . A hybrid

with V . simplex (V. x blanchardi) has not

been collected here (Fig. 1 ).

3. Verbena bracteata Lag. & Rodr.

Creeping Vervain, Large-bracted

Vervain Map 5.

Verbena bracteosa Michx.

Herb, perennial. Stem prostrate-ascend¬

ing, diffusely branched from a thickened tap

root, to 6 or more dm long, 5-20 cm tall,

often forming large, round mats, coarsely

hispid-hirsute. Leaves 1-4.5 cm long, 0.8-

3.2 cm wide, lanceolate to ovate, pinnately

incised or 3-lobed, gradually narrowed into

a short petiole. Spikes few to many, elon¬

gating in fruit, densely flowered and conspic¬

uously bracteate. Bractlets 7-13 mm long

and much longer than the calyx, foliaccous,

spreading to recurved in age, hirsute. Calyx

2.9-4. 3 mm long, white-hispid, the acumi¬

nate lobes connivent. Corolla lavender-pur¬

ple to pink, 2. 5-4.0 mm broad, the tube

about 4 mm long and slightly exsert from

the calyx. Nutlets linear, 2-2.5 mm long.

2n = 14 (Jackson 1960).

Wide ranging from Maine to British Co¬

lumbia, south to Florida, Texas, and Mexico

(cf. map in Moldenke 1962b), and intro¬

duced in Western Europe, in Wisconsin

primarily south of the Tension Zone, a weed

(Curtis 1959) of droughty and sunny dis¬

turbed habitats as sandy or gravelly road¬

sides, railroad rights-of-way, quarries, open

woods, sandy river terraces, trails, and waste

places in towns.

Flowering from the first week in June,

fruiting from mid-June through October.

Once established, the ability of Creeping

Vcrvian to persist is exemplified by a Rich¬

land County collection (Nee 13,460, WIS)

where it is reported to be “common, in de¬

cumbent mats in heavily grazed hog lot . . .

now the only living plant.”

Although common and apparently indig¬

enous, it appears not to be part of any na¬

tive plant community (Swink 1974), except¬

ing perhaps sandy river terraces and dry

rocky exposed cliffs where it is rarely col¬

lected. These, supposedly undisturbed sites

include Brady’s Bluff Prairie (Trempealeau

County), Ferry Bluff (Sauk County), and

Observatory Hill (Marquette County), all

rocky, xeric “open” habitats with minimal

competition characterized by bedrock ex¬

posures and shallow soils.

But even the earliest Wisconsin collec¬

tions stress disturbance: “A roadside weed

very plentiful in some places” ( Lapham

s.n., Waukesha Co., 11 Aug. 1847, WIS) :

“. . . in waste places” ( Cheney s.n., Lafay¬

ette Co., 27 Aug. 1888, WIS); “On ballast

of C.M.St.P.&P. RR near University Farm.”

(Heddle 706, Dane Co., 1 Aug. 1907,

MIL). C. C. Parry (1852), who cataloged

the plants during a geological survey of Wis¬

consin, Minnesota and Iowa in 1848, char¬

acterized the habitat of V. bracteata as

“roadsides.” Although the plant probably

existed in Wisconsin prior to settlement cer¬

tainly its frequency has increased greatly

from the time of the first settlers.

Capable of autogamy, Verbena bracteata

shows a tendency for cross-fertilization, with

butterflies the primary visitors (Perkins et

al. 1975).

In Wisconsin, Verbena bracteata hybrid¬

izes with V. stricta (V . x deamii) and V.

urticifolia (V x perriana) (Fig. 1).

4. Verbena simplex Lehm.

Narrow-leaved Vervain Map 6.

Verbena angustifolia Michx.

Herb, perennial, 2.8-5 dm tall. Stems

erect, and simple, or ascending and some¬

times spreading branched, sparsely strigose.

Leaves 4. 5-7. 5 cm long, 3-11 mm wide,

88

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

linear to narrowly oblong or spatulate, ta¬

pering to a subsessile base, serrulate often

only toward the acute apex, sparsely strigil-

lose to scabrous on both surfaces. Spikes

3-24 cm long, slender, stiffly erect. Bract-

lets 2. 5-4.3 mm long, lanceolate-subulate,

shorter to as long as the calyx. Calyx 3-4.5

mm long, the five accuminate lobes erect to

spreading in fruit. Corolla light blue to pur¬

ple or rose, 5-7.1 mm broad, the tube 3.8-

4.8 mm long, slightly surpassing calyx. Nut¬

lets linear, 2.3-3 mm long, raised reticulate

above, striate below. 2n = 14 (Noack

1937).

Widespread from New Hampshire and

Massachusetts to Ontario, southern Minne¬

sota, and Nebraska, south to Florida and

Oklahoma (cf. map in Moldenke 1964c), in

Wisconsin only in the southern third, where

rare in open, dry, calcareous sites such as

gravelly moraines, roadsides, or railroad

rights-of-way.

Flowering from mid-June through mid-

August, fruiting late June to September.

In Wisconsin, V. simplex hybridizes with

V. stricta (V. x moechina ). Hybrids with

V. hastata ( V . x blanchardi ), and V. urti-

cifolia (V. x stuprosa) are not known from

Wisconsin (Fig. 1).

5. Verbena stricta Vent.

Hoary Vervain Map 7.

