: “JOURNAL OF THE Po ae
N EW ENGLAND BOTANICAL CLUB

CON TENTS:

aN vegetation rn fidtistie Siva of a created ate in southeastern r
New Hampshire. Donald J. Padgett and Garrett E. Cha ae .

me Observations on Seen in ie irianthophora (Orchidaceae).
Z Susan A. Willia ams

mee Caleareous fens se ‘western New England and adjacent ee York State. : .

a Chromosome ete oe some ae. American species BF Amis (Bew-

_ laceae). Guillermo. Restrepo and C Ca

; Evolution of a flora: early Connecticut valley Botanists. ‘€ doh Burk sn
~ New England Notes: : *

Lie oe eis (Poaceae) i in ‘New England, Jeanne E. Anderson and

Notes on the Rbode e Island Flora. Richard Te. Champlin oy ae By
_Desmonema wrangell (Ag. e Bornet et Flahault, anew ee for Maine
L.C. Co it

The biology’ age lec Onorch A ihe Powitaca oleracea E- complex i in
No orth. America: Ene lignes}, Matthew
“Reviews:

eee Sey Cetin Ah ae (WES Sie Ba A

- Rare \

Les Onan de Ve.

or ay-<

Index to Vi lum

‘THE NEW. ENGLAND BOTANICAL CLUB
P.O, Box 1897, Lawrence, Kansas 66044

1 i

eva

22 Divinity Avenue, Cambridge, Massachusetts 02138

VOL.96 — : January, 1994 No. 885

- 1 iN a8 land Bienes Club, Ine :

ce 2 Divinity Avenue, Cambridge, Massachusetts 02138

-RHODORA
a “GORDON, DeWOLF, IR., Editor-in-Chief ae

_ Associate Editors
DAVID S. CONANT
ay eee - LISA A. STANDLEY _ eae
ee peanaee assn 0035. 4902) Pics four times a ‘year. Ge
oo = April, July, and October) by The New England Botanical Club, 810

East 10th St., Lawrence, KS 66044 and printed by Allen Press, docs
1041 New Hampshire St., Lawrence, KS 66044. Second class postage

pending at Lawrence, KS. POSTMASTER: Hae address Waals too |

-RHODORA, BO: Box 1897, Lawrence, KS 660:

= RHODORA i is a journal of botany devoted eee to North pnenes
_ Authors are encouraged to submit manuscripts of scientific papers
and notes relating to the systematics, distribution, ecology, Ppaleo-
botany, or conservation biology of this or floristically related areas.

es SUBSCRIPTIONS: $40.00 per calendar year, net, postpaid, in funds ees

payable at par in the United States currency. Remittances payable io

66044- 8897.

DORA. Send to RHODORA, P.O. Box 1897, Lawrence, KS .

| MEMBERSHIPS: _ Regular $35: Family $45, Application fe form printed ae

o ‘kee VOLUMES AND SINGLE COPIES: Aare available: : informa: :
tion and prices will be furnished upon request to the Treas urer.

as aes CHANGES: In order to receive the next. nee a a . ae
HODORA, changes must be received by ss business Ges yoer Ue

re first day of January, ‘April, July, or Octo
INFORMATION FOR CONTRIBUTORS: Inside back cover, rd January
so eee me nd to: Visidon® DeWolf,

ee MP ch a eee Managing Editor, RHODORA

125 Long Hill Road
3 Mest Brookfield, MA 01585.

@ This paper meets the requirements of ANSI/NISO 239.48- 1992 (Permanence of Paper).

RHODORA

URNAL OF THE

NEW ENGLAND BOTANICAL CLUB
Vol. 96 January 1994 No. 885

RHODORA, Vol. 96, No. 885, pp. 1-29, 1994

A VEGETATION AND FLORISTIC ANALYSIS OF A
CREATED WETLAND IN SOUTHEASTERN
NEW HAMPSHIRE

DONALD J. PADGETT AND GARRETT E. Crow

ABSTRACT
The artificial creation of wetlands Has pecome a common means to mitigate
the effects of wetland alteration due t p t t. The einai of diverse,
functional plant ial and challenging p a wetland creation

project. This study analyzes plant community structure of a seven year old, a
tificially created wetland. One hundred and es Naseulge plant saan are eee
umented from the study it ew Hampshire
and the first state record for an introduced weedy species.

The computer program TWINSPAN used species abundance data to classify
the vegetation into seven cover types. The three plant associations of the open

effusus-Phalaris arundinacea cover type, and a Carex ‘stricta cover type. Each
cover type and associated habitat was described and delineated.

Key Words: floristics, wetlands, wetland mitigation, rare plants

INTRODUCTION

Encompassing such diverse systems as freshwater marshes,
swamps, bogs, sloughs, ponds and lakes, rivers and streams, coast-
al freshwater tidal marshes, and saltmarshes, wetlands collectively
constitute one of the most productive biomes on Earth (Good et
al., 1978; Mitsch and Gosselink, 1986). Beyond the obvious aquatic
productivity and associated food-chain support, wetland ecosys-
tems are vital in performing numerous other ecological functions
(Goodwin and Niering, 1974; Burke et al., 1988; Larson, 1988).
The rather late recognition of wetland values and functions over
the past two decades seems to have been catalyzed by the alarming

|

2 Rhodora [Vol. 96

rate at which wetlands are being destroyed. Most important, the
vital role wetlands play in the overall quality of the environment
has finally been recognized and a variety of efforts are being im-
plemented to provide a means for their conservation.

One strategy that has been developed to conserve wetlands is
termed wetland mitigation. Wetland mitigation reduces or com-
pensates for the negative impacts on aquatic ecosystems imposed
by development by artificially creating or restoring wetland areas
(Savage, 1986). Mitigation has become an integral part of wetland
protection and conservation policies. Wetland mitigation at-
tempts to restore to an area those functions of a wetland that are
impacted or destroyed.

The primary goal of wetland creation or restoration projects
typically aims at restoring lost wetland functions. Wetland cre-
ation efforts additionally aim at creating an area that is at least
the same size as the area lost and the establishment of certain
wetland vegetation types (Lowry, 1990; Kruczynski, 1990).

The establishment of hydrophilic vegetation is crucial in wet-
land mitigation projects if they are to imitate successfully natural
wetland ecosystems. Wetland vegetation provides some of the
important functions and values inherent in wetlands (Good et al.,
1978: Mitsch and Gosselink, 1986) and creates suitable habitats
for wildlife (Weller, 1978). The establishment of diverse, func-
tional wetland plant communities is a challenging part of miti-
gation projects (Padgett and Crow, in press).

The objective of the present study is to provide a detailed
vegetation analysis and total floristic inventory of a seven year
old, artificially created, freshwater wetland located in southeastern
New Hampshire.

Site History

During the winter of 1985-86, the Hospital Corporation of
America (HCA) created a freshwater wetland under the Section
404 guidelines of the U.S. Army Corps of Engineers to compensate
for the filling of a wetland at the construction site of a new regional
hospital in Portsmouth, New Hampshire. The land area desig-
nated for the creation of the compensatory wetland was an aban-
doned gravel pit located approximately 3 miles from the hospital
site in the southern end of Portsmouth, NH (Rockingham Co.)
(Michener et al., 1986). The size of the mitigation area (ca. 13

1994] Padgett and Crow— Wetland 3

acres) was similar to the size of the area lost due to the hospital
construction.

Wetland construction began in the fall of 1985 and continued
through the winter of 1986. Basins were excavated and graded to
form a sinuous configuration of open water pools and marsh areas
(Michener et al., 1986). The wetland soils and peat at the new
hospital site (impacted wetland) were excavated and transported
to the new wetland creation site. This material was stockpiled
through the winter and eventually spread as a 6-12” top dressing
over the excavated basins.

The primary strategy to vegetate the newly created basins was
to allow for the natural colonization of wetland plants. Therefore
there was no direct planting of wetland plant stock. The muck
top dressing was expected to serve as a created ‘“‘seed bank’’ nat-
urally holding large quantities of plant seeds and vegetative pro-
pagules (Michener et al., 1986).

Groundwater is the primary hydrological input into the site.
Groundwater flow is westward, discharging into the adjacent for-
ested swamp (Garlo, 1993). However, during the first summer
after construction, beavers dammed the small outflow channel.
Consequently, the water level was raised approximately two feet
above the design elevation.

Site Description

The H.C.A. Portsmouth wetland is a created freshwater wetland
consisting of primarily herbaceous aquatic plants. Some tree and
shrub areas exist on several upland islands and pool margins.
Open water areas form sinuous configurations around these is-
lands. The maximum depth of the open water areas is approxi-
mately two meters. The created wetland is surrounded by upland
forest communities on the northeastern and southern sides, and
a wetland forest community on the northwestern side. A dis-
turbed, open, sandy area with few woody plants surrounds the
wetland on the southeast side.

MATERIALS AND METHODS

Vegetation Analysis

Vegetation data were collected during July of 1992. Sampling
was done using a systematic sampling method (Mueller-Dombois

4 Rhodora [Vol. 96

and Ellenberg, 1974). Thirteen transects were established across
the wetland at 20 meter intervals. A single 0.5 x 0.5 m quadrat
was placed at every five meters of transect for a total of 284 sample
plots. Trees, shrubs and herbs occurring on the banks and upland
sites on the islands throughout the wetland were not included in
the sampling.

A visual estimate of percent cover was recorded to estimate
abundance. The primary focus of the vegetation analysis was
vascular plant species. However, because of their relative abun-
dance Chara cf. vulgaris and Ricciocarpus natans, two non-vas-
cular aquatic plants, were included in the vegetation analysis.
Cover was defined as the vertical crown or shoot-area projection
per species in relation to the reference area (Mueller-Dombois
and Ellenberg, 1974).

The data were analyzed using TWINSPAN (Two-Way Indi-
cator Species Analysis), a fortran classification analysis program.
The program first constructs a classification of samples and then
based on this classification, constructs a classification of species
according to their ecological preferences (Hill, 1979).

TWINSPAN groups the quadrat samples according to the flo-
ristic similarity of its members. Using these groups as a basis,
species are clustered to form hierarchial dichotomies. There are
three ordinations involved in determining a dichotomy: 1) a pri-
mary ordination, made by a method of reciprocal averaging; 2)
a refined ordination, using differential species determined from
the primary ordination; and 3) an indicator ordination, using
highly preferential species (Hill, 1979). Corresponding to the
Braun-Blanquet cover-abundance scale of 0, 5, 25, 50 and 75
percent (Mueller-Dombois and Ellenberg, 1974), abundance data
of TWINSPAN ‘“‘pseudo-species”’ were classified at cut levels of
1, 2, 3, 4 and 5 respectively. Ultimately, a two-way table is gen-
erated grouping similar quadrat samples across the top and similar
species down the left side (Padgett, 1993).

A vegetation map (Figure 2) was prepared based on cover types
of quadrats along the 13 transects to give a visual estimate of the
pattern of distribution of the seven cover types.

Floristic Study

An inventory of the vascular plants of the site was undertaken
during the 1992 field season. Voucher specimens were collected

1994] Padgett and Crow— Wetland 5

284
141 143
Open water lypcs Emergent types
SA 87 134
Chara cover type Eleocharis smaltii
cover Lype
68 19 55 79
8 } 2 ] Typha latifolia
cover lype cover type cover type
65 14
Juncus effusus Carex stricta
Phataris arundinacea cover type
cover lype

Figure 1. TWINSPAN analysis showing the seven plant cover types classified
at four hierarchical levels. Numerals represent the number of quadrats included
in each dichotomy.

and deposited in the Hodgdon Herbarium (NHA) at the University
of New Hampshire.

Nomenclature follows Aquatic and Wetland Plants of North-
eastern North America (Crow and Hellquist, in press) and Manual
of Vascular Plants of Northeastern United States and Adjacent
Canada (Gleason and Cronquist, 1991).

RESULTS AND DISCUSSION

The construction of the H.C.A. Portsmouth wetland has created
a unique environment, allowing for a natural colonization and
emergence of wetland plant species. The various micro-habitats
formed by the gradual slopes of the basins and integration of
raised islands provide a heterogeneous ecosystem. As a result,
concentric vegetation zones have developed according to the eco-
logical affinities of species characteristic of natural wetland sys-
tems. The vegetation patterns of the site are typical of those
described for inland freshwater wetlands (Mitsch and Gosselink,
1986; Hammer, 1992; Weller, 1978; Curtis, 1959). Generally, for
most inland wetlands, sedges (Carex) and rushes (Juncus) occupy

6 Rhodora [Vol. 96

the wet-meadow areas gradually passing into the shallow water
areas colonized by Cattails (7ypha), Bulrush (Scirpus) and Pick-
erel Weed (Pontederia cordata). The deeper, open water areas are
colonized by submerged species (Utricu/aria) and floating leaved
species (Potamogeton and Nuphar).

Plant Cover Types

The wetland vegetation was classified using TWINSPAN into
seven cover types (Figure 1). Data from 284 quadrats were an-
alyzed and 67 species were clustered at four hierarchical divisions
into the seven cover types which could be visually recognized in
the field.

The first dichotomy grouped the 284 quadrats into two major
groupings, with 141 quadrats representing the open water cover
types, and 143 quadrats representing the emergent cover types.
At the second level of clustering, 54 quadrats of open water types
were distinguished as the Chara cover type, with 87 quadrats
remaining to be sub-clustered. At the third level 68 quadrats
represented the Potamogeton natans cover type and 19 quadrats
defined the Potamogeton pusillus cover type.

TWINSPAN grouped the 143 quadrats in the emergent cover
type group into four cover types (Figure 1). At the second level
the Eleocharis smallii cover type was recognized (9 quadrats). At
the third level the 7ypha latifolia cover type was distinguished
(55 quadrats), while the Juncus effusus-Phalaris arundinacea (65
quadrats) and Carex stricta (14 quadrats) cover types were not
defined until the fourth level.

The clustering of the quadrats at the first divisional level clearly
correlates with the two primary habitats designed for the site
creation. The open water cover types occupy the areas where the
water 1s too deep to be colonized by emergent species. Most of
the open water areas were vegetated by floating-leaved, free-float-
ing and submerged species. However, some areas remained de-
void of vascular plants. For instance, one open pool was heavily
colonized by filamentous algae. Although three cover types were
recognized as the open water areas, these were not characterized
by distinct vegetation zones. Instead these cover types occurred
in a mosaic pattern.

The emergent cover types represent the marsh or wet meadow
regions designed for the wetland. These areas are defined by a

1994] Padgett and Crow— Wetland 7

topographic gradient beginning in the shallow fringe areas of the
open pools and extending back into the higher areas of the marsh.
Consequently, the emergent cover types are better characterized
by zonational patterns when compared to the open water cover
types. The emergent cover types are the most floristically diverse
areas of the wetland and are characterized by a predominance of
emergent species. However, the free-floating Lemna minor was
found in greater frequencies and abundance within the emergent
cover types than in the open water cover areas.

OPEN WATER COVER TYPES

Potamogeton natans Cover Type

The Potamogeton natans cover type is the largest plant asso-
ciation of the open water types (Figure 2), defined by the 68
quadrats clustered by TWINSPAN. Of the seventeen species that
characterize the community, the floating-leaved P. natans is dom-
inant with 35% cover and 90% frequency (Table 1). The sub-
merged Najas minor and Utricularia gibba are sub-dominants
with 11% and 9% covers, respectively, although U. gibba has a
significantly higher percent frequency. The delicate morphology
of U. gibba does not allow for high percent coverage.

The cover type is easily discernible by the presence of broad
floating leaves of Potamogeton natans that form an expansive
cover on the surface of the water. This floating-leaf layer was so
extensive in certain portions of the wetland, particularly in the
northwestern pool, that few open surface water areas were present.
Although the cover of P. natans was very extensive, the mean
percent cover of P. natans throughout the entire cover type was
only 35%. In areas that were sparsely populated with P. natans
submerged species, such as Najas minor, Utricularia gibba and
P. pusillus, were frequent.

Chara Cover Type

The Chara cover type is one of the larger plant associations of
the open water types (Figure 2). The community can be charac-
terized by 15 species, but the macrophytic alga Chara cf. vulgaris
dominates with 81% cover and 98% frequency (Table 2). Plants

[Vol. 96

Rhodora

Cor Ff
~ = VY. s
S = = O = a
~ 2 & Peg TY
; Ms cs 3 £ S88 E
{ ie ye Se ee es Se
ita a i? a SOmta QgessoO
: Sseeetatettath A ie BH neat ‘} i NI reneg aia SS
: 4: ?) 3 4 v a N eas \N
REE aes 8s i N a Ea He eh \

ae
‘aeietenaete eatatetaiets: “
sdtausencunausanans 3

39,

Ae
NN
ane

a34 gist

ee
— iS x 5 +2
_ ee saan

The distribution of the seven plant cover types of the H.C.A. Ports-

Figure 2.
mouth created wetland. CT

Cover type.

1994] Padgett and Crow— Wetland 9

Table 1. Mean percent cover and percent frequency for species in the Pota-
mogeton natans cover type (species with cover <1 percent are excluded).

Species % Cover % Frequency
Potamogeton natans 35 90
ajas minor 11 15
Utricularia gibba 9 82
otamogeton pusillus 15
Eleocharis acicularis 2 6
Lemna minor l 19
Ludwigia palustris 1 10

of Chara were commonly observed as monotypic colonies or with
few other species. However, Utricularia gibba and Potamogeton
natans are sub-dominant components of the community. The
remaining species of the community (Table 2), with the exception
of P. pusillus, are more characteristic of emergent areas.

e weedy nature of Chara has allowed it to successfully col-
onize the open water areas of the site. More typically associated
with hard water sites, Chara has been found to dominate deeper
waters of lakes producing large amounts of biomass (Rickett,
1921). The Chara community exists in the deeper pools of the
wetland, where the cover of floating leaved species is relatively

Potamogeton pusillus Cover Type

This cover type is characterized by the submerged Pondweed
species, Potamogeton pusillus (Table 3). This plant association

Table 2. Mean percent cover and percent frequency for species in the Chara
cover type (species with cover <1 percent are excluded)

Species % Cover % Frequency
Chara cf. vulgaris 81 98
Utricularia gibba 11 56
otamogeton natans 9 37
Eleocharis acicularis 7

Juncus articulatus
Typha latifolia

poms pam pest
& R40

10 Rhodora [Vol. 96

Table 3. Mean percent cover and percent frequency for species in the Pota-
mogeton pusillus cover type (species with cover <1 percent are excluded).

Species % Cover % Frequency
Potamogeton pusillus 36 95
Pontederia cordata 12 21
Eleocharis acicularis 5 26
Potamogeton natans of 47
Utricularia gibba 3 42
Lemna minor ] 37
Najas minor l 11
Potamogeton amplifolius 1 5

covers the smallest portion of the open water areas and was rep-
resented by only 19 quadrats. The Potamogeton pusillus cover
type occurs in scattered areas within the open water portions of
the wetland, most typically close to the emergent vegetation cover
types. Two of the largest areas where this cover type is predom-
inate are shown in Figure 2. Although, this cover type is primarily
composed of submerged and free-floating species, the emergent
Pontederia cordata is a sub-dominant component.

EMERGENT COVER TYPES

Juncus effusus-Phalaris arundinacea Cover Type

The Juncus effusus-Phalaris arundinacea cover type is the most
floristically diverse cover type of the entire wetland. Of the 55
species occurring in the cover type, Juncus effusus and Phalaris
arundinacea are co-dominant, with a collective cover value of
41% (Table 4). This cover type generally extends from the upland
shores or islands to the edge communities of the Typha latifolia
cover type (Figure 2). However, some pool margins are directly
surrounded by this cover type where Typha does not dominate.

The areas of the Juncus effusus-Phalaris arundinacea cover type
are rarely inundated for any extended period of time. Most por-
tions remain wet to damp, allowing for a wide range of wetland
species to exist. This mostly low-growing, herbaceous plant com-
munity has some woody components such as A/nus incana ssp.
rugosa, Acer rubrum and Cornus spp. occupying various portions.
Unique among cover types, both annuals and perennials are well
represented within this association. The Cyperaceae is best repre-

1994] Padgett and Crow— Wetland 11

Table 4. Mean percent cover and percent frequency for species in the Juncus
alia arundinacea cover type (species with cover <1 percent are ex-
cluded).

Species % Cover % Frequency

Juncus effusus 21 66
Phalaris arundinacea 20 83
Typha latifolia 6 31
Scirpus cyperinus 6 22
Carex scoparia 4 28
Alnus incana ssp. rugosa 4 9
Eleocharis acicularis 3 14
Lemna minor 2 28
Carex lurida 2 14
Acer rubrum 2 8
Cicuta bulbifera l 28
Galium palustre l 20
Polygonum amphibium 1 14
Onoclea sensibilis | 14
Carex vulpinoidea l 14
Lythrum salicaria 1 11
Polygonum punctatum l 11
Juncus articul 1 9
Typha latifolia 1 9
Ricciocarpus natans 1 8
Eleocharis obtusa | 6
Spiraea tomentosa 1 6
Eleocharis smallii 1 5
Vaccinium macrocarpon ] 3
Cornus stolonifera 1 3

ornus amor ! 2
Potamogeton pusillus ] 2
Solidago ru ] 2
Spiraea latifolia ] 2
Thelypteris palustris ] 2

sented in this community. The Juncus effusus-Phalaris arundinacea
cover type is similar in general physiognomy to the wet-meadow
zones or sedge-meadows discussed by Mitsch and Gosselink
(1986), by Hammer (1992) and by Weller (1978).

Typha latifolia Cover Type

The Common Cattail cover type represents the second largest
community association of the emergent types. Of the 27 species
that occur in the cover type, Typha /atifolia is dominant repre-

12 Rhodora [Vol. 96

Table 5. Mean percent cover and percent frequency for species in the Typha
latifolia cover type (species with cover <1 percent are excluded).

Species % Cover % Frequency
Typha latifolia 25 78
Juncus effusus 6 35
Lemna minor 4 82
Phalaris arundinacea 3 18
Ludwigia palustris 2 29
Utricularia gibba 2 22
Juncus articulatus 2 By)
Eleocharis acicularis ] 13
Najas minor ] 14
Acer rubrum l 2

sented by a 25% cover (Table 5). Sub-dominant components are
Juncus effusus and Lemna minor with covers of six and four
percent, respectively.

Typha latifolia and Lemna minor have the highest frequencies
of occurrence in the cover type. These two species were frequently
found associated with each other throughout the wetland, usually
with L. minor fronds stranded on the wet mud and rhizomes or
floating on the still water in between the 7. /atifolia plants. In a
survey of southern Wisconsin Cattail marshes, L. minor had a
frequency of 100%, although its presence was reported as insig-
nificant (Curtis, 1959).

The Typha latifolia cover type is the most easily recognizable
and discernible of the emergent cover types. This cover type can
be quite expansive at certain portions of the wetland, or exist as
narrow bands (Figure 2). Several isolated, nearly monotypic, Ty-
pha “islands” are also present. The dense rhizomatous network
produced by Typha latifolia, characteristic of the species (Mc-
Naughton, 1966), encroaches upon the deeper open water from
the very shallow edges. The 7ypha /atifolia cover type is similar
to the 7ypha-dominated zone typical of freshwater marshes dis-
cussed by Mitsch and Gosselink (1986) and by Weller (1978).

The cover type is characterized by species of many growth
forms. Important emergent species include Typha latifolia, Juncus
effusus and Phalaris arundinacea. Submergent species include
Utricularia gibba and Najas minor, while the free-floating Lemna
minor, and the amphibious Ludwigia palustris are also compo-
nents of the cover type.

1994] Padgett and Crow— Wetland 13

ble Mean percent cover and percent frequency for species in the Carex
stricta cover type (species with cover <1 percent are excluded).

Species % Cover % Frequency
Carex stricta 58 100
Juncus effusus 6 64
Phalaris arundinacea 6 57
Eleocharis acicularis 6 7
Cicuta bulbifera 4 43
Typha latifolia 3 14
Galium palustre 1 21
Triadenum fraseri l 7
Spiraea tomentosa l 7
Typha angustifolia l 7
Polygonum sagittatum l z
Carex lurida 1 7

Carex stricta Cover Type

The Tussock Sedge cover type is one of the smaller emergent
plant associations of the wetland. Of the 21 species associated
with this cover type Carex stricta is dominant, with a 58% cover
(Table 6). The raised, wide-spreading tussocks formed by C. stric-
ta create a unique, easily discernible cover type. The Carex stricta
cover type is scattered throughout the wetland, occupying small
areas (Figure 2).

Eleocharis smallii Cover Type

The Eleocharis smallii cover type occupies the smallest area in
the wetland (Figure 2). The cover type was defined by TWIN-
SPAN by the clustering of only nine quadrats and includes 14
species. The dominant species is E. smallii, with 27% cover (Table
7). Sub-dominant species are Typha angustifolia and Eleocharis
acicularis. This cover type is scattered throughout the northern
portions of the emergent wetland and is also represented by small
discontinuous areas intermixed with the Typha Jatifolia and Jun-
cus effusus-Phalaris arundinacea cover types.

This cover type is the most floristically unique portion of the
wetland. Species typically associated with peatlands, such as Dros-
era intermedia, Vaccinium macrocarpon and Sphagnum sp. are
part of this community. Dulichium arundinaceum and Viola lan-
ceolata, also components of the community, were not observed

14 Rhodora [Vol. 96

Mean percent cover and mean frequency for species in the Eleocharis
smallii cover type (species with cover <1 percent are excluded)

Species % Cover % Frequency
Eleocharis smallii 27 67
Typha angustifolia 23 56
Eleocharis acicularis 12 22
Vaccinium sical 9 11
Triadenum fr 4 56
Asclepias incarnata 2 11
Lysimachia terrestris l 33
Lemna minor | 33
Spiraea tomentosa ] 22
Lythrum salicaria ] 11

growing anywhere else in the wetland. Other species common to
this portion of the wetland but observed infrequently elsewhere,
are Asclepias incarnata, Triadenum fraseri, Bidens frondosa and
Aster racemosus. The soil of the area is very peaty and is slightly
inundated in spring and early summer, but as water table drops
later in the summer it becomes merely damp.

RARE PLANT DOCUMENTATION

Two plants listed by Storks and Crow (1978) and the New
Hampshire Natural Heritage Inventory (DRED, 1987) as rare for
New Hampshire occur at the H.C.A. created wetland, Spargani-
um eurycarpum (threatened) and Potamogeton foliosus (endan-
gered). The occurrence of a third, non-indigenous species, Najas
minor, constitutes the first record of this plant in New Hampshire.
Although rare in New Hampshire, both Sparganium eurycarpum
and Potamogeton foliosus are widely distributed throughout the
United States (Crow and Hellquist, in press). Both species rep-
resent new records for the city of Portsmouth.

Sparganium eurycarpum (Giant Bur-reed) is reported as com-
mon along the coastal plain and in western portions of New En-
gland (Crow and Hellquist, 1981). There are currently seven doc-
umented stations for Sparganium eurycarpum in New Hampshire
(Natural Heritage Program, pers. comm.). In a study of the rare
plants of coastal New Hampshire, S. eurycarpum was reported
from the towns of New Castle, Rye and Hampton (Dunlop and
Crow, 1985). In the study site, S. eurycarpum occurs in shallow

1994] Padgett and Crow— Wetland Be:

water areas along the edges of the pools. The plant was more
common along the western edge of the wetland. The water level
of this area dropped greatly later in the growing season, leaving
the plants growing in damp soils.

In New England Potamogeton foliosus (Leafy Pondweed) is
reported as being common in calcareous waters of northeastern
Maine, Vermont, western Massachusetts and Connecticut (Hell-
quist and Crow, 1980). Storks and Crow (1978) reported that P.
foliosus was known from a single station in New Hampshire, in
the town of Columbia, Coos County. There are currently five
known stations of P. foliosus in New Hampshire (Natural Heritage
Program, pers. comm.). Potamogeton foliosus occurs in the open
water areas of the study site. Plants were intermixed with the very
similar looking P. pusillus, therefore the exact extent of its oc-
currence is hard to estimate. Other open water macrophytes as-
sociated with P. foliosus were P. natans and Chara cf. vulgaris.
Because Pondweeds are heavily consumed by waterfowl (Martin,
1951; Fassett, 1957), P. foliosus was most likely introduced into
the study site by ducks or other waterfowl.

A third new record for the flora of Portsmouth, the adventive
Najas minor, is also the first record for New Hampshire (Padgett
and Crow, 1993). A European native, Najas minor (Eutrophic
Water-nymph) was first reported in America in the Hudson River,
in 1934 (Clausen, 1936). The North American range of this in-
troduced species has continued to spread steadily in the eastern
United States. By 1968, N. minor was known from New York,
Pennsylvania, Illinois, West Virginia, Tennessee, Alabama and
Florida (Merilainen, 1968). Its present North American distri-
bution extends from New England and New York west to Illinois,
south to Florida, Mississippi and Arkansas (Haynes, 1979; Crow
and Hellquist, in press). The first report for this plant in New
England was in 1965, occurring in Lake Champlain at New Ha-
ven, Vermont. Two reports of the species occurring in Bershire
County, Massachusetts followed in 1974 (Hellquist, 1977). Hell-
quist and Crow (1980) report the distribution of N. minor to be
infrequent in waters of extreme western New England. The present
discovery of N. minor in New Hampshire possibly represents the
northeastern-most range extension to date of the species for the

At the H.C.A. created wetland, Najas minor occurs in the open
water areas. A component of the strictly aquatic macrophyte com-

16 Rhodora [Vol. 96

munity of the site, N. minor is commonly associated with Pota-
mogeton pusillus, P. natans, Utricularia gibba and Chara cf. vul-
garis Later in the summer, many plants became fragmented and
consequently were floating at the water surface and near the pool
margins.

Najas minor is an annual, producing large quantities of seed.
The most important dispersal agent for N. minor is waterfowl
who feed heavily on the entire plant. Interestingly, the pattern of
distribution of N. minor is very similar to the Atlantic flyway
pattern of migrating waterfowl (Merilainen, 1968). Therefore it
is believed that N. minor was probably introduced into the site
by waterfowl. Najas minor has also been reported to invade re-
cently constructed artificial lakes and ponds (Wentz and Stuckey,
1971).

FLORA OF H.C.A. PORTSMOUTH CREATED WETLAND

The vascular flora of the H.C.A. Portsmouth created wetland
consists of 104 species, belonging to 69 genera and 45 families.
Fifty-six species are dicots and forty-four species are monocots.
The dominant families are the Asteraceae, Cyperaceae and Po-
aceae.

POLYPODIOPHYTA

DRYOPTERIDACEAE
Onoclea sensibilis L. Sensitive Fern
Abundant; throughout the wetland on wet soil. Padgett 59,
145.

OSMUNDACEAE
Osmunda regalis L. Royal Fern
Uncommon; on western edge of wetland growing on em-
bankment. Padgett 138.

THELYPTERIDACEAE
Thelypteris palustris Schott Marsh Fern
Occasional; along the shores of the wetland but locally
abundant along the western shore. Padgett 92, 160, 231.

1994] Padgett and Crow— Wetland 17

EQUISETOPHYTA
EQUISETACEAE
Equisetum arvense L. Common Horsetail
Common; throughout the wetland on wet or moist soils.
Padgett 56, 81.

MAGNOLIOPHYTA
DICOTYLEDONS
ACERACEAE
Acer rubrum L. Red Maple
Common; throughout the emergent wetland as seedlings,
but more occasional as young trees. Padgett 87.

APIA
een bulbifera L. Bulbiferous Water-hemlock
Common; throughout the wetland growing in emergent
areas. Padgett 68, 158, 183, 217.
Sium suave Walter Water-parsnip
Uncommon; growing in wet emergent areas. Padgett 216.

AQUIFOLIACEAE
Ilex verticillata (L.) A. Gray Winterberry
Occasional; growing in wooded areas within the wetland.
Padgett 235.

ASCLEPIADACEAE
Asclepias incarnata L. subsp. pulcra (Ehrh. ex. Willd.)
Woodson Swamp Milkweed
Occasional; throughout the emergent wetland, but locally
abundant in the western shore area. Padgett 86, 121.

ASTERACEAE
Aster lanceolatus var. simplex (Willd.) A. G. Jones Eastern
Lined Aster
Uncommon; along western side of wetland. Padgett 237.
Aster racemosus Ell. Small White Aster
ncommon; growing around the western shore area. Pad-
gett 252.

18 Rhodora [Vol. 96

Bidens connata Muhl. Purplestem Beggar-ticks
Ocassional; in emergent areas, but more common in the
western area. Padgett 218, 234, 245, 246, 253, 256, 261.
Bidens frondosa L. Devil’s Beggar-ticks
Uncommon; in emergent area of the western shore. Padgett
233

Eupatorium perfoliatum L. Boneset
Occasional; growing along wet shores and emergent areas
of the wetland. Padgett 198.
Euthamia graminifolia (L.) Nutt. Common Flat-topped
Goldenrod
Occasional; growing along the shores of the wetland. Pad-
gett 222.
Solidago rugosa Miller subsp. rugosa Wrinkle-leaved Gold-
enro
Occasional; growing along the shores and scattered in the
emergent areas. Padgett 18].

BALSAMINACEAE
Impatiens capensis Meerb. Orange Touch-me-not
Occasional; throughout the northern emergent areas and
outermost shores. Padgett 175.

BETULACEAE
Alnus incana (L.) Moench. subsp. rugosa (Du Roi)
Clausen Speckled Alder
Frequent; growing along outer shores and wooded areas.
Padgett 76, 101.

CALLITRICHACEAE
Callitriche verna L. Water Starwort
Uncommon; growing in shallow water areas along the
shores. Padgett 58, 187.

CLUSIACEAE
Hypericum boreale (Britt.) Bickn. St. John’s-wort
Uncommon, growing in emergent marsh areas. Padgett

228.
Hypericum canadense L. Narrow-leaved St. John’s-wort

1994] Padgett and Crow— Wetland 19

Uncommon; growing in emergent marsh areas. Padgett
204

Hypericum dissimulatum Bickn. St. John’s-wort
Locally abundant; in western open marsh area and along
northwestern shore. Padgett 227.

Hypericum ellipticum Hook. St. John’s-wort
Common; throughout the emergent areas and in shallow
water. Padgett 139, 156.

Hypericum mutilum L. Dwarf St. John’s-wort
Occasional; throughout the emergent marsh. Padgett 201.

Triadenum fraseri (Spach) Gleason Marsh St. John’s-wort
Common; throughout the emergent areas but more fre-
quent in the open area of the western side. Padgett 94, 146,
195, 260.

CORNACEAE

Cornus amomum subsp. obliqua (Raf.) J.S. Wilson Silky Dog-

wood
Uncommon; growing along outer southwestern shore.
Padgett 104.

Cornus stolonifera Michx. Red-stemmed Dogwood
Occasional; growing in emergent and wooded areas, but
more common in the western open marsh area. Padgett
69 457,

DROSERACEAE
Drosera intermedia Hayne Sundew
Locally abundant; growing in wet open area of western side
of the wetland. Padgett 95, 147.

ERICACEAE

Lyonia ligustrina (L.) DC. Maleberry
Occasional; in wooded areas throughout the wetland. Pad-
gett 230.

Vaccinium corymbosum L. Highbush Blueberry
Uncommon; in wooded areas on western side of wetland.
Padgett 155.

Vaccinium macrocarpon Aiton Cranberry
Locally abundant; in open area of western shore and grow-

20 Rhodora [Vol. 96

ing with Typha /atifolia in northern area of wetland. Pad-
gett 90, 140.

HALORAGACEAE
Proserpinaca palustris L. Common Mermaid-weed
Uncommon; in open water areas close to the shores. Pad-
gett 69, 125.

LAMIACEAE
Lycopus americanus Muhl. Bugle-weed
Occasional; throughout the emergent marsh. Padgett 205.
Lycopus uniflorus Michx. Bugle-wee
ommon; throughout the emergent marsh. Padgett 240.
Scutellaria galericulata L. Common Scullcap
Uncommon; growing in wet mud of northwestern shore.
Padgett 208.

LENTIBULARIACEAE
Utricularia gibba L. Creeping Bladderwort
Abundant; in open water areas throughout the wetland.
Padgett 57, 133.
Utricularia minor L. Lesser Bladderwort
Uncommon; growing in very shallow pools floating on
surface between other vegetation. Padgett 250.

LYTHRACEAE
Lythrum salicaria L. Purple Loosestrife
Common; throughout the emergent marsh areas. Padgett
75, 114, 124.

MYRICACEAE
Myrica gale L. Sweet Gale
Uncommon; in emergent wetland at island shore edge of
northern area. Padgett 19].

NYMPHAEACEAE
Nuphar variegata Durand Yellow Water-lily
Occasional; in open water areas close to shores. Padgett
126.

1994] Padgett and Crow— Wetland 21

ONAGRACEAE
Epilobium palustre L. Marsh Willow-herb
Uncommon; growing in emergent marsh areas. Padgett

Ludwigia palustre (L.) Elliott. Common Water-purslane
Frequent; growing throughout the open water areas and on
the muddy bottoms of the emergent areas. Padgett 63, 129.

POLYGONACEAE
Polygonum amphibium L. var. emersum Michx. Water
Smartweed
Common; throughout the marsh on the elevated islands
and emergent areas. Padgett 148, 180, 238, 247.
Polygonum arifolium L. Tearthumb
Occasional; growing with Polygonum sagittatum. Padgett

193.

Polygonum lapathifolium L. Pale Smartweed
Uncommon; growing on a beaver den in northern area.
Padgett 172.

Polygonum punctatum Ell. var. punctatum Water Smartweed
Frequent; throughout the open emergent wetland. Padgett
186, 190.

Polygonum sagittatum L. Arrow-vine

mon; at the edges of the open water areas. Padgett
179,36).

Rumex crispus L. Curly Dock
Uncommon; at the northern edge of the wetland. Padgett
177.

PRIMULACEAE
Lysimachia terrestris (L.) BSP. Bulbil-loosestrife
Frequent; throughout the emergent wetland. Padgett 152,
219, 239, 242.

ROSACEAE
SPIRAEA LATIFOLIA (Aiton) Borkh. Meadowsweet
Uncommon; throughout emergent areas of the wetland.
Padgett 85.
SPIRAEA TOMENTOSA L. Hardhack

22 Rhodora [Vol. 96

Common, throughout emergent areas, but more frequent
in open western area. Padgett 91, 137.

RUBIACEAE
Galium palustre L. Marsh Bedstraw
Common; growing in the lower parts of the emergent veg-
etation throughout the wetland. Padgett 83.

SALICACEAE
Salix nigra Marshall Black Willow
Uncommon; growing along the northern and western outer
shores. Padgett 200.

SAXIFRAGACEAE
Penthorum sedoides L. Ditch-stonecrop
Occasional; growing in open emergent areas and cattail
stands of western side of wetland. Padgett 207.

SCROPHULARIACEAE
Agalinis purpurea (L.) Pennell Purple Gerardia
Uncommon; growing in marsh area along the western lim-
its of the wetland. Padgett 214.
Mimulus ringens L. Monkey Flower
Occasional; growing in emergent marsh areas throughout
the northern half of the wetland. Padgett 165, 178.

SOLANACEAE
Solanum dulcamara L. Bittersweet
Occasional; throughout the emergent wetland. Padgett 62,
ee

URTICACEAE
Boehmeria cylindrica (L.) Sw. Bog Hemp
Uncommon; growing along outer edge of wetland in north-
west area. Padgett 257.

VERBENACEAE
Verbena hastata L. Blue Vervain

1994] Padgett and Crow— Wetland 23

Uncommon; growing along the outer edges of the wetland.
Padgett 203.

VIOLACEAE
Viola lanceolata L. Strap-leaved Violet
Locally abundant; growing in moist soil of western open
marsh area. Padgett 255.

MONOCOTYLEDONS

ALISMATACEAE
Alisma triviale Pursh. Northern Water-plantain
Occasional; growing in muddy substrate throughout the
wetland. Padgett 96, 166.
Sagittaria latifolia Willd. Common Arrow-head
Occasional; in shallow water at southwest end of wetland.
Padgett 97, 176.

CYPERACEAE

Carex comosa Boott Long-hair Sedge
Common; throughout the emergent areas of the wetland.
Padgett 67, 107, 144.

Carex lurida Wahl. Lurid Sedge
Common; Throughout the emergent areas of the wetland.
Padgett 98, 168.

Carex lupulina Willd. Hop Sedge

ncommon; in emergent area along western perimeter.

Padgett 258.

Carex pseudocyperus L. Cyperus-like Sedge

ommon; throughout the emergent areas of the wetland.

Padgett 132.

Carex scoparia Schk. Broom Sedge
Frequent; throughout the emergent areas of the wetland.
Padgett 70, 109.

Carex stricta Lam. Tussock Sedge
Occasional; along the outer shore edges and emergent marsh
areas. Padgett 72.

Carex vulpinoidea Michx. Fox Sedge

24 Rhodora [Vol. 96

Frequent; throughout the emergent areas of the wetland.
Padgett 143, 154.

Cyperus strigosus L. Umbrella Sedge
Uncommon; growing along western and northern shores.
Padgett 196.

Dulichium arundinaceum (L.) Britt. Three-way Sedge
Locally abundant; growing in open marsh area of western
shore. Padgett 215.

Eleocharis acicularis (L.) R.&S. Needle-rush

ommon; growing along muddy shores and throughout
the emergent marsh. Padgett 99, 173.
Eleocharis obtusa (Willd.) Shultes Blunt Spike-rush
ommon; growing throughout emergent marsh areas and
shallow pools. Padgett 169.

Eleocharis smallii Britt. Small’s Spike-rush
Occasional; throughout the emergent areas, but more lo-
cally abundant growing out of shallow pools of western half
of wetland. Padgett 73.

Eleocharis tenuis (Willd.) Shultes var. tenuis Slender Spike-

rush
Occasional; in wet mud open areas or below Typha latifolia
stands. Padgett 170.

Rhynchospora capitellata (Michx.) Vahl. Beakrush
Uncommon; growing in wooded area of northwestern re-
gion of wetland. Padgett 226.

Scirpus atrocinctus Fern. Black-girdle Bulrush

ncommon; growing along northern and southern outer
limits of wetland. Padgett 159.

Scirpus cyperinus (L.) Kunth. Wool Grass
Frequent; Growing in emergent areas throughout the wet-
land. Padgett 161, 189.

Scirpus hattorianus Makino
Uncommon; growing along outer limits of the wetland.
Padgett 150, 225.

Scirpus pungens Vahl. Three-square Bulrush
Uncommon; growing in wet soil along the northwestern
shore. Padgett 194.

Scirpus tabernaemontanii K.C. Gmel. (= S. validus Vahl) Great

Soft-stem Bulrush
Frequent; throughout the emergent wetland. Padgett 112,
122.

1994] Padgett and Crow— Wetland 25

IRIDACEAE
Tris versicolor L. Northern Blue Flag
ncommon; growing along the edges of open pools around
the islands. Padgett 115, 136.

JUNCACEAE
Juncus articulatus L. Jointed Rush
Frequent; throughout the emergent wetland. Padgett 131,
135,403,171; Joe.
Juncus canadensis J. Gay Canada Rush
Occasional; throughout the emergent wetland. Padgett 223,
224

Juncus effusus L. Soft Rush
Abundant; throughout the emergent wetland. Padgett 66,
60, 111, 117,

LEMNACEAE
emna minor L. Common Duckweed
Abundant; floating on surface throughout the open water
and Typha stands. Padgett 116.
Wolffia columbiana Karst. Water-meal
Uncommon; floating on surface in open water with Lemna
minor at north end of study site. Site Record.

NAJADACEAE
Najas minor Allioni. Eutrophic Water-nymph
Common; in open water with Chara cf. vulgaris Padgett
167, 236.

POACEAE

Calamagrostis canadensis (Michx.) Beauv. Reed Bent Grass
Common; in emergent areas around the perimeter of the
wetland. Padgett 120, 153, 251.

Echinochloa crusgalli (L) Beauv. Barnyard Grass
Uncommon; in emergent marsh areas. Padgett 213, 248.

Glyceria canadensis (Michx.) Trin. Rattlesnake Manna Grass

ncommon; along outer edge of the western shore. Padgett

259.

26 Rhodora [Vol. 96

Leersia oryzoides (L.) Sw var. oryzoides Rice Cut Grass
Common; in emergent marsh areas. Padgett 241, 249.
Panicum villosissimum Nash. Panic Grass
Uncommon; growing in open areas adjacent with upland
islands. Padgett 192.
Panicum rigidulum Nees. Panic Grass
Uncommon; in emergent marsh areas. Padgett 244.
Phalaris arundinacea L. Reed Canary Grass
Frequent; growing along the outer edges of the wetland.
Padgett 60, 65, 93, 108, 119, 162.
Poa palustris L. Fowl Bluegrass
Occasional; in open marsh areas. Padgett 174.

PONTEDERIACEAE
Pontederia cordata L. Pickerel-weed
Occasional; in shallow open water areas close to the shore,
forming dense stands. Padgett 102, 123.

POTAMOGETONACEAE

Potamogeton amplifolius Tuckerman Big-leaf Pondweed
Uncommon; in open water area with Potamogeton natans
at southeast end of wetland. Padgett 127.

Potamogeton foliosus Raf. var. foliosus Leafy Pondweed
Frequent; in open water areas throughout the wetland.
Padgett 164.

Potamogeton natans L. Floating Pondweed
Abundant; throughout open water areas. Padgett 57, 74,
130

Potamogeton pusillus L. var. tenuissimus Mert. and
Koch Slender Pondweed
Frequent; in open water areas throughout the wetland.
Padgett 118, 128.

SPARGANIACEAE
Sparganium eurycarpum Engelm. Giant Bur-reed
Common; in shallow water areas throughout the marsh,
but more frequent along the western edge of the wetland.
Padgett 53, 88.

1994] Padgett and Crow— Wetland 2]

TYPHACEAE
Typha angustifolia L. Narrow leaved Cattail
Occasional; growing intermixed with Typha latifolia along
pool edges and in thick stands. Padgett 77, 142.
Typha latifolia L. Common Cattail
Abundant; throughout the wetland, often encroaching the
open pools. Padgett 78.

In conclusion, the vegetation component of this relatively young
created wetland appears to be quite heterogeneous with a diverse
floristic composition comprising seven discernible vegetation cover
types. One hundred and four vascular plant species are docu-
mented from the wetland including two species listed as rare for
New Hampshire and the first state record for an introduced weedy
species. What makes this site remarkable is that although it is
floristically diverse there was no direct planting of wetland plants
during its construction. The challenge of establishing diverse,
functional wetland plant communities at this site appears to have
been met essentially by the transplantation of wetland muck. This
point should be stressed for future considerations in mitigation
project planning.

ACKNOWLEDGMENTS

We would like to thank Drs. A. Linn Bogle and Thomas D.
Lee, University of New Hampshire, for their helpful comments
on the manuscript. We are grateful to Kathleen M. McCauley for
her assistance in the field. Albert Garlo, Normandeau Associates
Inc., kindly provided mitigation information on the H.C.A. site.
Dr. Clotilde Straus, former chair of the Portsmouth Conservation
Commission, was instrumental in the selection of the site for
suitable wetland mitigation and providing additional buffer to an
important Atlantic White Cedar Swamp conservation area. She
also kindly provided us with her floristic information on the study
site which she prepared prior to any mitigation efforts.

This paper is Scientific Contribution Number 1823 from the
New Hampshire Agricultural Experiment Station.

LITERATURE CITED
Burke, D. G., E. J. Meyers, R. E. TINER AND H. GROMAN. 1988. Protecting

nontidal wetlands. eene Planning Assoc., Planning Adv. Serv. Report
No. 412/413.

28 Rhodora [Vol. 96

CLAUSEN, T. R. 1936. Studies in the genus Najas in the northern United States.
Rhodora 38: 333-345.

Crow, G. E. AND C. B. HELLQuisT. 1981. Aquatic vascular plants of New En-
gland: part 2. Typhaceae and Sparganiaceae. N. H. Agric. Exp. Sta. Bull. 517.

. In press. Aquatic and Wetland saa of Northeastern North
America. University of Wisconsin Press, Madiso

Curtis, J.T. 1959. The Vegetation of Wisconsin: An Orsnain of Plant Com-
munities. University of Wisconsin Press, Madison,

DRED. 1987. Protected plants of New Hampshire. el Heritage Inventory.
Res-N 306 plant listing. Dept. of Resources and Economic Development,

DunNLop, D. AND G. E. Crow. 1985. Rare plants of coastal New Hampshire.
Rhodora 87: 487-501

Fassett, N.C. 1957. A Manual of Aquatic Plants. The University of Wisconsin
Press, Madison, WI.

Garo, A. 1993. Wetland creation/restoration in gravel pits in New Hampshire,

. 54-62. In: F. J. Webb, Jr., Ed., Proceedings of the 19th Annual Conference

on eta Restoration and Creation. Hillsborough Community College,
Tamp

GLEASON, H AA ND A. CronguisT. 1991. Manual of vascular plants of north-
eastern United States and adjacent Canada, 2nd ed. New York Botanical
Garden, Bronx, NY

Goon, R. E., D. F. WHIGHAM AND R. L. Simpson. 1978. Freshwater Wetlands:
Ecological Processes and Management Potential. Academic Press, New York,

Soa. R. H. AnD W. A. NierinGc. 1974. Inland wetlands: their ecological
role and eee status. Ecol. Soc. Bull. 55(2): 2-6

Hammer, D. A. 1992. Creating Freshwater Wetlands. Lewis Publishers, Inc.,
Ann Arbor, MI.

Haynes, R. R. 1979. Revision of North and Central American Najas (Naja-
daceae). SIDA 8(1): 34-56

HELLQulisT, C. B. 1977 Obsenvationser some uncommon vascular aquatic plants

in ae England. Rider 79: 445-452.

ND G. E. Crow. 1980. Aquatic vascular plants of New England: part
l. Zosteraceae, Potamogetonaceae, Zannichelliaceae, and Najadaceae. N. H.

Hitt, M. O. 1979. TWINSPAN: a FORTRAN program for arranging multi-
variate data in an ordered two-way table by classification of the individual
and attributes. Department of Ecology and Systematics, Cornell University.
Ithaca, NY.

Kruczynskl, W. L. 1990. Options to be considered in preparation and evalu-
ation of mitigated plans, pp. 555-570. Jn: J. A. Kusler and M. E. Kentula,
Eds., Wetland Creation and Restoration: The Status oh the Science. Island
Press, Washington, DC.

Larson, J.S. 1988. Wetland creation and restoration: an outline of the scientific
perspective, pp. 73-79. In: J. Zelazny and J. S. Feierabend, Eds., Proceedings
of a Conference: se our Wetland Resources. National Wildlife Fed-
eration, Washin

Lowry, D. J. 1990. Restoration and creation of palustrine wetlands associated

1994] Padgett and Crow— Wetland 29

with riverine systems of the glaciated northeast, pp. 267-280. Jn: J. A. Kusler
and M. E. Kentula, Eds., Wetland Creation and Restoration: The Status of
the Science. Island Piece: Washington, DC.

Martin, A. C. 1951. Identifying pondweed seeds eaten by ducks. J. Wildl.
Manage. 15(3): 253-258.

McNauauton, S. J. 1966. Ecotype function in the 7ypha community-type.
Ecol. Mono. 36(4); 297-325

MERILAINEN, J. 1968. Najas minor All. in North America. Rhodora 70: 161-

5

Micuener, M. C., N. HANSEN AND J. BripGes. 1986. Final plan for mitigating
the loss of wildlife habitat on the site of the proposed hospital construction
in Portsmouth, NH, Hospital Corporation of America. Normandeau Asso-
ciates, Inc, Bedford, NH.

Mitscu, W. J. AND J. G. GossELINK. 1986. Wetlands. Van Nostrand Reinhold,
New York, NY.

MUELLER-Domsois, D. AND H. ELLENBERG. 1974. Aims and Methods of Veg-
etation Ecology. John Wiley and Sons, New York, NY.

Papcett, D. J. 1993. A comparison of created and natural wetlands of south-
eastern New Hampshire: flora and vegetation. M.S. thesis, University of New

AND G. E. Crow. 1993. Some unwelcome additions to the flora of New
Hampshire. Rhodora 95: 348-351.

AND n press. A comparison of floristi mposition and species
richness within and between . and natural wetlands of southeastern
New Hampshire. /n: F. J. Webb, Jr., Ed., Proceedings of the 20th Annual
Conference on Wetland Restoration and Creation. Hillsborough Community
College, Tampa, FL.

RicketT, H. W. 1921. A quantitative study of the larger aquatic plants of Lake
endota, Wisconsin. Trans. Wis. Acad. Sci. Arts Let. 20: 501 —527
Sioa N. 1986. The een predicament. Environ. Manag. 10(3): 319-

Storks, I. M. AND G. E. Crow. 1978. Rare and endangered vascular plant
species in New Hampshire. The New England fear me in cooperation
with the U.S. Fish and Wildlife Service. [Newton Corner, MA.]

WELLER, 7 W. 1978. Management of freshwater marshes fo wildlife, pp. 267-
284

n: R. E. Good, D. F. Whigham and R. L. Sim , Eds., Freshwater
cae Ecological Processes and WManasenient Boe ati Academic Press,
New York, NY.

Wentz, W. A. ANDR. L. Stuckey. 1971. The changing distribution of the genus
coe (Najadaceae) 1n Ohio. Ohio J. Sci. 71(5): 292-302.

DEPARTMENT OF PLANT BIOLOGY
UNIVERSITY OF NEW HAMPSHIRE
DURHAM, NEW HAMPSHIRE 03824

RHODORA, Vol. 96, No. 885, pp. 30-43, 1994

OBSERVATIONS ON REPRODUCTION IN
TRIPHORA TRIANTHOPHORA (ORCHIDACEAE)

SUSAN A. WILLIAMS

ABSTRACT

A small population of the orchid Triphora trianthophora in western Massachu-
setts was investigated over a six-year period (1988 through 1993). Observations
on the flowering habits and seed capsule production are described, as well as
vegetative reproduction by means of tuberoids. Triphora produces an abundance
of short-lived flowers yet very few capsules are initiated. The majority of Tri-
phora’s existence is spent underground reproducing asexually by means of new
tuberoids.

Key Words: Triphora trianthophora, orchid, tuberoids

INTRODUCTION

Triphora trianthophora is a small woodland orchid, elusive and
secretive, appearing abundantly one year and rare or absent for
many succeeding years (Lownes, 1920; Ames, 1948; Sheviak,
1974; Brackley, 1981). Triphora trianthophora also shows syn-
chronous flowering, the majority of its short-lived flowers appear
48 hours after a temperature drop (Brackley, 1981; Keenan, 1986,
1988; Sheviak, 1974). In fact, all species of 7riphora which are
not self-pollinating exhibit gregarious flowering (Dressler, 1981)
apparently to increase their chances of cross-pollination (Luer,
1975). Even so, Triphora trianthophora rarely sets seed (Lownes,
1920; Keenan, 1992).

The most unusual feature of this orchid, however, is its exis-
tence for years at a time in a subterranean, tuberous condition
(Lownes, 1920; Ames, 1948; Zavitz and Gaiser, 1956). Although
the tribe Triphoreae appears to be a relic group with no close
allies, it does share a unique feature with members of the tribes
Orchideae, Diseae and Diurideae in having these root-stem tuber-
oids (Dressler, 1981). I started observing the underground tuber-
olds of Triphora in part because of a statement made by Oakes
Ames in 1948 in that he supposed there was a “... maximum
size for the tubers that bear flowering stems.” Every plant of T.
trianthophora has from one to numerous tuberoids attached by
stolons. These tuberoids are thickened underground storage struc-
tures superficially similar to tubers, but structurally different. As
Dressler (1981) points out, true tubers are not found in orchids.

30

1994] Williams— 7riphora 34

The primary tuberoid contains the apical bud which may form
a new shoot in the growing season. Axillary buds form secondary
tuberoids at the end of slender stolons. These secondary tuberoids
continue to increase in size with the age of the plant, along with
increasing length of the stolon. The stolon tends to grow down-
wards into the leaf litter so that with accumulating litter accounts
for the depth of the primary tuberoids. These secondary tuberoids
on becoming detached from the parent plant form primary tuber-
oids of new plants with the same genotype as the parent.

Most plants, however, do not form flowering shoots but remain
in the leaf litter (pers. obs.), and these plants consist of primary
and secondary tuberoids as seen in Figure 4a and 4b.

This paper presents the observations gathered over six years
on: 1) sexual reproduction in Triphora trianthophora, in particular
the reason for low capsule set; and 2) vegetative reproduction in
Triphora, particularly the underground growth form with possible
correlations to population fluctuations, and the relationship be-
tween primary tuberoid size and the plant’s reproductive status.

METHODS

I became acquainted with 7riphora trianthophora in 1988, when
I began studies on a small population in western Massachusetts
for the Massachusetts Natural Heritage and Endangered Species
Program.

The study site comprises an area of approximately two acres
of northern hardwood forest dominated by Fagus grandifolia and
Acer saccharum. The site is on a southeast facing slope at an
elevation of 900-1100 feet. It is midslope being approximately
300 feet above the Deerfield River basin.

The study area was subdivided into distinct sites where the
plants were located. In 1988 there were 25 sites and in each
succeeding year I found additional sites until there were a total
of 72 sites in 1993, all within the general area. The cumulative
total of vegetative and reproductive plants observed over the six
year period was 1448.

I observed the plants on a daily basis for their entire above-
ground existence of approximately one month, from their first
emergence through the litter, to capsule set. Each site was mon-
itored for the total number of plants; number of vegetative plants;

ae Rhodora [Vol. 96

plants which produced buds but failed to blossom (in all cases
this was due to some type of herbivory); plants which blossomed
but did not initiate a seedpod; and plants which ripened seedpods.
Percent capsule set was determined by the ratio of capsules set
to total buds produced by the population.

While collecting data on each site, any pollinators or floral
visitors were noted, and also whether the plants had undergone
any type of herbivory.

A few plants that did not produce capsules were carefully re-
moved from the litter and the length of their primary tuberoid
was measured. This group included small primary tuberoids which
had never produced any stem; tuberoids with hyaline stems (frag-
ile, translucent stems remaining under the litter not producing
leaves or flowers but with one or more secondary tuberoids at-
tached); vegetative plants (those having above-ground stems with
one small leaflet but no flowers); and plants with one, two, three,
or four flowers.

As I started removing litter I also noted many other tuberoids
in the same vicinity that had produced neither leafy nor flowering
stems. These consisted of a primary tuberoid and one to many
secondary tuberoids. Some had hyaline stems with secondary
tuberoids but had never produced above-ground stems. Many
large tuberoids were also noted which had previously produced
above-ground stems but had not produced any in the current
year. These tuberoids were not measured. Due to the invasive
nature of measuring primary tuberoid length, only a small sample
was measured each year.

A yearly comparison was made of the total number of plants
for the original 25 sites as well as comparing fluctuations within
individual sites.

RESULTS

Observations on Sexual Reproduction

Throughout the six-year study period, the majority of Triphora
trianthophora plant produced flower-bearing stems; only 10% or
less failed to flower in any given year. (Figure 1). Most of the
plants bore one or two flowers; those having three of four were
much less frequent totaling less than 10% in any given year. This
pattern remained relatively constant over six years.

1994] Williams— 7riphora 33

60;
‘
31988
aol 1989
1990
301 m@ 1991
1992

L] 1993

Percent of Total Population

= NN
ee
me AS we. N = E B e SS rare ee <r)
1 flower 2 flowers 3flowers 4 flowers

Plant Reproductive Status

igure 1. Percentages of plants of differing reproductive status in a Triphora
population over a six-year period.

Figure 2 illustrates on a yearly basis the percentages of bud
loss, flowers produced and capsules ripened out of the total bud
population.

The large percentage of buds lost was caused by several factors.
Rarely, a few buds would be underdeveloped and tiny and these
would always drop off prior to flowering. Occasionally, entire
plants had been consumed by chipmunks (their tunnels were found
under clumps of Triphora plants) and no trace of plant or tuber-
oids could be found. Slugs were the most destructive herbivores
of Triphora. Stems were chewed through, usually at litter level,
and the plants would be lying on the ground. Sometimes, they
were chewed in half higher up and occasionally the buds were
half eaten. I noted slugs in the upper litter layers many times and
observed slugs on Triphora plants (including the tuberoids) sev-
eral times.

Also illustrated in Figure 2 is that capsule production relative
to total buds produced is very low. However, out of 110 capsules

34 Rhodora [Vol. 96

fj —buds lost
f§j-flowers produced
[ ]-seedpods ripened

Total Bud Population [%]
b
ro)

*89 ‘90 ’91 °92 ’93

Figure 2. Percentages of buds lost, flowers produced and seedpods ripened out
of the total bud population, compared yearly.

initiated 104 ripened, which is 95%. The other 6 did not ripen
because of loss of the plant due to herbivory.

It appears that few of the flowers are actually pollinated, sup-
ported by the fact that on only two occasions in six years did I
note any floral visitors. One appeared to be a small species of
bumblebee (Bombus) which entered the blossom for a few sec-
onds, then backed out with pollinia attached. The only other
flower visitor I observed was a much smaller and slender bee
probably belonging to the genus Hylaeus. I did not observe any
pollinia removal with its visit, but whether the bee was too small
or the pollinia had already been removed could not be deter-
mined.

Another factor contributing to low capsule set is the fact that
the Triphora blossom is available to pollinators for one day only.
The day following a bloom shows noticeable fading and drooping

1994] Williams— 7riphora 35

2 8+ e e
2.6T
2.47
2.2+ °
E 2.07 ° °
oO +
= 1.87 ee e *
6D 4. ee e
5S 1.67 oe
+ e ° ee
B 1.44
5 ug eee
S 1.24 eee
a + eee e
eB 1.07 °
S 4 e e
: + ee eooo °
& 0.87 es eoe000 ee
a e pie, ;
0.47 e ee0 ee e
O.2+ eeecsee
NS HyS | StP | 1FI 2 Fl 3Fl | 4Fl
a b b Cc d

Reproductive Status

Figure 3. Relationship of primary tuberoid length (cm) to geen status
Sea NS—No Stem; HyS— Hyaline Stem; StP—Sterile Plant; | Fl—one flower;
2 Fl—two flowers; 3 Fl—three flowers; 4 Fl—four flowers. Each dot ee an
individual plant. Asterisks represent the mean of the group. Different letters in-
dicate a significant difference between the means. (Bonferroni adjustment, P =

until, after an average of five days, the blossom falls off the plant.
If the flower is fertilized, the ovary starts to swell and the dried
corolla remains intact.

Observations on Vegetative Reproduction

Although the stimulus for shoot production in Triphora is as
yet unclear, Figure 3 shows the relationship between primary
tuberoid size and type of plant. Plants that consisted of a primary

36 Rhodora [Vol. 96

Table |. One-way analysis of variance of reproductive status. The length data
were log transformed to homogenize the variance and the log of length was then
used as the dependant variable.

Analysis of Variance

Sum-of- Mean-
Source Squares df Square F-Ratio P
Reproductive status 29.766019 5 5.953204 47.560594 0.000000
Error 8.511623 68 0.125171

tuberoid with no shoot development had the smallest primary
tuberoids; those plants with either hyaline stems or sterile leaflets
had slightly larger primary tuberoids; and primary tuberoid size
tended to increase with increasing numbers of flowers per plant.
A one-way ANOVA was done on the data with the log of length as
the dependent variable. The length data were log transformed in
order to homogenize the variances. As seen from the ANOVA table
(Table 1), the developmental classes showed significant variation
in tuberoid length. In order to analyze the variation between
classes, a Bonferroni Adjustment was done. The results (shown
by the letters in Figure 3) indicated which classes varied signifi-
cantly from which other class.

The primary tuberoid is generally the largest and the deepest
down in the litter from which the flowering stem and secondary
tuberoids arise (see Figure 4). The secondary tuberoid closest to
the primary tuberoid is the largest with smaller tuberoids ap-
pearing up the stem. There are many variations in the number
of secondary tuberoids and their arrangement on any individual
plant. Some plants have only one secondary tuberoid, in others
I have counted up to 17. However, all plants have at least one.
Occasionally the secondary tuberoids get quite large before be-
coming separated from the main plant. Several secondary tuber-
oids were even slightly larger than the primary tuberoid. There
doesn’t appear to be a specific size when the stolon disintegrates
between the main plant and the secondary tuberoids. More likely,
they separate as a result of physical forces since they are fragile
and near the litter surface. The tuberoids remain in the leaf litter
never reaching into the soil substrate.

Another distinct difference between primary and secondary tu-
beroids is that the primary tuberoid is a tan color whereas all
secondary tuberoids are waxy white. I found tiny primary tuber-

1994] Williams— 7riphora 37

Figure 4. Variation in primary and secondary tuberoid growth in successive
stages of Triphora development. (a) small primary tuberoids with secondary tuber-
oid development but no stem development; (b) hyaline stem production; (c) leaflet
of a non-flowering shoot; (d) flowering plants (arrows show swellings near primary
tuberoid which will develop into secondary tuberoids).

38 Rhodora [Vol. 96

Number of Plants

Year

Figure 5. Fluctuations in the total population from 1988 through 1993 on the
original 25 sites.

oids in the litter only | mm long, some of which were already
producing a new secondary tuberoid.

Figure 5 shows the yearly fluctuations in population size for
the original 25 sites found in 1988. It suggests a pattern of in-
creasing and decreasing population size, but more data is needed
to confirm this.

Table 2 shows the fluctuations of plants within each of the 25
sites over the six years. There is not only a great deal of variation
in plant numbers between sites but also within individual sites
ona year-to-year basis. What the Table doesn’t show is that plants
that reappear are not always in the same location as in previous
years even though they are in the same site. This indicates that
there are many primary tuberoids existing in the litter without
producing shoots for years at a time.

Due to this sporadic shoot production, it is very difficult to
assess population size and vigor. Their absence for many years
does not mean they are not there. I confirmed this by removing

1994] Williams— 7riphora 39

Table 2. Actual numbers of vegetative and flowering plants for each of the
original 25 sites and their yearly fluctuations.

Total Number of Plants

Site # 1988 1989 1990 199] 1992 1993
1] 4 0 0 0 0 0
2 8 0 6 4 9 11
3 4 0 0 0 0 1]
4 3 0 0 0 0 0
5 7 0 2 2 0 0
6 8 2 I 5 0 1
] 1 0) 0 ] 0 13
8 There is no site #8
9 14 2 0 8 5) 2

10 10 0 0 6 2 0

11 28 1 4 2 1] 10

12 0 0 0 0 0

13 2 l 0 0 0 0

14 3 0) 0 0 2 0

15 11 2 0 0 0 0)

16 19 5 7 9 20 18

17 13 0 1 2 8 14

18 20 3 0 22 35 36

19 22 7 l 4 2. 3

20 23 4 6 18 19 8

21 37 2a 30 61 113 93

22 47 0) 4 I 3 19

23 7 0 2 2 3 6

24 5 ] 0 1 0 0

25 3 2 l 0) 11 5

Total 300 51 65 148 243 240

litter in several potential sites in 1989 where I had never observed
Triphora. In three of the sites I found numerous tubers and in
1990 these areas contained 21 flowering plants.

DISCUSSION

Triphora trianthophora is a species of the climax hardwood
forest always associated with Fagus grandifolia (Lownes, 1920;
Zavitz and Gaiser, 1956; Sheviak, 1974; Crow and Stokes, 1980;
Brackley, 1981; Martin, 1983; Keenan, 1992). It initiates shoot
growth and flowers in August when the herbaceous environment
under a full canopy is characterized by very low light levels,

40 Rhodora [Vol. 96

highest soil temperatures, and lowest nutrient and water avail-
ability due mainly to uptake by trees (Mahall and Bormann, 1978).
Many of the ground herbs here are spring ephemerals which com-
plete their life cycles prior to canopy closure. Early leaf and flower
production may be the result of selection for completing these
processes while light intensity at the forest floor is high and could
be viewed as a method by which plants reduce interspecific com-
petition (Newell and Tramer, 1978).

It has also been shown that spring ephemerals lack the capacity
to modulate their photosynthetic and respiratory physiology in
response to decreased light levels accompanying canopy closure.
This lack of adaptability limits the ephemerals from exploiting
reduced light environments and consequently ephemerals revert
to a dormant condition soon after the canopy species leaf out
(Chabot and Mooney, 1985).

Shade tolerant species such as Triphora carry out most of their
growth and photosynthesis under a closed canopy. 7riphora may
be successful in this stressful habitat because its mycorrhizal as-
sociation meets its energy and nutrient needs making it an obligate
mycotroph for most of its existence. As a result, its leaves are
reduced (indicating reduced photosynthetic capability), and much
of its existence is spent beneath the leaf litter reproducing by
means of new tuberoids. Many of 7riphora’s associates are also
saprophytic or root parasites. Epifagus virginiana 1s especially
abundant. Coralorrhiza maculata, and Monotropa uniflora are
also found thinly scattered in the herb layer.

Because of the positive correlation with Fagus grandifolia, it
seems probable that there is a three-way connection from Tri-
phora through it’s mycorrhizae to the Fagus roots and that 77i-
phora may be indirectly receiving it’s nutrients and photosynthate
from the beech trees. A similar situation occurs in the Australian
orchid Gastrodia cunninghamii which lives almost entirely un-
derground. Its tubers are covered with a network of fungal mycelia
which penetrates the living roots of an adjacent tree, most often
Nothofagus (Withner, 1974). This three-way relationship is known
to occur in other orchids and plants (Harley, 1982; Harley and
Harley, 1987; Bernhardt, 1989).

Although sample size for tuberoid measurements was small due
to the invasive nature of sampling, analysis of the data indicates
that the plant’s reproductive status depends in part on primary
tuberoid size. Since the variability in primary tuberoid size was
quite large for 1 to 4 flowered plants and the fact that many

1994] Williams— 7riphora 4]

primary tuberoids of flowering size (1.e., greater than .7 cm) were
discovered in the litter near measured plants suggests that 77vi-
phora continues a productive life under the litter reproducing
vegetatively after it has produced a flowering shoot. Additional
data may show that a primary tuberoid may produce many flow-
ering shoots years apart.

It is not unusual for deciduous forest herbs to reproduce veg-
etatively as well as by seed. However, even for species that replace
themselves primarily vegetatively, seeds are necessary for the
establishment of new populations and nearly all species flower
(Bierzychudek, 1982). The number of seedpods actually formed
relative to the number of flowers produced may be largely con-
trolled by pollinator activity and by general environmental and
physiological conditions (Withner, 1974). This limitation of re-
productive output by pollinators seems to be a common phe-
nomenon in many species (Bierzychudek, 1981).

From personal observations, the woods in August have few
pollinating insects. In order to ensure pollination, many of T7ri-
phora’s associates remain in flower for an extended time, such as
Solidago caesia, Aster divaricatus, and Laportea canadensis. Im-
patiens pallida flowers continually throughout the summer with
the later flowers being cleistogamous. Epifagus virginiana, an
abundant saprophyte in the area, has sterile upper flowers, but
abundantly fertile cleistogamous lower flowers.

Two other members of the Orchidaceae found scattered over
the site include Epipactis helleborine and Corallorhiza maculata.
These both share the feature of long flower availability to polli-
nators by successive opening of long-lived flowers. Triphora, on
the other hand, only opens its flowers for one day, and that fact
coupled with low pollinator availability accounts for the low cap-
sule set. To offset the short duration of flower availability, how-
ever, Triphora flowers gregariously, i.e., many flowers open on
the same day. Gregarious flowering is obviously an advantage to
plants with short-lived flowers, as it gives a much greater chance
of pollination than would be the case if the flowers opened spo-
radically over a long period (Lawson, 1966). These massed floral
displays presumably increase the attractiveness to pollinators and
in Isotria verticillata, the larger clones did have higher pollination
percentages (Merhoff, 1983).

Triphora trianthophora appears well adapted to the habitat in
which it is found. Since pollination and seed ripening occur rarely,
Triphora utilizes vegetative propagation by secondary tuberoids

42 Rhodora [Vol. 96

to ensure future generations in the extant population. This type
of adaptation, however, may result in a loss of genetic variability
in the population, and can become detrimental in that Triphora
trianthophora, like many other plants, is sensitive to changes in
it’s habitat. Loss of it’s habitat would mean certain destruction
for an entire population. The fact that it may not be detected for
years at a time makes this unique orchid especially difficult to
protect.

ACKNOWLEDGMENTS

I would like to thank the Massachusetts Natural Heritage and
Endangered Species Program for partial funding of this project.
I very much appreciated the helpful comments on the manuscript
by P. Somers, review and help with the statistics by T. Lee, and
especially to N. Williams for her field (and other) assistance she
so freely gave over the past five years.

LITERATURE CITED

Ames, O. 1948. Orchids in Retrospect: A Collection of Essays on the Orchi-
daceae. Botanical Museum of Harvard University, Cambridge, MA

BERNHARDT, P. 1989. Wiley Violets and Underground Orchids. Random House,
Inc., NY.

BIERZYCHUDEK, P. 1981. Pollinator limitation of plant reproductive effort. Amer.
Naturalist 117: 838-840.

BIERZYCHUDEK, P. 1982. Life histories and demography of shade-tolerant tem-
perate forest herbs: a review. New Phytol. 90: 757-776.

BRACKLEY, F. 1981. Orchids of New Hampshire. M.S. thesis, University of New
Hampshire, Durham, NH

CHABOT, B. F. and H. A. Mooney. 1985. Physiological Ecology of North Amer-
ican Plant Communities. Chapman and Hall, New York.

Crow, G. ANDI. SToKEs. 1980. Rare and endangered plants of New Hampshire:

Seah eS tga viewpoint. Rhodora 82: 173-189.

Dresser, R. 1981. e Orchids: Natural History & Classification. Harvard
Caiversity Press, pate

Hap.ey, G. 1982. Orchid Vivcoraza: pp. 85-118. 7a: J. Arditti, Ed., Orchid
Biology, Reviews and Perspectives, Vol. I]. Cornell University Press, Lon-
don.

Har ey, J. L. AND E. L. HARLEY. 1987. A Check-List of Mycorrhiza in the
British Flora. New ciee 105: 1-102.

KEENAN, P. E. 1986. New stations for Platanthera flava and Triphora triantho-
phora and other observations. Rhodora 88: 411-412.

——. 1988. Three-birds orchids at Golden Pond. Amer. Orchid Soc. Bull. 57:
25-27.

. 1992. Anew form of Triphora trianthophora (Swartz) Rydberg, and Part

1994] Williams— 7riphora 43

3 of observations on the ecology of Triphora trianthophora (Orchidaceae) in
New Hampshire. Rhodora 94: 38-41.

Lawson, G. W. 1966. Plant Life in West Africa. Oxford University Press, Lon-
don

Lownes, A. E. 1920. Notes on Pogonia trianthophora. Rhodora 22: 53-55.

Luer, C. A. 1975. The Native Orchids of the United States and Canada. New
York Botanical Garden, NY. 361 pp.

Mana lt, B. E., AND F. H. BORMANN. 1978. A quantitative description of the
vegetative phenology of herbs in a northern hardwood forest. Bot. Gaz.
(Crawfordsville)139: 467-48 1.

Martin, L. 1983. Nodding Pogonia. Maine State Planning Office. Pamphlet.
Critical Areas Program

MeruorF, L. A. 1983. Pollination in the genus Jsotria (Orchidaceae). Amer. J.
Bot. 70: 1444-1453.

NEWELL, S. J. AND E. J. TRAMER. 1978. Reproductive strategies in herbaceous
plant communities during succession. Ecology 59: 2 34,

SHEVIAK, C. J. 1974. An Introduction to the Beioey of the Illinois Orchidaceae.
Scientific Papers XIV, Illinois State Museum, Springfield, IL, pp. 49-50.

WITHNER, C. 1974. The Orchids. John Wiley & Sons, New York

ZAVITZ, C. H. AND L. Gaiser. 1956. Notes on 7riphora trianthophora in Ontario.
Rhodora 59: 31-35

40 DEMERITT AVE.
LEE, NEW HAMPSHIRE 03824

RHODORA, Vol. 96, No. 885, pp. 44-68, 1994

CALCAREOUS FENS OF WESTERN NEW ENGLAND
AND ADJACENT NEW YORK STATE

GLENN MOTZKIN

ABSTRACT

This study presents a community characterization of calcareous fens of western
New England and adjacent New York State. Based on analyses of data from 24
oo seven vegetation types are defined: Carex lasiocarpa-Cladium mariscoides

, Carex aquatilis Type, Betula pumila Type, Carex lacustris Type, Carex
Bae Type, Typha angustifolia-Carex lasiocarpa Type, and Carex interior-Carex
leptalea-Carex flava Type. The distribution of these vegetation associations is
related to hydrologic and ionic gradients. Calcareous fens of the region are similar
to minerotrophically rich fens elsewhere in North America with respect to veg-
etation and environmental characteristics.

Key Words: wetlands, calcareous fens, vegetation classification, ordination

INTRODUCTION

Calcareous wetlands are ecologically significant communities
that support numerous rare or uncommon plant species as well
as several unusual animal species (McVaugh, 1957; Bernard et
al., 1983; Rawinski and Rooney, 1989; Weatherbee, 1990). His-
torically limited in distribution and extent throughout the north-
eastern United States, modern development pressure continues
to threaten the integrity of several sites.

Although calcareous wetlands are recognized as priorities for
conservation, most previous investigations of freshwater wetlands
in the northeastern United States have focused on peatlands char-
acterized by acidic, nutrient-poor surface waters (Moizuk and
Livingston, 1966; Hemond, 1980; Damman and French, 1987;
among others). This research trend reflects the predominance of
acidic bedrock throughout the region and has resulted in the gen-
eral absence of information concerning the vegetation and en-
vironmental characteristics of circumneutral to alkaline wetlands.
The present study marks the first comprehensive evaluation of
calcareous fens in western New England and adjacent New York
State, providing baseline information that is critical for conser-
vation and management planning, as well as for future more
detailed studies of these communities. The specific objectives of
this study are to: 1) characterize vegetation associations of cal-
careous fens in western New England and adjacent New York

44

1994] Motzkin—Calcareous Fens 45

State, 2) relate variation in fen vegetation to environmental pa-
rameters, and 3) compare fens of the northeastern United States
with those described from elsewhere in North America.

Calcareous fens are non-forested wetlands that are influenced
by base-rich groundwater and contain calcicoles (Rawinski, 1984).
In the present study, the term ‘fen’ is applied to both level and
sloping communities with or without peat deposits. Although
most studies of fens in North America have been restricted to
peatlands (Schwintzer, 1978; Slack et al., 1980; Sims et al., 1982;
Davis and Anderson, 1991), several European investigators have
recognized the floristic and functional similarities between sites
with peat and those that lack organic sediments (Wheeler, 1980;
Boyer and Wheeler, 1989; Peterson, 1989).

STUDY AREA

In western New England and adjacent New York State, cal-
careous fens occur primarily as small communities (< 1-50 ha)
throughout the Berkshire Valley Lowland and portions of the
adjacent Taconic Region where carbonate bedrock or surficial
deposits occur (Figure 1). Many calcareous fens in the region occur
on rocks of the Stockbridge formation, which consist primarily
of marble, limestone, dolostone and other carbonate-rich rock
types (Motts and O’Brien, 1981). A few fens overlie the Wal-
loomsac and related formations that contain limestone interbed-
ded with schists. A single fen in the Connecticut Valley of Mas-
sachusetts occurs on Mesozoic sedimentary bedrock. Although
most fens occur on calcareous till deposits, several have developed
on glaciofluvial or glaciolacustrine deposits.

Calcareous fens in the region occur at elevations that range
from 75 to 400 m a.s.l., with most sites between 150-300 m.
Climate throughout the region is continental, with cool winters
and warm summers (U.S.D.A., 1988). Average annual precipi-
tation of 110 cm is fairly evenly distributed throughout the year.

METHODS

The methods for this investigation are modified from those
established for prior surveys for the Massachusetts Natural Her-

46 Rhodora [Vol. 96

N
a
010 50 100 Miles

gure 1. Map showing the location of the study area and the occurrence of
anne bedrock (stippled) in the northeastern United States. Study sites are
indicated as dots on the enlarged inset (right), except for one site in the Connecticut
Valley that is indicated on the regional map (left). Modified from Moore (1935),
Denny (1982), and Isachsen et al. (1991).

1994] Motzkin—Calcareous Fens 47

itage and Endangered Species Program (Motzkin, 1991). Field
sites were selected in consultation with the staffs of the MA, CT,
and NY Natural Heritage Programs and the Eastern Heritage Task
Force of The Nature Conservancy. Twenty-four sites, believed to
represent most of the undisturbed, non-forested fens in the region,
were sampled for vegetation and environmental characteristics.

Fifty-five 100 m? relevés (Mueller-Dombois and Ellenberg,
1974) were sampled for vascular plant species-cover estimated
within height strata and percent cover of dominant bryophytes.
Relevés were subjectively located in areas of uniform vegetation
and distributed to encompass a range of vegetation types. General
observations of vegetation structure, landform type, site condi-
tion, disturbance history, and presence of rare species were also
recorded. Bryophytes and unidentified herbaceous species were
collected for identification. Nomenclature follows Seymour (1989)
for vascular plants, except for Carex section Stellulatae, which
follows Reznicek and Ball (1980), and Carex stricta complex,
which follows Standley (1987). Nomenclature for bryophytes fol-
lows Crum (1976).

Surface water pH and conductivity were measured at each re-
levé location. Water samples were collected, refrigerated, and later
analyzed at the University of Massachusetts using an atomic ab-
sorption spectrophotometer to determine concentrations of cal-
cium, magnesium, and phosphorus. Water color of filtered sam-
ples (related to concentration of humic acids) was estimated in
three classes: ‘clear’, ‘moderate’, or ‘dark’. Depth of organic sed-
iments at each relevé was determined by probing to a maximum
depth of 2 m. A soil auger was used to determine the degree of
decomposition of the upper one meter of organic sediments, es-
timated according to the Von Post scale (Clymo, 1983).

Relevé cover data were analyzed with cluster analysis and or-
dination techniques. Agglomerative cluster analysis (AGGLOM;
Orloci, 1967) served as the basis for classifying vegetation types
and detrended correspondence analysis (DCA; Hill, 1979) was
useful in identifying relationships among relevés. Canonical cor-
respondence analysis (CCA; ter Braak, 1986), correlation analysis,
and Duncan’s Multiple Range Tests were used to identify envi-
ronmental factors influencing the distribution of plant commu-
nities.

48 Rhodora [Vol. 96

Average Within-Group Dispersion, as Percent of Total

O 20 40 60 80 100
| _| I t | J
# of
releves
Carex lasiocarpa-Cladium mariscoides Type
TA
I
; Carex aquatilis Type IB
5 :
Betula pumila Type TLA1
IIA
Carex lacustris Type TLA2
IL
9 ;
Carex stricta Type I1B1
IIB
7 Typha angustifolia-C. lasiocarpa Type
TIB2
Other
-~ Carex interior-C. leptalea-C. flava Type IIA
Il

Other TB

Figure 2. Simplified AGGLOM dendrogram based on absolute distances for
species cover in 55 relevés. See text for description of vegetation types.

RESULTS
Vegetation Classification

Relevé data were clustered by AGGLOM using absolute dis-
tances into three main groups, composed of several smaller clus-

1994] Motzkin—Calcareous Fens 49

ters (Figure 2). Group I is characterized by Carex lasiocarpa, C.
aquatilis, Potentilla fruticosa, and Myrica gale. Cladium maris-
coides, Rhyncospora alba, Sarracenia purpurea, Andromeda glau-
cophylla, Pogonia ophioglossoides, Drosera rotundifolia, Men-
yanthes trifolia, and Utricularia spp. occur with greater frequency
in this group than in Groups II or III (Table 1). Sphagnum spp.
and Campylium stellatum are the common bryophytes. Uncom-
mon species associated with Group I include Carex limosa, C.
aquatilis, and others.

Group II is characterized by Betula pumila, Carex lacustris, C.
stricta, Thelypteris palustris, Salix candida, other Salix species,
and Spiraea latifolia. Rhus vernix, Phragmites communis, Typha
angustifolia, T. latifolia, Lythrum salicaria, and Lysimachia thyr-
siflora are also characteristic of this group. Galium labradoricum,
Betula pumila, and Salix candida are species characteristic of
Group II that are regionally uncommon.

Group III is characterized by Carex interior, C. leptalea, C.
flava, C. hystericina, Larix laricina, Parnassia glauca, Solidago
patula, S. purshii, and Thelypteris palustris, with Rhamnus al-
nifolia, Equisetum fluviatile, Equisetum spp., Ribes spp., and Cor-
nus spp. also important. Rare species include Carex sterilis, C.
tetanica, Equisetum scirpoides, Petasites palmatus, Lobelia kal-
mii, and Spiranthes romanzoffiana.

Within the classification hierarchy, subgroups are characterized
by a subset of the species listed for the major groups as well as
by additional species. Subgroups that are distinct with respect to
species composition and abundance may be considered separate
vegetation types (Table 1). Group I includes two such subgroups.
The Carex lasiocarpa-Cladium mariscoides Type (Subgroup IA)
is a sedge-dominated association characterized by Carex lasio-
carpa, Myrica gale, Potentilla fruticosa, Peltandra virginica, and
Cladium mariscoides. Other frequent species include Drosera ro-
tundifolia, Sarracenia purpurea, Hypericum virginicum, Rhyn-
cospora alba, Typha species, Vaccinium macrocarpon, and Utric-
ularia spp. The dominant bryophytes are Campylium stellatum,
Calliergonella sp., and Sphagnum spp.

The Carex aquatilis Type (Subgroup IB) is structurally similar
to Subgroup IA but is dominated by Carex aquatilis or by Carex
aquatilis and C. lasiocarpa and lacks species such as Peltandra
virginica and Hypericum virginicum.

Group II includes four subgroups. The Betula pumila Type
(Subgroup IJA1) is characterized by a shrub layer of Betula pumi-

50 Rhodora [Vol. 96

Table 1. Percent frequency of occurrence by Group and Subgroup. * = incom-
plete sampling of minor bryophytes; no frequencies calculated.

Group: I II II
Subgroup: IA IB HWAl WA2 JBI WB2A ITA

Trees

Chamaecyparis thyoides 14

Acer rubrum 14 20 100 14 57 40 40

Fraxinus spp. 20 29 20

Pinus strobus 14 70

Larix laricina 20 40

Tsuga canadensis 30

Betula alleghaniensis 14 20

Carpinus caroliniana 14 20

Populus tremuloides 20

Quercus alba 20

Betula papyrifera 10

Tilia americana 10

Shrubs

Chamaedaphne calyculata 14

Lonicera villosa 14

Aronia arbutifolia 14 20 10

Andromeda glaucophylla 14 40

Myrica gale 86 =: 100 40 14 40

Potentilla fruticosa 86 =: 100 100 71 43 20 80

Alnus spp. 43 40 40 14 86 60 20

Salix discolor 14 20 40 71 14 40

Spiraea latifolia 14 40 43 71 60 40

Cornus stolonifera 14 80 14 40 50

Cephalanthus occidentalis 14 20 14 14 60

Salix candida 20 20 57 14 80 10

Betula pumila 60 100 40

Rhus vernix 40 14 14 20 20

Rhamnus alnifolia 20 14 40

Ribes spp. 20 14 60

Salix serissima 40 43 14 50
] : 60 14 43 20 60

Rosa palustris 40 29 10

Viburnum recognitum 20 14 20 10

Rhamnus frangula 20

Tlex verticillata 20

Myrica pensylvanica 20

Cornus amomum 29 14

Cornus racemosa 14 14 20

Spiraea tomentosa 14 10

Lyonia ligustrin 14 40

Lindera benzoin 14 20

Amelanchier sp. 14 20

1994] Motzkin—Calcareous Fens al

Table 1. Continued.
Group: I II Ill
Subgroup: IA IB WAl IA2 IIB! IIB2A IIIA

Vaccinium corymbosum 20 10

Viburnum lentago 30

C * spp. 10

Ulmus sp. 10

5 ae angustifolium 10

Hamamelis virginiana 10

Rhamnus cathartica 10
Graminoids

Rhyncospora alba 43 10

Eriophorum gracile 14

Scirpus acutus 43 20 14 20

Scirpus hudsonianus 14 40

Cladium mariscoides 86 60

Carex lasiocarpa 100 60 14 100

Carex aquatilis 14. +100 40 20

Phragmites communis 20 40 40

Carex limosa 20

Carex prairea 20 40 14 14 40 10

C Serene canadensis 60 57. 29 20 10

Poa palustris 14 20

Carex lacustris 80 ~=—100 40 10

Eleocharis erythropoda 20 20 43 20

Carex stricta 40 57 100 60 20

Carex stipata 57

Scirpus atrovirens 43 30

Carex vulpinoidea 29

Dulichium arundinaceum 60

Eleocharis sp. 29 40 14 60 30

Carex comosa 20 14 14 20

Carex interior 60 29 14 100

Carex leptalea 40 29 100

Carex flava 20 40 29 40 90

Carex ec 29 43 80

Carex tetani 20 30

elon vn carinatum 14 14 50

Glyceria striata 20 43 20 50

Juncus sp. 14 14 20 20

Muhlenbergia glomerata 14 20 14 14 50

Carex scoparia 14 14

Phalaris arundinacea 14

Bromus ciliatus 20 14 30

Juncus dudleyi 14 30

Juncus effusus 14 20

a2 Rhodora [Vol. 96
Table 1. Continued.
Group: I II Ill
Subgroup: IA IB WA! WA2 IBl IIB2A ITA
Juncus nodosu 14 50
Eleocharis Se 14
Leersia oryzoides 14
Carex rostrata 14
Poa pratensis 29 10
Carex granularis 40
Anthoxanthum odoratum 20
Festuca spp. 20
oa compressa 10
Sphenopholis sp. 10
Carex gracillima 10
Carex laevivaginata 10
Carex spp 14 10
Andropogon scoparius 10
Danthonia spicata 10
Herbs
Utricularia spp. 29
Potamoget 29
Potentilla palustris 29 40
Peltandra virginic 71 40 14 40
Pogonia ophioglossoides 29 20 10
Drosera rotundifolia 86 20 20 50
Sarracenia purpurea 100 80
Vaccinium macrocarpon 100 80 20
Menyanthes trifolia 86 ~=-100 20 20
Utricularia intermedia 57 80 20 14
Typha latifolia 14 60 60 57 14 10
Equisetum fluviatile 100 43 14 40
Lysimachia thyrsiflora 100 14 40 10
Osmunda regalis 14 40 14 10
Sagittaria latifolia 40 14 20
Parthenocissus quinquefolia 20 14
Rubus s 80 29 29 40
Polygonum sp 20 29 29
Acorus calamus 29 10
Lemnaceae 14 14
Galium labradoricum 40 43 14 10
Tris versicolor 40 57 14 40 20
I Nes desi capensis 20 14 43
Viola sp. 20 19
Bes oo 29
Solidag 40 14 20
Typha eee 43 40 14 29 =100

1994] Motzkin—Calcareous Fens 53
Table |. Continued.
Group: I II Ill
Subgroup: IA IB WA! IIA2 IIBl IIB2A IIIA
Thelypteris palustris 29 100 86 71 100 90
Hypericum virginicum 71 20 14 5 80
Lythrum salicaria 60 14 29 60 20
Eupatorium maculatum 40 43 43 40
Onoclea sensibilis 20 14 29 20 20
Eupatorium perfoliatum 14 14 20 20
Aesclepias incarnata 20 29 40
Proserpinaca palustris 20 14 20
ampanula aparinoides 29 40
Solidago patula 43 20 90
Utricularia minor 20
Equisetu p 14 14 80
Parnassia glauca 43 80
Solidago purshii 14 20 20 43 70
Symplocarpus cei 40 43 14 20 50
Thalictrum polygamu 40 14 40
Solanum dulcamara 20 14
Galium spp. 29 43 40
; 29 14
Ludwigia palustris 14 14
Verbena hastat 14
Umbelliferae spp 29 10
Hydrocotyl americana 14 40
Achillea millefolium 14 20
Compositae $s 29 20
Lonicera s 14 20
Lysimachia ciliata 14 10
Apios americana 14 10
Nuphar variegatum 14
Boehmeria cylindrica 14
Desmodium sp. 14
Geum rivale 40
Senecio aureus 40
Prunella vulgaris 30
Petasites palmatus 20
Smilacina stellata 20
eed ills 20
Fragar 20
E nek rugosum 10
Tussilago farfara 10
Amphicarpa bracteata 10
Geranium maculatum 10
Houstonia sp. 10
Cardamine sp. 10

54 Rhodora [Vol. 96
Table 1. Continued.
Group: I II Ill
Subgroup: IA IB WAL WA2 IIBl IIB2A TIA
Sanicula marilandica 10
Spiranthes romanzoffiana 10
Taraxacum officinale 10
Chrysanthemum leucanthemum 10
Clematis s 10
C vee ny ye reginae 10
Adiantum pedatum 10
Aster puniceus 10
Aster schreberi 10
Aster stare 10
ryopteris c 10
Epilobium ae 14 10
Equisetum scirpoides 10
Lycopus uniflora 10
Lysimachia terrestris 20 10
Mentha arvensis
Unknown 14 20 14 29 20
Bryophytes
Sphagnum fimbriatum 14
Sphagnum subsecundum 14
var. contortum
Drepanocladus sp. 14 20 14 60
Calliergonella sp. 71 40 60 14 14 40 10
Campylium stellatum 86 =100 80 57 29 40 60
Spha sci warnstorfil 14 30
Sphagnum spp. 20
S eines magellanicum 10
Thuidium delicatulum *
Leptodictyum sp. 7
ulacomnium palustre -
Cratoneuron filicinum .
Scorpidium scorpioides
Tomenthypnum nitens =
Helodium blandowii -
Dicranum scoparium a
*

Thuidium recognitum

la, Potentilla fruticosa, Acer rubrum, Cornus stolonifera, and Salix
spp., amidst herbaceous species such as Carex lacustris, Cala-
magrostis canadensis, Typha angustifolia, T. latifolia, Lysimachia
thyrsiflora, Lythrum salicaria, Thelypteris palustris, Equisetum

1994] Motzkin—Calcareous Fens 55

IA = 1
400 IB = 2
IIA1 = 3
TA2=4
TIB1=5
| TIIB2 = 6
| TA =
Other = 0
300
= |
2 200 9
x
L<§
7
|
| 7
|
100
7
fa) O
oeesee | | | | |
if T T T
O 100 200 300 400 500 600

Axis 2

Figure 3. Plot of DCA sample ordination for axes | and 2. Numbers refer to
groups and subgroups defined by cluster analysis (Figure 2).

fluviatile, and Campylium stellatum. Rhus vernix is locally abun-
ant.