Verbena stricta Vent, forma albiflora

Wadmond, in Rhodora 34:19. 1932. (Type:

“southeastern Wisconsin,” in Gray Herbar¬

ium Harvard U. (?), not seen).

Herb, perennial, 2.9-11 dm tall. Stems

erect, robust, simple to branched, densely

pubescent. Leaves 3. 2-9. 4 cm long, 1.3-5

cm wide, narrowly elliptic, orbiculate, or

widely ovate, sessile or nearly so, sharply

serrate, biserrate or irregularly incised-ser-

rate, thick-textured, hirsute and rugose

above, densely canescent below. Spikes usu¬

ally 7, or 3 to 5, or more, thick and stiffly

erect, hirsute throughout, to 3.1 dm long.

Bractlets 3. 8-5. 6 mm long, lanceolate-subu¬

late, hirsute, shorter or equalling the calyx.

Calyx 3. 8-5. 6 mm long, densely hirsute, its

five lobes acuminate. Corolla purple, laven¬

der or blue (pink or rose in forma rosei-

flora Benke, white in forma albiflora

Wadmond) 7.5-9 mm wide, the tube 3.8-

6.4 mm long, surpassing the calyx. Nutlets

ellipsoid, 2.3-3 mm long, raised reticulate

above, striate below. 2n = 14 (Noack 1937).

A native of the Great Plains, from On¬

tario and Ohio to Montana, south to Texas,

Arizona, and Mexico (cf. map in Moldenke

1964c), in Wisconsin south of the Tension

Zone in many dry, sunny habitats such as

xeric and sandy prairies, limey or “goat

prairies” (here modal, Curtis 1959), abun¬

dant in heavily grazed, sandy or gravelly

pastures or abandoned fields, less frequently

in open oak or oak-jack pine woods, road¬

sides, and railroad rights-of-way.

Flowering from late June to early Octo¬

ber, fruiting from early July through Octo¬

ber.

In studying effects of grazing on thin soil

Wisconsin prairies, Dix (1959) gave V.

stricta a high negative grazing susceptibility

rating because of its greatly increased fre¬

quency on grazed hill prairies.

Capable of autogamy as are many weeds,

V. stricta shows a tendency for cross fertili¬

zation due to partial self-incompatibility.

Selfing is thwarted by the position of the

corolla tube which, horizontal and partially

closed by hairs, prevents pollen falling onto

a stigma unreceptive until anthesis. Insect

visitors are equally divided between Diptera,

Hymenoptera, and Lepidoptera (Perkins et

al. 1975).

In Wisconsin, V. stricta hybridizes with

each of the other four native species (Fig.

1), but most commonly with V. hastata ( V .

x rydbergii).

Section Glandularia Schauer

(Adventive Species) (See Umber 1979)

6. Verbena canadensis (L.) Britton

Rose Vervain

Herb, annual or perennial. Stems erect or

decumbent, often rooting at the nodes, gla-

1979]

Tans and litis— -Preliminary Reports— V erhenaceae

89

brous to spreading-hirsute. Leaves coarsely

incised, pinnatifid or 3 -cleft, glabrous to

hirsute on both sides. Spikes with many

large flowers in flat-topped, showy clusters.

Bractlets mostly shorter than calyx. Calyx

10-13 mm long, glandular-hirsute, with sub¬

ulate lobes. Corolla usually purplish-rose, 11-

15 mm broad , the tube about twice as long

as the calyx. Nutlets 3-3.5 mm long, 2n

3 0 (Dermen 1936).

Common from Pennsylvania, Tennessee

and Colorado south to Florida and Texas,

most abundant in the Ozarks, cultivated and

sometimes naturalized in Minnesota and

southern Michigan (cf. map in Moldenke

1962c), its seeds available commercially

and in Wisconsin occasionally planted in

cemeteries or flower gardens. Two speci¬

mens were observed: “in landscaped road¬

side gravel,” (Walworth County, T4N

RISE NE14SW14 Sect. 7, Sept. 4, 1975

Tans 1976); “dry roadside, south shore

Lake Wingra, UW-Arboretum,” (Dane

County, Sept. 5, 1971 (. Reardon 045, OSH).

7. Verbena x hybrida Voss ex Rumpler

Garden Verbena

Herb, annual or perennial. Stems freely

branched, procumbent, forming mats to 1 m

wide, densely hirsute or villous. Leaves 1.5-

8 cm long, 1.5-8 cm wide, dentate, trun¬

cate to cuneately narrowed to petiole, both

sides densely soft pubescent. Spikes a flat-

topped corymb, the flowers large and showy.

Bractlets 5-6 mm long, lanceolate, shorter

than calyx, densely soft pubsecent. Calyx

8-15 mm long, densely white-hirsute. Cor¬

olla showy, variously colored (blue, deep

violet, red, etc.), usually with a prominent

central white “eye,” 10-25 mm broad, the

tube 15-30 mm long.

A polymorphic hybrid of uncertain an¬

cestry, supposedly involving the South

American V. platensis, V. phlogiflora, V.

incisa, V. peruviana, and V. tenera (Mol¬

denke 1963b), commonly cultivated as an

ornamental throughout the world, known in

Wisconsin only from gardens since at least

the 1860’s (T. /. Hale s.n. WIS). 2n = 10,

20 (Furusato 1940).