The Carex lacustris Type (Subgroup IIA2) includes some of the
species found in the Betula pumila type, but lacks extensive woody
cover. Frequent species include Carex lacustris, C. stricta, Salix
candida, Potentilla fruticosa, Typha spp., and Thelypteris palus-
tris.

The Carex stricta Type (Subgroup IIB1) has higher frequency
and cover values for Carex stricta than other subgroups. Acer

56 Rhodora [Vol. 96

rubrum, Alnus sp., Scirpus atrovirens, Galium spp., Eupatorium
maculatum, and Thelypteris palustris are other characteristic spe-
cies.

The Typha angustifolia-Carex lasiocarpa Type (Subgroup IIB2)
is dominated by Typha angustifolia and Carex lasiocarpa, with
Salix candida, Cephalanthus occidentalis, Dulichium arundina-
ceum, Eleocharis spp., Campanula aparinoides, and Hypericum
virginicum frequently encountered. Potentilla fruticosa is less fre-
quent in this subgroup than in any other vegetation type identified
through cluster analysis. Two plots in this Subgroup are shown
as ‘Other’ in Figure 2. These plots are dominated by Scirpus acutus
with only low cover of species characteristic of the Typha an-
gustifolia-Carex lasiocarpa Type.

Group III includes two subgroups. The Carex interior-C. lep-
talea-C. flava Type (Subgroup IIIA) has sparse tree and shrub
strata containing Pinus strobus, Larix laricina, Potentilla fruti-
cosa, Salix serissima, Ribes spp., and Rhamnus alnifolia, and a
low herbaceous stratum characterized by Carex interior, C. lep-
talea, C. flava, C. hystericina, Juncus dudleyi, J. nodosus, Muh-
lenbergia glomerata, Parnassia glauca, Solidago purshii, and S.
patula. Relevés in Subgroup IJIB are shown as ‘Other’ in Figure
2 because these plots contain vegetation that appears transitional
between two or more of the types described above.

Figure 3 presents the results of the DCA ordination, with relevés
assigned according to the groups and subgroups defined by cluster
analysis. The grouping of relevés along the ordination axes dem-
onstrates the lack of overlap between types defined by cluster
analysis at both a coarse (major groups) and fine level (subgroups)
of classification.

Environmental Characteristics

The surface waters of the calcareous fens investigated are cir-
cumneutral to alkaline (pH: 6.0—-8.1), with relatively high con-
centrations of calcium (8-65 mg/liter) and magnesium (4-32 mg/
liter) (Table 2). Phosphorus levels are low, with only 8 out of 55
samples above the detection limit of .15 ppm (not shown in Table
2). Depth of organic sediments range from >200 cm in most
Group I sites to 0-15 cm in areas with visible groundwater seepage
(Group III). Where organic sediments occur, the degree of de-

able 2. Environmental characteristics of calcareous fens of western New England and adjacent New York State (average value with range
in ee Groups and vegetation types numbered as in Figure 3. Color refers to filtered samples (1 = clear, 2 = moderate, 3 = dark).

No. of Conductivity Depth* Von Post Color

Samples pH (umho) ee a (cm) (0-10) (1-3)
Group I 12 6.6 (6.0-7.1) 415 (87-1151) 31 (11-62) 10 (4-17) 200 3 (2-3) 2 (1-3)
Type | 7 6.5 (6.0-7.0) 213 (87-322) 22 (11-34) 10 (4-17) 200 3 (3-3) 2 (1-3)
Type 2 5 6.8 (6.5-7.1) 697 (225-1151) 43 (26-62) 10 (7-14) 200 3 (2-3) 2 (1-3
Group II 26 6.9 (6.5—7.7) 366 (208-581) 43 (8-65) 19 (7-32) 135 (20-200) 7 (3-9) 2 (1-3
Type 3 5 6.8 (6.7-6.8) 391 (251-505) 43 (34-52) 19 (11-26) 171 (55-200) 5 (3-8) 2 (1-2
Type 4 7 6.8 (6.5-7.4) 341 (225-461) 50 (30-62) 16 (10-22) 141 (65-200) 7 (3-9) 1 (1-3
Type 5 9 7.1 (6.7-7.7) 371 (248-515) 45 (16-65) 19 (7-32) 103 (20-200) 8 (3-9) 2 (1-3
Type 6 5 6.9 (6.7-7.2) 368 (208-581) 35 (8-50) 22 (15-29) 150 (75-200) 5 (3-9) 1 (1-2)
Group III 10 7.2 (6.9-8.1) 333 (158-478) 46 (26-62) 17 (8-23) 20 (0-70) 7 (3-9) 1 (1-2)
Type 7 10 7.2 (6.9-8.1) 333 (158-478) 46 (26-62) 17 (8-23) 20 (0-70) 7 (3-9) 1 (1-2)

* Maximum value of 200 cm.

58 Rhodora [Vol. 96

composition within the uppermost one meter of sediments varies
from largely undecomposed peat (Von Post scale: 2-3) in many
Group I plots, to highly decomposed muck (Von Post: 7-9) in
Group II sites.

Correlation analyses indicate a significant negative relationship
between depth of organic sediments and pH (R = —.63; P =
.OOO1), degree of pero pOsuon (R = —.45; P = .0017), and Mg
concentration (R = —.33; P = .02). Ca concentration is positively
correlated with Mg concentration (R = .58; P = .0001), and with
conductivity (R = .49; P = .0004), and both Ca and Mg are
negatively correlated with hydrogen ion concentration (Ca: R =
—.39, P= .01; Mg: R= —.39, P= .01). pH is positively correlated
with degree of decomposition of peat (R = .34; P =.03). Water
color is positively correlated with hydrogen ion concentration (R
= 36; P= .01).

Relationship Between Vegetation and Environment

Canonical correspondence analysis (CCA; ter Braak, 1986), a
form of direct gradient analysis in which the vegetation ordination
is constrained by environmental data, was used to relate vege-
tation variability to measured environmental variables. In the
resulting ordination (Figure 4), relevés separate primarily along
gradients of depth of organic sediments, pH, and degree of de-
composition of organics. Ca, Mg, conductivity, and water color
account for less of the variability in the vegetation. Combined,
these environmental factors account for a relatively low percent-
age of the total variability of the vegetation (~ 25%), reflecting
the fact that the fens investigated are highly variable with respect
to vegetation and water chemistry. ter Braak (1986) has noted
this pattern for other data sets, suggesting that CCA may still be
useful in cases where a relatively low percentage of the total vari-
ability is accounted for by the ordination diagram.

The hierarchical nature of cluster analysis allows one to test
the hypothesis that different environmental variables are asso-
ciated with vegetation variation at different levels of classification.
For example, in montane systems one might hypothesize that
elevation gradients are responsible for patterns of vegetation vari-
ation at a coarse level. Within a particular elevation zone, how-
ever, local edaphic factors (rather than elevation) might control
patterns of vegetation variation. Duncan’s Multiple Range Tests

1994] Motzkin—Calcareous Fens 59

‘ e ‘
‘si a
<50cm organic ..”” 50-200cm / >200cm organic sediments
sediments ‘ organic /
sediments + -0O5

Figure 4. Plot of CCA sample ordination for axes | and 2 Broken ines dis-
tinguish plots with > 200 cm, 50-200 cm, and <50 cm of organic Solid
wavy lines separate plots with pH values of >7.3, 7.0-7.3, and <7.0. An outlier
with respect to pH class (pH = 7.4) is indicated by an open circle.

were used to determine which environmental factors differ sig-
nificantly with respect to major vegetation groups and subgroups
defined by cluster analysis.

At the coarse level of classification, depth to mineral soil is the
only variable measured that differs significantly (P = .0001) be-
tween all three groups. Hydrogen ion concentration, Mg, and
degree of decomposition of organics (Von Post value) for Group

60 Rhodora [Vol. 96

I are significantly different (P = .01, P = .0004 and P = .0001,
respectively) from Groups II and III (which are not different from
each other). Calcium levels are different (P = .07) between Groups
I and III, but not between Groups I and II or between Groups II
and III. Conductivity and color of filtered samples do not vary
significantly among the three groups. These results suggest that
Groups I and III (defined by cluster analysis of relevé data) are
most different with respect to environmental parameters (e.g.,
depth of sediments, pH, Mg, degree of decomposition, and Ca)
and that Group II is intermediate with respect to these variables.
As with the CCA results, depth to mineral soil is the single variable
most strongly associated with vegetation patterns at this level of
classification.

Similar analyses were performed to determine environmental
differences between subgroups within each major cluster group.
Within Group I, Subgroup 1A differs significantly from Subgroup
1B with respect to Ca concentration (P = .0223) and conductivity
(P = .0098). Within Group II, only Von Post values differ between
the subgroups (P = .0767), with Subgroup IIB1 significantly dif-
ferent from Subgroup IIAI (but not from I[A2 and IIB2), and
Subgroups ITA], IfA2, and IIB2 not significantly different from
each other. No other environmental parameters tested differ sig-
nificantly between subgroups defined by cluster analysis. Group
III contains only one subgroup (IIIA) considered to be a distinct
vegetation type, and so was not included in this level of analysis.

DISCUSSION

Relationship of Vegetation to other North American Fens

The coarse level of classification determined through cluster
analysis of relevé data can be related to existing classifications of
calcareous wetlands. Group I corresponds to a Lake Basin Gra-
minoid Fen type described from the Northeast (Rawinski, 1984;
Weatherbee, 1990), and to a Rich Graminoid Fen type of New
York State (Reschke, 1990). Dominant species in this group are
also recorded from minerotrophic fens of northern Michigan
(Schwintzer, 1978). Group II contains elements of the Seepage
Swamps of highly calcareous soils described by McVaugh (1957)

1994] Motzkin—Calcareous Fens 61

and the Seepage Marsh of Rawinski (1984), which he describes
as having characteristics of both fen and marsh. Rich-fen types
from Britain are structurally similar to Group II vegetation
(Wheeler, 1980). Group III, which corresponds to Reschke’s (1990)
Rich Sloping Fen and to Rawinski’s (1984) Calcareous Sloping
Fen, contains several species considered to be ‘rich fen indicator
species’ in Ontario (e.g., Carex leptalea, Eriophorum viridi-cari-
natum, Scorpidium scorpioides; Sims et al., 1982).

Several of the vegetation types (subgroups) described here are
also related to types found elsewhere in North America. The
Carex lasiocarpa-Cladium mariscoides Type (Subgroup IA) has
been described from a Maine peatland (Rawinski and Rooney,
1989). The vascular vegetation of the Carex aquatilis Type is
apparently related to the Drepanocladus revolvens-Carex aquatilis
association found in patterned rich fens in western Alberta (Slack
et al., 1980), although none of the study sites investigated in our
region display the physiographic patterning typical of fens in bo-
real climates (Foster and King, 1984; Glaser, 1987). Additional
work is needed to evaluate the bryophyte composition of this

type.

The Betula pumila Type corresponds to a ‘larch-birch-sumac
swamp type’ described from a site included in this investigation
(McVaugh, 1957) and to the Rich Shrub Fens of New York
(Reschke, 1990). The Carex stricta Type 1s apparently similar to
the Sedge Meadow communities found in Maine and New York
(Reschke, 1990; Gawler, 1991). No published descriptions cor-
respond closely to the Typha angustifolia- Carex lasiocarpa Type.

The Carex interior-C. leptalea-C. flava Type contains species
described by numerous authors as typical of rich calcareous fens
(e.g., Sims et al., 1982; Vitt and Chee, 1990), although no specific
reference to this association occurs from outside the region.

Field observations suggest that two additional vegetation as-
sociations occur in fens of the region but were inadequately sam-
pled in the current study. A Scirpus acutus association was ob-
served in several of the largest peatland fens and along marl pond
shores. Also, extensive Phragmites communis-dominated areas
occur in association with several of the vegetation types described
above. Additional sampling is required to relate these associations
to the vegetation classification presented here and to those de-
veloped for other regions of North America.

62 Rhodora [Vol. 96

Environmental Characteristics

Several authors have related vegetation patterns to environ-
mental characteristics of graminoid-dominated peatland fens
elsewhere in North America (Schwintzer, 1978; Slack et al., 1980;
Sims et al., 1982; Foster and King, 1984; Vitt and Chee, 1990).
Comparison of environmental characteristics reported here (Ta-
ble 2) with those of previous investigations (Table 3) suggests that
the groups and subgroups defined in the current study are com-
parable to minerotrophically rich fens sampled throughout North
America. Of the types considered here, Group I is apparently the
least ‘rich’ (with respect to surface water chemistry), Group II
sites are intermediate, and Group III may be characterized as
extremely rich (Table 2).

Group I represents a peatland-fen type characterized by deep
organic sediments, permanently saturated conditions, and con-
solidated or floating, sedge-dominated organic mats. Although
little is known of the developmental history of these fens in our
region, preliminary observations of the macrofossil stratigraphy
of two sediment cores extracted from Kampoosa Fen in Stock-
bridge, MA indicate that since the development of a peat mat at
this site, fairly uniform sedge and shrub-sedge peat has accu-
mulated to a depth of 270 cm. This suggests that vegetation struc-
turally similar to the modern vegetation has occupied the site
since the time of peat mat establishment. Although the time it
has taken for these 270 cm of sediments to accumulate is not
known, an estimate of a few thousand years seems reasonable,
based on estimates of peat accumulation rates from throughout
northeastern North America. Thus, the vegetation at this site
appears relatively stable and there is no indication of rapid in-
filling or terrestrialization.

Sites supporting Group II vegetation are characterized by 50-
200+ cm of moderate to well-decomposed organic sediments
occurring in a variety of seasonally inundated depressions and
drainages. Vegetation characteristic of this group occurs in basins
that also support Group I vegetation, in canopy gaps within rich
forested swamps, in drainages with current or former beaver im-
poundments, and in level to slightly sloping sites adjacent to
Group III vegetation.

Group III sites are slightly to moderately sloping with ground-
water seepage frequently visible in distinct rivulets. In areas of

Table 3. Chemical properties of waters from North American wetlands (average values with range in parentheses). From Zoltai (1988).

Wetland No. of Conductivity MG
Class Samples pH (umho) (mg/liter) (mg/liter) Source
Bog 18 4.0 (3.7-4.4) — 2.3 (1.2-3.7) 0.4 (0.2-0.9) Schwintzer (1981)
13 (4.6-5.1) (35-62) (0.2-0.8) (0.1-0.2) Gauthier (1980)
10 (3.84.4) 31 0.2 | Foster and King (1984)
Fen (poor) 193 (4.6—5.2) (18-59) (0.44.8) (0.1-0.7) Gauthier (1980)
14 (4.7-5.5) 49 0.3 0.2 Foster and King (1984)
| 5.0 ~ 2.4 0.4 Vitt et al. (1975)
Fen (moder- 42 5.2 65 1.1 0.2 Gauthier (1980)
ately poor)
Fen (interme- 9 7.2 (6.8-7.9) 281 (140-456) 28 (18-37) 11 (4-28) Slack et al. (1980)
diate to rich) 2] 6.1 (5.2-6.9) 59 (33-128) 10 (4-18) _ Glaser et al. (1981)
5 6.5 (5.4-7.1) oa 43 (7-124) 10 (2-15) Schwintzer (1978)
Swamp (conif- 12 7.2 (6.9-7.8) ~ 40 (22-52) 12 (8-17) Schwintzer (1981)

64 Rhodora [Vol. 96

heavy groundwater discharge, mineral soil is typically exposed.
Elsewhere, small hummocks develop with shallow accumulations
of organic matter. Field observations suggest that the importance
of woody species in seepage fens is inversely related to distur-
bance, with more highly disturbed sites (i.e., those that are flooded
by beavers, grazed, or mowed) supporting only low, scattered
shrubs. A similar pattern has been described from European fens
(Regnéll, 1980; Peterson, 1989).

Environmental Gradients and Vegetation Distribution

CCA analyses indicate that depth to mineral soil is the envi-
ronmental parameter that accounts for the greatest portion of the
variability in the relevé data, with degree of decomposition of
organics and pH also important. Duncan’s Multiple Range Tests
also indicate depth as a significant variable, along with degree of
decomposition, pH, Ca, and Mg concentrations. In the study sites
considered, depth of peat probably reflects a gradient of ‘rate of
groundwater-flow’ ranging from ‘slow’ (and poorly aerated) in
well-defined basins with deep peat accumulations (Group I), to
relatively ‘rapid’ (and well aerated) in sloping (Group III) sites
that lack peat. Depth of peat may also indicate the relative avail-
ability to vegetation of nutrients or cations derived from contact
with mineral soil. The results of both CCA and Duncan’s Multiple
Range Tests suggest the importance of both hydrologic gradients
(as indicated by depth of organic sediments) and ionic gradients
related to alkalinity-acidity (pH, Mg, and Ca concentrations) in
determining the variability between the rich fen types considered
here,

Similar hydrologic and ionic gradients have been related to
vegetation distribution in fen types throughout North America.
Peat depth, pH, water levels, and water chemistry are related to
differences between fens in Canada (Slack et al., 1980; Sims et
al., 1982; Vitt and Chee, 1990). Vegetation variation in a peat
and marl fen in western New York is related to complex gradients
associated with hydrology, soil organic matter, and soil carbonate-
carbon concentrations (Bernard et al., 1983). In one of the few
detailed studies to consider non-peatland fens, Boyer and Wheeler
(1989) suggest that structural and compositional variation in seep-
age fens in Britain is related to differences in phosphorus avail-
ability between high and low groundwater discharge areas.

1994] Motzkin—Calcareous Fens 65

At a fine level of classification, subgroups within Group II do
not differ significantly with respect to most environmental pa-
rameters measured. This probably reflects the high degree of vari-
ability of the measured environmental parameters, a pattern not-
ed by several authors (e.g., Sims et al., 1982) for rich fens elsewhere
in North America.

Within Group I, relevés in the Carex lasiocarpa-Cladium mar-
iscoides Type differ significantly from those in the Carex aquatilis
Type with respect to Ca concentration and conductivity, but not
with respect to depth of organics or degree of decomposition. This
suggests an ionic (alkalinity-acidity) rather than hydrologic gra-
dient, with the Carex lasiocarpa-Cladium mariscoides Type rep-
resenting the ‘poorer’ type. This corresponds well with the ob-
servation that in the northeastern United States, Carex lasiocarpa
occurs at sites ranging from poor to rich, whereas Carex aquatilis
is restricted to minerotrophically rich sites. These results also
suggest that with respect to Group I, different environmental gra-
dients (hydrologic vs. minerotrophic) may be associated with veg-
etation variation at different levels of classification.

CONCLUSIONS

Calcareous fens of western New England and adjacent New
York State are similar to rich peatland fens throughout North
America with respect to species composition, water chemistry,
and the influence of hydrologic and minerotrophic gradients on
vegetation distribution. Future attempts to develop regional clas-
sifications of fens should more fully evaluate seepage and sea-
sonally inundated types. In cases where hydrologic and ionic (al-
kalinity-acidity) gradients do not appear to explain observed
variability in vegetation, factors such as nutrients, developmental
history, response to disturbance, or biotic interactions may be
significant.

ACKNOWLEDGMENTS

I thank D. Foster, K. Metzler, W. Patterson, L. Sneddon, B.
Sorrie, and P. Swain for support and helpful guidance, H. Crum,
N. Slack, and L. Standley for assistance with plant identification,
and P. Weatherbee and T. Zebryk for assistance with field work.
This study was improved by the critical reviews of D. Foster, F.

66 Rhodora [Vol. 96
Gerhardt, A. Lezberg, C. Mabry, W. Patterson, G. Whitney, and
B. Wilson

Funding for this project was provided by the Massachusetts
Natural Heritage and Endangered Species Program and the East-
ern Heritage Task Force of The Nature Conservancy.

LITERATURE CITED

BERNARD, J. M., F. K. SEISCHAB AND H. G. GAucu, Jr. 1983. Gradient aa
of the vegetation of the Byron-Bergen swamp, a rich fen in western
York. Vegetatio 53: 85-91.

Boyer, M. C. H. AND B. D. WHEELER. 1989. Vegetation patterns in spring-fed
calcareous fens: calcite precipitation and constraints on fertility. J. Ecol. 77:
597-609

CLymo, R. S. 1983. Peat, pp. 159-224. In: A. J. P. Gore, Ed., Ecosystems of
the World, 4A, Mires: Swamp, Bog, Fen, and Moor. General Studies. Elsevier,
Amsterdam

Crum, H. 1976. Mosses of the Great Lakes Forest. Univ. Herbarium, Univ. of
Michigan. Vol 10.

DAMMAN, A. W. H. AND T. W. FRENCH. 1987. The ecology of peat bogs of the
placiated northeastern United States: a community profile. U.S. Fish Wildl.
Serv. Biol. Rep. 85(7.16)

Davis, R. B. AND D. S. ANDERSON. 1991. The eccentric bogs of Maine: a rare
wetland type in the United States. Maine Agric. Exp. Station Tech. Bull. 146.
Univ. of Maine, Orono

Denny, C. S. 1982. Geononsholony of New England. U.S. Geol. Surv. Prof.
Paper 1208. a , DC.

Foster, D. R. AND G. A. Kinc. 1984. Landscape features, vegetation and de-
co apacee history of a patterned fen in southeastern Labrador, Canada.
Journal of Ecology 72: 115-143.

GaAuTHIER, R. 1980. La végétation des tourbieres at les sphaignes du parc des
Laurentides, Quebéc. Forest Ecology Laboratory, Laval Univ. Etudes éco-
logiques No. 3. Quebec, Canada.

GAWLER, S. (oot. Natural landscapes of Maine: a classification of ecosystems
and natural communities. Draft report. Office of nena Planning,
Maine Natural Heritage Program. Augusta, Mai

GLASER, P. H. 1987. The ecology of patterned boreal peatlands of northern
Minnesota: Rema profile. U.S. Fish Wildl. Serv. Rep. 85(7.14).

,G heeler, E. Gorham and H. E. Wright, Jr. 1981. The patterned
mires oe Red Lake peatland, northern Minnesota: vegetation, water chem-
istry and landforms. J. Ecol. 69: 575-599.

Hemonb, H. F. 1980. Biogeochemistry of Thoreau’s Bog, Concord, Massachu-

6

Hirt, M. O. 1979. DECORANA—A FORTRAN Program for Detrended Cor-
respondence Analysis and Reciprocal Averaging. Cornell University, Ithaca,
New York.

ISACHSEN, Y. W., E. LANDING, J. M. LAUBER AND L. V. RICKARD. 1991. Geology

1994] Motzkin—Calcareous Fens 67

of New York. NY State Museum Geological Survey. Univ. of the State of
New York, Albany

McVauGu, R. 1957. Flora of the Columbia County area, New York. NY State
Museum and pains Serv. Bull. No. 360. Albany,

Moizuk, G. A. AND R. B. LivinGsTon. 1966. Ecology of red seiple (Acer rubrum)
in a Massachusetts upland bog. Ecology 47: 942-950.
Moore, F. H. 1935. Marbles and limestones of Connecticut. State Geol. and

Nat. Hist. Surv. Bull. No. 56. Hartford, CT.

Motrts, W. S. AND A. L. O’BrIEN. 1981. Geology and hydrology of wetlands in
Massachusetts. Water Resources Research Center Publication No. 123. Uni-
versity of Massachusetts

Morzkin, G. 1991. Atlantic white cedar wetlands of Massachusetts. Massachu-
setts Agricultural Experiment Station Research Bull. No. 731. Amherst, MA.

MUELLER-Domsois, D. AND H. ELLENBERG. 1974. Aims and Methods of Veg-
etation Ecology. John Wiley & Sons, New York.

Or.oci, L. 1967. An agglomerative method for classification of plant commu-
nities. J. Ecol. 55: 193-206

PETERSON, P. M. 1989. Minerotrophic fens and wet meadows. Opera Botanica
96: 31-37.

RaAwinskI, T. J. 1984. New England natural community classification. New
England setae runes Program, The Nature Conservancy. Boston, MA.
Draft ae

RAwINsKI, T. J. oe co Rooney. 1989. A preliminary study of wetland
vegetation at sounant san Penobscot County, Maine. Unpubl. report.
Maine Natural Heritage Progra

REGNELL. G. 1980. A numerical on of successions in an abandoned, damp
calcareous meadow in S Sweden. Vegetatio 43: 123-130

RESCHKE, C. 1990. Ecological communities of New York State. New York Nat-
ural Heritage aay N.Y.S. Dept. of Environ. Conserv. Latham, N.Y.

Reznicek, A. A. AND P. W. BALL. 1980. The taxonomy of Carex geese a
lulatae in North America north of Mexico. Contr. Univ. Mich. Herb.

53-203.

sian Seg C. R. 1978. Vegetation and nutrient status of northern Michigan
s. Can. J. Bot. 56: 3044-3051.
——. 1981. Vegetation and nutrient status of northern Michigan bogs and
conifer swamps with a comparison to fens. Can. J. Bot. 59: 842-853.
Seymour, F.C. 1989. The flora of New England, 2nd ed. Phytologia Memoirs
V. Plainfield, NJ.

Sims, R. A., D. W. COWELL AND G. M. Wickware. 1982. Classification of fens
near southern James Bay, Ontario, using vegetational physiognomy. Can. J.
Bot. 60: 2608-2623.

SLACK, N. G., D. H. Vitr anp D. G. Horton. 1980. ones gradients of
minerotrophically rich nite in western Alberta. Can. J. Bot. 58: 330-350.

STANDLEY, L. A. 1987. Taxonomic revision the Carex ne (Cyperaceae)
complex in eastern North America. Can. J. Bot. 67: 1-14.

TER BRAAK, C. J. F. 1986. Can dee analysis: a new eigenvector
technique for multivariate direct gradient analysis. Ecology 67: 1167-1179.

U.S.D.A., Soil Conservation Service. 1988. Soil survey of Berkshire County,
Massachusetts. Amherst, MA

68 Rhodora ; [Vol. 96

Vitt, D. H., P. ACHUFF AND R. E. ANpRrus. 1975. The vegetation and chemical
properties of patterned fens in the Swan Hills, north central Alberta. Can. J.
Bot. 53: 2776-2795.

Virt, D. H. AND WalI-LIN CHEE. 1990. The relationships of vegetation to surface
water chemistry and peat chemistry in fens of Alberta, Canada. Vegetatio
89: 87-106

WEATHERBEE, P. B. 1990. Flora of Berkshire County, Massachusetts. Unpub-
lished M.S. thesis, University of New Hampshire

WHEELER, B. D. 1980. Plant communities of rich-fen systems in England and
Wales. J. Ecol. 68: 365-395.

ZOLTAI, S. C. 1988. Wetland Environments and Classification, pp. 1-26 Jn:
National Wetlands Working Group. Wetlands of Canada. Ecological Land
Classification Series, No. 24. Sustainable Development Branch, Environment
Canada. Ottawa, Ontario.

HARVARD FOREST
HARVARD UNIVERSITY
PETERSHAM, MASSACHUSETTS 01366

RHODORA, Vol. 96, No. 885, pp. 69-74, 1994

CHROMOSOME NUMBERS OF SOME LATIN
AMERICAN SPECIES OF ALNUS (BETULACEAE)

GUILLERMO RESTREPO AND CAMILLE GERVAIS

ABSTRACT

The chromosome numbers of A/nus acuminata H.B.K. ssp. acuminata, A. ac-
uminata ssp. arguta (Schlechtendal) Furlow and A. jorullensis H.B.K. ssp. jorul-
lensis were determined on material from 10 different localities of 4 Latin American
countries: Columbia (6 stations), Venezuela (1), Costa Rica (2) and Guatemala
(1). All the species were tetraploid (2n = 28), A. acuminata ssp. arguta and A.
Jorullensis ssp. jorullensis being cytologically studied for the first time.

Key Words: chromosome numbers, tropical highland forest, A/nus acuminata,
Alnus jorullensis, Columbia, Venezuela, Costa Rica, Guatemala

INTRODUCTION

The present paper is a part of a comprehensive study, initiated
in 1990 by the first author, on the genetic variation and ecology
of Alnus acuminata H.B.K. ssp. acuminata. in Columbia.

This species has a large geographical and ecological distribution
along the Andes cordillera, from western Venezuela to northern
Argentina (Furlow, 1979), In Columbia, A. acuminata behaves
as a pioneer species whose normal distribution is between 1700
and 3300 m in the Central and the Oriental cordilleras (Del Valle
Arango and Gonzalez Pérez, 1988). According to the classification
system of Holdridge (1967), A. acuminata is present in the fol-
lowing types of habitats: dry forest-lower montane (df-LM), moist
forest-lower montane (mf-LM), wet forest-lower montane (wf-
LM), moist forest-montane (mf-M) and wet forest-montane (wf-M).

In the Central cordillera the species presents itself as a tree
which can grow up to 40 m high and 70 cm in diameter. In the
Oriental cordillera, however, mostly shrubby specimens are ob-
served and only a few trees in certain habitats exceed 10 m

Because of these phenotypic (or possibly genetic) variations,
chromosome counts on some representative individuals of the
species in its natural and more septentrional area were carried
out. The only previous cytological determinations on Latin Amer-
ican material of this taxon were those of Giusti (1989) from Tu-
cuman, in Argentina, and those of Coba de Gutiérrez and Al-
varado de Coral (1989) from Manizales in Columbia. The
chromosome number 2n = 28 was reported in both cases. As a

69

70 Rhodora [Vol. 96

complement to the present chromosomes studies on the typical
subspecies (ssp. acuminata), additional counts were realized on
ssp. arguta, from Costa Rica, and on A. jorullensis from Guate-
mala.

The chromosome number 2n = 28 was the only one found in
the native species of A/nus in Canada and in the United States
(Furlow, 1979), though at least five ploidy levels are reported in
the genus. Besides the tetraploid level (2n = 28) which is the most
common, a diploid species (A. inocumae Mur & Kus. = A. hirsuta
Turkz. var. microphylla Kusaka) 1s indicated for Japan (Chiva,
1962) and many European and Asiatic species have 2n = 42 or
2n = 56 chromosomes (Gram et al., 1941; Chiva, 1966; Furlow,
1979; Hall and Maynard, 1979; Bousquet and Lalonde, 1990).
Finally, A. firma Sieb. & Zucc. and A. sieboldiana Matsum. from
Japan are reported to possess 2n = 112 chromosomes (Kodama,
1967, 1970), at least in the root nodule tissues.

A few species are reputed to present two or three polyploidy
levels: A. cordata and A. orientalis Decne. have both 2n = 28 or
42, A. hirsuta Turez. 2n = 28 or 2n = 56, A. glutinosa (L.) Gaertn.,
A. japonica Sieb. et Zucc. and A. subcordata Mey. 2n = 28, 42,
or 56 chromosomes (Gram et al., 1941; Chiva, 1966; Furlow,
1979; Hall and Maynard, 1979; Bousquet and Lalonde, 1990).
The chromosome number 2n = 42 could be interpreted as the
result of hybridization (Gram et al., 1941; Furlow, 1979).

MATERIAL AND METHODS

Nutlets of A. acuminata ssp. acuminata from 7 stations (6 from
Columbia, one from Venezuela) were collected between June and
August, 1990, by the first author. Additional seeds (4. acuminata
ssp. arguta from two stations in Costa Rica and A. jorullensis
from one provenance in Guatemala) were also received by ex-
change services.

The seeds were sterilized in 6% sodium hypochlorite for 10
minutes, washed in 70% ethanol for 30 seconds and rinsed in
sterile distilled water for another 30 seconds. Afterwards they
were sown on a 5 mm layer of silica in Petri dishes. The Petri
dishes were kept in a germination incubator with a 12 hour pho-
toperiod and 80% humidity. The day temperature was 23°C and
the night 18°C.

For the cytological study, seedlings were collected when the first

Table 1. Chromosome counts in A/nus acuminata H.B.K. and A. jorullensis H.B.K. from Latin America.

Chr.
Latitude Longitude Altitude No. of Number*
Cordillera Locality (N) (W) (m) Tree (2n)
A, acuminata H.B.K. ssp. acuminata

Columbia
Nudo de Pasto Santa Lucia, La Cocha (Narifio) 1°03’ TOS 2710 C18-1 28
Central Rio Piendam6o, Silvia (Cauca) 2°36’ 76°21’ 2300 C32-5 28
Central Rio Verde, Pijao (Quindio) 4°20' 75°36! 2290 C07-3 28
Central Cocora, Salento (Quindio) 4°45’ use 2600 C11-2 28
Central La Cristalina, Neira (Caldas) 5°21! 73°32! 2380 C02-3 28
Oriental Rio Pémeca, Arcabuco (Boyaca) 5°44’ 73°26' 2790 C30-5 28

Venezuela
Oriental Rio Chama, Tabay (Mérida) 8°38’ 71°04’ 2050 C27-2 28

A. acuminata ssp. arguta (Schlechtendal) Furlow

Costa Rica
Talamanca Jardin, Sta. Maria Dota (Cartago) 9°42’ 83°57’ 2000 CR-1 28
Volcanica Central Prusia, Llano Grande (Cartago) 9°56' 83°54’ 2200 L125-1 28

A. jorullensis H.B.K. ssp. jorullensis

Guatemala

Tierras Altas Siguampar (Sacatéquez) 14°35’ 90°48’ 2000 AJ91-1 28

* The chromosome numbers have been counted from seeds collected in situ.

oe) Rhodora [Vol. 96

Figures 1-3. Somatic chromosomes in A/nus. 1. A. acuminata ssp. acuminata;
metaphase (2n = 28) in young leaf tissue after cold treatment (7 hr. at 4°C);
Arcabuco, Columbia. 2. A. acuminata ssp. arguta, metaphase (2n = 28) in root
tip tissue after cold treatment of plantlet (7 hr. at 4°C); Santa Maria de Dot
Costa Rica. 3. A. jorullensis ssp. jorullensis; metaphase (2n = 28) in root tip fiscue
after cold treatment of plantlet (7 hr. at 4°C); Siguampar, Guatemala.

pair of leaves appeared and were deposited in cold water in a
refrigerator (4°C) for about 7 hours. This pre-treatment was nec-
essary to shorten the chromosomes before the fixation of the
seedlings in a 3:1 mixture of anhydrous alcohol and glacial acetic
acid. The root tips or very young leaves were used to count the
chromosomes after coloration in acetocarmine for at least 2 hours.
The drawings were done with the help of a camera lucida.

RESULTS AND DISCUSSION

The chromosome number 2n = 28 was observed for all of the
individuals studied in the three taxa (Table 1). This tetraploid

1994] Restrepo and Gervais— Alnus 73

number is actually reported for all of the chromosome counts
published for American species of A/nus (Furlow, 1979).

In our material, the chromosome being rather small (1.25 to
2.5 um; Figures 1-3), it was not possible to prepare a karyo-
gramme for any of the species. However, since the chromosomes
of A. jorullensis ssp. jorullensis, showed more details (Figure 3),
it could be tentatively assumed that 2 pairs are metacentric, 8
submetacentric, 3 acrocentric and one telocentric.

The chromosome counts for A. acuminata ssp. arguta and A.
Jorullensis ssp. jorullensis are apparently the first reports for these
two species. The other taxon, 4. acuminata ssp. acuminata, have
been studied in 7 different stations from three cordilleras, one of
the stations being in Venezuela and the others in Columbia (Table
1). The station of Neira, in the central cordillera, is not far from
Manizales (5°15'N, 75°30’W) where the chromosome number of
A, acuminata was earlier reported by Coba de Gutiérrez and
Alvarado de Coral (1989). It could be noted that the chromosome
number of this species was also found to be 2n = 28 by Giusti,
the same year (1989), in Argentina.

ACKNOWLEDGMENTS

The first author would like to express his gratitude to the Com-
pania Fosforera Colombiana and Servicios y Consultoria Ltda
for their financial and logistical support while collecting indige-
nous material in Colombia. The Instituto Tecnoldgico de Costa
Rica and Seed Export of Guatemala must also be sincerely thanked
for seeds received from their countries. The help of R. Trahan
for the chromosome counts and of Dr. A. Munson for the revision
of the English manuscript was also greatly appreciated.

LITERATURE CITED

BouSsQUET, J. AND M. LALONDE. 1990. The genetics of actinorhizal Betulaceae,
pp. 239-261. Zn: C. R. Schwintzer and J. D. Tjepkema, Eds., The biology
of Frankia and Actinorhizal Plants. Academic Press, San Diego, CA

Curva, S. 1962. Studies on the breeding of Betu/a and Alnus species. (1) On the
differences of morphological characters and chromosome number between
Alnus hirsuta and Alnus hirsuta var. microphylla. J. Jap. For. Soc. 44: 237-
243

. 1966. Studies on the tree improvement by means of artificial hybridation

74 Rhodora [Vol. 96

and polyploidy in A/nus and Populus species. Oji Paper Co., Kuriyama, Japan.
Bull. Oji Inst. For. Tree Impr. No. 1. 42 pp

CoBA DE GUTIERREZ, B. AND C. ALVARADO DE CoRAL. 1989. Numero de cro-
mosomas de A/nus gene H.B.K. age Colombiana 6: 45-46.

DEL VALLE ARANGO, J. I. AND H. GONZALEZ PEREZ. 1988. Rendimiento y cre-
cimiento del cerezo (/ slr jorullensis) en la region central andina, Colombia.
Revta. Fac. nac. Agro 91.

Fur.ow, J. J. 1979. The : sy ieinatice of the American species of A/nus (Betu-
laceae). Rhodora 81: 1-121, 151-248

Giusti, L. 1989. El namero de cromosomas de tres especies tucumanas. Lilloa
37: 153-154.

Gran, K., C. M. LARSEN, C. S. LARSEN AND M. WESTERGAARD. 1941. Contri-
butions to the cytogenetics a forest trees. I]. A/nus studies. K. Vet.-og
Landboheisk. Aarsskr. 1941:

HALL, R. B. AND C. A. MAYNARD. ca Considerations’ in the genetic say
ment of alder, pp. 95-110. /n: J. C. Gordon, C. T. Wheeler and D. A. P
Eds., Symbiotic nitrogen fixation in the management of temperate ao
Oregon State University Press, Corvallis, OR.

Hotpripce, L.R. 1967. Life zone ecology, revised ed. Tropical ps Center,
San José, Costa Rica. Photographic suppl. by J. A. Tosi, Jr. 206 pp.

KopaMA, A. 1967. Karyological studies on root nodules of three species of A/nus.
Bot. Mag., Tokyo 80: 230-232.

. 1970. Cytological and morphological studies on the plant tumors. II.

Root nodules of Mi species of A/nus. J. Sci. Hiroshima Univ., Series B,

Div. 2, 13: 261-264.

DEPARTEMENT DES SCIENCES FORESTIERES
FACULTE DE FORESTERIE ET DE GEOMATIQUE
UNIVERSITE LAVAL, SAINTE-FOY

QUEBEC, CANADA GIK 7P4

C.G.

LABORATOIRE DE CYTOLOGIE ENVIRONNEMENTALE ET
DES RESSOURCES PHYTOGENETIQUES DU MENVIQ
DEPARTEMENT DES SCIENCES FORESTIERES

FACULTE DE FORESTERIE ET DE GEOMATIQUE
UNIVERSITE LAVAL, SAINTE-FOY

QUEBEC, CANADA GIK 7P4

RHODORA, Vol. 96, No. 885, pp. 75-96, 1994

EVOLUTION OF A FLORA: EARLY
CONNECTICUT VALLEY BOTANISTS

C. JOHN BURK

ABSTRACT

The first known botanical studies of the Connecticut Valley in western Mas-
sachusetts were initiated during the decade 1810-19. Four area physicians, David
Hunt of Northampton (1773-1837), Jacob Porter of Plainfield (1783-1846), Den-
nis Cooley of Deerfield Gehan siseee and pteppen West Williams of Deerfield
(1790-1855), kept ext d with Amos Eaton
at Williams College and aa Silliman at Yale. Porter, Cooley, and gran
published phenological accounts of local vegetation and other scientific obse
vations. Beginning around 1816, Edward Hitchcock (1793-1864) ene noe
botanical studies at Deerfield that culminated in the first flora of the region,
published at Amherst in 1829. Hitchcock acknowledged Cooley and Williams as
“early coadjutors in this work,” which served as a basis for Hitchcock’s 1833
compilation of the Massachusetts flora.

Key Words: botanical history, Connecticut Valley, Massachusetts, David Hunt,
Jacob Porter, Dennis Cooley, Stephen West Williams, Edward Hitch-
cock, Massachusetts flora

INTRODUCTION

The early 19th Century in both Europe and North America is
‘remarkable for the simultaneous appearance of groups of highly
intelligent leaders with such broad interests from politics to phi-
losophy and science that their impact on the future outweighed
their numbers” (Billings, 1985). Nonetheless, sources of infor-
mation on the development of botany in the Connecticut Valley
in western Massachusetts throughout this period are scattered,
fragmentary, occasionally contradictory, and often accessible only
from archival sources and special collections. The purpose of this
study is to assemble and synthesize information on botanical
activities in the region from the earliest known studies through
the completion of its first complete flora, Edward Hitchcock’s
“Catalogue of Plants Growing Without Cultivation in the Vicinity
of Amherst College’’ (Hitchcock, 1829), a list which served as a

ajor source for the first compilation of the Massachusetts flora
(Hitchcock, 1833)

Hitchcock, by then a prominent geologist who had served as
president of Amherst College, observed in his reminiscences

75

76 Rhodora [Vol. 96

(Hitchcock, 1863) that Amos Eaton, who had lectured in Amherst
and Northampton, Massachusetts in 1816 and 1817, was “the
chief agent of introducing a taste for [natural history] into the
Connecticut Valley” and that at the time “Dr. Stephen Williams,
Dr. Dennis Cooley, and myself, all of Deerfield, took hold of
mineralogy and botany with great zeal.”’ Widely regarded as a
charismatic teacher (Ballard, 1897), Eaton had in 1817 just re-
cently published the first edition of A Manual of Botany for the
Northern States (Eaton, 1817). This work described and classified
all genera known to occur north of Virginia according to a mod-
ified version of the Linnean system and included brief descriptions
of the more common species with notes on their occurrence in
the vicinity of Williams College and elsewhere.

Despite Hitchcock’s recollections, botanical work had been in
progress in the Connecticut Valley for some years prior to the
time of Eaton’s lectures, and when Eaton published a much ex-
panded and enlarged second edition of his manual in 1818, he
was able to include within its catalogue localities for at least 91
species, indicated by the letter ““N.,”’ that occurred along the Con-
necticut River between Northampton and Deerfield. These were
attributed to “Dr. D. Hunt, and Drs. Williams and Cooly (sic) of
Deerfield” (Eaton, 1818).

The work of Eaton, which was largely done outside the area,
and the geological contributions of Edward Hitchcock have been
studied in some detail (Ballard, 1897; Foose and Lancaster, 1981;
Knowlton, 1897; Smallwood, 1941). In addition, Jenny (1980,
1987) has recently described the ‘“‘rediscovery” of the herbarium
of Stephen Williams and the “‘small but highly competent group
of botanists” who lived in Deerfield around 1816. However, the
various and inter-related botanical contributions of Hunt, Wil-
liams, and Cooley, as well as those of Jacob Porter, whose plant
lists from the area were later acknowledged both by Eaton (1822,
1824) and by Torrey and Gray (1838-40), are less well known,
as is Hitchcock’s early work in botany. In this paper, Hunt, Porter,
Cooley, Williams, and Hitchcock are discussed chronologically
according to their dates of birth. This chronological order is rough-
ly correlated with the increasing importance of their contributions
to the development of botanical science in the region and with
the greater amount of information available on the later individ-
uals.

1994] Burk—Early Botanists Td
DAVID HUNT

Information regarding the botanical activities of David Hunt
(1773-1837) is particularly scarce. Hunt was a member of a well-
established Northampton family and sufficiently prominent as a
physician to be included by Stephen Williams in the American
Medical Biography (Williams, 1845). Hunt was also described,
in a tribute written long after his death, as “ta botanist and min-
eralogist of considerable note” (Springfield Union, 1904). Wil-
liams, who knew Hunt personally, reported that Hunt maintained
a correspondence with Benjamin Silliman and other academic
scientists, belonged to several medical and scientific societies, and
possessed a “cabinet of minerals” that was “‘rare ... and large,
for a private individual’ (Williams, 1845). The Centennial
Hampshire Gazette (1886) reprinted William’s tribute, including
a portrait of Hunt (Figure 1), along with additional biographical
materials by Hunt’s son Seth, who recalled that his father pos-
sessed a “large herbarium ... very neatly and scientifically ar-
ranged and labelled, by his own hands” (Centennial Hampshire
Gazette, 1886). Both David Hunt and his father Ebenezer Hunt
were among the sponsors of Eaton’s 1817 course of lectures in
Northampton and signed a letter of commendation, included in
the introduction of Eaton (1818), expressing their “entire satis-
faction” with Eaton’s work there. Although Porter (1821) reported
at least one of Hunt’s observations, that ‘““Doctor David Hunt...
has found at Northampton, numerous specimens of the sarracenia
with yellow blossoms,” Hunt apparently published nothing re-
lated to botany under his own name. Williams (1845) recalled
only a single paper by Hunt in a medical journal, a paper on a
case of lead poisoning, and concluded that “the productions of
his pen were not numerous.”

Hunt was married for nearly 42 years to Wealthy Dickinson,
a descendent of one of the first settlers of Hadley, Massachusetts.
The couple had thirteen children, ten of whom reached adulthood
(Centennial Hampshire Gazette, Sept. 6, 1886.) In the first of his
major geological papers, Edward Hitchcock acknowledged his
“particular obligations” to Hunt and to Benjamin Silliman
(Hitchcock, 1818), and Hunt’s nonprofessional interests almost
surely were centered more in geology than botany.

The location of Hunt’s herbarium, if it still exists, is at present

78 Rhodora [Vol. 96

x “i, Sif oy “ pig

Yip yg

.

..- Sts
“oS SS
= Ss aes

vy 4
Z\Ni
4 ‘

~

SS

"/ A - oe
"f ‘4 tL OLE
fe (po UTE ; ral ae +
1: f ff if; a)
J hi é / ‘4 <<
/ SAY f Jf 2S a“ =
ie / a |
Lf

Dr. David Hunt.

Figure 1. Dr. David Hunt, reprinted from Williams (1845) in Centennial
Hampshire Gazette (1886). Courtesy of Historic Northampton.

1994] Burk—Early Botanists 79

unknown. Although the Amherst College “cabinet” during Hunt’s
lifetime included the only institutional herbarium in the area.
Hitchcock (1863) does not mention Hunt’s specimens among the
botanical materials in the collections and I have not been able to
identify any Hunt specimens in the Amherst College materials
incorporated in MASs in the 1950s.

JACOB PORTER

Jacob Porter (1783-1846) was born in Abington, Massachu-
setts. He attended Williams College, but in 1802 transferred to
Yale University, receiving an A.B. degree there in 1803 (Dexter,
1885-1912). He then moved to Plainfield, Massachusetts where
he practiced medicine for the remainder of his life. He assembled
a collection of plants and minerals from the vicinity (Porter, 1834)
and, according to a local historian (Dyer, 1891), “although a
highly educated man, gave his attention mostly to literary pur-
suits, being well versed in botany and mineralogy.” In addition
to poetry, translations of religious tracts, and a history and geology
of Plainfield (Porter, 1834), Porter published two accounts of
phenological phenomena (Porter, 1818, 1821). These were among
a number of “floral calendars” published with the encouragement
of Benjamin Silliman (Silliman, 1818) in the early volumes of
The American Journal of Science. Silliman included these ac-
counts in response to a memoir by Jacob Bigelow setting out
research into “the comparative forwardness of the spring season
in different parts of the United States” (Bigelow, 1818).

Plainfield is situated at an elevation of 1684 feet on the western
edge of the Connecticut Valley watershed (U.S.G.S. topographic
map, Plainfield Quadrangle, Franklin Co., Massachusetts), and
Porter’s first calendar, was intended to “show that vegetation is
considerably later on the range of mountains, on which this place
is situated, than in the level parts of our country.”” The paper was
possibly also inspired by the well-publicized observations of
Humboldt and Bonpland relating altitude to climate (Humboldt
and Bonpland, 1805; Billings, 1985). Though neither author men-
tioned the other, Porter’s data could be compared with similar
though more extensive observations from Deerfield, Massachu-
setts in the Connecticut Valley to the east reported by Stephen
Williams (1819) in the subsequent issue of The American Journal
of Science. Williams, for example, found Hepatica triloba (H.

80 Rhodora [Vol. 96

americana) and Erythronium (E. americanum) in flower on April
26, 1818 in Deerfield, Massachusetts at an elevation of 140 feet
(U.S.G.S. topographic map, Greenfield Quadrangle, Franklin Co,
Massachusetts), while Porter reported these taxa first blooming
on May | of that year at the higher elevation.

Porter’s second and more extensive calendar covers the 1819
season and 1820 through July 30 and includes the observations
on Sarracenia by David Hunt with notes from a trip to North-
ampton. The work was abbreviated by Silliman in response to
“objections to highly detailed floral calendars” (editorial com-
ment in Porter, 1821); nonetheless, the entries for 1819 alone
include observations of more than 280 taxa of higher plants. These
are predominantly native species along with a number of crop
plants and ornamentals. For most taxa, common names are care-
fully assigned and often include part of'a Latin binomial; the entry
for May 24, 1820, for example, includes “Caulophyllum, glaucous
kalmia, painted trillium and water cress” in bloom. Porter’s style
is occasionally “literary” (“the petals of the roundleaved violet,
in particular, resemble specks of gold scattered around the path”
on May 6, 1819 in one instance), and he sometimes suggests herbal
remedies, including a tincture of goldenrod root in brandy as a
proven cure for spitting blood. In addition to its phenological
information, the value of the work lies in its listing of a wide
range of species from a diversity of natural habitats and from
locations, such as the banks of the Westfield River and a bog in
the town of Hawley, which are still botanically important.

Porter knew and corresponded with several academic scientists
of the period; these include Hitchcock, who cites several species
on Porter’s authority (Hitchcock, 1829), Eaton, who, in the third
(1822) and later editions of his Manual mentions Porter’s “‘im-
provements in the localities of plants ...” from “the mountain
range in Plainfield, Hawley, and Cummington, Mass.” and John
Torrey and Asa Gray, who include him in a list of contributors
of plants from Massachusetts (Torrey and Gray, 1838-40). He
was a member of several scientific societies, including the Lyceum
of Natural History of New York and the Massachusetts Society
for Promoting Agriculture. His first wife, Betsey Mayhew, of Wil-
liamsburg, Massachusetts, died a few months after their marriage
in 1813; his second wife, Sally Reed, outlived him for many years
(Dexter, 1885-1912).

1994] Burk—Early Botanists 81
DENNIS COOLEY

Two short biographical sketches of Dennis Cooley (1789-1860)
were published in the decades following his death. The first and
longer of these (Kenaston, 1863) consists of a two page account
of Cooley’s life in connection with a set of resolutions accepting
Cooley’s herbarium, which his widow had donated to the Mich-
igan State Agricultural College (now Michigan State University,
East Lansing). This account is consistent with genealogical data
in Sheldon (1895-96). The second, a two paragraph “‘sketch” by
Beal (1901) is inaccurate in some details, including the year of
Cooley’s birth.

Dennis Cooley was born in the Bloody Brook section of Deer-
field, Massachusetts in 1789 (Sheldon, 1895-96). The Cooley
family had been prominent in that area for many years and like
Hunt, Porter, and Williams, Cooley was educated as a physician.
By 1816, he was studying medicine in the office of William Stod-
dard Williams and his son Stephen West Williams (Williams,
1849). He received a degree from the Berkshire Medical Insti-
tution in 1822 (Beal, 1901), and reportedly, “‘. . . his leisure hours,
during his whole course of study, were spent in pastures, woods
and swamps in pursuit of botanical specimens” (Kenaston, 1863).
These botanical activities led to the 1818 contributions to Eaton
and a “floral calendar”’ for Deerfield (Cooley, 1820) which, though
brief, included a chart comparing seven phenological phenomena
over a five year interval, beginning in 1815.

Correspondence (now in the Amherst College Archives) be-
tween Hitchcock and Silliman and between Hitchcock and John
Torrey indicates that by the early 1820’s, Cooley and Hitchcock
had jointly prepared a manuscript catalogue of the Deerfield flora.
The fate of the Cooley/Hitchcock manuscript is considered later
in this paper. However, in a letter to John Torrey (Edward Hitch-
cock to John Torrey, Nov. 25, 1819 in President Edward Hitch-
cock Papers, Box 6, Folder 12, Amherst College Archives) Hitch-
cock commented that Cooley was not appreciated by the people
of Deerfield, who felt that he neglected his medical practice for
his botany; and subsequently, Cooley moved to Monticello, Geor-
gia “to seek his fortune” according to Hitchcock, who urged Sil-
liman to provide him with introductions, if possible, to southern
botanists (Edward Hitchcock to Benjamin Silliman, Oct. 17, 1822,

82 Rhodora [Vol. 96

Box 6, Folder 4, President Edward Hitchcock Papers, Amherst
College Archives).

In Georgia, Cooley practiced medicine for about three years, a
period “‘turned to a good account in his rapidly growing Herbar-
ium’ (Kenaston, 1863). For reasons of health, he returned North
and eventually relocated to Washington Township, Macomb
County, Michigan in 1827, taking his plant collections with him
then or at a later date. Cooley’s Massachusetts specimens alone
represented, according to Hitchcock (1829), “nearly all the plants
hitherto found in this district.”’ In Michigan, Cooley began a study
of the flora of Macomb County, sending lists of plants from that
area to Eaton as early as 1829 (Eaton, 1829, 1833, 1836; Eaton
and Wright, 1840). He was married to Elizabeth Anderson of
Deerfield in 1830; the couple had two children, both of whom
died in early childhood. Elizabeth Anderson died in 1834, and
in 1836, Cooley married Clarissa A. Andrews, who survived him
(Kenaston, 1863; Sheldon, 1895-96). Cooley apparently spent
most of his later years in Michigan; the Memorial Libraries at
Deerfield possess a single letter in his hand (Dennis Cooley to Eli
Cooley, June 2, 1850, Cooley Family Papers, Pocumtuck Valley
Memorial Association Library, Deerfield, Massachusetts) in which
he says that he had thought of visiting Deerfield, “but 1 suppose
that folks are not enlitened there and shall not come this year.”
Cooley continued the practice of medicine until 1856 and served
as postmaster of Washington township until 1859 (Beal, 1901).
In his reminiscences, Hitchcock (1863) reflected that Cooley “be-
came an excellent botanist, and even at a recent date, when he
died in Michigan, has pursued the subject with zest.” His her-
barium when presented to the Agricultural College was estimated
to contain more than 20,000 specimens and described as “‘es-
pecially rich in our indigenous flora’’ with ‘‘a large collection of
tropical, Californian, and Australian species . . . Many of the plants
were obtained by exchanges with Dr. Torrey, W. S. Sullivant, Dr.
Dewey, John Carey, and many other celebrated Botanists, ...”
(Kenaston, 1863). Beal (1901) stated the herbarium contained
around 4000 specimens, and Jenny (1987) noted that the contra-
diction in these estimates had not been resolved. Possibly Beal
may not have included the specimens obtained by exchange in
his estimate.

The Beal-Darlington herbarium at Michigan State University
now contains the Cooley collection, including material from Deer-

1994] Burk—Early Botanists 83

field collected between 1817 and 1821 (Jenny, 1987; Martha Case,
pers. corresp.).

STEPHEN WEST WILLIAMS

Stephen West Williams (1790-1855) was also a physician and
a member of a prominent Deerfield family. He attended Deerfield
Academy and then studied medicine as an apprentice to his father,
William Stoddard Williams, spending one winter as a student at
Columbia College in New York (Huntington, 1881; Viets, 1936).
By 1816, he was “‘enamoured” with the field of botany, inspired
by readings in the works of Barton, Bartram, Bigelow, Cutler,
Eaton, Elliott, Muhlenberg, Pursh, and others (Williams, 1849).
He recorded botanical observations as early as 1811 (Williams,
1819); and, along with Cooley and Hitchcock, began collecting
plants about 1816. By 1817 he had compiled an herbarium that
included around 500 pressed specimens (Jenny, 1980, 1987).

This 1817 herbarium was privately held until recently when its
botanical interest was recognized by Roberta Poland and its iden-
tity established by William Jenny (Jenny, 1980, 1987). The her-
barium is now preserved at the Pocumtuck Valley Memorial As-
sociation Library at Deerfield, Massachusetts. It contains
specimens of at least 360 taxa, ranging from common garden
flowers to local rarities such as Asplenium ruta-muraria. These
are mounted, often several to a sheet, in a bound volume, and
vary from whole pressed plants to small portions of a single shoot
or even single flowers. The specimens are identified by Latin and/
or common names either written directly on the sheet or on a
slip of paper attached to the specimen. Collection data, including
localities and dates, are generally lacking.

In 1818, Williams married Harriet Goodhue of Deerfield, the
daughter of Joseph Goodhue, an army surgeon (Viets, 1936).
About this time, Williams prepared a companion volume to his
herbarium, a handwritten manuscript now owned by the Flynt
Library at Historic Deerfield (Jenny, 1987). It contains indices to
the common and scientific names of the specimens in the her-
barium, an outline of the Linnean system, and a description of
each plant listed, its medicinal properties and other characteris-
tics. Cultural directions for garden plants such as hyacinth and
lilac are given, as are materials from various authors, including

84 Rhodora [Vol. 96

a passage from Bartram describing Sarracenia. Harriet Goodhue
drew and painted “from nature, and from other sources” a num-
ber of original illustrations for this interesting compilation, which
also includes illustrations clipped from contemporary publica-
tions such as The Monthly Flora.

Also around 1818, Williams sent Benjamin Silliman a box of
specimens of minerals to be identified, and, in responding, Sil-
liman encouraged both Hitchcock and Williams to contribute to
The American Journal of Science (Benjamin Silliman to Edward
Hitchcock, March 1, 1818, President Edward Hitchcock Papers,
Box 6, Folder 5, Amherst College Archives). In 1819, Williams
published his accounts of phenological phenomena for the years
1811, 1812, and 1818. This more original work (Williams, 1819)
seemingly has been confused by Viets (1936) with the manuscript
companion volume to the herbarium in the Flynt Library, which
was prepared at roughly the same period. The contrast between
Williams’s observations in the period 1811-12 and his obser-
vations in 1818 is striking and reflects his increasing taxonomic
skills and interests. In 1811, he reported a number of phenological
phenomena, particularly observations upon “the time of the ger-
mination, foliation, florification and frutification” of around 60
plant taxa, providing common names for all of these and scientific
names for approximately half. Of the taxa listed, over two thirds
were in Cultivation as either ornamentals or crop or garden plants,
while the remainder were natives such as bloodroot, Sanguinaria
canadensis, or non-native ruderals, such as the dandelion (as
Leontodon taraxacum). In 1812, he repeated observations on
many of these and included information on an additional 14 taxa.
In 1818, however, in addition to continued observations on the
plants previously noted, he provided data on an additional 92
species, listing all of these by Latin binomials, with common
names for most. The greater number of the 92 new entries are
native species taken from a diversity of natural habitats, including
wetlands (for instance skunk cabbage, as Pothos foetida) or upland
woods (Epigaea repens, Trillium cernuum).

Although, his career centered on teaching medical jurispru-
dence, chiefly at the Berkshire Medical Institution with stints at
Willoughby University in Ohio and the Dartmouth Medical School
(Viets, 1936), Williams retained a lively interest in botany for the
remainder of his life. He, along with Cooley, was acknowledged
by Hitchcock as ‘‘among my early coadjutors” in the 1829 Cat-

1994] Burk—Early Botanists 85

alogue, and he reciprocally (Williams, 1849) recognized the im-
portance of Cooley and Hitchcock and the early “‘botanical in-
vestigations” of Orra White, later Mrs. Hitchcock, to the
development of botany in Massachusetts. He remained in com-
munication with Amos Eaton and prepared etymologies for ge-
neric and subgeneric names that appeared in the 6th and later
editions of the Eaton Manual. Throughout his career, he wrote
extensively on the subject of medicinal plants (Deane, 1855);
among the most important of these works was the catalogue of
319 indigenous medicinal plants of Massachusetts in Williams
1849).

As Professor of Medical Jurisprudence at the Berkshire Medical
School at Pittsfield, Massachusetts, he became known as the lead-
ing medical historian and biographer of his day. He and Harriet
Goodhue had four children, two daughters, both of whom pre-
pared illustrations of medicinal plants for their father’s lectures,
and two sons, one of whom died in infancy. The second son was
trained as a physician in his father’s office and later settled in
Laona, Illinois. Stephen Williams himself retired in 1853 to Laona,
where, according to his daughter (Huntington, 1881), he spent his
remaining days in study and “riding over the beautiful prairie,
looking for new specimens of flowers, animals or birds.”

EDWARD HITCHCOCK

Edward Hitchcock (1793-1864), was born at Deerfield, Mas-
sachusetts, and attended Deerfield Academy. Because of weak-
ened eyesight, he abandoned plans to attend Harvard University
and took a position as principal of Deerfield Academy, a post he
held from 1815-19. By his own account (Hitchcock, 1863), he
began his studies in natural history about the time of Eaton’s
1816 lecture in Amherst, collecting, with Dennis Cooley, “‘nearly
all the plants, phenogamous and cryptogamous in the valley.” In
1817, he also initiated his lifelong correspondence with Benjamin
Silliman at Yale (Robinson, 1979).

In 1818, Hitchcock informed Silliman that he had collected
over 800 plants in the vicinity and that 150 of these had been
painted by “‘Miss White” (Edward Hitchcock to Benjamin Silli-
man, Sept. 28, 1818, President Edward Hitchcock Papers, Box
6, Folder 4, Amherst College Archives). Orra White (1796-1863)

86 Rhodora [Vol. 96

was a native of Amherst, Massachusetts who was teaching botany,
among other subjects, at Deerfield Academy and who had already
achieved some degree of success as a botanical illustrator (Jenny,
1987; Worman, 1989). In 1818, Hitchcock also undertook a course
of studies for the ministry.

In 1819, Hitchcock initiated a correspondence with John Tor-
rey, requesting help with the identification of a specimen of Po-
tentilla (Edward Hitchcock to John Torrey, July 20, 1819, Pres-
ident Edward Hitchcock Papers, Box 6, Folder 12, Amherst College
Archives). The following year he and Dennis Cooley promised
Torrey a list of plants “growing spontaneously in Deerfield’’ (Ed-
ward Hitchcock to John Torrey, April 28, 1820, President Edward
Hitchcock Papers, Box 6, Folder 12, Amherst College Archives)
which had been sent by the end of the summer (Edward Hitchcock
to John Torrey, September 8, 1820, President Edward Hitchcock
Papers, Box 6, Folder 12, Amherst College Archives).

Shortly before his ordination to the ministry, Hitchcock and
Orra White were married and after ordination, the couple moved
to Conway, Massachusetts where Hitchcock served as pastor from
1821-25 (Foose and Lancaster, 1981). Hitchcock continued to
pursue scientific interests at Conway, preparing both geological
and botanical papers for The American Journal of Science. In
1822, he expressed concern to Silliman that Thomas Nuttall, who
was visiting the area, was “exploring the country minutely” and
that “it would be a little mortifying if the facts that I have been
collecting should appear in Philadelphia first’ (Edward Hitchcock
to Benjamin Silliman, September 22, 1822, President Edward
Hitchcock Papers, Box 6, Folder 12, Amherst College Archives).

At this time, Hitchcock and Dennis Cooley were preparing the
catalogue of Deerfield plants intended for the American Journal
of Science (Edward Hitchcock to John Torrey, November 3, 1822,
President Edward Hitchcock Papers, Box 6, Folder 12; Edward
Hitchcock to Benjamin Silliman, October 17, 1822, President
Edward Hitchcock Papers, Box 6, Folder 4, Amherst College
Archives). Cooley then moved from Deerfield to Georgia, leaving
the catalogue for Hitchcock to finish (Edward Hitchcock to Ben-
jamin Silliman, October 17, 1822, President Edward Hitchcock
Papers, Box 6, Folder 4, Amherst College Archives). Silliman
delayed in publishing the manuscript, however, “lest objections
be made that too much space was allowed to communications
from a particular quarter” (Benjamin Silliman to Edward Hitch-

1994] Burk—Early Botanists 87

cock, December 5, 1822, President Edward Hitchcock Papers,
Box 6, Folder 5, Amherst College Archives), and in 1823 returned
the manuscript to Hitchcock, leaving “‘any further disposition of
it to your better judgement” with the suggestion that the work
should be revised as two papers with cryptogams treated sepa-
rately (Benjamin Silliman to Edward Hitchcock, May 5, 1823,
President Edward Hitchcock Papers, Box 6, Folder 5, Amherst
College Archives). At least two manuscript volumes of Orra White
Hitchcock’s botanical work survive from this period, a collection
of watercolors of grasses and other flowering plants entitled Her-
barium parvum, pictum dated “*1817,18” and now in the Archives
of Deerfield Academy (Jenny, 1987; Worman, 1989) and a sketch-
book of fungi entitled Fungi, selecti picti, dated 1821 and now in
the Smith College Archives. Figure 2 reproduces a representative
page from the latter manuscript; thirteen different fungi are de-
picted on the page; all but one is identified at least to genus. The
individual fungi are numbered, but the numbers are not always
consecutive, ranging on this page from 14 to 61; the numbers
thus may refer to numbered specimens in a collection or included
in a list.

Meanwhile, in a second major paper on the geology of the
region, Hitchcock (1823a) included “occasional Botanical notic-
es” that related a number of plant species to underlying rock types
and landforms, including a series of cryptogams associated with
mica slate. His discovery in a hilly pasture of Botrychium simplex
(Hitchcock, 1823b) was the “‘first new fern described by an Amer-
ican” (Tryon, 1969), and he also published observations on Gy-
ropodium coccineum, a new genus and species of fungus described
by Schweinitz and first seen at Deerfield by Dennis Cooley (Hitch-
cock, 1825). Orra White Hitchcock continued her work in sci-
entific illustration, including drawings to accompany her hus-
band’s publications and the drawings for Chester Dewey’s
publications on the genus Carex. Contrary to Fernald (1950),
Carex hitchcockiana was named by Dewey (1826) in honor of
both Edward and Orra White Hitchcock, the “lady, to whom I
am so greatly indebted for the figures which accompany this Car-
icography.”

In 1825, Hitchcock left the ministry, taking an appointment as
the first Professor of Natural History and Chemistry at Amherst
College. At Amherst, Hitchcock’s interests turned increasingly to
the field of geology, a subject that, throughout his life, he at-

88 Rhodora [Vol. 96

dt 5 het ia:
Spariced By om ted Lparcent

5 4
A
cofrccitled Lert,

Figure 2. Representative page from Orra White Hitchcock’s manuscript
sketchbook Fungi, selecti picti (1821). Courtesy of Smith College Archives.

tempted to reconcile with the tenets of religion (Foose and Lan-
caster, 1981; Robinson, 1979). Nonetheless, he continued botan-
ical studies and maintained a correspondence on his botanical
discoveries in the Amherst area with Torrey. Among the latter
were what he believed to be a new species of Malaxis (Edward
Hitchcock to John Torrey, July 5, 1827, President Edward Hitch-
cock Papers, Box 6, Folder 12, Amherst College Archives) that
he had found on Mt. Holyoke on a field trip with his botany class
and which he hoped to publish as Malaxis holyokea ‘in the hope
of exciting more zeal in natural history among the students here.”

In 1829, at the request of the members of the Junior class who
were attending his lectures in botany, Hitchcock published the
Catalogue of Plants Growing Without Cultivation in the Vicinity
of Amherst College. Apparently no copy of the Hitchcock-Cooley
manuscript that Silliman declined to publish now exists; hence it
is impossible to determine how that differed from the published

1994] Burk—Early Botanists 89

Hitchcock catalogue. The Hitchcock catalogue was intended to
include “all the indigenous and naturalized plants, that have been
discovered and ascertained, within forty or fifty miles of Amherst
College’ and, in addition, “such as are peculiar to the White
Mountains and the sea coast of New England” plus ‘‘a few others
mentioned on account of their being rare and interesting.’ Ex-
amples of the latter include the prickly pear, Cactus opuntia L.
(now Opuntia humifusa Raf.) from “East Rock, New Haven.”

Phenogamous plants, both angiosperms and gymnosperms, were
listed alphabetically by genus, with species alphabetized within
the genera. This resulted in such blatantly “‘unnatural’’ sequences
as genera 197, 198, and 199 which are Juncus, Juniperus, and
Kalmia respectively. Members of the class Cryptogamia were
organized alphabetically by genera under orders Filices, Musci,
Hepaticae, Algae, Lichenes, and Fungi with species alphabetized
by genera under their respective genera. Excluding species from
the White Mountains and the seacoast or included only because
they were rare or interesting, the Catalogue included by Hitch-
cock’s own count 1447 species within 531 genera that occurred
within 50 miles of Amherst. Of these, 997 species in 395 genera
were phenogamous while 450 species in 136 genera were cryp-
togams.

Localities were given for rare plants in the vicinity, including
sites attributed to Dewey, Eaton, and Porter. Some of these, in-
cluding Eaton’s record of Liquidambar styraciflua from North-
ampton, Porter’s Orontium aquaticum from Southwick, and an
unattributed report of Listera cordata from Plainfield, remain of
considerable interest.

In 1830, Hitchcock was appointed to head a survey of the
geology, mineralogy, botany, and zoology of Massachusetts (Foose
and Lancaster, 1981). The initial report of the survey (Hitchcock,
1833) contained a catalogue of the flora of Massachusetts. This
was organized according to a “Natural System of Prof. Lindley
as adapted to North American Plants by Torrey” and based on
Hitchcock’s 1829 Catalogue with additions and corrections along
with lists published by Dewey (1829) and by Bigelow (1824) and
a manuscript flora of New Bedford and Nantucket by Thomas
A. Green. W. S. Tyler, who knew Hitchcock both as a teacher
and as a colleague, observed (Tyler, 1873) that Hitchcock “‘did
but one great work at a time. But he was never afraid of having
too many smaller irons in the fire.’’ Nonetheless, the Massachu-

90 Rhodora [Vol. 96

es
a

near

cai

eee
a
- unt
lage a,

a gr Pag:
Bose osees
ag Og i gt

—

ange

me

aie w
a Seg sat tam Ss
- sap phe gee “at > Sc
she pad ail Se aa “ae ite ony
a ge an Sa < eS he
SSO See Oe oii stag tar On an eae
a ray ws Se gale
ra Ora Se rrr rR ae et net
soa eerie Se eS
pM Ti at a On ag cl agg regi Dae
a y oS SS <—S.
> ae coo
SO Oe
SsSsso es
Sos
aie
Se

Figure 3. Edward Hitchcock, engraving in Tyler (1873). Courtesy of Forbes
Library, Northampton, MA.

1994] Burk—Early Botanists 9]

setts Catalogue was apparently Hitchcock’s last botanical contri-
bution. The report of the survey (Hitchcock, 1841) dealt only with
the geology of Massachusetts, and in 1845, Hitchcock was ap-
pointed president of Amherst College, a position he occupied until
1854, when he returned to another decade of teaching, writing
and research. The engraving included in Tyler’s History of Am-
herst College (Figure 3) shows Hitchcock in his later years.

Hitchcock’s plant collections, however, formed the ‘“‘nucleus”
of the Amherst College Herbarium (Goodale, 1932). This her-
barium, if founded in 1829, the year of the publication of the
Catalogue, was according to Ewan (1969) the second institutional
herbarium established in the United States. [The first herbarium
in the United States was apparently at the Academy of Natural
Sciences in Philadelphia, an institution ‘“‘destined to receive the
nation’s oldest plant collections’ (Ewan, 1969).] Figure 4 isa view
of Amherst College published in 1839 (Barber, 1839); the her-
barium probably was located in the building with the tower. Day
(1901) noted that the Amherst College herbarium contained
“plants of local interest, many of which are no longer found grow-
ing in the vicinity,” and Stone (1913) reinforced Day’s statement
in his assertion that “Some of the plants observed by Dr. Hitch-
cock nearly a century ago and included in this list have not been
found since his time, and others have become rare.” Archival
materials related to the Amherst College Herbarium are virtually
nonexistent at either Amherst College or at the University of
Massachusetts, where the collections were incorporated in MASS
in the 1950’s. Those Hitchcock specimens I have seen have almost
certainly been remounted at some time. They consist of a pressed
plant or part of a plant taped to an herbarium sheet with a small
handwritten label on discolored stock usually pasted in the lower
left corner. The label contains the name of the plant and often a
date and/or location; the labels seem to be in Hitchcock’s hand,
as seen in his correspondence at Amherst College. In at least one
instance, the handwritten label is mounted attached to the spec-
imen itself. The collector’s name does not appear on the hand-
written slip, but a printed label attached in the lower right contains
the statement “ex Herb Edward Hitchcock, LLD.”

There are, however, in MASS a group of specimens from the
Amherst College herbarium that have been remounted in a man-
ner similar to those from the Hitchcock collection with older

92 Rhodora [Vol. 96

North-western view of Amherst College.

Figure 4. ‘North-western view of Amherst College,” engraving in Barber (1839).

discolored handwritten labels pasted on the left and printed labels
on the right. As in the Hitchcock specimens, the handwritten label
does not identify the collector, but unlike the Hitchcock speci-
mens, the printed label also bears no reference to collector. These
anonymous collections may well represent the “few hundred spe-
cies of plants in the vicinity ... previously prepared, I believe,
by members of the College” in Hitchcock’s 1863 account; if so,
they would also represent early 19th century plant collections
from the area.

Edward Hitchcock and Orra White had eight children, two of
whom died in infancy. One of their daughters, Emily Hitchcock
Terry (1838-1921), was a pioneer illustrator of the Minnesota
flora and an early contributor to Rhodora (Smith, 1992).

Hitchcock’s 1829 Catalogue was not reprinted or revised until
after his death; it then underwent three expansions and revisions
in less than 40 years (Tuckerman and Frost, 1875; Cobb, 1887;
Stone, 1913). It remains an essential document for understanding
both the flora and the development of botanical sciences in the
Connecticut Valley region of western Massachusetts in the early
19th Century.

1994] Burk—Early Botanists 93

ACKNOWLEDGMENTS

Iam particularly indebted to Ruth Mortimer, Karen Kukil, and
Sarah Black of the Smith College Rare Book Room for their
invaluable assistance in this and other projects; to John Lancaster
and the staff of the Amherst College Archives and Special Col-
lections; to David R. Proper of the Memorial Libraries at Historic
Deerfield; and to Elise Bernier-Feeley, reference librarian at Forbes
Library, Northampton, MA. I am grateful to Dario D’Arienzo,
Archivist of Amherst College, for permission to quote from the
papers of Amherst College President Edward Hitchcock, to David
R. Proper to quote from the Cooley Family papers in the Pocum-
tuck Valley Memorial Association Library, and to Margery Sly,
Smith College Archivist, for permission to include a represen-
tative page from the Orra White Hitchcock manuscript volume
Fungi, selecti picti in the Smith College Archives. Assistance and
helpful information was also provided by Martha Case at the
Beale-Darlington Herbarium; Leo Hickey at Yale University:
Karen Searcy and Roberta Lombardi at the University of Mas-
sachusetts, Amherst; Arvilla Dyer of the Plainfield Historical So-
ciety; and by Lynn Bassett and the staff of Historic Northampton.
In addition, the manuscript has profited greatly at various points
in its development from the comments of John Lancaster, Robert
McMaster, Ruth Mortimer, David Proper, Karen Searcy and Pa-
mela Weatherbee.

LITERATURE CITED

BALLARD, H. H. 1897. Amos Eaton. A pioneer of science in Berkshire County.
Collections of the Berkshire Historical and Scientific Society, Vol. 2. Press
of the Sun Printing Co, Pittsfield, M

Barser, J. W. 1839. Historical Collections, Being a General Collection of In-
teresting Facts, Traditions, Biographical Sketches, Anecdotes, &c.. Relating
to the History and Antiquities of Every Town in Massachusetts. Dorr, How-
land & Co, Worcester, MA.

BEAL, W.J. 1901. ee of Dr. Dennis Cooley. Report of the Michigan Academy
of Science, Vol.2: 108.

BIiGELow, J. 1818. tes serving to show the comparative forwardness of the
spring season in different parts of the United States. Memoirs of the Amer.
Academy of Arts and Sciences, Vol. IV, Part I: 77-8

1824. Florula bostoniensis, a Collection of Plants of Boston and its

Vicinity. Cummings, Hilliard, Boston.

94 Rhodora [Vol. 96

Brtuincs, W. D. 1985. The historical development of physiological plant ecology.
In: F. Cabot and H. A. Mooney, Eds., Physiological Ecology of North
American Plant Communities. Chapman and Hall, New k

CENTENNIAL HAMPSHIRE GAZETTE. Sept. 6, 1886. Sketches of prominent men
... Dr. David Hunt. p. 19.

Coss, N. A. 1887. A List of Plants Found Growing ee Within Thirty Miles
of Amherst. S. E. Bridgman & Co., Northampton,

Mass. fanee: Jour. Sci. II: 254.
Day, M. A. 1901. Herbaria of New England. Rhodora 3: 67-71.
DEANE, J. 1855. Biographical sketch of the late Steph. W. Williams. The Boston
Medical and Surgical Journal, Vol. LIII, no. 2. Thursday, Aug. 9, 1855
Dewey, C. 1826. Caricography, (continued). Amer. Jour. Sci. X: 274
_ 1829. Catalogue of plants. Jn D. D. Field, A History of the County of
Berkshire by Gentlemen of the County. Samuel Bush, Pittsfield, MA.
DEXTER, F. B. 1885-1912. Jacob Porter, pp. 603-606. Jn: Biographical Sketches
of the Graduates of Yale College: With Annals of the College History, Vol.
5. Holt, New York.
Dyer, C. N. 1891. History of the Town of Plainfield, Hampshire county, Mass.
Gazette Printing Co, Northampton, MA.
Eaton, A. 1817. A Manual of Botany for the Northern States. Websters and
Skinners, Albany, N
1818. A Manual of Botany for a Northern and Middle States, 2nd ed.
Websters and Skinners, Albany, N
1822. A Manual of Botany : the Northern as Middle States of
Amiencs: 3rd ed. Websters and Skinners, Albany, N
24. A Manual of Botany for the Northern a Middle States of
America, 4th ed. Websters and Skinners, Albany, NY.
1829. rage : Botany for North America, 5th ed. Websters and
Sinners: Albany,
——. 1833 “Mania a Botany for North America, 6th ed. Oliver Steele,

Albany, N
1836. ie of Botany for North America, 7th ed. Oliver Steele,

Albany. NY.

ND = WriGut. 1840. North American Botany, 8th ed. Elias Gates,

Troy, N
N, J., a. 196 9. hoe History of Botany in the United States. Hafner

“ehne Co., ork.

FERNALD, M. L. cn ae s Manual of Botany, 8th ed. American Book Co.,
New York.

Foose, R. M. AND J. LANCASTER. 1981. Edward Hitchcock: New England ge-
ologist, minister, and educator. Northeastern Geology, v. 3 (January, 1981):
13-17

Goopa_e, A. S. 1932. Notes from the Amherst College Herbarium. Rhodora
34: 34-37

Hircucock, E. 1818. Remarks on the geology and mineralogy of a section of
Massachusetts on Connecticut River, with a part of New-Hampshire and
Vermont. Amer. Jour. Sci. [: 105-116.

1994] Burk—Early Botanists 95

1823a. A sketch of the geology, mineralogy, and scenery of the regions
contiguous to the river Connecticut. Part I. Amer. Jour. Sci. VI: 1-86.
. 1823b. Description of a new species of Botrychium. Amer. Jour. Sci.
VI: 103-104.
1825. Physiology of the Gyropodium coccineum. Amer. Jour. Sci. IX:
56-60.
—. 1829. Catalogue of Plants a Without i sia in a Vicinity
of Amherst College. J. S. and C. Adams, and Co. Printers, Amherst, MA.
1833. Report on the — Miner ralogy, ie any, oe net le of
Neher tncens J. S. and C. Adams, and Co. Printers, Amherst, MA.
41. Final Report soa © coon of Massachusetts. J. S. ic Adams,
ana Co. Printers, Amher:
een ener a ies College, Historical, ae Bio-
graphical, and Autobiographical. Bridgman & Childs, Northam
Humpo pt, A. AND A. BONPLAND. 1805. Essai sur la Geographie te nee
Chez te ile Schoell et Compagne, Paris.
HuntTIncTon, H. H. 1881. Stephen West Williams. Memorial Biogs. New En-
gland Historical and Genealogical Soc.: 389-397.
jee W.W. 1980. American Herbarium of Stephen West oe er
Historic Deerfield Quarterly Vol. XIX, no. 1 (January, 1980) u
1987. “American Herbarium’: key to Deerfield’s cee caeeace
Historical Journal of Massachusetts 15: 61-69.
Kenaston, C. A. 1863. Second Annual Report of the Secretary . the State
Board of Agriculture of the State of Michigan. Lansing, MI. pp.
KNOwLTON, F.H. 1897. Some early American botanists. Amos aa The Plant

orld I, no 2
Porter, J. 1818. Floral cdiendax for Plainfield, Massachusetts. Amer. Jour. Sci.
I: 254-255

. 1821. Floral and Miscellaneous Calendar of Plainfield, Mass. Amer.
Jour. Sci. HI: 273-284.
ioe, Topographical Description and Historical Sketch of Plainfield, in
shire County, Massachusetts. Prince and Rogers, Greenfield, MA.
come. G. 1979. Edward Hitchcock, pp. 49-83. Jn: Leonard G. Wilson, Ed.,
Benjamin Silliman and his Circle: Studies on the Influence of Benjamin
Silliman on Science in America. Science History Publications, New York
SHELDON, G. 1972. A History of Deerfield, Vol. Il. Facsimile of the 1895-96
edition, New Hampshire Publishing Co., Somersworth, MA
SILLIMAN, B. 1818. Professor Bigelow, on the comparative forwardness of the
spring, in different parts of the United States. Amer. Jour. Sci. I: 76-77.
SMALLWoop, W. M. 1941. Natural History and the American Mind. Columbia
University Press, New York, NY.
SmitH, B.S. 1992. A Painted Herbarium: The Life and Art of gy Hitchcock
Terry (1838-1921). University of Minnesota Press, Minn
SPRINGFIELD UNION. May 29, 1904. Northampton Hunts ae for wealth,
learning and political power in early days of United States—among descen-
dents governors and statesmen
Stone, G. E. 1913. A List of Plants Growing Without Cultivation in Franklin,
Hampshire and Hampden Counties, Massachusetts. Press of Carpenter &
Morehouse, Amherst, MA.

96 Rhodora [Vol. 96

Torrey, J. AND A. Gray. 1838-1840. Flora of North America. Wiley and
Putnam, New York.

Tryon, R. 1969. Pteridology. pp. 97-102. Jn: J. Ewan, Ed., ee History of
Botany in the United States. Hafner Publishing Co., New

TUCKERMAN, E. AND C. C. Frost. 1875. A Catalogue of Plants a aon
Cultivation Within Thirty Miles of Amherst College. Edwin Nelson, Am-
herst, MA.

TyLer, W. S. 1873. History of Amherst College During Its First Half Century.
Clark W. Bryan and Co., Springfield, M

Viets, H. R. 1936. Stephen West Williams. Dictionary of American Biography
XX. Charles Scribner’s Sons, New York, 1963. p. 291

WILuaMs, S. W. 1817. Botanical Description and Medical, Culinary & Other
Uses of the Plants in the First Volume of My American Herbarium. Auto-
graph Manuscript, Henry N. Flynt Library of Historic Deerfield, Inc., Deer-
field, MA.

1819, Floral calendar kept at Deerfield, Massachusetts, with miscella-
neous remarks. Amer. Jour. Sci. I S12:

1845. Dr. David Hunt, pp. 298-299. Jn: American Medical Biography
or Memoirs of Eminent Physicians. Reprinted 1967, Milford House Inc.,
New York

Report on the indigenous medical botany of Massachusetts. The
Transactions of the American Medical Association, Vol. II: 863-927.

Worman, E. C. 1989. The watercolors and prints of Orra White Hitchcock. AB
Bookman’s Weekly, Vol. 83, no. 7: 646-668.

DEPARTMENT OF BIOLOGICAL SCIENCES
SMITH COLLEGE
NORTHAMPTON, MA 01063

RHODORA, Vol. 96, No. 885, pp. 97-101, 1994
NEW ENGLAND NOTE

GLYCERIA MAXIMA (POACEAE) IN NEW ENGLAND

JEANNE E. ANDERSON AND A. A. REZNICEK

Tall Mannagrass, G/yceria maxima (Hartman) Holmburg is a
stout perennial Eurasian grass found as a native species through-
out most of northern Eurasia from the British Isles east to Japan
and Kamchatka, albeit absent from the extreme north and most
of the southwest of this range. (Tutin et al., 1980; Freckmann and
Reed, 1979). Dore and McNeill (1980) provide one of the few
keys among North American manuals which distinguishes G.
maxima from the relatively similar G. grandis. They separate G.
maxima by the length of the lower glume (2-3 mm versus 1|.2-
1.5 mm in G. grandis) and by the ascending and stiff panicle
branches and rough sheaths versus nodding panicle branches and
smooth sheaths in G. grandis. Fernald (1950) also distinguishes
G. maxima, but under the name G. spectabilis Mert. & Koch.
The synonymy ascribed to G. maxima is thoroughly detailed by
Freckmann and Reed (1979).

Glyceria maxima was collected in the United States for the first
time in 1975 in Racine County, Wisconsin (Wilhelm & Schulen-
berg 2161 Mor). A second collection located approximately 12
miles east of the first collection in Oak Creek, Milwaukee County,
WI was made in 1979. (Reed s.n. Uwsp). Both collections were
reported in 1979 publications (Swink and Wilhelm, 1979; Freck-
mann and Reed, 1979).

Glyceria maxima has been twice reported erroneously from the
United States, Jordal (1951) reported G. maxima from near Fair-
banks, Alaska, but the specimen (Jordal 3539, Micu) is G. grandis.
More recently, an historical record of this species (as Glyceria
maxima subsp. maxima) from Mattituck, Long Island (Latham
35967, NYS) mapped in the Preliminary Vouchered Atlas of New
York State Flora (New York Flora Association, 1990) has proven
to be a large individual of G. canadensis (Gordon Tucker, pers.
comm., 1992).

Three populations of G/yceria maxima were discovered and
mapped by Marc Lapin in 1990 on the Massachusetts Audubon
Society’s Ipswich River Wildlife Sanctuary in the towns of Tops-
field and Wenham, Essex County, Massachusetts (Lapin, 1990).

97

98 Rhodora [Vol. 96

A fourth population, also in Wenham, was located by Tim Smith
in 1993. Voucher specimens were identified by A. A. Reznicek
as cited below. Specimens examined: Massachusetts. Essex Coun-
ty: Wenham, Ipswich River Wildlife Sanctuary, Great Wenham
Swamp (behind Perkins Island), 23 Jul 1992, Anderson s.n. (MICH),
Topsfield, Ipswich River Wildlife Sanctuary, Mile Brook Swamp,
6 Aug 1992, Anderson s.n. (MICH). An additional voucher from
the Wenham site has also been deposited: Anderson s.n. October
9, 1992 (NEBC). These collections are the first for New England
and represent the third documented occurrence in the United
States.

The Wenham population, within a high marsh system of the
Great Wenham Swamp, forms a near monoculture of about two-
thirds of an acre. There is a dense, but relatively shallow rhizome
system reaching about halfa meter in depth. Aerial shoots exceed
two meters in height. Only a few individuals of Cephalanthus
occidentalis, Fraxinus pennsylvanica, and Lythrum salicaria per-
sist within this patch. The grass is also growing in a smaller stand,
6.5 meters square, located roughly 60 meters southeast of the
larger Wenham population close to an edge of Perkins Island.

The Topsfield populations both occur within a shrub swamp
community along Mile Brook, a tributary of the Ipswich River.
The largest of these patches covers 2 acres. A portion of this
population exists as a floating mat of individual plants and rhi-
zomes which can be readily dislodged if disturbed. Downstream
along the stream channel, a smaller stand measures about 15
meters in diameter. Sampling completed in one section of this
swamp in 1990 determined that four non-native species, G/yceria
maxima (sub G. septentrionalis), Rhamnus frangula, Iris pseu-
dacorus, and Lythrum salicaria accounted for 67% of the relative
cover within this community. Of these, G. maxima was the most
prevalent species providing 39% of the relative cover (Anderson,
1991). This larger patch is discernible at least as far back as 1957
on black and white aerial photographs, however, G. maxima was
not reported for Essex County by Harris (1975).

Dore and McNeill (1980) and Dore (1947) note that the first
herbarium specimens of this species in North America were col-
lected from the vicinity of Hamilton, Ontario, Canada in 1943.
They also report that the species’ distribution in the New World
is concentrated primarily in Ontario. Other stands are known
from Somenos Lake, British Columbia (Ca/der & MacKay 31984

1994] New England Note 99

DAO) and St. Johns and Brigus, Newfoundland (Day, 1991; Rou-
leau and Lamoureux, 1992). The specimen reported in Scoggan
(1978) from northern Alberta has been determined to be G. gran-
dis (Susan Aiken, pers. comm., 1992).

Glyceria maxima is a large, aggressive species with the ability
to form huge stands in wetlands. Freckmann and Reed (1979)
noted that G. maxima dominated at least 15 acres of the Oak
Creek site in 1979. This invasive tendency has been evident in
Ontario over the past fifty years and has been a subject of study
in Europe (e.g., Buttery and Lambert, 1964). Even within its
native range, the ability of G. maxima to create virtual mono-
cultures under varying levels of disturbance is of conservation
concern. Burgess et al. (1990) report that the swamp community
type dominated by G. maxima more than doubled its area in the
Ouse Washes, England between 1972 and 1988, largely in re-
sponse to an overall increase in the incidence of summer flooding.
This spread of G. maxima has reduced plant species diversity on
the site, and as a consequence, reduced the number of seed-pro-
ducing plants (particularly within the Cyperaceae and Polygonace-
ae) available to wintering seed-feeding ducks. G. maxima is re-
ported to be a poor foodplant for wintering grazing waterfowl and
a poor nesting substrate for many common wetland species (Bur-
gess et al., 1990).

The spread of Glyceria maxima in the United States should be
carefully monitored as it has the potential to be a very serious
invader of natural wetlands. Vegetative dispersal 1s reported from
Canada, where 53 separate stands of G. maxima were located in
one thirteen mile stretch of the Mississippi River in Ontario (Gut-
teridge, 1954). The reliance on vegetative mechanisms of repro-
duction and dispersal is also suggested by Dore’s (1953) report
that only | to 9% of the florets set good grains. Further information
on the spread of existing North American populations, coupled
with additional data on dispersal mechanisms and control meth-
odologies should be gathered before the species becomes thor-
oughly entrenched.

ACKNOWLEDGMENTS

Weare grateful to Gordon Tucker and Susan Aiken for checking
the identity of the specimens supporting the New York and AI-
berta reports of Glyceria maxima respectively. Jacques Cayouette

100 Rhodora [Vol. 96

kindly sent us a great deal of information about the Canadian
distribution of this species. Bob Freckmann and Ted Cochrane
similarly us useful information on the Wisconsin sites. Pa-
mela B. Weatherbee alerted us to the literature on one of the
Wisconsin Meaceniie of G. maxima, for which we are thankful.
Thanks are also extended to Les Mehrhoff, Sally Rooney, Mark
Anderson, Michael Taylor, Roberta Lombardi, Jody Hall, and
Keith Killingbeck for review of vouchers at their respective her-
aria.

LITERATURE CITED

ANpbERSON, M. M. 1991. Population structure of Lythrum salicaria in relation
to wetland community structure. Master of Science thesis, University of New
Hampshire, Durham, New Hampshire.

Burcess, N. D., C. E. EVANS AND G. J. THoMAs. 1990. Vegetation change o
the Ouse Washes Wetland, England, 1972-88 and effects on their conser-
vation importance. Biol. Conserv. 53: 173-189.

Buttery, B. R. AND J. M. LAMBERT. 1964. Competition between G/yceria max-
ima and Phragmites communis in the region of Surlingham Broad. I. The
competition mechanism. J. Ecol. 53: 163-181

Day, R. 1991. Asecond oe locality for English Water Grass G/yceria
maxima. Osprey 22(1): 1

Dore, W. G. 1947. ae maxima in Canada. Canad. Field-Naturalist 61:
174

———. 1953. A Forage Species for Wet Land. Proceedings of the Fifth Annual
Meeting, Eastern Canadian Society of Agronomy, Agriculture School, Ste.
eieseeeas Quebec, pp. Al2-A19.

McNEILL. 1980. Grasses of Ontario. Monograph 26. Biosyste-

matics se Institute, Research Branch, Agriculture Canada. Ottawa,

FERNALD, M. L. 1950. Gray’s Manual of Botany, 8th ed. (corrected). D. Van
Nostrand Company, New Yor

FRECKMANN, W.R. AND D. M. REED. 979. Glyceria maxima: a new, potentially
troublesome wetland weed. Bull. . Club Wisc. 11(2&3): 30-35.

GUTTERIDGE, R. L. 1954. Glyceria maxima on the Mississippi River, Ontario,
1953. Canad. Field-Naturalist 68: 133-135.

Harris, S. K. 1975. The Flora of Essex County, Massachusetts. Peabody Mu-
seum, Salem, MA.

JorpDaL, L.H. 1951. Plants from the vicinity of Fairbanks, Alaska. Rhodora 53:
156-160

Lapin, M. 1990. Ipswich River Wildlife Sanct Ecological Management Plan,
Unpublished document. Massachusetts Audubon Society, Lincoln, MA.

New YorkK FLorA AssociATION. 1990. Preliminary Vouchered Atlas of New
York State Flora. Edition |. New York State Museum Institute, Albany, NY.

RoOULEAU, E. AND G. LAMOuREUX. 1992. Atlas of the Vascular Plants of the

1994] New England Note 101

Island of Newfoundland and of the Islands of Saint Pierre and Miquelon.
Saint-Henri-de-Levis, Quebec, Canada.

Scoccan, H. J. 1978. The Flora of Canada, Part 2—Pteridophyta, Gymnosper-
mae, Monocotyledonae. National Museum of Natural Sciences Publications
in Botany, No. 7(2), Ottawa, Canada.

Swink, F. AND G. WILHELM. 1979. Plants of the Chicago Region. The Morton
Arboretum, Lisle, Illinois.

Tutin, T. G., V. A. HEywoop, N. A. BurGcess, D. M. Moore, D. H. VALENTINE,
S. M. WALTERS AND D. A. Wess, Eds. 1980. Flora Europaea, Vol 5. Alisma-
taceae to Orchidaceae (Monocotyledons). Cambridge University Press.

MASSACHUSETTS AUDUBON SOCIETY
SOUTH GREAT ROAD
LINCOLN, MASSACHUSETTS

A.A.

THE aneeeae OF MICHIGAN HERBARIUM
NORTH UNIVERSITY BUILDING

ANN ARBOR, MICHIGAN 48109-1057

RHODORA, Vol. 96, No. 885, pp. 102-103, 1994
NEW ENGLAND NOTE

NOTES ON THE RHODE ISLAND FLORA

RICHARD L. CHAMPLIN

The following list includes plant species found and collected
by the author over a period of years, proving that even in the
smallest state one can turn up species new to the area or otherwise
of interest botanically. Specimens of all species have been sent to
the Herbarium of the New England Botanical Club.

Thelypteris palustris (Salisb.) Schott var. pubescens (Laws.) Fern.

A far cry from their usual habitat, specimens of the Marsh Fern
grow in Newport, RI. not in an open, sunny marsh, but next to
the mortar cementing together the granite stonewall of a private
estate. About three feet above the sidewalk they share the ex-
traordinary habitat with Woodsia obtusa and Asplenium platy-
neuron. In five years of observation I have failed to find these
Marsh Ferns fruiting, but the sterile fronds have all the diagnostic
traits of the species— forking veins, the frequent twist of the whole
frond, smooth green axis, black rootstock, etc.

Dryopteris filix-mas (L.) Schott

When Frederick Law Olmsted (1822-1903) designed the land-
scape in Newport for John W. Auchincloss in 1887 at the estate
known as Hammersmith Farm, he included a rock garden for
alpine plants. It consisted of a long trench blasted out of calcium-
bearing shale. A bed of the Dryopteris was planted along the north
ridge. Not only has the fern survived, it has flourished, and its
spores have developed new plants in various parts of Newport,
three of which have been located so far. Male fern, a mountain
dweller, has come down to altitudes of less than one hundred feet.

Panicum amarum EIl.

M. L. Fernald’s 8th Edition of Gray’s Manual of Botany, 1950,
puts Connecticut as the northernmost range of this grass. But it
has reached Rhode Island and was found along Narragansett Bay
at Plum Point, Saunderstown (Washington County) in 1989. It
grows on the beach with Prunus maritima, and occasional Geas-
ters.

102

1994] New England Note 103

Carex kobomugi Ohwi

This exotic carex occurs on Rhode Island sand dunes at East
Beach, Charlestown (Washington County), found and collected
in 1981.

Epipactis helleborine (L.) Crantz

This orchid has found a congenial habitat along a busy street
in the City of Newport. One of its several locations is under the
spread of another European species, the European Hornbeam,
Carpinus betulus.

Epifagus virginiana (L.) Bart.

An even more striking example of adaptability than the above
is the case of the Beech Drops. This saprophyte grows in eastern
North America under the American Beech (Fagus grandifolia).
In Newport, however, it has become established under a specimen
of European Beech (Fagus sylvatica) on an estate where many
European Beeches have come to maturity, but no American
Beeches. The nearest known occurrence of Beech Drops and
American Beech is about twelve miles away. The Epifagus is not
known to grow natively in Europe, which makes this choice of
habitat all the more surprising.

Morus rubra L.

Rhode Island does not fall within the natural distribution range
of the Red Mulberry. Yet in Foster (Providence County) on the
outskirts of the Lester Steere apple orchard grow without culti-
vation five specimens of the fruit-bearing tree. The probable ex-
planation? This land in the nineteenth century was a cherry or-
chard. In that age, amid their cherries, orchardists planted Red
Mulberries, which ripen simultaneously with the cherries. Thus
they hoped to draw the bird population away from the cherries
to the less valuable fruit. These specimens have apparently es-
caped to the surrounding, rich woods where they grow in company
with Christmas fern, Basswood, Butternuts, and white Baneberry.

PRISCILLA ROAD
JAMESTOWN, RHODE ISLAND 02835

RHODORA, Vol. 96, No. 885, pp. 104-108, 1994
NEW ENGLAND NOTE

DESMONEMA WRANGELLII (AG.)
BORNET ET FLAHAULT, A
NEW RECORD FOR MAINE

L. C. Cot, JR.

Key Words: Desmonema, Scytonema, freshwater, basalt

During a field trip in search of Micropterus dolomieui Lacepede
in south-central Maine, my collecting gear got caught on the rough
and irregular surface of a massive basalt rock partially submerged
in the water of Crawford Pond, Union, Maine. This pond in Knox
County has been previously described (Colt, 1977, 1985). This
rock surface lies on the east side of the large central island in the
northern section of the pond, and is identified by the presence of
several bronze plaques attesting to the unfortunate but accidental
deaths of several people in that area of the pond’s waters.

While retrieving my collecting gear I found that the entire sur-
face of the rock below the water line was covered by a brownish
felt-like mat of microscopic plants. Several samples of the plants
were gathered and preserved in Transeau’s Fluid (Prescott, 1962)
since the proper equipment for examination of the samples was
not immediately available.

Subsequent examination of the plants with a light microscope
in my laboratory at the University of Massachusetts Dartmouth
revealed that the plants were a cyanophyte alga, and using Smith
(1950), the plant was identified as Desmonema wrangellii (Ag.)
Bornet et Flahault 1887. The characteristics of the Maine plants
matched the description by Smith, and subsequently, the descrip-
tive material in Tilden (1910). Smith notes that the plant has
parallel trichomes within common sheaths, and penicillate tufts
which readily dissemble upon removal from the substrate. Tilden
adds that heterocysts are basal, the plant mass caespitose and
penicillate, and the filaments somewhat dichotomously divided.
The summary environmental characteristics provided by
VanLandingham (1982) generally fit. Although my plants were
collected in a pond, there is frequent wind-aided water drift in
the vicinity of the rock surface, along with small wave action,
thus moving water is present, a characteristic of Desmonema
habitat reported to frequently occur.

104

1994] New England Note 105

Drouet, in his monumental Revision of the Nostocaceae with
Cylindrical Trichomes (1973), placed Desmonema in synonymy
with Scytonema hofmannii Ag. According to Drouet, the plant
identified by Bornet & Flahault as Desmonema is the same as the
plant named Thorea by Agardh in 1812, as Oscillatoria by Agardh
in 1813, as Calothrix by Bornet et Flahault in 1887, as Coleo-
desmium by Borzi in 1887, and as Dillwynella by Kuntze in 1891.

Drouet (1973) reduced the Nostaceae with cylindrical tri-
chomes to three genera by characterizing the morphology of the
terminal cells in the following manner.

Scytonema: Terminal vegetative cells at first hemispherical,
becoming almost spherical

Calothrix: Terminal vegetative cells at first hemispherical,
becoming blunt-conical or cylindrical with rotund
tips.

Raphidopsis: Terminal vegetative cells at first hemispherical,
becoming more or less acute-conical.

Not all cyanophyte specialists agree with Drouet’s revision.
While Drouet examined an enormous number of specimens, more
than 30,000, in researching his revisions of the cyanophytes, he
came to depend on such characters as granulation (among the
Oscillatoriaceae) and the shape of the terminal cell of the vege-
tative portions of the plants in both the Oscillatoriaceae and the
Nostocaceae. Drouet (1973) states that the constant features of
morphological value in classification of the Oscillatoriaceae are
those of the protoplast alone. The characteristics of the external
sheath, according to Drouet, are the products of the history of
the environment and as a result, prove to be of no value for the
purposes of classification. He further notes, writing about the
Oscillatoriaceae, that the ‘‘end ofa fragmented trichome proceeds,
under favorable environmental conditions, to assume a config-
uration characteristic of the species”’.

There seems to be no disagreement among cyanophycologists
that there is a great deal of morphological variation among the
blue-green algae, and that this variation is primarily a function
of environmental or habitat variation. Baker and Bold (1970) in
their two-year culture studies in the Oscillatoriaceae note that the
vegetative terminal cells in their cultures showed considerably
more morphological variation than Drouet’s work would seem
to allow as permissible. Kathleen Baker of the University of Ha-

106 Rhodora [Vol. 96

waii (Pers. comm.), who was kind enough to respond to my query
on Desmonema, suggested that much work remains yet to be done
by microbiologists to fully establish the relationships among the
cyanophytes.

While my plants show the hemispherical-approaching-spheri-
cal cells of Scytonema, my experiences with desmid (Desmidi-
aceae) cells, where morphological changes in the new vegetative
semicells appear to be the result of environmental factors present
immediately prior to or during their development, leads me to
wonder if such dependence on the morphology of one cell is
appropriate. New desmid semicells can and do show distinct mor-
phological differences from the parent (semicell), and if these
changes persist into the next vegetative generation, subsequent
observations can lead to the erroneous conclusion that the “new”
semicells represent a new taxon. I suggest that what are needed
are discrete studies at the molecular level which would identify
the relationships among or the differences between the many al-
ready described cyanophyte taxa. J am aware that some studies
have been undertaken, but these few appear to represent only a
minute fraction of what is needed to clearly and accurately define
the cyanophytes.

When I checked the New England Algal Data Bank (NEADB,
1993) to determine whether or not this plant had been previously
collected in the New England Region, the only reported collection
was by W. A. Setchell (1895) in New Haven, Connecticut. Given
the large number of plants at the Maine site, the lack of other
collections in New England seemed curious. Geitler (1932) reports
Desmonema as being collected in England and the Pyrenees, and
Whitford and Schumacher (1969) and Cocke (1967) report Des-
monema from North Carolina, but I am not aware of any other
reports from New England.

Given Drouet’s assignment of Desmonema to Scytonema, it
seemed reasonable to determine whether or not Scytonema hof-
mannii had been reported in New England. According to data in
the NEADB Conn and Webster report S. hofmannii in Middlesex,
County, CT, in 1908, and Holden is reported in both Tilden
(1910) and Hylander (1925, 1928) to have collected it in Litchfield
County, CT. Setchell (1896) collected S. hofmannii in Norfolk
County, Massachusetts, and Collins (1904b) collected S. hof-
mannii in Grafton County, New Hampshire. The latter two col-
lections are noted in Tilden (1910). Bailey (1847b) reported S.

1994] New England Note 107

hofmannii from Providence County, Rhode Island, and Flint
(1916) collected S. hofmannii in Chittenden County, Vermont.
The question is, do these all represent the same species, Scyto-
nema, as reported by Drouet?

In that my own research is with certain of the chlorophytes, I
feel somewhat uneasy in making any final determination on the
systematic position of the cyanophyte plants which I collected in
Maine. Despite the conclusions of Drouet, however accurate as
they may eventually be shown to be, and given the uncertainties
expressed in the literature and by personal communication I be-
lieve the appropriate action at this time is to report the collection
of Desmonema wrangellii (Ag.) Bornet et Flahault for the second
time in New England and for the first time in Maine.

I am indebted to Robert K. Edgar of the University of Mas-
sachusetts Dartmouth for access to his personal library and for
his taking the time to assist in searching for clues to the systematics
of this cyanophyte enigma.

LITERATURE CITED

BAILEY, J. W. 1847b. Notes on Algae of the United States. Amer. J. Sci. Arts.
Ser. 2. 3(9):399-403
BAKER, AILSIE F. AND HAROLD C. Bo_p. 1970. Phycological Studies. X. Taxo-
nomic Studies in the Oscillatoriaceae. Univ. Texas Publication No. 7004.
Cocke, ELron C. 1967. The Myxophyceae of North Carolina. Wake Forest
Univ., Winston-Salem, NC.
Cotuins, F. S. 1904b. Algae of the Flume. Rhodora 6(69):18 1-182.
Cott, L. C., Jr. 1977. A new station for Coleochataceae in New England.
Rhodora 79(818): 300-304.
Vaucheria undulata Jao again in New England. Rhodora 87(852):
597-599.
1993. The New England Algal Data Bank, Library, Marine Biological
Laboratory, Woods Hole, M
Conn, H. W. AND L. W. WEBSTER. 1908. A Preliminary Report Upon the Fresh
Waters of Connecticut. Conn. Geol. Nat. Hist. Surv., Bull No. 10, Hartford,
Cr

DrouetT, FRANCIS. 1973. Revision of the Nostocaceae with Cylindrical Tri-
c s. Hafner Press, NY.

FLINT, L. “EL 1916. A First Partial List of a of Vermont. M. S. Thesis,
University of Vermont, Burlington, Ver

GEITLER, L. 1932. Cyanophyceae. In: es ene Flora von
Deutschland, Osterreich und der Schweiz. Band XIV. Leipz

HyLanper, C. J. 1925. The Algae of Connecticut. Ph.D. dion: Yale
University, New Haven, CT.

108 Rhodora [Vol. 96

. 1928. The Algae of Connecticut. Conn. State Geol. Nat. Hist. Surv.,
Bull. No. 42, Hartford, CT

Prescott, G. W. 1962. Algae of The Western Great Lakes Area. Wm. C. Brown,
Co., Dubuque, IA.

SETCHELL, W. A. 1895. Notes on some Cyanophyceae of New England. Bull.
Torr. Bot. Club 22(10): 424-431.

SMITH, G. M. 1950. The Fresh-Water Algae of the United States. McGraw-Hill
Book Co., NY.

TILDEN, J. 1910. The Myxophyceae of North America. Minnesota Geol. Surv

inneapolis, MN.

Vani deo SAM. 1982. Guide to the Identification, Environmental Re-
quirements and Pollution Tolerance of Freshwater Blue-Green Algae (Cya-
nophyta). United States Environmental Protection Agency, Environmental
Monitoring and Support Laboratory, Cincinnati, O

WHITFORD, L. A. AND G. J. SCHUMACHER. 1969. A Manual of Fresh-Water Algae
in North aie. North Carolina Agricultural Experiment Station, Raleigh,

bee

DEPARTMENT OF BIOLOGY
UNIVERSITY OF MASSACHUSETTS DARTMOUTH
N. DARTMOUTH, MA 02747

RHODORA, Vol. 96, No. 885, p. 109, 1994

THE BIOLOGY AND TAXONOMY OF THE
PORTULACA OLERACEA L. COMPLEX IN
NORTH AMERICA: ERRATUM

JAMES F. MATTHEWS

Figure 3, page 170, is incorrect. The seed surface shown is that
of a stellate tuberculate Portulaca pilosa L. The seed surface pat-
tern for the specimen labelled Portulaca retusa Engelm. as cited
in Table 3, page 178, is shown here.

MATTHEWS, JAMES F., DONNA W. KETRON, AND SANDRA F. ZANE. 1993. The
biology and taxonomy of the Portulaca oleracea L. complex in North Amer-
ica. Rhodora 95(882): 166-183.

DEPARTMENT OF BIOLOGY
UNIVERSITY OF NORTH CAROLINA AT CHARLOTTE
CHARLOTTE, NORTH CAROLINA 28223

109

RHODORA, Vol. 96, No. 885, p. 110, 1994
BOOK REVIEW

Cheryl McJannet, George Argus, Sylvia Edlund, and Jacques Cay-
ouette. 1993. Rare Vascular Plants in the Canadian Arctic.
Syllogeus Series No. 72. Canadian Museum of Nature. Pre-
paid orders available from the Canadian Museum of Nature,
Direct Mail Section, P.O. Box 3443, Station “‘D,” Ottawa,
Ontario, Canada K1P 6P4. (Price $17.95 U.S. and overseas,
$14.92 Canadian orders).

This 79 page booklet covers the 236 rare vascular plant taxa
found in the Canadian Arctic. A typical entry includes name,
synonymy, family, phytogeography, where found in the Canadian
Arctic, and its status in the various Provinces and the Northwest
Territories. Each entry also contains a dot range map.

This is a valuable publication for those interested in the arctic
flora. It follows the format for the other publications dealing with
rare plants in the Canadian Provinces. The price seems excessive
for such a publication. Mail orders must be prepaid by check,
Visa, or Mastercard.

C. BARRE HELLQUIST

DEPARTMENT OF BIOLOGY

NORTH ADAMS STATE COLLEGE
NORTH ADAMS, MASSACHUSETTS 01247

110

RHODORA, Vol. 96, No. 885, pp. 111-112, 1994
BOOK REVIEW

Bernard Boivin. 1992. Les Cypéracées de l’est du Canada. Pro-
vancheria 25, Herbier Louis-Marie, Universite Laval (230
pp., including 69 figures). In French. Available from Herbier
Louis-Marie, Faculté des sciences de l’agriculture et de I’al-
imentation, Université Laval, Quebec, Canada, G1K 7P4.

This treatment of the sedges of the six eastern provinces of
Canada (Ontario, Quebec, New Brunswick, Nova Scotia, Prince
Edward Island, and Newfoundland/Labrador) is the first segment
of the Flore de l’est du Canada, planned as a series in Provancheria
by Bernard Boivin, one of the outstanding botanists of this cen-
tury. A student of both Marie-Victorin and Fernald, his work in
floristics and taxonomy was wide-ranging and critical, and is best
represented by the Enumération des plantes du Canada (1967),
and in the 5-part Flora of the Prairie Provinces (1979). Written
by Boivin before his death in 1985 at age 68, the Cyperaceae
section was edited for publication by Paul Catling and Jacques
Cayouette of Agriculture Canada, who are responsible for an up-
to-date bibliography for the Cyperaceae in eastern Canada and a
list of new records for this region discovered after Boivin’s com-
pletion of the manuscript.

The flora includes two new taxa (one nothospecies, one new
form) and ten new combinations. The twelve genera (Hemicarpha
is included in Scirpus) and 217 species are represented, with an
additional thirty nothospecies. Carex includes 168 species and
29 nothospecies. Each taxon is provided with a short description
and a discussion, including distribution, ecology, and taxonomic
notes, that is one of the more valuable parts of the book. Keys
have been tailored to each genus and work very well. Primary
keys to some genera (Cyperus, Eleocharis) and sections (Carex
sect. Bracteosae) are geographical, while the majority are based
on morphology. Keys for Carex are particularly effective. An ex-
tended discussion is provided for the notoriously difficult Carex
section Ovales, which Boivin has divided into eight groups based
on distinctive morphological features.

The figures are perhaps the weakest aspect of the book. The
half-page figures showing from two to nine species provide good
overall illustrations of inflorescences for all species, but often omit
critical details of achenes, scales or perigynia. In particular, veins

Lit

hi Rhodora [Vol. 96

or nerves are not clearly shown in these figures. The placement
of figures is also poor, often separating related taxa or the de-
scription and illustration for a particular species. For example,
Fimbristylis and Bulbostylis, two easily confused genera, appear
on separate figures, as do the two species of Scleria. For Carex
nardina, the text and figure are separated by fifteen pages. Al-
though the illustrations have scale bars, it is not clear what mea-
surements these represent.

Boivin’s taxonomic approach is characteristically idiosyncratic.
He frequently combines species, often to an extent greater than
that of Cronquist (Gleason and Cronquist, 1991). As examples,
Scirpus longii is reduced to S. cyperinus var. brachypodus, and S.
acutus and S. validus are combined as S. /acustris. Many species
of Carex sect. Ovales have been reduced to varietal status (C.
bicknellii to C. brevior var. crawei, for example). Boivin makes
significant use of the taxonomic rank of ‘forma’, and recognizes
numerous nothospecies. Since it preserves Boivin’s understanding
of the family, this treatment does not recognize many recent
taxonomic revisions in Carex and Scirpus, including several by
the editorial team.

Despite some flaws, this volume is an excellent reference for
botanists interested in the sedge flora of northeastern North Amer-
ica. Descriptions are clear and concise, the keys work well, and
the discussions provide valuable habitat and taxonomic infor-
mation. The taxonomic treatment provides an interesting alter-
native to other recent works, and may stimulate further research
to resolve the conflicting opinions of two of the most notable
floristic botanists of our time.

LITERATURE CITED

Borvin, B. 1967. Enumération des plantes du Canada. V. Monopsides. Le Na-
turaliste Canadien 94: 131-157.

. 1979. Flora of the Prairie Provinces. Part IV. Monopsida. Provancheria
5. 251 pp

GLeason, H. A. AND A. Cronquist. 1991. Manual of Vascular Plants of North-
eastern United States and Adjacent Canada, 2nd ed. New York Botanical
Garden, Bronx, New York.

L. A. STANDLEY

VANASSE HANGEN BRUSTLIN, INC.
101 WALNUT STREET, P.O. BOX 9151
WATERTOWN, MA 02272

RHODORA

JOURNAL OF THE
NEW ENGLAND BOTANICAL CLUB

GORDON P. DEWOLF, JR., Editor-in-Chief

Associate Editors

DAVID S. BARRINGTON RICHARD A. FRALICK
A. LINN BOGLE GERALD J. GASTONY
DAVID E. BOUFFORD C. BARRE HELLQUIST
CHRISTOPHER S. CAMPBELL MICHAEL W. LEFOR
WILLIAM D. COUNTRYMAN ROBERT T. WILCE

GARRETT E. CROW

VOLUME 96

1994

The New Gngland Rotanical Club, Inc.

Harvard University Herbaria, 22 Divinity Ave., Cambridge, Mass. 02138

RHODORA, Vol. 96, No. 885, pp. 115-120, 1994

INDEX TO VOLUME 95

Acadia National Park 122-128

Achene micro-morphology as a sys-
tematic aid to the series placement of
Svenson’s undesi ages oe
(Cyperaceae) species.

Agoseris glauca var. ee 393,
396

Algal vegetation of the York River Es-
tuary and the adjacent open coast of
southern Maine 285-324

Allium geyeri var. chatterleyi 417-418

Alpine zone 52-75

Ambrosia sandersonii 396-397

Angelo, Ray. Some new records for New
England states. III. 188

Antennaria aromatica 261-276

Antennaria media 261-276

Antennaria pulvinata Greene: The le-

itimate name for 4. aromatica Evert

(Asteraceae: Inuleae). 261-276

Antennaria rosea 261-276

Antennaria umbrinella 261-276

Aquilegia grahamii 412-413

Arctomecon 197-213

Arenaria rubella var. filiorum 393

Arisaema ringens 254-260

Aristida beyrichiana 25-37, 189-190

Aristida stricta 25-37, 189-190

Artemisia campestris ssp. borealis var.
petiolata 397-398

Aster 236-240

Astereae 234-253

Astragalus toanus var. scidulus 403

Astragalus zionis var. vigulus 404

Bahama flora 369-391

Baskinger, S. M. 422-424

Belland, René J. and W. B. Schofield.
Salix vestita Pursh and Saxifraga op-
positifolia L.: arctic-alpine species new
to Nova Scotia. 76-78

Biology and taxonomy of the Portulaca
—— L. (Portulacaceae) complex

n North America. The, 166-183
iclionta 240

BOOK REVIEWS
Atlas of Ontario Mosses. 82-84
Women Botanists of Ohio: Born be-

Bregin, Candice L.

Britton, Donald M.

Bromeliaceae 342-347

Brunton, Daniel F. and Donald M.
Britton. Jsoetes prototypus (Isoeta-
ceae) in the United States. 122-128

Buffalo Academy of Sciences 137-154

Cahill, Jr., James F. 97-112

Camissonia bairdii 408-409

Camissonia claviformis var. crucifor-
mis 409

Campbell, Michael T. 11-20

Canada 129-136, 225-333, 234-253

Carex castanea x C. debilis, a new nat-
ural hybrid from Ontario. 129-136

Carex debilis 129-136

Carex praegracilis W. Boot (Cypera-
ceae) new to New England. 425-426

Carex section Sylvaticae 129-136

Catling, P. M. Carex castanea x C. de-
bilis, a new natural hybrid from On-
tario. 129-136

Catling, ue M. Rediscovery of the

ned strawberry, Fragaria
multe ae. one
Paul M. and Stu ay

“Phe “hybrid origin of Eleoc ae ma-
COUNIL.

Ceanothus greggi var. franklinii 413-
414

a orbiculatus 188
hang, Youngjune 11-20
ee 38-51, 352-368
Chamaesvyee deltoidea 38-51
Chamaesyce deltoidea ssp. adhaerens
48-49

Chamaesyce deltoidea ssp. deltoidea
4

Chamaesyce deltoidea ssp. pinetorum
49-50

115

116

Chamaesyce deltoidea ssp. serpyllum
50

Chemung County 137-154
Chionaster sp. 21

Chloromonas spp. 21-24

G. Antennaria pul-

or A. aromatica Evert (Asteraceae:
Inuleae). 261-276

Chromosome number determinations
in fam. Compositae, tribe Astereae.
V. Eastern North American taxa.
234-253.

Chromosome numbers 234-253

24]

Coastal seaweeds 285-324

Cogbill, Charles V. The interplay of
botanists and Potentilla robbinsiana:
discovery, systematics, collection, and
stewardship of a rare species. 52—75

Colorado 11-20, 155-157
olt, L. C., Jr., S. M. Baskinger, M. O.
Fontaine, T. W. Howie, K. Richards

d K. A. Simmons. Some new re-

cords for algae in ae Mas-
sachusetts. 422-4

Compositae hay

Connecticut 97-112

Conservation 122-128

Conservation biology 52-75

Contributions of William T. Gillis
(1933-79) to the of the Baha-
mas. The,

Conyza 240

Costa, Martin 285-324
Croptilon 240

Crow, Garrett E. 348-351

Cuscuta glomerata 158-165

Cyperaceae 85-96, 129-136, 214-224,
349, 425-426

Cytogeography 234-253

Cytology 166-183

Discovery of Subularia aquatica L. in
Colorado and the extension of its
range. 155-157

Disjunct populations 11-20

Distribution 11-20, 25-37, 38-51, 76-

Rhodora

[Vol. 96
78, 122-128, 158-165, 166-183,
184-187, 197-213, 225-233, 325-

341, 348-351, 425-426

Dominican Republic 325-341

Dr. Thomas F. Lucy: early botanist of
the Chmung River Valley, New York.
137-154

Duval, Brian. Snow algae in northern
New England. 21-24

Ecology 225-233

Eleocharis 85-96, 214-224

Eleocharis x macounii 85-96

Elmira Academy of Sciences 137-154

Elmira College Herbarium 137-154

Endangered species 38-51, 52-75

Endemism 225-233

Erigeron sionis var. trilobatus 398

Eriogonum corymbosum var. albiflo-
rum 410-411

Eriogonum corymbosum var. thomp-

411

Eriogonum jamesii var. higginsti 411-

412

Eriogonum racemosum var. nobilis
Erythronium bracteatum 119
Eshbaugh, W. Hardy 369-391
Estuarine seaweeds 285-324
Euphorbiaceae 38-51, 352-368
Euthamia 241

Fire and Cuscuta alr Choisy in
Ohio: a connectio 158-165

Florida 342-347, Tae:

Fontaine, M. O.

Fragaria multicipita 225-233

Gard, C. Luke 11-20

Germplasm protection 225-233
Gilia latifolia var. imperialis 409-410
Gillis, William T. 391

Glyceria acutiflora 188

Gulf of St. Lawrence 225-233

Gulf coastal plain 6-10

Gutierrezia 241

Haplopappus lignumyiridis 398
Harmon, William E. 155-157
Haufler, Christopher H. 11-20

1994]

Hehre, Edward J. 285-324

Hehre, Edward J. and Arthur C. Ma-
thieson. Porphyra amplissima (Kjell-
man) Setchell et Hus: new records of

oe. seaweed in Southern
e, New Hampshire and Mas-

serene 184-187

Heminway, Sarah W. 91-112

Herbarium collections 52-75

Herndon, Alan. A revision of the Cha-
maesyce deltoidea (Euphorbiaceae)
complex of southern Florida.

Herndon, Alan. Notes on Chamaesyce
(Euphorbiaceae) in Florida. 352-368

Heterotheca 241

Hispaniola 325-341

Historical ecology 52-75

Howie, T. W. 422-424

Hybrid 85-96, 129-136

Hybrid origin of Eleocharis macounii.
The, 85-96

Aymenoxys acaulis var. nana 398-399

Identity of Erythronium bracteatum
(Liliaceae). The, 119-121

Interplay of Botanists and Potentilla
robbinsiana: discovery, systematics,
collection and stewardship of a rare
species. The, 5

prea ae taxonomy and compari-

ns of nrDNA ITS-2 sequences of

oe ringens (Araceae). 254-
2

Tsoetes eas (Isoetaceae) in the
United States. 122-128

Isozyme ek 11-20

ITS-2 254-260

Judd, Walter S. and Eliza Karpook.
Taxonomic studies in the Miconieae
(Melastomataceae). V. Miconia sten-
obotrys, circumscription and _ rela-
tionships. 325-341

Karpook, Eliza 325-341

Kass, Lee B. 137-154

Kass, Lee B. and W. Hardy Eshbaugh.
The contributions of William T. Gil-

Index to Volume 95

1

lis (1933-79) to the flora of the Ba-
hamas. 369-391

Kelloff, Carol L. and Lee B. Kass. Dr.
Thomas F. Lucy: early botanist of the
Chemung River Valley, New York
137-154

Ketron, Donna W. 166-183

Ko, Sung Chul, Steve L. O’K.
and Barbara - Schaal. Intraspecific
taxonomy and comparisons of nr-
DNA rs sequences of Arisaema
ringens (Araceae). 254-260

Lane, Meredith A., Zhongren Wang,
Christopher H. Haufler, Philip A.

Schott, Greg Spielberg, Kathryn E.
Stoner, Mark B. Taylor, and Ilan Ya-
rom. Rhododendron albiflorum Hook.
(Ericaceae): one taxon or two? 11-20
Lepidium montanum var. claronensis
01-402

Liliaceae 119-121

Long Island Sound 91-112

Lucy, Thomas Francis 137-154

Luther, Harry E. A new record for Ti/-
landsia (Bromeliaceae) in Florida.

47
Lygodesmia grandiflora var. doloresen-
sis 399
Lygodesmia grandiflora var. entrada
399

MacDonald, Laura A. 97-112

Maine 112-128

Marine and brackish water species of
Vaucheria (Tribophyceae, Chryso-

phyta) from Connecticut. The, 97-
112

Massachusetts 113-118, 184-187,
285-324

Mathew, Brian. The identity of Ery-
thronium bracteatum (Liliaceae).
119-121

Mathews, James F., Donna W. Ketron
and Sandra F. Zane. The biology and
taxonomy of the Portulaca oleracea

118

L. (Portulacaceae) call in North
America. -18

Mathieson, Arthur A 184-187

Mathieson, Arthur C., Edward J. Hehre,
and Martin Costa. Algal ae
of the York River Estuary e
adjacent open coast of uae
Maine. 285-324

McCormac, James S. and Jennifer L.
Windus. Fire and Cuscuta glomerata
Choisy in Ohio: a connection? 158—
165

Melastomataceae 325-341

Menapace, Francis J. Achene micro-
morphology as a systematic aid to the
series placement of Svenson’s undes-
ignated Eleocharis (Cyperaceae) spe-
cies. -

Mentzelia goodrichii 407-408

sda ee var. coriacea 393

Mexico

Miconia Chaenopleura aa 341
Miconia stenobotrys 325-34
Micro-morphology oe
Mohave Desert 197-213

0)
Myriophyllum spicatum 250

Najas minor 348-349
Nelson, Deanna R. and Stanley L.
elsh. Taxonomic revision of Arc-

tomecon Torr. & Frem. 197-213

Nelson, Jody K. and William E. Har-
mon. Discovery of Subularia aqua-
tica L. in Colorado and the extension
of its range. 155-157

New England Notes 188, 422-424,
425-426

New Hampshire 52-75, 184-187, 285-
324

New record of Tillandsia (Bromeli-
aceae) in Florida. A, 342-347

New records for algae in southeastern
Massachusetts. Some, 422-424

New species 1-5, 277-284, 392-421

New taxa and new nomenclatural com-
binations in the Utah Flora. 392-
42]

New York 137-154

Rhodora

[Vol. 96

Notes on Chamaesyce (Euphorbiaceae)
in Florida. 352-368

Nova Scotia 76-78

nrDNA 254-260

Nuttall, Thomas 52-75

Oakes, William, 52-75
Ohio 158-165

O’Kane, Steve L. 254-260

Oldham, M. J. 425-426

Ontario —136

Orchidaceae 1-5, 277-284, 418-420

Pacific Northwest 11-20

Padgett, Donald J. and Garrett E. Crow.
So unwelcome additions to the
flora of New Hampshire. 348-351
diomelum aromaticum var. barnebyi

404-407
Peet, Robert K. A taxonomic study of
Aristida stricta and A. beyrichiana.
25-37
Peet, Robert K. A taxonomic study
Aristida stricta and Aristida eee
ana, adden 90
Pelexin asinus "379-282
Pelexia cundinamarcae 277-279
Pelexia sinuosa 281-284
Penstemon franklinit 414-416
Penstemon humilis var. desereticus 416
Phacelia cronquistiana 2
Phacelia heise var. atwoodii 403
Phacidium sp.
Phylogenetic species concept 325-341
225-233

Setchell et H

New Hampshire and Meccan.
184-187

Portulaca oleracea 166-183
Potamogeton crispus 349-350
Potentilla robbinsiana 52-75

Prairie 158-165

Québec 85-96, 225-233

1994]

Range extension 155-157

Rediscovery of the many-crowned
strawberry, Fragaria multicipita Fer-
nald. 225-233

Reviewers of Manuscripts 427

Revision of the Chamaesyce deltoidea
(Euphorbiaceae) complex of south-
ern Florida. A, 38-51

Reznicek, A. A. and M. J. Oldham.
Carex praegracilis W. Boot (Cyper-
a new to New England. 425-

Rhadendron albiflorum Hook. (Eri-
eae): one taxon or two? 11-20

nee DNA 254-260

Richards, K. 422-424

Salix vestita Pursh and Saxifraga op-
positifolia L.: Arctic-alpine species
new to Nova Scotia. 76-78

Salt marsh 97-112, 113-118

Sarracenia purpurea L. ssp. venosa
(Raf.) Wherry var. burkii Schnell
(Sarraceniaceae)—a new variety of the
Gulf coastal plain. 6-10

Saxifraga oppositifolia 76-78

Schaal, Barbara A. 254-260

Schiedeella romeroana (Orchidaceae,
Spiranthi nae), a new and interesting
species cn eee 1-5

Schneider, Craig W., Laura A. Mac-

nald, James F. Cahill, Jr., and Sar-

ah W. Heminway. The marine and

brackish water species of Vaucheria

(Tribophyceae, ee es from
Connecticut. 97-1

aa Donald E. ae

L. ssp. venosa (Raf.) Wherry var. bur-

kii Schnell (Sarraceniaceae)—a new

variety of the Gulf coastal plain. 6-10

Schott, Gary 11-20

Scirpus fluviatilis 349

Scotiella eryophila 21

Seaweeds 184-187, 285-324

Seed morphology 166-183

Selenotila sp. 21

SEM 214-224

Semple, John C., Jie Zhang, and Chun-
sheng Xiang. Chromosome number

Index to Volume 95

Lig

determinations in fam. Compositae,
tribe Astereae. V. Eastern North
American taxa. 234-253

Senecio castoreus 399-400
Senecio musiniensis 400-401

Simmons, K. A. 422-424
Snow algae in northern New England,
4

1-2

Solidago 241-248

Some new records for New England
states. III. 18

Southeastern U.S. 6-10, 25-37

Southern Florida 38-51, 342-347

Species biology 166-183

Spielberg, Greg 11-20

Spiranthes romanzoffiana var. diluvi-
alis 419-420

Stoner, Kathryn E. 11-20

Strawberry 225-233

Subspecies complex 38-51

Subularia aquatica 155-157

Susquehana Valley 137-154

Szlachetko, Dariusz L. Schiedeella
romeroana (Orchidaceae, Spiran-
thinae), a new and interesting species

T

Szlachetko, Dariusz L. Three new spe-
cies of the genus Pelexia (Orchida-
ceae, Spiranthinae) from Columbia.
277-284

Taxonomic revision of Arctomecon
Torr. & Frem. 197-21 3
Taxonomic

obotrys, circumscription and _ rela-
tionships. 325-341

Taxonomic study of Aristida stricta and
A. beyrichiana. A, 25-3

Taxonomic study of Aristida stricta and
Aristida beyrichiana; addendum. A
189-190

Taxonomy 25-37, 225-233

Taylor, Mark B. 11-20

Thompson, Philip A. 11-20

Three new species of the genus Pelexia
(Orchidaceae, Spiranthinae) from
Columbia. 277-284

Tillandsia fasciculata 342-347

120

Transcribed spacer 254-260
Tribophyceae 97-112
Trifolium friscanum 407
Tuckerman, Jr., Edward 52-75
Upper Susquehana flora 137-154
Utah flora 392-421
Vaucheria arcossonensis 101-103
Vaucheria compacta 103
Vaucheria compacta var. koksoakensis
103-104

Vaucheria coronata 104
104-105
5

Vaucheria nasuta 107-108

Vaucheria subsimplex Crouan from a
Massachusetts salt marsh: first Unit-
ed States record. 113-118

Vaucheria velutina 108-109

Vaucheria vipera 109

Vaucheriaceae 97-112

Virgulus 246

Wang, Zhongren 11-20

Rhodora

[Vol. 96

Webber, Edgar E. Vaucheria subsim-
plex Crouan from a Massachusetts salt

sh: first United States record.
113-118

Weeds 348-351

Welsh, Stanley L. 197-213

Welsh, Stanley L. New taxa and new
nomenclatural combinations in the

21

Windler Award winner announced,
Third Annual Richard and Minnie.
19]

Windus, Jennifer L. 158-165

Xiang, Chunsheng 234-253

Yarom, Ilan 11-20
York River, Maine 285-324
Yucca vespertina 417

Zane, Sandra F. 166-183
Zhang, Jie 234-253
Zuckia brandegei var. plummeri 393

Vol. 95, No. 883/884, including pages 197-428, was issued March 22, 1994.

Coes

a eae S: “Send to: Gordon p DeWolf,

Tt . Ni ew eiiceast Benin Club,

2 pina Avenue, Cambridge, Massachusetts 021 2138

_ RHODORA |
GORDON P. DEWOLE, JR, Raitor-in-Chiet

| Associate Editors: :
DAVID S. CONANT
ee “LISA A. STANDLEY . sae:
: “RHoDORA (ISSN 0035 -4902). Puplished: four times a year a |

East 10th St., Lawrence, KS 66044 and printed by Allen Press, Inc.,
1041 New ‘Hampshire St., Lawrence, KS 66044. Second class

- RHODORA, RO. Box 1897, Lawrence, KS 66044

: “RHODORA isa journal of bot any devoted pameely to Nowh es ; i Le

Authors are encouraged to submit manuscripts of scientific papers

-and notes relating to the dentate. distribution, ecology, paleo 4

~ botany, or conservation biology of this or floristically related. areas.

ee $40.00 per calendar year, net, postpaid, in funds ee
ao poet lee at aah in the United States currency. Remittances payable cae Es
ORA. Send to » RHODORA, wee Sag 1897, eee ee he ae ee

Deve: S897.

— MepERsies: Regular 935; ae Family $45. Applicat ae fom es ae

ee BACK joes AND SINGLE COPIES: Some available: oe aie

tion and prices will be furnished upon request to the Treasure oe

ce ADDRESS. CHANGES: In order to receive the nex number

~RHODORA, changes. must be received by be busi ness. office ay or 0
the first day of January, April, July, or Octob

INFORMATION Fc FOR CONTRIBUTORS: inside back k cover, ay o

and April.

anaging oe RHODORA
a as Long Hill Road ©
West Brookfield MA 01585.

tn én as . 4 SC ARICL/IANCC- 790 AR.

This paper meets the req ANSI/ . 1992 (Permanence of Paper).

April, July, and October) by The New England Botanical Club, 810 - ae
6

_ pending at Lawrence, KS. ! -OSTMASTER: Send address changes to el

RHODORA

OURNAL OF THE
NEW ENGLAND BOTANICAL CLUB

Vol. 96 April 1994 No. 886

RHODORA, Vol. 96, No. 886, pp. 121-169, 1994

THE ENVIRONMENT AND VASCULAR FLORA OF
NORTHEASTERN IOWA FEN COMMUNITIES

JEFFREY C. NEKOLA

ABSTRACT

While fen communities have been long known from northwestern Iowa, only
recently have they been found to occur in northeastern Iowa, south of the limit
of Wisconsinan glaciation. Since 1984, over 2333 potential fen sites were surveyed
in this region, and 160 were found extant. Extant fen sites can be divided into
one of five classes dependent upon surficial geology: pre-IIlinoian till, bedrock,
eolian sand, fluvial sand, and oxbows. Five of eight soil chemistry variables differed
significantly across these geologic classes, including organic matter, pH, available
phosphorus, calcium, and magnesium. Soil pH, organic matter, and available
magnesium were found to have fairly well-defined spatial gradients across north-
eastern Iowa fen sites. Four principal axes of climatic variation were also described
from northeastern Iowa, and include temperature, precipitation, growing season,
and summer precipitation. A total of 320 vascular plant taxa were located or have
been reported from this community, with seven being only known from historical
records. Fully 44% of the flora are considered rare in the region or state. Of these,
30% are rare in northeastern Iowa, 20% are rare in the state, 12% are listed for
state legal protection, 11% are newly known from the northeastern Iowa flora,
and 4% are newly known from the Iowa flora. Even though northeastern Iowa
fens currently cover only .01% of the northeastern Iowa land surface, they con-
tribute significantly to state and regional biodiversity by harboring approximately
28% of the regional and 18% of the total state flora.

Key Words: fens, Iowa flora, rare species, biodiversity

INTRODUCTION

Fens are peatland areas whose source water has been enriched
in nutrients by passage through the ground (Sjérs, 1952; Moore
and Bellamy, 1974). While covering a large extent of the boreal
North American landscape (Horton et al., 1979; Glaser, 1987),
along the glacial margin in North America fens are generally

LZ]

122

Rhodora

Table 1. States in the eastern U.S. where fens have been reported.

State Selected References
Connecticut LeFor, 1986
Illinois Sheviak, 1974
Swink Wilhelm, 1979
Moran 8
Indiana Bresner ae Potzger, 1946
Shuey, 1985
Wilcox et al., 1986
lowa Anderson, 1943
Nekola and Lammers, 1989
Kansas Horr and McGregor, 1948
Griger, 1973
ss egal Weatherby and Crow, 1992
Michig Cain and Slater, 1948
Schwintzer, 1978
Kron, 1989
Minnesota Glaser, 1987
Coffin and Pfannmuller, 1988
Missouri Steyermark, 193
Orzell, 1983
Nelson, 1985
Nebraska Kaul et al., 198
New Jersey Breden and Smith, 1988
New York Stewart and Merrell, 1937

North Carolina
North Dakota

McVaugh, 1958

Schafale and Weakley, 1990
Seiler and Barker, 1985
Malterer and Bluemle, 1988

Ohio Gordon, 1933
Andreas, 1985
Stewart, 1987
Pennsylvania Conard, 1952

South Dakota

Virginia
West Virginia

Wisconsin

Over,

Bartgis and Lang, 1984
Curtis, 1959
Carpenter, 1990

eed, 1985

[Vol. 96

considered one of the rarest habitats in the landscape (Reed, 1985;
Eggers and Reed, 1990). Fens or fen-like areas have been reported
from a number of central and eastern states (Table 1), where they
harbor a distinct flora including many rare species.

1994] Nekola— Northeastern Iowa Fens Oe

RET TE
IS Werth Mitchel | Howard ls
—

\
\
= Floyd
~

|

Black Hawk | Buchanan

Figure 1. Landforms and counties included in the study region. Landform
boundaries are based on Prior (1991).

Fens have long been the subject of research along the western
side of the Wisconsinan terminal moraine in northwestern Iowa
(Cratty, 1903; Wolden, 1926; Anderson, 1943; Hayden, 1943:
Conard, 1952; Thorne, 1952; Holte and Thorne, 1962; Holte,
1966; van der Valk, 1975; Lammers and van der Valk, 1979:
Moats, 1981). Until the mid-1980’s biologists and ecologists
thought that fens were restricted in Iowa to this region (e.g., Roosa
and Eilers, 1978).

However, spring-fed, hillside peat deposits (termed ‘mound
springs’) were also reported from northeastern Iowa in early geo-
logic surveys (White, 1870; Calvin, 1897, 1902). These habitats
remained biologically undescribed until 1984, when the flora of
a northeastern Iowa site was studied (Nekola and Lammers, 1989).
Like fens elsewhere in central and eastern North America, this
site was found to harbor a large number of regionally rare species.

To better assess the regional fen environment across north-
eastern Iowa, and its contribution to local biodiversity, recon-
naissance for all extant fen sites within the Iowan Erosional Sur-
face, Paleozoic Plateau, and portions of the Southern Iowa Drift
Plain and Mississippi River Alluvial Plain of Prior, 1991 (Figure
1) was conducted. Results from this survey will be used to address
three main topics: (1) What was the original extent of fens in
eastern Iowa, and how many remain extant? (2) What is the

124 Rhodora [Vol. 96

environmental variation of the extant sites in terms of their un-
derlying geology, soil chemistry, and regional climate? (3) What
is the species diversity of these sites, and how do they contribute
to state and regional biodiversity?

METHODS

A. Site Location

For purposes of this study, fens were defined as wetland areas
with saturated but not inundated soils which are fed by permanent
groundwater seepage. This definition differs slightly from those
typically used by European researchers (e.g., Sj6rs, 1952; Moore
and Bellamy, 1974) in that sites which have saturated soils, but
little or no peat accumulation are considered fens. The broader
definition of ‘fen’ used in this study follows the precedent of many
other U.S. fen researchers (e.g., Holte, 1966; Andreas, 1985;
Weatherbee and Crow, 1992).

Potential fen sites in northeastern Iowa were located through
use of county soil survey maps prepared by the USDA Soil Con-
servation Service (SCS). These modern soil survey maps have
been completed for all counties within the study region. The
location of potential sites was made possible as most occurrences
of muck or peat soils (variously named ‘Peat’, ‘Muck’, ‘Palms
Muck’, or ‘Houghton Muck’) are restricted to fen sites.

From 1984 to 1991 all occurrences of peat soil pedons occurring
on SCS county soil maps in the 29-county study region were
visited to determine the location of extant sites. Sites were con-
sidered extant if they possessed essentially undisturbed vegetation
and/or rare species. A number of undrained sites that had been
completely invaded by weedy plants such as Phalaris arundina-
cea, or had been grazed beyond recognition, were excluded from
this list. The boundaries of all pedons of fen soil mapped by the
Soil Conservation Service were digitized to provide a regional
perspective on the distribution of presettlement and extant sites.

B. Soil Analysis

Four soil samples were collected from each extant fen in an
attempt to capture the range of soil variation present. These sam-
ples were oven dried, and 50 g taken from each to make a single

1994] Nekola— Northeastern Iowa Fens 175

composite sample for each site. Determination of percent organic
matter, pH, NO,-N, extractable P, exchangeable K, extractable
SO,-S, exchangeable Ca, and exchangeable Mg in these samples
were conducted by Minnesota Valley Testing Laboratory of Ne-
vada, Iowa, and followed the methodology of Dahnke (1988).
These variables are referred to as percent organic matter, pH, P,
K, S, Ca, and Mg, respectively, in the remainder of the paper.
Water samples were not taken as standing water was not present
at all sites, and as water samples could not be stored for later
analysis. However, soil and water chemistry should covary as any
dissolved or suspended particulates will remain in the soil after
drying.

One-way ANOVA’s were used to test the significance of dif-
ferences for these soil variables across five classes of underlying
surficial geology. An approximation of the spatial distribution of
each variable across northeastern Iowa fen sites was estimated
through the interpolation technique of block kriging (Burgess and
Webster, 1980).

C. Climatic Analysis

Twenty-two climatic variables were calculated as 30-year means
from 96 recording stations across Iowa. Principal components
analysis was conducted on all variables to identify the most im-
portant axes of variation. These axes were related to the original
variables through use of a vari-max rotation. To document the
spatial distribution of each of these four main axes within the
study region, the interpolation technique of block kriging was
used to estimate the variation of each across the study region.

D. Floristic Analysis

Occurrence of all vascular plant species were noted for all extant
northeastern Iowa fens. In addition, a survey was made of the
northeastern Iowa botanical literature (Fitzpatrick, 1899: Barnes
et al., 1900; Pammel, 1908; Guldner, 1960; Cooperrider, 1962:
Hartley, 1966; Eilers, 1971) to identify all species that were his-
torically known from these habitats but for which no extant pop-
ulations could be located.

Five types of rarity within the flora were assessed. First, rarity

126 Rhodora [Vol. 96

within northeastern Iowa is based on classifications provided from
the two regional floras (Cooperrider, 1962; Eilers, 1971) whose
boundaries fall entirely within the region of this current study.
For a taxon to be considered rare within the region, it had to be
considered either rare in both of these floras, or rare in one and
absent from the other. Second, state-level rarity was assessed
following Howe et al. (1984). Third, the listing of the taxa legally
designated as ‘Endangered’, ‘Threatened’, and ‘Special Concern’
in Iowa was based on Roosa et al. (1989). Fourth, taxa new to
the regional flora are based on Guldner (1960), Cooperrider (1962),
and Eilers (1971). Lastly, the designation of taxa new to the state
is based on those not found in published county and regional
floras.

Voucher collections were made for all rare taxa occurrences by
top-snatching relevant material so that populations would not be
affected. Specimens were deposited at the R. V. Drexler Herbar-
ium at Coe College, Cedar Rapids, Iowa. Where populations were
large enough, duplicate specimens were also made and placed in
the University of lowa Herbarium, Iowa City, and the University
of North Carolina Herbarium, Chapel Hill.

RESULTS AND DISCUSSION

A. Distribution of Presetthement and Extant Sites

In total, 2333 potential sites were identified from SCS soil maps
in the 29 eastern Iowa counties surveyed (Figure 2a). Of these,
only 160 remained extant by 1991 (Figure 2b). Site names, sizes,
geologic class, and locations for these sites are given in Appendix I.

In presettlement, clustering of fen sites was present at a number
of differing spatial scales. At a regional scale, sites clustered into
two general groups: one paralleling the eastern margin of the Des
Moines Lobe, and the other paralleling the eastern border of the
Iowan Erosional Surface.

At smaller scales (approximately the size of counties), roughly
parallel northwest-to-southeast trending bands of high site density

—

Figure 2. Distribution of fen habitats in peeve ie based upon dig-
itization of SCS county soil maps. a: Presettlement. b:

1994] Nekola— Northeastern Iowa Fens iy

Figure 2a: Presettlement Fens

Figure 2b: Extant Fens - 19914

128 Rhodora [Vol. 96

approximately 10-15 miles wide and 75 miles long occurred. In
the western cluster, these bands seem to be closely related to
outwash sands and gravels along the Winnebago and Shell Rock
Rivers. In the east, longer and more diffuse bands are present
which seem to cut across local drainage patterns and topography.
One of the most marked of these concentrations begins in central
Chickasaw County and terminates in northwestern Cedar County.
It is possible that these bands of high site density correlate with
locations of pre-Illinoian moraines where the concentration of
upland gravel deposits are greater.

At even smaller spatial scales (roughly the size of a township)
individual sites were often grouped into linear strings. The most
intriguing of these are a parallel series of five such strings which
extend over a 30 mile extent in Chickasaw, Howard, Fayette, and
Winneshiek Counties. The underlying factors responsible for such
pattern are not known.

While diverse spatial structure existed in presettlement site
distribution, the loss of 93% of sites in the last 50-100 years has
led to the destruction of almost all of this pattern. The distribution
of extant sites now appears almost random, with even the large
regional clusters being almost unidentifiable. This loss of spatial
structure will probably lead to a decrease in the rate of immigra-
tion between sites, and alter the direction of species migration
through the landscape.

SCS maps, which were used to generate these data, should not
be considered infallible sources of information. Two caveats re-
garding their use in documentation of presettlement fen sites must
be made. First, not all fen sites present in the presettlement land-
scape have been mapped. This form of error has led to the ex-
clusion of some historic and extant sites from the published soil
maps. For instance, a site near Tipton described by White (1870)
is not mapped in the Cedar County soil survey. Also, eleven
percent of the extant sites listed in Appendix I were not mapped
as possessing fen soil pedons, and were located by visual inspec-
tion during field reconnaissance or from local contacts.

The second form of error in soil survey mapping of potential
fen sites occurs when marsh habitats with peat accumulations are
mapped as possessing soils characteristic of fen sites. Although
such wetlands may have muck or peat soils, they do not provide
proper habitat for fen species, and do not represent potential fen
habitats. The mapping of all muck soil pedons in such cases will

1994] Nekola— Northeastern Iowa Fens 129

overestimate the number of presettlement fen sites. This error is
most prevalent in the western counties bordering the Des Mcines
Lobe, where numerous pothole marshes with peat accumulations
were present prior to draining. Some of the mapped areas of peat
or muck soil in Clinton County may also represent former marsh
habitats. This form of error was minimized by only digitizing
sloping, hillside peat pedons on the Des Moines Lobe, thereby
removing potential marsh habitats from analysis. As it was not
possible to make any a priori rules for exclusion of potential
Clinton County marsh sites, all peat pedons in that county were
mapped. However, field observations suggest that less than 10%
of potential Clinton County sites represent former marsh sites,
indicating a relatively low rate of this form of error.

Although county soil maps represent an imperfect data source,
they are the best and only estimate available of the presettlement
extent of fens in northeastern Iowa. As both of these error rates
are relatively low (approximately 10%), the analysis presented
above should thus reflect the trends in site distribution once pres-
ent in presettlement landscape.

B. Environmental Variation Across Northeastern
Iowa Fen Sites

Geologic Classification

The majority of the region included in this study was ungla-
ciated during the Wisconsinan and subjected to intense periglacial
erosion which removed existing tills and exposed water-bearing
till or bedrock sequences, created stepped landscape surfaces, and
prevented deep accumulations of loess (Hallberg et al., 1978:
Prior, 1991). The sediment liberated through periglacial erosion
was deposited as eolian and fluvial sands along major river cor-
ridors (Prior, 1991).

Depth of loess accumulation appears to negatively covary with
abundance of fen sites. On the southern margin of the study region
in Tama, Benton, Cedar, and southern Linn Counties where peri-
glacial erosion was less intense and loess deposits accumulated
(Anderson, 1983), fens are much less frequent than in the till-
mantled or bedrock-controlled landscapes to the north which ex-
perienced higher erosion rates.

The 160 extant fens in northeastern Iowa were classified into

130

Rhodora [Vol. 96

A

C : ——_

Ce

LEGEND
ere SAND Peat ALLUVIUM
Loess. [&

©3] sano [ff water = [S4 timestone
GRAVEL TABLE (WITH KARST)

1994] Nekola— Northeastern Iowa Fens 131

five broad groups based upon their underlying geology: pre-Illi-
noian till, bedrock, eolian sands, fluvial sands, and oxbows. These
groups basically follow those of Thompson (1992) and Thompson
et al. (1992), with the exception that (1) ‘gravel ridge’ sites have
not been separated from other pre-Ilinoian till sites as they simply
represent pre-Illinoian till sites occurring in a more eroded land-
scape, and (2) sites developed in upland, eolian sands have been
differentiated from those found with fluvially-deposited sands and
gravels.

The most frequent geologic class of northeastern Iowa fens is
associated with deposits of pre-Illinoian sand and gravel (Figure
3A). These 119 sites are found throughout the region (Figure 4)
in association with water-bearing gravels deposited within and
between tills (Thompson, 1992). The majority of pre-Illinoian till
fens are found on footslopes along pediment margins (Hallberg
et al., 1978), although they may occur at any position in the
landscape where water-bearing gravels have been exposed.

Fens associated with bedrock aquifers (Figure 3B) are much
less frequent (13 sites), and are restricted to the northeastern and
northwestern parts of the study region (Figure 4) where most or
all of the till was removed through periglacial processes. Like pre-
Illinoian sites, bedrock fens are usually found on footslopes along
pediment edges. While peat usually has formed in these fens, some
sites have only minor peat accumulations, and have a soil surface
consisting of cold, wet, limestone rubble. Such sites contain some
of the most calcareous fen habitats found in the region.

Fens associated with eolian sands (14 sites) are restricted to the
southern third of the region, with the majority being found along
the Wapsipinicon and Cedar Rivers where extensive dune fields
have formed in the surrounding uplands (Figure 4). Some eolian
sand sites are also found along sand-mantled paha ridges distant
from rivers (for an explanation of pahas see Prior, 1991). Eolian
sand fens are often found along the base of dunes where the
downward flow of water is stopped by more impervious tills or
paleosols (Figure 3C). Many of these sites contain relatively low

—

Figure 3. Idealized cross-section diagrams of the five classes of northeastern
Iowa fens based on surficial geology. a: Pre-Illinoian till; b: Bedrock; c: Eolian
Sand: d: Fluvial Sand; e: Oxbow.

132 Rhodora [Vol. 96

‘ j
Legend: oO as
+ = Pre-Illinoian Till (119 sites) tt ;
* = Bedrock (13 sites)
@ = Eolian Sand (14 sites)

vy = Fluvial Sand (8 sites)

ce
VY

) = Oxbow (6 sites)

Figure 4. Distribution of the five geologic classes of northeastern Iowa fens.

amounts of organic matter and have margins consisting of wet,
acidic sands. The lowest percent organic matter, pH, Ca, and Mg
readings from eastern Iowa fens occur in these sites (see below).

Fens associated with fluvial sand and gravel terraces are rarer
yet (8 sites). Fluvial sand fens occur when water moving through
sands and gravels emerges to the surface along terrace margins
(Figure 3D). While most of these sites are associated with outwash
sands and gravels along the western section of the study region,
three sites were found along the Upper Iowa and Cedar Rivers
in Allamakee and Muscatine Counties (Figure 4). The largest
single presettlement sites identified in the region by SCS maps
occurred with fluvial sand and gravels along the Shell Rock and

1994] Nekola— Northeastern Iowa Fens 133

Winnebago Rivers. Only a small part of a single one of these sites
remained extant by 1991. Floristic lists from these sites prior to
draining (Pammel, 1908, 1917) suggest that they once contained
one of the most diverse fen floras in the state.

The most uncommon type of fen in northeastern Iowa are those
formed from hydric succession of oxbow channels along aban-
doned stream meanders (6 sites). While oxbow channels typically
have open water and would not support fen species, organic matter
accumulation can continue until the soil surface becomes equiv-
alent to the water table, creating the saturated but not inundated
conditions necessary for colonization of fen species (Figure 3E).
Fens occupying former oxbow channels are scattered throughout
northeastern Iowa along the Winnebago, Cedar, Wapsipinicon,
and Little Wapsipinicon Rivers (Figure 4). The greatest concen-
tration of this fen type once occurred along the Cedar River south
of Iowa City in the ‘Lake Calvin Basin’ where a number of suc-
cessive oxbow lakes have formed fen habitats. Baker et al. (1987)
have characterized the Holocene history of one of these sites
through macrofossil and pollen analysis.

Geologic History

The landscapes upon which northeastern Iowa fens are found
have formed through erosional processes within the last 20,000
years. The age of fen sites in this landscape was investigated by
Thompson (1992), who determined the dates of the basal peat
layer, using C,, isotope analysis, for ten sites. The age of basal
peats in each of these sites were found to differ greatly, ranging
between 10,900 and 2280 years B.P.

The interpretation of these dates is made more difficult by
interruptions in peat deposition (sometimes accompanied by ox-
idized peat) within sites having the oldest basal dates (Hall, 1971;
Van Zant and Hallberg, 1976). Fens without old (8000 year) basal
peat dates are underlain by oxidized clay, till, or occasionally
prairie soil (J. C. Nekola and R. G. Baker, unpubl. data). The
existence of oxidized layers of peat or mineral soil underlying fen
sites suggests that prior to modern peat development the site dried
out, exposing the peats and underlying substrates to oxidation
and erosion.

Fens in northeastern Iowa thus appear to be a Holocene phe-

134

Rhodora

[Vol. 96

Table 2. Comparison of means and standard deviations from soil chemistry
variables between all surficial geology groups, and over all sites. Standard devi-
ations are presented within parentheses under the mean for each group.

Geologic Group

Pre
Illinoia Eolian Fluvial
Variable Til Bedrock Sand Sand Oxbows All
% Organic 40.9 26.5 18.8 29.1 35.5 37.0
(12.3) (13.2) (16.2) (7.8) (15.4) (14.5)
pH 6.2 6.6 5.4 6.8 5.8 6.2
(.5) (.6) (.7) (3) (.7) (.6)
N 1.1 0.7 0.8 0.6 0.6 1.0
(1.3) (.4) (.6) (.4) (.3) (1.7)
P 26.6 20.9 37.6 10.6 29.0 26.5
(14.6) (15.8) (13.1) (9.2) (13.0) (15.0)
K 42.2 47.6 44.4 42.7 42.2 42.9
(17.6) (23.6) (22.6) (7.4) (18.9) (18.2)
S 186.4 208.1 100.6 126.0 195.7 178.4
(140.1) (159.7) (69.7) (78.9) (98.7) (135.2
Ca 4494.4 4364.8 2101.1 4507.1 3331.4 4224.1
(923.8) (996.1) (1038.1) (827.9) (1176.2) (1168.9)
Meg 411.2 560.5 220.7 480.1 339.9 406.6
(130.6) (146.9) (92.8) (119.7) (129.2) (146.9)

nomenon, with the onset of development generally coinciding
with the end of the Hypsithermal (Thompson, 1992). The creation
of new sites has apparently continued until present, with some
bedrock and eolian sand sites having less than eight cm of peat
accumulation. A similar pattern has been found in the boreal fens

Table 3. Summary statistics from a 1-way ANOVA comparing the differences
in soil chemistry variables across all geologic classes.

Soil Variable P r
% Organic O00 261
pH 000 210
N 433 024
P 001 110
K 813 .010
S 133 044
Ca 000 345
Mg .000 238

1994] Nekola— Northeastern Iowa Fens 135

of central Canada (Zoltai and Vitt, 1990) and pocosins of the
southeastern Coastal plain (Daniel, 1981).

The implication for this post-Hypsithermal development of fen
sites is that rare species which are found on them cannot be
postulated to represent Pleistocene relicts. Rather, all of the rare
taxa found in this community have had to disperse into this
landscape within the last 6000 years.

Soil Chemistry

Variation of soil chemistry as a
function of geology

The mean and standard deviation for eight soil chemistry vari-
ables within each of the five geologic classes of fens are presented
in Table 2. The results of one-way ANOVA’s on each of these
variables (Table 3) demonstrate that three variables (N, K, and
S) show no significant differences between the five geologic groups.
The remaining five variables (percent organic matter, pH, P, Ca,
and Mg) were found to possess significant differences across the
geologic groups. Organic matter was highest in pre-IIlinoian and
oxbow fens, and was lowest in sites developed in eolian sands.
Soil pH was highest on sites developed on bedrock and fluvial
terrace sands, and was lowest in eolian sand and oxbow sites.
Phosphorus content was highest in eolian sand and oxbow sites,
and lowest in fluvial sand and bedrock sites. Calcium concentra-
tion was highest in pre-IIlinoian till and bedrock sites, and was
lowest in eolian sand sites. Lastly, soil magnesium was highest in
bedrock sites, and was lowest in eolian sand sites.

Although significant differences were observed in the central
tendency of values for these soil chemistry variables between
geologic groups, this should not imply that there is little overlap
between the soil chemistry variables for sites in differing geologic
groups. For instance, while the mean percent organic matter of
pre-Illinoian till sites was twice that of eolian sand fens (Table
3), 75% of sites in these two groups have soil Ca found in the
range between the lowest recorded pre-IIlinoian till (1900 ppm)
and highest recorded eolian sand (4000 ppm) values. This fact is
also demonstrated in the r* values for the significant ANOVA’s
which ranged between 11% for P to 35% for Ca, implying that

136 Rhodora [Vol. 96

A:
% Organic
B:
H _
p ppm SO,4
C:
ppm NO.,-N
D: H:
ppm P Mg (2 fee.
soa F
—
on

Figure Spatial distribution of fen soil chemistry variables across north-
eastern fave a: Percent organic matter; b: pH: c: N; d: P; e: K; f: S; g: Ca; h: Mg.

no more than '4 of the observed variance in these variables can
be accounted for by geologic grouping. Thus, although the average
for percent organic matter, pH, P, Ca, and Mg differ significantly
between the geologic groups, the values overlap enough to make

1994] Nekola— Northeastern Iowa Fens es

it likely that any two sites will possess similar soil chemistry
variables, independent of their surficial geology.

Variation in soil chemistry as a function of
spatial location

How does soil chemistry vary in fen sites across the region? To
address this question, kriged response surfaces for each of the
eight soil variables were calculated for the study region (Figures
5A-5H). Relatively simple, approximately monotonic response
surfaces were observed only in pH, organic matter, and Mg. The
remaining soil chemistry variables have less clear spatial gradients
as they possess multiple maxima and minima over the study
region. The causes of all these gradients are undoubtedly complex,
and related to climatic patterns (see below), and the non-random
distribution of fens of differing geologic origin.

Climate

A final major class of environmental variation which defines
the northeastern Iowa fen environment is climate, which will not
only affect the distribution of species, but may also influence soil
chemistry and the size and nature of groundwater aquifers.

Four major trends in climatic variation were identified over
the study region through principal components analysis, account-
ing for 96% of the total variation. The first axis is related to
temperature, and is most closely correlated with Average Yearly,
Summer, Fall, and Spring Temperatures, Heat Stress Days/Year,
Heat Stress Degrees/Year, Heating Degree Days, Cooling Degree
Days, and Growing Degree Days. The variation of precipitation,
exclusive of the summer months, represents a second major axis
of variation. This axis is most closely correlated with Average
Yearly, Winter, Spring and Fall Precipitation, Average Winter
Temperature, and Maximum 24-hour Precipitation. The third
major axis of climatic variation is represented by growing season,
which is most closely correlated to 32°, 30°, 28°, 26°, and 24°
Growing Seasons (in degrees Fahrenheit), and number of Freezing
Days/Year. The last major climatic gradient is represented by
Average Summer Precipitation. The spatial variation of each of
these axes across the region was estimated by mapping the vari-
able most highly correlated to each axis through block kriging
(Figure 6A-6D).

138 Rhodora [Vol. 96

A: Heat Stress Days / Year C: 32° Growing Season

2

September-November D: June-August

Precipitation Precipitation

ae

||| i

igure 6. Spatial distribution of the main axes of climatic variation across

cipitation (as shown by average fall (September-November) precipitation/year).
c: Growing season (as shown by 32°F growing season). d: Summer (June-August)

PICK Pilalivil Bi

C. Biological Diversity of Northeastern
Iowa Fen Habitats

Enumeration of the Vascular Flora

A total of 320 vascular plant taxa were recorded from north-
eastern Iowa fen communities (Appendix II). The flora includes
approximately 23% which are incidental: their presence on fens

1994] Nekola— Northeastern Iowa Fens 139

Table 4. Summary of the northeastern Iowa fen vascular plant flora in terms
of major phylogenetic groups.

Native Taxa Exotic Taxa

Division Fen Incid. Tot. Fen Incid. Tot.
Pteridophytes 10 5 15 0 0 0
Gymnosperms 0 l 0 0 @)
Dicotyledons 145 49 194 4 4 8
Monocotyledons 84 13 97 3 2 5
Totals 239 68 307 7 6 3
Grand total 320

may be due to the mass effect (Shmida and Wilson, 1985) of
surrounding communities. Only 13 taxa are exotic to northeastern
Iowa, and almost half of these are incidental (Table 4). None of
the exotic taxa at present is a threat to overall biodiversity on
undisturbed sites, although some do invade disturbed areas.

The majority of the fen flora are dicotyledons, followed by
monocotyledons, pteridophytes, and gymnosperms (Table 4). Of
the families present in the fen flora, the two largest (in terms of
taxa represented) are the Compositae and Cyperaceae, followed
by the Gramineae, Salicaceae, Rosaceae, Labiatae, and Orchi-
daceae (Table 6). The best represented genera are Carex, Salix,
Aster, Cypripedium, and Solidago (Table 5).

Table 5. Enumeration of the largest families and genera present in the north-
eastern Iowa fen vascular plant flora.

Families Genera
(7 or more taxa) (5 or more taxa)
Number of Number of
Name Taxa Name Taxa
Compositae 41 Carex 27
Cyperaceae 40 Salix 14
Gramineae 27 Aster 6
Salicaceae 15 Cypripedium 6
Rosaceae 14 Solidago 6
Labiatae 13 Galium 5
Orchidaceae 13 Gentiana 5
Scrophulariaceae 9 Juncus 5
Leguminosae 8 Viola 5

Onagraceae 7

140 Rhodora [Vol. 96

Table 6. Summary of the number of rare vascular plants in the northeastern
Iowa fen flora for different rarity classifications.

Percent of Percent of
Source Number Total Flora Native Flora

Native species rare on Iowan sur-
face: based on Eilers (1971) and

Cooperrider (1962) 97 30% 32%
Rare in Iowa: Howe et al. (1984) 63 20% 21%
Endangered in Iowa: Roosa et al.

(1989) 21 7% 7%
Threatened in Iowa: Roosa et al.

(1989) 11 3% 4%
Special concern in Iowa: Roosa et

al. (1989) 6 2% 2%
New to Iowan surface: based on Ei-

lers (1971) and Cooperrider (1962) 34 11% 11%
New to state 13 4% 4%
Total native rare taxa 134 44% 42%
Total exotic rare taxa 4 1%

Stability and Distribution of the
Vascular Plant Flora

Only seven taxa (Carex diandra, Cassia marilandica, Eleo-
charis acicularis, Eriophorum gracile, Filipendula rubra, Salix
lucida, and Selaginella eclipes) which had been previously re-
ported from northeastern Iowa fens were not located during the
course of this study. Of these taxa, only Carex diandra is currently
considered extirpated from the state (Howe et al., 1984). Cassia
marilandica, Filipendula rubra, and Selaginella eclipes became
extirpated from the northeastern Iowa fen flora through site de-
struction, while the remaining four have become extirpated through
loss of populations on extant sites.

The distributions of most species seem relatively unaffected by
the environmental variation described above. A few taxa should
be pointed out, however, which do appear to be limited in their
distribution by environmental factors. For example, one-half of
the stations for Menyanthes trifoliata occur in oxbow fen sites,
even though such sites represent only 4% of extant fens in the
region. Some taxa, including Angelica atropurpurea, Artemisia
serrata, Lobelia kalmii, and Triglochin maritimum are largely
restricted to the northwestern portion of the study area where

1994] Nekola— Northeastern Iowa Fens 141

precipitation tends to be lowest and soil pH, Ca, and Mg tend to
be highest. Additional taxa, including Botrychium multifidum,
Botrychium simplex, Carex leptalea, Dryopteris x uliginosa, Eq-
uisetum x littorale, Eriophorum virginicum, Habenaria flava var.
herbiola, Ludwigia palustris, Osmunda regalis, Panicum boreale,
Rhexia virginica, Solidago patula, Viola lanceolata, and Viola
primulifolia are restricted to the south and southeast where they
are usually found in low-pH, low-organic matter, low-Mg, and
low-Ca eolian sand fens. This part of the study area also has the
highest temperature and precipitation.

Rare Taxa in the Northeastern Iowa Fen Flora

A remarkable number of the taxa included in the flora of north-
eastern Iowa fens are rare in the region and state. Approximately
44% of the flora (138 taxa) fall into one of the five categories of
rarity (Tables 6 and 7). Four of these are rare exotic weeds. Of
the native taxa, 97 (30% of the flora) are rare in northeastern
Iowa, 63 (20%) are rare in the state, 38 (12%) are listed by the
Iowa DNR for Iowa Endangered Species Act protection, 34 (11%)
are new to northeastern Iowa, and 13 (4%) are new to the state.
Approximately 12% of all state rare taxa (Howe et al., 1984), and
17% of all Iowa ‘endangered’, ‘threatened’, and ‘special concern’
species (Roosa et al., 1989) occur in northeastern Iowa fen sites.

While certainly important for rare species, fens also contribute
greatly to the total vascular plant biodiversity in the region and
state. Even though fen habitats originally occupied approximately
only .1% of the northeastern Iowa landscape (and currently cover
less than .01% of the land surface), these sites harbor approxi-
mately 28% of the total flora of the Iowan Erosional Surface (based
on estimates in Eilers, 1971) and 18% of the state’s total flora
(based on estimates from John Kartesz, pers. comm.).

CONCLUSIONS

Fens in northeastern Iowa represent a diverse collection of
habitats which vary in their surficial geology, soils, and climate.
These fen sites harbor a substantial proportion of both the total
diversity of the Iowa flora and the rare components of that flora.

Of all the results presented above, perhaps none is more 1m-

142 Rhodora

[Vol. 96

Table 7. List of rare Iowa plants found in northeastern Iowa fens.

Taxon

Iowa
Rare

New to
NE Iowa

New to
lowa

DNR
Rank

Alnus rugosa
Angelica atropurpurea
Artemesia serrata
Aster Junciformis
Aster punice
Betula pumila var. glandulifera
Betula x sanbergti

=
io)

Botrychium multifidum
Botrychium simplex

Cacalia suaveolens
Calamagrostis inexpansa

C allitriche fe
Carex

Ci ee!

Carex granularis

Carex lasiocarpa var. ammericana
Carex leptalea

Carex prairea

Carex rostrata var. utriculata
Carex sartwellit

Carex sterilis

Carex tetanica

Cirsium muticum

Equisetum pei

Equisetum x litt

Eriophorum engin
aie gracile

Eriophorum ae

Piped ubra

Galium labradoricum

pe trifidum

Gentiana crinita

~~ Ke Ke KK

i ~ Me Pe OM

Pal

em Pm OM

~~ eK KM KM KM

SC

™ ™

1994] Nekola— Northeastern Iowa Fens

Table 7. Continued.

Iowa New to New to DNR

Taxon Rare NE lowa Iowa Rank

Gentiana crinita procera Xx x

Gentiana proc

Habenaria pe var. herbiola
ea

>
AmHAmea

Liparis loeselii

mii

Melanthium virginicum
Menyanthes trifoliata

Mimulus glabratus var. fremontii
Muhlenbergia glomerata
Ophioglossum vulgatum
Osmunda regalis

Panicum boreale x
Parnassia glauca

Pilea fontana

Rhexia virgini

Rhynchospora capita

Rubus pubesce

Rumex as

Salix a

Salix luc

Salix pedicellaris var. Aypoglauca

a oo
N
MMA

2)
Ima

4 > tO OOO
~

Salix x subsericea

alix x cryptodonta
Salix eal x pedicellaris
Scleria verticillata <
Selaginella eclipes x
Solidago patula x
Solidago uliginosa X X
Spiranthes lucida X Xx E
Triadenum fraseri
Triglochin maritimum
Triglochin palustre
Valeriana edulis
Viola primulifolia

PE a a

™ mm ™

~ hm om
Pad
<j

144 Rhodora [Vol. 96

portant than the wholesale destruction of this habitat, which had
led to the loss of 93% of sites and of the complex spatial structure
once present from township to regional scales. Sites are now more
isolated than formerly, leading to an almost certain reduction in
the rate of immigration between sites and in the rate (and direction
of) migration of fen species across this landscape. It is essential
that site isolation not increase further if fen species are to be given
optimal chance of migration, as the past history of sites indicates
that their habitats may not survive global warming.

Only 14 sites out of the 160 extant (Brayton-Horsley, Cutshall
Access, Freeport, Elk Creek, Mark Sand Prairie, Nichols, North-
woods Park, Rowley, Rowley North, Slough Creek |, Split Rock
Park, Spring Creek 2, Wiese Slough, Worth Pond) are currently
owned and protected by private or public conservation organi-
zations. The remaining 146 sites are largely under private own-
ership. While most private landowners of fen sites have expressed
interest in protecting their sites, a few are endangering sites. Since
initiation of survey work, three sites (Grassley’s Folly, Mark Sand
South, and Temple Hill) have been badly damaged (and perhaps
destroyed) through draining or grazing.

To protect the remaining biodiversity of this habitat, and to
provide optimal migration ability for species in the event of cli-
mate change, all extant fens in northeastern Iowa must be afforded
some form of protection, either through private landowner action
or through the purchase of sites when they become endangered.
It is essential that conservation organizations in the state make
fen protection a priority to ensure that the physical and biological
diversity documented here will be protected into the 21st century.

ACKNOWLEDGEMENTS

A number of people have been instrumental in this research
and must be thanked. John and Gretchen Brayton suggested to
me in 1984 the question of the diversity and extent of eastern
Iowa peat ‘bogs’, and are in large part responsible for my interest
in these sites. Bob Moats also played a critical role by showing
me northwestern Iowa fens in the summer of 1985. Dick Baker,
Willard Hawker, Louise McEarchen, John Nehnevaj, Dean Roosa,
Arlan Kirchman, Dennis Schlicht, Bill Thomas, Robert Thom-
son, Dave Wendling, and Herb Wilson all provided me with
locality information for extant fen sites, as well as able field as-

1994] Nekola— Northeastern Iowa Fens 145

sistance. Fred Nekola provided invaluable support and field as-
sistance throughout the project. Herb Wilson of the Soil Conser-
vation Service furnished soil survey maps for all counties within
the study region. Richard Carlson of the Department of Agron-
omy at Iowa State University kindly provided climatic data for
Iowa. Dean Roosa and Terrence Frest helped finance field work
during the 1988 season. The substantial financial support needed
for completion of soil analyses was provided by the McElroy
Foundation of Waterloo, Iowa. Field work was also funded in
part through a National Science Foundation Graduate Fellow-
ship. Peter White, Robert Peet, A. Linn Bogle, and Gordon DeWolf
read and made useful comments regarding earlier drafts.

LITERATURE CITED
ANDERSON, W. A. 1943. A fen in northwestern Iowa. Amer. Midl. Nat. 29: 787-
791

ANDERSON, W. I. 1983. Geology of Iowa. Iowa State University Press, Ames.
Anpreas, B. K. 1985. The relationship between Ohio peatland distribution and
uried river valleys. Ohio J. Sci. 85(3): 116-125.

Baker, R. G., D. G. Horton, H. K. Kim, A. E. SULLIVAN, D. M. Roosa, P. M.
WITINOK AND W. P. PUSATERI. 1987. Late Holocene paleoecology of south-
eastern Iowa: development of riparian vegetation at Nichols Marsh. Proc.
Iowa Acad. Sci. 94(2): 51-70

BARNES, W. D., F. REPPERT AND A. A. MiLter. 1900. The flora of Scott and
Muscatine counties. Proc. Davenport Acad. Natural Sciences 8: 199-287.

Bartais, R. L. AND G. E. LANG. 1984. Marl wetlands in eastern West Virginia:
distribution, rare plant species, and recent history. Castanea 49: 17-25.

Burcess, T. M. AND R. WepsTER. 1980. Optimal interpolation and isarithmic
mapping of soil properties. II. Block kriging. J. Soil Sci. 31: 333-341.

IN, S. A. AND J. V. SLATER. 1948. The vegetation of Sodon Lake. Amer. Midl.
Nat. 40(3): 741-762.

CaALvin, 8. 1897. Geology of Buchanan County. Annual Rep. Iowa Geol. Surv.
8: 291-339.

. 1902. Geology of Chickasaw County. Annual Rep. lowa Geol. Surv. 13:
257-292.

CARPENTER, O. J. 1990. Hydrology and vegetation on a calcareous peat mound
fen in southeastern Wisconsin. Masters thesis, University of Wisconsin, Mad-

ison.

CorFFIN, B. AND L. PFANNMULLER. 1988. ef ace Endangered Flora and
Fauna. University of Minnesota Press, Minneapo

ConaARD, H.S. 1952. The vegetation of Iowa. Stud. ao Hise Iowa Univ. 19(4):
1-166

COOPERRIDER, T. S. 1962. The vascular plants of Clinton, Jackson, and Jones
Counties, Iowa. Stud. Nat. Hist. lowa Univ. 20(5): 1-76.

146 Rhodora [Vol. 96

CratTty, R. I. 1903. The flora of Emmet County, Iowa. Proc. lowa Acad. Sci.
11: 201-251.
Curtis, J.T. 1959. The Vegetation of Wisconsin. University of Wisconsin Press,

DaAHNKE, W.C. 1988. Recommended chemical soil test procedures for the north-
central region. North Dakota Agricultural Experiment Station Bulletin No.
499. North Dakota State University, Fargo.

DanieEL, C. C. II. 1981. Hydrology, geology, and soils of pocosins: a comparison
of natural and altered systems, pp. 69-108. /n: C. J. Richardson, Ed., Pocosin
Wetlands. eee Ross Publishers, Stroudsburg, Pennsylvania

Eccers, S. D. AND D. M. REED. 1990. Wetland plants and plant communities
of Minnsst and Wisconsin. U.S. Army Corps of Engineers, St. Paul, Min-

Eners, “ 1. 1971. og vascular flora of the Iowan area. Stud. Nat. Hist. lowa
Univ. 21(5): 1

moe T. J. i Manual of the flowering plants of Iowa. Parts I: Polypet-

and II: ate Sani Author, Lamoni, Iowa.

oa R.C. AND J. E. PotzGer. 1946. The Savi Creek raised bog, Randolph
County, fndiana Butler Univ. Bot. Stud. 8: 24-43,

GLASER, P. H. 1987. The ecology of boreal ewe of northern Minnesota: a
eerie profile. U.S.D.I., Fish and Wildlife Service Biological Report
85(7.

GorDOoN, a B. 1933. An unique raised bog at Urbana, Ohio. Ohio J. Sci. 33:
453-459

GruGER, J. 1973. Studies of the late-Quaternary end history of north-
eastern Kansas. Geol. Soc. Am. Bull. 84: 239-2

GuLpner, L. F. 1960. Vascular plants of Scott and ashe Counties. Dav-
enport Public Museum Botanical Publication #1.

Hatt, S. A. 1971. Paleoecological interpretation of bison, mollusks, and pollen
from the Hughes peat bed, Linn County, Iowa. Masters thesis, University
of Iowa, Iowa City.

HALLBERG, G. R., T. E. Fenton, G. A. MILLER AND A. J. LUTENEGGER. i
The Iowan Erosion Surface: an old story, an peasy lesson, and some ne
wrinkles, pp. 2-1 to 2-94. Im: R. Anders Ed., 42nd Annual Tri- ene
Geological Field CouRrenke Geology of East. Senta Iowa. Iowa Geological

urvey, Iowa Cit

HARTLEY, T. G. 1966. The flora of the “‘Driftless Area.’’ Univ. Iowa Stud. Nat.
Hist. 21(1): 1-174

HAYDEN, A. 1943. A hotaneat survey in the Iowa Lakes region of Clay and Palo
Alto counties. Iowa State Coll. J. Sci. 17: 277-416.