8. Verbena bipinnatifida Nutt.

Cutleaf Verbena

Herb, annual or perennial. Stems erect or

procumbent, diffusely branched, hispid-hir¬

sute. Leaves 2-6 cm long, ovate in general

outline, delicately bipinnatifid or tripinnati-

fid with the divisions deeply cleft. Spikes,

canescent, with showy flowers in dense flat-

topped clusters. Bractlets 8-9 mm long, nar¬

rowly lanceolate, surpassing the calyx. Calyx

6-7.5 mm long with 2-3 mm long setaceous

teeth. Corolla pink to rose, purple or blue,

8-10 mm wide, the tube about 10 mm long

and 1.3-1. 5 times as long as the calyx. Nut¬

lets cylindric, 2-3 mm long. 2n = 30 (Sol-

brig 1959).

Primarily a Great Plains and southwestern

United States species, ranging from South

Dakota to Missouri and Georgia, west to

Texas and Arizona (cf. map in Moldenke

1962a), the only Wisconsin collection

from a Green County roadside (T3N R9E

Sect. 21, May 24, 1969, Maurer 311 OSH).

Hybridization Among Native Species

of Verbena

The Vervains native to Wisconsin (Sect.

VERBENA) are all wide ranging sympatric

species interfertile with one another. All are

diploid, with x = 7 (2n=14). Ecological

factors, no doubt, were responsible for iso¬

lating species and preventing hybridization

in pre-settlement times. Now, however, Ver¬

vains are common in many disturbed sites,

especially where human activities have

eradicated some or all of the native flora,

as for example in pastures. Of the three

Verbena hybrids in Kansas studied by Poin¬

dexter (1962), all were found in over-

grazed pastures, roadsides or waste places.

Hybridization is enhanced not only by

their tolerance of disturbance, with the re¬

sultant breakdown of ecological isolation,

but also by their extended flowering periods

(2 Vi to 3Vi months), allowing any two spe-

Fig. 4. Middle stem leaves of Verbena urticifolia, V. bracteata, V. stricta and V. simplex

and their hybrids native to Wisconsin.

Fig. 5. Middle stem leaves of Verbena urticifolia, V. hastata and V. stricta and their hybrids

native to Wisconsin.

1979]

Tans and litis — Preliminary Reports — Verbenaceae

91

cies to flower contemporaneously, and by

the wide variety of pollinators from many

insect orders.

Even though first generation hybrids are

relatively common in Verbena, the vast

majority of Vervains encountered in the field

are not hybrids. Introgression, apparently, is

limited, for the distinctness of the parental

species has not been reduced as, for exam¬

ple, in Tradescantia, Iris, or Helianthus. As

the parental species themselves are well

adapted to disturbed sites, such introgres-

sants may have no competitive advantage,

and, due to partial sterility, may have rela¬

tively short survival.

The following six hybrids occurring in

Wisconsin often can be recognized by inter¬

mediate leaf shape (Figs. 4, 5), pubescence,

flower size, and inflorescence, as well as re¬

duced fertility, and the nearby presence of

their parental species.

Verbena X deamii Moldenke

Beam’s Vervian Map 8.

Intermediate between V. bracteata and

V. stricta. Ascending to procumbent, re¬

sembling V. bracteata in habit, but with

larger, less laciniate, and reticulate leaves,

larger flowers, and bractlets usually not

foliaceous, the spikes more slender than in

V. stricta, often poorly fruited, the leaves,

stems, branches and inflorescences softly

villous.

Known from the central United States

where the parental species are sympatric, in

Wisconsin, from Brown County (“Bay Set¬

tlement” July 19, 1883, Schuette s.n., DS),

this erroneously attributed by Moldenke

(1963a) to Pierce County, perhaps because

the town of “Bay City” is located there.

Verbena X engelmannii Moldenke

Engelmann’s Vervain Map 8.

Intermediate between V. hastata and V.

urticifolia. Resembling V. urticifolia but

with denser, often poorly-fruited spikes usu-

aly with overlapping calyces, and blue to

purple flowers, the leaves coarsely serrate to

biserrate, larger than those of V. hastata,

the leaves, stem, branches and inflorescences

from nearly glabrous to evenly pubescent.

Corolla tube and calyx intermediate in

length, and pollen fertility reduced (Poin¬

dexter 1962).

Occurring in northeastern United States

where the parental species are sympatric, in

Wisconsin known from some 17 collections

from moist open woods, roadsides, stream

edges and wet meadows.

Verbena X illicita Moldenke

Bastard Vervain Map 9.

Intermediate between V. stricta and V.

urticifolia . Resembling V. urticifolia in habit,

the spikes denser with some of the mature

calyces overlapping, often poorly fruited,

but similar to V. stricta in its densely pu¬

bescent leaves, stems, branches and inflores¬

cences. Nutlet length and ribbing and calyx

and corolla lengths intermediate, the pollen

fertility greatly reduced (Poindexter 1962).

Known primarily from the central United

States where the parental species are sym¬

patric, in Wisconsin only from Crawford

(A. R. Moldenke 995, KANU, U), Iowa

(A, R. Moldenke 1003 KANU) (Moldenke

1965) and Winnebago Counties (Oshkosh,

July 30, 1909, M. K. Clemens s.n., POM)

(Moldenke 1964a).

Verbena X moechina Moldenke Map 9.