Ho te, K.E. 1966. A floristic and ecological analysis of the ee
in pea Iowa. Ph.D. dissertation, University of Iowa, low

AND R. F. THORNE. 1962. Discovery of a calcareous fen oe in
sonwestrn Iowa. Proc. Iowa Acad. Sci. 69: 54-60.

Horr, W. H. AnD R. C. McGrecor. 1948. A raised marsh near Muscotah,
Kansas. a Kansas Acad. Sci. 51(2): 197-200.

Horton, D. G., D. H. Vitt AND N. G. Stack. 1979. Habitats of circumboreal-
subartic sphagna: I. A quantitative analysis and review of species in the
Caribou Mountains, northern Alberta. Can. J. Bot. 57: 2283-2317.

Houau, M. Y. 1983. New Jersey Wild Plants. Harmony Press, Harmony, NJ.

1994] Nekola— Northeastern Iowa Fens 147

Howe, R. W., M. J. Huston, W. P. PusaTeri, R. H. LAUSHMAN AND W. E
eee 1984. An inventory of significant natural areas in Iowa. lowa
Conservation Commission, Des Moines.

Hutton, E. E., C. MILLER AND C. Conrap. 1968. A marl marsh natural area
in eastern West Virginia. Castanea 33: 241-246.

Kartesz, J. T. 1994. A synonymized checklist of vascular flora of the United
States, Canada, and Greenland. Timber Press, Portland, Ore.

Kaul, R. B., G. E. KANATAK AND S. P. CHURCHILL. 1988. The Niobrara River
valley, a postglacial migration corridor and refugium of forest plants and
animals in the grasslands of central aes America. Bot. Rev. 54(1): 44-81.

LAmMmers, T. G. AND A. G. VAN DER VALK. 1979. A checklist of the aquatic and
wetland vascular plants of Iowa: ferns, rt allies, and monocotyledons. Proc.
Iowa Acad. Sci. 85: 121-163.

Macterer, T. J. AND J. P. BLuUEMLE. 1988. Peat initiation and accumulation
during the Holocene, Denbigh Fen, McHenry County, North Dakota. Proc.
North Dakota Acad. Sci. 42: 30.

McVauau, R. 1958. Flora of the Columbia County area, New York. New York
State Museum and Scientific Service Bulletin #360, Albany.

Moats, R. 1981. Fens. lowa Conservationist 40(3): 14

Moore, P. D. AND D. J. BELLAMY. 1974. Peatlands. Springer-Verlag, New York.

Moran, R. C. 1981. Prairie fens in northeastern Illinois: floristic composition
and disturbance, pp. 164-168. Jn; R. Stuckey and K. J. Reese, Eds., P
ceedings of the Sixth North American Prairie Conference. Ohio State Uni-
versity, Columbus.

Nexo 1a, J. C. 1991. A western range extension for Selaginella eclipes Buck in
north-central Iowa. Amer. Fern J. 81: 67.

G. Lammers. 1989. Vascular flora of the Brayton-Horsley prairie:
a remnant prairie and spring fen complex in eastern Iowa. Castanea 54(4):
238-254.

Ope, D. J. 1985. Rediscovered plants in fen habitats of South Dakota. Proc.
Acad. Sci. 64: 217

Octe, D. W. 1989. Barnes Chapel Swamp: an unusual Arbor-vitae (Thuja oc-
cidentalis L.) site in Washington County, Virginia. Castanea 54(3): 200-203.

OrzeLL, S. L. 1983. Natural area inventory and floristic analysis of fens in
selected southeastern Missouri counties. Masters thesis, St. Louis University,

Over, W. H. 1932. Flora of South Dakota. University of South Dakota Press,
Vermillion.

PAMMEL, L. H. 1908. Flora of northern Iowa peat bogs. Annual Rep. Iowa Geol.
Surv. 19: 739-777.

1917. Flora of the bogs of Cerro Gordo and Worth counties, p. 171. Jn:
Iowa Parks, State Board of Conservation, Des Moines

Prior, J.C. 1991. Landforms of Iowa. University of Iowa Press, lowa City

REED, D. M. 1985. Composition and distribution of calcareous fens in etch
to environmental conditions i in southeastern Wisconsin. Masters thesis, Uni-

Roosa, D. M. AND L. J. Ercers. 1978. Endangered and threatened Iowa vascular
plants. Special report No. 5, lowa hi Preserves Advisory Board, lowa
Conservation Commission, Des Moin

LEOSCHKE AND L. J. EILERS. 1989. Distribution of Iowa’s endan-

148 Rhodora [Vol. 96

gered aan threatened vascular plants. lowa Department of Natural Resources,

ee M. P. AND A. S. WEAKLEY. 1990. Classification of the natural com-
munities of North Carolina. North Carolina Natural Heritage Program, De-
partment of Environment, Health, and Natural Resources, Raleigh.

SCHWINTZER, C. A. 1978. Vegetation and nutrient status of northern Michigan
fens. Can. J. Bot. 56: 3044-3051

SEILER, G. J. AND W. T. BARKER. 1985. Vascular flora of ayes, Richland,
and Sargent Counties, North Dakota. Prairie Nat. 17(4): 193-24

SHEVIAK, C. J. 1974. An introduction to the ecology of the Illinois Orchdacse
Scientific Papers 14. Illinois State Museum, Springfield.

SHMIDA, A. AND M. V. WILSON. pe Biological determinants of species di-
versity. J. Biogeography 12: 1-

SyOrs, H. 1952. On the relation oe vegetation and electrolytes in North
Swedish mire waters. Oikos 2: 241-258.

STEWART, P. A. AND W. D. MeRRELL. 1937. The Bergen Swamp: an ecological
study. Proc. Rochester Acad. Sci. 8: 209-262.

sAintdaaie J. A. 1938. Plants new to Missouri. Rhodora 40(475): 249-255.

Sw. DG. WILHELM. 1979. Plants of the Chicago region. Morton Ar-
Cae Lisle, Illinois.

THOMPSON, C. 1992. Origins and development of Iowa fens. Report to the
State Preserves Advisory Board, Iowa Department of Natural Resources,
aie Mol

E. A. Bettis III AND R. G. BAKER. 1992. Geology of Iowa fens. Jour.
iowa Acad. Sci. 99(2-3): 53-59.

THORNE, R. F. 1952. The 1952 foray of the Central States Section, Botanical
Society of America, at lowa Lakeside peas ory. III. Aquatic and bog veg-
etation of the Lake Okoboji area. Asa Gray Bull. new ser. 1: 307-308.

TOLSTEAD, W. L. 1938. A flora of Winneshiek and Allamakee counties and
Clayton County in the vicinity of McGregor. Masters thesis, Iowa State

VAN DER VALK, A. G. 1975. Floristic composition and structure of a fen in
northwest Iowa. Proc. Iowa Acad. Sci. 82: 113-118

VAN ZANT, K. L. AND G. R. HALLBERG. 1976. A late-glacial pollen sequence
from northeastern Iowa: Sumner Bog revisited. lowa Geol. Surv. Tech. Infor.
Bull. Series #3

WEATHERBEE, P. B. AND G. E. Crow. 1992. Natural plant communities of Berk-
shire County, Massachusetts. Rhodora 94: 171-209.

WHEELER, W. A. 1900. A contribution to the knowledge of the flora of south-
eastern Minnesota. Minn. Bot. Stud. 24: 353-416

Wuite, C. A. 1870. Report on the geological survey apie state of Iowa to the
Thirteenth General Assembly, Vol. 2. Mills, State Printer, Des Moines.

Witcox, D. A., R. J. SHEDLOCK AND W. H. HENpRICKSON. 1986. Hydrology,
water chemistry, and aaa relations in the raised mound of Cowles Bog.
J. Ecology 74: 103-11

Wo LpEN, B. O. 1926. A flora of northern Iowa. Amer. Bot. 32: 6-10.

ZOLTAI, S. C. AND D. H. Vitr. 1990. Holocene climate change and the distri-
bution of peatlands in western interior Canada. Quat. Res. 33: 231-240

1994] Nekola— Northeastern Iowa Fens 149
CURRICULUM IN ECOLOGY

229 WILSON HALL, CB #3275

UNIVERSITY OF NORTH CAROLINA

CHAPEL HILL, NC 27514

APPENDIX I. INVENTORY OF EXTANT NORTHEASTERN IOWA
FEN SITES.
Geo-
Size logic
Site Name (ha) class Site Location
Allamakee County
Clear Creek | 139 F NW%SE%w NW" Sec. 27, T. 1OO N, R. 5 W
Clear Creek 2 4 F SW'%s NE’ NW'4 Sec. 27, T. 100 N, R. 5 W
Benton County
Boar Power 0.3 T SE’ NE% NE Sec. 12, T. 86 N, R. 9 W
Elberon 12 T SE%4NW'%4 Sec. 17, T. 83 N, R. 12 W
Gilchrist 1.8 T SE%4NW'%4 SE'% Sec. 12, T. 86 N, R. 9 W
Mt. Auburn 28.8 T SNE Sec. 23, T. 86 N, R. 11 W
Black Hawk County
Dunkerton SE | 12.1 T SE%NE'’% NE Sec. 11, T. 89 N, R. 11 W
Dunkerton SE 2 15.5 T NW%SE'% Sec. 2, T. 89 N,R. 11 W
Hammond Road 1.6 T NE” NW'4 SE Sec. 15, T. 87 N, R. 13 W
Mark Sand North 12 E SW%4NW'4 Sec. 19, T. 90 N, R. 14 W
Mark Sand South 3.6 E S%SE%SW'4 Sec. 19, T. 90 N, R. 14 W
St. Francis 0.4 T NWANW4 NW'4 Sec. 8, T. 89 N, R. 11 W
St. John 6.3 T NE”WNW'4 Sec. 27, T. 90 N, R. 12 W
Bremer County
Brayton-Horsley 47 T N*NW'4 SE Sec. 2, T. 92 N, R. 11 W
Bushing 20 T SWANEW NW Sec. 36, T. 93 N, R. 14 W
Frederika 0.9 T NW% NEW SW'4 Sec. 13, T. 93 N, R. 13 W
Horsley 0.3 T NW'%4 SE SE Sec. 2, T. 92 N,R. 11 W
Northwoods Park 3.4 QO SW'%4 NE" SE'4 Sec. 13, T. 93 N, R. 11 W
Plainfield 0.3 T SE% NE NE Sec. 31, T. 93 N, R. 14 W
Waverly North 0.66 T SE% NE’ NE Sec. 14, T. 92 N, R. 14 W
Buchanan County
Amish 8.8 T NW% NE SW'4 Sec. 18, T, 90 N, R. 9 W
Cutshall Access 40 O NE NE’ SW'4 Sec. 6, T. 89 N, R. 10 W
Masonville 10 T SE’ NE” NW'4 Sec. 13, T. 89 N, R. 7 W
Otterville 15 T WANE“ NE Sec. 14, T. 89 N, R. 10 W
Rowley 2.2 T NW%4SE'% SE Sec. 2, T. 87 N,R.9 W
Rowley North 0.1 T NE% NE SE% Sec. 2, T. 87 N, R.9 W
Rowley West 6.7 T SW'% SEs SE Sec. 5, T. 87 N, R. 9 W

150

Rhodora [Vol. 96

APPENDIX I. Continued.

Geo-
Size logic
Site Name (ha) class Site Location
Walker Sand Ridge
North 0.5 E SE’ NW'4 SW'4 Sec. 31, T. 87 N,R. 8 W
Walker Sand Ridge
0.9 E SW'%4SW'4 SW'4 Sec. 31, T. 87 N, R. 8 W
Winthrop West 40 T N*%SW% NW'4 Sec. 33, T. 89 N, R. 8 W
Butler County
Allison East | 14 TT SE NE'% SE Sec. 35, T. 92 N, R. 16 W
Allison East 2 10 T SW'%NE'% SE% Sec. 35,T. 92 N,R. 16 W
Allison East 3 2.7 TO NW'%4NE'% SE Sec. 35, T. 92 N,R. 16 W
Allison West 0.7 T SW%NW'4 NE Sec. 27, T. 92 N, R. 17 W
Austinville 0.1 B SW%4NW'% NW'44 Sec. 19, T. 90 N, R. 18 W
ig Ro 15 T NW'%4NE'%™ NW" Sec. 22, T. 92 N, R. 16 W
Clarksville NE 0.9 T NW'%4 SE% NW‘ Sec. 35, T. 93 N, R. 15 W
Clarksville South 0.6 T SW'%4SW'%4 SE'% Sec. 19, T.92 N,R. 15 W
Dumont 12 T SE SE%4 SE Sec. 18, T. 92 N, R. 18 W
Ft. Sumpter 1 7.7 TO S'%NW'4 NE Sec. 8, T. 91 N, R. 17 W
t. Sumpter 2 0.8 T SW%SW% SW'% Sec. 5,T.91 N,R.17W
Ft. Sumpter 3 7.3. T SE%NW'4 SE Sec. 10, T. 91 N, R. 17 W
New Albion 10 TT SE SW SW Sec. 9, T. 90 N. R. 16 W
Pilot Rock 0.7 T SE%SE% NE Sec. 21,T.92N, R. 17 W
Cedar County
Rochester South 0.9 E SW% NE’ SE'% Sec. 25, T. 79 N, R. 3 W
Cerro Gordo County
sea Slough 12.6 O E”%SW%4 NE Sec. 34,T.97N,R. 20 W
1 6.1 T NE’ NE% NW4 Sec. 6, T. 97 N, R. 21 W
Neutn 10 TO NE%NE'%s SW'4 Sec. 15, T. 97 N, R. 21 W
Pop 0.3 T NE%™SW'4 NW Sec. 9, T. 97 N, R. 21 W
scceen 0.9 B NW'%SW'4SW'4 Sec. 10, T. 94 N, R. 20 W
Wheelerwood 2.3 B SE% SE% NE Sec. 15, T. 97 N, R. 21 W
Winnebago | 04 T SW NE'% SE Sec. 5,T.97N,R. 21 W
Winnebago 2 05 T NW SE SE Sec. 9,T. 97 N, R. 21 W
Chickasaw County
Alpha East 0.3 T SE% SE%4 SE Sec. 1,T.94.N,R.11 W
Barker Station 12 T NE“”SW'%SW'% Sec. 25,T. 94.N,R. 12 W
Boyd 26.3 T SE NE'% NE Sec. 33, T.95 N,R. 12 W
Boyd North 0.3 T NW'%4 SE’ SE'% Sec. 21, T.95 N, R. 12 W
Boyd South 3.4 T NEW’ NE SW'4 Sec. 34,T.95 N,R. 12 W
Bradford North 15 TT SE SE'% SE% Sec. 4, T. 94N, R. 14 W
Chickasaw | 18.4 T SW%NW'4SW'4 Sec. 28, T. 95 NR. 14 W
Chickasaw 2 3.5 T SE’ NE NW'4 Sec. 28, T. 95 N,R. 14 W
East Fork 3.1 T NE’NW'% SW'4 Sec. 5, T. 95 N, R. 12 W

1994] Nekola— Northeastern Iowa Fens 151
APPENDIX I. Continued.
Geo-
Size logic
Site Name (ha) class Site Location
Grassley’s Folly 13.0 T SW%4SW'%4 SW'4 Sec. 13, T. 95 N, R. 12 W
Kleiss 16 T SW'%SE% NE Sec. 26, T. 95 N, R. 11 W
Lawler East 49 T NE”W’NW%4 NW'4 Sec. 11, T. 95 N, R. 11 W
Lawler NW 20.3 T SWASW'% SW'4 Sec. 32, T. 96 N, R. 11 W
Lawler S 6.5 T NW%4 NE NE Sec. 22, T. 95 N, R. 11 W
Marsh Creek 1.1 T NE”NE“ NW4 Sec. 21, T. 94.N, R. 11 W
New Hampton
East 36.7 T S*SE% SW Sec. 8, T. 95 N, R. 12 W
New Hampton
North 22 T NW'%4SW'4 SE Sec. 32, T. 96 N, R. 12 W
Simpson Creek 1 3.7 T NE“”SW'%4 SE Sec. 21, T. 95 N, R. 11 W
Simpson Creek 2 5.5 T SW NE NE Sec. 28, T. 95 N, R. 11 W
Simpson Creek 3 2.3 T NE NE% SW'4 Sec. 27, T. 95 N, R. 11 W
Split Rock 10 T SW%4SW'%4 SW'% Sec. 35, T. 94 N, R. 12 W
Stapleton Church 0.8 T NW SE%4 SE Sec. 23, T.95 N, R. 11 W
Two Mile Creek 18 T SW%4 SW SE'% Sec. 36, T. 95 N, R. 14 W
Clayton County
Postville 37.5 T NW Sec. 21, T. 95 N,R.6 W
Strawberry Point 97 T SW%4 NENW Sec. 25, T. 91 N, R. 6 W
Clinton County
rid 0.2 jE SE4NW% NE Sec. 36, T. 82 N,R. ILE
Springeponde 83 T NE”NW4% SW'4 Sec. 23, T. 81 N,R.2E
Toronto | 0.1 E SE“%’SW% NW" Sec. 24, T. 82 N,R. 1E
Toronto 2 13 E NE% SE'%4 NE Sec. 23, T. 82 N,R.1E
Delaware County
Hawker 7.3 T NW'%4NE% NW Sec. 22, T. 89 N, R. 6 W
Robinson 99 T SE4NW% SW'4 Sec. 19, T. 87 N, R. 6 W
Dubuque County
Epworth 0.2 B SW NE SE'% Sec. 4, T. 88 N, R. 1 W
Farley 0.2 B NE“”NW%4 NW Sec. 4, T. 88 N, R. 1 W
Fayette County
2.7 T NW SE% NW'%4 Sec. 29, T. 95 N, R. 10 W
Alpha SE 1 0.1 T SW% SE% SE Sec. 9, T. 94.N, R. 10 W
Alpha SE 2 6.1 T NESE SW'4 Sec. 9, T. 94 N, R. 10 W
Hawkeye 26.9 T SW'%4 SEY SW Sec. 11, T. 94. N, R. 10 W
Hunter Creek 0.9 T NE%SW'% SE'% Sec. 30, T. 91 = R.9 W
Oelwein West 43 T ENE SE% Sec. 13, T. 91 N 10 W
Otter Creek 20 T NW%4NW'4 NW'4 Sec. 31, T. N, R. 9W
Smithfield
Township Hall 3.5 T SW'%4 SE NE Sec. 16, T. 92 N, R. 8 W

152

Rhodora [Vol. 96

APPENDIX I. Continued.

Geo-
Size logic
Site Name (ha) class Site Location

St. Lucas 2.4 B S'rSW%4 SW'4 Sec. 4,T.95 N,R.9 W

Sumner SE | 6.4 T NE SE SE Sec. 6, T. 92 N,R. 10 W

Sumner SE 2 2.6 T NE'%4SW'4 SE Sec. 6, T. 92 .N,R. 10 W

Sumner SE 3 0.66 T SW SEY NW'4 Sec. 32, T. 93 N,R. 10 W

Turner Creek 1 2.1 T N%NW'4 SEM Sec. 11, T. 94 N, R. 9W

Turner Creek 3 0.2 T NE’ NE'% SE'%4 Sec. 11, T.94N,R.9 W

Wadena NW 0.8 B NW'4SE% SE% Sec. 18,T.93 .N,R. 7 W

West Union

South 7.1 TO SW'% NE SW'4 Sec. 5, T. 93 N, R. 8 W

Floyd County

Charles City East 10.0 TO NE%WSW'%4SW'4 Sec. 36, T. 96 N, R. 15 W

Charles City West 0.5 To NW'4 NE SE Sec. 34,T.96N,R. 16 W

Rockford 2.3 O NE’ NE'%4 SW'4 Sec. 10, T. 95 N, R. 18 W

Slough Creek 1 9.3 T NE%SW'% NE Sec. 21, T. 97 N, R. 16 W

Slough Creek 2 0.8 T NE'% NE'% NE Sec. 22, T. 97 N, R. 16 W

Slough Creek 3 10 T SEY NEY NE'%4 Sec. 22,T.97N,R 16 W

Slough Creek 4 12 T NW SE's NE Sec. 22, T. 97 N, R. 16 W

Slough Creek 5 49 T NW'% NE’ NW'4 Sec. 26, T. 97 N, R. 16 W
Franklin County

Maynes Creek 10 Bo NW'% NE SE% Sec. 12, T.91 N,R. 19 W

Spring Creek 1 11.9 F NE’™NW'%4 SE'% Sec. 26, T. 92 N, R. 21 W

Spring Creek 2 0.8 F SE%NE'% SE Sec. 24, T. 92 N,R. 21 W

Spring Creek 3 13 F NW'%4 SE'% SW'4 Sec. 24, T. 92 N, R. 21 W
Grundy County

Morrison 16 T N'%SE% NW'4 Sec. 15, T. 87 N,R. 16 W

New Hartford 14 T NE%4NW'‘4 NE" Sec. 5, T. 89 N, R. 15 W

Stout 0.6 T SW’ NW NE' Sec. 18, T. 89 N,R. 15 W

Wellsburg 4.1 T NW'%NW'%4 SEM Sec. 12, T. 88 N, R. 18 W
Howard County

Lime Springs 1.8 T SW'%SE% NE'% Sec. 27, T. 100 N, R. 12 W

Riceville North 12 T SW SW'4 SW'4 Sec. 8, T. 99 N, R. 14 W

Schley 21.5 T WhSW¥%4 NW'4 Sec. 16,7. 98 N, R. 11 W

Staff Creek 2.4 T NW'%4 SE’ SW'4 Sec. 28, T. 100 N, R. 14 W

Turkey River 85 T NW%4NE%SW'4 Sec. 31, T. 98 N, R. 11 W
Johnson County

Coralville 0.2 T SE% NE SE Sec. 20, T. 80 N, R. 6 W
Jones County

Olin 2.1. E NW'%4 SW'4 SE% Sec. 6, T. 83 N, R. 2 W

Temple Hill 0.3 E NE’4™NW'%4 NE'4 Sec. 17, T. 85 N, R. 1 W

1994] Nekola— Northeastern Iowa Fens 153
APPENDIX I. Continued.
Geo-
Size logic
Site Name (ha) class Site Location
Walnut Creek | 0.8 T NE%NE% SE Sec. 32, T. 84 N,R.4W
Walnut Creek 2 11 TT SE’NW'% SW Sec. 33, T. 84, N, R. 4 W
White Oak Creek 1.0 T NE SE% NE Sec. 26, T. 83 N, R. 4 W
Linn County
Central City 0.3 T NE”ANW4%4 NE Sec. 9, T. 85 N, R. 6 W
Loupee 15.9 T ENE NE Sec. 14, T. 83 N, R. 6 W
atus 18.5 T SE%NE'% Sec. 36, T. a N, R. 6 W
Moses Road 0.2 T NW'<4SE% SE Sec. 6, T. 86 N, R. 8 W
Paris 11 E SE”’NW'%4SW'4 Sec. 30, T. 86 N, R. 6 W
Patton 0.8 T ENW'% NE Sec. 6, T. 86 N, R. 6 W
Western College 0.66 T SE% NE SE'% Sec. 35, T. 82 N, R. 7 W
Whittier 16 T NE“™SW%NW'4 Sec. 8, T. 84.N, R. 5 W
Windy Oaks
ort 04 E SW'% SEs NE Sec. 29, T. 85 N, R.5 W
Windy Oaks
South 0.6 E NW%SW% NE Sec. 29, T. 85 N,R. 5 W
Mitchell County
Little Cedar 0.7 T SW% SE% SE% Sec. 9, T. 99 N, R. 16 W
McIntire 0.5 T SE%SE% SE% Sec. 35, T. 100 N, R. 15 W
Osage West 12.4 T NE%NW'% NE Sec. 34, T. 98 N, R. 18 W
Mona 1.4 T SW% SE SW'% Sec. 14, T. 100 N, R. 18 W
Riceville NW 2.1 T SW'% SE% Sec. 15, T. 99 N, R. 15 W
t. Ansgar 2.0 B NW%SW%4 SW Sec. 13, T. 99 N, R. 18 W
Sie North 24 T NW%4SW'4 SE'% Sec. 30, T. 100 N, R. 16 W
Stacyville SE 0.5 T N%SE% NW Sec. 9, T. 99 N, R. 16 W
Stone School 1.1 T NW% SE% NE'% Sec. 20, T. 99 N, R. 15 W
Muscatine County
Conesville 3.00 F SE% NE“ NE'4 Sec. 20, T. 76 N, R. 4 W
i 49.3 O NW*%4 SE Sec. 23, T. 77 N, R. 4 W
Wiese Slough 19.9 OO E* SEs NE Sec. 13, T. 78 N, R. 3 W
Tama County
Sand Hill 0.4 E NE%NW'4 SE Sec. 2, T. 82 N, R. 15 W
Winneshiek County
Dry Run 3.9 B SE%NE% SW'4 Sec. 35, T. 98 N, R.9 W
Freeport 14 B NE“ NE NE Sec. 14, T. 98 N, R. 8 W
Jackson Junction 14 T NE SW'4 NE Sec. 20, T. 96 N, R. 10 W
Kendallville South 5.1 To NW%sSW'% SE Sec. 9, T. 99 N, R. 10 W
Madison
Church | 2.8 B SW'% NE NE Sec. 16, T. 98 N, R. 9 W

154 Rhodora [Vol. 96

APPENDIX I. Continued.

Size logic

Site Name (ha) class Site Location
Madison
Church 2 46 B SE%4NW'4 NEA Sec. 21, T. 98 N, R.9 W
Worth County
Elk Creek 2.1 F SW'4SE% NW'4 Sec. 18, T. 99 N, R. 21 W
Worth Pond 50.1 F SW'4 NE NW'4 Sec. 26, T. 99 N, R. 20 W

Legend for geologic classes: T = pre-Illinoian till site, B = Bedrock site, E =
Eolian sand site; F = Fluvial sand site, O = Oxbow site.

APPENDIX I: ANNOTATED CATALOGUE OF
VASCULAR PLANT SPECIES FROM
NORTHEASTERN IOWA FENS

This annotated catalogue is based on collections and observations made from
1984 to 1991 by J. C. Nekola, and from a survey of the northeastern Iowa botanical
literature as detailed under Methods. The nomenclature follows that of Swink and
Wilhelm (1979). Where names differ from those of Kartesz (1993), the Kartesz
synonym follows in brackets. Following each binomial is a common name, rarit
in the northeastern Iowa and entire Iowa flora [using the following abbreviations:

= Rare in Iowan Erosional surface (Cooperrider, 1962; Eilers, 1971);
= Listed as rare Iowa plant by Howe et al. (1984);

Listed as Endangered by Roosa et al. (1989);

= Listed as Threatened by Roosa et al. ne

SC = Listed as Special Concern by Roosa et al. (19

NI = New to Iowan Surface flora (Cooperrider, oo ke. 1971);

NS = New to state flora]

omrnv
Il

Taxa incidental in the fen flora are also noted, as is an approximation of the
distribution and frequency of each taxa across this region. For selected rare taxa,
the number of sites of occurrence is recorded. All exotic taxa are preceded by an
Ook?

PTERIDOPHYTES
Equisetaceae

aren arvense L. (Field Horsetail): Incidental. Infrequent throughout on mar-

ae fluviatile L. (Water Horsetail): R, L Scattered throughout. (14 sites)

Equisetum sylvaticum L. (Wood Horsetail): R, L, E Incidental. Rare in extreme
north on margins. (1 site)

Equisetum x littorale Kuhlewein ex Rupr.: Rare in extreme east on an eolian sand
fen. (1 site

1994] Nekola— Northeastern Iowa Fens 155

Ophioglossaceae

a multifidum (Gmel.) Trev. (Leathery hi Fern): NI, L, E Incidental.
n sandy margins in south central. (2 sites)
Bowne simplex E. Hitche. (Small Sue Fer R, L, E Incidental. Rare on
san rgin in the south central. (1 s
Lee vulgatum L. [O. pusillim a idee Tongue Fern): R, L, SC
Scattered in central. (13 sites)

Osmundaceae

Osmunda regalis L. var. spectabilis (Willd.) Gray (Royal Fern): R, L, E Incidental.
are in southeast on moist margins of eolian sand and oxbow fens. (3 sites)

Polypodiac
(including Dry oma. oe ee

Athyrium filix-femina (L.) Roth (Lady Fern): Frequent in east-central.
cee fragilis (L.) Bernh. var. fragilis (Fragile Fern): NI Rare in north. (1

Dr ee cristata (L.) Gray (Crested Shield Fern): R Frequent in eastern half.
(26 sites)

cc x peta ss Braun ex Dowell) Druce: NI Rare on a margin in the
reme south. (1 s

pe sensi vsibill is L. ee Fern): Abundant and dominant throughout.

Thelypteris palustris Schott (Marsh Fern): Abundant and dominant throughout.

Selaginellaceae
Selaginella eclipes Buck (Northern Marsh Spike Moss): R, L, E Reported (and
single site in Muscatine Pore! (Guldner, 1960).
Although Thee from a calcareous depression in Worth County (Nekola,
1991), no extant fen populations have been observed.

GYMNOSPERMS
Cupressaceae
Juniperus virginiana L. (Red Cedar): Incidental. Uncommon throughout on dry
hummocks and margins.
DICOTYLEDONS
Aceraceae
Acer negundo L. (Box Elder): Incidental. Uncommon throughout in thickets and
0

n margins.
Acer saccharinum L. (Silver Maple): Incidental. Rare on margins.

156 Rhodora [Vol. 96

Anacardiaceae
Rhus radicans L. si aaa lie radicans (L.) Kuntze] (Poison Ivy): Incidental.
re on margin
Apocynaceae
Apocynum sibiricum Ag (4. cannabinum L.] (Indian Dogbane): Incidental. Un-
common throughou
Asclepiadaceae

Asclepias incarnata L. (Swamp Milkweed): Common throughou
Asclepias syriaca L. (Common Milkweed): Incidental. Dee throughout on
margins

Balsaminaceae
Impatiens capensis Meerb. (Spotted Jewelweed): Common throughout in discharge

zones.
Impatiens pallida Nutt. (Pale Jewelweed): Rare in discharge zones.

Betulaceae
Alnus rugosa (Regel) Fern. [A/nus i (L.) Moench ssp. rugosa (DuRoi) Clausen]
(Speckled Alder): L Uncommon i in north aaa where it is locally dominant.

(5S si es

Betula pumila L. var. lta Regel (Bog Birch): R, L, SC Locally common
in north central. (20

Betula x sandbergii Britt. ae s Birch): NI, NS Rare in north central. (1 site)

Callitrichaceae
Callitriche heterophylla Pursh (Large Water Starwort): R,L,T Incidental. Rare in
cold water of discharge streams. (2 sites)
Campanulaceae

Campanula aparinoides Pursh (Marsh Bellflower): Abundant throughout.
Lobelia kalmii L. (Kalm’s Lobelia): NI, L, T Rare in northwest on low vegetation

mats. (4 sites)
Lobelia siphilitica L. (Great Blue Lobelia): Frequent throughout.

Caprifoliaceae

Sambucus canadensis L. (Elderberry): Infrequent throughout in thickets.
Viburnum lentago L. (Nannyberry): Infrequent throughout in thickets.

1994] Nekola— Northeastern Iowa Fens 157

Caryophyllaceae

Arenaria lateriflora L. [Moehringia lateriflora (L.) Fenzl.] (Wood Sandwort): In-
cidental. Uncommon in eastern counties.

area vulgatum L. [Cerastium fontanum Baumg. ssp. vulgare (Hartman)
Greuter & Burdet] (Mouse-eared Chickweed): Incidental. Throughout on
heavily grazed margins

aed nivea (Nutt.) Muhl. ex Otth. (Snowy Campion): R Scattered in southern

Saige longifolia Muhl. ex Willd. (Stitchwort): Frequent in eastern half.

Compositae

pense millefolium L. (Yarrow): Dry hummocks and margins throughout.
— rosia artemisiifolia L. (Common Ragweed): Incidental. Rare on dry hum-

ocks.
A ms trifida L. (Giant Ragweed): Locally dominant throughout on disturbed

sol

diene serrata Nutt. (Saw-toothed Sage): L Uncommon on margins in north-
est. (5 sites)

Aster ee ‘ (Heath Aster): Incidental. Throughout on dry humm ;

Aster ellie Rydb. [4. borealis (Torr. & ee Prov.] (Rush renee R,L,E

ees third on cold, wet soil. (4 s1

Aster novae-angliae L. (New England Aster): Se throughout.

Aster praealtus (Willow Aster): Scattered in east half.

Aster puniceus L. (Swamp Aster): L Abundant Sea (152 sites)

Aster umbellatus P. Mill. (Flat-top Aster): Comm aici Most material is
referable to the doubtfully distinct Aster Vee Cro

Pee cernua L. (Nodding Bur Marigold): Uncommon cain on exposed
peat.

Bidens coronata (L.) Britt. R (Tall Swamp Marigold): Frequent throughout.

Bidens frondosa L. (Common Beggar’s Ticks): Uncommon on disturbed soils.

Boltonia asteroides (L.) L’Her. (False Aster): Common throughout

Cacalia Benge L. [Synosma suaveolens (L.) Raf. ex Britt.] (Sweet Indian

P : R, L, T Scattered and rare on margins and shrub thickets. (2 sites)
Cirsium re mre (L.) Hill (Tall Thistle): R Incidental. Scattered on disturbed

soils.

*Cirsium arvense (L.) Scop. (Canadian Thistle): Incidental. Infrequent on disturbed
margins.

Cirsium discolor (Muhl. ex Willd.) Spreng. (Pasture Thistle): Incidental. Infrequent

on margins.
Cirsium muticum Michx. (Swamp Thistle): R, L Common in northwest, rare and
elsewhere. (30 sites)

Erechtities hieraciifolia (L.) Raf. ex DC. (Fireweed): R Scattered throughout on
exposed peat.

Eupatorium maculatum L. (Spotted Joe-pye Weed): Common throughout.

Eupatorium perfoliatum L. (Common Boneset): Frequent throughout on mineral-
rich soil

158 Rhodora [Vol. 96

pee rugosum Houtt. [Ageratina altissima (L.) King & H. E. Robins. var.
altissima] (White Snakeroot): Incidental. Rare in south.

pial autumnal L. (Sneezeweed): ee throughout. One of the most

s late-summer flowerin

Helianthus erosserats aes eae Sunflower Common throughout, be-
coming most abundant on disturbed areas

Lactuca pene L. (Wild Lettuce): ae 1. Infrequent on margin

Liatris je sania (Nels. ) Schum. (Northern Blazingstar): R, L ficidenial anes in

aire eel aor es Michx. (Prairie Blazingstar): pene in north half.

Rudbeckia hirta L. (Black-eyed Susan): Abundant throughou

Senecio aureus L. (Golden Ragwort): R Frequent locally in nee half.

ics pauperculus Michx. (Common Ragwort): Incidental. Rare on dry hum-

Mar ins

Sian perfoliatum L. (Cup Plant): Incidental. Rare on margin

Solidago sie ssima L. [S. canadensis L.] is ee a throughout.
Increases in dominance with distur

Sanaa? eigantea Ait. (Late ae ener throughout north half.

L.) Salisb. [Euthamia graminifolia(L.) Nutt.] (Grass-leaved

Goldenrod): “Common througho

Solidago Aes see ex Willd. R, 7 e (Broad-leaved Goldenrod): Rare in ex-
treme south. (1 s1

Solidago riddelli ee ex Riddell (Riddell’s Goldenrod): R Frequent throughout,

ss common in southeast. (62 sites

Solidago uliginosa Nutt. (Bog Goldenrod): NI, NS Rare in northeast. (2 sites)

*Sonchus asper (L.) Hill. (Sow Thistle): R Rare on disturbed areas throughout.

*Taraxacum officinale Weber (Common een Frequent throughout. This
is the most frequently encountered exotic s

Vernonia fasciculata Michx. (Common Tronweed). rredueil throughout on mar-
gins

Convolvulaceae
(including Cuscutaceae)

Convolvulus sepium L. aaa sepium (L.) R. Br.] (Hedge Bindweed): Inci-
denta

are on mar
Cuscuta cephalanthi enecin. (Buttonbush Dodder): R Infrequent on tall herba-
ceous plants
Cornaceae

Cornus alternifolia L. f. (Alternate-leaved Dogwood): Incidental. Rare in thickets.
Cornus obliqua Raf. [C. amomum P. Mill. ssp. obliqua (Raf.) J. S. Wilson] (Pale
ogwood): Frequent throughout in thickets
Cornus stolonifera Michx. [C. sericea L. ssp. sericea] (Red-osier Dogwood): R, L
Scattered throughout, but most common in north half in thickets.

1994] Nekola— Northeastern Iowa Fens 159

Cruciferae

Cardamine bulbosa (Schreb. ex Muhl.) BSP (Bulbous Cress): Abundant through-

out.
*Nasturtium officinale R. Br. [Rorippa nasturtium-aquaticum (L.) Hayek] (Water
Cress): R Scattered throughout in cold, flowing water of discharge streams.

Cucurbitaceae

Echinocystis lobata (Michx.) T. & G. (Wild Cucumber): Incidental. Rare vine of
thickets.

Euphorbiaceae

Acalypha rhomboidea Raf. [A. virginica L. var. rhomboidea (Raf.) Cooperrider]
(Three-seeded Mercury): Frequent throughout on dry hummocks.

Gentianaceae

Gentiana andrewsii Griseb. (Bottle Gentian): Common throu

Gentiana crinita Froel. Saeeaec crinita (Froel) Ma] (Large ae Gentian):
R, L Locally common in south half. (34 sites)

Gentiana procera Holm ensanop re (Holm) ee (Small Fringed Gen-
tian): R, L, T Locally common inn half. (21 sites)

Gentiana quinquefolia L. (Gentianella nea (L.) a (Stiff Gentian): Scat-
tered throughout.

Gentiana crinita Froel. x Gentiana procera Holm: NI, NS Rare in north central.
(1 site)

Hypericaceae

ee ign (Gray) Britt. ue St. John’s Wort): L Scattered throughout
n hummocks and disturbed soil.
pean oan Lam. ae St. John’s Wort): Incidental. Rare in mar-
i

gins.
Hypericum halacsadl Ait. [Hypericum ascyron L.] (Great St. John’s Wort):
Frequent in east
Hypericum pe ennai Michx. (Round-fruited St. John’s Wort): Scattered
throughout.
Triadenum fraseri (Spach) Gl. (Marsh St. John’s Wort): R, L Locally common
except in northwest. (54 sites)

Labiatae

sac a Muhl. ex W. Bart. (Common Water Horehound): Abundant
thro ut,

Lycopus Gaies Michx. (Northern Bugleweed): Less common, but found
throughout.

Lycopus virginicus L. (Bugleweed): R Scattered throughout.

160 Rhodora [Vol. 96

Mentha arvensis L. (Wild Mint): Scattered throughout.
Monarada fistulosa L. (Wild Bergamot): Incidental. Frequent on margins
Physostegia virginiana (L.) Benth. (Obedient Plant): Scattered oe
*Prunella vulgaris L. (Self Fea oo Throughout on margins, dry hum-
mocks and disturbed, dry s
grin ocl virginianum aa & aes ex B. L. Robins & Fern.
mmon Mountain Mint): Abundant throughou
Sc stab ae . (Marsh Skullcap): R Sa in south half, scattered
ere. (33 sites
eae lateriflora . (Mad-dog Skullcap): Uncommon on margins.
Stachys palustris L. (Woundwort): Frequent.
Stachys tenuifolia Willd. (Smooth Hedge Nettle): Occasional in south half.
Teucrium canadense L. (Wild Germander): Rare in east central.

Leguminosae

Amorpha fruticosa L. (Indigo Bush): Occasional in east half in shrub thickets.

Amphicarpa bracteata (L.) Fern. (Hog Peanut): Incidental. Rare on margins.

Apios americana Medik. (Ground Nut): R Incidental. Rare on margins.

Cassia marylandica L. [Senna marilandica (L.) Link] (Wild Senna): R Incidental?
Reported from a single, destroyed Muscatine County site (Guldner, 1960).

sido canadense ee (Showy Tick Trefoil): Incidental. Scattered on
dry mocks and m

Lathyrus aah L. (Marsh Vetchling) Common throughou

Lathyrus venosus Muhl. ex Willd. (Veiny Pea): Incidental. ee on dry margins.

Strophostyles helvula (L.) Ell. (Trailing Wild Bean): Incidental. Rare on dry mar-

gins.

Lythraceae

Lythrum alatum Pursh (Winged Loosestrife): Abundant throughout. Often reaches
highest dominance in zones of low vegetation

Melastomataceae

_ virginica L. (Meadow Beauty): R, L, E ae Rare on an eastern
ian sand site, in proximity to a vernal pool. (1 s

Menyanthaceae

Menyanthes trifoliata . (Buckbean): R, L, E Uncommon throughout, preferring
very wet soil. (6 sites)

Oleaceae

Fraxinus nigra Marsh. (Black Ash): Incidental. Occasional in shrub thickets and
ry margins

1994] Nekola— Northeastern Iowa Fens 161

Onagraceae

Epilobium coloratum Biehler (Cinnamon Willow Herb): Frequent throughout.

at seen Lehm. [E. ciliatum Raf. ssp. glandulosum (Lehm.) Hoch
& Raven] (Northern Willow ao R Scattered throughout. Often difficult
to se mle from the preceding taxa.

Epilobium leptophyllum Raf. (Fen saise Herb): R Frequent throughout. (58

sites
asc es Muhl. ex Spreng. (Downy Willow Herb): NI, NS Rare in

al. (5 sites)
Lui a (L.) Ell. (Marsh Purslane): R Incidental. Rare on wet sand of
arge streams in east.
Geen biennis L. (Common Evening Primrose): Incidental. Rare on dry hum-

s.
Oenothera pilosella Raf. (Prairie Sundrops): Rare in extreme south.

Oxalidaceae

Oxalis violacea L. (Violet Wood Sorrel): Incidental. Rare on margins.

Polemoniaceae

Phlox maculata L. (Wild Sweet William): Common throughout

Phlox pilosa L. (Prairie Phlox): Incidental. Occasional on dry: hummocks and
margins in west half.

Polemonium reptans L. (Jacob’s Ladder): Frequent in east half.

Polygonaceae

Polygonum amphibium L. (Water Knotweed): R Scattered in east half.
Polygonum pensylvanicum L. (Pennsylvania Knotweed): Common throughout.
be ee L. (Arrow-leaved Tear-thumb): pene! throughout, be-
nant with disturban
es orbits Gray (Great ee ee: R, L Frequent throughout in very
wet S

Primulaceae

Dodecatheon meadia L. (Shooting Star): Incidental. Rare in east central on dry
margins.

Lysimachia ciliata L. (Fringed Loosestrife): Incidental. Scattered in centr

Lysimachia quadriflora Sims as leaved me fe): Abundant aioe
becoming most dominant in low etatio

Lysimachia terrestris (L.) BSP. Sane. Cc ee A aaa. Rare in margins.

Lysimachia thyrsiflora L. (Tufted Loosestrife): R Scattered throughout, often in
very wet soil. (13 sites)

162 Rhodora [Vol. 96

Ranunculaceae

Anemone canadensis L. (Canada Anemone): Incidental. Frequent on margins.
Anemone quinquefolia L. (Wood Anemone): Incidental. Rare on dry hummocks.

Ranunculus pensyivanicus L. f. (Bristly Buttercup): Frequent } in east half.
Thalictrum dasycarpum Fisch. & Lall. (Common Meadow Rue): Common
throughout.

Rosaceae

Agrimonia parviflora Ait. (Swamp Agrimony): R Frequent throughout.

Filipendula rubra (Hill) B. L. Robins. (Queen of the Prairie): R, L, E No extant
populations seen on fen habitats although an old report exists from a Mus-
catine County s

Geum aleppicum iat wv. ellow Avens): R Frequent throughout. More common
than following species

Geum laciniatum Murr. (Rough Avens): Occasional in dry m ns.

Potentilla norvegica L. (Rough Cinquefoil): Frequent in png

Potentilla palustris (L.) Scop. [Comarum palustre L.] (Marsh Cinguefi R Rare
throughout, absent from southeast. Prefers very wet soils. )

pire; fees Michx. (Common Cinquefoil): Incidental. Beas on mar-

nd dry hummocks.
Prunus eee L. (Choke Cherry): Incidental. Rare in margins and thickets.
Rosa blanda Ait. (Early wild Rose): Scattered throughout.

1 1
o J MUU ARS

s throu ghout.
ces owe L. (Black Raspberry): Incidental. Rare in shrub thickets and

Rubus pier te Raf. (Dwarf Raspberry): NI, L Scattered and rare throughout.

sites
Spiraea alba DuRoi (Meadowsweet): Common throughout.

Rubiaceae

Galium pune - (Northern Bedstraw): Frequent in east half on dry hummocks
an
mee laradricum (Wieg.) Wieg. (Bog Bedstraw): R, L, E Rare in north half.

Galium ralnd Bigelow (Wild Madder): Common throughou
Galium tinctorium (L.) Scop. (Stiff Bedstraw): Occasional ae
rasige : ment . (Small Bedstraw): R, L Rare in north half in ete wet soil. (2

1994] Nekola— Northeastern Iowa Fens 163
Salicaceae
Populus he dos Michx. (Trembling Aspen): Occasional in northeast in thick-
ets and m
Salix bebbiana a (Beaked Willow): Common in north half.
Salix oe Fluegge ex Willd. (Sage Willow): R, L, SC Locally frequent except
i me east, where absent. (50 sites)
Salix dieu Muhl. (Pussy Willow): Common throughou
Salix interior Rowlee. [S. exigua Nutt.] (Sandbar iat Incidental. Scattered
ins and discharge streams.
Salix lucida Mubhl. (Shining Willow): R, L, E Rare in north half. No extant pop-
ulations observed, but historical collections exist from at least three of the
nes inventoried (Fitzpatrick, 1899; Tolstead, 1938; Hartley, 1966; Eilers,

cae pedicelaris Pursh (Bog Willow): R, L, T Scattered and rare in north half.
(11 sites)

Salix petiolars Sm. (Petioled Willow): Common throughou
Salix rigida Muhl. [S. eriocephala Michx.] (Heart- ne Willow): Common
oughout.
Salix sericea Marsh. (Silky Willow): NI, L Scattered in central. (5 sites)
Salix x clarkei Bebb. (Clark’s Willow): NI, NS Scattered over range of S. candida.
(23 sites)

Salix x cryptodonta Fern.: NI, NS Rare in west central. (2 sites)

eat x — Bebb. (Red Willow): NI, NS Scattered over range of S. candida,
less frequent than preceding. (13 sites)

Salix sve (Anders.) Schn. [S. petiolaris Sm.]: NI A few perplexing —.

es between S. petiolaris and S. sericea have been located in the no
ca may represent this taxon

i say Fluegge x Salix pedivellaris Pursh: NI, NS Very rare in sites of

ental co-occurrence. (2 sites)

Santalaceae

rie richardsiana Fern. [Comandra umbellata o poaike var. umbellata] (Bas-
Toadflax): Incidental. On dry hummocks and margins.

Saxifragaceae
(including Grossulariaceae)

Heuchera richardsonii R. Br. (Prairie Alum Root): Incidental. Scattered througt
hummocks an

aswaes glauca Raf. Pep Sain R, L Frequently scattered throughout.
(39 sites)

a sedoides L. (Ditch Stonecrop): Incidental. Occasional on muddy, dis-

margins.

FES americanum P. Mill. (Wild Black Currant): Scattered in north half in shrub

thicket

Saxifraga pel ireres L. (Swamp Saxifrage): Common in all but northwest.

164 Rhodora [Vol. 96

Scrophulariaceae

Chelone glabra L. (Turtlehead): R Common throughout. Six sites support pop-
ulations of the fen Secs butterfly Euphydryas phaeton phaeton, which
obligately feeds on this sp

Agalinis pat (Gray) ei (Fea False Foxglove): Scattered in central. Pop-
ulations growing in very wet, cold soil approach var. borealis Pennell. (16

sites

seer ei aba (L.) Pennell (Purple False Foxglove): Scattered in southeast

is perhaps best to lump this and the preceding species, although G.
pau var. borealis is quite disti

Agalinis renuifolia (Vahl) Raf. (Slender False Foxglove): Incidental. Rare in ex-

Gratolia neglecta Torr. (Clammy Hedge Hyssop): Incidental. Rare in south third

Mimulus glabratus HBK var. fremontii (Benth.) Grant. [M/. g. var. jamesii (Torr.
Gray ex Benth.) Gray] (Yellow Monkeyflower): R, L, E Rare in discharge
streams. (3 sites)
Mimulus ringens L. (Monkeyflower): Common throughout on margins.
Pedicularis lanceolata Michx. (Fen Betony): Common throughout.
Veronicastrum virginicum (L.) Farw. (Culver’s Root): Frequent throughout.

Solanaceae

* Solanum dulcamara L. (Bittersweet Nightshade): R Incidental. Rare on disturbed
margins.

Ulmaceae

*U/mus pumila L. (Siberian Elm): R Scattered throughout on dry hummocks and
shrub thickets.

Umbelliferae

Angelica atropurpurea L. (Giant Angelica): R, L, SC Frequent in northwest, very
e elsewhere. (8 sites)
Cicuta bulbifera L. (Bublet-bearing Water Hemlock): R Scattered throughout in
soil.

Cicuta maculata L. (Water Hemlock): Frequent throughou

Oxypolis rigidior (L.) Raf. (Cowbane): Common hee This is the most
ubiquitous fen umbel.

Sium suave Walt. (Water Parsnip): R Uncommon throu

Zizia aurea (L.) W. P. J. Koch (Golden Alexanders): ae inoue on dry
hummocks and margins.

Urticaceae

Boehmeria cylindrica (L.) Sw. (False Nettle): Most frequent in extreme east.

1994] Nekola— Northeastern Iowa Fens 165
Parietaria pensylvanica Muhl. ex Willd. (Pellitory): Incidental. Rare on hum-
mocks.
Pilea fontana (Lunnell) Rydb. (Spring Clearweed): R, L Frequent throughout. (63
sites
Pilea pumila (L.) Gray (Clearweed): Incidental. Rare on disturbed, wooded mar-
gins.
Urtica dioica L. (Tall Nettle): Frequent in north half. Becomes dominant following
disturbance.
Valerianaceae
Valeriana edulis Nutt. ex T. & G. (Common Valerian): R, L, SC Scattered in
central. (17 sites)
Verbenaceae

Verbena hastata L. (Blue Vervain): Common throughout, particularly on disturbed
margins

Violaceae

Viola lanceolata L. (Lance-leaved Violet): R Rare in south third on wet, sandy

ma
Viola sha ale Greene (Northern Bog Violet): Abundant throughout. The
st ubiquitous fen violet.

vane en (Banks ex DC.) Brainerd [V. macloskeyi F. E. Lloyd ssp. pallens
(Banks ex DC.) C. L. Hitche.] (Smooth White Violet): R Scattered in central.
(15 sites)

Viola primulifolia L. (Primrose Violet): R, L Rare in east. (1 site)

Viola sagittata Ait. (Arrow-leaved Violet): Incidental. Infrequent on dry margins.

Vitaceae

Parthenocissus vitacea (Knerr) A. S. ae (Thicket Creeper): Scattered through-
out on hummocks and m
Vitis riparia Michx. (Wild Grape): ae on shrub thickets.
MONOCOTYLEDONS
Alismataceae

Alisma subcordatum Raf. (Common Water Plantain): Occasional on muddy mar-

gins.
Sagittaria latifolia Willd. (Common Arrowhead): Frequent throughout in wet soil.

(including Acoriaceae)

Acorus calamus L. (Sweet Calamus): Scattered throughout.

166 Rhodora [Vol. 96

Symplocarpus foetidus (L.) Salisb. ex Nutt. (Skunk Cabbage): R Scattered in ex-
treme east. (3 sites)

Cyperaceae

Carex annectens (Bickn.) Bickn. (Yellow-fruited Sedge): Scattered throughout.

Carex bebbii Olney ex Fern. (Bebb’s Sedge): Scattered throughout

Carex buxbaumii Wahlenb. (Buxbaum’s Sedge): Frequent throughout.

Carex comosa Boott (Bristly Sedge): R Scattered throughout in very wet soil.

Carex cristatella Britt.: Uncommon on margins.

Carex diandra Hootie R, LA single a collection only (Eilers, 1971), no extant

pulations obser

Carex frankii Kunth (Frank s Sedge): R, L Rare in southeast on dry fen margins.

Carex ik aierset Muhl. ex Willd.: NI Scattered in north and northwest, preferring
neral rich soils. (9 sites

eae Ae ie Muhl. ex Willd. (Bottlebrush Sedge): oo throughout.

Carex interior Bailey (Inland Sedge): Common througho

Carex lacustris Willd.: R Scattered in north half.

Carex laeviconica Dew.: R, L Rare in southeast. (1 site)

Carex lanuginosa Nice’ (Woolly Sedge): Common througho

Carex sods Ehrh. var. americana Fern.: NI Scattered in central on very wet
s. (6 sites

ea ees Wahl.: NI, L, E Rare in south third, typically associated with eolian

sa

Carex normalis Mackenz.: Scattered throughout on margins.
Carex prairea Dew. ex Wood. (Fen Sedge): R, L Common throughout. (92 sites)
Carex rostrata Stokes var. utriculata (Boott) Bailey [C. utriculata Boott]: R, L

Carex scoparia Schkuhr ex Willd. (Pointed Broom Sedge) scattered seeNehONE
Carex sterilis Willd.: NI, NS Rare in central. (7 sites)

Carex stipata Muhl.: Uncommon in southeast on muddy margins.

Carex stricta Lam. (Tussock Sedge): Abundant throughout.

Carex suberecta (Olney) Britt.: R Absent from northwest (?), common elsewhere.
Carex tetanica Schkuhr: > L Scattered in north half. (16 sites)

Carex tribuloides Wahl.: R Uncommon throughout.

Carex vulpinoidea eerie (Fox Sedge): Common throughout on disturbed mar-

gins.
Cyperus rivularis Kunth. [C. bipartatus Torr.] (Brook Sedge): R Scattered and rare
along discharge streams and in low vegetation mats.

Bleochars acicularis (L.) R. & S. (Needle Spike Rush): R Incidental. Collected
nce (Eilers, 1971), but no extant populations observe

Gleocharis elliptica Kunth. (Fen Spike Rush): NI Frequent ‘Hrouehout: This is

ost ubiquitous fen oo (S1 sites)

Eleocharis palustris (L.) R. & S.: Scattered throughou

Eleocharis smallti Britt. a s Spike Rush): NI ete and rare

sli angustifolium Honckeny (Narrow-leaved Cotton Grass): R, SC Ab-
ent from extreme southeast, frequent elsewhere. (68 sites)

1994] Nekola— Northeastern Iowa Fens 167

Eriophorum gracile W. D. J. Koch (Slender Cotton Grass): R, L, E No extant
populations observed, however, old collections exist (Eilers, 1971).

mene ee L. (Tawny Cotton Grass): NI, NS Very rare in central in

m dominated fen on eolian sand ridge. (1 site)

Rhynchospor capillacea Torr. (Hairy Beak Rush): NI, L, T Rare and scattered
in north half, as mats of very low vegetation. (8 sites

Scirpus cyperinus (L.) Kunth. (Wool Grass): Frequent throughout

—_ fluviatilis nous Gray (River Bulrush): R Incidental. Rare in extreme

ut
serous valid Vahl. [S. tabernaemontani K. C. Gmel.] (Great Bulrush): Frequent
out. Often found in fens with mats of low vegetation
Sera ei Muhl. ex Willd. (Vanilla Nut Rush): NI, L, E Scattered and
rare in central. Is restricted to mats of low vegetation. (6 sites

Gramineae

* Agrostis alba L. [A. gigantea Roth.] (Redtop): Frequent throughout.

Agrostis hyemalis (Walt.) BSP. (Tickle Grass): Occasional throughout.

Andropogon gerardii Vit. (Big Bluestem): Uncommon in west half on dry hum-
mocks

Bromus ciliatus L. (Fringed Brome): R Occasional in north half. (7 sites)

Calamagrostis canadensis (Michx.) Beauv. (Blue Joint Grass): Abundant through-
out.

is ee inexpansa Gray [C. stricta (Timm) Koel. ssp. sess (Gray) C.
W. Greene] (Bog Reed Grass): NI Rare and scattered in extrem h.

* Digitaria sanguin (L.) Scop. (Smooth Crabgrass): Incidental. ae on dis-
turbed marg

Echinochloa crus- eal (L.) Beauv. (Barnyard Grass): R Incidental. Rare on dis-
ur

Festuca paradoxa Desk (Large Fescue): R Scattered throughout. (6 sites)

Glyceria grandis S. Wats. (Reed Manna Grass): R Occasional in very wet soil
along discharge streams.

Glyceria striata (Lam.) A. S. Hitchc. (Fowl Manna Grass): Abundant throughout.

Hierochloe odorata (L.) Beauv. (Vanilla Grass): R Scattered in north half.

*Hordeum jubatum L. (Squirrel-tail Grass): Incidental. Rare on dry hummocks.

Leerzia oryzoides (L.) Sw. (Rice Cut Grass): Frequent throughout, often in dis-
turbed sites.

Muhlenbergia glomerata (Willd.) Trin. (Fen Wild Timothy): R, L Common
throughout. (102 sites

Muhlenbergia mexicana (L.) Trin. (Leafy Satin Grass): R Common throughout.

a pales Nash iDiconthelae boreale (Nash) Frackmann] (Northern Panic

: NI, E Rare in south central, preferring dry hummocks and moist sand

ae (4 st se

Panicum implicatum Scribn. see aa acuminatum (Sw.) Gould & C. A.
Clark var. ae (Torr.) Frackmann]: Common throughout on dry
h ks and low vegetation mats.

Phalaris arundinacea L. (Reed Canary Grass): Native species extensively planted
for pasture ‘improvement’ on disturbed sites, where it becomes dominant.

168 Rhodora [Vol. 96

Rare and scattered in north third
a compressa L. (Canadian Bluegrass): Occasional throughout on margins.
Poa palustris L. (Marsh Bluegrass): Rare in wet soil.
*Poa pratensis L. (Kentucky Bluegrass): Common throughout on margins.
Sorghastrum nutans (L.) Nash (Indian Grass): Incidental. Rare on dry hummocks
and margins

Phragmites communis Trin. [P. australis (Cav.) Trin. ex Steud.] (Common Reed
Grass)

Spartina pectinata Link (Slough Grass): Scattered throughout on margins.
Sphenopholis intermedia (Rydb.) Rydb. (Slender Wedge Grass): R Common

throughout.
Sphenopholis obtusata (Michx.) Scribn. (Prairie Wedge Grass): R Rare and scat-
tered throughout.

Tridaceae

Iris virginica L. var. shrevei (Sm.) Anders. (Blue Flag): Common throughout.

Juncaceae

Juncus canadensis J. Gay ex Laharpe (Canadian Rush): R Incidental. Occasional
on margins.
Juncus dudleyi Wieg. (Dudley’s Rush): Common throughout
Juncus nodosus L. (Joint Rush): R Absent from extreme southeast on
elsewhere where it is most typically found in mats of low vegeta
Juncus tenuis Willd. (Path Rush): Incidental. Occasional on disturbed ‘ait margins.
Juncus torreyi Coville (Torrey’s Rush): R Incidental. Rare on fen margins.

Juncaginaceae

Trigloc be maritimum L, (Common Bog Arrow Grass): NI, L, T Rare in northwest.

(2 sites)
oan ees palustre L. (Slender Bog Arrow Grass): NI, L, T Rare in central. (2
sites)

Lemnaceae

Lemna minor L. (Small Duckweed): Incidental. Uncommon in discharge streams.
Spirodela polyrrhiza (L.) Schleid. (Great Duckweed): Incidental. Uncommon i
outlet streams.

Liliaceae

Hypoxis hirsuta (L.) Cov. (Yellow Star Grass): Incidental. Occasional on dry
u and margins
at cee la Farw. “vdchisan Lily): Incidental. Rare on dry hummocks

argins
Wa vireinicum L. (Bunchflower): R, L Incidental. Observed only once
on badly degraded site in south central.

1994] Nekola— Northeastern Iowa Fens 169

Smilacina stellata (L.) Desf. sacniaipageds stellatum (L.) Link] (Starry False
Soloman’s Seal): Incidental. Rare
Zygadenus elegans Pursh (White poe R Scattered in north half on margins.

Orchidaceae

desea candidum a ex Willd. (White Lady’s Slipper): R, L, T Rare in
st half. (3 sites)

cipnpeiom ae ns Salisb. (Small Yellow Lady’s Slipper): NI, NS Rare in
sites

Copripedu eee Willd. (Large Yellow Lady’s Slipper): R, L Occasional in

compan reginae Walt. (Showy Lady’s Slipper): R, L, E Rare in central. (1

nn, x andrewsii A. M. Fuller (Andrew’s Lady’s Slipper): NI, L Rare in
central. (1 site

Cypripedium ~ favillianum Curtis (Favill’s Lady’s Slipper): NI, NS Rare in central.
(1 site)

Habenaria flava (L.) R. Br. var. herbiola (R. Br. ex Ait. f.) Ames & Correll [Pla-
settle flava (L.) Lindl. var. herbiola (R. Br. ex Ait. f.) aaa (Green Orchid):
are in south third on wet, sandy margins. (2
pres hyperborea an R. Br. [Platanthera hyperborea cL, ) Lindl.] (Northern
Fringed Orchid): NI, L, T Very rare in north. (1 site)
eee praeclara Ren & Bowles) Cronq. [Platanthera praeclara Sheviak
wles] (Prairie Fringed Orchid): R, L, E Incidental. Rare in central. (1

site
aes psycodes (L.) Spreng. [Platanthera psycodes (L. ) Lindl.] (Purple Fringed
:R, L, T Scattered and rare throughout. ( )
Liparis eas (L.) Rich. (Fen Twayblade): R, L Scattered throughout. (12 sites)
Spiranthes cernua (L.) Rich. (Nodding Lady’s Tresses): Scattered throughout
Spiranthes lucida (H. H. Eat.) Ames (Early Lady’s Tresses): NI, NS, E Very rare
in northeast. (1 site)

Typhaceae
(including Sparganiaceae)
Sparganium eurycarpum pec ex Gray (Common Bur Reed): R Occasional
throughout in very w
Typha angustifolia L. oe ieawed Cattail): R struee thro ughout.
Typha latifolia L. (Common Cattail): Common throughou

RHODORA, Vol. 96, No. 886, pp. 170-178, 1994

NOMENCLATURAL NOTES IN NYMPHAEACEAE
FOR THE NORTH AMERICAN FLORA

JOHN H. WIERSEMA AND C. BARRE HELLQUIST

ABSTRACT

In conjunction with a study of the Nymphaeaceae in North America, the tax-
onomy and nomenclature of three taxa is reviewed. Two new combinations are
provided at subspecific rank for two taxa sometimes treated as species, one for
Nymphaea tuberosa Paine and one for Nuphar rubrodisca Morong, and both are
lectotypified. A neotype is selected for Nymphaea advena Aiton which serves to
maintain usage of Aiton’s epithet for a widespread taxon of Nuphar.

Key Words: Nuphar, Nymphaea, water-lilies, cow-lilies, spatterdock

Research on the genera Nymphaea and Nuphar of the Nym-
phaeaceae has revealed that the following nomenclatural adjust-
ments are necessary for a flora of North America treatment.

Nymphaea odorata Aiton subsp. tuberosa (Paine) Wiersema &
Hellquist, comb. nov.

Nymphaea tuberosa Paine, Annual Rep. State Cabinet Nat. Hist. New York
18: 184 (Cat. pl. Oneida Co. 132). 1865. TYPE: UNITED STATES.
New York: S. shore of Lake Ontario, 1865, Paine s.n. (LECTOTYPE:
K). See discussion.

Nymphaea odorata, which is distributed throughout eastern
North America, 1s a polymorphic species. In and around the Great
Lakes region, where the plants here designated as subsp. tuberosa
are found, two predominate forms can be observed. In the south-
ern part of the range of subsp. tuberosa where subsp. odorata is
absent, e.g., in Iowa, Illinois, Indiana, Ohio, and somewhat to
the north and east, plants of subsp. twberosa are easily distin-
guished morphologically from subsp. odorata (see accompanying
key). Further north where their ranges overlap occasional popu-
lations are intermediate in morphology or more rarely popula-
tions may include plants referable to both subspecies as well as
intermediate plants. The intermediates exhibit a range of varia-
tion spanning the morphological gap between the two subspecies
and, in some cases at least, display no evidence of reduced fertility.
Although traditional treatments distinguished the two forms at
specific rank, several recent floristic works (e.g., Voss, 1985; Glea-

170

1994] Wiersema and Hellquist— Nymphaeaceae 171

son and Cronquist, 1991) have combined them into one variable
species without further distinction. While calling attention to this
taxonomic problem, field studies from within this region (Mon-
son, 1960; Williams, 1970; Bayly and Jongejan, 1982) have not
sufficiently accounted for the observed variation. These studies
suggest that some variability may be induced by environmental
conditions; however, we have observed both extremes growing
together under seemingly identical conditions. Such populations
require more detailed study before this variation is fully under-
stood. Artificial hybridization studies and/or molecular approach-
es may also aid in clarifying this relationship.

Based on existing knowledge, we believe the geographic pat-
terning of the overall variation and the usefulness of retaining a
separate status for those forms previously classified as Nymphaea
tuberosa justifies the recognition of two subspecies, as distin-
guished below. While useful in separating the two extremes in
this morphological continuum, the key is of limited use in iden-
tifying intermediate plants. Compounding the problem of iden-
tification is the fact that key characters are often poorly repre-
sented on herbarium material. Populations containing
intermediate plants are known from Minnesota, Wisconsin,
Michigan, New York, Vermont, and southern Ontario and Que-
bec and until better understood are best treated as Nymphaea
odorata without regard to subspecies.

1. Petioles not striped; blades usually reddish-purple (occasion-
ally green) abaxially; seeds 1.5-2.5 mm long ............

Se eT eT Te Te Terns ee Seer eee eee ree eee subsp. odorata

1. Petioles with brown-purple stripes; blades green or faintly pur-
ple abaxially; seeds mostly 2.8-4.5 mm long ............

bea eae a4 4s ase ee ee ae oe subsp. tuberosa

Paine (1865) cited anumber of localities for Nymphaea tuberosa
but failed to designate a holotype. To fix the application of his
name it is appropriate to select a lectotype. Conard’s (1905) listing
of ““Nymphaea tuberosa Paine (1865), fid. specimen coll. Paine
on S. shore of Lake Ontario, from hb. A. Gray, in hb. Kew” is
considered not to represent an effective lectotypification as it does
not satisfy the requirements of Article 8.3 of the International
Code of Botanical Nomenclature (Greuter et al., 1988). The spec-

Ly2 Rhodora [Vol. 96

imen at Kew contains a leaf, two flowers, and a developing fruit
and matches our concept of subsp. tuberosa; it is here formally
designated as lectotype.

Nuphar lutea (L.) Smith subsp. rubrodisca (Morong) Hellquist &
Wiersema, comb. nov.

Nuphar rubrodiscum Morong, Bot. Gaz. 11: 167-168. 1886. TYPE: UNIT-
ST S. Vermont: at the mouth of Lewis Creek, Lake Champlain,
Ferrisburgh, 5 Aug 1885, Morong s.n. (LECTOTYPE: NY). See dis-
cussion. Nymphaea rubrodisca (Morong) E. Greene, Bull. Torrey Bot.
Club 15: 84. 1888

While working with the genus Nuphar Smith, we became aware
of the need for a new combination in addition to those at sub-
specific rank previously made by Beal (1956). Nuphar lutea subsp.
rubrodisca, which Beal treated under the hybrid formula N. /utea
subsp. pumila (Timm) E. O. Beal x N. lutea subsp. variegata
(Durand) E. O. Beal and most likely of hybrid origin, is producing
viable seed and is found in areas far removed from either of the
probable parents. It differs from the other two subspecies in a
number of characteristics, as detailed in our flora treatment, and
in accordance with Article H.3.4 Note | and Example 3, we prefer
to treat this as an additional subspecies. As no name at subspecific
rank exists for this taxon, Morong’s epithet, which has most com-
monly been applied to it, can be retained.

As Morong failed to designate a holotype, a lectotype is selected.
In his original publication, Morong discussed his study of this
taxon during the summer of 1885 along Lake Champlain at Fer-
risburgh, Vermont near the mouths of Lewis and Little Otter
creeks. Following the description, he lists ‘““Lake Champlain, Vt.”
as the type locality. Three sheets of this taxon from Morong’s
original herbarium now at NY pertain to this study. One sheet
stamped ““MORONG HERBARIUM” contains 3 leaves and 3
mounted and several unmounted fruits and bears two labels: 1)
“Ivs. of N. rubrodiscum, Ferrisburgh, Vt., Aug. 5, 1885” and 2)
“N. rubrodiscum fruit, Ferrisburgh, Aug. 11.’ A second sheet
stamped “MORONG HERBARIUM” and “BRITTON HER-
BARIUM” contains 4 leaves, 1 mounted flower, and 3 mounted
and some unmounted fruits and bears 4 labels: 1) ““Leaves of N.
rubrodiscum, Lewis Creek, Ferrisburgh, Vt.’’; 2) identical with /;

1994] Wiersema and Hellquist— Nymphaeaceae 173

3) “N. rubrodiscum Morong, Lewis Creek, Ferrisburgh, Vt.; Aug.
6, 1885"; and 4) “N. rubrodiscum, Lewis Creek, Ferrisburgh, Vt.,
Aug. 11, 1885.” On all the above labels the epithet /uteum has
been overwritten with rubrodiscum. The third sheet stamped
“MORONG HERBARIUM, property of BARNARD COL-
LEGE” contains | leaf, 1 flower, and 2 fruits and bears a single
label: ‘“Nuphar rubrodiscum Morong, N. luteum Sm.?. At the
mouth of Lewis Creek, Lake Champlain, Ferrisburgh, Vermont.
Leg. T.M. 18 5/8 85.” This sheet is accompanied by a lengthy
note written by Morong the content of which is entirely repro-
duced in the protologue. On this sheet the provisional name Nu-
phar rubrum has been replaced with N. rubrodiscum both on the
label and the note. While all three sheets match the original de-
scription, this third sheet is the only one which appears to rep-
resent a single gathering. It is also the most completely labelled
and contains both flowers and fruits. It has been selected as lec-
totype.

Nuphar lutea (L.) Sm. subsp. advena (Aiton) Kartesz & Gandhi,
Phytologia 67: 463. 1989, “advenum.”

Nymphaea advena Aiton, Hort. kew. 2: 226. 1789. TYPE: UNITED STATES.
Pennsylvania: Philadelphia, tidal marsh along Darby Creek in John
Heinz National Wildlife Refuge at Tinicum, 24 July 1993, J. H. Wier-
sema & A. E. Schuyler 2372 (NEOTYPE: PH; ISONEOTYPES: US,
BM). See discussion. Nuphar advena (Aiton) W. T. Aiton, Hort. kew.
ed. 2, 3: 295. 1811.

In conjunction with this study, we have been investigating the
typification of Nymphaea advena Aiton, which has commonly
been applied to a taxon of North American Nuphar, as Nuphar
advena (Aiton) W. T. Aiton or Nuphar lutea subsp. advena (Aiton)
Kartesz & Gandhi. The original publication (Aiton, 1789) is a
“catalogue of the plants cultivated in the Royal Botanic Gardens
at Kew.” As is typical of most early botanical publications, no
specimens were directly cited in the protologue of Nymphaea
advena. The native distribution was given as “North America,”
the phrase name Nymphaea floribus flavis Clayton in Gronovius,
Flora virginica 164. 1743 was cited, and it was said to have been
introduced by Mr. William Young in 1772. According to Stafleu
and Cowan (1976), type material for both Hortus kewensis and

174 Rhodora [Vol. 96

Flora virginica is deposited at BM. However, from correspon-
dence with the herbaria of both Kew (G. L. Lucas, pers. comm.)
and British Museum (R. Vickery, pers. comm.) it is clear that no
material of the Kew cultivation or the John Clayton collection,
which served as the basis for the Gronovius phrase name, can be
located.

The original description appears to combine characteristics of
two taxa, Nuphar advena and Nuphar variegata Durand, which
we distinguish at the rank of subspecies following Beal (1956).
The traits of semiterete petioles and purple-colored sepals and
stamens best apply to Nuphar lutea subsp. variegata (Durand) E.
O. Beal while the emergent leaves clearly indicate subsp. advena.
Most early users of Aiton’s name (Poiret, 1798; Willdenow, 1799,
1809; Michaux, 1803; Sims, 1803; Martyn, 1807; de Candolle,
1821; Torrey and Gray, 1838; Planchon, 1853; Morong, 1886)
did not recognize the distinctions between the two taxa. Pursh
(1814) distinguished the two, but misapplied the European Nu-
phar lutea to what was named Nuphar variegata in 1866. Bigelow
(1824) and Hooker (1829), while noting the differences between
the northern and southern plants, continued to treat both as a
single taxon. Hooker’s comment that ‘Dr. Graham and myself
have long observed that the N. advena, as cultivated in our gar-
dens, has the leaves sometimes floating, sometimes rising above
the water” indicates that both taxa were introduced to Europe.
Whether or not this was the case some 40 years earlier when the
original description was published by Aiton is not known. In any
event, none of these early authors succeeded in typifying Nym-
phaea advena.

The relationship between the two taxa was clarified by Miller
(1902). Though he stated that the type locality of Nymphaea
advena was probably Philadelphia, Miller failed to lectotypify the
name. In their revision, Miller and Standley (1912) listed the type
locality as “vicinity of Philadelphia, Pennsylvania” but cited no
type specimen. Beal’s 1956 revision accepted Nuphar lutea subsp.
variegata for the floating-leaved northern taxon and Nuphar lutea
subsp. macrophylla (Small) E. O. Beal for the emergent-leaved
southern one, listing Aiton’s name as a partial synonym of both.
For nomenclatural reasons, Kartesz and Gandhi (1989) replaced
subsp. macrophylla with subsp. advena (Aiton) Kartesz & Gandhi.
To this day, however, Aiton’s name has never been properly
typified. Since Miller it has consistently been applied to the emer-

1994] Wiersema and Hellquist— Nymphaeaceae 175

gent-leaved taxon and to preserve this usage the name must be
typified on that element.

The supposition that the type locality should be Philadelphia
was presumably based on that being the home of the William
Young referred to by Aiton as having introduced the plant to
England in 1772. According to Harshberger (1917), William
Young, Jr. (1742-85) of Philadelphia, in his capacity as botanist
to the Queen of England, departed Philadelphia for England in
November of 1771, no doubt carrying the Nuphar material he is
credited with having introduced the following year. A 1772 letter
from Dr. John Fothergill of London to Humphrey Marshall of
Philadelphia, reported by Rhoads (1916), reports receipt of ma-
terial of Ne/umbo from William Young, Jr. that same year. Young
is known to have carried material from the Carolinas abroad but
this was in 1768 and 1769 (Harshberger, 1917). As he did not
apparently make any further trips to the southern states following
his return to Philadelphia in 1770, his 1772 introductions to
England would seem to have been collected near to his home in
Philadelphia.

In his flora of the Philadelphia area, William Barton (1818)
applied Nuphar advena to an emergent-leaved taxon which was
said to be abundant “on the marshy shores of the Delaware,
Schuylkill, and all other waters in our neighborhood, covering the
shores for miles together in extent” and in terms of current usage
this appears to be the desired application of the name. Young’s
estate was reportedly adjacent to that of John Bartram, which
bordered the Schuylkill River (Harshberger, 1917). A recent field
trip in the company of A. E. Schuyler provided an opportunity
to study Nuphar in the Philadelphia area on both sides of the
Delaware River near the mouth of the Schuylkill. Large popu-
lations still exist in some protected areas, probably remnants of
the formerly extensive distribution. Much of the area remains
under tidal influence as would have been the case in Young’s
time.

The populations consist of strongly emergent plants, almost
completely so at low tide, which have flowers mostly with parts
variously tinged with reddish-purple. Such coloration is lacking
over most of the range of the taxon commonly referred to as N.
advena, but is characteristic of N. variegata, which is found at
nearby sites in southern New Jersey. A full range of intermediate
plants for those characters which normally distinguish the two

176 Rhodora [Vol. 96

taxa can be observed in other southern New Jersey localities.
Though the populations in the immediate vicinity of Philadelphia
display some degree of intermediacy, their overall morphology
compares well with N. advena as the name has been applied by
most authors. One of our collections near the mouth of the Schuyl-
kill thus serves as a suitable neotype.

ACKNOWLEDGMENTS

We are grateful to G. L. Lucas and R. Vickery for researching
type material at K and BM, M. Beasley for supplying information
on illustrations in the Young Collection at BM, A. E. Schuyler
for help with field work in and around Philadelphia, and G. P.
DeWolf, Jr., D. H. Nicolson, J. L. Reveal, A. Y. Rossman, E. L.
Schneider, J. W. Thieret, and two anonymous reviewers for re-
viewing earlier drafts of this manuscript.

LITERATURE CITED

Aiton, W. 1789. Hortus kewensis; or, a catalogue of the plants cultivated in the
Royal Botanic Garden at Kew. George Nicol, London.

Barton, W. P. C. 1818. Compendium Florae Philadelphicae: Containing a
Description of the Indigenous and Naturalized Plants Found Within a Circuit
of Ten Miles Around Philadelphia. Vol. 2. M. Carey & Son, Philadelphia.
234 pp

Bay Ly, I. L. AND K. JONGEJAN. 1982. A morphological and ecological variant
of the tuberous water lily, Nymphaea tuberosa Paine, from the Jock River,
Ottawa, Ontario. Canad. Field-Naturalist 96: 301-

BEAL, E.O. 1956. Taxonomic revision of the genus Nuphar Sm. of North Amer-
ica and Europe. J. Elisha Mitchell Sci. Soc. 72: 317-346.

BiGELow, J. 1824. Florula Bostoniensis: A Collection of Plants of Boston and
its Vicinity, 2nd ed. Cummings, Hilliard, & Co., Boston. 422 pp.

CANDOLLE, A. P. De. 1821. Regni Vegetabilis Systema Naturale. Vol. 2. Treuttel
et Wii

Conarp, H. S. 1905. The waterlilies: a monograph of the genus Nymphaea.
Publ. Sada Inst. Wash. 4: 1-279.

G eason, H. A. AND A. CRONQUIST. 1991. Manual of Vascular Plants of North-
eastern United States and Adjacent Canada, 2nd ed. New York Botanical
Garden, New York. 910 pp.

GREUTER, W., H. M. Burpet, W. G. CHALONER, V. DEMOULIN, R. GROLLE, D.
L. HAwkswortH, D. H. Nicoison, F. A. STAFLEU, E. G. Voss, AND J. MCNEILL.
1988. International code of botanical nomenclature, adopted by the Four-
teenth International Botanical Congress, Berlin, July-August 1987. Regnum
Veg. 118.

1994] Wiersema and Hellquist— Nymphaeaceae 177

HARSHBERGER, J. W. 1917. William Young, Jr., of Philadelphia, Queen’s Bot-
anist. Torreya 17: 91-99

Hooker, W. J. 1829. Flora Boreali-americana. Vol. 1. Henry G. Bohn, London.

51 pp

Kartesz, J. T. AND K. N. GANDHI. 1989. Nomenclatural notes for the North
American flora. I. Phytologia 67: 461-467.

Martyn, T. 1807. Philip Miller's Gardener’s and Botanist’s Dictionary. Law
and Gilbert, London.

MicHAux, A. 1803. Flora Boreali-americana. C. Crapelet, Paris.

MILterR, G.S., JR. 1902. The large yellow pond lilies of the northeastern United
States. Proc. Biol. Soc. Wash. 15: 11-13.

, AND P. C. STANDLEY. 1912. The North American species of Nymphaea.
Contr. U.S. Natl. Herb. 16: 63-108.

Monson, P. H. 1960. Variation in Nymphaea, the white waterlily, in the Itasca
State Park region. Proc. Minnesota Acad. Sci. 25-26: 26-39.

Moron, T. 1886. Revision of the North American species of Nuphar. Bot.
Gaz. 11: 164-169.

PAINE, J. A., JR. 1865. Catalogue of plants found in Oneida County and vicinity.
Ann. Rep. State Cabinet Nat. Hist. New York 18: 53-205.

PLANCHON, J.-E. 1853. Etudes sur les Nymphéacées. Ann. Sci. Nat., Bot. ser. 3,

19: 17-63.

Porret, J. L. M. 1798. Nenuphar; Nymphaea, pp. — In: Lamarck, En-
cyclopédie Méthodique. Botanique. Tome Quatrieme. H. Agasse, Paris.
Pursu, F. 1814. Flora Americae Septentrionalis. a Cochrane, and Co.,

London.

Ruoaps,S.N. 1916. Preface to Botanica Neglecta. William Young, Jr. ‘“Botaniste
de Pensylvanie” and his long-forgotten book ‘‘Catalogue d’Arbres Arbustes
et Plantes Herbacees d’Amerique” published in Paris in 1783. Privately
printed, Philadelphia.

Sims, J. 1803. Nymphaea advena. Three-colored water-lily. Bot. Mag. 17: t. 684.

STAFLEU, F. A. AND R. S. Cowan. 1976. Taxonomic Literature. Volume 1: A—
G. Bohn, Scheltema & Holkema, Utrecht.

TorREY, J. AND A. Gray. 1838. A Flora of North America. Wiley & Putnam,

Voss, E.G. 1985. Michigan flora, Part II. Dicots (Saururaceae-Cornaceae). Cran-
brook Institute of Science and University of Michigan Herbarium, Ann Ar-

WILLDENOW, C. L. 1799. Carolia Linné Species Plantarum... Tomus II. G. C.
Nauk, Berlin.

1809. Enumeratio Plantarum Horti Regii Botanici Berolinensis. Taberna
Libraria Scholae Realis, Berlin.

WILLIAMS, G. R. 1970. eae in the white waterlilies (Nymphaea) of
Michigan. Michigan Bot. 9: 72-86.

J. H.W.

UNITED STATES DEPARTMENT OF AGRICULTURE
AGRICULTURAL RESEARCH SERVICE

BUILDING 011A, BARC-WEST

BELTSVILLE, MARYLAND 20705-2350

178 Rhodora [Vol. 96

Ge eal = &

NORTH ADAMS STATE COLLEGE

BOX 9145

NORTH ADAMS, MASSACHUSETTS 01247

RHODORA, Vol. 96, No. 886, pp. 179-189, 1994

LIFE HISTORY OF SHOOTS OF
CAREX COMOSA F. BOOTT.

JOHN M. BERNARD AND FRANZ K. SEISCHAB

ABSTRACT

Carex comosa shoots have a life history which varies depending on their time
of emergence. Shoots that emerge in spring, summer or early autumn develop
flower initials before winter, flower the following year, then die in August, having
lived 240-420 days. Shoots that emerge in late autumn do not develop flower
imitials, remain vegetative the next summer, then develop flower initials in the
second autumn. These shoots flower and die the following summer, having lived
for up to 700 days. Few shoots live their maximum time and mortality 1s 90% in
some cohorts. Mortality patterns indicate that shoots die at a constant rate through-
out the year.

Key Words: Carex, life history, ramets, central New York

INTRODUCTION

Many temperate zone wetland ecosystems are dominated by
clonal Carex plants. Such plants (genets) are capable of producing
large numbers of shoots (ramets) through continuous reiteration
of modules. Each shoot is typically separated by a spacer organ,
usually a rhizome.

Different clonal sedge species have evolved different life history
adaptations. Bernard (1990), in his review of Carex, noted three
basic patterns of growth based on behavior of the spacer organ.
He pointed out that some species have spacers which are short,
resulting in tight clumps of shoots, while others have long rhi-
zomes and place shoots at a distance from the parent shoot. The
third and probably most common type is found in those shoots
that produce both long and short rhizomes, resulting in loose
clumps of shoots connected to other loose clumps through the
rhizomes.

There has been a relatively large number of studies on Carex
species exhibiting the latter two strategies, among them those by
Bernard (1975, 1976), Bernard and Gorham (1978), Bernard and
MacDonald (1974), Callaghan (1976, 1990), Costello (1936) and
Ratliff (1983) but few on those that produce only short spacers
and grow in clumps. Examples are the studies of David and Kelcey
(1985) on the C. muricata L. complex, Schmid (1984) on C. flava

179

180 Rhodora [Vol. 96

L., Taylor (1956) on C. flacca Schreber, and Bernard and Fiala
(1986) on C. comosa Boott.

Carex comosa is a large clump-forming species whose shoots
produce short rhizomes only. It is found scattered in disturbed
areas of wetlands. Bernard and Fiala (1986) found that the small-
est clones were made up only of vegetative shoots, but as the
clones grew larger the flowering percentage increased to a maxi-
mum of almost 70% in the largest 100-shoot clones. The shoot/
root ratio also increased as the clones expanded in size. From
their study, they labelled C. comosa a long-lived fugitive species
which invades gaps in wetlands, grows vegetatively for the first
years, then flowers heavily when larger. This adaptation enables
C. comosa to invade a gap, increase in ramet numbers and hold
the gap for a decade or more while producing large crops of seeds,
some of which may find new gaps to invade.

The purpose of this study was to determine life history and
demographic patterns in shoots of Carex comosa and to determine
if there are phenological differences in plants growing in different
sites.

METHODS

Data were collected from three different study areas, two in the
Cumming Nature Center, Naples, New York and one in Inlet
Valley, Ithaca, New York. The Inlet site was the same one used
by Bernard and Fiala (1986) and was dominated by Sparganium
eurycarpum Engelm. Botanical nomenclature follows Gleason and
Cronquist (1991). A small stream flowed through the site with a
row of willows (Salix sp. L.) bordering one side. One of the Nature
Center sites was just downstream from an active beaver (Castor
canadensis) dam. The water table was just at the surface of the
soil during the growing season; the vegetation was a mixture of
many species including Typha latifolia L., Carex lacustris Willd.
and Leersia oryzoides (L.) Swartz. There was a considerable num-
ber of trees at the site and tree branches were common on the
ground. The second Nature Center site was upstream from the
beaver dam along the pond shore. This site had less herbaceous
cover with a few scattered alders (4/nus incana L. Moench.). The
water table was somewhat lower and the organic soil more con-
solidated than at the site below the dam.

All shoots of Carex comosa in the 14 clones growing in the two

1994] Bernard and Seischab— Carex comosa Shoots 181

sites at Cumming Nature Center were tagged with a numbered
plastic tag in the autumn of 1983 or spring of 1984. The length
of all shoots/clone were measured then and at subsequent visits
at approximately monthly intervals during the growing season
until autumn 1985. All shoots tagged on each date became part
of a cohort. Notes on the health of shoots and whether they were
in vegetative or flowering condition were made at each visit. New
shoots were tagged as they emerged. In order to determine if
differences in mean number of shoots per plant above and below
the beaver dam at the Nature Center were significant, a sample
t-test was conducted.

In a separate part of the study, all shoots in three clones were
tagged on November 5, 1983 in Inlet Valley. There were 64 shoots
in the three clones varying in length from 3.5 cm to over 100 cm.
All shoots were placed in four cohorts, defined differently than
those above; cohort | was made up of the smallest shoots which
had just emerged above ground plus four additional shoots that
emerged after November 5. Cohorts 2 and 3 were shoots which
had emerged in summer and early autumn and grew to heights
of 10-20 cm by November. Cohort 4 was made up of shoots over
30 cm long; most had a cluster of dead leaves at the shoot bases.
These clones were visited four times between November 5 and
June 22, 1984. At each visit, other non-tagged clones were har-
vested. These were brought back to the laboratory, individual
shoots measured, then dissected to determine if the shoots had
flowering initials present. All shoots were dried and weighed after
dissection.

RESULTS

Table 1 presents data on the total number of shoots in each of
nine clones located at Site | at the Nature Center, downstream
from the beaver dam. Shoot numbers in five of the clones in-
creased from 80 to 400%, one clone remained the same, and three
declined over the 25-month period of study. The total number
of shoots in all clones went from 77 to 134 after one year, then
increased slightly to 139 after the second year, for a total gain of
80%. Shoot number during both summers increased about 35%
but varied dramatically during the two winter periods. From Sep-
tember 1983 to May 1984, shoot numbers increased 28%, but
from September 1984 to May 1985, they decreased by 22%. Thus,

182 Rhodora [Vol. 96

Table 1. Number of shoots/clone and total numbers in nine clones of Carex
comosa sampled over a two-year period. The site was located in a wetland just
downstream from a beaver dam (Site 1).

1983 1984 1985 9/83-10/85
Clone 9/19 21 9/17 5/24 10/17 % Change
1 5 10 13 17 25 400
2 ef 15 15 10 25 21
3 8 10 17 16 19 137
4 15 13 28 21 28 87
> 15 20 29 22 on 80
6 4 El 10 5 4 0
7 9 9 8 6 7 on
8 6 q 9 4 2 — 66
9 8 7 5 3 2 =75
Totals pe 98 134 104 139 81%

although the average shoot percentage gain was 80% over two
years, almost all the gain was during the first year; little change
occurred during the second year. It is important to note that four
of the clones were probably dying, two having two shoots re-
maining, and two having the same number or fewer shoots than
at the beginning of the study.

Table 2 presents data for five clones growing at Site 2 along the
edge of the pond upstream from the beaver dam. All clones 1n-
creased at least 220% in total shoot numbers from May 1984 to
October 1985, the average being 445%. Although mortality in
winter was 18%, almost the same as clones at Site 1, emergence
in summer was heavy; clones gained 152% from May to October
1985, a rate similar to the 164% growth of the previous summer.

Table 2. Number of shoots/clone and total numbers in five clones of Carex
comosa at Site 2, located along the edge of a beaver pond.

1984 1985 5/84-10/85
Clone 5/21 9/17 ag we 10/17 % Change
l 5 20 16 57 1040
2 5 ibe, 16 29 500
3 6 10 13 34 466
4 6 13 1] 20 233
5 ) 20 11 29 Zee
Totals 31 82 67 169 445%

1994] Bernard and Seischab— Carex comosa Shoots 183

Table 3. Cumulative gains and losses and percentage change of Carex comosa
shoots in nine clones at Site | and five clones at Site 2. Note: births and deaths
are cumulative for each period: for example, between 5/21/84 and 7/3/84, 34 new
shoots emerged and 12 died.