Intermediate between V. simplex and V.

stricta. Resembling V. simplex in habit, with

oblanceolate leaves, but with a dense, short

pubescence on leaves and stem, as well as

smaller flowers and shorter bractlets, often

poorly fruited. i

In Wisconsin known only from Rock

County “on north side of St. Lawrence Ave.,

1.4 miles west of Paddock Road, TIN

RUE Section 36,” ( Souter & Rice 1692,

July 31, 1972 BELC UWJ), ( Tans 1431,

August 23, 1975 WIS), a small population

in roadside gravel and in an abandoned

limestone quarry, growing with V. stricta

and such prairie plants as Asclepias verticil-

92

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

lata, Andropogon gerardi, Eragrostis specta-

bilis. Kuhnia eupatorioides, Ratibida pin-

nata, and Solidago nemoralis . While the

other parent, V. simplex, could not be found

nearby, a notation on the sheet of V. simplex

collected from Rock County ( Furnish s.n.,

July 24, 1971, UWJ) by W. Rice suggests

that it may be mislabeled, its probable col¬

lection site being a small gravel pit in Sect.

36, Newark Township, this, in fact, the site

of V. x moechina.

Verbena X perriana Moldenke

Perry’s Vervian Map 10.

Intermediate between V. bracteata and V.

urticifolia. Sometimes a large, diffuse plant,

stouter than V. bracteata and erect when

young, often decumbent at maturity, the

leaves much larger than in V. bracteata,

laciniate, the larger often with two basal

lobes characteristic of V. bracteata, the

bracteata, the bractlets equalling to surpass¬

ing the calyx.

Known from eastern and central United

States where the parental species are sym-

patric, in Wisconsin from Sauk Co. (Bara-

boo in 1861, Hale s.n., WIS); Dane Co.

(1858, Shears s.n., WIS; Madison in

1860’s?, Hale s.n., WIS); Waukesha Co.

(Aug. 22, 1891, Dunlap 9163, MIL); and

Marquette Co. (roadside, T14N R10E

NW1/4 Sect. 31, Oct. 3, 1958, litis 12,361,

WIS). According to Moldenke (1964b),

known from La Crosse Co. (La Crosse, July

20, 1887, L. H. Pammel s.n., ISC); Lafay¬

ette Co. (Shulsburg, July 19, 1883, Man¬

ning s.n., NY); and Milwaukee Co. (/. S.

Douglas s.n., WJC) .

Verbena X rydbergii Moldenke

Rydberg’s Vervian Map 10.

Intermediate between V. stricta and V .

hastata and resembling either parent. Leaves

wider and more deeply serrate than V. has¬

tata, but reticulate and velutinous beneath,

fruits irregularly produced. Intermediate in

leaf length/width ratio, nutlet length, cor¬

olla and calyx lengths with pollen fertility

greatly reduced (Poindexter 1962).

Occurring in central and eastern United

States where the parental species are sym-

patric, in Wisconsin known from some 15

collections from river bottom forests, dry

prairies and, most frequently, roadsides and

pastures.

2. Phyla Lour. Fog-fruit

A small, mostly tropical New World

genus of 10 species with only the follow¬

ing in Wisconsin:

1. Phyla lanceolata (Michx.) Greene

Fog-fruit Map 1 1 .

Lippia lanceolata Michx.

Lippia lanceolata Michx. var.

recognita Fern. & Grisc.

Herb, perennial. Stems creeping, to 8 dm

long, simple or branched at the base and

rooting at the nodes, occasionally with as¬

cending flowering branches, evenly pubes¬

cent with short malpighiaceous (attached at

their center) hairs. Leaves 2-8 cm long,

6-30 mm wide, lanceolate to narrowly ovate,

sharply serrate above the middle, evenly

pubescent on both surfaces. Heads dense,

borne singly or in pairs from middle and

upper leaf axils on 4-12 cm long pedun¬

cles, at first globose, later to 15 cm long and

cylindrical. Flowers sessile, solitary, in the

axils of short bractlets 2-3 mm long. Calyx

membranous, 2 mm long, compressed and

keeled, 2-fid. Corolla white or flushed with

purple, 2.5-3 mm long, salverform the slen¬

der tube cylindrical, surpassing the calyx,

the limb oblique, somewhat 2-lipped, 4-

lobed. Stamens included to barely exserted.

Fruit globose, 1-1.2 mm long, dividing into

2 nutlets. 2n = 32 (Smith 1966); 2n = 36

(Lewis 1961).

Widespread across the United States and

northern Mexico, in Wisconsin south of the

Tension Zone, a “tropical element” of open

habitats such as sunny, dry to wet sands or

1979]

Tans and litis— Preliminary Reports — V erhenaceae

93

silt along major streams, most frequent along

the lower Wisconsin and Mississippi Rivers

on gravel bars and sandy mudflats, upstream

as far as Columbia and Pierce Counties, also

along the lower Kickapoo River, Galena

River in Lafayette County, Sugar and Rock

Rivers and Fox and Wolf Rivers in Winne¬

bago and Waupaca Counties, occasionally

on bare lake shores.

Flowering from the second week of July

through the first week in October, fruiting

from late July through October.

Literature Cited

Beale, G. H. 1940. The genetics of Ver¬

bena. I. J. Genetics 40:337-358.

Correll, D. S., and M. C. Johnston. 1970.