Site 1 Site 2

Total % Turn- Total %
Date _ Births Deaths Shoots over Births Deaths Shoots Change

9-19-83 77

5-21-84 36 15 98 20 a1

7-3-84 70 2) 120 18 ) l 39 25
7-30-84 86 52 111 =5 18 3 46 18
8-21-84 127 77 127 14 51 6 76 65
9-17-84 160 103 134 6 65 14 82 7
5-24-85 186 159 104 —22 80 44 67 —18
6-24-85 202 163 116 12 136 55 112 67
7-23-85 = 252 170 [59 37 159 66 124 10
8-26-85 286 pale 152 —4 229 88 172 38
10-17-85 339 v5 | 139 =f 308 170 169 oa

In comparison to Site 1 clones, the number of shoots in all Site
2 clones had at least doubled and one had increased over 10 times.

Shoot turnover for both sites is shown in Table 3 which gives
the cumulative gains and losses during the sampling period. At
Site 1, from September 1983 to October 1985, the total shoots
increased from 77 to 139 but a total of 339 new shoots had
emerged and 277 had died. At Site 2, total shoot numbers were
31 in May 1984 and increased to 169 in October 1985. Again,
significant shoot turnover occurred with 308 shoots emerging and
170 dying.

Shoot mortality during the two winter periods varied during
the study. During the first winter at Site 1, only 15 shoots died
but 36 new shoots emerged, resulting in a 20% increase. In con-
trast, during the second winter, plants had high mortality with
relatively low emergence resulting in a total loss of 22% of all
shoots in the nine clones.

Clone numbers at Site 2 (Table 3) showed a somewhat different
picture. Plant mortality over winter was high (54%), but emer-
gence during the following spring was somewhat higher, resulting
in a decline of 18% in total shoot numbers over winter.

Figure | shows the time of emergence for all cohorts of shoots
in nine clones making up the population at Site 1 on July 23,

184 Rhodora [Vol. 96

Shoot Numbers

9/19/83 5/21/84 7/3/84 7/30/84 ~=«-8/21/84. «9/17/84 = «5/24/85 =~: /2.4/85
Date of Cohort Emergence
. Age structure of 109 shoots in nine clones of Carex comosa on July
23, 1985, The total number of shoots oar tagged (shaded bars) 1 in each of

(open bars) shoots still alive.

1985. Bars represent the original number of shoots in each cohort
(shaded) and the number of vegetative (black) and flowering shoots
(open) still remaining at the end of the study. The oldest cohorts
(9/1/83, 5/21/84, 7/3/84, 7/30/84) had very high mortality; few

100 |
90 + |
B % ve
| 80 | is | |
| 70 | [1% Flow. | |
¢ eo}; ——— |
o |
B 507 |
a 40 |
30 + |
20 + |
10 +
‘2 oe oe a ee es
fe) a oO oO =- ae wv wv
5 N = & $ hi N N
[o>] w Lop} wo oO

Date of Cohort es

Figure 2 centage of vegetative (black bars) and flowering shoots (open
bars) of peri comosa remaining on July 23, 1985 from original number in the
cohorts.

1994] Bernard and Seischab— Carex comosa Shoots 185

Table 4. Number of shoots in four cohorts of overwintering clones of Carex
comosa at the Inlet Valley site.

1984 1985
Cohort 1/15 3/10 4/29 6/22
1 11 15 15 14*
2 25 26 26 26*
a 15 15 15 iS
- 13 13 12 6*
Totals 64 69 68 61

* Denotes shoots in flower.

shoots remained from the original number that had emerged.
Mortality was less in the second 1984 cohort, and most that lived,
flowered. None of the 1985 shoot cohorts flowered. Figure 2 gives
percentages of vegetative and flowering shoots in the eight cohorts
still living on July 23, 1985. The two oldest cohorts of autumn
1983 and May 1984 had over 90% shoot mortality; about half
the survivors flowered. The rest of the 1984 cohorts had higher
percentages living and most were flowering shoots. The youngest
shoots of the 1985 cohorts were all vegetative and although having
emerged only 1-2 months before July had considerable mortality,
the June cohort having a 7% mortality.

Recently emerged shoots were all vegetative, but older cohorts,
with the exception of the September 1983 cohort, had mostly
flowering shoots. At the conclusion of the study, two shoots two
years old were still in a vegetative condition, and 55% of shoots
in all clones had flowered. Mortality patterns described a type II
survivorship curve, a pattern to be expected of an opportunistic
early successional species (Deevey, 1947).

In a separate part of the study, three clones were tagged in late
autumn in the Inlet Valley site near Ithaca and followed during
the winter and early spring. There were 64 shoots in the three
clones tagged in November (Table 4); four additional shoots
emerged in cohort 1 by March 10 and one additional larger shoot
was tagged and placed in cohort 2. Of the total 69 shoots, 61
survived until June 22, with one shoot in cohort | and six of the
original 13 in cohort 4 dying. The shoots that survived until June
in all four cohorts averaged about 102 cm in length and all flow-
ered.

In addition to the shoots placed in the four cohorts, almost all
shoots in cohorts 2 and 3 produced one or more small axillary

186 Rhodora [Vol. 96

shoots, all less than 3 cm long in March; at harvest in June these
axillary shoots accounted for the largest percentage of vegetative
shoots in each clone since none flowered. Cohort | and cohort 4
shoots did not produce any axillary shoots over winter during
this study. We found that 37 of the 107 shoots harvested (35%)
produced axillary shoots after November 5, a percentage about
twice that of shoots in cohorts 2 and 3 of the tagged clones. Some
of the harvested shoots which were old enough to have produced
axillary shoots did not. These were all either smaller than average
size, had brown meristems, or were infected with Diptera larvae.

DISCUSSION

Maximum length of life for single shoots of various Carex
species 1s extremely variable. Some, such as Carex lacustris, live
12 months (Bernard, 1975); others such as C. rostrata may live
for about 18-24 months in the temperate zone (Bernard, 1976)
but longer, up to 60 months, in oligotrophic sites in northern
Sweden (Solander, 1983). Many arctic and alpine tundra species
also live for up to 60 months (Alexeev, 1988; Hultgren, 1988),
as do some montane species. A particularly good example of long-
lived shoots was found in C. nebraskensis Dewey, some of which
lived for 96 months in alpine habitats of the Sierra Nevada Moun-
tains in California (Ratliff, 1983).

Few Carex shoots live their maximum possible time and annual
mortality is usually at least 80% in temperate sites although lower
in arctic sites. Life spans are variable even in shoots in the same
cohort, ranging, for example, from only 12 days to over 750 days
in C. /asiocarpa Ehrh. (Bedford et al.. 1988). Bedford et al. (1988)
found, as did we, that shoots die at a constant rate, a situation
also shown in other studies of Carex (Bernard, 1990). Shoots that
emerge in late autumn or early winter have the longest possible
life span in comparison to spring or summer emerged shoots but
these differences are not as different as they at first appear. If one
subtracts the approximately 140 days of winter from the autumn-
emerged shoots the average life span becomes less variable. For
example, shoots emerging in late autumn live for 670-700 days
or so, the longest of all possible life spans, while spring shoots
live for perhaps 550 days. Subtracting the 140 days of winter from
the autumn shoots gives them approximately the same life span
as spring shoots. It is the same situation described by Bedford et

1994] Bernard and Seischab— Carex comosa Shoots 187

al. (1988), who noted that both autumn and spring emerged shoots
of C. lasiocarpa were part of the same cohort, the shoots not
emerging until spring were already formed but were dormant
below ground during the winter.

There is extremely heavy shoot turnover depending in part on
whether or not individual clumps grew well. This has also been
found in other Carex species. Bartlett and Noble (1985), Noble
et al. (1979), Bernard (1990) and Callaghan (1976) all noted that
at a time of favorable growth conditions, more shoots are pro-
duced than are needed for replacement; when the environment
worsens, they die back. Thus, the better the environment for
growth, the more emergence. When conditions become less suit-
able, greater than average mortality occurs. This mortality does
not happen in more northern sites; Solander (1982) found little
mortality in C. rostrata in her arctic study, owing probably to the
low nutrient status of her site.

Mortality patterns did not match well in the years of study.
During the first winter, clones showed low shoot mortality at both
Nature Center sites but during the second winter clones had very
high mortality, especially at Site 1. This may have been due to a
high water table at Site | during the second winter; the water
table at Site 2 did not vary to the same extent. The high second-
year mortality at Site 1 had a profound effect on individual clones,
some of which did not recover.

Flowering percentages in clumps varied for two reasons. First,
size of clumps was important. Bernard and Fiala (1986) found
that shoots in the smallest (and probably youngest) clones did not
flower while the largest clones had as many as 67% of their shoots
in flower. Second, we found in this study that during winter veg-
etative meristems of some shoots were either dead or affected by
Diptera larva. These conditions obviously affect expected flow-
ering percentages.

Bernard and Fiala (1986) noted that Carex comosa had the
characteristics of an opportunistic early successional species, in-
vading open sites, growing rapidly and then flowering. It also had
characteristics that allowed it to live for a long time in a site
favorable for growth. The shoot life history patterns found in this
study allow this species to both compete for favorable sites while
at the same time producing flowering shoots which will distribute
seed each year. Thus, while larger clones produce many flowering
shoots, there are always vegetative shoots and short rhizomes

188 Rhodora [Vol. 96

present to carry on the genet in autumn after flowering shoots

SUMMARY AND CONCLUSIONS

Carex comosa shoots may live for up to 700 days, those emerg-
ing in autumn usually being longer-lived than shoots emerging in
spring and summer. Shoots that emerge in summer and early
autumn will develop flower initials during autumn and flower the
next June. These flowering shoots also develop axillary shoots
during late autumn and early spring which grow and provide a
population of young vegetative shoots during the summer. These
emergence and growth patterns ensure that mature clones are
always a mixture of flowering and vegetative shoots, the former
for producing seed to be disseminated to new sites, the latter to
provide the clone with a young shoot population after death of
the flowering shoots in August. This is important because C.
comosa acts as a fugitive species, only occupying a site for a certain
time so seed production is essential to its establishment in new
disturbed sites (Bernard and Fiala, 1986). Other species which
grow by both long and short rhizomes may dominate an area;
they tend to flower much less and instead reproduce mostly by
vegetative means.

LITERATURE CITED

ALEXEEV, Y. E. 1988. ae in Carex species. Aquatic Bot. 29: 39-48.
BARTLETT, N. R. AND J. C. Nos 1985. The population biology of plants with
ne growth. III. petits ie mortality in Carex arenaria. J. Ecol. 73:
-10.

eae B. L., N. R. RAPPAPORT AND J. M. BERNARD. 1988. A life history of
Carex lasiocarpa Ehrh. ramets. Aquatic Bot. 29: 63-80.
BERNARD, J. M. 1975. The life history of shoots of Carex lacustris. Canad. J.
Bot. 53: 256-260.
1976. The life history and population dynamics of shoots of Carex
rostrata. J. Ecol. 64: 1045-1048.
1990. Life history and vegetative reproduction in Carex. Canad. J. Bot.
68: 1441-1448.
ND K. Fiara. 1986. The life history strategy reflected in standing crop
ee biomass allocation patterns of Carex comosa Boott.: a clump-forming
wetland sedge. Ekol. CSSR 5: 247-259,
AND E. GorHAM. 1978. Life history aspects of primary production in
wetland sedes pp. 39-51. In: R. E. Good, D. F. Whigham, and R. L

1994] Bernard and Seischab— Carex comosa Shoots 189

Simpson, Eds., Freshwater Wetlands: Production Processes and Management

Potential. Academic Press, New Yor

AND J. G. MACDONALD Jr. 1974. Primary production and life history
of Carex lacustris. Canad. J. Bot. 52: 117-123.

CALLAGHAN, T. 1976. Growth and population dynamics of Carex bigelowiti
in an alpine say ronment Oikos 27: 402-413.

1990. Aspects of clonality in the arctic: a comparison between Lyco-
podium annotinum and Carex bigelowii, pp. 131-151. Jn: J. Groenendael
and H. deKroon, Eds., Clonal Growth in Plants: Regulation and Function.

B Academic Publishing, The H

CosTELLo, D. F. 1936. Tussock oe in southeastern Wisconsin. Bot. Gaz.
97: 610-647.

Davin, R. W. AND J. G. Ketcey. 1985. Biological flora of the British Isles. Carex
muricata L. aggregate. J. Ecol. 73: 1021-1039.

Deevey, E. S., Jr. 1947. Life tables for natural populations. Quart. Rev. Biol.
22: 283-314.

GLEASON, H. A. AND A. CronquistT. 1991. Manual of Vascular Plants of North-

eastern United States and Adjacent Canada. 2nd ed. New York Botanical
Garden, New York.

Huttacren, A. B.C. 1988. A demographic study of aerial shoots of Carex rostrata
in relation to water a el. Aquatic Bot. 29: 81-93.

Nose, J. C., A. D. BELL AND J. L. Harper. 1979. The population biology of
plants with clonal aon I. The morphology and structural demography of
Carex arenaria. J. Ecol. 67: 983-1008.

RaT irr, R.D. 1983. Nebraska sedge (Carex nebraskensis Dewey): observations
on shoot life history and management. J. Range Managem. 36: 429-4
ScumMip, B. 1984. Life history in clonal plants of the Carex flava group. J. Ecol.

72: 93-114.

aa D. 1982. Production of Mg oe ae in two small, subarctic lakes
n northern Sweden. pp. 181-186. Jn: J. J. Symoens, S. S. Hooper and P
ene Eds., Studies in renee Vascular Plants. Proc. ea Col-
loquium on eumae Vascular Plants, oo Belgium
ass and shoot production of Carex rostrata and Equisetum
Huviaiile in unfertilized and fertilized eect lakes. Aquatic Bot. 15: 349-

366.
Tayior, F. J. 1956. Carex flacca. J. Ecol. 44: 281-290.

DEPARTMENT OF BIOLOGY
ITHACA COLLEGE
ITHACA, NY 14850

Bi Re:

DEPARTMENT OF BIOLOGY

ROCHESTER INSTITUTE OF TECHNOLOGY
ROCHESTER, NY 14623

RHODORA, Vol. 96, No. 886, pp. 190-194, 1994

A COMPUTER METHOD FOR PRODUCING
DOT DISTRIBUTION MAPS

RAY ANGELO

ABSTRACT

eu, enemnOd or progueiie dot Cast IOEGOn maps Sauer from a database
software and p main mapping software
is ea: on sample mans are provided.

Key words: atlas, distribution map, MicroCAM, New England, WordPerfect 6.0

While preparing the first installment of an atlas of the vascular
plants of New England, a method was developed using various
computer software programs to produce dot distribution maps
automatically from a database file. This note described the meth-
odology in general terms (but with references to specific software
programs) and provides two sample distribution maps produced
with this method. More technical details will appear in the first
installment of the New England atlas of vascular plants, which is
in preparation.

Three computer software programs were used. The mapping
software used to produce a customized map of New England with
county borders was MicroCAM (Version 3.1). This software, which
is for computers that use MS-DOS (most versions), was developed
by Dr. Scott A. Loomer (now a professor of geography at the U.S.
Military Academy at West Point). Since it is an adaptation of a
federally funded mainframe mapping program (CAM), it is in the
public domain. It may be obtained at nominal cost from:

Microcomputer Specialty Group of the Assoc.
of American Geographers

% Department of Geography

Indiana University of Pennsylvania

Indiana, PA 15705-1087

The MicroCAM software includes maps of countries, U.S. states,
county boundaries, Canadian provinces, coastlines, major rivers,
major lakes and major islands derived from World Data Bank II
and U.S. Census Bureau boundary files.

190

1994] Angelo— Distribution Maps 191

Equisetum arvense Equisetum pratense

Figure 1. Sample distribution maps produced using the method of this note.

The maps created with MicroCAM can be saved in files in a
variety of formats. One of these is Hewlett-Packard Language
(HPGL), a standard vector graphics format. The HPGL map files
can be imported into WordPerfect 6.0 image files which can then
be manipulated. Maps from other mapping programs that can be
imported into WordPerfect 6.0 can also be used in this method.

The database software used was Paradox (version 4.0 for DOS)
produced by Borland International, Inc. However, any database
software that can export its files into ASCII delimited text files
(or files of other formats that will be recognized by WordPerfect
6.0 as data files) can be used.

The word processing software WordPerfect 6.0 (product of
WordPerfect Corporation) is probably the most vital element of
the method. Its ability to manipulate graphics images inside
graphics boxes (rectangular areas embedded in a page of text) and
its flexibility in allowing the printing of superimposed graphics
boxes are critical to the viability of this method. Also, the avail-
ability of a scalable, sharp dot in the WordPerfect character sets
and a Hewlett-Packard laserjet printer are important. Dots (cir-
cular disks) printed by dot matrix printers do not have perfectly
circular margins. Similarly, dots produced by some graphics soft-

192 Rhodora [Vol. 96

ware (MicroCAM, for example) are actually many-sided polygons
and can be recognized as such.

The basic idea of the method is to perform a merge between a
database file containing distribution records and a form file con-
sisting of superimposed graphics boxes. One graphics box contains
the outline map (hereafter called a map box). Two other graphic
boxes, superimposed over the map box, place dots at specific
locations within the box (hereafter called a dot box), depending
on the contents of the database fields. A merge is the process that
businesses use to produce individualized letters from a customer
list (database file) and a form letter (form file).

For the atlas, each taxon is a record in the database file. Each
record consists of fields for each component of the scientific name
and for each county in New England. Additional fields are pro-
vided for the arbitrary subdivisions of some larger counties and
for Mt. Desert, Maine and Block Island, Rhode Island. If a her-
barium record exists for the taxon in a given county, a letter code
for the particular herbarium appears in the field for the county.
Otherwise, the given county field is empty. Most of the herbarium
records (perhaps 85% to 90%) in the database file are based upon
specimens in the New England Botanical Club herbarium.

A map of New England was created using MicroCAM and saved
in a file in HPGL format. Then a graphics box (the map box) was
designed in WordPerfect 6.0 with the New England map file named
as the contents. The image was rotated, shifted and scaled to fit
neatly in the graphics box using the image editing capabilities of
WordPerfect 6.0. This map box was then saved as a file.

A WordPerfect macro was then created for each New England
county. A macro is a sequence of keystrokes that can be executed
by pressing one (or a few) preassigned key(s). Each macro created
for this method places a dot of a given size at a specific position
inside a graphics box (dot box) which is the same size as the map
box. The position of the dot is chosen to correspond to the position
of the given county in the map box when the dot box is super-
imposed over the map box. The advantage of defining the macros
with respect to the margins of the dot box is that the box may be
placed at any position on a page (along with the associated map
box) without the need to create a new set of macros.

The form file consists of at least one set of three superimposed
graphics boxes. The lowermost is the map box. The next box
above (a dot box) contains a set of merge commands that execute

1994] Angelo— Distribution Maps 193

the macros for the counties in Maine, New Hampshire and Ver-
mont for each corresponding nonempty county field in the da-
tabase file. The uppermost box (a dot box) contains a set of merge
commands for the counties in Massachusetts, Rhode Island and
Connecticut and a set that produces a caption (the name of the
taxon). Two dot boxes were required to hold the merge commands
for the counties of New England since one graphics box (of the
size used) would not accept the entire set of merge commands.
Also, the contents of a graphics box in WordPerfect 6.0 must
either be a graphics image or text. There is limited ability to add
text to a graphics box containing a graphics image. Since the dots
are considered text, it was not possible to place the required merge
commands directly into the map box.

Figure | 1s an example of two distribution maps produced using
this method with actual data for Eguisetum pratense Ehrhart and
Equisetum arvense L. The map was created and imported into
WordPerfect 6.0 on a Dell 386 notebook computer. The merge
was executed on a Dell 386 desktop computer. The printer used
was a Hewlett-Packard Laserjet 4.

This author is unaware of any single, inexpensive software pack-
age that accomplishes the results of the method described here.
A sophisticated Geographic Information System (GIS) type of
software package, PC Arc/Info, made by Environmental Systems
Research Institute, no doubt can accomplish the results of this
method (except possibly the sharp dots), but its capabilities far
exceed what 1s necessary. Other software programs for IBM com-
patible computers that might be capable of achieving the results
of the method described here are Atlas GIS (Strategic Mapping
Inc.), GIS Plus (Caliper Corp.), and MapInfo (MapInfo Corp.).
These packages (including PC Arc/Info) are somewhat expensive.
Readers interested in geographic data systems for plants should
consult the electronic article, ““Design Criteria for a Plant Geo-
graphic Information System” by Jim Sullivan published in “Flora
Online” (Issue #18, Sept. 13, 1988, Text file 018GISIO.TXT,
53,805 bytes), an electronic publication of TAXACOM.

My coworker on the atlas project is David E. Boufford of the
Harvard University Herbaria, who I thank for suggestions that
improved the wording of this article, for aiding my access to a
computer and printer at the Harvard University Herbaria, and
for providing a copy of the electronic article referred to in the
preceding paragraph.

194 Rhodora

CURATOR OF VASCULAR PLANTS

NEW ENGLAND BOTANICAL CLUB

HARVARD UNIVERSITY HERBARIA
22 DIVINITY AVE.

CAMBRIDGE, MA 02138

[Vol. 96

RHODORA, Vol. 96, No. 886, pp. 195-203, 1994

ADDITIONS TO THE FLORA OF
NEWFOUNDLAND. JI.

STUART G. HAy, ANDRE BOUCHARD,
AND Luc BROUILLET

ABSTRACT

Five significant additions of arctic and subarctic species have been made to the
native vascular flora of the island of Newfoundland as a result of botanical ex-
plorations on the Great Northern Peninsula: Minuartia biflora (Wats.) Schinzl. &
Thell., Primula stricta Hornem., Sagina caespitosa (J. Vahl) Lange, Sagina sa-
ginoides (L.) Karst. and Salix argvrocarpa Anderss. Important range extensions
on the island are also reported for Cerastium terrae-novae Fern. & Wieg., Dan-
thonia intermedia Vasey and Salix arctica Pallas.

Key Words: vascular flora, arctic and subarctic additions, Newfoundland

Studies on the flora of the island of Newfoundland have taken
an important turn recently with the publication of a repertoire
with distribution maps of the “‘Rare Vascular Plants of the Island
of Newfoundland” (Bouchard et al., 1991), and a general “Atlas
of the Vascular Plants of Newfoundland and the Islands of Saint-
Pierre-et-Miquelon”’ (Rouleau and Lamoureux, 1992). Knowl-
edge of the flora of this insular region of Atlantic Canada continues
to progress, but as both of these publications point out, there is
still room for work to be done to gain a fuller understanding of
the region.

We have already made evident in earlier papers on new records
to the flora of Newfoundland (Hay et al., 1990, 1992) that many
parts of the island are relatively inaccessible, and have yet to be
thoroughly surveyed botanically. Further exploration of these ar-
eas is bound to result in exciting new discoveries, range exten-
sions, and to fill in gaps in the presently-known distribution pat-
terns of plants on the island.

In 1992, we pursued our floristic investigations of Newfound-
land, particularly of the rare plants, in two areas located at the
tip of the Great Northern Peninsula. This area is already known
for its important arctic-alpine flora, much of which is rare because
of its very restricted distribution on the island. During our field
investigations, five remarkable arctic or subarctic species were
collected for the first time in Newfoundland. This report includes
notes on their distributional range, their habitat, and the phyto-

Jee)

196 Rhodora [Vol. 96

geographical significance of the new records. All five species make
significant additions to both the list of rare plants of Newfound-
land (Bouchard et al., 1991) and the recent atlas of the flora of
Newfoundland (Rouleau and Lamoureux, 1992).

L’ Anse-aux-Meadows

Two species new to Newfoundland were discovered in L’ Anse-
aux-Meadows National Historic Park, on Sacred Bay, where we
conducted a survey of the rare vascular plants for Parks Canada
(Bouchard et al., 1993). The park 1s classified as a World Heritage
Site by UNESCO because of its cultural significance as the earliest
known Norse landing and settlement site in the New World.

Primula stricta Hornem.

SPECIMENS. Strait of Belle Isle Dist.: L’Anse-aux-Meadows
National Historic Park, between Upper Quarter Deck Cove and
Rudder Cove, 1992/08/08, Bouchard, Hay & Brouillet 92018
(CAN, COCO, MT).

The distribution of Primula stricta extends throughout the
southern islands of the Canadian Arctic Archipelago, into the
Hudson Bay region and along parts of coastal Ungava and north-
ern Labrador. It is an arctic amphi-Atlantic species that ranges
eastward through Greenland and Iceland to Scandinavia (Hultén,
1958; Kelso, 1991). Confusion with other allied species in section
Aleuritia, namely Primula anvilensis S. Kelso, P. borealis Duby
and P. incana M. E. Jones, have, until recently, obscured the true
distribution of this species (Kelso, 1991). The only reports of it
from Labrador are restricted to the northern coastal area of the
Peninsula (Morisset and Payette, 1987; Kelso, pers. comm.).

At the L’Anse-aux-Meadows site, plants of Primula stricta were
discovered in an open, partly herbaceous Empetrum health com-
munity that covered a flat, supralittoral, gravel terrace behind the
landwash along a seashore beach. This appears to be typical of
the habitat generally observed for the species throughout its range.
The plants were past anthesis when collected and were mistaken
for efarinose specimens (f. chlorophylla Fern.) of Primula lauren-
tiana Fern. These two species can be confused, especially where
their ranges overlap. No field information was recorded on the
size of the colony.

1994] Hay et al.—Newfoundland Flora 197

The L’Anse-aux-Meadows record of Primula stricta is an im-
portant extension southward of the species’ range along the At-
lantic coast, in comparison with localities in northern Labrador.
The origin of the present-day, amphi-Atlantic distribution pattern
is still unclear. P. stricta is a 14-ploid species that could have
derived from either European or North American progenitors.
The polyploid complex stems from the postglacial migration and
concatenation of ancestral populations following Pleistocene gla-
ciation (Kelso, 1991).

Salix argyrocarpa Anderss. (Syn.: S. ambigua Tuckerm., S. /a-
bradorica Schw.)

SPECIMENS. Strait of Belle Isle Dist.: L7 Anse-aux-Meadows
National Historic Park, along boardwalk between interpretive
center and Norse archaeological site, 1992/08/07, Bouchard, Hay
& Brouillet 92001 (CAN, GH, MT)

Endemic to northeastern North America, the range of Salix
argyrocarpa lies mainly in the northern subarctic regions of the
Québec-Labrador Peninsula, with only a few, more southerly lo-
calities extending from the eastern James Bay coast to the Céte-
Nord of the Gulf of St. Lawrence (Morisset and Payette, 1987;
Raup, 1943). Rare and disjunct occurrences appear southward in
alpine and subalpine situations in the Shickshock Mountains of
the Gaspé Peninsula in Québec and in the New England states of
Maine (Mt. Katahdin) and New Hampshire (Mt. Washington)
(Crow, 1982; Raup, 1943).

At L’Anse-aux-Meadows, a single colony of this willow was
discovered growing in an alder-willow thicket in a zone of heavy
snow accumulation. The scrub community was found on sloping,
soggy ground below a low, northeast-facing cliff in the bedrock.
Less than ten clones were counted in the population.

The Newfoundland population of Salix argyvrocarpa marks the
eastern limit of the species’ compass. Although the type of snowbed
habitat where it was found may be ecologically restrictive, it seems
surprising that this willow has never been recorded on the island.
With further search, we would expect it to be found elsewhere in
suitable alpine and subalpine sites in the Long Range Mountains
and at the tip of the Northern Peninsula.

198 Rhodora [Vol. 96

White Hills Mountains

Three species new to Newfoundland, all Caryophyllaceae, have
been discovered in the White Hills Mountains to the west of St.
Anthony. The White Hills form two adjoining plateau-like massifs
of ultramafic, serpentinized peridotite and dunite bedrock that
lie close to the north shore of Hare Bay. The eastern massif forms
a high, barren tableland (alt. 230-300 m) covering a vast area of
approximately 100 km?. The northern and eastern rims of the
plateau are flanked by steep talus slopes that fall away into a chain
of long lakes, two of the largest of which are Eastern Long Pond
and Western Long Pond. Snowbeds, some of which are late-lying,
are a conspicuous feature of these talus slopes.

The White Hills are of the same geological origin as the ser-
pentine complex in the Bay of Islands and Bonne Bay areas that
lie approximately 250 km further south on the west coast. The
latter tablelands are already well known for their unique serpen-
tine flora that includes some of the rare and more interesting
arctic-alpine and endemic elements of Newfoundland (Bouchard
et al., 1986, 1991; Dearden, 1979; Roberts, 1992). The White
Hills on the Great Northern Peninsula, however, have been less
well explored. Three species, Cerastium terrae-novae Fern. &
Wieg., Danthonia intermedia Vasey and Salix arctica Pallas, pre-
viously known only from the Bay of Islands and Bonne Bay ser-
pentines, were collected here for the first time, and make signif-
icant range extensions on the island.

The three Caryophyllaceae new to Newfoundland are:

Minuartia biflora (Wats.) Schinzl. & Thell. (Syn.: Arenaria sa-
janensis Willd. ex Schlecht., A. biflora Wats., not L.)

SPECIMENS. Strait of Belle Isle Distr.: White Hills, west of
St. Anthony, southwest of Western Long Pond, 1992/08/17, Bou-
chard, Hay & Brouillet 92386 (CAN, MT).

Minuartia biflora is a circumpolar, arctic-montane species. In
North America, in the western half of its range, it extends from
the lower islands of the Canadian Arctic Archipelago southward
through Alaska, Yukon and the Rocky Mountains of British Co-
lumbia to Colorado and Montana. In the eastern half of its range,
M. biflorais found from the high Arctic to the Hudson Bay region,
and is scattered through the arctic and subarctic region of the

1994] Hay et al.—Newfoundland Flora 199

Québec-Labrador Peninsula (Morisset and Payette, 1987; Hultén
and Fries, 1986). Only one southward disjunct locality has been
recorded from the serpentine summit of Mt. Albert on the Gaspé
Peninsula.

On the Eastern White Hills, Minuartia biflora was discovered
at a single site on the eastern rim of the tableland barrens. Only
a few plants were found on patches of exposed, moist gravel on
an east-facing slope above a small ravine with late-melting snow.
Surprisingly, we did not encounter /. marcescens (Fern.) House,
the rare, northeastern endemic species that - well known from
the Bonne Bay and Bay of Islands serpentin

The new record on the Great Northern cue for this arctic-
alpine species adds a second major, disjunct, southern outpost to
the one already known on Mt. Albert in the Shickshock Mountains
of the Gaspé. The historical reasons why rare vascular plants and
bryophytes with arctic-alpine or Cordilleran affinities have be-
come isolated from their main distribution areas have long been
the subject of opposing theories by phytogeographers and ecol-
ogists. The question as to whether these plants survived the last
glaciation as relicts in in situ, ice-free refugia (nunataks) around
the Gulf of St. Lawrence, or whether they are post-glacial im-
migrants from southern or coastal refugia following ice retreat,
that subsequently persisted in these unique habitats, is still open
to debate (Belland et al., 1992; Bouchard et al., 1986). Their
survival in narrow-niche habitats, such as these serpentine bar-
rens, can be attributed to the extreme ecological conditions that
tend to reduce competition, favoring the persistance of rare and
endemic species at their range limits. Long-distance dispersal is
also a possible mechanism that may explain the disjunct distri-
bution pattern of some species.

Sagina caespitosa (J. Vahl) Lange (Syn.: Sagina nivalis (Lindbl.)
ries var. caespitosa (J. Vahl) Boivin)

SPECIMENS. Strait of Belle Isle Dist.: St. Anthony, 4-5 mi.
west of town, Eastern White Hills, 1951/06/29, Savile & Vail-
lancourt 1971 (mt); 1951/08/08, Savile & Vaillancourt (MT).

Sagina caespitosa has a typical arctic amphi-Atlantic distri-
bution pattern. It occurs as a coastal and montane species in the
eastern Arctic of North America, radiating eastward across the
Atlantic through Greenland, Iceland and Scandinavia (Crow, 1978;

200 Rhodora [Vol. 96

Hultén, 1958: Scoggan, 1978-79). In the arctic region of the Qué-
bec-Labrador Peninsula, populations are widespread but rare
(Bouchard et al., 1983; Morisset and Payette, 1987; Rousseau,
1974),

The discovery of this remarkable new Sagina for Newfound-
land results from our revision of older herbarium sheets of col-
lections made in 1951 from the Eastern White Hills tableland,
erroneously identified as Minuartia rubella (Wahlenb.) Hiern.
These Savile and Vaillancourt collections have only recently been
made available for consultation and have escaped revision until
now. The label information on the herbarium specimens gives
only a vague description of the habitat as serpentine gravel barrens
at 200 m alt. At the time of our foray on the tableland, we were
as yet unaware of the existence of these older specimens, and we
did not discover it ourselves during our own collecting.

The new record from the tip of the Northern Peninsula rep-
resents a major disjunct outlier to the south. Crow (1978) has
discussed the present-day distribution patterns of the different
species of Sagina in context of Pleistocene glaciation events. He
has argued that eastern Arctic populations of S. caespitosa may
have survived in coastal mountain refugia and cites as evidence
their distinct affinity with probable nunatak populations of this
species in western Greenland. He is, nonetheless, careful to point
out that long distance, airborne dispersal may have played an
important role, allowing species of Sagina, because of their dust-
like diaspores, to become established 1n suitable habitats follow-
ing glaciation.

Sagina saginoides (L.) Karst. (Syn.: Sagina linnaei Presl)

SPECIMENS. Strait of Belle Isle Distr.: White Hills, west of
St. Anthony, southwest of Western Long Pond, 1992/08/17, Bou-
chard, Hay & Brouillet 92349 (CAN, MT).

Sagina saginoides has a circumpolar distribution, correlating
almost entirely with montane regions of the Northern Hemi-
sphere. In North America, it is almost exclusively a western alpine
species, ranging through the Cordillera from Alaska south to Ar-
izona and New Mexico (Crow, 1978; Hultén and Fries, 1986). It
is of rare occurrence in eastern North America, where only seven
disjunct locations have been previously recorded from the arctic

1994] Hay et al.—Newfoundland Flora 201

and subarctic region of the Québec-Labrador Peninsula and on
the Gaspé Peninsula serpentines of Mt. Albert (Bouchard et al.,
1983; Crow, 1978; Morisset and Payette, 1987; Scoggan, 1978-
79).

On the Eastern White Hills serpentine plateau, Sagina sagi-
noides was discovered along the eastern rim of the tableland where
several of the talus slopes are snow-covered until late in the sea-
son. Snowbed ecosystems have a particularly high incidence of
rare plants in Newfoundland (Bouchard et al., 1991). Their special
ecological conditions account for the presence of this new species
and other rare, characteristic elements of arctic-alpine affinity
such as Cassiope hypnoides (L.) D. Don, Epilobium anagallidi-
folium Lam., E. lactiflorumHausskn., Gnaphalium norvegicum
Gunn., Phieum alpinum L., Salix herbacea L., Veronica worm-
skjoldii Roem. & Schultes, and Viola palustris L. The populations
of Cassiope hypnoides, Salix herbacea and Viola palustris are new
records for the area, being formerly known only from more south-
ern alpine summits in the Long Range Mountains, between the
Bay of Islands and the Highlands of Saint John (Bouchard et al.,
1991),

Sagina saginoides was one of the arctic-alpine species that Fer-
nald (1925) cited as evidence of a disjunct relictual flora that may
have survived glaciation in ice-free, nunatak refugia, such as the
summit of Mt. Albert on the Gaspé. More recently, however,
Crow (1978) has reexamined the phytogeography of this species,
and he feels that, in light of today’s evidence, the rare and widely
isolated eastern populations in Québec and Labrador are better
regarded as resulting from long distance dispersal and/or of a
pleistocene fragmentation of what was once a more wide-ranging
arctic-circumpolar distribution.

ACKNOWLEDGMENTS

Funding for the study of the rare plants of L’ Anse-aux-Meadows
National Historic Park came from Parks Canada, Ottawa. G.
Argus (Canadian Museum of Nature, Ottawa) kindly identified
specimens of Salix. G. Crow (University of New Hampshire,
Durham) confirmed our identifications of Sagina and Minuartia.
Primula specimens were revised by S. Kelso (Colorado College,
Colorado Springs).

pAOy Rhodora [Vol. 96

LITERATURE CITED

BELLAND, R. J., W. B. SCHOFIELD AND T. A. HEDDERSON. 1992. Bryophytes of
the Mingan yee eae National Park Ree Québec: a boreal flora with
arctic and alpine components. Can. 70: 2207-2222.

BOUCHARD, A., D. BARABE, M. Dun MAIS AND < . Hay. 1983. The rare vascular
plants of Québec/Les plantes vasculaires rares du Québec. Syllogeus 48.

S. G. Hay, C. GAUVIN AND Y. BERGERON. 1986. Rare vascular plants
of Gros Morne National Park, Newfoundland (Canada). Rhodora 88: 481-
502.

—, L. BRouILtet, M. JEAN AND I. SAuciER. 1991. The Rare Vascular
Plants of the Island of Newfoundland/Les plantes vasculaires rares de I’ile
de ap Neuve. Syllogeus

———., L. BROUILLET AND S. G. HAy. 1993. The rare vascular plants of L’Anse-
aux-Meadows National Historic Park. Parks Service, Environment Canada,
Ottawa. unpubl. report.

Crow, G. E. 1978. A taxonomic revision of Sagina (Caryophyllaceae) in North
America. Rhodora 80: 1-91.

. 1982. New England’s rare, threatened, and endangered aes U.S. Dept.
of the Interior, Fish and Wildlife Service, Northeast Regi

— P. Re Some factors influencing the composition one location of

on a serpentine bedrock in western Newfoundland. Jour-
nal biogeoeaph 6: 93-104.

FERNALD, M. 1919. The American representatives of Arenaria sajanensis.
Rhodora a 12-17.

———. 1925. The persistence of plants in unglaciated areas of boreal America.
Mem. Amer. Acad. Arts Sci. 15: 239-342,

Hay, S. G., A. BouCHARD AND L. BROUILLET. 1990, Additions to the flora of
the island of Newfoundland. Rhodora 92: 277-293.

AND M. JEAN. 1992. Additions to the flora of Newfound-

baa. I. cee 94: 383-386.

Hu ten, E. 1958. The amphi-atlantic plants and their phytogeographical con-
nections. Almqvist & Wiksell, Stockholm.

. Fries. 1986. Atlas of North European i dees Plants, north of
the Tropic of Cancer. Koeltz Scientific Books, Fed. Rep. Germany. 3 vols.

Ketso, S. 1991. Taxonomy of Primula sects. Aleuritia a Armerina in North
America. Rhodora 93: 67-99

Morisset, P. AND S. PAyettTe, Eds. 1987. Flore du Québec nordique et des
Territoires adjacents. Centre d’études nordiques et Herbier Louis-Marie,
Univ. Laval, Québec. 3 vols.

Raup, H. M. 1943. The willows of the Hudson Bay region and the Labrador
Peninsula. Sargentia 4: 81-135.

Roserts, B. - 1992. The serpentinized areas of Newfoundland, Canada. Jn:
A. J. Baker, J. Proctor and R. D. Reeves, Eds., The Maca of Ultra-
mafic (Serpentine) Soils. Intercept Scientific Publications, Fra

ROULEAU, E. AND G. LAMOUREUXx. 1992. Atlas of the Vascular "Plants of the
Island of Newicunciand and of the Islands of Saint-Pierre-et-Miquelon/Atlas
des Plantes Vasculaires de I’Ile de Terre-Neuve et des [les de Saint-Pierre-
et-Miquelon. Fleurbec Editeur, Québec.

1994] Hay et al.— Newfoundland Flora 203

Rousseau, C. 1974. Géographie Floristique du Québec-Labrador. Presses de
I’Université Laval, Ste-Foy.

Scoccan, H. J. 1978-79. The Flora of Canada. 4 parts. Natl. Mus. Canada,
Natl. Sci. Publ. Bot. no. 7. Ottawa.

INSTITUT DE RECHERCHE EN BIOLOGIE VEGETALE
JARDIN BOTANIQUE DE MONTREAL

4101 EST, RUE SHERBROOKE

MONTREAL, QC, CANADA H1X 2B2

INFORMATION FOR CONTRIBUTORS TO RHODORA

Submission of a manuscript implies it is not being considered
for publication simultaneously elsewhere, either in whole or in
part.

Manuscripts should be submitted in triplicate (an original and
two xerographic copies) and must be double-spaced (at least *”)
throughout including tables, figure legends, and literature cita-
tions. The list of legends for figures and maps should be provided
on a separate page. Footnotes should be used sparingly. Do not
indicate the style of type through the use of capitals or under-
scoring, particularly in the citation of specimens. Names of genera
and species may be underlined to indicate italics in discussions.
Specimen citations should be selected critically, especially for
common species of broad distribution. Systematic revisions and
similar papers should be prepared in the format of ““A Monograph
of the Genus Malvastrum,” S. R. Hill, Rhodora 84:1-83, 159-
264, 317-409, 1982, particularly with reference to indentation of
keys and synonyms. Designation of a new taxon should carry a
Latin diagnosis (rather than a full Latin description), which sets
forth succinctly just how the new taxon 1s distinguished from its
congeners. Papers of a floristic nature should follow, as far as
possible, the format of “Annotated List of the Ferns and Fern
Allies of Arkansas,” W. Carl Taylor and Delzie Demaree, Rho-
dora 81: 503-548, 1979. For bibliographic citations, refer to the
Botanico-Periodicum-Huntianum (B-P-H, 1968) which provides
standardized abbreviations for journals originating before 1966.
All abbreviations in the text should be followed by a period, except
those for standard units of measure and direction (compass points).
For standard abbreviations and for guidance in other matters of
biological writing style, consult the CBE Style Manual, 5th ed.
(original title: Style Manual for Biological Journals). In preparing
figures (maps, charts, drawings, photos, etc.) please remember
that the printed plate will be 4 x 6 inches; be sure that illustrations
are proportioned to reduce correctly, and indicate by blue pencil
the intended limits of the figures. (Some “‘turn-page”’ figures with
brief legends will be 32 x 6 in.) Magnification/reduction values
given in text or figure legends should be calculated to reflect the
actual printed size. An Abstract and a list of Key Words should
be supplied at the beginning of each paper submitted, except for
a very short article or note. All pages should be numbered in the
upper right-hand corner. Brevity is urged for all submissions.

205

THE NEW ENGLAND BOTANICAL CLUB
22 Divinity Avenue
Cambridge, MA 02138

The New England Botanical Club is a non-profit organization
that promotes the study of plants of North America, especially
the flora of New England and adjacent areas. The Club holds
regular meetings, has a large herbarium of New England plants,
and a library. It publishes a quarterly journal, RHODORA, which
is now in its 95th year and contains about 400 pages a volume.

Membership is open to all persons interested in systematics
and field botany. Annual dues are $35.00, including a subscription
to RHODORA. Members living within about 200 miles of Boston
receive notices of the Club meetings.

To join, please fill out this membership application and send
with enclosed dues to the above address.
Regular Member $35.00
Family Rate $45.00
For this calendar year —————
For the next calendar year ————

Name

Address

City & State Zip

Special interests (optional):

Vol. 96, No. 885, including pages 1-120, was issued September 7, 1994.

206

THE NEW ENGLAND BOTAN ICAL CLUB

vg - eoso Officers ad Poinck Members te 1994 1995

Presiden ¢ Barre Helanict, Box 9145, Dee ee of Biolo-
a North Adams State ee: North Adams, Massachusetts: :
01247 — ‘|

04578

* - setts 02154.

setts 01247 . .
Revontine Secretary: ‘Lisa A. Sidley - Be
_ Curator of Vascular Plants: Raymond Riedie eB
Assistant Curator of Vascular Plants: Pamela Weatherbee.
| Curator of Non-Vaseule Plants: eM Reid. ne |
ast Librarian: Paul Somers at.

at

On Sos FRE Oe LOT of Rhodora and 4
ee oe _ Councillors : Leslie J. Mehthoff (Past President)
ee _ William E. Brumback’ 952
Thomas Mione *96_ a
- GarettE, Crow 97 _ ee
ee : Donald dans (Graduate Student fc
oe ae .

Vee Pee on “Pros gram Chair): : L Hones Hutson: Go. | is
- Chewonki_ ‘Foundation, RR2, Box. 1200, Barareh ae

is Core Reon. ‘ ney M. -Eyster-Smith, Department Lael
~~ of Natural Sciences, Bentley ee ae Massachu- :

CY Da ‘Harold G. Ec wan Box 9092, Depee eal of “Biol ak : a
: ogy, North Adams State College, a Adams, ae i"

ones - Consisting of the. Blecied Officers, Associate Curator 1s | a

THE NEW ENGLAND BOTANICAL CLUB

_P. O. Box 1897, Lawrence, Kansas 66044 _
ie innit Avenue, Cambridge, M: husetts 02138
me No. 887 .

| VOL. 96

"July, 1994

es JOURNAL OF" THE : ae: |
NEW ENGLAND ) BOTANICAL CLUB i:
4
CONTENTS:
ea : : A comparison of the marine cigiear from the Goleta Slough nn adjacent :
to open coast. of Goleta/Santa_ Barbara, California with those PRG so
southern gulf of Maine. Arthur C. Mathieson and Edward J. Hehre... 207
“ ooey poets biology: and Population BEneHCS of eee
4 vulgan rene
A new tri- ‘hybrid lycopod, Diphasiastram ee gitatum x sabinifolium : ps
ae Arthur V, Gilma PAA Ree eG
ee Membership Fos Bee a ct 294 .
ae Peat and C il Memt ; inside back cover

: 2 Divinity Avenue b ambridge Mas ac u ets
-RHODORA _ .

“GORDON P, DeWOLF, JR. Editor-in Chiet oa Oe

a - Associate Editors arg
-DAVIDS. CONANT :
LISA A. STANDLEY .

-RHODORA (ssn 0035- ale Published ae times a year oer
oe July, and October) by The New England Botanical Club, 810.
- East 10th St., Lawrence, KS 66044 and printed by Allen Press, Inc.,
1041 New Hampshir e St., Lawrence, KS 66044. Second class postage
4 ee at Lawrence, ee POSTMAS1 ER: Send address aie fo
_ RHODORA, P.O. Box 1897, I KS 66044, of
“RHODORA | isa journal of botany Gone primarily t to North America. - |
_ Authors are encouraged to submit manuscripts: of Scientific paperssa 2
_ and notes relating to the systematics, s, dis , ecology, paleo-
- botany, or conservation biology: of this or floristical y related areas.
SUBSCRIPTIONS: $40.00 per. calendar year, net, postpaid, in funds
payable at par in the United States currency. Remittances payable to
RHODORA. Send to ) RHODORA, P.O: Box 1897, eens: KS”
ee 66044- 8897. .
_ MEMBERSHIPS: Regular $35: Family S45. Applic a f n printed. ae

o
a

ee ties AND SINGLE COPIES: Some available; informa. :
tion and prices will be furnished upon reques st to the Treasu B
ADDRESS. CHANGES: In order to receive the” next | ae of
- RHODORA, changes must be received by o business office pie to.

~ the first day of January, April, July, or Octobe eas
INFORMATION FOR CONTRIBUTORS: “Tosi back cover, r Tanaay mae
and April. . ; ate
"MANUSCRIPTS: “Send to: ‘Goren Rh DeWo If ars

-Managi ng Editor, RHODORA
: tat 125 Long Hill Road
a ree Broomnens, mts OTSSS:

.

_ @ This paper meets the requi ts of ANSI/NISO Z39.48- -192 (Permanence of Paper).

oS

RHODORA

OURNAL OF THE
NEW ENGLAND BOTANICAL CLUB

Vol. 96 July 1994 No. 887

RHODORA, Vol. 96, No. 887, pp. 207-258, 1994

A COMPARISON OF THE MARINE ALGAE FROM THE
GOLETA SLOUGH AND ADJACENT OPEN COAST OF
GOLETA/SANTA BARBARA, CALIFORNIA WITH
THOSE IN THE SOUTHERN GULF OF MAINE!

ARTHUR C. MATHIESON AND EDWARD J. HEHRE

ABSTRACT

The seaweed se from the Goleta Slough and adjacent coastal sites near Santa
Barbara, Californ d with several coastal and estuarine habitats with-
in the southern Gulf of Maas. One ne fifty-two taxa are described from the
Goleta-Santa Barbara area, consisting of 104 Rhodophyta, 22 Phaeophyta and 26
Chlorophyta. Four of these taxa represent modest range extensions within the
state (1.e., Farlowia conferta, Giffordia hincksiae var. californica, Lomentaria ca-
seae and Prionitis australis, while two others found within the Slough (Capsosiphon
fulvescens and Microspora pachyderma) are significant additions to the marine
flora of the Pacific Coast. The patterns of species richness and composition at four
contiguous southern California sites showed strong contrasts due to local envi-
ronmental variability, with the highest numbers of taxa (117 taxa or ~77%)
occurring at the sand-abraded nearshore Goleta Point site and the lowest within
the shallow Slough (26 taxa or ~17%). The depauperate flora of this small arid
salt marsh habitat is dominated by ephemeral green algae (16 species) and Sali-
cornia virginica, with ten other seaweeds (i.e., 7 reds and 3 browns) only occurring
Just upstream from its mouth. By contrast, estuarine seaweed floras within New
England usually have much higher numbers of red algal taxa, their “open coastal”
floras often extend much farther inland (~2.0-8.5 miles) and seaweed standing
stocks are usually dominated by fucoid brown algae. The similarity of the Slough’s
green algal flora to New England’s estuarine vegetation is striking, with most of
the latter sites exhibiting floristic affinities of 50% or more. The analogous dis-
tributional and abundance patterns of ephemeral green algae (Cladophora sericea
and Enteromorpha compressa) and the California horn snail (Cerithidea califor-

' Scientific Contribution from the New Hampshire Agricultural Experiment
Station; also issued as contribution Number 287 from the Jackson Estuarine
Laboratory and the Center for Marine Biology.

20d

208 Rhodora [Vol. 96

nica) suggest several important interactions between seaweeds and snails, plus
several other invertebrates and birds.

Key Words: seaweeds, estuarine, coastal, Goleta Slough, southern California, New
Englan

INTRODUCTION

Even though the marine flora of California has a long history
of investigations (Papenfuss, 1976), there is a lack of information
about seaweed communities at individual sites and man’s effect
on them (Murray, 1974; Murray and Littler, 1989). In southern
California, Dawson (1959a, 1959b, 1965) conducted some of the
earliest investigations of intertidal seaweeds during 1956-59, pro-
viding a ““semi-quantitative baseline” of floras at 44 stations be-
tween Point Conception and the U.S.-Mexican border. He con-
cluded that a major reduction of species richness had occurred
at some sewage-impacted sites within Los Angeles County since
the early 1900’s, and an increase of selected articulated corallines.
After resurveying 15 of Dawson’s southern California sites be-
tween Point Dume and Dana Point during 1968-70, Widdowson
(1971) found a further reduction in species richness. He suggested
that sewage discharge, along with foot traffic and air pollution,
were important factors causing an abundance of gelidioid turf
algae (Gelidium coulteri/G. pusillum) and articulated corallines
(Thom and Widdowson, 1978). Subsequent to the Santa Barbara
oil spill in 1969, several investigators (e.g., Foster et al., 1971;
Nicholson and Cimberg, 1971; Straughn, 1971) documented a
further loss of some species and an increase of others (e.g., U/va,
Corallina and Gelidium) which were capable of uptaking dis-
solved amino acids (Murray, 1974; North et al., 1970). In dis-
cussing short-term plant loss immediately after the spill, Foster
et al. (loc. cit.) noted a strong interaction between oil dispersal,
storms and sand fluctuations. Three and one-half years later, Cim-
berg et al. (1973) re-investigated several intertidal beaches near
Santa Barbara and stated that periodic inundations by sand were
more important than oil in altering rocky epibenthic communities
(Emerson and Zedler, 1978; Littler, 1980; Littler et al., 1991;
Nicholson, 1972; Nicholson and Cimberg, 1971; Stewart, 1983;
Thom and Widdowson, 1978; Widdowson, 1971). In ranking the
effects of the Santa Barbara oil spill, Nicholson (loc. cit.) com-
mented that 1s was only one of several abuses to these coastal

1994] Mathieson and Hehre—Goleta Slough 209

waters, many of which dated back to the 1920’s. In summarizing
the effects of chronic long-term pollution near a natural oil seep
in Goleta (i.e., Coal Oil or Devereaux Point), Littler et al. (1991)
stated that several seaweeds either failed to recruit or recolonized
very slowly, while Foster et al. (1971) and Nicholson and Cimberg
(1971) noted diminutive and/or depauperate floras. In comparing
the ecology of a sewage polluted site on San Clemente Island,
Murray and Littler (1974) found pronounced changes in species
richness and reduced spatial heterogeneity versus nearby controls.

Until recently, estuaries have been the least studied coastal
environments in California (Josselyn, 1983; Macdonald, 1977:
Macdonald and Barbour, 1974; Zedler, 1982b; Zedler and Nord-
by, 1986). In addition, most studies of estuarine seaweeds have
emphasized productivity and ecological evaluations (Rudnicki,
1986; Shellem and Josselyn, 1982; Zedler, 1977, 1980, 1982a,
1982b), while there have been relatively few detailed assessments
of species composition (Josselyn and West, 1985: Norris, 1970:
Ripley, 1969; Silva, 1979). Several factors have contributed to
the limited knowledge of estuarine seaweeds: (1) the more local-
ized occurrence of estuarine versus rocky and sandy habitats
(Chapman, 1977; Littler et al., 1991; Macdonald, 1977; Onuf,
1987; Reimold, 1977); (2) the muddy and often eutrophied con-
ditions within estuaries versus the clean sand and pounding surf
of the open coast (Silva, 1979); (3) the apparent dominance of
flowering plants (Mason, 1957); (4) the perceived occurrence of
a limited and/or cryptic algal flora and; (5) the massive destruction
and alterations of such habitats, particularly within southern Cal-
ifornia (Macdonald, 1977; Zedler, 1982b). In describing the loss
of coastal wetlands in southern California, Zedler (loc. cit.) states
that there are about thirty wetlands within this geography with a
total area of ~5000 ha (12,500 acres). She also emphasizes that
this acreage 1s ~ 25% of what existed upon the arrival of European
man (Speth 1969). In discussing the functional role of estuarine
seaweeds, Mann (1972) states that they provide valuable habitats
for a myriad of organisms; they also contribute to organic carbon
and detrital cycling (Josselyn and Mathieson, 1978, 1980). With
increased eutrophication, massive algal blooms may occur within
these sheltered habitats (Bach and Josselyn, 1978: Johnson, 1971:
McComb et al., 1981; Norton and Mathieson, 1983; Sawyer,
1965; Sewell, 1982; Silva, 1979: Wilce et al., 1982; Wilkinson,
1980).

210 Rhodora [Vol. 96

T T T | T T T qT T
/ GAVIOTA SANTA BARBARA PT
lL 30° / GOLETA PT./BAY & SLOUGH , 7

Bek ;
Po / CARPENTERIA

PT ance yh PT. LAS PITAS
DEVEREAUX PT \ /
, - / [eee ae tI
20 Pree TEER: VENTURA
y 4 MILES

10’ aot PORT HUENEME 4
POINT ARGUELLO TO POINT DUME

° ° , -
34 120°10' 119° 10 PT. OUME
l 1 l it ! i i mls = |

Figure 1. The southern California coastline between Point Arguello and Point
Dume, showing the locations of the Goleta Point, Bay and Slough sites, plus
Naples Reef.

T T
PT. ARGUELLO

The present study was initiated to document the open coastal
and estuarine seaweeds near Goleta and Santa Barbara, California
(Figure 1). We have attempted to summarize the species com-
position and distributional patterns of marine macrophytes with-
in this arid, Mediterranean geography (Macdonald, 1977; Onuf,
1987; Onuf and Zedler, 1988; Zedler, 1982b; Zedler and Nordby,
1986), comparing them with thirteen New England coastal-es-
tuarine habitats having diverse topographies and hydrographic
conditions.

METHODS AND MATERIALS

Seasonal collections and observations of seaweed populations
were made during 1979 at four coastal-estuarine habitats between
Goleta and Santa Barbara, California (Figure | and Appendix).
Estuarine collections were made within the Goleta Slough (Figure
2), while open coastal samples were obtained at Goleta Point and
within Goleta Bay near the University of California, Santa Bar-
bara Campus (UCSB), as well as at Naples Reef, offshore from
Santa Barbara Point. A synopsis of each site’s location, general
topography and substrata is given in the Appendix. Overall, the
most detailed spatial sampling was conducted within the shallow
Goleta Slough, while the most comprehensive seasonal collections
were made at Goleta Point. Twenty-seven Slough sites were eval-
uated during the spring and summer, while monthly collections
(9 of 12 months) were made along an intertidal-shallow subtidal

1994] Mathieson and Hehre—Goleta Slough 211

SAN PEDRO CREEK

SANTA BARBARA =

GOLETA

UNIVERSITY OF CALIFORNIA
AT SANTA BARBARA

GOLETA PT GOLETA BAY

Figure 2. Twenty-seven study sites within the Goleta Slough and the adjacent
open coast.

transect at Goleta Point. Periodic subtidal samples (— 10 to —30
feet) were made near an experimental farm site in Goleta Bay
(Wheeler et al., 1981), while analogous subtidal collections were
made at Naples Reef (— 15 to —40 feet). All conspicuous seaweeds
were collected within the intertidal (on foot) and subtidal zones
(by SCUBA) at each site when both habitats were present. The
samples were sorted and identified using Abbott and Hollenberg
(1976), Scagel et al. (1986) and Stewart (1991) as primary refer-
ences. Several other synopses were also consulted, including some
dealing with North Atlantic seaweeds (e.g., Bliding, 1963, 1968;
Collins, 1909, 1912; Dawson, 1946; Scagel, 1966; Silva, 1979;
Smith, 1969; South and Tittley, 1986; Sparling, 1971, 1977; Tay-
lor, 1957; Webber and Wilce, 1971; Wilkinson, 1980). Approx-
imately 550 herbarium voucher specimens are deposited in the
Albion R. Hodgdon Herbarium (NHA) of the University of New
Hampshire to document the region’s flora.

A comparison of the number of species in common to the
Goleta Slough and thirteen New England estuarine habitats is
given, using a variety of published synopses (see Mathieson et al.,
1993). The nature of these California and New Aaa ea is
compared using Cheney’s (1977) floristic ratio, (R + C

paw: Rhodora [Vol. 96

# of Rhodophyta and Chlorophyta species
# of Phaeophyta species

With this calculation, a value of < 3.0 indicates a temperate or
“cold-water” flora, while intermediate values and those > 6.0
indicate “mixed” and “warm-water” floras, respectively. The mean
number of taxa/site from the Goleta Slough and thirteen New
England estuarine habitats was also compared.

Monthly records of surface water temperatures and salinities
were recorded at Goleta Point during 1979, while hydrographic
surveys of the Slough were made during the spring and summer
of the same year. A hand-held thermometer and refractometer
were used for these measurements, having accuracies of 0.1°C and
0.5%o (1.e., parts per thousand), respectively. Monthly records of
sand levels within the intertidal zone at Goleta Point were doc-
umented by drawing out a metered line to a uniform distance
between two rock outcrops with permanent bench marks (cement
nails). Sand levels were determined by plumbing down to the
sand/rock interface at ~1.0 m intervals along the transect line
(Daly and Mathieson, 1977). All elevations were corrected to
mean low water (MLW) after multiple measurements from these
bench marks. Monthly comparisons of species richness and sand
levels were made.

Abundance patterns of several green algae and the California
horn snail, Cerithidea californica (Haldeman), were documented
from three shallow pans within the Slough. A quadrat frame (1/
16 m’) was positioned at ~0.9 m intervals and the contents sam-
pled. Individual seaweeds were dried (60°C for ~40 hours) and
quantified as g dry weight/m?’; snail density was enumerated as
#/m?.

HABITAT DESCRIPTION

As outlined by Ricketts et al. (1985), most of the southern
California coastline is exposed to moderate wave action due to
an offshore archipelago—the Channel Islands. Even so, a stron
gradient of water motion may exist (Wheeler et al., 1981), ranging
from Naples Reef (most exposed) to Goleta Point, Bay and Slough
(most sheltered). A pronounced gradient of rocky substrata occurs
here, with Naples Reef having the most abundant rocky substrata
and Goleta Slough the least. The broad sandy beaches at the

1994] Mathieson and Hehre—Goleta Slough 213
S 207
af
=
rs}
+
E
-_
ae
o
Lu
<=
= ‘
. a . ’
or 1.0 T T T T r T — 88:2 T r 1
0 5 10 15 20 25 eo

HORIZONTAL DISTANCE (m)

Figure 3. Variation of sand levels at Goleta Point between two major rock
outcrops (i.e., at O and 10-1 ).

Goleta Point and Bay sites have scattered rock outcrops and
boulders (i.e., Monterey shale/siltstone), which are often buried
and extensively scoured (Littler et al., 1991). Figure 3 illustrates
the variation of sand levels on the Goleta Point transect. Although
rather irregular, sand elevations were usually highest during sum-
mer (July) and lowest in winter (February). J” situ organisms
growing in such habitats may be removed, scoured or buried by
a meter or more of sand during mid-summer.

Goleta Point (Figure 1) is located downstream from a natural
oil seep at Coal Oil (Devereaux) Point (Littler et al., 1991). In
discussing the effects of the seep, Foster et al. (1971) noted that
30-60% of the rock surfaces at Goleta Point were covered by tar,
causing a conspicuous reduction in abundance and stature of ben-
thic organisms. Although we did not quantify the distribution and
abundance of this “‘tar’’ on the intertidal transect, we concur with
Foster et al. (loc. cit.) that it must, at the very least, reduce the
availability of rocky substrata for benthic organisms

Goleta Slough 1s a small, shallow, arid salt marsh habitat located
at ~34°25’N and 119°50'W (Figure 2). As noted by Onuf and
Zedler (1988), the average annual precipitation within this locale
is < 40 cm (Baldwin, 1974) and the only substantial streamflow
occurs just after a major storm. Because of such climatic extremes
and the Slough’s shallow topography, evaporation usually exceeds
precipitation after March and monthly saturation deficits often

214 Rhodora [Vol. 96

exceed 10 cm. Tidal limits of the Slough extend ~1.0-1.2 miles
inland from its mouth near Goleta State Beach Park. Like the
Tijuana Estuary in San Diego, it is a wetland-dominated habitat,
with no major embayment and a relatively narrow ocean con-
nection (Zedler and Nordby, 1986). The Slough is located midway
between Santa Barbara’s Municipal Airport and the UCSB cam-
pus. Thus, it is an urbanized and highly modified estuary (Norris,
1970; Zedler and Nordby, 1986), with its northeastern perimeter
drastically altered due to airport expansion, while the construction
of overpasses (i.e., for gas pipelines and roads) and bike paths has
changed its topography. In addition, a large part of the Slough is
in the form of channels, which are artificially deepened for flood
control and separated from the adjacent salt marsh by dikes (Onuf,
1987). Currently, the west-central part of the Slough is dominated
by three major tidal channels, which bifurcate ~0.52 miles up-
stream from its mouth near the Clarence Ward Memorial Boule-
vard overpass. Further upstream (~0.94 miles inland), the source
of these tidal channels is evident (1.e., Glen Annie Creek and an
unnamed channe! running parallel to Aero Camino Road). In the
east-central portion of the Slough, three creeks converge between
~0.42-0.51 miles inland, with two originating from S and SW
flowing flood channels (i.e., San Pedro and San Jose Creeks, re-
spectively) and one arising from a tributary due east near Atas-
cadero. Twenty-four study sites were established along the west
central tidal channels, two near the Slough’s mouth and one at
the tidal dam near Atascadero (Figure 2 and the Appendix).
Most of the Goleta Slough consists of salt marshes, shallow
pans and a series of channels (Appendix). Scattered boulders and
artificial structures (i.e., bridge abutments, pilings, broken pieces
of asphalt, etc.) are most abundant near the Slough’s mouth and
occur sporadically upstream. Sand periodically blocks the mouth,
restricting tidal exchange and contributing to enhanced summer
salinities (Figure 4; Macdonald, 1977: Onuf, 1987; Onuf and
Zedler, 1988; Zedler, 1982b). As outlined by Onuf and Zedler
(loc. cit.), Salicornia virginica L. is the dominant salt marsh plant
within this tidally restricted habitat (94% frequency of occur-
rence), with only patchy amounts (1-2%) of Cotula coronopifolia
L., Frankenia grandifolia Chamiso et Schlechtendal and Lolium
perenne L., Zedler (1982b) lists eleven additional flowering plants
from the Slough: Atriplex watsonii A. Nelson, Cressa truxillensis,
H. B. K., Cuscuta salina Engelmann, Distichlis spicata (L.) Greene,

1994] Mathieson and Hehre—Goleta Slough 215

—a—_ March

35 7 °
=~ ~~~ - June /\
=
Ww
cc
=)
a
<x
cc
Wu
ao
=
WW
_

10 T T T T qT 1

80 4
eo
<
a 40 5
=
a
G 207

0 are T T . r {} T 4

0.0 0.2 0.4 0.6 0.8 1.0 1.2
DISTANCE INLAND a
Figure 4. Surfac e., temperature and salinity)

at Caen Point ai ana Slough during March and June of 1979.

Jaumea carnosa (Lessing) Gray, Lasthenia glabrata Lindley, Li-
monium californicum Heller, Monanthochloé littoralis Engel-
mann, Salicornia subterminalis Parish, Suaeda californica Wat-
son and Triglochin concinnum Davy. Although we primarily
studied the macroalgal communities within the Slough, a variety
of Vaucheria and microscopic algal mats (i.e., blue-green algae
and diatoms) also occur on muddy surfaces (Onuf, 1987; Zedler,
1980, 1982a, 1982b). As outlined by Onuf (loc. cit.), the overall
coverage of seaweeds is relatively small versus that of emergent
vascular plants. Seaweeds, therefore, grow attached to widely scat-
tered solid substrata, as entangled masses on the banks of tidal
channels and as epiphytes upon salt marsh vegetation. The dom-
inant macro-invertebrate within the Slough is Cerithidea califor-

216 Rhodora [Vol. 96

nica, the California horn snail (Onuf, 1987; Zedler, 1982b; Zedler
and Nordby, 1986).

Figure 4 illustrates surface water temperatures and salinities at
Goleta Point and within the Slough during March and June of
1979. Temperatures tended to increase from outer (1.e., site 1) to
inner estuarine sites (e.g., 21 and 22), ranging from ~ 14.0°-24.0°C
during March to ~17.0°-35°C in June. Salinities at Goleta Point
were relatively constant at 30—3 1 %o, while those within the Slough
varied from 0 to ~27 %o in March to ~12-68%0 in June. During
March, twelve Slough sites had salinities of < 2%o. Three months
later (June) the two lowest salinities were 10 and 13%, while
seven sites had salinities > 28%. A maximum of 68%o was re-
corded in one shallow pan with abundant green algae and horn
snails (see ABUNDANCE PATTERNS).

SPECIES COMPOSITION AND NEW DISTRIBUTIONAL RECORDS

One hundred fifty-two seaweed taxa were recorded from the
four study sites, including 103 Rhodophyceae, 22 Phaeophyceae
and 26 Chlorophyceae (Table 1). The highest numbers of taxa or
percent contribution to species richness were recorded at Goleta
Point (117 taxa or ~77%) and the lowest within the Slough (26
taxa or ~17%). Most taxa from Goleta Point occurred on the
shallow transect (i.e., 94 taxa or ~62%), while the remainder were
collected immediately offshore from the UCSB campus (1.e., 23
taxa or ~ 15%). Floristic patterns at Naples Reef and at the off-
shore Goleta Bay site were somewhat intermediate, with 52 taxa
(~34%) occurring at the former and 39 (~ 26%) at the latter.

The numbers (%) of red, brown and green algal taxa/site were
highly variable (Table 1). Green algae were maximal at Goleta
Point (19) and within the Slough (16), while the lowest numbers
occurred at the offshore Goleta Bay (2) and Naples Reef (1). Brown
and red algae were highest at Goleta Point (17 and 81 taxa) and
lowest within the Slough (3 and 7 taxa). Green algae dominated
the Slough (~ 62%), were intermediate at Goleta Point (16%) and
lowest at the offshore Naples Reef (2%) and Goleta Bay sites (5%).
The percentage contribution of brown algae was minimal within
the Slough with its depauperate flora (~ 12%), while it ranged
from ~15-33% at the Goleta Point and Bay sites. Red algae
dominated at Naples Reef (~85%), were intermediate at the Go-

1994] Mathieson and Hehre—Goleta Slough 217

leta Point and Bay sites (~ 62-69%) and lowest within the Slough
(~27%).

Floristic diversity is further documented using Cheney’s (R +
C)/P ratio. The highest value was recorded for the Slough (~7.7),
followed by the transect (~6.8) and total floras at Goleta Point
(~5.9), Naples Reef (~7.7), and Goleta Bay (2.0). Thus, most of
the floras would be interpreted as “‘warm-water,” except for the
offshore Goleta Bay site.

Six of the 152 taxa recorded here (Table 1) represent new geo-
graphical records. Two were collected from Goleta Slough (Cap-
sosiphon fulvescens and Microspora pachyderma) and four from
the open coast (Farlowia conferta, Giffordia hincksiae var. cali-
fornica, Lomentaria caseae and Prionitis australis). Capsosiphon
Julvescens was previously known from northern British Columbia
and Vancouver Harbor (Garbary et al., 1982), while the only
previous “marine” records for the freshwater green algae .
pachyderma (Collins, 1909: Prescott, 1962) were from several
western North Atlantic estuaries (Mathieson and Hehre, 1986;
Mathieson and Penniman, 1986a, 1991; Mathieson et al., 1993).
The records of G. hincksiae var. californica, F. conferta and P.
australis are modest extensions beyond San Luis Obispo County
(i.e., Point Conception), while the occurrence of Lomentaria ca-
seae 18 a northern expansion of its known range from Del Mar
within San Diego County to Isla Guadalupe, Baja California (Ab-
bott and Hollenberg, 1976; Scagel et al., 1986; Stewart, 1991).

DISTRIBUTIONAL PATTERNS AND FLORISTIC COMPARISONS

Seaweeds at the four study sites exhibited three local distri-

butional patterns (Figure 5 and Table 1)

(1) Coastal—restricted to the nearshore open coast (126 taxa or
~ 83%).

(2) Cosmopolitan—present in estuarine and open coastal envi-
ronments (20 taxa or ~ 13.0%).

(3) Estuarine—restricted to Goleta Slough (6 taxa or ~4.0%).

All six estuarine seaweeds were green algae, including Capsosi-
phon fulvescens, Cladophora microcladioides, Enteromorpha in-
testinalis, Microspora pachyderma, Rhizoclonium riparium and
Ulvaria oxysperma. In contrast, coastal taxa consisted of 97 Rho-
dophyceae, 19 Phaeophyceae and 10 Chlorophyceae, while the

Table 1. Seaweed taxa from three open coastal sites and one estuarine habitat near Goleta, California.

la lb 2 3
Chlorophyta
Acrochaete viridis (Reinke) R. Nielsen X X
Blidingia minima (Nageli ex Kiitzing) Kylin % 3 x
Bryopsis corticulans Setchell x x
Bryopsis hypnoides Lamouroux Xx X
Capsosiphon fulvescens (C. eat a et Gardner x
Chaetomorpha aerea (Dillwyn) Kit X X x
Cladophora columbiana Collins x X
Cladophora graminea Collins X Xx
Cladophora microcladioides Collins X
Cladophora sericea (Hudson) Kiitzing x X X
Codium sp. (basal filaments) Xx Xx
Codium fragile (Suringar) Hariot
Enteromorpha clathrata (Roth) Greville Xx x x
Enteromorpha compressa (L.) Greville x X X
Enteromorpha flexuosa (Roth) J. Agardh ssp. flexuosa X x X
Enteromorpha intestinalis (L.) Lin X
Enteromorpha linza (L.) J. Agardh Xx Xx Xx X
Enteromorpha prolifera (Miller) J. Agardh xX x X
Microspora pachyderma (Wille) Lagerheim x
Rhizoclonium implexum (Dillwyn) Kiitzing x X
Rhizoclonium riparium (Roth) Harvey x
Ulva californica Wille Xx x
Ulva costata (Howe) Hollenberg x
Ulva lactuca L. x X Xx Xx

BIC

eIOPpOYyY

96 [OA]

Table 1. Continued.

la lb 2 a
Ulva taeniata (Setchell) Setchell et Gardner x x x
Ulvaria oxysperma (Kiitzing) Bliding X
Total green algal taxa/site 18 19 16
Total green algal taxa (26) la Ib 3
Phaeophyta
Colpomenia sinuosa (Roth) a et Solier x X
Cylindrocarpus rugosus Oka x x
Cystoseira osmundacea CTumer) ec Agardh x Xx
Desmarestia ligulata (Lightfoot) Lamouroux var.
jirma (C. Agardh) J. Agardh
Desmarestia ligulata (Lightfoot) Lamouroux va. /igulata me X
esmarestia viridis (Miller) Lamouroux X
Dictyota binghamiae J. Agardh
Ectocarpus parvus RR Hollenberg Xx Xx x x
Egregia menziesii (Turner) Areschoug x x X
Endarachne a J. Agar Xx x 4 X
Giffordia hincksiae (Harvey) Hamel
r. californica Hollenberg et Abbott X
Giffordia sandriana (Zanardini) Hamel X X X
Haplogloia andersonii (Farlow) Levring X
Hincksia granulosa (J. E. Smith) Hamel Xx X x
Laminaria farlowii Setchell X Xx
Macrocystis pyrifera (L.) C. Agardh X X
Pterygophora californica Ruprecht x

ysnojs eIaJOH—a1YdH«~pue uosaryepy [p661

617

Table 1. Continued.

Scytosiphon dotvi Wynne
Scytosiphon lomentaria (Lyngbye) J. Agardh
Sphacelaria didichotoma Saunders
Taonia lennebackerae J. Agardh
Zonaria farlowii Setchell et Gardner

Total brown algal taxa/site

Total brown algal taxa (22)

a ince
orium venulosum (Zanardini) Kylin

dine fastigiata (Postels et Ruprecht) Makienko
Antithamnion defectum Kylin
Asterocolax gardneri (Setchell) Feldmann et Feldmann
Audouinella daviesii ees n) Woelkerling

angia vermicularis Harv
Bossiella orbigniana Deeicid ee

ssp. dichotoma (Manza) Johans
Bossiella orbigniana (Decaisne) in ssp. orbigniana

Botryocladia pseudodichotoma aa. Kylin
C ce haha cheilosporioides M

Callithamnion biseriatum Kylin

Callophyllis firma (Kylin) R. Norris

~~ PM PP OK

ms

— pC Od ot Ot Oe

~~

sm oO

eed

OCT

eIOPOYY

96 TOA]

Table 1. Continued.

Callophyllis flabellulata Harvey

Callophyllis pinnata ee et Swezy
Callophyllis violacea J. Agardh

Centroceras clavulatum Cc. jen Montagne
Ceramium californicum J. Agardh

Ceramium eatonianum (Farlow) DeToni
Ceramium sinicola Setchell et Gardner
Ceramium zacae Setchell et Gardner
Chondria nidifica Harve

y
Corallina officinalis L. var. on (Decaisne) Kiitzing

Corallina vancouveriensis Yen

Cryptopleura ps nea aaa
Cryptopleura cr

Cryptopleura I (J. Agardh) Kylin
Cryptopleura violacea (J. Agardh) Kylin

Cumagloia andersonii (Farlow) Setchell et Gardner
Erythrotrichia carnea (Dillwyn) J. Agard

Farlowia conferta (Setchell) Abbott

Fryeella gardneri (Setchell) Kylin

Gastroclonium subarticulum (Turner) Kiitzing
Gelidiocolax microsphaerica Gardner

Gelidium coulteri Harve

Gelidium purpurascens Gardne

Gelidium pusillum (Stackhouse) Le Jolis

Gelidium robustum (Gardner) Hollenberg et Abbott

mm OO

~~ ee OK

i i i i i i ir a!

~~ PK OOO

ysno[sg B1IVJOH—aiyoH{~pue uosarpiepy [661

CC

I

Table 1. Continued.

Gigartina canaliculata Harv
Gigartina exasperata ee et Bai
Gigartina harveyana (Kiitzing) Sethe et Gardner
Gigartina leptorhynchos J. Aga
Gigartina ornithorhynchos J. he if
Gigartina volans (C. Agardh) J. Agardh
Gracilaria papenfussii Abbott
Gracilaria textorii (Suringar) J. Agardh
var. cunninghamii (Farlow) Dawson
Gracilariophila oryzoides Setchell et Wilson

Halymenia californica Smith et ams:

Halymenia hollenbergii Abbot

Herposiphonia plumula (J. | an Hollenberg

Herposiphonia secunda (C. Agardh) Ambron
forma tenella (C. Agardh) Wynne

Herpo nada verticillata 7 Kylin

Heterosiphonia japonica

Holmesia californica ae Dawson

Janczewksia pees oe

Laurencia pacifica

Laurencia sere Posts et Ruprecht

Lomentaria caseae Dawso

~ em OM

Se i i!

a

~~ OK

~ PO

~~ em eK

> od td

GG

vIOPOYY

96 ‘IOA]

Table 1. Continued.

Mastocarpus papillatus (C. Agardh) Kitzing
Melobesia marginata Setchell et Foslie
Melobesia mediocris (Foslie) Setchell et Mason
Mticrocladia coulteri Harve
Nemalion helminthoides (Velley) Batters
Nienbergia andersoniana (J. Agardh) Kylin
Nitophyllum northii Hollenberg et Abbott
Opuntiella californica (Farlow) Kylin
Phycodrys isabelliae R. Norris et Wynne
Platythamnion ee _ Agardh) J. Agardh
Platythamnion Acie
Platythamnion villos
Pogonophorella californica 7 Agardh) Silva
Polyneura latissima (Harvey) Kylin
Polysiphonia hendryi Gardner
Polysiphonia pacifica Hollenberg
Polysiphonia scopulorum Harve

var. villum (J. Agardh) Hollenberg
Porphyra perforata J. Agardh
Prionitis angusta (Harvey) Okamura
Prionitis australis (J. Agardh) J. eg
Prionitis lanceolata (Harvey) H
Pterochondria woodii (Harvey) Hollenber erg
Pterocladia capillacea (Gmelin) Bornet et Thuret
Pterosiphonia baileyi (Harvey) Falkenberg

~ OK

~~ PP OOK

em em OM

~~ Ke em Ke KK

Yysnojs eIaJOH—oaIY9H{~pue uosaryleypy [F661

TCC

Table 1. Continued.

la lb ) 3 4
Pterosiphonia dendroidea Gaerne Falkenberg x X x X
Rhodoglossum affine (Harvey) Kylin 4 x
Rhodoglossum ae oh Agardh) Abbott X X
Rhodoglossum roseum (Kylin) G. M. Smith x
Rhodoptilum plumosum (Harvey et Bailey) Kylin X
Rhodymenia arborescens Dawson Xx
Rhodymenia californica Kylin var. attenuata (Dawson) Dawson x
Rhodymenia californica Kylin var. californica x X
Rhodymenia pacifica Kylin x x x x
Sarcodiotheca furcata (Setchell et Gardner) Kylin x x
Sarcodiotheca gaudichaudii (Montagne) Gabrielson X x Xx X
Scinaia articulata Setchell X Xx X
Scinaia confusa (Setchell) Huisman x Xx
Schizmenia pacifica Kylin X Xx x
Smithora naiadum (Anderson) Hollenberg X X
Stenogramme interrupta (C. Agardh) Montagne Xx 4
Stylonema alsidii (Zanardini) Drew X Xx x
?Tenarea dispar (Foslie) Adey Xx
Total red algal taxa/site 64 81 24 7 44
Total red algal taxa (104)
Total taxa/site 94 117 39 26 52

Total taxa (152)

#1a = Goleta Point (transect, intertidal & shallow subtidal); #1b = Goleta Point (total flora, intertidal & subtidal); #2 = Goleta Bay (offshore-
subtidal); #3 = Goleta Slough (intertidal & subtidal); #4 = Naples Reef (offshore-subtidal).

VCC

eIOPOY Y

96 [OA]

1994] Mathieson and Hehre—Goleta Slough 225

cosmopolitan component contained 7 Rhodophyceae, 3 Phaeo-
phyceae and 10 Chlorophyceae. Approximately 31% of the coastal
taxa occurred at two or more sites (39 taxa), while the remaining
87 were restricted to one of the following sites: (1) Goleta Point
transect—9 Chlorophyceae, 4 Phaeophyceae and 45 Rhodophy-
ceae; (2) Offshore Goleta Point—1 Chlorophyceae and 4 Rho-
dophyceae; (3) Goleta Bay—3 Phaeophyceae and 4 Rhodophy-
ceae; (4) Naples Reef— 1 Chlorophyceae, 2 Phaeophyceae and 17
Rhodophyceae. One cosmopolitan taxa occurred at all four sites
(1.e., Ectocarpus parvus), while seven were found at three sites
(i.e., Enteromorpha linza, Ulva lactuca, Endarachne binghamiae,
Chondria nidifica, Gigartina canaliculata, Polysiphonia hendryi
and Rhodoglossum affine). See Table 1 for specific details.

A comparison of the floras from Goleta Slough and the adjacent
open coast with two analogous New England habitats (1.e., the
York River and Great Bay Estuarine areas) shows varying patterns
of local distribution (Figure 5). For example, 131 taxa occur with-
in the York River Estuary and adjacent open coast of southern
Maine, consisting of 59 coastal, 45 cosmopolitan and 27 estuarine
(Mathieson, et al., 1993). The flora within the Great Bay Estuary
and adjacent open coast is more diverse (i.e., 194 taxa), consisting
of 21 coastal, 149 cosmopolitan and 24 estuarine taxa (Mathieson
and Hehre, 1986). Thus, the estuarine reduction pattern is most
conspicuous within Goleta Slough (1.e., 126 coastal, 20 cosmo-
politan and 6 estuarine), intermediate within the York River Es-
tuary, and least pronounced within the Great Bay Estuary. Flo-
ristic patterns within the Goleta Slough and York River are similar,
except for differences in the numbers of total taxa and brown and
red algae. The prevalence of cosmopolitan taxa within the Great
Bay area (77%) should be emphasized, as it represents a very
different pattern than that found within either the York River
(34%) or Goleta Slough (13%).

The number of seaweed taxa/site within Goleta Slough is sum-
marized in Figure 6 and Table 2. The highest number of taxa was
recorded at site 1 (22), while reduced and irregular numbers oc-
curred upstream (i.e., 1-13). Green algae dominated the Slough,
with Enteromorpha clathrata and Rhizoclonium riparium exhib-
iting occurrences of 85.2 and 81.5%, respectively. All seven red
algae and two of the three browns were restricted to site | (1.e.,
3.7% occurrence), which is just 0.09 miles inland from the mouth.
Only one brown alga (Ectocarpus parvus) extended upstream be-

226

NO. OF SEAWEED TAXA NO. OF SEAWEED TAXA

NO. OF SEAWEED TAXA

Rhodora [Vol. 96

GOLETA SLOUGH AREA (CALIFORNIA)

lls, ——
nai)

WMG,
ee

LA

N

YORK RIVER ESTUARINE AREA (MAINE)

| SEReaaESeRASE

REE

ETRE AIRC REE)
CZZZIZIZIIZIZA)
a Tey

TOTAL COASTAL COSMOPOLITAN ESTUARINE

1994] Mathieson and Hehre—Goleta Slough 227

d

BEES

— RSSSSASSSSSSSSSSSSS

SSN

SOA

Se)

AAS

SY

AAAAAAAAAAAA
Rakin

NO. OF SEAWEED TAXA
rs
GEES ANANAAAAAARAARAARARAARRS

SSVV—sVvTwwweo

GB ANAAAAAANARAARARRARRRE
2B ANAMARARARAARRRS

DISTANCE INLAND (MILES)

Figure 6. Number of seaweed taxa/site (green, brown and red) at twenty-seven
ene Slough stations.

yond site 1, occurring also at sites 5 and 6 (i.e., 11.1% occurrence).
See Table 2 for further details.

The paucity of seaweed taxa within Goleta Slough is further
documented by comparing its species composition and mean
number of taxa/site (Figure 7 and Table 2) with thirteen New
England estuaries. Only 26 seaweeds were recorded from the
Slough, with a mean of just 5.1 + 4.3 taxa/site. Inner riverine
habitats like the Winnicut River (4 taxa and 1.3 + 1.6) have less
diverse floras, while outer and mid-estuarine sites Piet have
greater number of species: e.g., Oyster (49 taxa and 12.6 + 7.9),
York (72 taxa and 21.4 + 3.2) and Piscataqua Rivers (143 and
25.3 + 24.9). Figure 8 compares the green algal taxa within the
Slough (16 taxa) and thirteen New England estuaries. The number
of taxa ranged from 24-37 within outer and mid-estuarine New
England sites versus 12-20 for inner locations. The % of Slough
greens within the same New England sites varied from 25-81%,
with most habitats, other than the Winnicut and Merrimack Riv-
ers, having floristic affinities of 50% or more. None of the browns

_

Figure 5. A comparison of species richness and local distribution of seaweed
taxa from the Goleta Slough and adjacent open coast (California) with the York
River (Maine) and the Great Bay Estuarine areas (Maine-New Hampshire).

Table 2. Summary of seaweed taxa at twenty-seven sites within the Goleta Slough.

BCC

1 2 3 4 5 6 7 8 10 11
Chlorophyta

Blidingia minima X x x x Xx X x
Capsosiphon ak eSCens x x x
Chaetomorpha X

Cladophora i ere Xx
Cladophora sericea x x X x X
Enteromorpha clathrata x Xx Xx X X X X
Enteromorpha compressa x X x x
E bale las flexuosa

. flexuosa Xx x x x Xx

Patines intestinalis X X X
Enteromorpha linza X
Enteromorpha prolifera X Xx Xx X
Microspora pachyderma X X X X
Rhizoclonium riparium X Xx X x Xx x x
Ulva lactuca x x x
Ulva taeniata X
Ulvaria oxysperma x Xx x x x X

# Green algal taxa/site 12 10 8 7 12 8 2 3

Mean # (+ SD) green algal taxa/site (4.7 + 3.1

Mean % occurrence (+ SD) green algal taxa (28.9% + 24.6%)

Total # green a taxa (16)

Phaeophyta

Ectocarpus parvus x x x

vIOpoyy

96 ‘1OA]

Table 2. Extended.

12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 %*
x K x X 40.7
x X 18.5
3.7
3.7
x x x 29.6
X x x Xx Xx X X x X Xx x x x x 85.2
x 18.5
x x Xx x X x 40.7
x 14.8
3.7
x x x x x x 40.7
x X Xx 33.3
x x x x x x X x 4 Xx x x x x 81.5
11.1
3.7
x Xx Xx 33.3
3 4 3 4 3 3 4 4 4 1 5 2 3 2 3 9

ysnojs klajJOH —a1yoH pue uosaipiepy [r661

6CC

Table 2. Continued

1 Z 3 4 5

Endarachne binghamiae Xx
Taonia lennebackerae a
# Brown algal taxa/site 3 0 0 0 ]
Mean # (+ SD) brown algal taxa/site (0.19 + 0.61)
Mean % occurrence (+ SD) brown algal taxa (6.2% + 3.5%)
Total # brown algal taxa (3)
Rhodophyta
Chondria nidifica
Gelidium coulteri
Gigartina canaliculata
Gigartina leptorh: is
Polysiphonia hendry
Pterosiphonia aes
Stylonema alsidii
# Red algal taxa/site 7 0 0 0 0)
Mean # (+ SD) red algal taxa/site (0.26 + 1.32)
Mean % occurrence (+ SD) red algal taxa (3.7% + 0%)
Total # red algal taxa (7)

i a a)

Total seaweed taxa/site 10 8 7 13
Mean # (+ SD) total seaweed taxa/site ee | + 4.4)
Total seaweed taxa (26)

* The % values represent calculations based upon 27 Goleta Slough stations.

OE

RIOPOYY

96 ‘IOA]

Table 2. Extended. Continued.

19

20

21

22

ae)
3g.

ysnojs e19J0H —a149H{s pure uosaryiepy [p661

lec

yao) Rhodora [Vol. 96

#GREENS
(1) #BROWNS
Mi #REDS

fe SSS

LW

NO. OF SEAWEED TAXA
a

Q 64
ip) 4
a
<q
~<
<
j=
ie
Oo
Oo
=
=
<< f
4 f
ec 2 % ¢€& &€& &£ &€ ¢c &¢§ & ¥ e& FE
es «8 B= 8 SS & rE 5§ 2 ¥* 2
ina =) 5 a s i Wu Wu = = fod oO °
o Go 4 < < r tC Fe hUmE!lCO o mo « @G
> ¢« F-F uw A gg & & FF 0 FB = =
KE = x = Pa = ~ 77) = Ww = a
< 5 O Ww o < ° z = = o Fb
) a 4 o < = 3 § w
Yn s —_
= = @& o = O
z Oo He o
” =
<q
x=

Figure 7. A comparison of the Goleta Slough flora and thirteen other New
England estuarine habitats, expressed as the total number of green, brown and
red algae/site as well as the mean (+ SD) number of total taxa/site.

and only one red alga (Sty/onema alsidii) occurred in both the
Slough and New England estuaries.