Manual of the Vascular Plants of Texas.

Texas Research Foundation, Renner. 1881 p.

Cronquist, A. 1968. The Evolution and

Classification of Flowering Plants. Hough¬

ton Mifflin Co., Boston. 396 p.

Curtis, J. T. 1959. The Vegetation of Wis¬

consin. University Wis. Press, Madison.

657 p.

Deam, C. C. 1940. Flora of Indiana. Indi¬

ana Dept. Conservation, Indianapolis. 1236

p.

Derman, H. 1936. Cytological study and

hybridization in two sections of Verbena

Cytologia 7:160-175.

Dix, R. L. 1959. The influence of grazing

on the thin-soil prairies of Wisconsin. Ecol¬

ogy 40:36-49.

Federov, A. A. (ed.). 1969. Chromosome

Numbers of Flowering Plants. Leningrad.

926 p. Reprint 1974, Otto Koeltz Science

Publishers, Koenigstein, West Germany.

Fernald, M. L. 1936. Plants from the outer

coastal plain of Virginia. Rhodora 38:414-

452.

— - 1950. Gray’s Manual of Botany.

8th Ed. Amer. Book Co., New York. 1632

P-

Furusato, K. 1940. Polyploid plants pro¬

duced by colchicine. Bot. and Zool. (Tokyo)

8:1303-1311.

Gleason, H. A. 1963. The New Britton and

Brown Illustrated Flora of the Northeastern

United States and Adjacent Canada. The

New York Botanical Garden, New York.

Vol. 3:125-139.

- , and A. Cronquist. 1963. Manual

of Vascular Plants of Northeastern United

States and Adjacent Canada. D. Van No¬

strand Reinhold Co., Princeton. 810 p.

Hartley, T. G. 1966. The Flora of the

“Driftless Area.” Univ. Iowa Stud. Natural

History 21:1-174.

litis, H. H. 1957. Preliminary reports on

the flora of Wisconsin. No. 39. Phrymaceae-

Lopseed Family. Trans. Wis. Acad. 46:105.

Jackson, R. C. 1960. Documented chromo¬

some numbers of plants. Madrono 15:220.

Jennings, O. E. 1953. Wild Flowers of

Western Pennsylvania and the Upper Ohio

Basin. Univ. Pitt. Press, Pittsburgh. Vol. 1,

574 p.

Jones, G. N., and G. D. Fuller. 1955. Vas¬

cular Plants of Illinois. Univ. Ill. Press. Ur-

bana. 593 p.

Junell, S. 1934. Zur Gynaceummorphologie

und Systematik der Verbenaceen und La-

biaten. In Federov, A. (ed.). Chromosome

Numbers of Flowering Plants. 1969. Lenin¬

grad. 926 p.

Kanda, M. 1920. Field and laboratory

studies of Verbena. Bot. Gaz. 69:54-71.

Lewis, W. H. 1961. Index to plant chromo¬

some numbers for 1961. M. S. Cave, ed.

Univ. North Carolina Press, Chapel Hill.

1962.

Lewis, W. H. and R. L. Oliver. 1961. Cyto-

geography and phylogency of the North

American species of Verbena. Amer. J. Bot.

48:638-643.

Moldenke, H. N. 1937. Notes on new and

noteworthy American plants. Revist. Sud-

amer. Bot. 4:15-20.

— - . 1940. Verbenaceous novelties.

Amer. Midi. Nat. 24:750-754.

— - — — 1958. Hybridity in the Verbena-

ceae. Amer. Midi. Nat. 59:333-370.

- - -. 1961-65. Materials toward a mono¬

graph of the genus Verbena . I-XXIX. Phyto-

logia, vols. 8-11 (a series of 29 consecutively

numbered publications) as follows:

1962a. III. 8:175-216.

1962b. V. 8:274-328.

1962c. VI. 8:371-384.

1963a. X. 9:59-97.

94

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

1963b. XII. 9:189-238.

1964a. XIV. 9:351-407.

1964b. XX. 10:271-319.

1964c. XXV. 11:155-213.

1965. XXIX. 11:435-507.

- . 1966-77. Additional notes on the

genus Verbena. Phytologia vols. 13-36. (A

series of 27 consecutively numbered publi¬

cation.)

- . 1971. A Fifth Summary of the

Verbenaceae, Avicenniaceae, Stilbaceae, Di-

crastylidaceae, Symphoremaceae, Nyctan-

thaceae, and Eriocaulaceae of the World as

to Valid Taxa, Geographic Distribution and

Synonymy. Braun-Brumfield, Inc., Ann Ar¬

bor, Mich. 2 Vol., 974 p.

- . 1972-77. A fifth summary of the

Verbenaceae, Avicenniaceae, Stilbaceae, Di-

crastylidaceae, Symphoremaceae, Nyctan-

thaceae, and Eriocaulaceae of the world as

to valid taxa, geographic distribution, and

synonymy. Phytologia 23-36 (a series of 7

consecutively numbered supplements to the

fifth summary).

- . 1974. Notes on new and note¬

worthy plants. LXIX. Phytologia 28:401-

404.

- .and A. L. Moldenke. 1946. A brief

historical survey of the Verbenaceae and re¬

lated families. Plant Life 2:13-98.

Mulligan, G. A. 1961. Chromosome num¬

bers of Caandian weeds. III. Canad. J. Bot.