PHENOLOGICAL PATTERNS AT GOLETA POINT

Table 3 summarizes monthly occurrences of seaweed taxa on
the Goleta Point transect. No conspicuous pattern was evident,
except that the minimum number of taxa occurred in April (27)
and the highest in February and November (45). Green algae

1994] Mathieson and Hehre—Goleta Slough 233

Mi Total #
100 > % Slough Green

iN
SSO Oooo

IN
XMAS

iN

be
SSA SA MS_-_'/S-s—_—'5|

SSS SSS

WN

SS

GREEN ALGAL TAXA (# & %)

0 Gu Ga G4 G4 Ga Gu Ga G Gi A
a: ee ec cc ec et e& ff a na 5
<x 8] ao > oO > [om 7p) =
x 5S - &§ 8 & & SEs 2 ¥ B
Co YY = «< e FF Oo ma « a
2) z —! Wi a x o n < oO ov rT)
> £ FF &« 3 8 > ub wm ZF tf =F &
F nu O 2 0 s = YW =
<= oO <q z ” x
Oo o 0 *§ Oo =< §& f roa bi
a = 2 = zu
oe =z oO rF = 6G
uO a
=
<
ag

Figure 8. A comparison of the green algal floras of the Goleta Slough and
thirteen other New England estuarine habitats. expressed as the total number of
green algae as well as the % of Goleta Slough taxa.

varied from 5-12 taxa/month versus 3—9 for the browns and 23-
29 for the reds. The mean number of taxa/month was 39.2 +
6.0, with reds contributing 24.7 (+ 2.8), greens 8.4 (+ 2.1) and
browns 6.1 (+ 2.4). Four species were collected each month (1.e.,
Gelidium coulteri, Gigartina canaliculata, G. leptorhynchos, Pte-
rosiphonia dendroidea). Thirteen others occurred during eight of
nine months (90% occurrence), including Chaetomorpha aerea,
Cladophora columbiana, Ulva taeniata, Egregia menziesii, En-
darachne binghamiae, Scytosiphon dotyi, Centroceras clavulatum,
Ceramium eatonianum, Corallina vancouverlensis, Cryptopleura
violacea, Gastroclonium subarcticulatum, Gigartina volans and
Polysiphonia hendryi. By contrast, 30 of the 94 taxa were only
found once (~32% of total flora).

ABUNDANCE PATTERNS WITHIN GOLETA SLOUGH

Figure 9 illustrates the abundance of green algae and the Cal-
ifornia horn snail (Cerithidea californica) within three shallow

Table 3.

Monthly records of seaweed taxa on Goleta Point transect.

J F M A M J J A S O ior
Chlorophyta

Acrochaete viridis x 11.1
Blidingia minima x x Xx x 44.4
Bryopsis corticulans Xx 11.1
Bryopsis hypnoides X 22.2
Chaetomorpha aerea Xx Xx x Xx Xx x Xx 88.9
Cladophora columbiana Xx X x X x X Xx 88.9
Cladophora graminea Xx 11.1
Cladophora serice X X x Xx 55.6
Codium sp. (basal filaments) Xx 11.1
Enteromorpha clathrata x Xx X X X 66.7
Enteromorpha compressa x x x x x x 77.8
Enteromorpha flexuosa ssp. flexuosa X Xx X 33.3
Enteromorpha linza Xx Xx X X X 66.7
Enteromorpha prolifera X Xx X X Xx X 66.7
Rhizoclonium implexum a 11.1
Ulva californica Xx x Xx 33.3
Ulva lactuca X X X X 55.6
Ulva taeniata x x x x Xx x Xx 88.9

# Green algal taxa/month 7 1] 7 5 8 7 10 12

Mean # (+ SD) green algal taxa/month (8.4 + 2.1)

Total green algal taxa (18)

Phaeophyta

Colpomenia sinuosa Xx Xx 22.2

PoC

BIOpoyy

96 IOA]

Table 3. Continued.

J M A M J A N %*

Cylindrocarpus rugosus x x X 44.4
Ectocarpus parvus x x x x 55.6
Egregia menziesii x x X 4 x x x 88.9
Endarachne binghamiae Xx X X X X X 88.9
Giffordia sandriana X Xx Xx X 44.4
Hincksia granulosa X 222
Scytosiphon dotyi X X Xx X Xx x x 88.9
Scytosiphon lomentaria x X 22.2
Sphacelaria didichotoma x X 33.3
Taonia lennebackerae X x x 44.4
Zonaria farlowii x X Xx x 55:0

# Brown algal taxa/month 9 9 3 5 5 6 8

Mean (# (+ SD) brown algal taxa/month (6.1 + 2.4)

Total brown algal taxa (12)

Rhodophyta
Ahnfeltia fastigiata X X x X Xx x 77.8
Antithamnion defectum x 11.1
Asterocolax gardneri ee |
ae lia 11.1
angia v cularis Xx 11.1

oo. ein ssp. dichotoma X 33:3
Bossiella orbigniana ssp. orbigniana x ie eat
Centroceras clavulatum x x X Xx X Xx 88.9
Ceramium californicum x 11.1

ysno[s k1BJOH—o1YydH pue uosarylepy [p661

SEC

Table 3. Continued.

J ig M A M J A %*
Ceramium eatonianum Xx x Xx Xx X X 88.9
eramium SIMI Xx 11.1
Chondria nidifi x Xx Xx Xx 55.6
Corallina officinalis var. chilensis x 11.1
Corallina vancouveriensis 4 X x X x X X 88.9
aera ea 11.1
topleur X 11.1
C ie ies Xx x X X x x 88.9
Cumagloia andersonii X Xx X 44.4
Erythrotrichia carne x X 33.3
Gastroclonium subarticulatum x X x x x Xx 88.9
Gelidiocolax microsphaerica Xx 11.1
Gelidium coulteri x Xx X x X x X 100
Gelidium pusillum x X x 33.3
Gigartina canaliculata x Xx X x x Xx X 100
Gigartina leptorhynchos X Xx X x Xx x Xx 100
Gigartina ornithorhynchos x x x 33.3
Gigartina volans X Xx x x x Xx 88.9
Gracilaria papenfussi X Xx Xx x x 66.7
Gracilaria textorii var. -cunninghar x x x X 55.6
Gracilariophila oryzo x 11.1
Grateloupi ae Xx x X Xx 55.6
Gymnogonegrus chitor 11.1
Gymnogongrus leptophyllus X 22.2
Herposiphonia secunda forma tenella X 11.1

vIopoyYy

96 ‘1OA]

Table 3. Continued.

M A M J %*
Herposiphonia verticillata 11.1
Holmesia californica 11.1
Janczewskia lappacea X 33.3
Laurencia pacifica 22.2
Laurencia spectabilis x X Xx 66.7
Mastocarpus papillatus x x 22.2
Melobesia marginata 11.1
Melobesia mediocris 11.1
Microcladia coulteri X X 22.2
Nemalion helminthoides X X 55.6
Nienbergia andersoniana 22.2
Pogonophorella oo X 11.1
Polysiphonia hendry X Xx X 88.9
Polysiphonia pac cifica eae
Polysiphonia scopulorum var. villum 11.1
Po ra perfo x X X X 66.7
Prionitis lanceolata Xx 11.1
Pterochondria woodii 22.2
Pterosiphonia baileyi X X 44.4
Au spe dendroidea X Xx X X 100
Rhodoglos. X X 55.6
Rhodes californicum x 11.1
Rhodymenia pacifica X 22.2
pied aed Se ieee X X X 55.6

Scinaia articulata

yBnojs RIaJOH—a1yaHpue uosaryiepy [p661

LEG

Table 3. Continued.

J F M A M |

A N %*

Scinaia confusa X 11.1
Schizymenia pacifica X Xx 22.2
Smithora naiadum X X 4 X X x 66.7
Stenogramme ee 11.1
Stylonema alsi x x x 22.2

# Red algal taxa/month 23 26 24 19 24 23 29 26 28

Mean # (+ SD) red algal taxa/month (24.7 + 2.8)

Total red algal taxa (64)

Total # taxa/month 39 45 40 27 37 32 44 44 45

Mean # (+ SD) total taxa/month (39.2 + 6.0)

Total seaweed taxa (94

* The % values represent calculations based upon a total of nine monthly samples.

BET

elOpoyYy

96 ‘10A]

1994] Mathieson and Hehre—Goleta Slough 239

Pan #1
1000 5 300
dl F
--- #Cerithidea i
800 - a F 250
——*— _ Plant Biomass 1 oN L

—_
w Q
= £
<—— Pan #2 —
3p 1000 r 300 oD)
— t —
> ep)
= w)
Oo) <q
> =
i O
i co
al aed
< <
—- O
7p) —!
= §
Pan #3
1400 - 300
1200 ae + 250
Pe. rig f
1000 + 7 sO + 200
J ‘ “J 5
/
800 + , + 150
7 _-« ,
600 4 oo NS Fe ie
my ff
400 4 a 50
200 a a 0
0 100 200 300 400 500 600 700

HORIZONTAL DISTANCE (m)

Figure 9. Spatial patterns of green algal biomass (g dry weigh/m?) and horn
snail densities (#/m?) within three shallow Goleta Slough pans.

240 Rhodora [Vol. 96

CERITHIDEA
1000

w

E 800 4

Ras

600 +

n :

=

A 400-5

eS | 2

= 200 - GY 7

° Vs

#1
100 5
E. compressa

a 80 + CU. oxysperma

E

= HC. sericea

= eos

2p)

79)

= 40 4

Q

[ea]

PAN NUMBER

Figure 10. Mean (+ SD) snail density (#/m*) and biomass of Cladophora
sericea, Enteromorpha compressa and Ulvaria oxysperma (g dry weight/m?) within

three shallow Goleta Slough pans.

pans. A strong similarity was evident between plant and snail
abundances, except for one modest deviation within pan #3. Up-
wards of 800-1300 snails/m? were recorded, with maximum plant
biomass ranging from ~50-250 g dry weight/m?*. Cladophora
sericea and Enteromorpha compressa were the dominant sea-
weeds with mean biomass values of ~73 (+ 26.4) and ~82 (+
5.8) g dry weight/m’; mean snail densities ranged from 225 (+
101) to 826 (+ 82)/m? (Figure 10).

1994] Mathieson and Hehre—Goleta Slough 241
DISCUSSION

Patterns of species composition and richness of seaweeds at
our study sites show strong contrasts (Figure 5), which may be
associated with pronounced habitat and local environmental vari-
ability (Littler et al., 1991; Murray and Littler, 1989; Thom, 1976,
1980; Thom and Widdowson, 1978). Strong gradients of water
motion, substratum availability and depth are evident from the
offshore Naples reef site into the Goleta Slough, with deep water
and abundant rocky substrata (former) versus muddy, shallow
sedimented habitats (latter). The Goleta Point and Bay sites are
exposed to contrasting (i.e., diminishing) patterns of water motion
and sand fluctuation (Wheeler et al., 1981), with the latter often
causing extensive erosion and burial at Goleta Point (Figure 3)
and periodically blocking off the Slough (Macdonald, 1977; Onuf
and Zedler, 1988). Contrasting patterns of light penetration are
evident between Goleta Point and the Slough due to enhanced
sedimentation (Wheeler et al., loc. cit.), while hydrographic dif-
ferences between coastal and Slough habitats are most pro-
nounced during summer (Figure 4). As would be expected, the
most depauperate flora occurs within the Slough, a shallow, turbid
and hydrographically variable environment (Zedler, 1982b). The
most diverse flora was found at Goleta Point, which had a variety
of intertidal and subtidal habitats, as well as intermediate con-
ditions of light penetration, sediments and hydrographic vari-
ability. The Goleta Point transect was also the most extensively
studied habitat (Table 3). The subtidal zone within the Slough is
limited and shallow, while it is more expansive on the adjacent
open coast. Typically, temperate subtidal habitats are dominated
by a variety of red and brown algae, with ratios of greens to reds
decreasing with depth (Dawson et al., 1960; Edelstein et al., 1969;
Kain, 1960; Lamb and Zimmerman, 1964; Mathieson, 1979;
Neushul, 1965; Sears and Wilce, 1974). By contrast, the Slough’s
flora is dominated by a few ephemeral green algae (Figure 6).

As noted previously, diversity of seaweeds near Goleta can be
further documented using Cheney’s (R + C)/P ratio. Three of the
four study sites are interpreted as having “warm water”’ floras
(~5.9-7.7), except for the offshore Goleta Bay site (2.0). Obvi-
ously, these floristic ratios are influenced by variability of habitat
(see above) and species composition, with the highest value oc-
curring within Goleta Slough (~ 7.7) where only three brown algae

242 Rhodora [Vol. 96

were recorded. In comparing these ratios for New England sites,
eleven outer and mid-estuarine locations had values of < 6.0,
indicating ‘“‘mixed”’ or “‘cold-water”’ floras, while three inner riv-
erine habitats (i.e., Lamprey, Salmon Falls and Squamscott Riv-
ers), with reduced numbers of brown algal taxa, had ratios of 7-
14 (cf. Mathieson and Penniman, 1986a, 1991). Adjacent open
coastal sites in New England have much lower ratios (~ |.72-3.0)
and an overall mean value of ~2.5 (Mathieson and Penniman,
1986b). Such patterns confirm the usefulness of Cheney’s ratio
primarily for open coastal populations.

Based on a variety of detailed studies at 22 open coastal inter-
tidal sites in southern California, Littler et al. (1991) recorded
224 taxa (i.e., 149 red, 47 brown and 23 green algae), while we
recorded 152 at four primary sites (i.e., 104 red, 22 brown and
26 green algae). Our documentation of higher numbers of green
algae and six new geographical records (see above) probably re-
sulted from more extensive seasonal sampling of estuarine and
subtidal habitats. In discussing distributional patterns of individ-
ual taxa, Littler et al. (loc. cit.) state that only 12 of the 224
seaweeds (i.e., exclusive of Cyanophyceae) occurred at all 22 sites,
with these consisting of corallines, sheet-like species and small
filamentous algae (Stewart, 1982, 1983). Most other taxa had
restricted distributions, reflecting the high degree of local envi-
ronmental variability between habitats (Murray and Littler, 1989).
Both studies document a diverse warm-temperature flora dom-
inated by red algae (Liining, 1990; Thom, 1976, 1980) with re-
stricted occurrences of several taxa. Littler and colleagues (loc.
cit.) note that subclimax assemblages occur where there is a lack
of environmental constancy (e.g., fluctuating sand) or some form
of physiological stress (e.g., domestic pollution). In such habitats,
opportunistic species readily occupy newly opened space (Em-
erson and Zedler, 1978; Murray and Littler, 1978; Wilson, 1925).
If disturbance is periodic and not too severe, intertidal popula-
tions exhibit increased densities due to coexistence of both early
and late successional species (Sousa, 1979). Such patterns prob-
ably occur at Goleta Point, where a large number of epiphytic
and “turf’ forms (Stewart, 1982, 1983), as well as the highest
number of taxa/site, are found.

Our floristic records for open coastal taxa can be compared
with several previous studies near Goleta and throughout south-
ern California (e.g., Dawson, 1959a, 1959b, 1965; Murray and

1994] Mathieson and Hehre—Goleta Slough 243

Littler, 1989; Nicholson and Cimberg, 1971). For example, Daw-
son and Nicholson and Cimberg (loc. cit.) recorded intertidal
species composition at three contiguous points near Goleta (Fig-
ure 1), with the former recording 38 and 47 seaweed taxa/site at
Goleta and Santa Barbara Points, respectively and the latter, 49
taxa at Devereaux Point. Murray and Littler (loc. cit.) documented
seasonal collections at 21 rocky intertidal sites between Govern-
ment Point (Santa Barbara County) and Ocean Beach (San Diego
County), finding a range of 51-107 taxa/site and a mean of 76.3
+ 15.5. In addition, they recorded 93 taxa at Devereaux Point,
consisting of 69 red, 15 brown and 9 green algae. Dawson’s col-
lections of “dominant intertidal seaweeds” at Goleta Point (38
taxa) were based upon three seasonal collections (winter, spring
and fall), while ours (i.e., 94 taxa) were made during nine monthly
collections, which included all conspicuous taxa from an intertidal
and shallow subtidal transect. Thirty of these 94 taxa (1.e., ~32%)
were only collected once (Table 3), giving a mean value of 39.2
+ 6.0 taxa/month. The latter value is almost identical to Daw-
son’s report. Overall, our records from Goleta Point are very
similar to those of Murray and Littler (loc. cit.) from Devereaux
Point (~93 taxa), while they are much higher than Dawson’s from
Goleta Point. Unfortunately, further comparisons are difficult
because of varying collecting methods and sampling frequencies.
Even so, eight of Dawson’s “common intertidal” taxa from Goleta
Point were never found by us at any of our four sites (1.e., Ani-
socladella pacifica, Callophyllis obtusifolia, Chondria decipiens,
Corallina pinnatifolia, Cryptopleura lobulifera, Gracilaria lema-
neiformis, Laurencia splendens and Pleonosporium squarrulo-
sum), while Murray and Littler (loc. cit.) found half of these taxa
at Devereaux Point. Thus, four of Dawson’s common taxa may
have become less so, while other specific changes may have oc-
curred (Widdowson, 1971).

Estuarine reductional and compositional patterns observed
within the Goleta Slough (Figure 6) are fairly typical (Coutinho
and Seeliger, 1984; Josselyn and West, 1985; Ketchum, 1983;
Mathieson and Penniman, 1986a; Wilkinson, 1980); even so, the
rate of species loss upstream and dominance by ephemeral green
algae are more pronounced within the Slough than in San Fran-
cisco Bay (Josselyn and West, loc. cit.; Silva, 1979) and most New
England estuaries (Figure 7). For example, most open coastal red
and brown algae within the Slough dropped out after ~0.09 miles.

244 Rhodora [Vol. 96

By contrast, Silva (loc. cit.) found that 125 of the 156 total taxa
recorded from San Francisco Bay penetrated the central Bay,
while the remainder were protected water species (estuarine) that
were more prevalent in the Bay than on the open coast. In New
England open coastal taxa extend inland about one mile on the
tidal portions of the Merrimack River in Massachusetts (Ma-
thieson and Fralick, 1973), ~2-3 miles on the York River in
Maine (Mathieson et al., 1993) and ~8.5 miles on the Piscataqua
River in New Hampshire (Mathieson et al., 1983; Reynolds and
Mathieson, 1975). Of the three New England sites, the Piscataqua
River has the strongest currents (~5.0 knots) and most abundant
rocky substrata, while industrial development and eutrophication
are limited on the York, moderate on the Piscataqua, and maxi-
mal on the Merrimack (Mathieson et al., 1993). Several physical
factors are probably responsible for the pronounced truncation
of species within Goleta Slough. Both the range of salt content
and maximum salinities are greater (i.e., ~0-68%o, Figure 4) than
those found within the entire Great Bay Estuary (i.e., all ten
subareas), which vary from ~0-34%o (Mathieson and Hehre, 1986:
Norall et al., 1982). Goleta’s arid climate and the Slough’s re-
stricted tidal exchange (i.e., via sand deposition) no doubt con-
tribute to these high salinities (Macdonald, 1977: Onuf, 1987:
Onuf and Zedler, 1988; Zedler, 1982b). Maximum surface water
temperatures within the Great Bay system are less than those in
the Slough, but they exhibit a greater seasonal amplitude, varying
from ~—2.0°-27.0°C in the former (Mathieson and Hehre, loc.
cit.) to ~14°-35°C in the latter (Figure 4). The Slough’s shallow
topography and limited solid substrata also contribute to its re-
duced and patchy flora, as well as its dominance by loose-lying
and epiphytic seaweeds (Josselyn and West, 1985; Norton and
Mathieson, 1983).

In contrast to the depauperate flora of the Goleta Slough (i.e.,
26 taxa and 5.1 + 4.3 taxa/site), most New England habitats are
more diverse, particularly outer- and mid-estuarine sites like the
York (72 taxa and 21.4 + 3.2) and Piscataqua Rivers (143 and
25.3 + 24.9) and Little (130 and 34.1 + 29) and Great Bays (90
and 22.5 + 18.0). Depauperate floras dominated by green algae
may also occur within inner riverine sites like the Cocheco, Salm-
on Falls and Winnicut Rivers (i.e., 1.3 + 1.6 to 4.5 + 4.3 taxa),
as well as within the outer tidal portions of the Merrimack River
(3.5 + 5.2). The latter is badly polluted and one of the largest

1994] Mathieson and Hehre—Goleta Slough 245

sources of freshwater discharge into the Gulf of Maine (Anon.,
1984, 1987; Appolonio, 1979; Jerome et al., 1965; Lyons et al.,
1982: Mathieson and Fralick, 1973; Miller et al., 1971). Com-
paring estuarine macroalgal communities within San Francisco
Bay, Josselyn and West (1985) found a range of 28-61 taxa/site
beyond its central portion, with lowest numbers occurring in South
Bay. They noted that six of the eight most widely distributed
species were green algae (Enteromorpha clathrata, E. intestinalis,
E. linza, Ulva angusta, U. lactuca and Cladophora sericea) and
two reds (Antithamnion kylinii and Polysiphonia denudata). Our
studies of the Goleta Slough (Figure 8 and Table 2), plus those
of Zedler (1982a, 1982b) from the Tijuana Estuary near San Diego
and Norris (1970) from Elkhorn Slough near Carmel Bay, confirm
a similar dominance of many of these same ephemeral green algae.
Dawson (1962) found many of the same taxa (i.e., 9 green, 4
brown and 29 red algae) within the more saline Bahia de San
Quintin. In summary, the Goleta Slough’s flora is primarily dom-
inated by ephemeral green algae and contains no perennial reds
or browns; thus, it is analogous to several shallow embayments
(estuaries) along the Pacific Coast of the U.S.A. and in New En-
gland (Figure 8).

As noted by several investigators (Anon., 1964; Clokie and
Boney, 1980; Daly and Mathieson, 1977; Edwards, 1972; Littler,
1980; Littler et al., 1991; Patrick, 1963, 1964; Round, 1981;
Wilkinson, 1980; Zedler, 1982b), patterns of low species richness
are typical responses to stress, often allowing only a few tolerant
species to dominate in both numbers and biomass. For example,
the abundance of such ulotrichalean green algae as Ulva /actuca,
Enteromorpha and Ulvaria (Monostroma) spp., typifies many eu-
trophied estuarine habitats, including the Goleta Slough (Cotton,
1910; Fritsch, 1956; Sawyer, 1965). The latter are tolerant of
extremes of pollution and gross fluctuations in hydrographic con-
ditions. Similar patterns are found in several estuaries having
pronounced hydrographic variability and high sedimentation, in-
cluding the Chesapeake Bay (Orris, 1980), the Merrimack River
(Mathieson and Fralick, 1973) and many British and Dutch es-
tuaries (Hartog, 1967; Nienhuis, 1975; Wilkinson, 1980). Within
such turbid habitats, there is a relatively small pool of common
(cosmopolitan) estuarine species, including Cladophora, Entero-
morpha, and Ulva spp., together with several ceramialean red
algae (Munda 1969, 1972; Orris, 1980; Wilkinson, 1980). The

246 Rhodora [Vol. 96

similarity of the Slough’s green algal flora to New England estu-
arine vegetation is striking, with most sites exhibiting floristic
affinities of 50% or more (Figure 8). In discussing salt marsh
vegetation (i.e., angiosperms) within southern California, Zedler
(1982b) describes a rather limited and highly stressed community
from several hypersaline marshes, while Onuf and Zedler (1988)
state that six of the seven species and all of the genera listed by
Reimold (1977) from the eastern United States are in common.
Thus, the estuarine chlorophycean and angiosperm floras of the
two regions exhibit reduced numbers of taxa but strong floristic
similarities. Only one other seaweed, the red alga Sty/onema al-
sidii, occurred both within the Slough and in New England es-
tuarine habitats, while 18 other cosmopolitan and coastal taxa
from the Goleta and Santa Barbara area were in common to New
England (~13.0% of 146 taxa). These include 13 green (Acro-
chaete viridis, Blidingia minima, Chaetomorpha aerea, Cladoph-
ora sericea, Codium fragile, Enteromorpha clathrata, E. com-
pressa, E. intestinalis, E. linza, E. prolifera, Rhizoclonium
riparium, Ulva lactuca and Ulvaria oxysperma), 2 brown (Des-
marestia viridis and Scytosiphon lomentaria) and 3 red algae (Au-
douinella daviesii, Erythrotrichia carnea and Nemalion helmin-
thoides).

Describing seasonal variability of southern California sea-
weeds, Widdowson (1971) notes a conspicuous maximum in
March-June and a minimum during August-September, due to
changes of individual species that are present year-round. We
found a rather irregular pattern on the Goleta Point transect (Ta-
ble 3), which may have resulted from pronounced sand fluctua-
tions (Figure 3). Dawson (1965) states that no conspicuous sea-
sonal differences in quantity and quality of southern California
seaweeds occur. Littler et al. (1991) emphasize that local-scale or
even site-specific conditions predominate and they often obscure
broad climatic effects. Thus, major stochastic abiotic disturbances
may be more important than subtle seasonal patterns, including
heavy precipitation and flooding, extreme atmospheric conditions
during daylight tidal emersions, storm generated waves and sand
burial/scouring (see above). More pronounced seasonal patterns
are often evident within cold-temperate areas like New England
(Mathieson, 1989; Mathieson and Penniman, 1986a).

As noted by Zedler (1980, 1982a, 1982b), algal mats are often
conspicuous and productive components of southern California

1994] Mathieson and Hehre—Goleta Slough 247

tidal marshes, owing to the open canopies and short stature of
their vascular plants. Two green algae (i.e., Enteromorpha clath-
rata and Rhizoclonium riparium), plus a variety of microscopic
diatoms and blue greens, dominate the Tijuana Estuary near San
Diego (Zedler, loc. cit.), while an analogous benthic microflora
plus Enteromorpha and Ulva are abundant within the Mugu La-
goon near Port Hueneme (Onuf, 1987). According to Rudnicki
(1986), maximum standing crops of dominant green algae within
the Tijuana Estuary average 15 g dry wt./m?, which is much less
than the above ground biomass (~ 100-1100 g) of associated vas-
cular plants (Zedler, 1982b). Green algal mats within the Goleta
Slough were primarily composed of Cladophora sericea and E.
clathrata, with mean biomass values of ~73 and 82 g dry wt./
m/?, respectively (Figure 10). Such values are ~5.0—5.5 times greater
than equivalent green algal mats within the Tijuana Estuary (Rud-
nicki, loc. cit.), and those of E. clathrata from San Francisco Bay
(Shellem and Josselyn, 1982); they are also ~ 20-60 times greater
than mid-estuarine Enteromorpha spp. and Ulva lactuca popu-
lations from New England (Chock and Mathieseon, 1983). By
contrast, standing crop values for Goleta Slough macroalgae are
approximately the same as those found within New England inner
riverine sites such as the Oyster River (Mathieson, unpubl. data).
Fucoid algae typically dominate New England mid-estuarine hab-
itats, with Ascophyllum nodosum ecad scorpioides and A. nodosum
having mean biomass values of ~300 and 500 g dry wt./m?,
respectively (Chock and Mathieson, loc. cit.). Thus, green algal
standing stocks within the Slough are substantial but less than
those of southern California vascular plants and New England
estuarine fucoid algae.

The high densities of the California horn snail Cerithidea cal-
ifornica within the Goleta Slough (~225-826 snails/m2) should
be emphasized as well as the similarity of distributional patterns
to Cladophora sericea and Enteromorpha clathrata (Figure 9).
Among others, McCloy (1979) and Macdonald (1969) record high
densities of horn snails (250 to > 1000/m2) from southern Cal-
ifornia marshes, while Morris et al. (1980) state that it is probably
the most common estuarine snail within this geography, often
forming dense aggregations under debris and amongst plants on
high lagoonal mudflats. Several investigators (e.g., Morris et al.,
loc. cit.; Onuf, 1987; Whitlatch and Obrebski, 1980) suggest that
C. californica primarily feeds upon microorganisms and fine or-

248 Rhodora [Vol. 96

ganic detritus. By contrast, Dawson (1962) states that Entero-
morpha and other green algae serve in considerable part as food
for this extremely abundant snail, while Zedler (1982b) describes
an enhancement of macroalgal biomass within the Tyuana Es-
tuary after the exclusion of horn snails (1.e., via cages). She further
states that the removal of horn snails by shore birds results in
increased algal mats, while in situ patchiness of snails and inter-
tidal algal mats may be due to feeding by birds and other carni-
vores. The distributional patterns of C. californica are analogous
to those of the common mud snail //yanassa obsoleta, which
occurs in aggregated masses on mudflats and other estuarine hab-
itats on the east coast of North America (Morris, 1947; Smith,
1964). Similarly, the removal of J. obso/eta from intertidal mud-
flats in Georgia resulted in enhanced algal and microbial biomass
(Pace et al., 1979). Whether the abundance of C. californica within
Baja and southern California marshes (Barnard, 1962; Dawson,
1962; Morris et al., 1980; Zedler and Nordby, 1986) indicates a
consistent and similar interaction with seaweeds like that of J/.
obsoleta and Littorina spp. on the east coast (Mathieson et al.,
1991; Pace et al., loc. cit.), should be better clarified by exclusion
and manipulative experiments. In a discussion of the shore crab
Hemigrapsus oregonensis, Onuf (1987) states that it is capable of
shredding live macrophytes and will grow well in the laboratory
provided solely with Enteromorpha (Kuris and Mager, 1975).
However, in nature it feeds on diatoms (Morris et al., 1980) and
is a scavenger and predator (Chapman et al., 1982). Lastly, the
interrelationship between green algae and some surface-feeding
shorebirds (e.g., coots) within Mugu Lagoon should be noted, as
they primarily feed on thick mats of Enteromorpha (Onuf, 1987).

ACKNOWLEDGMENTS

The first author would like to express his deep appreciation to
the late Mike Neushul for providing space and resources during
a sabbatical leave at the University of California (Santa Barbara).
Michael’s insights and broad range of interests were a constant
source of stimulation. James Woessner and Miles Anderson are
acknowledged for their assistance with a variety of diving collec-
tion. Chris Onuff and Jack Loman are acknowledged for helping
to gain access to the Goleta Salt Marsh, which is located on Santa
Barbara’s Municipal Airport. Lastly, the first author would like

1994] Mathieson and Hehre—Goleta Slough 249

to acknowledge the encouragement and help of his wife, Myla
Mathieson.

LITERATURE CITED

ABBOTT, I. A. AND G. J. HOLLENBERG. 1976. Marine Algae of California. Stanford

Univ. Press, Stanford.

ANONYMousS. 1964. An investigation of the effects of oe wastes on kelp.

State Water Quality Control Bd. (California) Publ. 26.

. 1984. National Status and Trends Program he Marine Environmental
Quality. NOAA, Washington, D.C.

1987. Strategic Assessment of Near Coastal Waters, pee! Case
Study. EPA/NOAA Team on Near Coastal Waters, Washington, D.C.
APPOLONIO, S. 1979. The Gulf of Maine. Courier of Maine sis Rosen
Bacu, S. D. AND M. N. JossELYN. 1978. Mass blooms of the alga Cladophora

prolifera Seg nae Cladophorales) in Bermuda. Mar. Poll. Bull. 9: 34-37
BALDwIn, J. 1974. Chmates of the United States. National Oceanic and

oe Administration, Washington, D.C.

BARNARD, J. L. 1962. Benthic marine exploration of Bahia de San Quintin, Baja,

California, 1960-61. No. 6. General. Pac. Nat. 3: 250-274
BuipING, C. 1963. A critical survey of European taxa in Uivales. Part I. Cap-

sosiphon, Percursaria, Blidingia, Enteromorpha. Op. Bot. Soc. Bot. Lund 8:

1-160.

. 1968. A critical survey of European taxa in Ulvales, Part II. Ulva,

Ulvaria, Monostroma, Kornmannia. Bot. Notiser 121: 535-629.
CHAPMAN, J. W.,C. P. ONUF AND M. L. QUAMMEN. 1982. Effects of fish predation

on peracarid crustaceans in Mugu Lagoon, California, pp. 274-285. In: G.

M. Cailliet and C. A. Simenstad, Eds., Proceedings of the Third Pacific

Workshop on Fish Food Habits Studies, Washington Sea Grant Prog. Publ.,

€.

CHAPMAN, V. J. 1977. Ecosystems of the World, Vol. 1: Wet Coastal Ecosystems.
Elsevier, Amsterdam

CHENEY, D.P. 1977. R & C/P—a new and improved ratio for comparing seaweed
floras. J. Phycol. 13 (Suppl.): 12.

Cuock, J.S. AND A. C. MATHIESON. 1983. Variations of New England estuarine
seaweed biomass. Bot. Mar. 26: 87-97

CIMBERG, R., S. MANN AND D. STRAUGHN. 1973. A reinvestigation of southern
California rocky intertidal beaches three and one-half years after the 1969
Santa Barbara oil spill: a preliminary report. Proc. Joint. Conf. Prev. Contr.
Oil Spills, oe D.C., pp. 697-702.

CrioxiE, J.J. P. anp A. D. Boney. 1980. The assessment of changes in intertidal
eco sieae ec major reclamation work: framework for interpretation
of algal-dominated biota and the use and misuse of data, pp. 609-675. In:
J.H. Price, D. E. G. Irvine and W. F. Farnham, Eds., The Shore Environment,
Vol. 2: Ecosystems, Academic Press, New York.

Co.uins, F.S. 1909. The green algae of North America. Tufts Coll. Stud. 2: 79-
480.

250 Rhodora [Vol. 96
12. The green algae of North America, first supplement. Tufts Coll.
9-109

Corron, A. D. 1910. On the growth of Ulva latissima in water polluted by
sewage. ee Misc. Inform. Roy. Bot. Gard. Kew, pp. 15-19.

COUTINHO, R. AND U. SEELIGER. 1984. The horizontal distribution of the benthic
algal flora in the Patos Lagoon Estuary, Brazil, in relation to salinity, sub-
stratum and wave exposure. J. Exp. Mar. Biol. Ecol. 80: 247-257.

Day, M. A. AND A. C. MATHIESON. 1977. The effects of sand movement on
intertidal seaweeds and selected invertebrates at Bound Rock, New Hamp-
shire, USA. Mar. Biol. 43: 45-55

Dawson, E. Y. 1946. A guide to the literature and distributions of the marine
algae of the Pacific Coast of North America. Mem. So. Calif. Acad. Sci. III:
1-134

1959a. A primary report on the benthic marine flora of southern Cali-
fornia. /n: Oceanographic Survey of the Continental Shelf Area of Southern
California. Publ. Calif. State Water Poll. Cont. Bd. 20: 169-218.
——. 1959b. Benthic marine vegetation. Jn; Oceanographic Survey of the
Sate eee of Southern California. Publ. Calif. State Water Poll.
nt. Bd. 20: 219-2
1962. Benthic marine exploration of Bahia de San Quintin, Baja, Cal-
oni. 1960-61. No. 7. Marine and marsh vegetation. Pac. Nat. 3: 275-
280.

. 1965. Intertidal algae. Jn: An Oceanographic and Biological Survey of
the Southern California Mainland Shelf. Publ. Calif. State Water Quality
Contr. Bd. 27: 220-231, 351-438.

——. M. NEUSHUL AND R. D. WILDMAN. 1960. Seaweeds associated with kelp
beds along southern oe 3 ae Mexico. Pac. Nat. 1: 1-81.

EDELSTEIN, T., J. S. CRAIGIE AND J. LACHLAN. 1969. Preliminary survey of
es sublittoral flora of a en J. Fish. Res. Bd. Canada 26: 2703-

713.

Epwarps, P. 1972. Benthic ne in polluted estuaries. Mar. Poll. Bull. 3: 55-60.
EMERSON, S. E. AND J. B. ZEDLER. 1978. —— of intertidal algae: an
experimental study. Mar. ‘Biol. 44: 315-3
M., M

Foster, M., M. NEUSHUL AND R. ZINGMARK. ee The Santa Barbara oil spill
Part II: initial effects on intertidal and kelp bed organisms. Environ. Poll. 2
115-134

FritscH, F. E. 1956. The Structure and Reproduction of the Algae, Vol. I.
Cambridge Univ. Press, —

GARBARY, D. J., L. GOLDEN AND R. F. SCAGEL. 1982. Capsosiphon fulvescens
(Capsosiphonaceae, Sion). rediscovered in the northeastern Pacific.
Syesis 15: 39-42.

Hartoa, G. Den. 1967. Brackish water as an environment for algae. Blumea
15: 31-43.

JEROME, W. C., JR., A. P. CHESMORE, C. O. ANDERSON, JR. AND F. Grice. 1965.
A Study of the Marine Resources of the Merrimack River Estuary. Mono-
graph Ser. No. 1, Div. of Mar. Fish., Dept. Natl. Res., Commonwealth of
Massachusetts.

Jounson, C. S. 1971, Macroalgae and their environments. Proc. Roy. Soc. En-
gland (B) TH: 195-207.

1994] Mathieson and Hehre—Goleta Slough 251

JossELYN, M. N. 1983. The Ecology of San Francisco Bay Tidal Marshes: A
Community Profile. U.S. Fish and Wild. Serv., Biol. Serv. Program, Wash-
ington, D.C., Rept. FWS/0BS-83/23.

, AND A.C. MATHIESON. 1978. Contribution of receptacles from the fucoid
Ascophyllum nodosum to the detrital pool of a north temperate estuary.
Estuaries 1: 258-261.

AND A. C. MATHIESON. 1980. Seasonal influx and decomposition of
alochiionGus macrophyte litter in a north temperate estuary. Hydrobiologia
71: —208.

. AND J. A. West. 1985. The distribution and temporal dynamics of the
estuarine macroalgal community of San Francisco Bay. Hydrobiologia 129:

Karn, J. M. 1960. Direct observations on some Manx sublittoral algae. J. Mar.
Biol. Ass. U.K. 39: 609-630.

Ketcuum, B. H. 1983. Estuarine characteristics, pp. 1-14. 7m: B. H. Ketchum,
Ed., Ecosystems of the World, Vol. 26: Estuaries and Enclosed Seas. Elsevier
Sci. Publ. Co., Amsterdam.

Kuris, A. M. AND M. MaAcer. 1975. Effect of limb regeneration on size increase
of molt of the shore crabs Hemigrapsus oregonensis and Pachygrapsus cras-
sipes. J. Exp. Zool. 193: 353-360.

Lams, M. AND M. H. ZIMMERMAN. 1964. Marine vegetation of Cape Ann, Essex
County, Massachusetts. Rhodora 66: 217-254.

Litter, M. M. 1980. Southern California ge ies ecosystems: methods,
community structure and variability, pp. 565-608. Zn: J. H. Price, D. E
Irvine and W. F. Farnham, Eds., The Shore eee nea Vol. 2: Poses.
Academic Press, New York

, D. S. Lirtter, 8. N. Murray AND R. R. Seapy. 1991. Southern Cali-
fornia rocky intertidal ecosystems, pp. 273-296. Jn: A. C. Mathieson and P.
H. Nienhuis, Eds., Ecosystems of the World, Vol. 24: Intertidal and Littoral
Ecosystems. Elsevier, Amsterdam.

Lumina, K. 1990. Seaweeds: Their oe Biogeography, and Ecophys-
iology. Wiley- sy nae Publ.,

Lyons, W. B., P. B. ARMSTRONG, R. _ omen AND H. E. GAUDETTE. 1982.
Trace aictal discharge into oe Bay Estuary, New Hampshire. Univ. New
Hampshire Sea Grant Tech. Rept., UNH-SG-176, Durham.

MACDONALD, K. B. 1969. Quantitative ee of salt marsh mollusc faunas
from the North American Pacific Coast. Ecol. Monogr. 39: 33-60.

1977. Plant and animal ae of Pacific North American salt

marshes, pp. 167-191. In: V. J. Chapman, Ed., Ecosystems of the World,

Vol. 1: Wet Coastal Ecosystems. Elsevier, Amsterdam.

AND M.G. BARBouR. 1974. Beach and salt marsh vegetation of the North
American Pacific Coast, pp. 175-233. Jn: R. J. Reimold and W. H. Queen,
Eds., Ecology of Halophytes. Academic Press, Inc., N.Y.

Mann, K.H. 1972. Macrophyte production and detritus food chains in coastal
waters. Mem. Ist. Ital. Idrobiol. 29: 353-383.

Mason, H. L. 1957. A Flora of the Marshes of California. Univ. California
Press, Berkeley.

saan: A.C. 1979. Vertical distribution mae hoe of subtidal seaweeds

northern New England, U.S.A. Bot. Mar. 20: 511-520.

252 Rhodora [Vol. 96

——. 1989. Phenological patterns of northern New England seaweeds. Bot.
Mar. 32: 419-438.

—— AND R.A. Fra.icx. 1973. Benthic algae and vascular plants of the lower
Merrimack River and adjacent shoreline. Rhodora 75: 52-64.

AND E. J. HEHRE. 1986. Asynopsis of New Hampshire seaweeds. Rhodora
88: 1-139.

——, —— AND M. Costa. 1993. Algal vegetation of the York River Estuary
and the eee open coast of southern Maine. Rhodora 95: 285-324.

FUS AND C. EMERICH PENNIMAN. 1983. Benthic ecology in an

sctuariae een rapid. Bot. Mar. 26: 213-230.

AND C. A. PENNIMAN. 1986a. Species composition and seasonality of

New acer seaweeds along an open coastal-estuarine gradient. Bot. Mar

29: 161-

AND ——— A phytogeographic interpretation of the marine flora

from the isles . Shoals, U.S.A. Bot. M 9: 413-434.

AND 91. Floristic patterns and numerical classification of New

England estuarine and open coastal seaweed populations. Nova Hedwigia

52: 453-485

AND L. G. Harris. 1991. as Atlantic rocky shore ecology,
pp. 109-191. Jn: A. C. Mathieson and P. H. Nienhuis, Eds., Ecosystems of
the World, Vol. 24: Intertidal and Littoral Sea Elsevier, Amsterdam.

McCtoy, M. J. 1979. Population regulation in the deposit feeding mesogastro-
pod Cerithidea californica as it occurs in San Diego salt marsh habitat. M.Sc.

1ego.

thesis, San Diego State Univ., San D

McComas, A. J., R. ig ATKINS, P. B. BincH, D. M. GORDON AND R. J. LUKATELICH.
1981. Eutrophication in the Peel-Harvey estuarine system, Western Aus-
tralia, pp. 323-324. In: B. Nielson and A. Cronin, Eds., Nutrient Enrichment
in Estuaries. Humana Press, New Jersey

Miter, B., D. A. NORMANDEAU, G. PIEHLER, P. HALL, A. MATHIESON, R. FRALICK,
D. TurGEoN, P. MAHONEY AND W. Owen. 1971. Ecological Study Mer-
rimack River Estuary- oe Normandeau Associates Inc. and
Wadsworth Owen Vast

Morris, P.A. 1947. A Field 1Guide to the Shells of Our Atlantic and Gulf Coasts.
Houghton Aaa Boston.

Morris, R. H., D. P. peta DE.C. HADERLIE. 1980. Intertidal Invertebrates
of California. Stanford Univ. Press, Stanford.

Munpa, I. 1969. Differences in the algal vegetation of two Icelandic fjords,
Dyrafjérdur (West Iceland) and Reydafjérdur (East Iceland), pp. 255-261.
In: R. Margalef, Ed., Proceedings of the Sixth International Seaweed Sym-
posium, Santiago de Compostela, Spain, Subsecretaria de la Marine Mer-

cante, Madrid.
. 1972. General features . — benthic algal zonation around the Icelandic
Coast. Acta nat. Islandica 21: 1-34.

MURRAY, : N. 1974. Benthic ae and — pp. 9.1-9.61. Jn: M. D. Dailey
B. Hill and N. Lansing, Eds., A Summary of Knowledge of the Sorter
ae Coastal Zone and Offshore sas Vol. II. Biological Environment,
Bureau of Land Management, U.S. Dept. of Interior, Washingto

AND M.M. Litter. 1974. Se ae of Intertidal Gommnniice
Near the U.S. Navy Sewage Outfall, Wilson Cove, San Clemente Island,
California. Technical Paper No. 396, Naval Undersea Center, San Diego.

1994] Mathieson and Hehre— Goleta Slough 250

——— . 1978. Patterns of algal succession in a perturbated marine

ead community. J. Phycol. 14: 506-512.
1989. Seaweeds and seagrasses of southern California: dis-

tributional lists for twenty-one rocky intertidal sites. Bull. So. Calif. Acad.
Sci. 88: 61-79.

NEUSHUL, M. 1965. SCUBA diving studies of the vertical distribution of benthic
marine plants. Botanica Gothobergensia 3: 161-176.

NicHotson, N. L. 1972. The Santa Barbara oil spills in perspective. Calif. Mar.
Res. Comm., CalCOFI Rept. 16: 130-149

AND R. L. CimsBerG. 1971. The Santa Barbara oil spills of 1969: a post-
spill survey of the rocky intertidal, pp. 325-399. In: D. S. Straughn, Ed.,
Biological and Oceanographical Survey of the Santa Barbara Channel Oil
Spill 1969-1970, Vol. I: Biology and Bacteriology. Allan Hancock Foun-
dation, Univ. South. Calif., Los Angeles

NieNHuIs, P. H. 1975. Biosystematics a ecalosy of Rhizoclonium riparium
(Roth) Harv. (Chlorophyceae: Cladophorales) in the estuarine areas of the
rivers Rhine, Meuse and Scheldt. Doctoral thesis, Univ. Gronin

Nora, T. L., A. C. MATHIESON AND C. E. PENNIMAN. 1982. Nutrient and
A goane data for the Great Bay estuarine ecient , New Hampshire-—
Mai : September, 1973—December, 1975. Publ. UNH-D/TR-83-1,
Univ. ae Hampshire Mar. Program, Durham.

Norris, J. N. 1970. Studies on the Taxonomy and Distribution of Some Ulvales
from Elkhorn Slough, California. M.A. thesis, San Francisco State College,
San Francisco.

North, W. J., G. C. STEPHENS AND B. B. NortH. 1970. Marine algae and their
relations to pollution problems. FAO Tech. Conf. Poll., Rept. MP/70/R-8,
pp. 1-22.

Norton, T.A.ANDA.C. MATHIESON. 1983. The biology of unattached seaweeds,
pp. 333-386. Jn: F. E. Round and D. J. Chapman, Eds., Progress in Phy-
cological Research, Vol. 2. Elsevier Sci. Publ., B. V., Amsterdam

OnuF, C. P. 1987. The Ecology of Mugu Lagoon, California: An Estuarine Profile.
U.S. Fish Wildl. Serv., Biol. Rept. 85 (7.15), Washington, D.C.

AND J.B. ZEDLER. 1988. Pattern and process in arid-region salt marshes—
southern California, pp. 570-581. 7m: D. D. Hook, W. H. McKee, Jr., H. K.
Smith, J. Gregory, V. G. Burrell, Jr.,. M. R. DeVoe, R. E. Sojka, S. Gilbert,
R. Banks, L. H. Stolzy, C. Brooks, T. D. Matthews and T. H. Shear, Eds.,
The Ecology and Management of Wetlands, Vol. 1: Ecology of Wetlands.
Timber Press, Portland

Orris, P. K. 1980. A revised species list and commentary on the macroalgae
of the Chesapeake Bay in Maryland. Estuaries 3: 200-206

Pace, M. L., S. SHIMMEL AND W. M. DarLey. 1979. The effect of grazing by a
gastropod, Nassarius obsoletus, on the benthic microbial community ofa salt
marsh mudflat. Estuarine Coast. Mar. Sci. 9: 121-134.

PapenFuss, G. F. 1976. Landmarks in Pacific North American marine phycol-
ogy, pp. 21-46. 7a: I. A. Abbott and G. J. Hollenberg, Marine Algae of
California. eo line so Stanford.

Patrick, R. 1963. Th J ing ecological
conditions. Ann. New York Acad. Sci. 108: 359-364.

. 1964. A discussion of natural and abnormal diatom communities, pp.

185-204. Jn: D. E. Jackson, Ed., Algae and Man. Plenum Press, New York.

254 Rhodora [Vol. 96

PrescotT, G. W. 1962. Algae of the Western Great Lakes Area. Wm. C. Brown
o. Inc., Dubuque

REIMOLD, R. J. 1977. — ce salt marshes of eastern United ee pp.
157-166. In: V. J. Chapman, Ed., eae of the World, Vol. 1: Wet
Coastal Ecosystems. Sieg aes: Aa

REYNOLDS, N. B. AND A. C. MATHIESON. a: Seasonal occurrence and ecology
of marine algae in a New Hampshire tidal rapid. Rhodora 77: 512-533

Ricketts, E. F., J. CALvin, J. W. HEDGPETH, AND D. W. PHILLips. 1985. Between
Pacific Tides (Sth ed.). Stanford Univ. Press, Stanford.

Riptey, J. D. 1969. A floristic and ecological study of Angel Island State Park,
Marin County, California. M.A. thesis, San Francisco State College, San

Francisco.
Rounb, F. E. 1981. The Ecology of Algae. Cambridge Univ. Press, Cambridge.
Rupnicki, R. 1986. Dynamics of macroalgae in Tijuana Estuary: response to
simulated wastewater addition. M.Sc. thesis, San Diego State Univ., San

Diego.

SAWYER, C. M. 1965. The sea lettuce problem in Boston Harbor. J. Water Poll.
Cont. Fed. 37: 1122-1133.

ScaGEL, R. F. 1966. Marine Algae of British Columbia and Northern Wash-
Boe Part. I. Chlorophyceae (Green Algae). Bull. Nat. Mus. Canada 207.

D. J. GARBARY, L. GOLDEN AND M. W. Hawkes. 1986. A Synopsis of
the Benthic Marine Algae of British Columbia, Northern Washington and
Southeast Alaska. Dept. Botany, Univ. British Columbia, Vancouver

SEARS, J.R.ANDR.T. Witce. 1974. Sublilttoral, benthic marine algae of southern
Cape Cod and adjacent islands: Ree periodicity, associations, diversity
and floristic composition. Ecol. Monog. 45: 337-365.

SEWELL, P. L. 1982. Urban groun sa er as a possible nutrient source for an
estuarine benthic ae bloom. Estuarine Coast. Shelf Sci. 15: 569-576.
SHELLEM, B. H. AND M. N. JossELyn. 1982. Physiological ecology of Entero-
morpha clathrata (Roth) Grev. on a salt marsh mudflat. Bot. Mar. 25: 541-

549

Sitva, P.C. 1979. The eich algal flora of central San Francisco Bay, pp. 287—
345. In: T. J. Conomos, Ed., San Francisco Bay—The Ubanized Estuary.
Pacific Division AAAS, oy Francisc

SmitH, G.M. 1969. Marine Algae of the ere Peninsula, California. Second
Edition incorporating the 1966 Supplement by G. J. Hollenberg and I. A.
Abbott. Stanford Univ. Press, Stanford.

SMITH, R. I. 1964. Keys to Marine Invertebrates of the Woods Hole Region.
Contr. No. 11, Systematic-Ecology Program, Mar. Biol. Lab., Woods Hole.

Sousa, W. P. 1979, Disturbance in marine intertidal boulder fields: the non-
equilibrium maintenance of species diversity. Ecology 60: 1225-1239.

SoutH, G. R. AND I. Titrtey. 1986. A Checklist and Distributional Index of
the Benthic Marine Algae of the North Atlantic Ocean. Publ. Huntsman Mar.
Lab and British Museum (Natural Hist.), St. Andrews and London.

SPARLING, S.R. 1971. Recent records of marine algae in San Luis Obispo County,
California. Phycologia 10: 235-240.

1977. An Annotated List of the Marine Algae (Chlorophyta, Phaeophyta,
Rhodauhyta) of San Luis Obispo County, California, With Keys to Genera
and Species. Blake Printery, San Luis Obispo.

1994] Mathieson and Hehre—Goleta Slough 295

SPETH, J. W. 1969. The fuss over coastal wetlands. Outdoor Calif. 30: 6-7

STEwarT, J.G. 1982. Anchor species and epiphytes in intertidal algal turf. Pac
Sci. 36: 45-59.

Fluctuations in the quantity of sediments trapped among algal

thalli on intertidal rocky platforms in southern California. J. Exp. Mar. Biol.

Ecol. 73: 205-211.

. 1991. Marine Algae and Seagrasses of San Diego County. California Sea
Grant College Publ. No. T-CSGCP-020, La Jolla.

STRAUGHN, D. S. 1971. Biological and Oceanographical Survey of the Santa
Barbara Channel Oil Spill 1969-1970, Vol. I: Biology and Bacteriology. Allan
Hancock Foundation, Univ. South. Calif., Los Angeles.

Taytor, W. R. 1957. The Marine Algae of the Northeastern Coast of North
America. Univ. Michigan Press, Ann Arbor.

THoM, R. M. 1976. Changes in the intertidal ses of southern California main-
land. M.A. thesis, California State Univ. each.

——. 1980. A gradient in benthic intertidal a assemblages along the south-
ern California coast. J. Phycol. 16: 102-108.

——— AND T. B. Wippowson. 1978. A resurvey of E. Yale Dawson’s 42 in-
tertidal transects on the southern California mainland after 15 years. Bull.
So. Calif. Acad. Sci. 77: 1-13.

Wesser, E. E. AND R. T. Witce. 1971. Benthic salt marsh algae at Ipswich,
Massachusetts. Rhodora 73: 262-291.

WHEELER, W. N., M. NEUSHUL AND B. W. W. HARGER. 1981. Development of
a coastal marine farm and its associated problems, pp. 631-636. Jn: T.
Levring, Ed., Proceedings of the Xth International Seaweed Symposium,
Goteborg, pines, Walter de Gruyter, Berlin.

WHITLATCH, R. AND S. OBREBSKI. 1980. Feeding selection and coexistence in
two deposit aca gastropods. Mar. Biol. 58: 219-226.

Wippowson, T. B. 1971. Changes in the intertidal algal flora of the Los Angeles
area since the survey by E. Yale Dawson in 1956-1959. Bull. So. Calif. Acad.

i. 70: 2-16

WILcE, R. T., C. W. SCHNEIDER, A. V. QUINLAN AND K. VANDEN BoscuH. 1982.
The life history and morphology of free-living Pi/ayella littoralis (L.) Kjellm.
(Ectocarpaceae, Ectocarpales) in Nahant Bay, Massachusetts. Phycologia 21:

-354

WILKINSON, M. 1980. Estuarine benthic algae and their environments: a review,
25-486. In: J. H. Price, D. E. G. Irvine and W. F. Farnham, Eds., The
Shore Environment, Vol. 2: Ecosystems. Academic Press, London.
Witson, O. T. 1925. Some experimental observations of marine algal succes-
sions. Ecology 6: 303-311
ZEDLER, J. B. 1977. Salt marsh community structure in the Tijuana Estuary,
California. Estuarine Coast. Mar. Sci. 5: 39-53.
1980. Algal mat productivity: comparisons in a salt marsh. Estuaries 3:
122-131.
1982a. Salt marsh algal mat uae spatial and temporal com-
parisons. Bull. So. Calif. Acad. Sci. 81: 41-50.
982b. The Ecology of Southern Lane Coastal Salt Marshes: A
Community Profile. U.S. Fish Wildl. Serv., Biol. Serv. Program, Washington,
D.C., FWS/OBS-8 1/54.

256 Rhodora [Vol. 96

— anp C. S. Norpsy. 1986. The Ecology of Tijuana Estuary, California:
An Estuarine Profile. U.S. Fish Wildl. Serv., Biol. Rep. 85 (7.5), Washington,
1oA6

DEPARTMENT OF PLANT BIOLOGY
JACKSON ESTUARINE LABORATORY
UNIVERSITY OF NEW HAMPSHIRE
DURHAM, NEW HAMPSHIRE, 03824

APPENDIX: STUDY SITES

Open Coast

GP: Goleta Point: A semi-exposed sandy habitat also called Campus Point
located at the west end of Goleta Bay (Dawson, 1959a, 1959b). Scattered rock
outcrops (Monterey shale/siltstone) and boulders occur on a broad sandy beach,
many of which are extensively scoured. The point is exposed to greater longshore
currents and wave action than within Goleta Bay (Wheeler et al., 1981). Subtidal
collections (— 10 to —30 feet) were made just offshore from the Point (= 0 miles
inland; 30-32%, 14.0°-18.0°C)

GB: Goleta Bay: A coastal site having limited wave action and currents (see
above). Substratum primarily consists of sand with scattered boulders; sand often
covers in situ benthic populations. Collections were made from —5 to —S0 feet
near two experimental farm sites (cf. Wheeler et al., 1981) (= just offshore from
ei ee 32%0, 14.0°-18.0°C).

aples Reef: An offshore reef site having abundant rocky substrata (Mon-
terey le sits on reduced levels of sediment and sand. Collections were
made from —15 to —40 feet (= ~2.0 miles offshore from Santa Barbara Point).

Goleta Slough

1: Near the mouth of the Goleta Slough, just downstream from Goleta Beach
State Pier; a sandy berm-like area on the northern bank of the channel. Collections
ere made from an old pier piling as well as on cliff facings with scattered mussel
populations. The mouth of the slough is periodically blocked off by sand deposition
during the summer, restricting tidal exchange (= 0.087 miles inland; 13-32%c,
15.0°-20.0°C).

2: A muddy area upstream from the mouth of the slough and near the entrance
(i.e., bridge) to Goleta Beach State Park. Collections were made on the south bank
of the channel, next to the park ranger’s house. The substrata consists of muddy
surfaces, as well as scattered and broken pieces of asphalt (= 0.311 miles inland,
27-30%0, 15.2°-21.5°C).

muddy area on the south bank of the channel adjacent to the entrance to
Goleta a State Park. Plants were collected from bridge pilings, scattered
boulders, and as entangled masses on extensive muddy shorelines (= 0.392 miles
inland; 2—25.5%o, 13.0°-15.3°C)

4: A muddy area near the Clarence Memorial Boulevard overpass (~0.125

1994] Mathieson and Hehre—Goleta Slough 250

miles) and upstream from the entrance to Goleta Beach State Park (~0.0623
miles). Collections were made on the south bank of the channel (= 0.448 miles
inland; 1-33%o, 13.5°-20.5°C).

5: A muddy area on the south bank of a channel, ~0.075 miles from the
Clarence Ward Memorial Boulevard Oye Pes and near : Dieyele haa A
series of gas pipelines were also present. Collections asses
of seaweeds on muddy surfaces, while epiphytic and entangled masses ie algae
also occurred within a small shallow pan (= 0.517 miles inland; 2—32%0o, 15.2°-
28.5°C).

6: Two shallow pans near station 7. Collections were made of free-floating and
entangled masses of plants within the pans and on muddy surfaces around the
periphery of the pan (= 0.622 miles inland: 11-42%, 18.0°-32.0°C).

7: The south bank of a narrow muddy channel at the end of a large shallow
pan, ~0. 150 miles Upstream from we aCe aes Memional ans over-

pass. Collections dy
ae epiphytic populations on Salicor us virginica and attached mseiee on ‘dcift
wood (= 0.647 miles inland; 22%, 15.4°-24.5°C).

8: A narrow tidal channel located ae 0.3 miles upstream from the
Clarence Ward Memorial Boulevard overpass on the south bank of the channel.
The site is located at the end of an auxilliary road and just SW from the end of
Santa Barbara’s Municipal Airport runway 33L. Collections were made of epi-
phytic plants and entangled masses on muddy substrata (= 0.772 miles inland;
9-25%0, 17.0°-25.0°C)

uddy channel approximately 0.375 miles upstream from the Clarence
Ward Memorial Boulevard overpass on the south bank of the channel. The site
is next to a pipe crossing. Collections were made on muddy vertical bankings (=
0.778 miles inland; 0-25%0, 15.5°-29.0°C).

10: One end ofa large pan just downstream from the junction of two channels.
Collections were made from emergent and tangled masses of seaweeds, which
grew amongst vascular plants (= 0.996 mules inland; 0-12%o0, 17.5°-26.0°C).

11: Junction of two tidal channels ~0.48 miles upstream from the Clarence

Ward Memorial Boulevard overpass. Collections were made from entangled mud-
dy plants (= 0.878 miles inland; 0—28%o0, 17.8°-32.0°C)

12: A narrow muddy tidal channel, ~0.555 miles upstream from the Clarence
Ward Memorial Boulevard overpass. Collections were made from entangled mass-
es amongst Salicornia virginica, etc. (= 0.94 miles inland; 0—29%o0, 23.0°-29.0°C).

13: A tidal channel ~0.2 miles downstream from a spill gate; considerable
freshwater runoff was evident, particularly during spring. Collections were made
along an embankment area lined with Salicornia virginica (= 1.058 miles inland:
0-29%o, 18.5°-30.5°C).

14: A large shallow pan next to the Goleta Water District facility. Stations 10
and 14 represent the extremities of the same pan. Collections were made from
free-floating and entangled masses of seaweeds growing amongst vascular plants,
as well as from entangled masses on muddy surfaces on the periphery of the pan
(= 1.27 miles inland; 2-13%o0, 19.5°-27.0°C).

15: A small pan area located just north of the spill gate adjacent to the Goleta
Water District facility. Collections were made from free-floating and entangled
masses of seaweeds, growing amongst vascular plants and on muddy marginal
surfaces around the pans (= 1.114 miles inland; 6—> 32%, 21.0°-> 30.0°C).

258 Rhodora [Vol. 96

16: A shallow pan area, partially dried up during summer and forming cracked
pavement-like masses of blue-green algae. Entangled masses of green algae were
also present around the margin of the pan, growing amongst Salicornia vir eee
many dead bivalve and periwinkle shells were found near these pans (= 0.977
miles inland; 13-68%, 18.5°-32.0°C).

17: A shallow pan just upstream from station 5. Collections were made from
entangled and free-floating masses of seaweeds growing in and around the pan (=
0.572 miles inland; 20-48%c).

18: A shallow pan area just upstream from station 17. Collections were similar
to those at 17 (= 0.641 miles inland: 11-42%, 18.0°-32.0°

19: A muddy channel bank next to station 18. Collections were made from
entangled masses on muddy surfaces and around the bases of vascular plants (=
0.654 miles inland; 1-35%o, 14.5°-16.0°C).

20: A mney beaut sabes ne was aime! ane and widened. Collec-
tions ts on the muddy channel surfaces
and bankings (= 0.734 miles inland; 11-32%o, 14, 5°29, 2 °C),

21: A variety of shallow pans similar to station 16, although ofa more extensive
area (= 0.828 miles inland; 1 1—50%e0, 24.0°-35.0°C

22: A small channel downstream from station 23, approximately 15 feet wide
with conspicuous freshwater drainage at low tide. Collections were made from
small sticks and from entangled and tufted masses of seaweeds growing on muddy
surfaces (= 0.942 miles inland; 12—25%o, 23.5°-32.0°C),

A muddy channel with abundant masses of green algae and vascular plants
(= 0.840 miles inland; 35%o0, 32.0°C).

24: The south bank of a wide, artificially dug channel near station 16, having
limited seaweed populations (= 0.840 miles inland; 0-22%0, 22.5°-32.0°C).

25: A shallow pan area just north of and parallel to a bridge carrying pipes over
a channel. Collections consisted mostly of entangled masses of seaweeds, growing
on and amongst Salicornia virginica (= 0.940 miles; 0-> 32%, 23.8°-> 30.0°C).

26: The northern bank of Glen Annie Creek, opposite station 24 and adjacent
a pipe crossing. A channel 25-30 feet wide with muddy surfaces covered with
Se reer (= 1.145 miles inland; 24%, 30.0°C).

27: Headwaters of the easternmost tidal channel in the Goleta Slough (Atas-
cadero). It originates from San Pedro Creek and adjacent to the Clarence Ward
Memorial Boulevard Highway. Collections were made near a concrete dam, and
on adjacent scattered rocks and muddy bankings (= 0.846 miles inland; 0-10%«,
12.0°-28.5°C).

RHODORA, Vol. 96, No. 887, pp. 259-286, 1994

ECOLOGY, REPRODUCTIVE BIOLOGY AND
POPULATION age a
OPHIOGLOSSUM VULGA

(OPHIOGLASSACEAE) IN Case

ROBERT T. MCMASTER

ABSTRACT

The northern adder’s-tongue fern, Ophioglossum vulgatum var. pseudopodum
(Blake) Farw. (Ophioglossaceae), occurs in five known populations in Massachu-
setts and has been designated as threatened by the Massachusetts Natural Heritage
Program. The autecology, population biology and population genetics of O. vul-
gatum var. pseudopodum were analyzed during the summer of 1992. The distri-
bution of each population was mapped, associated vegetation sampled, and per-
manent plots established for long-term study. Isozyme electrophoresis was employed
to assess variability within and between populations. The historical distribution
of the species in Massachusetts was examined based upon herbarium specimens.
Each of the known Massachusetts populations is in an early successional site with
a unique history of disturbance; sites include pasture, a power line right-of-way,
and abandoned beaver meadows. Four of the five populations are small, ranging
_ 1 to 109 sporophytes. All specimens subjected to isozyme analysis were

nomorphic for all loci examined. The implications of the spatial and temporal
distribution, population size, and low genetic variability of SA ne ur
gatum var. pseudopodum are discussed. Management considerations for the S-
ervation of this and other early successional species are suggested.

Key Words: Ophioglossum vulgatum, Ophioglossaceae, fern reproductive biolo-
gy, population genetics, early successional species, rare plant species,
self-fertilization, Massachusetts

INTRODUCTION

Ophioglossum vulgatum var. pseudopodum (Blake) Farw., the
northern adder’s-tongue fern (Ophioglossaceae), has a broad dis-
tribution, occurring throughout its range in a variety of wet and
mesic habitats. In Massachusetts, O. vu/gatum var. pseudopodum
is known from over one hundred historical sites and as late as
1978 was thought to occur in all counties of the state (Coddington
and Field, 1978). By 1985, however, with only four extant stations
known, it was placed on the Massachusetts list of threatened
species, a status which it holds at present (Massachusetts List of
Endangered, Threatened and Special Concern Species, 1994). The
species was listed as rare and endangered in Rhode Island in 1978
(Church and Champlin, 1978) and officially designated as state
endangered by the Rhode Island Natural Heritage Program in

209

260 Rhodora [Vol. 96

1992 (Rhode Island Natural Heritage Program, 1992). Ophio-
glossum vulgatum was listed as threatened in Connecticut in 1991
(Connecticut Department of Environmental Protection, 1991). It
has not been granted special status by the other New England
states.

This study was undertaken to investigate the ecology, repro-
ductive status and population genetics of Ophioglossum vulgatum
var. pseudopodum and its present distribution in Massachusetts.
Specific objectives included (a.) examination of all known pop-
ulations of O. vu/gatum in the state, mapping of their distribution
and estimation of population sizes; (b.) identification of associated
vascular plant species in each site; (c.) analysis of mating systems
and assessment of genetic variability for each population through
starch gel electrophoresis: (d.) reassessment of the status of O.
vulgatum in Massachusetts and identification of possible reasons
for its apparent success in some locations and its decline in others;
and (e.) development of management strategies for preserving the
existing populations.

BIOLOGY OF OPHIOGLOSSUM VULGATUM

Gross Morphology, Taxonomy and Distribution

The Ophioglossales have traditionally been considered plants
of “‘ancient lineage and extreme conservatism practically unpar-
alleled in the other fern groups” (Clausen, 1938). Their sporo-
phytes are homosporous, perennial herbs with a single, erect stem
bearing one or more small sterile leaves and a single fertile branch
with sporangia. The gametophytes are subterranean and non-
chlorophyllous (Tryon and Tryon, 1982).

Ophioglossum includes about 30 species grouped in four sub-
genera and distributed worldwide from the tropics to the Arctic
(Tryon and Tryon, 1982). Sporophytes of all Ophioglossum spe-
cies have a short, erect stem 1-12 cm long. The stem bears a
single leaf with a mature sterile lamina 1.6-25 cm or more in
length. The leaf is usually entire and narrowly elliptic, but in a
few species it may be palmately lobed or dichotomously branched.
Many species of Ophioglossum display reticulate or areolate ve-
nation. The fertile branch usually arises at or below the base of
the sterile lamina, often in the form of a spike with two rows of
large sporangia.

1994] McMaster— Ophioglossum in Massachusetts 261

Plate 1. Ophioglossum eal var. apeca (Northern Adder’s- -Tongue
Fern), natural size. Illustration by Susan Alix Williams, reproduced with the
permission of the artist. oe published in Susan Alix Williams, Wildflowers
of Rowe, 1991, = el by the Rowe Historical Society.

262 Rhodora [Vol. 96

Ophioglossum 1s characterized by extremely large chromosome
numbers. Ophioglossum reticulatum L., with n = 631, is believed
to possess the largest number of chromosomes of any organism
known. Other Ophioglossum species range from n = 120 ton =
566 (Tryon and Tryon, 1982).

Ophioglossum vulgatum L. 1s typical of its genus. The single
sterile leaf is elliptic, 2.5-10 cm long and 1-4 cm wide, with a
tapered base and reticulate venation. The stalk of the fertile spike
is 4-23 cm long with sporangia arranged in two parallel rows | .5-
5 cm long at the tip. From the erect rhizome, numerous slender
roots diverge. The gametophyte is subterranean and mycopar-
asitic (Wagner et al., 1985; Goswami, 1987). The roots of O.
vulgatum, like those of the other members of the Ophioglossales,
are hairless, and some botanists have assumed that mycorrhizae
must play an important role in the uptake and transport of water
and nutrients (Wagner et al., 1985). Khandelwal (1990) has sug-
gested a chromosome number for O. vulgatum of n = 120.

Ophioglossum vulgatum now occurs in North America, Europe,
and Asia (Clausen, 1938). This wide range has led botanists to
conclude that O. vu/gatum itself is ancient and was probably
widely dispersed in Laurasia before the separation of North Amer-
ica from Eurasia. While it normally occurs at lower elevations,
specimens of O. vulgatum in the Gray Herbarium have been
collected at 4500 feet in Nepal and at 10,000 feet in China.

Fernald (1939) proposed two North American varieties of
Ophioglossum vulgatum: var. pseudopodum, the northern adder’s-
tongue fern, and var. pycnostichum, the southeastern adder’s-
tongue fern. Ophioglossum vulgatum var. pseudopodum has a pale
green sterile blade that tapers at the base. Ophioglossum vulgatum
var. pycnostichum Fern. is distinguished by its dark green sterile
blade with a rounded base (Fernald, 1950). Wagner (1971) be-
lieved spore size was another differentiating characteristic of these
two varieties, the spores of O. vulgatum var. pseudopodum av-
eraging 20% larger than those of O. vulgatum var. pycnostichum.
McAlpin (1971) argued that the presence of a persistent sheath
in var. pycnostichum and its absence in var. pseudopodum was
an important taxonomic character for distinguishing the two va-
rieties. Lellinger (1985) argued that the two varieties are sufh-
ciently distinct and sufficiently different from O. vulgatum L. to
justify two separate North American species, O. pusillum Raf.
and O. pycnostichum (Fern.) Love & Léve. More recently Wagner
and Wagner recognized Rafinesque’s O. pusil/um as the northern

1994] McMaster— Ophioglossum in Massachusetts 263

form and included O. pycnostichum (Fern.) Love & Léve within
the Linnaean O. vulgatum (Wagner and Wagner, 1993).

The present study follows the broader and more traditional
treatment of O. vulgatum. However, all plants investigated rep-
resent Rafinesque’s O. pusillum and Fernald’s O. vulgatum var.
pseudopodum (Blake) Farw. and are referred to as O. vulgatum.

Life History

Spore Development, Dispersal and Germination

Sori of Ophioglossum vulgatum are produced on the single fer-
tile sporophyll with spores maturing in late July or early August
in New England. The fully mature sporangia dehisce to release a
cloud of yellow, dustlike haploid spores. No estimates have been
made of spore production in O. vulgatum; however, other ferns
with spores of similar size produce between 7.5 x 105 and 7.5 x
10° spores per fertile frond (Page, 1979).

Fern spores are often carried long distances by air currents.
However, an increase in atmospheric moisture may result in a
sudden, local descent of spores in rainstorms and hence peaks of
fern spore deposition may occur during the first rainfall following
a warm, dry spell (Page, 1979).

After landing, the spores of Ophioglossum vulgatum often per-
colate several centimeters through porous soil before reaching a
stable substrate (Edwards, 1982; Wagner et al., 1985). Germi-
nation may commence immediately and proceed rapidly if con-
ditions are suitable; otherwise, development may extend over one
or more years (Foster and Gifford, 1974) since prothalli are ca-
pable of overwintering (Page, 1979). The spores may also remain
dormant for extended periods (Goswami, 1987). When the gam-
etangia mature, motile sperms are released and may move several
centimeters either by mass flow of ground water or by flagellar
motion (Wagner et al., 1985).

Since the gametophytes of Ophioglossum and Botrychium lack
chloroplasts and grow beneath the soil surface, intragametophytic
selfing may be favored because it is unlikely that two spores will
land sufficiently near to one another for cross-fertilization to occur
and soil particles may impede sperm (Klekowski and Baker, 1966;
Tryon and Tryon, 1982; Peck et al., 1990).

Wagner et al. (1985), however, contend that species with sub-
terranean gametophytes may cross-fertilize as frequently as spe-

264 Rhodora [Vol. 96

cies with surficial gametophytes because of the crowding of
gametophytes that results when “spore showers” settle out. In
addition, herbivores have been observed browsing sporangial
clusters and Wagner et al. (1985) hypothesize that spores of O.
vulgatum may resist digestive processes, be deposited in feces,
and germinate in clusters (Wagner et al., 1985).

Nonetheless, isozyme electrophoresis of various Botrychium
species suggests that intragametophytic selfing predominates (Sol-
tis and Soltis, 1986; McCauley et al., 1985; Watano and Sahashi,
1992).

Vegetative Reproduction

Ophioglossum vulgatum frequently reproduces vegetatively
through sporophytes arising directly from the roots of parent plants
(Foster and Gifford, 1974). Edwards (1982) studied vegetative
growth in a population of O. vu/gatum in southern England and
traced the root system of a single plant for 1.8 m at a depth of
18-25 cm in the soil. Two clonal sporophytes were produced
within this distance and the roots may have been connected with
other more distant fronds.

Ecology of Ophioglossum yvulgatum

Ophioglossum vulgatum has been found in such diverse sites
as bogs, fens, damp sands, pastures, wet meadows, grassy swales,
moist woods, rich swamplands, mud creeks, and cedar swamps.
Occasionally it occurs on dry, sandy beaches (Clausen, 1938) or
hillsides (SCHN, B. A. Phelps and A. V. Osmun, Cornwall Bridge,
CT, 1902) and the subterranean gametophyte may be an adap-
tation to seasonal drying and/or fire (Wagner et al., 1985).

Overall, Ophioglossum vulgatum favors early successional sites
(Coddington and Field, 1978; Tryon and Tryon, 1982) and its
occurrence often appears to be related to disturbance such as the
disruption of soil by grazing herbivores.

METHODS

Herbarium Research and Field Study

In order to determine historical and/or present sites, all Mas-
sachusetts specimens of Ophioglossum vulgatum were examined

1994] McMaster— Ophioglossum in Massachusetts 265

at the Smith College Herbarium (SCHN), the Herbarium of the
University of Massachusetts-Amherst (MASS), the Gray Her-
barium (GH), the New England Botanical Club Herbarium
(NEBC), and the Yale University Herbarium (Y). In addition, the
collection records of Harry Ahles and Roberta Poland, both of
whom collected extensively in Franklin County, were examined
at MASS. The Massachusetts Natural Heritage Program of the
Massachusetts Division of Fisheries and Wildlife (MNHP) in
Boston also provided records of present and historical popula-
tions.

Since 1961, Ophioglossum vulgatum has been recorded in Mas-
sachusetts at 13 sites. Two sites were not visited due to difficulties
in gaining access, although MNHP records indicate that other
workers have recently sought O. vu/gatum at both of these sites
without success. The remaining eleven Massachusetts sites were
visited between June | and August 30, 1992.

Ateach site where Ophioglossum vulgatum was relocated, every
sporophyte was flagged and recorded on a detailed site map. The
reproductive status of the sporophyte, whether sterile or fertile,
was also recorded. Because all populations occurred in open areas
surrounded by forest, three species lists were compiled for each
site: (1) the dominant trees of the surrounding forest, (2) all vas-
cular plant species in the area where O. vulgatum occurred, and
(3) associated species, 1.e., those within | m of any specimens of
the population of O. vulgatum. Nomenclature of vascular plant
species follows Fernald (1950) unless otherwise noted.

Property owners, conservation agents and other researchers were
consulted to determine the history of each property, including
disturbances caused by beaver occupation, logging, fires, agricul-
tural practices and applications of herbicides.

Laboratory Investigations

Starch gel electrophoresis was carried out on tissue samples
from four of the five Massachusetts sites where Ophioglossum
vulgatum occurred after permission to collect this threatened spe-
cies was obtained from MNHP. Because of the very small size of
its O. vulgatum population, no live material was collected at the
Brewster site. Because of the large size of the population and the
clustered occurrence of the sporophytes at the Conway site, live
material was taken from 108 sporophytes, 12 from each of 7
clusters distributed around the site and 24 taken randomly from

266 Rhodora [Vol. 96

around the site. At each of the remaining three Massachusetts
field sites, leaf segments were collected from about 12 sporo-
phytes. For purposes of comparison, live material was also ob-
tained from a site at Mt. Sunapee, New Hampshire. Segments
ranged from 40 mg to 100 mg depending on the size of the blade.
Care was taken not to uproot or otherwise disturb plants from
which leaf segments were taken. Each leaf segment was assigned
a number and placed in a small plastic bag, its location recorded
on a map of the site.

Live material was placed on ice immediately and kept on ice
from a few hours to a maximum of five days before processing
in the laboratory. Except where noted, the protocol for electro-
phoresis of ferns by Soltis et al. (1983) was followed.

In the laboratory each specimen was weighed and placed in a
mortar chilled in ice. The phosphate grinding buffer-polyvinyl-
pyrrolidone (PVP) solution was prepared in advance except for
2-mercaptoethanol which was added immediately before grind-
ing. The solution was then added to the mortar in the ratio of .4
ml to 100 mg of live material and the mixture ground into a
homogenate. Filter paper wicks with dimensions 5mm x 7 mm
were saturated with homogenate, then placed in labeled micro-
tubules. In many instances wicks were utilized immediately. Any
wicks held longer than 4 hours were placed at —80°C until im-
mediately before use, a maximum of 15 weeks.

Starch gels with dimensions 13.25 cm x 24.75 cm x 7.5 mm
were prepared using 30 g of potato starch (SIGMA S-4501), 50
g of sucrose and 400 ml of gel buffer. Normally, the solution was
heated for about |.5 minutes after reaching the gel point, aspirated
for | minute, then poured into gel forms, wrapped in plastic wrap,
and refrigerated overnight. With gel buffer #6 (Soltis et al., 1983)
the solution was heated for 1.75 minutes, then aspirated for 1.5
minutes.

Microtubules containing wicks to be run were thawed in ice
water for approximately 15 minutes. Wicks were removed from
the microtubule and inserted in one of twenty-four wells along
the center of each gel.

Gels were run under refrigeration at approximately 4°C for from
3 to 5 hours with a current between 40 and 65 milliamperes. All
enzymes migrated anodally. Two bags of crushed ice and water
were placed on top of each gel to prevent overheating and were

1994] McMaster— Ophioglossum in Massachusetts 267

changed as needed. When a gel was finished, the power was turned
down to 10 milliamperes until ready for slicing. Three to four
slices approximately 1.9 mm thick were made from each gel.

One hundred ml of stain was placed on each gel slice which
was then kept in the dark at room temperature for from | hour
to 21 hours depending on the enzyme. Gels were scored and some
photographed. Gel slices were then labeled, wrapped in plastic
and refrigerated for future reference.

An effort was made to develop a protocol for electrophoresis
of Ophioglossum vulgatum for each of the following enzymes:
Acid phosphatase (APH, EC 3.1.3.2), Aconitase (ACON, EC
4.2.1.3), Aldolase (ALD, EC 4.1.2.13), Aspartate aminotransfer-
ase (AAT, EC 2.6.1.1), Catalase (CAT, EC 1.11.1.6), Esterase
(EST, EC 3.1.1.-), Fructose-1,6-diphosphatase (F1,6DP, EC
3.1.3.11), Glutamate dehydrogenase (GDH, EC 1.4.1.2), Glucose-
6-phosphate dehydrogenase (G6PD, EC 1.1.1.49), Glyceralde-
hyde-3-phosphate dehydrogenase (G3PD, EC 1.2.1.12), Hexo-
kinase (HK, EC 2.7.1.1), Isocitrate dehydrogenase (IDH, EC
1.1.1.42), Malate dehydrogenase (MDH, EC 1.1.1.37), Malic en-
zyme (ME, EC 1.1.1.40), 6-Phosphogluconate dehydrogenase
(6PGD, EC 1.1.1.44), Phosphoglucoisomerase (PGI, EC 5.3.1.9),
Phosphoglucomutase (PGM, EC 2.7.5.1), Shikimate dehydroge-
nase (SkKDH, EC 1.1.1.25), and Triosephosphate isomerase (TPI,
eh Oa ne Fel ee

Protocols were developed that produced observable bands for
ten enzymes and 26 loci:

. Acid phosphatase (APH-1, APH-2, APH-3)

. Esterase (EST-1, EST-2, EST-3)

Glucose-6-phosphate dehydrogenase (G6PD-1, G6PD-2,
G6PD-3)

Isocitrate dehydrogenase (IDH)

Malate dehydrogenase (MDH-1, MDH-2, MDH-3)
Malic enzyme (ME-1, ME-2, ME-3)

6-Phosphogluconate dehydrogenase (6PGD-1, 6PGD-2)
Phosphoglucoisomerase (PGI-1, PGI-2)
Phosphoglucomutase (PGM-1, PGM-2)

. Triosephosphate isomerase (TPI-1, TPI-2, TPI-3, TPI-4).

For G6PD, IDH, MDH, 6PGD and PGM, buffer system 2 was
utilized. For APH and ME, buffer system 6 was used. For EST,

SOPNDAR We

—

268 Rhodora [Vol. 96

PGI and TPI, buffer system 8 was used. When more than one
locus was observed for an enzyme, loci were numbered with the
locus migrating farthest identified as number 1.

A set of gels was run for each of the seven clusters of sporophytes
in the Conway site 1n order to assess genetic variability within
clusters. Another set of gels was then run with one or two spec-
imens from each of the seven Conway clusters to assess variability
between clusters. A third set of gels was run using twelve spo-
rophytes from around the Conway site outside of the clusters
already sampled. A fourth set of gels was run for each of the other
five populations using a Conway specimen as a standard. Finally,
a set of gels was run with representatives of each of the populations
to assess variation between populations. Included on this final
gel were specimens of a tropical species, Ophioglossum reticula-
tum, obtained from the Lyman Plant House at Smith College,
and specimens of Botrychium dissectum and B. dissectum var.
obliquum obtained from a local woodland. The same Conway
specimen was used as a standard.

RESULTS AND DISCUSSION

Ophioglossum yulgatum in Massachusetts:
Historical Status

The earliest reference to Ophioglossum vulgatum in Massachu-
setts appears in Josselyn’s New England’s Rarities (1672), where
the species is reported to occur “upon dry hill grounds [and] in
places where the water hath flood all Winter... .”

Ophioglossum vulgatum is listed in Hitchcock’s Catalogue of
Plants Growing Without Cultivation in the Vicinity of Amherst
College (1829) and in the first complete flora of Massachusetts
(Hitchcock, 1833) with a location “in the vicinity of Amherst.”’

MNHP lists 105 current or historical records for Ophioglossum
vulgatum. Collection dates range from 1848 to 1991 from 89 cities
and towns in all 14 counties (Figure 1). The five herbaria consulted
for this project contain 101 dated Massachusetts specimens from
70 cities and towns in all 14 counties. Collection dates range from
1848 to 1986. Ophioglossum vulgatum was most widely distrib-
uted in Massachusetts from 1870 to 1939 and more widely col-
lected in western Massachusetts (Table 1). Collections in eastern
Massachusetts peaked during the period 1870 to 1919. Only five

1994] McMaster— Ophioglossum in Massachusetts 269

/

a Current site

e Historic site

Figure 1. Current distribution and recorded historical sites of Ophioglossum
vulgatum by city or town in Massachusetts.

specimens have been collected east of Worcester County since
1919. No more than 14 collections exist for any decade and in
some decades O. vu/gatum has not been collected at all.

Coddington and Field (1978) considered Ophioglossum vul-
gatum widespread but rare in Massachusetts and recommended
its inclusion in a proposed list of rare and endangered plant species
for the state, noting “... few field botanists have seen it. Char-
acteristic of early successional habitats.’’ Massachusetts collec-
tions of O. vulgatum range from sea level to 1200 feet, in habitats
ranging from bogs, abandoned beaver meadows and pastures to
old fields and power line rights-of-way.

Field Sites and Population Histories

Current Sites in Massachusetts

Five populations of Ophioglossum vulgatum were relocated,
one each in Conway and Sunderland (Franklin County), Lenox
(Berkshire County), Boylston (Worcester County), and Brewster
(Barnstable County).

Conway, MA. The Conway, MA site is located northwest of
Conway State Forest in Franklin County. It occupies .96 ha at an

270 Rhodora [Vol. 96

Table 1. Dated records of Ophioglossum vulgatum from Mass. Natural Her-
itage Program; records for eastern and western Massachusetts grouped by decade.
Eastern Massachusetts includes Essex, Middlesex, Suffolk, Norfolk, Bristol, Plym-
outh, Barnstable, Dukes and Nantucket Counties. Western doates eda includes
Berkshire, Hampden, Hampshire, Franklin and Worcester Countie

Time Period Eastern Mass. Western Mass. Total
1840-1849 0 1 l
1850-1859 0 0 0
1860-1869 0 2 2
1870-1879 9 5 14
1880-1889 9 4 13
1890-1899 2 ] 3
1900-1909 5 7 12
1910-1919 10 4 14
1920-1929 2 12 14
1930-1939 0 9 9
1940-1949 0) l l
1950-1959 0) 0 0)
1960-1969 l 2
1970-1979 1 2 3
1980-1989 1 2 3
1990-present 0) 0 0
Total 40 51 9]

elevation of 450 m above sea level and slopes gently from south-
west to northeast.

Aerial photographs indicate that low shrub vegetation covered
the site in 1952 and 1958. Beavers were active in the area in the
early 1960's (C. J. Burk, pers. comm.) and by the mid-1960’s the
site was covered with water to a depth of over 1 m. The pond
was abandoned in the late 1960’s, probably as a result of trapping.
Aerial photographs taken in 1971 show that half of the site was
covered with open water and the rest with grass and sedge veg-
etation. By July, 1980, the open water occurred only in front of
the dam (Mosher, 1981). Standing water now covers much of the
site at frequent intervals during the growing season.

Dominants of the surrounding forest are typical components
of the northern hemlock-hardwood forest including Acer sac-
charum, Betula lutea, Fagus grandifolia, Pinus strobus and Tsuga
canadensis.

The current Conway population was first discovered by Nancy
D. Mosher (Nancy D. McMaster) in 1980 (Mosher, 1981) and

1994] McMaster— Ophioglossum in Massachusetts 251

Plate 2. Abandoned beaver meadow in Conway (Franklin County), MA, an
open marsh dominated by Typha /atifolia, Spiraea latifolia and S. tomentosa and
a current site for Ophioglossum vulgatum.

reported to MNHP in 1985. In July, 1980, when the vegetation
in the site was first sampled, 26 sporophytes of Ophioglossum
vulgatum were present (Mosher, 1981). Approximately 40 spo-
rophytes of O. vulgatum were found in the site in 1985 (N. Mc-
Master, 1989) and approximately 50 in 1990 (N. McMaster, 1990,
unpubl. data). By July, 1991, the population had increased sharply
to a total of 901 sporophytes (R. McMaster, 1991), and by August
and September, 1992, had declined slightly to 866 sporophytes.

Most sporophytes in 1991 occurred in clusters of 10-27 stems.
Clusters were roughly circular with a radius of about .5 m. Fre-
quently, clusters grew on patches of somewhat elevated substrate,
either around the bases of shrubs such as 4/nus rugosa, Salix spp.,
or Spiraea spp., or on submerged logs covered with bryophytes
in the wetter soils of depressions. A few clusters occurred in lower
areas, however, and some sporophytes were not clustered, with
individual stems occurring .5 m or more from the nearest member
of the population. An estimated 61% of the sporophytes were
fertile (R. McMaster, 1991).

Ophioglossum vulgatum was not found where open water or
mud was present, nor in drier areas around the margins where
trees and shrubs predominated. Its frequency decreased abruptly

212 Rhodora [Vol. 96

on the southwestern edge of the site which supported a dense
cover of sedges and grasses.

In the immediate vicinity of the Ophioglossum vulgatum pop-
ulation, associated shrubs and small trees included Acer rubrum,
Alnus rugosa, Betula populifolia, Pinus strobus, Salix discolor, S.
sericea, Spiraea latifolia and S. tomentosa. Associated herbaceous
species in the immediate vicinity included Carex lurida, Drosera
rotundifolia, Dulichium arundinaceum, Eleocharis tenuis, Epi-
lobium strictum, Equisetum arvense, Eupatorium maculatum, EF.
perfoliatum, Galium tinctorium, Glyceria canadensis, Hydrocotyle
americana, Hypericum virginicum, Impatiens capensis, Juncus
effusus, Juncus sp., Leersia oryzoides, Liparis loeselii, Lycopus
virginicus, Mentha arvensis, Muhlenbergia mexicana, Onoclea
sensibilis, Osmunda regalis, Polygonum sagittatum, Sagittaria la-
tifolia, Scirpus cyperinus, Solidago canadensis, S. graminifolia,
Sparganium americanum, Thelypteris palustris Schott, Typha la-
tifolia, and Viola pallens (McMaster, 1991).

Sunderland, MA. The Sunderland, MA (Franklin County) site
is located in an electrical transmission line right-of-way at an
elevation of 220 m. The site slopes gradually from east to west
and is seasonally wet.

The right-of-way was established in about 1912 and is main-
tained by cutting, pruning and periodic chemical treatment of any
trees or shrubs that might interfere with the operation of the
transmission lines. The site was last cut back or treated in 1989
(R. Farrell, pers. comm.).

On both sides the site adjoins northern hardwood forest dom-
inated by Acer saccharum, Betula papyrifera, B. populifolia, Carya
ovata, Pinus strobus, and Quercus rubra.

The population of Ophioglossum vulgatum was first reported
in 1986 by Bruce Sorrie of the MNHP and Robert Ruhful of the
University of Massachusetts-Amherst. A total of 80 plants of O.
vulgatum were located over a .4~-.8 hectare area.

At this site in August, 1992, the population of Ophioglossum
vulgatum had decreased to 21 sporophytes occurring throughout
a roughly rectangular area 8 m Xx 2 m. There were only 2 fertile
sporophytes. Most leaf blades were small with atypically brown
or yellow patches. Associated shrubs and small trees in the 1m-
mediate vicinity included Acer rubrum, Betula lenta, Cornus ra-
cemosa, Hamamelis virginiana, Lindera benzoin, Lyonia ligus-
trina, Rubus sp. and Salix sp. Associated herbaceous species

1994] McMaster— Ophioglossum in Massachusetts 273

included Achillea millefolium, Carex scoparia, C. stipata, Cirsium
arvense, Equisetum sylvaticum, Fragaria virginiana, Galium as-
prellum, Geranium maculatum, Hypericum ellipticum, Juncus ef-
fusus, Onoclea sensibilis, Osmunda cinnamomea, Platanthera fla-
va (L.) Lindl., Polygonum sagittatum, Prunella vulgaris, Solidago
graminifolia, Thelypteris palustris, and Viola spp.