39:1057-1066.

Noack, K. L. 1937. Die Chromosomenzahlen

einiger Verbena. Cited in Federov, A.

(ed.). Chromosome Numbers of Flowering

Plants. 1969. Leningrad. 926 p.

Parry, C. C. 1852. Systematic catalogue of

plants of Wisconsin and Minnesota made in

connection with the geological survey of the

Northwest, during the season of 1848. In

Owen, David Dale. 1852. Report of a Geo¬

logical Survey of Wisconsin, Iowa, and Min¬

nesota; and Incidentally of a Portion of Ne¬

braska Territory. Pp. 606-622. Lippincott,

Grambo Co., Philadelphia.

Patermann, H. 1935. Beitrage zur Zytologie

der Verbenaceen. Cited in Federov. (ed.).

Chromosome Numbers of Flowering Plants.

1969. Leningrad. 926 p.

Perkins, W. E., J. R. Estes, and R. W. Thorp.

1975. Pollination ecology of interspecific hy¬

bridization in Verbena. Bull. Torrey Bot.

Club 102:194-198.

Perry, L. M. 1933. A revision of the North

American species of Verbena. Ann. Mo.

Bot. Gard. 20:239-362.

Poindexter, J. D. 1960. Documented chro¬

mosome numbers of plants. Madrono 15:

220.

- . 1962. Natural hybridization among

Verbena stricta, V . hastata, and V. urticifolia

in Kansas. Trans. Kansas Acad. Sci. 65:409-

419.

Radford, A. E., H. E. Ahles, and E. R. Bell.

1968. Manual of the Vascular Flora of the

Carolinas. Univ. North Carolina Press,

Chapel Hill. 1183 p.

Smith, E. B. 1966. IOPB Chromosome

numbers reports VII. 1966. Taxon 15:155-

163.

Solbrig, O. T. 1959. Documented chromo¬

some numbers of plants. Madrono 15:51.

Steyermark, J. A. 1963. Flora of Missouri,

Iowa State Univ. Press, Ames. 1725 p.

Swink, Floyd. 1974. Plants of the Chicago

Region. Ed. 3. The Morton Arboretum,

Lisle, Ill. 474 p.

Thorne, Robert F. 1976. A phylogenetic

classification of the Angiospermae. Evol.

Biol. 9:35-106.

Umber, Ray E. 1979. The genus Glandu-

laria (Verbenaceae) in North America. Sys¬

tematic Bot. 4:72-102.

Wadmond, S. C. 1932. Notes from south¬

eastern Wisconsin. Rhodora 34:18-19.

Zufall, C. J., and W. O. Richtmann. 1943,

1944. A Pharmacognostical study of cer¬

tain American species of Verbena. Pharma¬

ceutical Archives 14:60-96; 15:1-9.

A MASS BALANCE OF NITROGEN IN WISCONSIN

Dennis R. Keeney

Department of Soil Science

University of Wisconsin-Madison

A bstract

A mass balance of nitrogen (N) in Wisconsin, with emphasis on the flows

and reservoir sizes in agriculture, was constructed. The model assumes steady

state and is on an annual basis. The data used were available statistics and litera¬

ture values. The calculations show that fixation of N by legumes is an important

component of the agricultural nitrogen balance, estimated at about 2.5 times the

inputs of N from fertilizer. The N fixed by legumes, when cycled through animals

and returned as wastes or when plowed under as residues, is a major input to the

soil organic N, and eventually available N pools. Further evaluation of this phase

of the cycle, particularly the net mineralization estimate, is needed. Such evalua¬

tion is required to improve the efficiency of use of N fertilizer and minimize agri¬

cultural impacts on the environment.

Introduction

Nitrogen is a key element in crop produc¬

tion; of the essential elements supplied from

the soil or fertilizers, it is required in the

highest amounts and is the element most

often inadequate (Viets, 1965). Recent in¬

creases in fertilizer nitrogen prices resulting

from rising energy costs have focused at¬

tention on the need for a better understand¬

ing of N in agriculture. Of equal importance

are the potential environmental impacts of

the increased flow of fixed N from man’s

activities. These include: high nitrate in

potable waters, associated with toxicides

from methemoglobinemia (NRC, 1978);

excessive productivity (eutrophication) of

surface waters (Keeney, 1972); formation

of carcinogenic nitrosamines (NRC, 1978);

and ozone depletion in the stratosphere

linked to production of nitrous oxide during

denitrification (Crutzen and Ehalt, 1977).

Evaluation of human influence on the

flows and storage of N in the environment

is difficult because of the numerous diffuse

sources of N, and its many biological and

chemical conversions. A first and important

step in this evaluation is the construction of

a materials (mass) balance. A mass balance

permits study of the behavior of an element

or compound in a defined system. This is

contrasted to the more common pollution

monitoring methods, which are expensive

and do not detect serious impacts until after

they have occurred. The mass balance

method emphasizes interactions between

components in the system and thus facili¬

tates prediction of future impacts. Further,

it serves as an organizational tool and per¬

mits evaluation of information gaps and data

needs.

The mass balance approach also has for¬

midable problems. Usually the data base is

insufficient, and verification by independent

method is difficult. The degree of aggrega¬

tion of various pools and transformations is

also important; usually aggregation increases

as the complexity of the system increases.