Lenox, MA. The Lenox, MA (Berkshire County) site is located
on the edge of an open meadow at an elevation of 370 m. The
seasonally wet site slopes gently from west to east and has been
described as a “sloping graminoid fen’ (P. Weatherbee, pers.
comm.).

The area was used for pasture until 1939 and aerial photographs
show it has been kept open, probably by mowing or haying, since.
During the 1980’s a stream to the west was dammed by beavers,
causing surface water to flow across the site, which was last mowed
in 1989 (P. Weatherbee, pers. comm.).

The forest surrounding the meadow is dominated by Acer rub-
rum, A. saccharum, Carya ovata, Malus sp., Pinus strobus and
Populus tremuloides with an understory of Betula papyrifera,
Fraxinus americana, Prunus virginiana, and Viburnum recogni-
tum. Celastrus orbiculatus and Vitis spp. frequently overtop the
canopy to the west. 4/nus sp. dominates the wetter eastern bound-
ary of the meadow.

The population of Ophioglossum vulgatum was discovered and
reported to MNHP by Pamela B. Weatherbee in 1991. A total of
205 sporophytes occurred in a roughly rectangular area measuring
11m Xx 26m. Twenty-two percent of the sporophytes were fertile.

Within this area in August, 1992, the population of Ophio-
glossum vulgatum had declined to 62 sporophytes. Approximately
33% of these were fertile. Most were growing beneath low shrubs,
either Cornus stolonifera, Salix serissima or Salix discolor. Other
associated species in the shrub layer included Potentilla fruticosa,
Viburnum recognitum, Rhamnus cathartica, Salix serissima, S.
discolor and Vitis sp. Herbaceous species included Achillea mil-
lefolium, Carex stricta var. strictior, Equisetum arvense, Eupa-
torium maculatum, Festuca sp., Fragaria virginiana, Galium tri-
Hlorum, Geum rivale, Juncus canadensis, J. dudleyi, Lysimachia
ciliata, Onoclea sensibilis, Parthenocissus quinquefolia, Phleum
pratense, Prunella vulgaris, Rudbeckia hirta, Scirpus atrovirens,
Spiraea latifolia, Thelypteris palustris, Trifolium repens, T. agrar-
ium, and Valeriana officinalis.

274 Rhodora [Vol. 96

. Two sporophytes of Ophioglossum vulgatum, one with sporophyll, in

Plate 3
Lenox (Berkshire County), MA

1994] McMaster— Ophioglossum in Massachusetts 215

Boylston, MA. The Boylston, MA (Worcester County) site is
located at the eastern edge of a meadow which slopes from west
to east at an elevation of 170 m. Occasionally, shallow standing
water occurs on the lowest part of the meadow about 50 m north
of the site (J. Wright, pers. comm.).

The site was used for pasture before 1986. The meadow was
cut for hay in 1986 and 1987 and mowed once a year in October
in 1988, 1989, 1990 and 1992.

The adjacent forest includes a canopy of Acer rubrum, Carya
sp., Fraxinus americana, Malus sp., Pinus strobus, Prunus sero-
tina and Quercus velutina and an understory of Prunus virginiana
and Rhamnus frangula.

Shrubs and small trees found in the adjacent upland meadow
include Acer rubrum, Cornus amomum, Prunus serotina, Rham-
nus frangula, Salix bebbiana, Spiraea latifolia, S. tomentosa, and
Viburnum recognitum. Herbaceous species of the upland meadow
include Daucus carota, Eupatorium maculatum, Galium tincto-
rium, Onoclea sensibilis, Rhus radicans, Rosa nitida, Solidago
canadensis, S. rugosa, Trifolium hybridum, T. pratense, Vicia
cracea and Vitis labrusca.

The “very vigorous” population of Ophioglossum vulgatum
consisted of 30-40 plants when first reported in 1989 by Marc
Larocque. When the population was surveyed in August, 1992,
it had increased to 109 sporophytes located in a level area about
12m x 20 m. However, only 18% of the sporophytes were fertile.
Many sporophytes occurred beneath shrubs of Cornus amomum
and Salix spp. between .3 m and 1.0 m high. Associated shrubs
and tree seedlings included Acer rubrum, Alnus rugosa, Betula
populifolia, Cornus amomum, Malus sp., Prunus serotina, Quer-
cus rubra, Rhamnus frangula, Rhus radicans, Rosa nitida, Spiraea
latifolia, S. tomentosa, Vaccinium corymbosum, and Viburnum
recognitum. Herbaceous species included Achillea millefolium,
Agrostis stolonifera, Asclepias syriaca, Asparagus officinalis, Carex
stricta, Daucus carota, Galium tinctorium, Onoclea sensibilis, Os-
munda regalis, Oxalis montana, Solidago canadensis, S. gigantea,
S. rugosa, Trifolium hybrida, T. pratense, Rubus sp., and Vicia
cracca.

Brewster, MA. The Brewster, MA (Barnstable County) site is
located in a small clearing adjacent to a pond at an elevation of
8 m. The site is seasonally wet and was probably kept clear by

216 Rhodora [Vol. 96

periodic mowing until 1985. There has been no clearing since
1985 (R. Finch, pers. comm.).

The adjacent woodland is dominated by Acer rubrum and Nyssa
sylvatica.

The population of Ophioglossum vulgatum included 19 plants
when first reported by Richard LeBlond in 1985. Mario Di-
Gregorio, then Conservation Agent for the Town of Brewster,
observed the population from 1986 to 1989 and reported finding
six sterile plants and four to five fertile plants each year. He had
been unable to relocate any plants since 1989.

In July, 1992, only a single sterile sporophyte was located.
Associated shrubs and small trees included Acer rubrum, Leu-
cothoé racemosa, Lyonia ligustrina, Nyssa sylvatica, Rhododen-
dron viscosum, Vaccinium corymbosum and V. macrocarpon.
Herbaceous species included Maianthemum canadense, Rubus
hispidus, and Trientalis borealis. Extensive areas were covered by
Sphagnum spp. and other mosses.

Recent Historical Sites in Massachusetts

Six historical Massachusetts sites were visited but no popula-
tion of Ophioglossum vulgatum found. These occurred at Lanes-
boro (Berkshire County), Rowe and Gill (Franklin County),
Northampton (Hampshire County), Plymouth (Plymouth Coun-
ty), and Falmouth (Barnstable County).

Lanesboro, MA. The Lanesboro, MA population of Ophio-
glossum vulgatum was located by Pamela B. Weatherbee in 1984
when 15 sterile and 3 fertile sporophytes were found at the eastern
base of an outcrop of rock in an old field, an area underlain by
limestone. No sporophytes have been found on the site since 1984
(P. Weatherbee, pers. comm.).

Rowe, MA. The Rowe, MA population was located by Nancy
Williams and Susan Williams in 1983 when approximately 150
sporophytes were found in a “‘borrow pit.” In 1973 sand and
gravel were removed from the site which was subsequently col-
onized by herbaceous vegetation, small shrubs and tree seedlings.
The population of Ophioglossum vulgatum was seen yearly from
1983 to 1988 but not since in spite of repeated observations of
the area (N. Williams, pers. comm.).

Gill, MA. The Gill, MA population was found in 1961 ina
“marshy field.’ I was unable to relocate the population in June,

1994] McMaster— Ophioglossum in Massachusetts pA

1992. The wet meadow apparently had been disturbed recently
by mowing and excavation.

Northampton, MA. The Northampton, MA population was
located by Harry Ahles in 1978 in what was described as a wet
hay field. Bruce Sorrie searched unsuccessfully for the population
in 1986. I was unable to relocate the population in July, 1992.

Plymouth, MA. The Plymouth, MA population was located
by Bruce Sorrie in 1983 in an acid seep. I was unable to relocate
the population in June, 1992.

Falmouth, MA. The Falmouth, MA population was located
by H. K. Svenson in 1971 in the bottom of a glacial kettle near
the town landfill. The site was apparently destroyed in the ex-
pansion of the landfill and my efforts to relocate the population
in June, 1991 were unsuccessful.

1993 Observations

All four western Massachusetts populations of Ophioglossum
vulgatum that were extant in 1992 were monitored during the
hot, dry summer of 1993. The Conway population was thriving
throughout the summer of 1993. Stem counts in eight permanent
plots showed a 64% increase over similar counts made in 1991.
The Sunderland population had dwindled to one small, withered
blade when observed on August 5, 1993. No sporophytes of the
Lenox population were in evidence on July 24, 1993. Some spo-
rophytes of the Boylston population were observed to be in good
condition in early July, 1993, but had died back by August 18,
1993 (J. Wright, pers. comm.).

Laboratory Investigations

A total of 157 individuals of Ophioglossum vulgatum was as-
sayed at 26 loci (Table 2). All individuals in all populations were
found to be homozygous and monomorphic for all observable
loci.

Additional confirmation of the observed monomorphism among
the specimens assayed was obtained by electrophorescing material
from all five populations of Ophioglossum vulgatum (four from
Massachusetts, one from New Hampshire) with material of O.
reticulatum, Botrychium dissectum and B. dissectum var. obli-
quum on a single gel. Results for the three New Hampshire spec-

278 Rhodora [Vol. 96

Table 2. Total sample size and number of specimens for which satisfactory
band resolution was achieved at each Pee in each site, , CON Way: pero ane
LENox, BOYIston and MT. Sunapee. (—)

Locus CON SUN LEN BOY MTS TOTAL
Total sample size 108 13 11 13 12 157
APH-1 59 —_ — — = 59
APH-2 33 — — _ — 33
APH-3 24 — _ — _ 24
EST-1 16 — _ — — 16
EST-2 12 — _ _ = 12
EST-3 12 _ — — — 12
G6PD-1 59 7 7 11 l 85
G6PD-2 52 — — — — 52
G6PD-3 16 _ — — 16
IDH 1S 13 11 | 12 115
MDH-1 85 13 11 11 128
MDH-2 79 12 11 11 3 116
MDH-3 21 - — — — 21
ME-1 85 13 5 | — 104
ME-2 38 13 5 — 57
ME-3 1 1] 5 3 - 10
PGI-1 93 13 11 11 12 140
PGI-2 69 13 11 10 12 115
PGM-|1 77 13 11 13 8 122
PGM-2 29 — _ — —_ 29
6PGD-1 72 — 3 — 75
6PGD-2 50 — = a = 50
TPI-1 39 1 11 1 7 59
TPI-2 22 ] 1] 2 7 43
TPI-3 19 l 7 — 7 34
TPI-4 6 _ — — 7 13

imens of O. vulgatum were inconclusive and warrant additional
study. Bandings for O. reticulatum, B. dissectum and B. dissectum
var. obliquum were distinct from each other and from O. vul-
gatum.

The number of loci assayed represents only a small portion of
the total genome of Ophioglossum vulgatum. The significance of
complete monomorphism for these loci, however, 1s enhanced by
the fact that the same 26 loci found to be monomorphic for
specimens of O. vu/gatum have been widely assayed for other
homosporous ferns and frequently have been observed to be poly-
morphic (McCauley et al., 1985; Soltis and Soltis, 1986; Soltis
and Soltis, 1988; Watano and Sahashi, 1992). No other published
data on genetic variation in Ophioglossum vulgatum was found.

1994] McMaster— Ophioglossum in Massachusetts 279

Lack of genetic variability often has been cited as a factor in
the decline in population size and, in some cases, in the extinction
of rare plant species (Frankel and Soulé, 1981; Simberloff, 1986;
Barrett and Kohn, 1991; Huenneke, 1991). Data from the elec-
trophoresis of specimens from four of the Massachusetts popu-
lations of Ophioglossum vulgatum show a complete absence of
genetic variability at all loci assayed. Any explanation for this
observation must account for three phenomena: (1) the apparent
lack of genetic variability within sporophytes; (2) the apparent
lack of genetic variability among individuals within each popu-
lation; and (3) the apparent lack of genetic variability among
populations.

If intragametophytic selfing occurs, all heterozygosity will be
lost in one generation but there still may be some differentiation
within a population. Reductions in genetic variability may also
occur as a result of small effective population size. Decline in
effective population size may result when few individuals repro-
duce, when the number of progeny produced by individuals varies
widely, and/or when a population bottleneck, an extreme reduc-
tion in effective population size, occurs resulting in the trans-
mission of only a small subset of the total genetic variability of
the population to the next generation (Frankel and Soulé, 1981).
A population with a small effective population size is also sus-
ceptible to further loss of alleles due to genetic drift. Edwards
(1982) observed a population of Ophioglossum vulgatum in south-
ern England, for example, in which only 1-2% of the sporophytes
bore fertile fronds. Similarly, in the Sunderland population less
than 10% of the sporophytes found in the summer of 1992 had
fertile fronds. In addition, populations of clonal species such as
O. vulgatum may contain numerous sporophytes of a single ge-
notype.

New populations established by dispersal of spores from a ge-
netically monomorphic population will normally be of the same
homozygous genotype. If the species is capable of rapid vegetative
growth, it is possible for a number of large, homozygous, mono-
morphic populations to develop in a relatively short period of
time.

One adaptive compensation for the loss of genetic variation in
a population may be the purging of deleterious recessive alleles
through repeated episodes of in-breeding (Frankel and Soulé, 1981;
Lesica and Allendorf, 1992). By this mechanism, a predominantly
self-fertilizing species such as Ophioglossum vulgatum character-

280 Rhodora [Vol. 96

ized by low genetic variability may have evolved a complex of
alleles co-adapted to a very specific set of environmental condi-
tions. While it is unable to adapt successfully to changes in those
conditions, it is suited to sites where those conditions prevail.

The four western Massachusetts populations may have arisen
from a single monomorphic population. Gene flow from geneti-
cally distinct populations apparently has not occurred or has oc-
curred at sufficiently low frequencies that any variability intro-
duced has been lost through repeated intragametophytic selfing,
population bottlenecks and drift. Electrophoretic analysis of pop-
ulations of Ophioglossum yvulgatum in neighboring states could
shed additional light on these questions.

CONCLUSIONS AND MANAGEMENT RECOMMENDATIONS

The five current Massachusetts sites for Ophioglossum vulga-
tum vary significantly in geography, climate, soil and surrounding
forest type, but share three important characteristics:

(1) Each site has a history of disturbance that has reduced
competition from shrubs and small trees. The Conway site was
occupied by beavers. The Sunderland site has been subject to
periodic pruning or application of herbicide for maintenance of
the power line right-of-way. The Lenox site was occupied by
beavers and has been mowed occasionally. The Boylston site was
formerly pasture and has been mowed or hayed regularly for many
years. The Brewster site was probably maintained irregularly as
a lawn.

(2) Each site is characterized by wet conditions for at least a
part of the year. The Conway site is saturated throughout the
growing season with standing water at frequent intervals. The
Sunderland, Lenox, Boylston, and Brewster sites are subject to
seasonal flooding, usually in spring and fall.

(3) Each site supports a characteristic flora. Although no single
plant species occurs with Ophioglossum vulgatum in all five sites,
Acer rubrum and Onoclea sensibilis are abundant in four of the
sites and Achillea millefolium, Spiraea latifolia and Thelypteris
palustris, are abundant in three of the sites. Sporophytes of O.
vulgatum often are found growing in the shade of small shrubs,
usually A/nus spp., Cornus spp., Salix spp., or Spiraea spp. In
the Brewster site the single O. vu/gatum sporophyte occurred
beneath Leucothoé racemosa.

1994] McMaster— Ophioglossum in Massachusetts 281

Observation of populations over several growing seasons sug-
gest that where the shrub layer is completely lacking, aggressive
grass, sedge and rush species predominate and Ophioglossum vul-
gatum declines. The shrub layer, when present, may help to con-
trol competition from those species. Ophioglossum vulgatum seems
to prefer sites where the shrub layer 1s present but constrained by
pruning or periodic flooding.

Populations of Ophioglossum vulgatum are capable of sudden
and dramatic increases or declines. The current Conway site ap-
pears to have increased from 26 sporophytes in 1980 to a max-
imum of 901 in 1991, the Lenox site from 0 to 205 in under 12
years, and the Boylston site from 30 or 40 to 109 in three years.
Conversely, the Rowe population of approximately 150 sporo-
phytes disappeared completely after the 1988 growing season and
the Brewster population of 19 in 1985 had declined to only one
in 1992. Five additional more gradual extinctions have been doc-
umented in the state since 1961.

The low levels of genetic variability revealed by electrophoresis
may be important factors in the decline of Ophioglossum vulga-
tum in Massachusetts. A combination of intragametophytic self-
ing, small population size, low numbers of reproductively mature
sporophytes, founder effect and specialized habitat requirements
may be responsible for the loss of genetic variability within and
among populations. These processes may in turn lead to an in-
ability to adapt to environmental change in an inherently volatile
habitat. While O. vulgatum seems to be well adapted to compete
in early successional sites in Massachusetts, it may lack the genetic
heterogeneity that would allow it to persist as these sites change.
Only at the Conway site where succession is inhibited by inter-
mittent flooding does O. vu/lgatum appear to thrive. Hence its
long-term survival in Massachusetts may be a function not of
adaptation to changing conditions in a site, but of its ability to
disperse to new, sometimes distant sites and to found new pop-
ulations successfully.

The habitat requirements and reproductive biology of Ophio-
glossum vulgatum described suggest a number of measures for
the management of existing populations.

(1) Plant species competing with Ophioglossum vulgatum should
be controlled. Shrubs and small trees in the immediate vicinity
should be selectively pruned to provide moderate levels of filtered
light. The surrounding forest canopy may also require periodic

2) Rhodora [Vol. 96

thinning. Because the effects of herbicides on O. vulgatum are not
known, their use to suppress surrounding vegetation is not rec-
ommended. However, most populations of O. vu/gatum are small
enough that competing vegetation can be controlled by mechan-
ical means such as a portable ““weed-wacker” or even hand clip-
pers. The invasion of grasses, sedges and rushes may be restricted
by mowing late in the growing season, probably after October 1.
Care should be taken to protect shrubs and small trees from
damage during mowing.

(2) Disruption or compaction of soil should be avoided. Heavy
mowing equipment should not be used to clear Ophioglossum
vulgatum sites. Footpaths and vehicle rights-of-way should be
diverted to provide a generous buffer zone around protected pop-
ulations.

(3) Proper hydrological conditions in the immediate vicinity
of the Ophioglossum vulgatum population should be maintained.
Extended inundation or desiccation of the substrate during the
growing season which may result from beaver activity or anthro-
pogenic activities such as draining and filling often associated
with agriculture and highway construction should be avoided.

The size of an Ophioglossum vulgatum population may fluc-
tuate dramatically from year to year depending on availability of
light, soil moisture, disturbance and other conditions. Because of
the extensive root network and the ability of a population to
expand vegetatively, it may also be important to protect and
manage a site even after all sporophytes have disappeared. It is
possible that a subsequent reversal of environmental conditions
could result in the development of new sporophytes from the
dormant root system. Germination of spores or gametophytes
may also occur after an extended period of dormancy.

Habitat for Ophioglossum vulgatum also may be provided in
sites where it does not occur at present, particularly when those
sites are near existing populations. Minimum requisite conditions
should include a saturated substrate and moderate to high light
levels. Sites where such conditions occur naturally may include
seeps, wet meadows and abandoned beaver meadows. Areas sub-
ject to some human disturbance also may be appropriate including
highway shoulders or median strips, golf courses, utility corridors,
or pastures.

While historical data suggest that Ophioglossum vulgatum has
never been abundant in Massachusetts, its present status is pre-

1994] McMaster— Ophioglossum in Massachusetts 283

carious. Of the five known populations, the Brewster population
with its single sporophyte is probably destined for extinction,
while the Boylston, Lenox and Sunderland populations are mar-
ginal. The Conway population, while relatively large and healthy
in 1991 and 1992, could easily be subject to the kind of sudden
decline that has been observed at other sites.

Continued availability of disturbed, early successional sites may
be important to the survival of Ophioglossum vulgatum in Mas-
sachusetts. These sites may result from naturally occurring dis-
turbances such as wildfires, windstorms, debris slides and insect
predation. Some anthropogenic factors such as highway and dam
construction may favor the creation of disturbed sites while oth-
ers, such as the decline of agriculture and suppression of wildfires,
may result in less disturbance. While the reintroduction of beavers
into the state in the 1930’s created new early successional sites,
the subsequent expansion of beaver populations may have had
the opposite effect, forcing beavers to reoccupy abandoned sites
before early successional species were able to colonize them. Thus
human activity can be either beneficial or harmful to the survival
of O. vulgatum in Massachusetts.

Efforts should be made to locate additional populations of
Ophioglossum vulgatum in Massachusetts. Hagenah (1966) writes
“ Ophioglossum vulgatum] blends with the grasses and other veg-
etation, but after the first one is noticed search may reveal scores
or even hundreds more.” Similarly, Wagner (1971) observes that
“Many botanists and naturalists regard the adder’s-tongue as a
rare plant, but this is probably due largely to the fact that it 1s
overlooked.’ Roberta Poland (Mass) located at least 11 stations
for O. vulgatum in the town of Deerfield alone during the 1950’s,
an achievement which should inspire continued persistence in the
search for this ancient and intriguing plant.

ACKNOWLEDGMENTS

My original interest in Ophioglossum vulgatum resulted from
field trips to the Conway site with my wife, Nancy D. McMaster,
to whose memory this paper is dedicated. I also wish to express
my appreciation to C. John Burk, my thesis advisor, for his guid-
ance and support in the completion of this project, to Stephen G.
Tilley for his advice and assistance with starch gel electrophoresis,
and to Alan H. Bornbusch for his advice and review of this manu-

284 Rhodora [Vol. 96

script. The artwork of Susan Alix Williams and the photographic
services of Richard M. Fish are also gratefully acknowledged. This
project was supported by research grants from the Margaret A.
Walsh Grantham Research Fellowship, Department of Biological
Sciences, Smith College, the Massachusetts Natural Heritage Pro-
gram and the Connecticut River Watershed Council.

LITERATURE CITED

Barrett, S.C. H.ANDJ.R. KOHN. 1991. Genetic and evolutionary consequences

. small population size in plants: implications for conservation, pp. 3-30.
n: D. A. Falk and K. E. Holsinger, Eds., Genetics and Conservation of Rare

an Oxford Univ. Press, New York

CHurcH, G. L. AND R. L. CHAMPLIN. 1978. Rare and Endangered Vascular Plant
Species in Rhode Island. The New England Botanical Club in cooperation
with the U.S. Fish and Wildlife Service, Cambridge, MA.

CLAUSEN, R. T. 1938. ae on the OuhiGoc eee. Mem. of the Torrey
Bot. Club 19: 1-17

Coppincton, J. A a G. Fietp. 1978. Rare and Endangered Vascular Plant
Species in ee The New England Botanical Club in cooperation
with the U.S. Fish and Wildlife Service, Cambridge, MA

CONNECTICUT DEPARTMENT OF ENVIRONMENTAL PROTECTION. 1991, Regulation
of Department of Environmental Protection Concerning Endangered and
Threatened Species and Species of Special Concern. Connecticut Department

Environmental Protection, Hartford, CT

Epwarps, P. 1982. Root connections in a a colony of Ophioglossum vulgatum in
southern Be ces Fern Gazette 12: 241-242.

FERNALD, M.L. 1939. Last survivors in the flora of Tidewater Virginia. Rhodora
41(490): pee 504, 529-560

1950. Gray’s Manual of Botany. American Book Co., New York.

Foster, A. S. AND E. M. GIFForpD. 1974. — nee of Vascular
Plants. W. H. ma and Company. San Frar
FRANKEL, O. H. AND M. SouLeé. 1981. Population oe s and conservation,

pp. 31-77. iO. H. Frankel and M. Soulé, Eds., Conservation and Evolution.
Cambridge University Press, Cambridge, Englan
Goswami, H. K. 1987. Ophioglossales I: an overview. eon re 7: 47-89.
HAGENAH, D. J. 1966. Notes on ee pteridophytes, II: distribution of the
phioglossaceae. Amer. Fern J. 56: 150-163.

Hircucock, E. 1829. Catalogue of ee late Without Cultivation in the
Vicinity of Amherst College. Adams a Amherst, MA.

33. Report on the Geology. Minera Botany and Zoology of Mas-
sachusetts. Adams and Co., Amherst, MA.

HUENNEKE, L. F. 1991. rea ions of genetic variation in plant
populations, pp. 31-44. /n: D. A. Falk and K. E. Holsinger, Eds., Genetics
and Conservation of see aa Oxford Univ. Press, New York.

JOssELYN, J. 1672. New-England’s Rarities. G. Widdowes, London. Reprint,
1865, William Veazie, Boston.

1994] McMaster— Ophioglossum in Massachusetts 285

KHANDELWAL, S. 1990. Chromosome evolution in the genus Ophioglossum L.
Bot. J. Linn. Soc. 102: 205-217

KLekowskI, E. J. AND H. G. BAKER. 1966. as i significance of poly-
ploidy in the Pteridophyta. Science 153: 305-

LELLINGER, D. B. 1985. A Field Manual of the ee and Fern- — of the
United States and Canada. Smithsonian Institution Press, Washington, D.C.

Lesica, P., AND F. W. ALLENDORF. 1992. Are small populations of oe work
preserving? Cons. Biol. 6: 135-139.

NATURAL HERITAGE AND ENDANGERED SPECIES PROGRAM. 1994. Massachusetts
List of Endangered, Threatened and Special Concern Species. Natural Her-
itage and Endangered Species Program, Boston

McALPIn, B. 1971. Ophioglossum leaf sheaths: development and morphological
nature. Bull. Torrey Bot. Club 98: 194-199.

McCAauLey, D. E., D. P. WHITTIER AND L. M. Rertty. 1985. Inbreeding and the
rate of self- aaa in a grape fern, Botrychium dissectum. Amer. J. Bot.
72: 1978-1981

McMaster, N. D. (MosHER). 1989. The floristics and synecology of fifteen
abandoned beaver assent in Western Massachusetts. Ph.D. dissertation,
Smith College, Northampto

McMaster, R.T. 1991. Asurvey ae Drees vulgatum in Conway (Frank-
lin County), Massachusetts. Unpublished report, Mass. Natural Heritage
Program, Boston,

MosHer, N. D. 1981. Floristics and successional development of three aban-
doned beaver Svea in Western Massachusetts. M.A. thesis, Smith Col-
lege, Northampto

Pace, C. N. 1979. eal aspects of fern ecology, pp. 551-589. In: A. F.
Dyer, Ed., Experimental Biology of Ferns. Academic Press, London.

Peck, J. H., C. J. Peck AND D. R. FARRAR. 1990. Influences of life history
attributes on formation of local and distant fern populations. Amer. Fern J.
80: 126-142

Ruope IsLAND NATURAL HERITAGE PROGRAM. 1991. Rare Native Plants of
Rhode Island. Rhode Island Department of Environmental Management,
Providence, RI.

sai are D. 1986. The proximate causes of extinction, pp. 259-276. In: D.

p and D. Jablonski, Eds., Patterns and Processes in the History of
Life. Spinsee Verlag, Berlin

Sortis, D. E., C. H. HAUFLER, D. C.D ARROW, AND G. J. GAstony. 1983. Starch
gel electrophoresis of ferns: a compilation of grinding buffers, gel and elec-
trode buffers, and staining schedules. Amer. Fern J. 73: 9-27

bp P. S. Sottis. 1986. Electrophoretic evidence for inbreeding in the

fern pate virginianum (Ophioglossaceae). Amer. J. Bot. 73: 588-592.

Soxtis, P. S. AND D. E. Sottis. 1988. Genetic variation and population structure
in the fern Blechnum spicant (Blechnaceae) from western North America.
Amer. J. Bot. 75: 37-44.

Tryon, R. M. AND A. F. Tryon. 1982. Ferns and aie Plants with Special
Reference to Tropical America. Springer-Verlag, New k

Wacner, W. H., Jr. 1971. The Southeastern Adder’s- ieee Ophioglossum
vulgatum var. pycnostichum, found for the first time in Michigan. The Mich-
igan Bot. 10: 67-7

286 Rhodora [Vol. 96

AND F. S. WAGNER. 1993. Ophioglossaceae, pp. 85-106. Jn: Flora of
North America North of Mexico. New York: Oxford University Press.
——., F. S. WAGNER AND J. M. BEITEL. 1985. Evidence for interspecific hy-
bridisation in pteridophytes with subterranean mycoparasitic gametophytes.
P Edinb. 86B: 273-281

WaTANO, Y. AND N. SAHASHI. 1992. Predominant inbreeding and its genet
consequences in a homosporous fern genus, Sceptridium Opie
Syst. Bot. 17: 486-502.

DEPARTMENT OF BIOLOGICAL SCIENCES
SMITH COLLEGE
NORTHAMPTON, MA 01060

RHODORA, Vol. 96, No. 887, pp. 287-293, 1994

A NEW TRI-HYBRID LYCOPOD,
DIPHASIASTRUM DIGITATUM x SABINIFOLIUM

ARTHUR V. GILMAN

ABSTRACT

A putative tri-specific hybrid Diphasiastrum (Lycopodiaceae) is described from
a collection in northern Vermont and is figured in a line drawing. The hybrid is
morphologically intermediate between its probable parents, D. digitatum and D.
x sabinifolium. This is the first flat-branched hybrid lycopod believed to involve
three parental genomes. In light of its limited occurrence (a single clone), it is not
given an epithet.

Key Words: Diphasiastrum (Lycopodium), hybrid, genome analysis, species con-
cept, Vermont

INTRODUCTION

Five species, and at least five hybrids, of the ‘‘flat-branched”
lycopods (Diphasiastrum Holub) are known in northeastern North
America (Table 1). The taxonomy of this group has to date relied
almost entirely on morphology. Hybrids generally display char-
acters intermediate between their parents, and are usually distin-
guishable in the field and herbarium on morphological grounds
(Wilce, 1965). Cytological work has indicated that all North
American species and hybrids share a chromosome number 2n
= 46 (F. S. Wagner, 1992), with apparent normal pairing behavior
and spore formation in the hybrids (the plant described here
produces well-formed spores). Additionally, the difficulty of ger-
minating spores i” vitro and growing the resulting plants has so
far precluded artificial crossing experiments in Diph trum,
and the genus as a whole has not been analyzed on 1 the basis of
isozymes or DNA. On morphological grounds, Wilce (1965) de-
termined D. xsabinifolium, which had long been considered a
species, to be the hybrid D. sitchense x tristachyum, a treatment
which has gained acceptance (Beitel, 1979; Cody and Britton,
1989; Wagner and Beitel, 1993). Except for D. x issleri, which is
known in our area from a single collection, and D. complanatum
x digitatum, which may be quite rare, the hybrids are sometimes
locally common (Cody and Britton, 1989; W. H. Wagner, pers.
comm.), and D. x sabinifolium sufficiently so to be regarded as a
nothospecies. The hybrids appear to be fertile (F. S. Wagner,
1992), yet no hybrid progeny or backcrosses with parent species

287

288 Rhodora [Vol. 96

able 1. Species and hybrids of Diphiastrum in northeastern North America
(all 2n = 46).

pecies

. digitatum (Dill.) Holub
. complanatum (L.) Holub
tristachyum (Pursh) Holub
sitchense (Rupr.) Holub
. alpinum (L.) Holub

Hybrids
D. x sabinifolium (Willd.) Holub (sitchense « tristachyum)
D. x habereri (House) Holub (digitatum * tristachyum)
D. complanatum x digitatum
D. x zeilleri (Rouy) Holub (complanatum x tristachyum)
D. xissleri (Rouy) Holub (alpinum x tristachyum)

boosd

have been described. This paper discusses a new hybrid, D. dig-
itatum x sabinifolium, at present known from a single station in
northern Vermont. It is the first hybrid known to have a hybrid
parent.

HYBRID DESCRIPTION AND ANALYSIS

While collecting a series of pteridophytes in Caledonia County,
Vermont, I discovered, in a montane (ca. 540 m) east-facing
pasture, an unusual clone of Diphasiastrum which was not readily
referable to any described taxon. It was growing with D. digitatum
and D. tristachyum but was clearly neither. Collections were made
in August and November, 1993, from the same clone.

This plant has the general aspect of D. x sabinifolium with a
congested, tufted habit rather than a pseudo-monopodial habit
such as characterizes D. digitatum and D. tristrachyum (Figure
|). This is due to the branching of upright stems near their bases
and to the length of the lower branches which reach beyond the
bases of upper branches. The plant is also similar to D. x sabi-
nifolium in its short, usually unbranched peduncle with one or
two sporophylls on the peduncle below the strobilus. A hybrid
origin, however, was suspected because the lateral branchlets are
strongly dorsiventrally flattened, the lateral leaves are appressed
rather than spreading, and the lower (ventral) leaves are reduced,
all characteristics that do not occur in D. x sabinifolium. Fur-

1994] Gilman— New Tri-Hybrid Lycopod

(\ ()
mh bY
a 1A
mY
A Oia O50
' TAs Oy
Kai my MY
i vay Vy
YY a Wy
1A) 2) v
(0) wy |
Vy] ty)
i) y \)
Va \ |
nd () |
Vy y
'Y wy |
‘om
nt x \
|
VY a \ df
\ iy
at YR Y
‘ Se YY iy
Ny \ YY LD
AW
i \, X y \)
rh Wy \y
WANS Nt} ANN
Wy SQ. NW ANN
Y NA " 4 \ iy
wx ant /, } -
SR a KR Hh Pi
SN ~ Wy i ‘ is WH DEES
SAN Ka BY LIES
OQ A WY Y
YT ERY A
EN 2a IANS
RQ “ UN Sea) NY
SAN YS; Ni! Mh YES Ry g
Sh DR Ny OSS
MA LA, SS Wy WB WA
m aN SO WY PONY
\\ DR PN Le
AN ONY Wipe ts
\ SY. A Ne Wp? y
X Rs ee
NN SSS TRY ae GLEE rs
N = Ay SH AY Mi a AP <<
BN NN HPRY
Skye RY “A
SSSSSS |
eae

a e — ——\~

SS Tom

Figure 1. Diphasiastrum digitatum x ae The influence of D.
binifolium is indicated by the low branc the indeterminate growth ae
hown by annual constrictions, a

and eae peduncle with sporophylls
scattered at the base of the strobilus ee a The broad, flattened branches,

reduced lower leaves and broad sporophylls are evidence of D. digitatum par-
entage.

thermore, the plant overall is larger and stouter than typical D
x sabinifolium.

A morphological analysis was undertaken to determine the
identity of this puzzling plant. With such tests of hybridity as
garden experiments, isozyme or DNA analysis lacking at present

290 Rhodora [Vol. 96

able 2. Hybrid index comparing ten oo characters of the eg
hybrid D. digitatum = sabinifolium and its parents. (2 = character as in D. xs
binifolium, 1 = intermediate; 0 = character as in D. digitatum).

Character (from Wilce, 1965) sab dig x sab dig

. Habit congested due to branching low
on upright stem, vs. pseudomonopo-
dia

to

. Branches indeterminate aie 2 or more
years, as show y anniial canctrictinne 2 2 0)

3. Branches stioktly an VS. aetiee
green l 0

4. Branches not or only slightly flattened
=i Ren 2 0 0)
D5 r leaves not reduced, vs. reduced 2 l 0)
6. whe leaves spreading vs. appressed 2 0 0)

7. Lateral leaves with narrow base, VS.
broad base 2 0 0)

8. Fertile branch round with equal leaves
(vs. flattened with reduced lower leaves) 2 2 0

9. Peduncle usually unbranched (vs. nor-
mally branched) 2 2 0
10. Sporophyll narrow (vs. broad) 2 0 0)
Total 20 9 0

therefore, morphological comparisons give the strongest evidence
of this plant’s hybrid status.

A series of hybrid indices (modified from Wilce, op. cit., pp.
112-114) were prepared to compare this plant with potential
parents. From these, it was determined that D. digitatum and D.
x sabinifolium are the most likely parents. Table 2 is the index
prepared on the basis of ten qualitative characters by which these
probable parents differ. Of the ten, four character states (irregular,
tufted habit; indeterminate branchlet growth; rounded fertile
branch with isomorphic leaves; and unbranched peduncle) are
closest to D x sabinifolium. Six characters (green coloration; dor-
siventrally flattened and appressed lateral leaves; broad bases of
lateral leaves, reduced lower leaves, broad sporophylls) are closer
to D. digitatum. The putative hybrid has an index score of 9,
intermediate between the scores of D. digitatum (0) and D. x sa-
binifolium (20).

Quantitative characters give a somewhat less clear picture (Ta-
ble 3), since the hybrid is not medial in every character, but
approaches one parent or the other, or exceeds the parents in

1994] Gilman—New Tri-Hybrid Lycopod 291

Table 3. Comparison of mean quantitative morphological characters, each
based on 30 measurements of dried plants (D. digitatum, Gilman 93241, Whee-
lock, VT; D. digitatum x sabinifolium, Gilman 93246, Walden, VT and Gilman
92069, Marathon, Ont.). All measurements in mm. Standard deviations in pa-
rentheses

Character sab dig x sab dig

1. Branch width 1.42 (0.25) 2.01 (0.68) 2.52 (0.57)
Lower leaf width (base of free

ii)

on) 0.71 (0.09) 0.84 (0.12) 0.53 (0.12)
twee leaflength (free portion) 2.28 (0.28) 1.32 (0.15) 1.05 (0.14)

3.

4. Ratio oflowerleafwidth/length 0.311 0.636 0.505

5. Peduncle length 24.6 (7.2) 28.8 (6.24) 48.5 (7.6)
6. Strobilus length 21.6 (4.24) 30.6 (4.78) 22.0 (1.40)
7. Sporophyll width 2.28 (0.28) 2.02 (0.14) 1.88 (0.17)
8. Sporophyll length 2.08 (0.08) 1.98 (0.10) 1.84 (0.09)
9

. Aue of sporophyll width/
ength 1.09 1.02 1.02

a few cases. Branchlet width at 2.01 mm is intermediate between
the parents. The free portions of the lower leaves are intermediate
in length, but are somewhat wider than either parent. The pe-
duncle length is intermediate, but closer to D. x sabinifolium.
Strobilus length is greater than either parent, indeed greater than
average for any native species (Wilce, 1965); this may be an
anomaly of the individual plant or may be the result of hybrid
vigor. The sporophyll is intermediate in width and length, but
approaches D. digitatum in ratio of width to length (the “com-
planatum type” of Wilce).

DISCUSSION

Accepting the hypothesis that D. x sabinifolium is of hybrid
origin, then this hybrid has three ancestral genomes, D. digitatum,
D. sitchense, and D. tristachyum, presumably present in the pro-
portion 2:1:1.

Deviations from some expected intermediate character states
may be inherited from individually variable parents. D. x sabi-
nifolium is itself quite variable (Cody and Britton, 1989) and
variations in certain characters may have been passed on to the
hybrid. Alternatively, variations may be due to unequal contri-
butions of the ancestral genomes, D. digitatum, D. sitchense and
D. tristachyum. In particular, D. sitchense seems to be strongly

292 Rhodora [Vol. 96

represented in the short unbranched peduncle (as opposed to the
long and regularly forked peduncles of D. digitatum and D. trista-
chyum), and in the low-branched, tufted habit (as opposed to the
pseudomonopodial habit of D. digitatum and D. tristachyum).

Another potential hybrid combination, between D. sitchense
and D. xhabereri would have the same genomes in different
proportion (1:2:1). On analogy with D. x sabinifolium, such a
hybrid should display D. sitchense characters even more strongly,
1.¢., 1t would be expected to have non-flattened branches, spread-
ing lateral leaves, and unreduced lower leaves. Furthermore, this
alternative is unlikely because neither parent is known in the
immediate vicinity, whereas D. digitatum is abundant in the area
and D. x sabinifolium is known to occur nearby (D.S. Barrington,
pers. comm.).

As noted above, species and hybrids in Diphasiastrum are mor-
phologically distinct, and the hybrids are apparently fertile. The
barriers that prevent the hybrids from back-crossing and rehy-
bridizing to the point of swamping the parent species, which
maintain their distinctions over broad ranges, have not been iden-
tified. The discovery of the hybrid discussed here is significant
because it suggests at least the possibility of introgression and the
formation of hybrid swarms. On the other hand, both its dis-
tinctiveness and its rarity confirm the current concepts of species
within the group.

Although its spores are well-formed and may be viable, this
hybrid is as yet known only from one site. There is no direct
evidence that it is reproductively competent beyond surviving
and slowly spreading as a single clone. For this reason I have
chosen not to provide an epithet.

MATERIAL EXAMINED

Vermont: Caledonia County: Wheelock, highest pasture on E
side of Ide Mountain, elevation ca. 540 m, 4. V. Gilman 93240,
21 August and 27 November 1993 (vt, NEBC, MICH).

ACKNOWLEDGMENTS

The author thanks Warren H. Wagner Jr. and David S. Bar-
rington for discussions concerning the nature of this plant, and
Martha S. Brown for preparation of Figure 1; also two anonymous
reviewers for comments.

1994] Gilman—New Tri-Hybrid Lycopod 293

LITERATURE CITED

Berret, J. M. 1979. The clubmosses Lycopodium sitchense and L. sabinifolium
in the upper Great Lakes region. Michigan Bot. 18: 3-13.

Copy, W. D. AND D. M. Britton. 1989. Ferns and Fern Allies of Canada.
Agriculture C. anada, Ottawa.

WacneRr, F. S. 1992. Cytological problems in Lycopodium sens. lat. Ann. Mis-
souri Bot. on 79: 718-729.

Wacner, W.H. AND J. M. BEITEL. 1993. Lycopodiaceae, pp. 18-37. Jn: Flora
of North ree Vol. 2. Oxford University Press, New Yor

Witce, J. H. 1965. Section aici of the genus Icopodnin: Nova Hed-
wigia Beih. 19. Weinhei

P.O. BOX 82
MARSHFIELD, VERMONT 05658

THE NEW ENGLAND BOTANICAL CLUB
22 Divinity Avenue
Cambridge, MA 02138

The New England Botanical Club is a non-profit organization
that promotes the study of plants of North America, especially
the flora of New England and adjacent areas. The Club holds
regular meetings, has a large herbarium of New England plants,
and a library. It publishes a quarterly journal, RHODORA, which
iS now in its 95th year and contains about 400 pages a volume.

Membership is open to all persons interested in systematics
and field botany. Annual dues are $35.00, including a subscription
to RHODORA. Members living within about 200 miles of Boston
receive notices of the Club meetings.

To join, please fill out this membership application and send
with enclosed dues to the above address.
Regular Member $35.00
Family Rate $45.00
For this calendar year
For the next calendar year

Name

Address

City & State Zip

Special interests (optional):

Vol. 96, No. 886, including pages 121-206, was issued January 6, 1995,

294

THE NEW ENGLAND BOTANICAL CLUB
leo end Council Menibers for 1004-1905

President: C. Barre Hellquist, Box 9145, Department of Biolo-
gy, North Adams State College, North Adams, Massachusetts

me MIZE.

Vice-President (and Program Chair): W. Donald Hudson, Jr.,
Chewonki Foundation, RR2, Box 1200, Wiscasset, Maine
04578

_ Corresponding Secretary: Nancy M. Bysiee -Smith, Department

of Natural Sciences, Bentley Colle: Waltham, Massachu-

setts 02154 .

Treasurer, Harold G. Brotzman, Box 9092, Department of Biol-

~~ ogy, North Adams State College, North Adams, Massachu-
setts 01247

Recording Secretary: Lisa A. Standley

Curator of Vascular Plants: ee Angelo

Assistant Curator of Vascular Plants: | Pamela Weatherbee

Curator of Non-Vascular Plants: Anna M. Reid

Librarian: Paul Somers

Council: Consisting of the Elected Officers, Associate Curator,

| Editor of Rhodora and —

. Councillors: Leslie J. Mehrhoff (Past President)

William E. Brumback ’95

Thomas Mione ’96

Garett E. Crow ’97

Donald J. Padgett (Graduate Student
“Member) 95 _-

JOURNALOFTHE =~
NEW ENGLAND BOTANICAL CLUB

is CONTENTS:
The taxonomy of Carex ae (espersecee) and oes species in
North America. Peter W. Bal 295
Pollen and pubescence. S auchnipas ay s Ovals grandis Small. Darrell ~~
Moore, Dallas Mullins and Foster 311
Floristic diversity of a disturbed western Ohio fen. James S. MeCormae
and Gregory J. Schneider 327

Genetic variation in Massachusetts populations of Cypripedium arietinum
; in Ait. and C. acaule Ait. (Orchidaceae). an H.
Bornbusch, Lesley A. Swender, and Deborah L. Hoogerwerf .............. 354
Exine reduction in underwater flowering Callitriche (Callitrichaceae): im-
plications for the evel yes of hypohydrophily. no Thomas Philbrick
and eee M. Osbor. . 370

research award | . : 389.

PT OmGse coun for Callitriche (Callitrichaceae) in North America.
(5.76

THE NEW ENGLAND BOTANICAL CLUB
P.O. Box 1897, Lawrence, Kansas 66044 —
22 peso a one Cambridge, Massachusetts 02138

VOL.9%6 —_— October, 1994 No. 888

Vas The New ae Boranicnt Club, he
ae a6 2 Daal Avenue, Cambridge, Massachusetts 021 38

_ RHODORA —
_ GORDON P, DrWOLF, RR. Editor-in- Chiet

Sar cociale Raisers
DAVID S. CONANT

TICA A STANDILEY

shat RHODORA (ssn 0035- 4902). Published four times a a year ala anvary,
ae July, and October) by The Ne England Botanical Club, 810—
_ Bast 10th St., Lawrence, KS 66044 and printed by Allen Press, Inc.,
104) New Hampshire St., Lawrence, KS 66044. Second class postage i
pending at Lawrence, KS. POSTMASTER: Send ae ad LOM aie
~RHODORA, RO; Box. 1897, Lawrence, KS 66044. et ee

= , RHODORA i isa journal of botany devoted primarily to N orth aie
Authors are encouraged to submit manuscripts of scientific papers —
and notes relating to the oe distribution, ecology, ee,

1

ne botany, or conservation biology of thi:
A SUBSCRIPTIONS: $40.00 per sents year, net, pooae. in fuls 3
payable at, par in the United States currency. Remittances payable to
~RHODORA. ae to RHODORA, ae Box ADT ee KS.
66044-8897.

a ‘MEMBERSHIPS: "Regular $35: eee $45, Application form printed .
< herem, .

BACK VOLUMES AND SINGLE COPIES: one ie informa- :
~~... tion and prices will be furnished upon request to the Treasur rer. oe
- ADDRESS CHANGES: In order to receive the next Thumber of
RHODORA, changes must bs received by the business Ear Bae to
7 the first day of January, April, July, or October .
ein INFORMATION FOR CONTRIBUTORS: Inside c back cover, »Fanuary
Sati nd April. Poe ar ;
a MANUSCRIPT sr “Send to: ‘Gpidon D > De Wol
3 Managing Editor, RHODORA
acy . 425. Long Hill Roa
t ey, West Bre yokfield, MA 01585

This paper meets the requirements of ANSI/NISO 239.48-1 992 (Permanence of Paper). Z

RHODORA

JOURNAL OF THE
NEW ENGLAND BOTANICAL CLUB

Vol. 96 October 1994 No. 888

RHODORA, Vol. 96, No. 888, pp. 295-310, 1994

THE TAXONOMY OF CAREX PETRICOSA (CYPERACEAE)
AND RELATED SPECIES IN NORTH AMERICA

PETER W. BALL AND MARGARET ZOLADZ

ABSTRACT

In North America the Carex petricosa group has generally been considered to
consist of three species, C. petricosa and C. franklinii in western Canada and
Alaska, and C. misandroides in eastern Canada. In addition a number of other
species and varieties have been described in the past fifty years, but these have
generally not been given formal recognition by subsequent authors. A morpho-
logical study of herbarium specimens of this group led to the conclusion that all
or these pri should be treated as a single species. ape western and eastern pop-

iffered in the proportion of flowers that tigmatic. Western plants
were ae tristigmatic or had not more than 50% Pear flowers. Eastern
plants were mostly distigmatic or had more than 50% distigmatic flowers. Con-
sequently two varieties were recognised in C. petricosa, var. petricosa from the
west, and var. misandroides (Fernald) Boivin from the east.

Key Words: Carex sect. Aulocystis, taxonomy, distribution, North America

INTRODUCTION

Carex petricosa Dewey is a member of the large section Au-
locystis Dumortier. The section has been known under a variety
of names, in particular section Ferrugineae (Mackenzie, 1935)
and section Frigidae (Kiikenthal, 1909). Kiikenthal (1909) recog-
nised over 50 species in the section, of which only 6 occurred in
North America. In his monograph of North American Carex
Mackenzie (1935) recognised 10 species in the section, of which
5 were endemic to the western mountains of the United States,
just extending into southern British Columbia, 2 were widespread
circumboreal arctic-alpine species, and 3 were subarctic-montane
species endemic to Canada and Alaska. These latter 3 species, C.

27

296 Rhodora [Vol. 96

petricosa, C. franklinii Boott and C. misandroides Fernald, form
the group of species which is the subject of this paper.

The Carex petricosa group of species is difficult to define mor-
phologically in an unambiguous way. The perigynia are more or
less erect when mature, sparsely setose on the surface and tapering
at the apex to an indistinct beak. The distribution of staminate
and pistillate flowers in the inflorescence is extremely irregular,
some individuals having a single terminal staminate spike and
several lateral pistillate spikes, while others have an androgynous
terminal spike (staminate at the apex, pistillate at the base), with
some pistillate and some androgynous lateral spikes, often ar-
ranged in an irregular manner. The stigma number is also variable
as all three species can have 2 or 3 stigmas, although 3 is pre-
dominant in C. petricosa and C. franklinii and 2 is predominant
in C. misandroides. When Mackenzie prepared his monograph
these species were known from very few collections. Carex frank-
linii and C. petricosa were known from a very few stations in the
Rocky mountain range in Canada, while C. misandroides was
known from a single station in Newfoundland and another in the
Gaspé peninsular in eastern Quebec. Subsequent collections have
extended the range of C. franklinii and C. petricosa to Alaska and
the western Northwest Territories, and additional sites for C.
misandroides have been found in Newfoundland and Quebec.

Since the publication of Mackenzie (1935) the North American
species of the group have been the subject of two revisions. In
both of these a number of new or additional taxa were recognised
or described. Boivin (1948) described one new species, C. disti-
chiflora, and two new varieties, C. franklinii var. nicholsonis, and
C. petricosa var. edwardsii. Not to be outdone Raymond (1952)
added two new species, C. magnursina and C. lepageana, and the
first North American record of the Asian species C. stenocarpa
Turcz. ex Besser. However, Raymond included C. distichiflora in
the synonymy of C. franklinii, and treated C. misandroides as a
variety of C. franklinii. Subsequent authors have tended to take
a more conservative view of the group and have usually recog-
nized two species in the west, C. franklinii and C. petricosa, with
C. misandroides being included as a synonym or variety of C.
Jranklinii (e.g., Hultén, 1968; Porsild and Cody, 1980; Packer,
1983). Boivin (1967) and Scoggan (1978) recognised only a single
species with four varieties, and Welsh (1974) treated C. franklinii
as a synonym of C. petricosa. The report of C. stenocarpa from

1994] Ball and Zoladz— Carex petricosa 297

Alaska (Raymond, 1952) seems to have been overlooked as it is
not mentioned in Hultén (1968) or Welsh (1974). In addition to
North American names, Hultén included the Asian species C.
macrogyna Turcz. ex Besser as a synonym of C. franklinii. An-
other, unpublished, name also appears on herbarium sheets. As
noted by Raymond (1952), Fernald had named the earlier col-
lections from Lac Mistassini, Quebec, as C. mistassinica, but this
name was never validly published and Raymond treated the spec-
imens so named as C. franklinii var. misandroides.

Initially the investigation was aimed at determining the status
of C. misandroides, whether it could be retained as a taxon distinct
from the western species, and also whether it was most similar
to C. franklinii or C. petricosa. As the investigation progressed it
became apparent that these questions could not be adequently
addressed without a more thorough examination of the status of
the western taxa of the group, including those published since
Mackenzie (1935).

METHODS

The study was based entirely on herbarium specimens collected
from North America and a small number from north east Asia.
Approximately 250 specimens were examined from the following
herbaria: ALA, CAN, DAO, MT, TRT, TRTE (acronyms according to
Holmgren et al. 1990). The North American specimens repre-
sented a high proportion of the known stations of the taxa under
investigation.

It proved difficult to classify specimens on the basis of mor-
phology, so all plants from Newfoundland and Quebec were as-
signed to C. misandroides, except for the specimens named C.
mistassinica, which were provisionally treated as a separate taxon.
The post-1935 taxa are known only from specimens cited in the
protologue and these were also treated as separate taxa, although
in some cases the specimens were too immature to include in the
morphologic analyses. Carex petricosa and C. franklinii proved
difficult to distinguish based either on the names on specimens,
or on the characters utilized in such works as Packer (1983). The
specimens were eventually separated arbitrarily based on the width
of the mature perigynium, specimens with the perigynium wider
than 1.7 mm being allocated to C. franklinii, those with the peri-
gynium narrower than 1.7 mm being allocated to C. petricosa,

298 Rhodora [Vol. 96

and those with the perigynium 1.7 mm wide being allocated ac-
cording to the name on the label.

Characters were selected based on those utilized by Mackenzie
(1935), Boivin (1948) and Raymond (1952) to separate the taxa
that they recognized. A few proved impossible to record satis-
factorily, particularly the “curved” basal leaves emphasized by
Raymond (1952) in his diagnosis of C. magnursina, and the color
of the perigynia (dark brown versus yellowish brown) which to
some extent depends on the age and state of development of the
inflorescence. The number of stigmas was found to be quite vari-
able on one individual. This was recorded by randomly choosing
5 perigynia in an inflorescence and reporting the number which
had 2 stigmas. For specimens which had shed their stigmas the
stigma number was based on the cross-sectional shape of the
achene, biconvex for 2 stigmas and triangular for 3 stigmas. Be-
cause examination of the achenes is destructive only 3 perigynia
were examined in these specimens. In a few specimens a large
number of perigynia were available with visible stigmas and these
were used to confirm that the method outlined above gave a
reasonable estimate of the frequency of the two stigma numbers
on one individual.

A total of 20 characters were recorded from the specimens
examined, but not all could be utilised in the numerical analyses.
Specimens were selected based on the presence of reasonably well-
developed perigynia. Many specimens were collected too early in
the season to have reached this stage of development, conse-
quently it was sometimes necessary to use duplicate specimens
to obtain a reasonably large sample for analysis. A total of 101
specimens were eventually included in the data set, 29 specimens
of C. petricosa (including 2 determined as var. edwardsii), 27
specimens of C. franklinii, 24 specimens of C. misandroides, plus
2 specimens determined as C. mistassinica, 15 specimens of C.
macrogyna, 2 specimens of C. /lepageana, and | each of C. dis-
tichiflora, C. magnursina and C. stenocarpa (from Alaska).

Principal components analysis (PCA) and discriminant func-
tions analysis (DFA) were performed using SYSTAT 5.0 (Wil-
kinson, 1990). PCA was used to obtain a broad overview of the
variation in the data set and to determine whether there was any
tendency to form groups. The data used in this analysis were
standardised to zero mean and unit variance. All characters listed
in Table 1 were included in the analysis except for STS which

1994] Ball and Zoladz— Carex petricosa 299

Table 1. List of morphological characters recorded for the Carex petricosa
oup.

er
Character
Character Abbreviation
Stem height (cm) STHT
Bract length ( ) BRL
Bract sheath length (mm) BRSL
Sex of the terminal sp STS
Number of pistillate spikes FSPN
Number of staminate spikes MSPN
mber of hermaphrodite spikes HSPN

Pistillate spike width (mm) FSPW
Pistillate spike length (mm) FSPL
Width of the staminate portion of the terminal spike (mm) MSPW
Length of the staminate portion of the terminal spike (mm) MSPL
Staminate scale length ( STSCL
Staminate scale width (mm) STSCW
Pistillate scale length (mm) PSCL
Perigynium length (mm) PERL

erigynium width (mm) PERW
Perigynium beak length (mm) PERBL
Perigynium nerve number PERNN

Stigma number

was considered to be a variable that was not independent from

DFA was used to obtain maximum separation of the four major
taxa, C. franklinii, C. macrogyna, C. misandroides and C. petri-
cosa. PERW was not included in this analysis as this character
was used to define C. franklinii and C. petricosa. The other two
taxa were defined using distribution as the main criterion. A clas-
sification procedure was used to assign the remaining taxa to one
of the four groups.

RESULTS
Taxonomic Characters
It is not intended to review all the characters included in this

study, but rather to consider those on which considerable em-
phasis has been placed in the past. Except as discussed below

300 Rhodora [Vol. 96

none of the characters examined showed any significant difference
between the four main taxa included in this study.

1. STEM HEIGHT. This character has been used to distinguish
C. franklinii and C. petricosa (e.g., Porsild and Cody, 1980, Pack-
er, 1983), and C. franklinii and C. misandroides (Raymond, 1952).
Stem height was very variable and was poorly correlated with
other diagnostic characters, particularly perigynium width (cor-
relation coefficient 0.252).

2. TERMINAL SPIKE. Kiikenthal (1909) placed C. franklinii and
C’. petricosa in different subsections of sect. Frigidae based on
differences in the terminal spike. Carex franklinii was described
as having an androgynous terminal spike, while C. petricosa was
said to have a mainly staminate terminal spike. As pointed out
by Raymond (1952) this character is very variable. Boott (1839)
in his diagnosis of C. franklinii noted that the terminal spike
could be both staminate and androgynous. Two collections from
a population near Jasper, Alberta, close to the area from which
the type of C. franklinii was collected, emphasize the point. Por-
sild 22537 (CAN) has an androgynous terminal spike (with peri-
gynia 1.7 mm wide), and Porsild 22538 (CAN) has a staminate
terminal spike (with perigynia 2.2 mm wide). In both C. petricosa
and C. misandroides a staminate terminal spike is more frequent,
but in both taxa androgynous spikes occur in about 50% of the
specimens examined. This variation may occur in a single clone
Suggesting that this character is not under direct genetic control.
There is a tendency for robust plants to have an androgynous
terminal spike and for small plants to have a staminate terminal
spike, but this is not at all consistent (the two Porsild collections
cited above have stem heights of 48 cm and 51 cm respectively).

3. LATERAL PISTILLATE SPIKE SIZE. Raymond (1952) empha-
sized this character when describing C. /epageana and C. mag-
nursina, stating that these two species had pistillate spikes 8—25
x 3-4 mm, while C. petricosa, C. franklinii and C. misandroides
had pistillate spikes 8-15 x 6-9 mm. Carex petricosa was found
to have a spike length of 7-29 mm (mean 15 mm), while C.
misandroides and C. petricosa had a similar mean but narrower
range. Spike width showed a similar pattern, with C. petricosa
having spikes 3-7 mm wide. All three taxa had a mean width
close to 5 mm.

4. PERIGYNIUM SIZE. Perigynium width is the character most
frequently used to separate C. petricosa and C. franklinii. Ray-

1994] Ball and Zoladz— Carex petricosa 301

COUNT

08 10 12 14 #116 %118 20 22 24
PERW

Figure 1. Histogram of perigynium width (mm) of western North American
specimens of the Carex petricosa group.

mond (1952) gave the perigynium width of C. petricosa as |.5—
1.75 mm and that of C. franklinii and C. misandroides as 2-2.5
mm. Other authors, such as Mackenzie (1935), gave a similar
difference. Porsild and Cody (1980) claim a difference in perigy-
nium length giving 5-6 mm for C. franklinii and 4-5 mm for C.
petricosa. In this investigation perigynium length showed no clear
difference between the three main North American taxa. Carex
petricosa was the most variable with a range of 3.5-5.5 mm, while
C. franklinii had a range of 3.8-5.3 mm and C. misandroides 3.9-
5.2 mm. Perigynium width was eventually used to arbitrarily
separate C. franklinii from C. petricosa. The clear disjunction in
perigynium width claimed by Raymond (1952) and others does
not exist. A histogram of perigynium width (Figure 1) for the

302 Rhodora [Vol. 96

Figure 2. Histograms of the aaa of flowers, on each specimen, with
two stigmas. A. Western North Ameri

western North American taxa shows a somewhat bimodal dis-
tribution with peaks at about 1.3 mm and 1.7 mm but there is
no discontinuity. Carex misandroides has a perigynium width
matching that of C. petricosa rather than that of C. franklinii.

5S. STIGMA NUMBER. This character was emphasized by Fer-
nald (1915) and Mackenzie (1935) in their descriptions of C.
misandroides. Both acknowledged that there was some variation
in the stigma number, but no data were presented to indicate the
extent of the variation. In this study it was found that the eastern
populations have a preponderance of flowers with two stigmas,
while the western populations have a preponderance of flowers
with three stigmas (Figure 2). Approximately 10% of the western
specimens examined either had a preponderance of flowers with

1994] Ball and Zoladz— Carex petricosa 303

20° | ne js
15 +
|
eu |
Z|
5 10
‘e)
B
|
a
oe.
0.0 0.2 0.4 0.6 08 1.0

TN

Figure 2. Continued. B. Eastern North America.

two stigmas or had flowers with two and three stigmas about
equal in frequency. In the east about 20% of plants had a pre-
ponderance of flowers with three stigmas or had flowers with two
and three stigmas about equal in number. No western specimens
were entirely distigmatic and no eastern specimens were entirely
tristigmatic. The Asian species, C. macrogyna, was similar to the
western North American taxa in this character.

Multivariate Analyses

PCA showed a more or less normal distribution of specimens
in the space defined by the first three components, with no ten-
dency to form clusters (Figure 3). The first component accounted
for 23.7% of the variance in the data set, with a large number of

304

[Vol. 96

Rhodora
3
m
2
m m
m i, Pp a
f
i ae
mp ™ Pp
p Pomp ~ny ¥ p
0 . sfP p
'S) y yy a
Ao y 28 <4
sa pp ¥ i Pp J
y y ; Py
fp
9 y y a P
p
<3
p
ai | |
| aay! =| 0 l p) 3
PC |

Figure 3. Bivariate plot of OTU’s of the Carex petricosa group on principal
components | and 2. Letters identify taxa according to Table

size variables having high loadings. Most specimens of C. frank-
linii had a positive score on this axis, while C. macrogyna, C.
misandroides and C. petricosa had individuals with both high
positive and high negative scores. The second component ac-
counted for 10.6% of the variance in the data set, with staminate
spike number, stigma number, perigynium beak length and pis-
tillate spike number having high loadings. Carex misandroides
(together with C. mistassinica) had positive scores on this axis,
but they were mixed with individuals from the other main taxa.
The third component showed no useful groupings. The specimen
determined as C. stenocarpa was placed well within the main
cluster, and was neither marginal nor disjunct from the other

taxa.

1994] Ball and Zoladz— Carex petricosa 305

DFA gave only a slightly better grouping of specimens than the
PCA (Figure 4). Carex misandroides formed a reasonably discrete
group on the first DF, with only 4 of 24 specimens being mis-
classified (Table 2). The two specimens labelled as C. mistassinica
were classified in this group, as might be expected. Stigma number
was the most important discriminating variable, with bract sheath
length as a secondary character. The second DF weakly separated
C. franklinii, C. macrogyna and C. petricosa. With C. petricosa
only 16 of 29 specimens were correctly classified, the remaining
13 specimens being evenly distributed among the other three taxa.
Carex franklinii had 19 of 27 specimens correctly classified, and
C. macrogyna had 10 of 15 correctly classified. The type of C.
magnursina was Classified with C. misandroides, but was close to
the boundary with C. petricosa. A specimen from Nome, Alaska,
determined by Raymond as C. stenocarpa was classified as C.
franklinii, although it was close to the boundary with C. petricosa.
The specimens of C. /epageana were also classified as C. franklinii.

DISCUSSION

The complete failure of the multivariate analyses to satisfac-
torily differentiate the taxa included in this study 1s strong evi-
dence to support the view that there is only one species in this
group. The exclusion of perigynium width from the DFA might
account for some of the problems with group separation in that
analysis, but it is clear from Figure | that there is no clear dis-
junction in this character as suggested by previous authors. Carex
lepageana, C. magnursina and C. distichiflora fall well within the
range of variation exhibited by the group as a whole. Likewise
the specimen determined as C. stenocarpa appears to be a part
of this group. Although no reliably determined Asian specimens
of C. stenocarpa were examined, this species 1s described as having
the larger leaves 4-6 mm wide {Kreczetowicz, 1935), while the
Alaskan specimens have leaves not exceeding 3 mm wide, the
same range as the C. petricosa group. The status of C. macrogyna
is still unclear. The analyses support Hultén’s (1968) view that
C. macrogyna should be treated as a synonym of C. petricosa,
especially given the broader circumscription of C. petricosa pro-
posed above. However, most of the specimens included in the
analyses came from the subarctic regions of Russia, while C.
macrogyna was described from the mountainous area bordering

306 Rhodora [Vol. 96

Table 2. Classification of specimens in the Carex petricosa group based on
Rows are the groups to which specimens were assigned: columns are the
classification according to the DFA.

P M F Y Total
C. petricosa 14 3 7 5 29
Pr
C. misandroides 1 20 2 1 24
C. franklinii 1 1 19 5 26
F
C. macrogyna 2 l 0 11 14
Y
C. lepageana 0 0 2 0 2
L
C. mistassinica 0 3 0 0) 3
I
C. magnursina 0) 1 0 0 1
G
C. stenocarpa 1 0 0 0) |
S
C. distichiflora 1 0 0 0 |
D
Total 20 29 30 22 101

Mongolia. Without a more extensive study of the group in Asia
it seems prudent to not include C. macrogyna as a synonym of
C. petricosa at this time. It does seem clear that the Asian subarctic
populations are not distinct from C. petricosa, and these should
be included in that species.

The final question concerns the status of C. misandroides. This
taxon is disjunct from all the others by several thousand kilo-
meters. Between 80% and 90% of specimens can be correctly
classified using stigma number alone, without knowledge of the
origin of the specimen. No other characters can be used to dis-
tinguish C. misandroides from the western taxa. Bract sheath
length, which had a high value on the first DF together with stigma
number, is much too variable. Carex misandroides has the highest
mean value for this character, but the range, 3-38 mm, completely
includes all the western taxa. Stem height shows some slight dif-
ference, with C. misandroides tending to be shorter than other
taxa. Most plants of C. misandroides have stems in the range 20-
35 cm high, with the extremes in the range 10-55 cm. In the
western taxa the normal stem height is in the range 20-SO cm,

1994] Ball and Zoladz— Carex petricosa 307

3 = _ —
L EF
F LF
2 7
Fp P FM
| FF Ff F M
1 pos M
poar’ om
Pp Y SF
0. > PP P P Gt M MM |
Ss) F oY 5 : I M
ra | - ae Pp pa nf -
—] _ pP M M M M
Y FY Y ”
| Yy P
2.4 Y I
| vy
Pp
3
Y
—4
—4 a2 0 2 4 6
DF(1)

Figure 4. Bivariate plot of OTU’s of the Carex petricosa group on discriminate
functions | and 2. Letters identify taxa according to Table 2.

with an extreme range of 10-80 cm. The lack of any consistent
morphological difference between C. misandroides and the other
taxa in the group precludes the recognition of this as a distinct
species. On the other hand the geographical isolation combined
with a morphological feature which reliably classifies over 80%
of individuals suggests that recognition at infraspecific rank 1s
justified. The distribution of Carex petricosa in North America
is given in Figure 5.

TAXONOMIC SUMMARY

1. Carex petricosa Dewey, Amer. J. Arts Sci 29: 246. 1836. Sum-
mits of Rocky mountains (probably west of Jasper, Alberta).

308 Rhodora [Vol. 96

tes °
¥ . °°,
: ie) “es |
% ° ad ° g e
%° of e
Pa
% e ‘
Sol
| EA e
* , a a 4
er e
$ t
° « A

Figure 5. Distribution of Carex petricosa in North America. Circles = var.
petricosa, triangles = var. misandroides. Solid symbols indicate specimens ex-
amined in this study; open symbols indicate literature records.

Drummond 283 (HoLotyPe: Herb. Dewey, GH, photo CAN!,
MT!, ISOTYPE: K, photo CAN!).

a. varlety petricosa

Carex franklinii Boott in Hooker, Fl. bor.-am. 2: 217. 1839. Rocky
mountains (near Jasper, Alberta). Drummond 293 (HOLOTYPE:
K, photo CAN!, MT!).

Carex distichflora Boivin, Naturaliste Canad. 75: 206. 1948.
NORTHWEST TERRITORIES: Canol road, Mackenzie range, Sek-
wi river, mile 174E, pump station 5, 6 Sept. 1944. Porsild
and Breitung 11848 (HOLOTYPE: DAO!).

Carex mangursina Raymond, Canad. Field-Naturalist 66: 100.
1952. NORTHWEST TERRITORIES: Great Bear Lake, Harrison
river, McTavish arm. 30 July 1948. Steere et al. 3228
(HOLOTYPE: MT!).

Carex lepageana Raymond, Canad. Field-Naturalist 66: 101. 1952.
ALASKA: Anvil Hill, Nome. 15 August 1948. Lepage 24031
(HOLOTYPE: MT!).

1994] Ball and Zoladz— Carex petricosa 309

Carex stenocarpa sensu Raymond, Canad. Field-Naturalist 66:
100. 1952. non Turcz. ex Besser.

b. variety misandroides (Fernald) Boivin, Phytologia 43(1): 83.
1979

Carex misandroides Fernald, Rhodora 17: 158. 1915. NEw-
FOUNDLAND: Table Mountain, Port 4 Port Bay. Fernald and
St. John 10801 (HOLOTYPE: GH, photo cAn!). Carex franklinii
var. misandroides (Fernald) Raymond, Canad. Field-Natu-
ralist 66: 102. 1952.

Boivin first used this combination in 1967, but did not validly
publish it until 1979 in Flora of the Prairie Provinces 4.

ACKNOWLEDGMENTS

We thank the curators of ALA, CAN, DAO, and mT for the loan
of specimens and for answering queries. Particular thanks to Paul
Catling for tracing the place of publication of var. misandroides.
This work was supported by National Research Council of Canada
Grant A6494.

LITERATURE CITED

Borvin, B. 1948. eee a ae Canadiennes. II Etudes caricologique.
Naturaliste Canad. 75: 208.
1967. Srrs ee plantes du Canada. VI—Monopsides (2 éme
partie). Naturaliste Canad. 94: 471-528.
Boortt, F. 1839. Carex. Jn W. J. Hooker, Flora boreali-americana 2(11). Treuttel
and Wiirtz, London
FERNALD, M. L. 1915. a new Carex from Newfoundland. Rhodora 17: 158-

159.

HouLmaren, P. K., N. H. HOLMGREN AND L. C. BARNETT. 1990. Index Herba-
riorum. Part I: The Herbaria of the World, 8th ed. New York Botanical
Garden, New York, NY

HUuLtTEN, E. 1968. Flora of Alaska and Neighbouring Territories. Stanford Uni-
versity Press, Stanford, CA.

KreczeTowicz, V. I. 1935. Carex. Jn Flora URSS 3, V. L. Komarov, Ed. Len-
ingrad.

UKENTH 909. Cyperaceae-Caricoideae. In; A. Engler, Ed., Das Pflan-
ares 4(20). nee

Mackenzig, K. K. 1935. Cyperaceae-Cariceae. Jn North American Flora 18(5—
6). New Vork a Garden, New York, NY.

310 Rhodora [Vol. 96

Packer, J. G. 1983. Jn: E. H. Moss, Ed. Flora of Alberta, 2nd ed. University
of Toronto, Toronto.

PorsiLp, A. E. AND W. J. Copy. 1980. Vascular Plants of the continental North-
west Territories. National Museums of Canada, Ottawa.

RaymMonpb, M. 1952. The identity of Carex misandroides Fern. with notes on
the North American Frigidae. Canad. Field-Naturalist 66: 95-103.

ScoGGAN, H. J. 1978. The Flora of Canada. 2. National Museums of Canada,

Ottawa.
WELSH, S. L. 1974. Anderson’s Flora of Alaska. Brigham Young University,
Provo, UT.
WILKINSON, L. 1990. SYSTAT: The System for Statistics. Evanston, IL.
DEPARTMENT OF BOTANY

ERINDALE CAMPUS, UNIVERSITY OF TORONTO
MISSISSAUGA, ONTARIO, L5L 1C6, CANADA

RHODORA, Vol. 96, No. 888, pp. 311-326, 1994

POLLEN AND PUBESCENCE CHARACTERISTICS
OF OXALIS GRANDIS SMALL

DARRELL Moore, DALLAS MULLINS AND FOSTER LEVY

ABSTRACT

Pubescence in Oxalis grandis Small wa lyzed as a constellation of characters
describing the numbers, types, orientations, and lengths of hairs on different regions
of the plants. Pubescence is a complex character whose components vary inde-
pendently with no obvious geographical or ecological correlates. Hirsute plants
occur sporadically throughout the range of the species, but extreme expression of
the entire suite of pubescence characters is restricted to one population. A cor-
relation analysis indicates that pubescence in Oxalis grandis is most likely under
complex genetic control and is therefore valuable as a taxonomic character. An
analysis of pollen size utilizing plants from two populations representing extremes
in pubescence shows significant differences between flower morphs. For corre-

These studies have demonstrated high levels of intraspecific polymorphism within
Oxalis grandis.

Key Words: character analysis, heterostyly, Oxalis, pollen, pubescence

INTRODUCTION

The genus Oxalis comprises species with heterostylous and
homostylous flowers. Oxalis grandis Small, a resident of the
southern Appalachian Mountains and Ohio River Valley, is tn-
stylous with marked size differences among the style lengths of
its three flower morphs (Eiten, 1963). Each style morph has two
whorls of functional stamens. Members of the species possess
characteristics of subsection Strictae, section Corniculatae: a
creeping rootstock, spreading pedicels and septate trichomes. Ox-
alis grandis is further distinguished by relatively large leaflets
bordered by a purple margin (Wiegand, 1925). All individuals
from a population inhabiting lower and middle slopes of bluffs
along the Watauga River in Sullivan and Washington Counties
of northeastern Tennessee are extremely hirsute on all parts of
the plant and thus are distinguished easily from those of more
typical populations (Levy and Moore, 1993). These morpholog-
ical differences persist when plants are transplanted to a common
greenhouse environment and in a growth chamber when new
ramets are propagated from root cuttings (F. Levy and D. Moore,
unpublished data).

The extreme intraspecific variation encompassed by the Wa-

ple

ae Rhodora [Vol. 96

tauga River population in comparison to more typical populations
provided an opportunity for an in-depth examination of the pu-
bescence phenotype. Our objective was to provide a full descrip-
tion of the morphological differences related to pubescence and
to conduct a herbarium search for Oxalis grandis specimens with
extreme amounts of pubescence. We then performed a character
correlation analysis to assess whether pubescence is under simple
or complex genetic control. Specific goals of the study were to
determine whether (i) the extremely pubescent phenotype is unique
to the Watauga River population, (ii) there are geographic or
ecological correlates associated with pubescence, (ili) expression
of components of pubescence vary independently. Consequently,
we have dissected pubescence into several qualitative and quan-
titative components and examined the expression of each in in-
dividual plants.

In the most recent treatment of subsection Corniculatae, Eiten
(1963) considered qualitative components of epidermal vestiture.
Eiten’s implicit assumption was that each component represented
an independent character. If pubescence is a simple character with
a simple genetic basis, or if it is environmentally determined, then
it has minimal taxonomic value (Rollins, 1958). On the other
hand, if pubescence represents the expression of several indepen-
dent characters then it merits more attention in assessing degrees
of differentiation. In this study, we use correlation analysis to
assess the complexity of the pubescence phenotype and as a test
of Eiten’s assumption of independence.

A second objective of the study is an analysis of pollen sizes
of the three flower morphs of O. grandis to determine whether
pollen diameter differs among flower morphs and whether pollen
size differs between populations. For this analysis we compared
pollen from two populations, the Watauga River population and
a population in a more typical habitat for the species (mesic forest
at Iron Mountain Gap, Unicoi County, TN).

SECTION I—PUBESCENCE

METHODS

Pubescence was measured on ten plants from each of two study
populations. Of these ten plants, four and three were fresh and
the remainder recently collected and dried, from the Iron Moun-

1994] Moore et al.—Oxalis grandis 313

tain Gap and Watauga River populations, respectively. Plants
were carefully chosen to encompass the range of mature plant
size, habitat and geographical location within each population.
Only mature, flowering or fruiting specimens were included. The
numbers of nonseptate hairs (no septate hairs occur on leaflets)
were counted within 5 x 5 mm squares on the abaxial and adaxial
leaflet surfaces. These grids were displaced a minimum of 2 mm
from the midvein of one leaflet on each plant. For a subset of
plants, hairs were counted on three different leaves, but prelim-
inary analysis showed little variation and so, for the remainder,
only one grid was counted per plant. Nonseptate and septate hairs
also were counted on the circumference of one centimeter lengths
of (i) the middle of a petiole, (ii) a section of the stem from the
upper one-third of the plant (upper stem), and (iii) a section of
the stem from the lower third of the plant (lower stem). Also
measured were the maximal lengths of leaflet hairs (all nonseptate)
and the septate and nonseptate hairs on stems. All counts and
measurements were conducted at 40x under a dissecting micro-
scope.

The analysis identified a suite of eleven pubescence characters
by which the two populations may be differentiated (Table 3).
We measured the same set of characters in a survey of herbarium
specimens to determine (i) whether the extremely pubescent phe-
notype of the Watauga River population was unique or if a con-
tinuum of variation existed within the species, and (i1) whether
individual characters related to pubescence vary independently.
For the detailed analysis, we chose only herbarium specimens
possessing some hairs on their leaves, that is, specimens showing
at least a slight tendency toward a hirsute phenotype. The screen
for pubescent phenotypes included specimens from the following
five herbaria: UNC (51 sheets), TENN (31 sheets), vpr (25 sheets),
wv (57 sheets), and the collection at EAST TENNESSEE STATE
UNIVERSITY (10 sheets).

DATA ANALYSIS. For each of the five regions for which
hairs were counted (abaxial and adaxial leaf blades, petioles, upper
and lower stems), separate Kruskal-Wallis tests were conducted
to compare the number of hairs between populations. This non-
parametric test was chosen because of heteroscedasticity of var-
iances among populations. Separate analyses were carried out for
septate and nonseptate hairs. G tests were used to compare pro-
portions of septate and nonseptate hairs. These tests were con-

314 Rhodora [Vol. 96

ducted first within populations to determine whether proportions
varied in different regions of plants and then corresponding regions
of plants were compared among populations.

Associations between pairs of pubescence characters were es-
timated by Pearson Correlation Coefficients (CORR procedure,
SAS Institute, 1982). These correlations were computed between
all possible pairs of characters within: (i) the sample of 10 plants
from the Iron Mountain Gap population, (ii) the sample of 10
plants from the Watauga River population and, (iii) the sample
of 14 herbarium specimens exhibiting hairs on the leaves. Eight
characters were included for the Iron Mountain Gap and Watauga
River samples (there was no variation for any of the three hair-
length characters within either population), and eleven characters
in the herbarium sample (Table 3).

RESULTS

The Iron Mountain Gap population exhibited the complete
absence of hairs on the abaxial and adaxial leaf surfaces, except
for sparse nonseptate hairs along the midvein and leaflet margins.
This contrasted strongly with the abundant nonseptate hairs on
both leaf surfaces ofall plants from the Watauga River population
(Table 1). Although the midvein of leaflets from both populations
had nonseptate hairs, those of the Watauga River population were
longer (1.0-1.5 vs 0.7 mm) and were patent in contrast to the
appressed hairs of the Iron Mountain Gap population. Both sep-
tate and nonseptate hairs occurred throughout the vegetative axis
(stems and petioles) of plants from both populations, but the mean
number of hairs was significantly greater in the Watauga River
population for each hair type in each region of the plant (Table
1). Furthermore, the ranges of variation in the number of non-
septate hairs on petioles, as well as on upper and lower stems,
did not overlap between the two populations (Table 1).

The ratio of septate : nonseptate hairs increased significantly on
the lower stems within both populations, but did not differ be-
tween upper stems and petioles (Table 2). Comparisons between
populations showed that the relative contribution of septate hairs
on petioles was similar, but for upper and lower stems, it was
significantly higher in the Iron Mountain Gap population (Ta-
bie 2),

The unusual Watauga River population can be characterized

Table 1. Pubescence characteristics of two populations of Oxalis grandis. Entries represent the mean, standard deviation (SD) and range
for numbers of septate and non-septate hairs in different regions of plants. The Kruskal-Wallis statistics and their associated probabilities
(* <0.05, *** <0.001) are given for comparisons among populations.

[r66l

Stem
Upper Lower Petiole Leaflet
Non-septate Septate Non-septate Septate Non-septate Septate Abaxial Adaxial

Iron Mountain Gap

Mean (SD) 53.3 (17.6) 25.4 (9.9) 22.2 (18.6) 69.1 (32.8) 45.7 (14.0) 15.0 (7.3) 0 0

Range 29-89 9-44 8-57 39-149 0-78 8-30 0 0
Watauga River

Mean (SD) 375.0 (143.7) 67.8 (65.0) 250.2 (191.8) 203.3 (122.7) 209.1 (87.0) 39.9 (24.4) 79.9 (26.5) 131.3 (60.9)

Range 190-585 235 123-780 46-470 —410 11-73 30-119 49-225
Kruskal-Wallis 14,3*** 5.0* }4.3*** 8.3*** 14.3*** 6.2"

SIPUDAS S1]DXCQ —"[e 19 d1OO|W

Cle

316 Rhodora [Vol. 96

by a suite of traits that includes; abundant, long, patent, nonsep-
tate hairs on the abaxial and adaxial leaf surfaces; and abundant.
long, patent septate and nonseptate hairs on the stems. Thus,
these plants have an easily observed dense covering of relatively
long hairs throughout the plant; a hand lens is not necessary to
distinguish this phenotype.

CHARACTER CORRELATIONS. In the Iron Mountain Gap
population, four of the 28 possible correlations between pairs of
characters were significant (Table 3). Several significant character
correlations (seven of the possible 28) occurred in the Watauga
River population (Table 3), and five of these were highly signif-
icant. The correlation matrix for the sample of herbarium spec-
imens consisted of eleven characters (three hair-length characters
were included in addition to the eight hair type characters). Six
of the 55 possible correlations were significant (Table 3). All six
significant correlations were positive and concerned characters
related to numbers of nonseptate hairs. Six of the ten possible
nonseptate/nonseptate correlations were significant. There were
no significant correlations involving septate/nonseptate pairs of
characters or any combination of hair length characters. In sum-
mary, there were significant correlations involving septate and
nonseptate hairs within and between organs in the Watauga River
sample, but in the herbarium sample, the only significant asso-
ciations involved nonseptate hairs.

SECTION II — POLLEN

METHODS

Oxalis grandis is morphologically tristylous, that is, “the stig-
mas occur at a level different from the level of either of the two
rings of anthers: either below both (short-styled), between them
(midstyled), or above both (long-styled)”’ (Eiten, 1963). For most
flowers, pollen was sampled from both anther whorls of each
flower. Slides of pollen samples were prepared from fresh material
in the field. A freshly dehisced anther was touched to a glass slide
that was then stained with lactophenol aniline-blue dye. Because
Oxalis grandis propagates clonally from rhizomes, sampled plants
were separated by a minimum of 3 m. The samples from each
population spanned the physical extent of each population and
encompassed an area of approximately 5 km? and 20 km? in the

1994] Moore et al.—Oxalis grandis 317

Table 2. Relative contribution of septate hairs (expressed as percent septate)
to pubescence on different parts of Oxalis grandis in two populations. The value
of the G statistic is given for each population comparison, asterisks refer to
probabilities associated with each value of G. M = Iron Mountain Gap, W =
Watauga River, ns = non-significant, ** <0.01, *** >0.001.

Population @ awenh

M WwW populations)
Petiole 24.7% 16.0% 2.4 ns
Upper stem 32.3% 15.3% 10.4**
Lower stem 75.7% 44.8% 30.2***
Upper vs. lower stem

(within populations)
33.4*** 95.8***

Iron Mountain Gap and Watauga River populations, respectively.
The diameters of ten pollen grains were measured from each
anther using an ocular micrometer at 400.

DATA ANALYSIS. For each style morph in each population,
a separate one-way analysis of variance (ANOVA) was used to
compare pollen diameters between the two anther whorls of a
morph. None of these comparisons showed significant differences
and therefore, anther samples from the two whorls of a flower
morph were pooled for comparisons among style morphs within
and between populations. Pollen diameters among style morphs
within populations were compared by one-way ANOVA followed
by Student-Newman-Keuls (SNK) comparisons among means
(SAS Institute, 1982). To compare pollen diameters between pop-
ulations, each morph was treated separately ina one-way ANOVA
followed by SNK comparisons among means.

RESULTS

Differences in pollen diameter between anther whorls were not
significant for any of the three flower morphs within either pop-
ulation, although sample sizes for some comparisons were low
(Table 5). Nevertheless, there were significant differences in pollen
diameter among flower morphs within each population. In both
populations, pollen from the short-styled morph was significantly
larger than pollen from the two other morphs (Table 6). For all
three flower morphs, pollen from plants in the Watauga River
population was larger than pollen from the corresponding morph

318 Rhodora [Vol. 96

able 3. Character correlations within (A) the Iron Mountain Gap population,
(B) the Watauga River population, and (C) a sample of selected herbarium spec-
imens of Oxalis grandis. For each significant correlation, the Pearson Correlation
Coefficient and its associated probability is shown. Character correlations that
were significant in two of the sample groups are denoted by a common symbol.

Shared
Correlated Characters Coefficient P Correlations

A. Iron Mountain Gap

StUpNS:StUpS 0.68 0.03 -

StUpNS:StLoNS 0.73 0.02 #

StLoNS:StUpS 0.70 0.02

StLoS:PetS 0.68 0.03 +
B. Watauga River

AbLf:AdLf 0.81 0.004 fe)

AbLf:StLoNS 0.65 0.04

StLoNS:PetNS 0.84 0.002

StLoNS:StLoS 0.82 0.004

StLoS:PetS 0.80 0.006 +

StLoNS:StLoS 0.82 0.004

StUpNS:StUpS 0.63 0.05 =
C. Selected Herbarium Sample

AbLf:AdLf 0.55 0.04 fe)

AdLf:PetNS 0.54 0.05

AdLf:StUpNS 0.84 0.0001

AdLf:StLoNS 0.82 0.0003

PetNS:StUpNS 0.72 0.004

StUpNS:StLoNS 0.58 0.03 #

Key to Acronyms: number of ru adaxial leaflet surface, AdLf; number of
hairs, abaxial leaflet surface, AbLf,; number of nonseptate hairs, petiole, PetNS:

lower stem, StLoNS; number of septate hairs, lower stem, StLoS: nonseptate hair
length, abaxial leaflet surface, LfHr; nonseptate hair length, stem and petiole,
StHrNS; septate hair length, stem and petiole, StHrS.

in the Iron Mountain population (Table 6) and these differences
were highly significant for the mid- and me style aul (ANO-
VA results: short-styled, F = 2.96; df = 1, 12; P = 0.11; mid-
styled, F = 26.6; df = 1, 15; P=0. 0001: nae a F= 8. 5: dr.
= 1, 31; P= 0.006).

DISCUSSION

PUBESCENCE. In a study of Oxalis section Corniculatae,
Wiegand (1925) noted extreme intraspecific variation in pubes-

1994] Moore et al.— Oxalis grandis 319

Table 4. Characteristics of herbarium specimens? of Oxalis grandis that showed
some leaf hairs. An ““X”’ indicates expression of the character was within the range
of variation observed hn the Watauga River population. ““Total W” indicates
the number of Watauga River-like characters of the eight characters analyzed on
each plant. “‘ # Plants” refers to the number of test plants (max. = 14) that show
the Watauga River-like pacity De for a cheraer C1 and C2 serve as controls
since they were Watauga River site sp llected prior to the current study.
Acronyms as in Table 3

Character Total
o & & 2 op
ml bar i=
a 3 2 2 3 & 3

Plant < < ow 4) n = n an W

Li x 1

ae xX ]

Be x 1

4. x 1

>. x x 2

6. x x 2

Fe x xX 2

8. x x x x 4

9, x x x x x 5
10. xX x x x x 5
Ide xX x x Xx x 5
12; x xX x x xX x 6
13. x x x xX 4 x 6
14. x x x x x xX 6
Cl: x x x x x xX x x 8
CO. Xx x x xX x xX x 5.4 8
# Plants 11 6 7 7 7 ) 0

a ees to herbarium specimens: |. vIRGINIA, Scott Co., summer 1958, E. Elliott
TENNESSEE STATE UNIVERSITY); 2. VIRGINIA, Wythe Co., open woods, 28

1 une i910. FSH s.n. (vpt); 3. KENTUCKY, Rockcastle Co., road bank along Rock-
castle River, 17 June 1961, H. E. Ahles 54501 with H. Smith (Unc); 4. KENTUCKY,
Pulaski Co., mixed hemlock-hardwoods, Daniel Boone National Forest, 23 July
1970, E. M. Browne and E. T. Browne Jr. 70K15.19 (UNC); 5. TENNESSEE, Unicoi
Co., Unaka Island, 8 June 1974, C. L. Shepard 168 (EAST TENNESSEE STATE
UNIVERSITY); 6. WEST VIRGINIA, Hancock Co., Newmans Bridge, 21 June 1963, J.
Bonar s.n. (WVA); 7. KENTUCKY, Menifee Co., Red River Gorge, rich moist woods,
28 May 1969, P. D. Higgins 1401 (UNC); 8. WEST VIRGINIA Raleigh Co., 25 May
1940, J. P. Tosh s.n. (WVA); 9. TENNESSEE, Pickett Co., north-facing bluff of Wolf
River, 3 May 1984, G. L. Walker, E. E. C. Clebsch, Z. E. Murrell 017 (TENN);
10, KENTUCKY, Edmonson Co., mesophytic woods beside Bylew Ck., 7 June 1968,
K. A. Nicely and H. W. Elmore 1718 (unc); 11. INDIANA, Floyd Co., shale knobs,
18 May 1963, A. C. Koelling 1065 (TENN); 12. TENNESSEE, Cocke Co., roadside
near Douglas Lake, 3 May 1966, B. Allen s.n. (TENN); 13. ALABAMA, Colbert Co.,
rich north-facing slope over sandstone, 24 May 1974, R. D. Whetstone and T.