For example, a global mass balance for N

would have much higher degree of aggrega¬

tion than a confined laboratory system. As¬

signing ecosystem boundaries is also a prob¬

lem. Since N transfers are linked to atmo¬

spheric and water transport vectors, political

boundaries are seldom realistic. However,

they have the dual advantage that the data

base is more readily defined from available

95

96

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

YEAR

Fig. 1. Nitrogen fertilizer sales in Wisconsin,

1930-1975. Data from Wisconsin Department of

Agriculture, October 1975.

statistics, and practical decisions on material

flows are often made at the political level.

My colleagues and I often deal with dif¬

ficult questions about the sources, fate and

impact of N in Wisconsin agriculture. The

exercise reported here is an attempt to con¬

struct a meaningful statewide N mass bal¬

ance at low cost. Data are from published

statistics and the scientific literature. A

steady-state system is assumed, and the flows

are on an annual basis.

The State

Wisconsin is largely a rural state, with a

mixed cash grain, dairy, vegetable crop, and

general farm-based agriculture. The total

area is 14.64 X 106 ha, with 54 percent in

farmland. In 1974, there were 105,000

farms, averaging 75.6 ha. The population

Table 1. Area in corn grown for grain, rate of

application of nitrogen fertilizer, crop yield, and

nitrogen removal in grain for Wisconsin, 1964-

1974.

was 4.6 million, with a heavy concentration

around Milwaukee and its surrounding

counties. About 179,000 workers are em¬

ployed on farms. In 1973, gross farm cash

receipts were $2,540 million, with dairy

products accounting for $1,198 million and

livestock products for $690 million of the

gross income (Wisconsin Agricultural Sta¬

tistics, 1975).

Nitrogen Inputs

Fertilizer

The use of nitrogen fertilizer has increased

rapidly in Wisconsin in the past decade

(Fig. 1). In 1974, Wisconsin used 127

X 106 kg, about 1.5 percent of the U.S.

consumption of 8.5 million metric tons.

The area in corn grown for silage, about

400,000 ha, has remained roughly constant

in recent years. However, the area devoted

to corn that is grown for grain increased

more than 59 percent in the past decade,

and the average fertilizer N rates have more

than doubled (Table 1). The statewide

average yields of corn (about 80 percent of

the corn grain is raised in the southern and

eastern portions of the state) reflect the ef¬

fects of weather, however, more than they

do N fertilizer rates. For example, the sum¬

mer of 1970 was hot and dry, and corn

yields suffered from lack of moisture. In

1974, a late spring coupled with an early

killing frost resulted in a short growing sea¬

son and low yields. Hence, in 1974 (the

year I chose for this balance) N removal by

corn was below average.

Nitrogen Fixation

Symbiotic N fixation, mainly by alfalfa,

is a major part of Wisconsin’s agricultural

N budget (Table 2), but the amount of N

added to the soil when an alfalfa stand is

plowed under is difficult to estimate. There

are no data on accumulation of N by plant

roots for stands of alfalfa more than 6 to

9 months old. Several long-term experi¬

ments with crop rotations under conditions

found in Wisconsin indicate that an alfalfa

1979]

Keeney — Mass Balance of Nitrogen

97

Table 2. Nitrogen removal by legumes

in Wisconsin.

“From estimates in Boone and Welch (1972),

Johnson et ah (1975), Welch (1972), and Wis¬

consin Agricultural Statistics (1975).

b Largely red clover and timothy mixture.

meadow will supply the equivalent of about

120 to 135 kg/ha of fertilizer N (Rohweder

and Powell, 1973; Shrader and Pierre,

1966). The availability of N in legume resi¬

dues ranges from 28 to 50 percent of the

equivalent amount of fertilizer N (Shrader

and Pierre, 1966). Assuming that the up¬

take of fertilizer N is about 70 percent, the

residues of N in alfalfa amount to about

170 kg/ha. Since these residues accumu¬

late over the lifetime of the average crop

rotation (about 3 years in Wisconsin), I

estimated the input of alfalfa-N to the soil

reserve at 56 kg N/ha-yr and of mixed hay

systems at half this amount (28 kg N/ha-

yr). Soybeans fixed about two-thirds of the

N they remove (Johnson, Welch and Kurtz,

1975).

The rate of nonsymbiotic N fixation in

various ecosystems was estimated by Burns

and Hardy (1975) as follows: cropland,

5 kg N/ha-yr; grassland, 15 kg N/ha-yr;

forests, 10 kg N/ha-yr. Fixation in wetlands

and surface waters was estimated at 20 kg

N/ha-yr and includes fixation in sediment as

well as that by algae in the water (Mac-

gregor and Keeney, 1975).

Precipitation

A recent survey of N in precipitation in

Wisconsin (Hoeft, Keeney and Walsh,

1972) gave a weighted average flux of 3.5

kg N03-N, 3.5 kg NH4-N, and 7.5 kg par¬

ticulate N (total N minus inorganic N) per

hectare per year. Particulate N is assumed

to be derived largely from windblown soil

material, and thus its net contribution is

zero. Undoubtedly some of the N in precipi¬

tation comes from agricultural activities

through volatilization from fertilizers and

manure. In this sense, the estimated inputs

of nitrogen in Wisconsin precipitation in¬

volve some double accounting (that is, an

input is counted twice).

Total Inputs

Total inputs of N for various Wisconsin

land uses in 1974 are summarized in Table

3. Agriculture (categories 1-5) received

585 X 106 kg N/ha-yr, with 21 percent

from precipitation, 62 percent from biologi¬

cal fixation, and 10 percent in feed supple¬

ments.

Nitrogen Removal By Crops and

Nitrogen Transfers

Nitrogen removal by nonleguminous crops

in Wisconsin was calculated on the basis of

N contents of various crops (Agricultural

Research Service, 1971; Boone and Welch,

1972; Welch, 1972) and on the basis of

reported yields (Wisconsin Agricultural Sta¬

tistics, 1975). An estimated 144 X 106 kg

of N were removed in 1974 (Table 4).

Some of the crop N is transferred out of

the state in grain and vegetables and in ani¬

mal products; some is used in the state and

reappears in the production of wastes by the

animal and human populations. A large

amount of protein concentrate is imported.

Estimates for these N transfers are given in

Tables 4 and 5.

Analysis of Input, Output and

Transfer Errors

Of the N inputs listed in Table 3, the

quality of the estimates of fertilizer and pre¬

cipitation sources can be considered good.

In contrast, there is undoubtedly some error

in the estimates of symbiotic and nonsymbi-

98

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

Table 3. Estimated nitrogen inputs to Wisconsin, 1974.

“Estimates from Agricultural Research Service (1971), Boone and Welch (1972), Welch

(1972), Wisconsin Agricultural Statistics (1975). Corn grain, 138 kg N/1,000 kg; com silage,

3.3 kg N/1,000 kg; sweet corn. 35 kg N/1,000 kg; wheat, 31 kg N/1,000 kg; barley and rye,

18 kg N/1,000 kg; oats, 20 kg N/1,000 kg; potatoes, 2.8 kg N/1,000 kg.

1979]

Keeney — Mass Balance of Nitrogen

99

Table 5. Nitrogen as human or animal wastes

produced annually in Wisconsin.

a From estimates in National Research Council

(1972) and Taiganides and Stroshine (1971).

otic N fixation. However, inputs from non-

symbiotic N fixation are relatively small,

particularly in the agricultural sector. Inputs

from symbiotic fixation have been estimated

from data on crop harvests (Table 2). Fixa¬

tion of N by alfalfa is by far the largest com¬

ponent of such inputs. A range of ±10 per¬

cent in the N content of alfalfa is likely in

practice, giving a range in removal of 197 ±

20 X 106 kg N. A further source of error is

the allocation of residual N from alfalfa, an

important value because it affects the size

of the pool of available N in the balance, as

discussed later. Alfalfa also takes up some

N from the soil; accounting for this transfer

would reduce the net value of the contribu¬

tion by legumes through fixation. The esti¬

mate used in Table 2 is based on observa¬

tion of the relative yield response of corn

and alfalfa fields the first year after plowing.

The actual value will depend on weather

and farm management practices and it is

therefore difficult to estimate possible error.

Nitrogen production and transfer esti¬

mates (Table 4) were made by multiplying

accepted values for the N content of each

product by the appropriate production, con¬

sumption, or export data, available from

Wisconsin Agricultural Statistics (1975).

Similarly, N waste production values were

obtained by multiplying accepted per-unit N

excretion values by the populations. The N

transfer and waste production values (Ta¬

bles 4 and 5) are thought to be reliable,

with the possible exception of the data for

pastures (Table 4), where N uptake was

equated to N fixation. This estimate would

be low if the N in the excreta of grazing

cattle were considered, and high if consider¬

able immobilization of N were occurring in

the topsoil. On balance, the estimate seems

low, since uptake of 22 kg N/ha (40 X 10G

kg N/1.8 X 106 ha of pastureland; Tables

3 and 4) is equivalent to the production of

only about 1500 kg (1.5 percent N in herb¬

age) /ha of dry matter. However, since ex¬

tensive data on the productivity of Wiscon¬

sin pastures are lacking, no better estimates

are available.

Agricultural Nitrogen Balance

In Wisconsin

Estimates of the sizes of pools and fluxes

of N in Wisconsin agriculture are presented

in Figure 2. Two models were examined for

their applicability in this exercise. One in¬

volved consideration of the total soil N as

the receptor of the annual N flux, assuming

steady state in this pool. The pool of or¬

ganic N in the soil in Wisconsin was esti¬

mated, assuming 0.15 percent N (4000 kg

N/ha in 20 cm of soil), for a total of 22,500

X 106 kg N. This pool is several orders of

magnitude larger than any of the N inputs

(see Table 3 and Fig. 2). Thus, the steady-

state assumption with respect to soil organic

N would have to be extremely accurate to

permit estimation of an output by differ¬

ence (Kohl, Shearer and Vithayathil, 1977).

The problem can be overcome in part by

considering the pools of available N and

slowly-available organic N in the soil to be

dynamic. The steady-state assumption can

be applied with greater accuracy to the soil

available N pool because this would not be

expected to vary markedly on an annual ba¬

sis. This pool, which is essentially the inor¬

ganic N in the root zone, also is of the same

magnitude as the annual N fluxes. Experi-

100

Wisconsin Academy of Sciences, Arts and Letters

[Vol. 67

Fig. 2. The flow of nitrogen in Wisconsin agriculture in 1974.

Pool sizes and fluxes are expressed in 106 kg N.