320 Rhodora [Vol. 96

Table 5. ANOVA comparisons of pollen diameters (microns) between anther
whorls within style morphs of Oxalis grandis. n = number of anthers sampled.

Morph Anther Whorl n Mean F tag
Iron Mountain Gap
Short mid 2 35.45 0.07 0.80
long 3 34.9]
Mid short 6 29.85 0.22 0.65
long 6 30.53
Long short 12 32.90 1.70 0.21
mid 11 31.42
Watauga River
Short mid | 41.60 NA —
long 2 37.12
Mid short 2 39.81 5.20 0.11
long 3 35.54
Long short 4 34.14 2.01 0.19
mid 6 35.41

cence that he assumed arose from environmental and soil factors.
The exception, O. europea Jord. (= O. stricta L.), had stem and
pedicel pubescence that Wiegand attributed to the environment,
but on the basis of geographical correlates, he postulated that
hairiness on the abaxial leaf surface represented racial differen-
tiation. In the most recent and thorough treatment of the section,
Eiten (1963) contradicted Wiegand’s views and relied upon vari-
ability in pubescence as a key diagnostic character in dividing
section Corniculatae DC into subsections Corniculatae and Stric-
tae Eiten. The diagnostic characteristics of the latter subsection
were (1) the presence of septate hairs on stems, petioles and ped-
icels, and (11) stems that arose singly from thin underground rhi-
zomes (Eiten, 1963)

In contrast to our observations, prior studies have concluded
that leaf hairs in Oxalis grandis are rare. Wiegand’s (1925) de-
scription of O. grandis included presence of sparing pubescence

_
Atkinson 3198 (UNC); 14. TENNESSEE, Sinking Creek, moist woods, 11 June 1955,
J. Pearman S.n. (EAST TENNESSEE STATE UNIVERSITY); Cl. TENNESSEE, Washington
Co., shale; open woods, shale barren-like,l16 May 1990, K. Renzaglia s.n. (vpt);
C2. TENNESSEE, Washington Co., Watauga Flats, 5 May 1978, 7. Bruce s.n. (EAST
TENNESSEE STATE UNIVERSITY).

1994] Moore et al.—Ovxalis grandis 321

Table 6. ANOVA comparisons of pollen diameters (microns) among style
morphs of Oxalis grandis with SNK comparisons of mean diameters. Means
followed by the same superscript were not significantly different (P < 0.05).

Iron Mountain Gap

Source df MS F P
Among morphs 2 74.69 9.83 Q.0003
Error 43 7.60

Morph Hn Mean
Short ll 35.258
Long 12 32.19
Mid 23 30.19%

Watauga River

Source df MS F P
Among morphs 2 19.85 4.44 0.03
Error 15 4.47

Morph n Mean
Short 3 38.614
Mid 5 37.25%
Long 10 34.91°

on the stems, villous petioles and peduncles, and glabrous leaves
except for a few hairs on the underside; he did not distinguish
septate from nonseptate hairs. Similarly, in an exhaustive study
of section Corniculatae, Lourteig (1979) concluded that pubes-
cence is rarely found on the leaf blades in O. grandis. Both studies
noted a specimen (Cincinnati, C. G. Lloyd, 1882) that was un-
usually hairy on the stem with strigose abaxial leaf surfaces. Our
results indicate that leaf blade pubescence occurs sporadically
throughout the range of the species because nonseptate hairs were
seen on abaxial leaf surfaces in 16 of 174 herbarium specimens
examined.

Our results suggest that in Oxalis grandis, four different types
of pubescence characters contribute to the extremely hirsute phe-
notype —an increase in the number of (1) septate and (2) nonsep-
tate hairs, (3) elongation of both types of hairs throughout the
vegetative axis and leaves, and (4) reorientation of hairs from
appressed to patent. All except the last of these distinguishing
characteristics are quantitative, and in only one of these (length
of nonseptate hairs on the vegetative axis) is the extreme phe-

322 Rhodora [Vol. 96

notype (long hairs) restricted to the Watauga River population.
For the remaining quantitative characters, intermediate and ex-
treme phenotypes occur either as isolated characters, or in various
combinations.

Long, nonseptate hairs on stems occurred only in the Watauga
River population (Table 4, StHrNS, specimens C1, C2). Consid-
ering the sample of herbarium specimens, there were no corre-
lations involving the other two hair-length characters (LfHr;
StHrS). However, in individual specimens, these two characters
occurred singly (Table 4, specimen #4), in combination with each
other (Table 4, specimen #12), or in various, apparently random
combinations with other characters. From the apparent indepen-
dence of these characters, we conclude that the hair length com-
ponent of the Watauga River phenotype is a complex character.

In the screened herbarium sample, none of the character cor-
relations involving septate hairs was significant. Nevertheless, the
hirsute (Watauga River) phenotype was observed in eight speci-
mens for number of septate hairs on petioles, six specimens for
number of septate hairs on upper stems, and six specimens for
number of septate hairs on lower stems (data not shown). In two
specimens (Table 4, specimens #10, #13), septate hairs were ab-
sent on the upper stem, indicative of an extreme Iron Mountain
Gap phenotype, but the number of nonseptate hairs was clearly
typical of the Watauga River phenotype. In other herbarium spec-
imens, nonseptate hairs were sparse (Iron Mountain Gap phe-
notype) but septate hairs were abundant (Watauga River phe-
notype) in corresponding regions of the plants. Two lines of
evidence suggest that an increased number of septate hairs is
independent of other pubescence characters. First, there were no
significant character correlations involving septate hairs in the
herbarium sample. Second, hirsuteness for septate hairs occurred
in combination with extreme variation (encompassing both phe-
notypes) in the numbers of nonseptate hairs.

In the herbarium sample, there were no obvious ecological or
geographical correlates associated with the occurrence of pubes-
cent plants. Because herbarium sheets rarely contain more than
a few individual plants, it is not possible to know if pubescent
plants represent unusual plants or if, like the Watauga River pop-
ulation, the entire population is fixed for some increased degree
of hairiness. Nevertheless, the observation that plants represent-
ing both extremes of hairiness retained their characteristic phe-

1994] Moore et al.—Oxalis grandis 323

notypes after propagation in a common environment indicates a
genetic basis (Levy and Moore, 1993).

Although character correlations exist within the Iron Mountain
Gap sample (Table 4), these plants showed little tendency toward
the hirsute phenotype. Therefore, the meaning of these particular
correlations is unclear. The Watauga River population provided
the standard by which we defined the extreme hirsute phenotype
with multiple positive associations among pubescence characters.
However, the sample of selected herbarium specimens may pro-
vide a key to understanding the complexity of pubescence in
Oxalis grandis. These specimens were selected because each
showed a tendency towards hairiness. Within this sample, all
characters involving nonseptate hairs were positively correlated,
suggesting simple genetic control may underlie the increased num-
ber of nonseptate hairs. On the other hand, numbers of septate
hairs appeared independent of all other characters, and therefore
is probably under independent genetic control. Furthermore, nei-
ther of the two hair-number characters were correlated with in-
creased length of either type of hair.

If one considers only qualitative differences between phenotypic
extremes (e.g., patent versus appressed hairs; short versus long
nonseptate hairs), then it could be concluded that pubescence is
a simple character. For example, in Dithyrea (Rollins, 1958) and
Linum (Rogers, 1968) hairiness is controlled by a single genetic
locus. However, in Oxalis grandis we have provided several lines
of evidence involving qualitative as well as quantitative pubes-
cence attributes suggesting a more complex character. In the her-
barium sample we found (i) only one significant character cor-
relation in common with the Watauga River population, (11) no
significant correlations between septate and nonseptate hairs and,
(111) no individuals with long nonseptate hairs on stems or petioles
(unlike their ubiquitous occurrence in Watauga River). The ap-
parent complexity of the pubescence character in Oxalis grandis
endows it with significant value as an indicator of taxonomic
relationship but, the contrast between Dithyrea and Oxalis high-
light the necessity of a detailed character analysis for each taxon
under study.

POLLEN SIZE. Citing data from his own crosses, as well as
from Muller and Hildebrand, Darwin (1877) convincingly showed
that illegitimate pollination of several heterostylous Oxalis spe-
cies resulted in dramatically reduced seed production compared

324 Rhodora [Vol. 96

to legitimate pollinations. In O. regnellii, O. speciosa, and O.
valdiviana, pollen diameter was correlated with the stamen whorl
from which it was produced. The largest grains were produced in
long stamens, whether these stamens occurred on short- or mid-
styled flower morphs; the smallest pollen was found in short an-
thers. Pollen in Oxalis grandis also is heteromorphic. In both
populations of Oxalis grandis examined, short-styled morphs pro-
duced the largest pollen (Table 5). However, pollen diameter in
this tristylous species is largely a function of the style morph on
which it was produced; pollen from different anther whorls within
a style morph are similar in size.

Several heterostylous species in Oxalis subsection Corniculatae
have lost the self-incompatibility response and others show evi-
dence of loss of intermorph morphological differences (Ornduff,
1972). The three closely related species that constitute Oxalis
subsection Strictae (O. grandis, O. stricta, O. suksdorfii) are all
moderately to strongly self-compatible (Ornduff, 1964, 1972).
Moreover, pollen size in this section of the genus is not clearly
related to the length of the anther whorl. For example, long anthers
on short- and mid-styled morphs in O. suksdorfii produced the
largest pollen, but there was no difference in pollen size between
anther whorls in long-styled morphs. The size of pollen grains
from corresponding anthers on different style morphs differed
within long and mid anther whorls (Ornduff, 1964). Mid-styled
flowers showed strong pollen size differences between anther whorls
that was accompanied by physiological differences in the incom-
patibility reaction (Ornduff, 1964).

In some species of Oxalis, differences in pollen size have been
noted between populations as well as among anther whorls. For
example, pollen diameter differed among anther whorls from two
style morphs within each of three populations of O. dillenii ssp.
filipes, but the morph that lacked pollen size differentiation among
anther whorl types was not the same in each population (Ornduff,
1972). Oxalis alpina (section Ionoxalis) consists of di- and tri-
stylous populations. In the former, style morphs clearly differed
in the size of pollen produced, but there were no size differences
among anther whorls within style morphs (Weller, 1976). Pollen
appeared trimorphic in tristylous populations, but the magnitude
of differences between morphs was not as pronounced as in di-
stylous populations (Weller, 1976). A detailed analysis of the
trimorphic pollen of Oxalis pes-capre showed variation among

1994] Moore et al.—Oxalis grandis 325

populations when pollen from equivalent anthers was compared
(Ornduff, 1987). In the present study, pollen size was remarkably
different between the Watauga River and Iron Mountain Gap
populations within each flower morph. Pollen from the Watauga
River population always is larger and, for the mid- and long-
styled morphs, these differences are highly significant.

CONCLUSIONS

Two lines of evidence suggest that, in Oxalis grandis, pubes-
cence 1s genetically controlled. In prior studies, distinct pubes-
cence phenotypes retained their respective suites of characteristics
in a common environment. The current study uncovered the
sporadic occurrence of plants with varying degrees of a pubescent
phenotype, but these plants did not share obvious geographical
or ecological factors. Furthermore, the correlation analysis has
shown that components of pubescence vary independently in na-
ture. While a determination of the genetic basis of pubescence is
premature without a formal genetic analysis, the current findings
suggest a genetic architecture more complex than a single locus.
We postulate that the constellation of characters leading to an
extremely pubescent phenotype may arise from either the com-
bined actions of a minimum of three loci with major effects or it
may result from several polygenic characters. The complexity of
the phenotype tends to lend credence to Eiten’s emphasis on
pubescence as a diagnostic character in Oxalis section Cornicu-
atae.

Copious, long patent hairs and relatively large pollen grains
indicate the Watauga River population is an extreme variant of
Oxalis grandis. This population also differs in other attributes
from more typical representatives of the species. For example,
there were distinct differences from the Iron Mountain phenotype
with respect to the pattern of its circadian rhythm of leaflet sleep
movements as well as its responses to light level fluctuations (Levy
and Moore, 1993) which suggest some fundamental physiological
diversity. Finally, the Watauga River population occupies an un-
usual habitat—steep slopes underlain by easily eroded shale of
the Sevier formation. The resultant substrate and habitat is rem-
iniscent of the shale barrens of Virginia and West Virginia. These
sites are known to support regionally rare species and some areas
recently have been afforded protection. Future studies may show

326 Rhodora [Vol. 96

that the extremely pubescent phenotype is worthy of varietal sta-
tus but, for the present, we have chosen to adopt a conservative
approach and retain Oxalis grandis as a monotypic species but
to recognize its morphological and physiological diversity.

ACKNOWLEDGMENTS

We gratefully acknowledge support from the East Tennessee
State University Research Development Committee. We grate-
fully acknowledge the courtesy of the curators of TENN, UNC, VPI,
and wva for allowing examination of specimens from their col-
lections.

LITERATURE CITED

Darwin, C. 1877. The Different Forms of Flowers on Plants of the Same Species.
John Murray, London
Erren, G. 1963. Taconomy and regional variation of Oxalis section Corniculatae.
I. Introduction, keys and synopsis of the species. Amer. Mid]. Naturl. 69:
57-309

Levy, F. AND D. Moore. 1993. Population variation of leaflet sleep movements
in Oxalis grandis (Oxalidaceae). Amer. J. Bot. 80: 1482-1493.

LourtTeic, A. 1979. Oxalidaceae extra-austroamericanae. Phytologia 42: 57-
198.

OrRNDuFF, R. 1964. The breeding system of Oxalis suksdorfti. Amer. J. Bot. 51:
307-314.

1972. The breakdown of trimorphic incompatibility in Oxalis section
Comncuaiae Evol. 26: 53-65
Reproductive systems and chromosomal races of Oxalis pes-
caprae L. and their bearing on a noxious weed. Ann. Mo. Bot. Gard. 74:
Rocers, C. M. 1968. ae kak Aeclhiaa ee penne 70: 439-441.
Rotuins, R.C. 1958. in Dithyrea
(Cruciferae). are ake 60: 145-152.
SAS Institute. 1982. SAS User’s Guide: Statistics. Research Triangle Park, NC.
WELLER, S. G. 1976. Breeding system polymorphism in a heterostylous species.
Evol. 30: 442-454.
WIEGAND, K. M. 1925. Oxalis corniculata and its relatives in North America.
Rhodora 27: 113-124, 133-139.

DEPARTMENT OF BIOLOGICAL SCIENCES
EAST TENNESSEE STATE UNIVERSITY
JOHNSON CITY, TENNESSEE 37614

RHODORA, Vol. 96, No. 888, pp. 327-353, 1994

FLORISTIC DIVERSITY OF A DISTURBED
WESTERN OHIO FEN

JAMES S. MCCORMAC AND GREGORY J. SCHNEIDER

ABSTRACT

During the 1990-1992 growing seasons vegetation was surveyed in a Logan
County, Ohio fen. This 23 ha site we ee to De Toney disturbance
o 1985. Thousands
of cubic meters of peat and aa were enbied from the i and virtually the
entire surface area was affected by dredging activities. Surveys of the fen prior to
mining indicate that the site was occupied mostly by shrub communities domi-
nated by Cornus spp., Rosa palustris, and other woody species. However, shrubs
are infrequent at the present time. Post-disturbance surveys documented a total
of 208 vascular plant taxa, including 22 species listed as rare in Ohio. Several
species are present which are disjunct from areas in the state where they normally
occur, or are not known elsewhere in western Ohio.

Key Words: bog fen, disjuncts, peat mining, prairie fen, rare plants, substrate
disturbance

INTRODUCTION

Fens are alkaline peatlands which survive as relict wetland
communities scattered throughout glaciated Ohio. Prior to Eu-
ropean settlement, peatlands were estimated to cover 74,000 ha
in Ohio (Dachnowski, 1912). Recent work by Andreas and Knoop
(1992) indicates that approximately 98% of the state’s peatlands
have been destroyed. The primary cause of this loss has been
conversion of land for agricultural purposes, accounting for 85%
of the total loss of peatlands. Additional factors contributing to
the disappearance of these wetlands are recreation (vacation cot-
tages, camps, etc.), modification of hydrology, mining and other
development. Other more insidious factors causing a decrease in
diversity and an eventual elimination of typical fen associations
are natural succession into woody plant communities and inva-
sion by non-native plants. Alien species, in particular Rhamnus
frangula L. (European Buckthorn), have detrimentally impacted
many of Ohio’s fens. Left unchecked, this species can proliferate
to the point of eliminating much of the native flora.

Numerous rare plant taxa, and a few species of rare animals,
are confined to or exist primarily in fens in Ohio. Many of the
plants require the combination of marl (calcium carbonate pre-

2)

328 Rhodora [Vol. 96

cipitate derived from artesian spring waters) substrate, cool sub-
surface soil temperatures, and saturated soil conditions. Most of
the species which are exclusive to fens in Ohio occur 1n a wider
variety of habitats elsewhere. In Ohio, many of these species are
on the edge of their range, and therefore require the specialized
combination of habitat parameters in fens to gain a competitive
advantage over more generalized wetland species which are un-
able to tolerate the harsh conditions within fens. Rare plants
exclusive to fens in Ohio include: Carex flava L., Carex sterilis
Willd., Cladium mariscoides (Muhl.) Torr., Eleocharis pauciflora
(Lightf.) Link., Eriophorum viridicarinatum (Engelm.) Fern.,
Triglochin maritimum L., Triglochin palustre L., Tofieldia glu-
tinosa (Michx.) Pers., Zigadenus elegans Pursh var. glaucus (Nutt.)
Preece, Cypripedium calceolus L. var. parviflorus (Salisb.) Fern.,
Spiranthes romanzoffiana Cham., Deschampsia caespitosa (L.) P.
Beauv., Cacalia plantaginea (Raf.) Shinners, Solidago ohioensis
Riddell, Utricularia cornuta Michx., Myrica pensylvanica Mirbel.,
Salix myricoides (Muhl.) J. Carey, and Valeriana edulis Nutt. var.
ciliata (T. & G.) Cronq. (Anon., 1992). Uncommon animals often
found in fens include Clemmys guttata Schneider (Spotted Turtle)
and Sistrurus catenatus Rafinesque (Massasauga Rattlesnake).

STUDY AREA

The study site, known as McCracken Fen, is located in west-
central Ohio near the city of Bellefontaine, county seat of Logan
County (Figure 1). This region is characterized by low, rolling ©
hills formed by a series of kames and eskers deposited by the
Wisconsin ice sheet. McCracken Fen is situated on a large terminal
moraine which spans most of Logan County. Artesian springs
which emanate from deep gravel deposits are frequenct in this
area, as evidenced by names of local towns; Bellefontaine (French
for beautiful fountain), Big Springs, and Springhills. The abun-
dance of springs had led to the formation of numerous fens in
the Champaign-Clark-Logan county region (Schneider, 1992), al-
though many of these now have been destroyed (Andreas and
Knoop, 1992).

McCracken Fen is situtated in a bowl-like depression surround-
ed by rather abruptly sloping gravel ridges on three sides. The
south side is bordered by a small tract of Elm-Ash swamp forest
which grades into agricultural fields. The fen was mined for peat

1994] McCormac and Schneider—Ohio Fen 329

Pe eT
aL oe ClevSland_|
ce er Te
ge Sg

“pet | J -
: a S

a r 7
incinnati _t [ 2

Figure 1. Location of McCracken Fen, Logan County, Ohio.

and marl products over a period of approximately five years,
beginning in 1980. Many tons of substrate were removed from
nearly all areas of the peatland. This dredging caused intensive
disturbance to the surface of the wetland, and is still obvious in
the form of ditches, furrows, open flats, and artificial deepwater
pools (Figure 2). In 1985, the mining company apparently went
bankrupt, and operations abruptly ceased. Heavy equipment such
as draglines, cranes, tractors, and a small peat packaging plant
were abandoned and are still present at the site.

The hydrology of the fen is maintained by artesian springs
entering the wetland from the east. Although mining activities
altered the natural openings of these springs, they continue to flow

330 Rhodora [Vol. 96

Figure 2. Post-disturbance aerial photograph of McCracken Fen, Logan Coun-
ty, Ohio. Taken June, 1990, by ODNR.

ata rapid rate. Attempts to drain McCracken Fen via ditches and
drainpipes were unsuccessful, as the volume of water entering the
fen appears to have offset the amount which was drained off.

METHODS

Field surveys of McCracken Fen began in September of 1989.
A chance visit to the site by McCormac resulted in the discovery
of Scirpus smithii Gray (Smith’s Bulrush), which at that time was
known from only one small, possibly extirpated, population in
Ohio. During 1990, 1991 and 1992, eleven collecting trips were
made to the fen, between April and October. Two hundred and
eight species of vascular plants were collected and voucher spec-
imens were deposited in the following herbaria: CLM, KE, MICH,
Mu, and os (Holmgren et al., 1990). An effort was made to record
frequency and abundance of all plant taxa, based on field obser-
vations, and using codes supplied by Reznicek and Catling (1989).
Detailed information regarding plants listed as rare in Ohio (Anon.,
1992) is on file in the Ohio Natural Heritage database.

Standard regional manuals were used for identification of spec-

1994] McCormac and Schneider—Ohio Fen ao

imens, particularly Fernald (1950), Gleason and Cronquist (1991),
and Voss (1972, 1985). Nomenclature follows Gleason and Cron-
quist (1991).

RESULTS AND DISCUSSION

Rare Plants. Of the 208 species of vascular plants collected
in McCracken Fen, one is considered endangered in Ohio, seven
are threatened, and fourteen are potentially threatened (Anon.,
1992)

Many of these species are rare in Ohio due to their specificity
of habitat, and the majority are obligate fen species. As most of
the state’s original fen communities have been destroyed or great-
ly altered, many of the plants typically associated with this type
of wetland have become rare and local. Rare species are placed
into categories based on guidelines established by the Division
of Natural Areas and Preserves (Anon., 1992). Species classified
as endangered in Ohio are restricted in distribution to an area
delineated by three or fewer U.S. Geological Survey 7.5 minute
topographic maps. Threatened species occur on between four and
seven quadrangle maps or have fewer than ten total populations
statewide. Potentially threatened species are believed to be de-
clining in abundance in Ohio and have been placed in this “watch”
category due to their dependence on fragile ecosystems which
have become increasingly rare.

The presence of species with “‘threatened” (T) and “endan-
gered” (E) status in McCracken Fen is noteworthy, and these
plants are discussed individually below:

Betula pumila L. (T): This northern shrub is known from only
six extant populations in Ohio, mostly in the northern part of the
state. The nearest population to McCracken fen is at Cedar Bog
in adjacent Champaign County, a well-studied site (Frederick,
1974) which is floristically similar to McCracken Fen.

Carex bebbii (L.H. Bailey) Fern. (T): Another northern species
known from ca. eight sites in northern Ohio, most of which are
on the lake plain of Lake Erie. The McCracken Fen population
is disjunct from the nearest Ohio site by 120 km, and is one of
the most southerly populations in the midwest.

Carex sartwellii Dewey. (T): This plant is known from about
ten sites in Ohio, mostly in the western half of the state. Many
years these populations remain largely sterile, thus rendering de-

LEe Rhodora [Vol. 96

tection difficult. Field observations by the authors in recent years
suggest that fire, and possibly other forms of disturbance, may
induce flowering.

Eleocharis flavescens (Poiret) Urban, var. olivacea (Torr.) Glea-
son. (T): This species is very habitat specific in Ohio, as all eight
extant populations occur in seasonally exposed, saturated peat,
usually on the drying shores of lakes or ponds. With the exception
of a site in Champaign County, the nearest population to Mc-
Cracken Fen is ca. 130 km to the north.

Pogonia ophioglossoides (L.) Ker Gawler. (T): This was one of
the most exciting discoveries, as this orchid was reported only
once before on the till plains of western Ohio, from a peat bog
which straddled the Champaign-Logan county line and has since
been destroyed. Only five small populations of Rose Pogonia are
known to be extant in Ohio, primarily in the northeastern quarter
of the state.

Scirpus smithii A. Gray. (E): One of the rarest plants in Ohio,
Smith’s Bulrush was known from only one site in Ohio, on the
shore of Lake Erie in Ottawa County. The plant has not been seen
at the Lake Erie site in recent years; therefore, the McCracken
Fen population may be the only extant site in Ohio. It should be
noted that we treat Scirpus smithii A. Gray and S. purshianus
Fern. as distinct species; Ohio material of the former is easily
separated from the latter. Gleason and Cronquist (1991) submerge
these taxa under the name Scirpus smithii.

Sphenopholis obtusata (Michx.) Scribn. var. obtusata (T): Cur-
rently known from only five sites in Ohio, most records of this
easily overlooked and/or misidentified species are from the west-
ern half of Ohio.

Utricularia intermedia Hayne. (T): Eight populations are known
in Ohio, from glacial lakes in northern Ohio, and saturated marly
ground of a few fens in west central Ohio.

Aliens. Non-native species accounted for 8% of the vascular
flora of McCracken Fen, or 18 species. However only two of these,
Rhamnus frangula and Solanum dulcamara L. have become es-
tablished within the fen, although they are not yet common. The
remaining alien species persist locally in the transition zone be-
tween upland and fen, or are occasional on drier hummocks in
the fen. Whereas non-native plants are well-known colonizers of
disturbed ground (Muenscher, 1935), the low incidence of aliens
in the flora of McCracken Fen, in spite in the recent heavy dis-

1994] McCormac and Schneider—Ohio Fen 333

turbance, may be attributable to three factors. The substrate of
fens are consistently low in temperature, low in oxygen avail-
ability, and are nutrient deficient (Van der Valk, 1977). These
factors create an environment suitable only for an assortment of
plants which have adapted to these conditions, and may account
for the relative lack of colonization by non-native species.
Disjuncts. Many species in the flora of McCracken Fen are of
a northern and/or coastal plain affinity. A number of these plants
reach the southern limits of their range in Ohio in the extreme
northern and northeastern sections of the state, and also occur
sparingly as disjuncts in the fens and bogs of central Ohio. Ex-
amples of this type of distribution which are found at McCracken
Fen include Betula pumila, Carex bebbii, Eleocharis flavescens
var. olivacea, Eriophorum viridicarinatum, Rhynchospora alba
(L.) Vahl., Scirpus smithii, and Utricularia intermedia (Figure 3).

PRE-DISTURBANCE PLANT COMMUNITIES

Information regarding the condition of McCracken Fen prior
to the mining operation 1s scarce; however the site was known to
botanists and was visited several times. Pehaps the best docu-
mentation of pre-disturbance conditions are aerial photographs
of the fen taken by the Ohio Department of Natural Resources
(ODNR) in 1978 (Figure 4). These pictures indicate that the fen
was largely vegetated by woody shrubs such as Cornus spp. and
Rosa palustris Marshall. An extensive stand of Scirpus acutus
Muhl. is also evident. When the 1978 photograph (Figure 4) is
compared with a photo taken in 1992 (Figure 2), the change in
composition of vegetation caused by the mining operating is ev-
ident.

Botanists Allison W. Cusick and Guy L. Denny, both employees
of ODNR, also visited McCracken Fen in the mid-1970’s, prior
to mining. They state (pers. comm.) that the fen was practically
impenetrable due to the dense shrub zones, and with the exception
of Potentilla fruticosa L., no unusual or state-listed plants were
observed.

POST-DISTURBANCE PLANT COMMUNITIES

Six distinct zones of vegetation exist in present-day McCracken
Fen. With the exception of the open water habitat, and weedy

334 Rhodora [Vol. 96

Betula pumila

Carex bebbu

Figure 3. Ohio distribution of species which occur as disjuncts in western Ohio
fens, including McCracken Fen, Logan County, Ohio

peripheral zone, these vegetation zones resemble those which
occur naturally in many other undisturbed Ohio fens. As these
zones were artifically created as a side-effect of the peat-mining
operations, they differ in some respects from naturally occurring

abitats. Within several of the zones there are also recognizable
microhabitats.

1994] McCormac and Schneider—Ohio Fen 335

— ge

—

Figure 4. Pre-disturt ial photograph of McCracken Fen, Logan County,
Ohio. Taken August, 1978, by ODNR.

Open Water Areas. This habitat shows the most visible effect
of the mining activities. Two large areas, each ca. 1.5 ha in size,
were dredged to a much greater depth than other areas of the fen,
leaving ponds which vary from 30 cm to 5 m in depth. Fed by
artesian springs, these ponds have exceptionally clear water and
support twelve species of aquatic plants, as well as a number of
species dependent on the open, marshy ground bordering the
ponds. It is interesting that the composition of flora inhabiting
the open water areas is similar to that which is found in nearby
glacial lakes. In addition to the two large ponds, a number of
much smaller open water areas exist in the fen, usually in the
form of deeply dredged ditches.

Aquatic plants characteristic of the open water habitat include
Brasenia schreberi J.F. Gmelin, Lemna minor L., Najas flexilis
(Willd.) Rostkov & Schmidt, Nuphar advena (Aiton) Aiton i;
Nymphaea odorata Aiton, Potamogeton illinoensis Morong, P.
pectinatus L., and Utricularia vulgaris L.. Species typical of sat-
urated open soil bordering water areas include Alisma subcor-
datum Raf., Carex comosa F. Boott., C. lurida Wahlenb., Eleo-
charis palustris L., Eupatorium perfoliatum L., Leersia oryzoides

336 Rhodora [Vol. 96

(L.) Swartz, Mimulus ringens L., Sagittaria latifolia Willd., Scir-
pus acutus, and Verbena hastata L.

Mar! Flats. Open marl flats provide habitat for some of the
rarest obligate fen plants in Ohio. Marl flats are invariably as-
sociated with outflows of spring water containing calcium and
magnesium salts. These water-borne minerals precipitate after
reaching the surface and combine with peat to form a substratum
of marl throughout the fen.

Water flow through McCracken Fen has been altered to the
point where naturally forming marl flats are not present. However,
marly peat flats have been created by the peat excavating process.
In several areas, small flats were created when the overlying peat
layer was removed, exposing the underlying marl. These artificial
marl flats are saturated to the surface by ground water, and are
habitat for an unusual assortment of plants, including many rare
or uncommon species. Plant taxa typical of the marl flats include
Cyperus bipartitus Torr., C. flavescens L., Eleocharis tenuis (Willd).
Schultes var. borealis (Svenson) Gleason, Fimbristylis autumnalis
(L.) Roemer & Schultes, Hypericum majus (A. Gray) Britton,
Juncus articulatus L., J. brachycephalus (Englem.) Buchenau, J.
canadensis J. Gay., J. nodosus L., Rhynchospora alba, R. capil-
lacea Torr., Scleria verticillata Muhl., and Spiranthes cernua (L.)
Rich. Noteworthy is the almost complete dominance of two fam-
ilies, the Cyperaceae and Juncaceae. The frequency of these two
groups in soils that were heavily disturbed by mining activities
suggests a strong capability to store seeds in a viable state for an
extended period deep within the substratum.

Another type of flat occurs within McCracken Fen and is quite
different in composition of substrate and cause of origin, than the
above described marl flats. Autumnally exposed flats occur along
the margins of the open water areas, due to a slight (+15 cm)
drop in water levels, which begin to recede in mid-summer. By
September these areas are fully exposed. The substrate of these
shoreline flats is composed of unconsolidated, saturated peat which
is very unstable, and low in diversity of flora. However, two of
the rarest species present in the fen, Eleocharis flavescens var.
olivacea and Scirpus smithii, occur in this habitat. Other species
commonly found in this zone are: Bidens cernua L., Cyperus
odoratus L., and Ludwigia palustris (L.) Elliott.

Potentilla Fruticosa Meadow. This is the largest plant com-
munity in the fen, occupying ca. 5 or 6 ha. This habitat appears

1994] McCormac and Schneider—Ohio Fen 337

to be nearly a monoculture of Potentilla when viewed from a
distance, although it is easily the most diverse zone in McCracken
Fen. Slight changes in moisture regime influence supporting veg-
etation, as does the presence of Sphagnum. The driest areas of
the meadow are relatively low in species diversity, while wetter
areas support an impressive array of plants. Some of the dominant
species in this community include Aster puniceus L., A. umbel-
latus Miller, Campanula aparinoides Pursh, Cladium maris-
coides, Eupatorium maculatum L., Galium tinctorium L., Lobelia
kalmii L., Muhlenbergia glomerata (Willd.) Trin., Potentilla fru-
ticosa, Scutellaria lateriflora L., Solidago ohioensis, S. patula
Muhl., S. rugosa Miller, Thelypteris palustris Schott., and Tox-
icodendron vernix (L.) Kuntze.

ne small section in the wettest area of the Potentilla meadow
contains a lush growth of Sphaghum sp., and harbors several
vascular plants found nowhere else in McCracken Fen. These are
Calopogon tuberosus (L.) BSP., Drosera rotundifolia L., Epilobi-
um leptophyllum Raf., Eriophorum viridicarinatum, and Pogonia
ophioglossoides.

Graminoid Meadow. This is the rarest vegetation zone, oc-
cupying only a small (<5%) area of the total fen. Two different
types of meadow are present, one dominated by grass (Poaceae)
and the other composed primarily of sedges (Cyperaceae). The
grass dominated meadow is ca. 0.4 ha in size, and is a virtual
monoculture of Ca/lamagrostis stricta (Timm.) Koeler. Species
diversity in this meadow is quite low.

The sedge meadows are quite small, averaging only a few square
meters. These occur primarily within the Potentilla fruticosa
meadow, as small, randomly scattered openings. The dominant
species found in the sedge meadows include Aster borealis Prov.,
Carex flava, C. interior L. Bailey, C. leptalea Wahlenb., Epilobium
coloratum Biehler, Scutellaria galericulata L., and Triadenum
fraseri (Spach) Gleason.

Shrub Zone. In contrast to conditions in the fen prior to min-
ing, shrub zones now cover a relatively small area. Two distinct
areas of this habitat, covering ca. 3-4 ha, are present. One of the
shrub zones is dominated primarily by Cornus amomum Miller
and Rosa palustris. This area is one of the least diverse habitats
in the fen, as the dense, virtually impenetrable thickets exclude
most herbaceous growth. The other shrub zone is dominated by
a variety of woody species and shade tolerant herbaceous plants.

338 Rhodora [Vol. 96

Dominant woody species include Betula pumila, Cornus amo-
mum, Fraxinus pennsylvanica Marshall, [lex verticillata (L.) A.
Gray, Populus deltoides Marshall, Rosa palustris, Salix discolor
Muhl., S. sericea Marshall, and U/mus rubra Muhl. The under-
story herbaceous stratum of this thicket 1s quite diverse. Some of
the dominant species include Agrimonia parviflora Aiton, Aster
lateriflorus (L.) Britton, Boehmeria cylindrica (L.) Swartz, Clema-
tis virginiana L., Cuscuta gronovii Willd., Galium tinctorium,
Glyceria striata (Lam.) A. Hitche., Jmpatiens capensis Meerb.,
Onoclea sensibilis L., Osmunda regalis L., and Solidago gigantea
Aiton,

Weedy Peripheral Zone. Few non-native plant species have
attained a foothold within McCracken Fen, the two notable ex-
ceptions being Rhamnus frangula and Solanum dulcamara. As
most of the perimeter of this fen grades abruptly into dry, gravelly
slopes, much of which has been disturbed by mining activities,
there exists a narrow zone characterized by non-native species
bordering the fen. Most of these alien plants do not penetrate far
into the fen, and for the most part, do not occur in heavy con-
centrations. Two species which have become locally abundant in
wetter areas of this zone are Phalaris arundinacea L. and Typha
angustifolia L.

Some of the more common non-native plants occurring in the
weedy peripheral zone include Achillea millefolium L., Brassica
nigra L., Chrysanthemum leucanthemum L., Convolvulus arvensis
L., Daucus carota L., Nepeta cataria L., and Setaria viridis (L.)
P. Beauv.

FLORISTIC COMPARISON WITH
OTHER OHIO FENS

Stuckey and Denny (1981) analyzed the floristic affinities of
Ohio’s fens. They concluded that two distinctive types of fens
occurred in the state. They used the term prairie fen to describe
fens in west-central and south-central Ohio. These fens have a
distinctive floristic affinity with the wet, tall-grass prairies of the
midwestern United States. The term bog fen was introduced to
describe fens which occur in northeastern and extreme north-
western Ohio. These fens share many species with Ohio’s kettle
bogs. These species have northern affinities and are often abun-
dant in the boreal fens of northern Michigan, northern Minnesota
and Canada. Stuckey and Denny (1981) recognized that Cedar

1994] McCormac and Schneider— Ohio Fen 339

Bog, a fen in west-central Ohio, was unique to the state in that
its flora included both the prairie and boreal elements. McCracken
Fen, which is located 32 km north of Cedar Bog, was relatively
unknown until recently and was not discussed by Stuckey and
Denny. This fen has some of the boreal species which otherwise
occur mostly in northern fens or at Cedar Bog. These species
include Betula pumila, Eriophorum viridicarinatum, Rhyncho-
spora alba and Pogonia ophioglossoides. Another interesting facet
of McCracken Fen’s flora is the absence of many of the species
of wet prairie affinity which are so characteristic of all of the other
fens in this part of the state. Examples of species typical of Ohio
prairie fens which do not occur at McCracken Fen include An-
dropogon gerardii Vitman, Cacalia plantaginea (Raf.) Shinners,
Coreopsis tripteris L., Liatris spicata (L.) Willd., and Silphium
terebinthinaceum Jacq.

CONCLUSIONS

Although no thorough surveys were made of McCracken Fen
prior to the mining operation, evidence strongly suggests that the
study area was in an advanced state of succession. Pre-mining
photographs and accounts of observers indicate that shrub zones
blanketed most of the fen, greatly reducing diversity. Mining ac-
tivities removed most of the shrub zones, as well as some of the
substrate. During the mining process, soils in the fen were heavily
disturbed by the heavy equipment used to excavate the peat. It
would seem likely that the combination of eliminating most of
the shrub zones, thus restoring open fen meadow/marl flat habitat,
and stimulating the seedbank by disrupting the substrate, resulted
in increased diversity. Many of the plant taxa now present in
McCracken Fen are characteristic of fens in Ohio, yet it is unlikely
that they colonized the site from outside sources, as there are no
other fens in the immediate vicinity. The dramatic shift in veg-
etation resulting from the mining operation suggests that fen soils
are capable of storing large, viable seedbanks, and that restoration
of fens in advanced successional states is possible.

ANNOTATED LIST OF THE VASCULAR PLANTS OF
MCCRACKEN FEN, LOGAN COUNTY, OHIO

All species included on this list are represented by voucher
specimens deposited in herbaria, as indicated following each spe-

340 Rhodora [Vol. 96

cies. Nomenclature and phylogenetic order follow Gleason and
Cronquist (1991). Each species is assigned a frequency code, as
follows: rare, occasional, or abundant; and local, scattered or
widespread (Reznicek and Catling, 1989). Habitat(s) in which
each species occurs is given, using the following abreviations: GM
(Graminoid Meadows), MF (Marl Flats), OW (Open Water), PM
(Potentilla fruticosa Meadow), SZ (Shrub Zones) and WP (Weedy
Peripheral Zone). Finally, our collection numbers for all taxa are
given. Superscripts preceding species’ names indicate: 1, non-
native; 2, endangered; 3, threatened; and 4, potentially threatened.

PTERIDOPHYTES

EQUISETACEAE
Equisetum fluviatile L.—Occasional and local in SZ (3675 os).

OSMUNDACEAE
Osmunda regalis L.—Occasional and scattered in SZ (1692 os;
2724 MU).

ASPLENIACEAE
Thelypteris palustris Schott.—Abundant and widespread in PM,
GM, and SZ (31/&/ os).
Dryopteris cristata (L.) A. Gray—Occasional and local in SZ
(4065 MU).

ONOCLEACEAE
Onoclea sensibilis L.—Occasional and local in SZ (365/ os).

DICOTYLEDONS

NYMPHAEACEAE
Nuphar advena (Aiton) Aiton f.—Abundant and local in OW
(2415 MU, OS).
Nymphaea odorata Aiton—Abundant and local in OW (2620
CLM).

CABOMBACEAE
Brasenia schreberi J.F. Gmelin— Occasional and local in OW
(2417 MU).

1994] McCormac and Schneider— Ohio Fen 341

RANUNCULACEAE
Caltha palustris L.—Rare and local in SZ (4525 CLM, MU).
Anemone virginiana L.—Rare and local in PM (3917 KE).
Clematis virginiana L.—Occasional and local in SZ (2880 mu,

OS).

Ranunculus recurvatus Poiret— Rare and local in SZ (3650 os).
R. abortivus L.—Rare and local in SZ (3649 os).

BERBERIDACEAE
Podophyllum peltatum L.—Rare and local in SZ (4527 Mv).

PLATANACEAE
Platanus occidentalis L.—Rare and local in SZ (39/5 os).

ULMACEAE
Ulmus rubra Muhl.—Occasional and local in SZ (3666 os).

URTICACEAE
Urtica dioica L. var. procera (Muhl.) Wedd.— Occasional and
scattered in WP (3/73 os).
Boehmeria cylindrica (L.) Swartz— Occasional and scattered in
OW, PM, and SZ (3179 mv).
Pilea pumila (L.) A. Gray—Occasional and local in SZ (3/82
OS).

BETULACEAE
3Betula pumila L.—Occasional and local in PM and SZ (2616
CLM, OS).

PORTULACACEAE
Claytonia virginica L.—Rare and local in SZ (4528 mv).

POLYGONACEAE
'Rumex crispus L.—Rare and local in WP (367/ os).
Polygonum lapathifolium L.—Occasional and scattered in OW,

WP (3165 os, 3347 Os).
P. pensylvanicum L.—Occasional and scattered in OW (3162

Os).
'P. hydropiper L.—Rare and local in OW (3190 os).

342 Rhodora [Vol. 96

P. punctatum Elliott—Occasional and scattered in OW (3050
OS).

'P. persicaria L.— Rare and scattered in OW and WP (3049 os).

P. sagittatum L.—Occasional and scattered in SZ (3174 Mu).

P. scandens L.—Rare and local in SZ (3049 os).

CLUSIACEAE

Hypericum punctatum Lam.—Rare and local in MF (2883 os;
3028 OS).

H. multium L.—Occasional and local in MF (1694 KE; 2876
MICH; 3029 MICH).

4H. majus (A. Gray) Britton— Occasional and local in MF (2877
MICH; 3043 os).

Triadenum fraseri (Spach) Gleason— Occasional and scattered
in GM, OW, and PM (/684 os).

DROSERACEAE
4Drosera rotundifolia L.—Occasional and local in PM (26/1]
OS).

VIOLACEAE
Viola sororia Willd. — Occasional and scattered in PM, SZ, and
WP (4535 MU).
V. macloskeyi F. Lloyd—Occasional and local in SZ (4526 KE,
MU).

SALICACEAE
Populus grandidentata Michx.—Rare and local in SZ (26/4 os).
P. deltoides Marshall— Occasional and scattered in SZ and WP

(3643 os).
Salix amygdaloides Andersson—Rare and local in WP (3645

Os).

S. lucida Muhl.— Occasional and local in SZ (3674 MU).

S. exigua Nutt.— Occasional and local in SZ and WP (26/7 mu;
3644 Os).

S. sericea Marshall— Occasional and local in SZ (3655 os).

S. discolor Muhl.— Occasional and local in SZ (3/9/ os; 3656
CLM).

1994] McCormac and Schneider—Ohio Fen 343

BRASSICACEAE
‘Brassica nigra L.—Rare and local in WP (4785 Mv).
Rorippa palustris (L.) Besser— Rare and local in MF (4769 mv).

PRIMULACEAE
Lysimachia quadriflora Sims.—Abundant and widespread in
GM, MF, and PM (4766 Mv).
L. thyrsiflora L.—Rare and local in SZ (2420 KE).

GROSSULARIACEAE
Ribes americanum Miller—Occasional and local in SZ (3661
os; 4529 MU).

SAXIFRAGACEAE
Penthorum sedoides L.—Occasional and scattered in GM, OW,
and PM (31/64 os).

ROSACEAE
Fragaria virginiana Duchesne—Rare and local in WP (4531

MU).
Potentilla norvegica L.—Rare and local in WP (3654 os; 3916

KE).

P. fruticosa L.— Abundant and local in PM (1672 mu; 1706 os).

Geum canadense Jacq.— Rare and local in SZ (3038 os).

G. laciniatum Murray—Occasional and scattered in PM, SZ,
and WP (3037 MU).

Rubus occidentalis L.—Occasional and scattered in SZ (4771
MU).

Agrimonia parviflora Aiton—Occasional and scattered in PM
and SZ (3176 Mv).

Rosa palustris Marshall— Abundant and scattered in SZ (4783

MU).
Amelanchier spicata (Lam.) K. Koch.—Rare and local in SZ
(4534 os).

LYTHRACEAE
Lythrum alatum Pursh— Occasional and scattered in MF (3346
KE, OS

344 Rhodora [Vol. 96

Epilobium leptophyllum Raf.—Rare and local in PM (3039 mu;
3160 os).

E. coloratum Biehler— Occasional and scattered in GM, OW,
and PM (3/89 os).

Oenothera biennis L.—Rare and local in MF and WP (2881
MU).

CORNACEAE
Cornus amomum Miller— Abundant and scattered in SZ (3032
OS; 3662 MU).
Cornus sericea L.—Occasional and scattered in SZ (2418 Mu,
Os).

AQUIFOLIACEAE
Tlex verticillata (L.) A. Gray — Occasional and local in SZ (1685
OS; 3163 KE; 3652 MU).

RHAMNACEAE
'Rhamnus frangula L.—Occasional and scattered in PM and
SZ (4061 MU).

VITACEAE
Parthenocissus vitacea (Knerr) A. Hitchc.—Rare and local in
PM (4778 Mu, OS).
Vitis riparia Michx.— Occasional and local in SZ (3665 os).

ANACARDIACEAE
Rhus glabra L.—Rare and local in WP (4765 Mv).
Toxicodendron vernix (L.) Kuntze—Occasional and scattered

in PM and SZ (1/703 os).

OXALIDACEAE
Oxalis stricta L.—Occasional and scattered in MF, PM, and

WP (3033 MU).

BALSAMINACEAE
Impatiens capensis Meerb.—Abundant and scattered in OW
and SZ (4066 Mu; 4780 MU).

1994] McCormac and Schneider—Ohio Fen 345

APIACEAE
Sanicula gregaria E. Bickn.—Rare and local in SZ (3668 os).
' Daucus carota L.—Occasional and scattered in WP (3025 os).
Cicuta bulbifera L.—Occasional and local in OW (3/68 0s).

GENTIANACEAE
Gentiana andrewsii Griseb.— Occasional and local in PM and
SZ (3175 Mu; 3331 MU).

APOCYNACEAE
Apocynum cannabinum L.—Occasional and scattered in PM
and WP (3657 os).

ASCLEPIADACEAE
Asclepias incarnata L.—Occasional and scattered in GM, PM,
and OW (3027 cLM).

SOLANACEAE
1Solanum dulcamara L.— Occasional and scattered in OW, PM,
and SZ (3653 os).

CONVOLVULACEAE
1Convolvulus arvensis L.—Rare and local in WP (3647 os).

CUSCUTACEAE
Cuscuta gronovii Willd. — Occasional and scattered in OW, PM,
and SZ (1695 MU, OS).

BORAGINACEAE
1Cynoglossum officinale L.—Rare and local in WP (3659 OS).

VERBENACEAE
Verbena hastata L.—Occasional and scattered in GM, OW,
PM, and WP (2722 os; 3024 Os).

LAMIACEAE
Scutellaria lateriflora L.—Occasional and widespread in GM,
OW, PM, SZ, and WP (3/85 mv).

346 Rhodora [Vol. 96

S. galericulata L.—Occasional and scattered in GM, OW, and
PM (2618 cLm; 2725 MU).

S. nervosa Pursh—Occasional and local in OW and WP (4781
MU).

Lycopus virginicus L.—Occasional and scattered in GM, OW,
and PM (3/70 mu).

L. americanus Muhl.—Occasional and scattered in GM, OW,
and PM (3/88 os).

'Nepeta cataria L.—Rare and local in WP (39/4 KE).

Prunella vulgaris L.—Occasional and scattered in PM, SZ, and
WP (3030 os).

OLEACEAE
Fraxinus pennsylvanica Marshall—Rare and local in SZ (4770
MU).

SCROPHULARIACEAE
Mimulus ringens L.—Occasional and scattered in OW and PM
(2879 os).
Agalinis purpurea (L.) Pennell—Occasional and scattered in
M, MF, OW, and PM (1699 os).

LENTIBULARIACEAE
‘Utricularia intermedia Hayne—Rare and local in OW (3334

KE).
U. vulgaris L.—Abundant and local in OW (262/ cLM:; MU).
U. gibba L.—Occasional and local in OW (3042 os).

CAMPANULACEAE
Campanula aparinoides Pursh— Abundant and scattered in GM
and PM (2723 mu; 4776 Mv).
Lobelia kalmii L.—Occasional and widespread in GM and PM
(1700 os).
L. siphilitica L.—Occasional and scattered in OW, PM, SZ, and
WP (3186 MU).

RUBIACEAE
Galium trifidum L.—Occasional and local in PM (2885 os; 3034
MU).

1994] McCormac and Schneider— Ohio Fen 347

G. tinctorium L.—Occasional and scattered in GM, OW, PM,
and SZ (39/8 CLM).

G. triflorum Michx.—Rare and local in SZ (392/ KE).

G. aparine L.—Rare and local in SZ and WP (3648 os).

CAPRIFOLIACEAE
Viburnum lentago L.—Rare and local in SZ (3664 os).
Sambucus canadensis L.—Occasional and local in SZ (3810
MU).

VALERIANACEAE
Valerianella umbilicata (Sulliv.) A. Wood—Rare and local in
OW and WP (26/3 os).

ASTERACEAE

Rudbeckia hirta L.—Occasional and local in WP (3040 mv).

Bidens cernua L.—Occasional and scattered in MF, OW, PM,
and WP (/690 mv).

B. coronata (L.) Britton— Occasional and scattered in MF, OW,
PM, and WP (1/675 mu; 1683 os).

Ambrosia artemisiifolia L.—Rare and local in WP (3159 mu).

‘Achillea millefolium L.—Occasional and local in WP (4774

1):
‘Chrysanthemum leucanthemum L.—Occasional and local in
WP (4777 MU).
Senecio aureus L.—Occasional and scattered in PM and WP
(4530 MU).
Erechtites hieracifolia (L.) Raf.—Rare and local in WP (3333
Os)

Solidago patula Muhl.— Occasional and scattered in PM (1707

MU).

S. rugosa Miller—Occasional and scattered in PM and WP
(1696 os).

S. gigantea Aiton—Occasional and scattered in PM and WP
(3046 Mu; 4069 KE).

S. canadensis L.—Occasional and local in WP (4986 mv).

4S. ohioensis Riddel— Abundant and widespread in PM (1677
MU).

348 Rhodora [Vol. 96

S. riddellii Frank— Rare and local in PM (1702 CLM).

Euthamia graminifolia (L.) Nutt.— Occasional and local in WP
(3158 Mv).

Aster borealis Prov.—Rare and local in PM (1/705 os).

A. puniceus L.—Abundant and widespread in GM, OW, PM,
and WP (1674 os).

A. praealtus Poiret— Occasional and scattered in PM (/68/ os).

A. lateriflorus (L.) Britton—Occasional and scattered in GM,
PM, and WP (/701 os).

A. pilosus Willd.—Occasional and local in WP (/687 os).

A. novae-angliae L.—Rare and local in WP (4985 mu).

A. umbellatus Miller—Abundant and widespread in GM and
PM (1704 os).

Erigeron strigosus Muhl.—Rare and local in WP (3036 Os).

Conyza canadensis (L.) Cronq.—Rare and local in WP (3167
MU).

Eupatorium maculatum L.—Occasional and scattered in GM,
OW, and PM (/682 mv).

FE. perfoliatum L.—Occasional and scattered in OW and PM
(3184 MU).

Cirsium muticum Michx.—Occasional and widespread in PM
(1697 os).

'C. vulgare (Savi) Tenore—Occasional and scattered in WP
(3035 os).

'Taraxacum officinale Weber—Rare and scattered in PM and
WP (4532 MU).
'Cichorium intybus L.— Occasional and local in WP (4784 Mv).

MONOCOTYLEDONS
ALISMATACEAE
Alisma subcordatum Raf.—Occasional and scattered in OW
(3023 CLM).

Sagittaria latifolia Willd.—Occasional and scattered in OW
(3026 MU).

POTAMOGETONACEAE
Potamogeton pectinatus L.—Abundant and widespread in OW
(4060 os).
P. illinoensis Morong— Abundant and widespread in OW (2728
Os, CLM; 2886 OS, CLM; 3049 Os).

1994] McCormac and Schneider—Ohio Fen 349

NAJADACEAE
Najas flexilis (Willd.) Rostkov & Schmidt—Abundant and
widespread in OW (3992 MU).

LEMNACEAE
Lemna minor L.—Abundant and scattered in OW (3336 os).

COMMELINACEAE
'Commelina communis L.—Rare and local in WP (3157 Mv).

JUNCACEAE

Juncus canadensis J. Gay — Occasional and scattered in MF and
PM (1676 os; 3172 MU).

J. brachycephalus (Engelm.) Buchenau— Occasional and scat-
tered in MF (1679 os; 1686 MU).

J. torreyi Cov.— Occasional and scattered in GM, MF, and PM
(3057 CLM).

J. nodosus L.—Occasional and local in MF (4782 Mv).

J. articulatus L.—Occasional and scattered in MF (3056 CLM;
4988 Os).

J. tenuis Willd. var. dudleyi (Wieg.) F.J. Herm.—Occasional
and widespread in GM, MF, OW, PM, and WP (3192 mu;
3660 MU).

CYPERACEAE
2Scirpus smithii A. Gray — Abundant and scattered in MF (91/671
os; 1698 MICH; 316] MU; 3345 KE).
S. acutus Muhl.— Occasional and local in OW (3045 os; 4779

MU).

S. validus Vahl.— Occasional and local in OW (3183 os).

S. fluviatilis (Torr.) A. Gray—Rare and local in GM (4767 Mv).

S. atrovirens Willd.—Occasional and scattered in OW (2727
MU; 3055 os).

S. pendulus Muhl.—Rare and local in GM (4769 Mv).

S. cyperinus (L.) Kunth.—Occasional and local in OW (3171

MU).

4Eriophorum viridicarinatum (Engelm.) Fern.— Rare and local
in PM (26/2 CLM, Os).

Eleocharis tenuis (Willd.) Schultes var. borealis (Svenson) Glea-

350 Rhodora [Vol. 96

son—Occasional and scattered in GM and MF (2623 cL”,
os; 3058 os; 4070 micH).

E. palustris L.—Occasional and scattered in OW (3332 Mv).

3F. flavescens (Poiret) Urban var. olivacea (Torr.) Gleason—
Abundant and local in MF (3166 os; 3348 KE).

Fimbristylis autumnalis (L.) Roemer & Shultes— Abundant and
scattered in MF (J688 os; 1693 Mu; 3044 CLM).

*Rhynchospora alba (L.) Vahl.—Occasional and local in MF
(2882 MU, OS).

R. capillacea Torr.—Abundant and scattered in MF (/69/ Mv).

*Cladium mariscoides (Muhl.) Torr.—Occasional and local in
PM (2884 os).

CG ee us strigosus L.—Occasional and scattered in OW (3051

LM).

Cc. ee L.—Occasional and scattered in MF (4073 Mu).

C. flavescens L.—Occasional and local in MF (4073 mv).

C. bipartitus Torr.—Occasional and local in MF (1678 os).

Dulichium arundinaceum (L.) Britton— Occasional and local in
78 MU)

*Scleria verticillata Muhl.— Occasional and scattered in MF and
PM (3041 os).

Carex leptalea Wahlenb.— Abundant and scattered in GM and
PM (2419 cLM, os).

3C. sartwellii Dewey —Rare and local in PM (900] os).

C. vulpinoidea Michx.—Occasional and scattered in OW and
WP (2427 Mu, OS; 2622 Os).

C. stipata Muhl.—Occasional and scattered in OW and WP
(2425 MU).

*C. diandra Schranck.—Rare and local in PM (3658 os).

C. interior L. Bailey—Occasional and scattered in GM and PM
(2430 MICH; 2431] os).

°C. bebbii (L.H. Bailey) Fern.—Rare and local in PM (2627
MICH; 3672 OS).

4C. suberecta (Olney) Britton—Rare and scattered in GM and
PM (2421 MICH, MU; 2628 MICH).

‘C. alata T. & G.—Rare and local in PM (2629 micu; 3673 os).

C. tetanica Schk.—Occasional and scattered in GM and PM
(2426 Os).

C. granularis Muhl.— Occasional and scattered in OW and WP
(2424 MU).

*C. flava L.—Occasional and widespread in GM and PM (2423
CLM, MU, OS).

1994] McCormac and Schneider—Ohio Fen 351

C. crinita Lam.—Occasional and scattered in OW and WP
(3350 mv).

C. comosa F. Boott— Occasional and scattered in OW and WP
(2625 Os; 3663 MU).

C. lacustris Willd. — Occasional and scattered in OW (2429 os).

4C. utriculata F. Boott— Occasional and scattered in OW (2626
OS; 3667 MU).

C. lurida Wahlenb.— Occasional and scattered in OW and WP
(2428 Mu, OS; 3919 KE).

POACEAE

Leersia oryzoides (L.) Swartz— Occasional and scattered in OW
and WP (31/78 os).

Festuca subyerticillata (Pers.) E. Alexeev.—Rare and local on
hummocks in SZ (3670 os).

3§phenopholis obtusata (Michx.) Scribn. var. obtusata—Rare
and local in PM (3669 Mv).

Phalaris arundinacea L.— Abundant and local in WP (2721 os;
4775 MU).

4Calamagrostis stricta (Timm) Koeler— Abundant and local in
GM (3920 mv).

Agrostis hyemalis (Walter) BSP. var. scabra (Willd.) Blomq.—
Occasional and scattered in MF, PM, and WP (4072 os).

'Phleum pratense L.—Rare and local in PM (4772 Mv).

Muhlenbergia mexicana (L.) Trin.—Occasional and scattered
in PM (3177 os).

M. glomerata (Willd.) Trin.—Occasional and scattered in PM
(4068 MU).

Panicum dichotomiflorum Michx.—Occasional and scattered
in WP (407] Mv).

P. flexile (Gattinger) Scribn.— Occasional and scattered in MF,
PM, and WP (3054 os).

P. lanuginosum Elliott var. implicatum (Scribn.) Fern.—Oc-
casional and scattered in PM and WP (3053 os).

Echinochloa muricata (P. Beauv.) Fern.— Occasional and scat-
tered in OW and WP (3187 mu; 4987 os).

TYPHACEAE
Typha latifolia L.—Abundant and scattered in OW and WP
(3059 MU).
'T. angustifolia L.—Occasional and scattered in WP (4773 Mu).

52 Rhodora [Vol. 96

IRIDACEAE
Tris virginica L.—Occasional and scattered in OW and PM
(2416 MU, OS).

ORCHIDACEAE
Spiranthes cernua (L.) Rich.— Occasional and local in MF (3/67
CLM, OS).
Habenaria lacera (Michx.) Lodd.—Rare and local in SZ (2720

Os).

*Pogonia ophioglossoides (L.) Ker Gawler— Occasional and lo-
cal in PM (2609 cL, os).

*Calopogon tuberosus (L.) BSP.—Rare and local in PM (2719

OS).
Liparis loeselii (L.) Rich.—Rare and local in SZ (2610 cLM).

ACKNOWLEDGMENTS

The authors thank Allison W. Cusick, Guy L. Denny, and Mar-
shal A. Moser for information regarding the condition of Mc-
Cracken Fen prior to mining. Stanley J. Stine provided assistance
with Sa/ix, and Anton A. Reznicek offered help with identification
of the more confusing material, particularly in Carex. Two anon-
ymous reviewers made numerous suggestions which greatly im-
proved the manuscript. Finally, we are grateful to the Ohio Di-
vision of Natural Areas and Preserves for supporting our research.

LITERATURE CITED

ANDREAS, B. K. AND J. D. KNoop. 1992. 100 years of changes in Ohio peatlands.
Ohio J. Sci. 92(5): 130-138.

ANON. 1992. Rare native Ohio plants. Ohio Dept. Natur. Res., Division of
Natural Areas and Preserves. Columbus, OH. 20

DacuHnowskI, A. 1912. Peat deposits of Ohio: Their origin, formation and uses.
Bull. Geol. Surv. Ohio, 4th Ser., Bull. 16, Div. Geol. Surv., Columbus, OH.
424 pp

FERNALD, M. L. 1950. ae s manual of botany, 8th Ed. American Book Co.,
New York, NY. 1632 p

FREDERICK, C. M. 1974. A ae, history study of the ed flora of Cedar
Bog. Champaign County, Ohio. Ohio J. Sci. 74: 65-116.

ied H. A. AND A. CronguistT. 1991. Manual of ee plants of North-
eastern United States and adjacent Canada, 2nd Ed. The New York Botanical
Garden, Bronx, NY. 910 pp

1994] McCormac and Schneider— Ohio Fen 353

HoL”maren, P. K., N. H. HOLMGREN, AND L. C. apie Eds. 1990. Index
herbariorum, part 1, 8th ed. Regnum Veg. 120: x 3.
MUENSCHER, W. C. 1935. Weeds. The MacMillan . <a York, New York.

Reznicek, A. A. AND P. M. CATLING. 1989. Flora of Long Point, regional mu-
nicipality of Haldimand-Norfolk, Ontario. Michigan Bot. 28: 99-17

SCHNEIDER, G. J. 1992. Cyperaceae of Ohio fens: floristic analysis and phyiee
eat ae relationships. M.S. thesis, The Ohio State University, Colum-
bus, pp.

STUCKEY, “i L. AND G. L. Denny. 1981. Prairie fens and bog fens in Ohio:
floristic similarties, differences and geographical affinities, pp. 1-33. In: R.

. Romans, Ed., Geobotany II. Plenum Press, NY.

VAN DER VALK, A. G. 1977. The effects of fertilization on Iowa fen communities.
Paper presented at the Wauseba Wetlands Conference, June 2-5, 1977, 1 LE.S.,
University of Wisconsin, Madison

Voss, E.G. 1972. Michigan flora, Pt. 1. Gymnoperms and monocots. Cranbrook
By Sci. Bull. No. 55, Bioomild Hills, MI. 488 pp.

. Michigan flora, Pt. 2. D (S iraceae-Cornaceae). Cranbrook
ssh Set i No. 59 and the University of Michigan Herbarium, Ann
Arbor, MI. 724 p

OHIO DEPARTMENT OF NATURAL RESOURCES
DIVISION OF NATURAL AREAS AND PRESERVES
1889 FOUNTAIN SQUARE, BUILDING F
COLUMBUS, OHIO 43224

RHODORA, Vol. 96, No. 888, pp. 354-369, 1994

GENETIC VARIATION IN MASSACHUSETTS
POPULATIONS OF CYPRIPEDIUM ARIETINUM
ROWN IN AIT. AND C. ACAULE
AIT. (ORCHIDACEAE)

ALAN H. BoRNBUSCH, LESLEY A. SWENDER,
AND DEBORAH L. HOOGERWERF

ABSTRACT

Horizontal starch gel electrophoresis of isozymes was conducted on samples
from the single currently known Massachusetts population of Cypripedium arie-
tinum (ram’s head lady’s slipper) and two populations of C. acaule (pink lady’s

er). No genetic variation was detected at 27 presumptive loci in the C. ari-
etinum population and only slight variation among 26 and 19 presumptive loci,
respectively, in the C. acaule populations. Levels of polymorphism and hetero-
zygosity in all populations were exceedingly low for animal pollinated obligate
outcrossers. Genetic distance between the C. acaule populations was also low.
Yearly and possibly overall sana population sizes in C. acaule and C. arietinum

be low due to low flower production and/or fruiting rate. Low levels o
intrapopulation aon may eee result from genetic drift and founder effect,
as proposed for C. acaule by Gill (1989). Low variation between C. acaule pop-
ulations, though, indicates that genetic bottlenecks in ancestral populations may
also have contributed to low species-level variation, as proposed by Case (1994)
for Michigan populations of C. acaule, C. arietinum, C. candidum, and C. reginae.
These results suggest that low genetic variation in declining Cypripedium popu-
lations may not be anthropogenic. These populations may have been genetically
depauperate throughout much of their existence prior to human disturbance.

short term, more significant threats to Cypripedium populations than is a relative
lack of genetic variation.

Key Words: Cypripedium arietinum, Cypripedium acaule, lady’s slipper orchids,
genetic variation, conservation

INTRODUCTION

How much variation exists in natural populations and what
factors contribute to lower than expected intraspecific variation
are questions of longstanding interest to evolutionary geneticists.
These questions have also more recently assumed significance for
conservation biologists. Low levels of variation in populations
experiencing anthropogenic declines can represent ‘“‘genetic de-
terioration” from higher levels that existed prior to disturbance
(Simberloff, 1986). Low variation, though, can result from factors
unrelated to modern anthropogenic disturbances. In plants, these
factors include bottlenecks in ancestral populations due to climate

354

1994] Bornbusch et al.— Cypripedium 355

induced range contractions, adaptation to specialized and stable
environments, evolution of primarily selfing mating systems with
selection against homoygotes for certain alleles, and reproductive
strategies that lower effective population sizes (Barrett and Kohn,
1991; Karron, 1991). Assessing the relative contributions of an-
thropogenic and natural factors to lowered genetic variation in a
rare species can help define proximate threats to the species’ sur-
vival and thus guide formulation of appropriate conservation
strategies.

The genus Cypripedium (lady’s slipper orchids) in North Amer-
ica includes 12 species (Luer, 1975; Wiard, 1987). Some species
(e.g., C. arietinum R. Brown in Ait., C. candidum Muhl. ex Willd.,
C. reginae Walter) are experiencing rangewide declines and are
listed by states as endangered, threatened or of special concern.
Recent studies, based on Virginia or Michigan populations, have
proposed that C. acaule Ait., C. arietinum, C. candidum, and C.
reginae are depauperate in genetic variation due to life history
characteristics and/or ancestral bottlenecks (Case, 1994; Gill,
1989). Here, we expand the database for assessing these hypoth-
eses by documenting low levels of variation in Massachusetts
populations of C. acaule and C. arietinum.

Lady’s slipper orchids are noted for elaborate flowers that at-
tract pollinators that enter the pouch-like labellum and can escape
only by a pathway that facilitates cross-pollination. With only
two known exceptions (Cypripedium dickinsonianum Hagsater,
C. passerinum Richardson in Franklin), the complex floral struc-
tures also prevent self-pollination in the absence of pollinators
(Catling, 1990; Stoutamire, 1967). Thus, although the flowers
appear to offer no nectar reward or other food to pollinators, most
Cypripedium species are obligate outcrossers dependent on insect
pollination for sexual reproduction (van der Pil and Dodson,
1966; Stoutamire, 1967).

Gill (1989) suggested that a dependence on insect pollination
in combination with attraction by deceit has significant demo-
graphic consequences for populations of Cypripedium acaule. In
a 10-year field study ofa Virginia C. acaule population, Gill (1989)
found the highest percentage of fruits produced by flowering plants
in a given year was 7.1%. Six out of ten years, fruiting rates were
below 1%, and in four of those years, no fruits were produced in
a population with a geometric mean of 680.6 plants. Gill (1989)
attributed these low fruiting rates to the failure of non-rewarding

356 Rhodora [Vol. 96

flowers to attract pollinators repeatedly. Low rates of fruiting in
Cypripedium have been observed in other populations, including
C. acaule in Massachusetts, New Hampshire and New Brunswick
(Barrett and Helenurm, 1987; Brackley, 1985; Davis, 1986; Plow-
right et al., 1980), C. arietinum in Wisconsin (J. Bender, pers.
comm., 1992) and C. candidum in Ontario (Catling and Knerer,
1980; Falb and Leopold, 1993). Moreover, some of these studies
and our observations indicate that annual flower production is
also low in some Cypripedium populations. Populations of some
Cypripedium species may thus have low effective population sizes
(N., Hartl and Clark, 1989) in a single year due to low flower
production and/or fruiting rates (Gill, 1989; Kimura and Crow,
1963). Furthermore, because establishment of a C. acaule pop-
ulation requires a multiplicative series of improbable events (e.g.,
seed dispersal and germination, seedling survival to reproductive
maturity, and cross-pollination between founders), Gill (1989)
suggests that C. acaule populations are subject to pronounced
founder effects. Because the severities of genetic drift and founder
effect are inversely proportional to N,., these factors can lower
genetic variation within populations with low N,’s (Hartl and
Clark, 1989). It has thus been hypothesized by Gill (1989) that
C. acaule populations should be particularly low in genetic vari-
ation because of reproductive patterns that substantially reduce
N,’s. We note here, though, and discuss more fully below, that
calculating overall N. for a Cypripedium population is problem-
atic. Overall N, may be higher than indicated by annually low
N.S:

Although genetic drift can reduce intrapopulation variation, it
can increase genetic variance among isolated populations, es-
pecially among populations that are founded by few individuals
(Hartl and Clark, 1989). Because drift would act independently
among isolated Cypripedium acaule populations, variation that
is partitioned among these populations is thus expected to be
significant. Case (1994), though, documented low variation within
and between Michigan C. acaule populations. Similar patterns
were observed in C. arietinum, C. candidum and C. reginae from
the same region. Case (1994) thus proposed that ancestral pop-
ulations of these species experienced genetic bottlenecks due to
habitat loss and fragmentation during Pleistocene glaciations.
Furthermore, extant conspecific populations in the Great Lakes
region may have been derived by recolonization of glaciated areas

1994] Bornbusch et al.— Cypripedium 357

from a single genetically depauperate source population. Popu-
lations so derived would be low in variation and genetically sim-
ilar to each other (Case, 1994).

This study is a genetic analysis of three Cypripedium popula-
tions: the only currently known Massachusetts population of C.
arietinum (ram’s head lady’s slipper), and two of C. acaule (pink
lady’s slipper). Although C. arietinum has been described as lo-
cally abundant at some sites, 1t has historically been rare through-
out its range, and anthropogenic habitat loss, collecting and biotic
succession within habitats are possible causes of recent rangewide
declines (Case, 1964; Luer, 1975; The Nature Conservancy, 1990).
Several states officially list C. arietinum as endangered (e.g., Mas-
sachusetts, Minnesota), threatened (e.g., Maine, New York) or of
special concern (e.g., Michigan) (Beaman et al., 1985; Coffin and
Pfannmuller, 1988; Dibble et al., 1989; Sorrie, 1989; Young,
1992). Cypripedium acaule, by contrast, is common throughout
its range, and is one of the most common orchids in North Amer-
ica: (Luer, 1975);

The objectives of this study were to: (1) measure and compare
genetic variation in the three Cypripedium populations using hor-
izontal starch gel electrophoresis of isozymes; and (2) determine
whether patterns of genetic variation observed by Case (1994) for
some Cypripedium species in the Great Lakes region extend to
these New England populations.

METHODS

The Cypripedium arietinum population 1s located in the vicinity
of Mount Toby, Sunderland, Franklin County, MA. Although the
history of this population is uncertain, C. arietinum has been
noted from the Mt. Toby area since at least the nineteenth century
(Tuckerman and Frost, 1875). Stone (1913, p. 9) described the
species as ‘often found ... quite plentifully” at a Mt. Toby site.
The present C. arietinum population occurs at a site that 1s sea-
sonally moist and includes eastern hemlock (7Tsuga canadensis),
and, as less abundant species, witch hazel (Hamamelis virgi-
niana), red maple (Acer rubrum), gray and black birches (Betula
populifolia, B. lenta), shagbark hickory (Carya ovata), American
basswood (T7i/ia americana), and chestnut oak (Quercus prinus).
A 1992 census identified 111 above-ground plants (ramets) (4
flowering) in the C. arietinum population (P. Martin-Brown, pers.

358 Rhodora [Vol. 96

comm., 1993). Fifty-seven plants, including all flowering plants
and plants representing the size range in the population, were
sampled by removal of several cm? of leaf from each plant.

Two Cypripedium acaule populations were sampled: one in
Northampton, Hampshire County, MA and another near the C.
arietinum population. The Northampton site includes as the most
abundant woody species mountain laurel (Ka/mia latifolia), east-
ern white pine (Pinus strobus), white (B. papyrifera) and gray
birches, witch hazel, and red and sugar (4. saccharum) maples;
other abundant species include Lycopodium sp. and hay-scented
(Dennstaedtia punctilobula) and Christmas (Polystichum acros-
tichoides) ferns. The Northampton site has experienced small-
scale brush dumping which does not appear to have significantly
affected the C. acaule population. We estimated the size of this
population in 1992 as at least 250-300 above-ground plants
(ramets). The Northampton population was sampled as follows.
From the approximate center of the population eight 50 m tran-
sects were run. The transects were oriented according to randomly
chosen compass headings; two headings were chosen in each of
the four 90° quadrants (i.e., 0-89°, 90-179°, etc.). At every five
meters along each transect, the nearest C. acaule plant not more
than 2.5 m from the transect was sampled. Thirty-four plants
were thus sampled. Additional plants were sampled within areas
of high plant density.

All observed individuals (40), including flowering and non-
flowering plants, were sampled from the Mt. Toby Cypripedium
acaule population. This population, like the nearby C. arietinum
population, occurs at a site dominated by eastern hemlock and,
as less abundant species, gray birch and witch hazel.

Cypripedium acaule plants were sampled by removal of leaf
tissue as for C. arietinum. All samples from the three populations
were immediately placed on ice in the field and later stored at
—80°C no longer than six months prior to preparation for elec-
trophoresis.

Samples were ground in one of three grinding buffers (1 mg
tissue per 0.005 ml buffer; Table 1). Grinding buffers were pre-
pared according to Soltis et al. (1983) with the exception that
2-mercaptoethanol was added during grinding. Wicks dipped in
ground tissue were used immediately for electrophoresis or stored
at —80°C for up to nine days prior to use. Storage of wicks had
no detectable effect on electrophoretic resolution.

1994] Bornbusch et al.— Cypripedium 359

able 1. Enzymes assayed from sampled Cypripedium arietinum and C. acaule
populations, and combinations of grinding and gel/electrode buffers used in elec-
trophoresis. Enzyme abbreviations (EC numbers in parentheses): acid phospha-
tase, APH (3.1.3.2); aconitase, ACN (4.2.1.3); aldolase, ALD (4.1.2.13); aspartate
aminotransferase, AAT (2.6.1.1); esterase, EST (3.1.1.-); fructose-1,6-diphospha-
tase, F-1,6-PD (3.1.3.11); glutamic dehydrogenase, GDH (1.4.1.2); isocitrate de-
hydrogenase, IDH (1.1.1.42); leucine aminopeptidase, LAP (3.4.11.-); malate de-
hydrogenase, MDH (1.1.1.37); malic enzyme, ME (1.1.1.40); peroxidase, PER
(1.11.1.7); phosphoglucoisomerase, PGI (5.3.1.9); phosphoglucomutase, PGM
(5.4.2.2); 6-phosphoglucose dehydrogenase, 6-PGD (1.1.1.44); shikimate dehy-
drogenase, SKDH (1.1.1.25); triosephosphate isomerase, TPI (5.3.1.1). Grinding
buffers named and gel/electrode buffer systems numbered as in Soltis et al. (1983).

Gel/Electrode
Buffer

Grinding Buffer

. arle-
tinum — C. acaule

Enzyme C. arietinum C. acaule

APH phosphate tris-HCl 2 2
ACN _ tris-HC1 “ 8
ALD _ phosphate a 1]
AAT tris-HCl tris-HC 1 8 8
EST phosphate tris-HCl 7 7
F-1,6-PD _ phosphate 1]
GDH phosphate phosphate 7 7
IDH tris-maleate tris-HCl 2 2
LAP phosphate tris-HCl 2 5
MDH tris-HCl tris-HCl 5 5
ME tris-maleate tris-maleate 7 5
PER tris-maleate tris-maleate 2 8
PGI phosphate tris-HCl 2 2
PGM phosphate phosphate 2 2
6-PGD tris-HCl tris-maleate 5 5
SKDH phosphate phosphate 2 2
TPI phosphate phosphate 2 2

Methods of gel preparation, loading and slicing followed in
general those outlined by Murphy et al. (1990). Details of our
methods can be provided by the senior author upon request.
Buffer systems were discontinuous and gel and electrode buffers
were prepared according to Soltis et al. (1983). Combinations of
grinding, gel and electrode buffers that provided satisfactory elec-
trophoretic resolution were established for 17 enzyme systems
(Table 1). Enzyme stains were prepared according to Soltis et al.
(1983); leucine aminopeptidase (LAP) was stained according to

360 Rhodora [Vol. 96

Wendel and Weeden (1989). Following staining, all gel slices were
fixed in 50% glycerol or ethanol and stored at 2°C for future
reference. Interpretations of observed banding patterns were based
on reviews of quaternary structures and isozyme numbers by
Gottlieb (1981) and Weeden and Wendel (1989). Isozymes were
numbered and alleles at a locus lettered sequentially cathodal to
anodal. Interpopulation comparisons for C. acaule were facili-
tated by co-electrophoresis with the same individual (from the
Northampton population) as a standard for both populations.

RESULTS

For C. arietinum, satisfactory electrophoretic resolution was
obtained for 27 presumptive loci (14 enzyme systems). The av-
erage sample size per locus was 36.0 individuals (Table 2). No
variation was detected at any locus in our sample (Figure 1).

For the Northampton C. acaule population, satisfactory elec-
trophoretic resolution was obtained for 26 presumptive loci (16
enzyme systems). The average sample size per locus was 36.7
individuals (Table 2). Allelic variation was observed at four loci
(allele frequencies in parentheses): LAP (a—0.04, b—0.96); PER-2
(a—0.22, b—0.78); PGI-1 (a—0.08, b—0.92, Fig. 2a); PGM (a—
0.01, b—0.99, Fig. 2b). The proportion of polymorphic loci was
15.4%; polymorphism by 95% criterion, 7.7% (Ayala, 1982); av-
erage number of alleles per locus (standard error in parentheses),
1.15 (0.072); average effective number of alleles per locus, 1.03
(0.021) (Hartl and Clark, 1989); and unbiased estimate of average
heterozygosity per locus, 0.023 (0.015) (Nei, 1978).

or the Mount Toby C. acaule population, satisfactory elec-
trophoretic resolution was obtained for 19 presumptive loci (15
enzyme systems). The average sample size per locus was 27.9
individuals (Table 2). Allelic variation was identified at one locus
(allele frequencies in parentheses): PGI-2 (a—0.015, b—0.985).
The proportion of polymorphic loci was 5.3%; polymorphism by
95% criterion, 0%; average number of alleles per locus (standard
error in parentheses), 1.05 (0.053); average effective number of
alleles per locus, 1.00 (0.002); and unbiased estimate of average
heterozygosity per locus, 0.0016 (0.0016).
e two C. acaule populations were isomorphic at all loci that
were invariant in both populations and no locus was polymorphic
in both populations. At a locus that was polymorphic in one

1994] Bornbusch et al.— Cypripedium 361

Table 2. Sample — iia scored from sampled populati of Cypripe-
dium arietinum and C. aca

C. acaule
ar n n
C. arietinum ene éveunt
Locus n Locus ampton) Toby)

T 38 AAT-1 36 40
APH-1 40 AAT-2 36 20
APH-2 40 ACN 33 —
APH-3 40 ALD-1 35 —
APH-4 40 ALD-2 32 --
EST-1 40 APH 42 34
EST-2 40 EST-1 42 36
EST-3 39 EST-2 42 35
EST-4 39 F-1, 6-PD _ 26
EST-5 36 GD 36 26
GDH 36 IDH-1 41 21
IDH 39 IDH-2 35 —
LAP-1 27 LAP 37 22
LAP-2 30 MDH-1 42 37
MDH-1 29 MDH-2 38 _
MDH-2 30 MDH-3 38 a
MDH-3 33 MDH-4 42 _
ME 40 ME 42 40
PER-1 21 PER-1 35 10
PER-2 39 PER-2 30 11
PGI-1 40 PGI-1 39 33
PGI-2 40 PGI-2 42 33
PGM 35 PGM 38 13
6-PGD 23 6-PGD 42 40
SKDH-1 40 SKDH 21 24
SKDH-2 40 TPI-1 32 29
TPI 39 TPI-2 27 —

population, the most common allele was fixed in the other pop-
ulation. Genetic distance between the two C. acau/e populations,
estimated at 18 loci according to Nei (1978), was thus low, 0.0028.

DISCUSSION

The levels of genetic variation measured from the three sampled
Cyp among the lowest measured for plants
by eae wrcihode: a are pone ous for animal pollinated
outcrossers. Hamrick and Godt (1990), for example, found that

302 Rhodora [Vol. 96

“2. "ae aeeeeree = «ant

A
peagen gel bate “2
44 Ped be og Be

Figure |. Electrophoretic patterns observed in Cypripedium arietinum popu-
lation for: A, esterase; B, phosphoglucoisomerase. Isozymes numbered on right
as in Table 2. Each gel shows eleven individuals, each loaded twice, and one
control individual. Anode is toward the top on each gel.

A

ae
fi

b
lt '
owen ooo sr *-

B

Figure 2. Electrophoretic patterns observed in Northampton a
acaule population for: A, phosphoglucoisomerase-1 (seven individuals show
arrows indicate two heterozygotes with homodimeric (aa, bb) and eee eat
(ab) bands); B, phosphoglucomutase (11 individuals, each loaded twice, and one
control individual shown; arrows indicate heterozygote with two monomeric (a,
b) bands). Anode is toward the top on each gel.

1994] Bornbusch et al.— Cypripedium 363

the average proportion of polymorphic loci among populations
of 80 monocots was 40.3% (standard error 3.0). Obligate out-
crossing is often associated with high levels of genetic variability,
and Hamrick and Godt (1990) found the average proportion of
polymorphic loci in animal pollinated outcrossers was 35.9% (1.8)
(see also Hamrick et al., 1991).

Loci examined in our study included some of the most variable
isozyme loci assayed in plants (e.g., esterases, phosphatases and
peroxidases; Gottlieb, 1981). Additionally, although our popu-
lation sample sizes were not unusually large, 19-27 presumptive
loci were assayed in each population and increasing the number
of assayed loci is considered more important that enlarging sam-
ple size when estimating genetic variation (Gorman and Renzi,
1979; Nei, 1978). Interpreting the number of loci (isozymes) for
an enzyme system with monomorphic banding can be difficult
and for some systems we may have overestimated by a few loci
the number of actual observed loci. This, however, would not
affect our overall finding of low genetic variation. Hence, although
our results are striking, we do not believe that they are artifacts
of locus choice or number.

Our findings are in agreement with those of Case (1994) who
found a complete lack of variation at 11 loci in four Cypripedium
arietinum populations and low variation in four C. acaule pop-
ulations in Michigan. Reduced levels of polymorphic loci and/or
heterozygosity were also identified in Michigan populations of C.
candidum and C. reginae, but C. calceolus, by comparison, had
unusually high levels of variation (Case, 1993, 1994). Klier et al.
(1991) reported similar findings for Iowa populations of C. can-
didum and C. calceolus.

Based on the demographics of Cypripedium acaule populations,
Gill (1989) speculated that repeated genetic bottlenecks and
founder effect, resulting from reproductive patterns that lower
N,’s, may lower intrapopulation variation in C. acaule. According
to Gill (1989), a C. acaule population may begin as a sample of a
source population, but, because of infrequent cross-pollination, only
a few of the founders contribute to future cohorts. Founder effect
may leave these cohorts with little variation and subsequent in-
breeding among closely related progeny may further reduce popu-
lation heterozygosity.

In an established Cypripedium population, population-level
changes in allele frequencies due to genetic drift are a function of

364 Rhodora [Vol. 96

the population’s overall N.. The N. of a Cypripedium population
in a given year 1s determined in part by the number of plants that
flower and are pollinated. The percentage of plants which have
flowered in the Massachusetts C. arietinum population has ranged
from 3.6% to 25.3% during the period 1985-1992, with a har-
monic mean of 11.7 plants. The geometric mean of the number
of above-ground plants, by contrast, was 141.6 (P. Martin Brown,
pers. comm., 1993). In the sampled Northampton C. acaule pop-
ulation, only five plants flowered among the 241 plants located
in 1992. In the C. acaule population studied by Gill (1989), the
geometric mean of above-ground plants over a ten-year period
was 680.6, but the harmonic means of flowering and fruiting
plants were only 33.0 and 3.2, respectively. As demonstrated by
these examples, established C. acau/e and C. arietinum popula-
tions may experience genetic bottlenecks between cohorts to an
extent not indicated by measures of population size N.
Population-level changes in genetic variation due to drift,
though, are more dependent on a population’s overall N. across
cohorts. Although annual N.’s may be low in Cypripedium acaule
and C. arietinum populations, overall N.’s may not be as low.
Kimura and Crow (1963) estimate the overall N. of a population
as the harmonic mean of the N,’s for each generation (hence, our
use of the harmonic mean above). However, for long-lived pe-
rennials that do not flower every year, like Cypripedium, this
method may underestimate, if only slightly, overall N.. Gill (1989)
showed that C. acau/le plants may enter a subterranean dormancy
for one to five years and there is little pattern to which individuals
appear above ground and flower or fruit in a given year. Thus,
although in a single year few individuals reproduce, reproduction
over several years in a Cypripedium population may be spread
across a larger number of different individuals. If, for example,
five individuals reproduce in one year and five different individ-
uals reproduce the following year, the N. for each year is five and
the harmonic mean estimate of N. across the two years is also
five. (We assume for this argument that progeny are produced in
equal proportions among all reproducers.) More than five indi-
viduals, though, have contributed in the two years to future co-
horts. The overall N. for the two years is thus greater than five,
but less than ten since the first five individuals had no chance to
mate with the second five. Our purpose in this argument is to
suggest that population-level losses in genetic variation due to

1994] Bornbusch et al.— Cypripedium 365

drift in C. acaule and C. arietinum can be overestimated if annual
N.’s are used as indicators of overall N.’s. The fact remains,
though, that the numbers of reproducing (fruiting) individuals in
a C. acaule or C. arietinum population are annual extremely low
and, even if flowering were random among individuals, overall
N, would still be unusually low.

Although founder effect and genetic drift may thus contribute
to low levels of intrapopulation variation in Cypripedium acaule
and C. arietinum, these forces should also act to increase variance
in allele frequencies between isolated populations. Our results and
those of Case (1994), though, indicate low species-level diversity
in C. acaule and C. arietinum. Furthermore, Case (1993, 1994)
and Klier et al. (1991) showed that C. calceolus, although pre-
sumably similar to other Cypripedium species in life history char-
acteristics, has unusually high levels of variation. Case (1994)
thus proposed that ancestral genetic bottlenecks caused by glacial
disturbance of habitats have contributed to lowered variation in
the relatively ecologically and geographically restricted C. acaule,
C. arietinum, C. candidum, and C. reginae. Case (1994), though,
cautioned that her results were based on populations in the Great
Lakes portion of Cypripedium ranges. Our study sampled New
England populations and indicates that, at least for C. acaule and
C. arietinum, the patterns of genetic variation documented by
Case (1994) extend beyond the Great Lakes region.

In some respects, the Massachusetts Cypripedium arietinum
population is a marginal population: it is at the margin of its
species’ range, and flower production, although variable in a ten-
year period, has been low overall (Luer, 1975; see Soulé, 1973
and Grant and Antonovics, 1978 for criteria of marginality). It
has been suggested that marginal populations may differ in genetic
composition (e.g., be less variable) from more central ones due
to greater drift, lowered immigration, or stronger and/or different
selection pressures (Soulé, 1973). However, isozyme studies of
plants have not revealed a consistent pattern in genetic variation
between geographically marginal and central populations (e.g.,
Levin, 1977; Tigerstedt, 1973). Studies of phenotypic and genetic
variation at ecologically marginal versus more central sites have
also shown no consistent pattern (Agnew, 1968; Linhart, 1974).
Furthermore, the low variation in the Northampton and Mount
Toby C. acaule populations, which are neither geographically nor
ecologically marginal for the species, indicates that it would be

366 Rhodora [Vol. 96

premature to attribute the low variation in the Massachusetts C.
arietinum population to its marginality.

An implication of our findings and those of Case (1994) is that
low genetic variation in declining Cypripedium populations may not
be the result of anthropogenic factors. These populations may not
be experiencing “genetic deterioration” (sensu Simberloff, 1986) from
previously high levels of variation. Rather, these populations may
have been genetically depauperate throughout much of their exis-
tence prior to human disturbance due to a combination of historical
events (Pleistocene bottlenecks) and reproductive patterns (low flow-
er and/or fruit production) that lower effective population sizes. We
suggest that habitat loss, plant collecting and demographic stochas-
ticities are, at least in the short term, more significant threats to
Cypripedium populations than is a relative lack of genetic variation.
Strategies that combine habitat protection and management and
enhancement of population size may be the most appropriate means
of conserving these populations. Conservation strategies could in-
clude: (1) maintenance of suitable habitat by partial canopy opening
to offset natural successional change (Falb and Leopold, 1993:
Mitchell and Sheviak, 1981; The Nature Conservancy, 1990); (2)
managing Cypripedium sites for a diversity of flowering plants that
offer rewards to attract and retain pollinators at the sites (Catling
and Knerer, 1980; Davis, 1986; Falb and Leopold, 1993); (3) and
although labor intensive, hand pollination to increase fruit produc-
tion (Davis, 1986; Gill, 1989; Barrett and Helenurm, 1987).

ACKNOWLEDGMENTS

We thank the Massachusetts Division of Fisheries and Wildlife
and the Natural Heritage and Endangered Species Program for
permission to collect Cypripedium arietinum material, The Na-
ture Conservancy (Massachusetts Field Office) for assistance in
obtaining permission to work at the C. arietinum site, and J.
Richburg for helping to locate the C. arietinum population. E.
Lyons and R. McMaster assisted in locating C. arietinum plants,
and we thank P. Martin-Brown for sharing his census data for
the C. arietinum population and J. Burk for information on the
history of C. arietinum on Mt. Toby. We thank R. Munson and
the Trustees of Christ United Methodist Church, Northampton
for permission to collect C. acaule material. R. Fish helped pre-

1994] Bornbusch et al.— Cypripedium 367

pare the figures, and J. Burk, M. Case, E. Lyons, R. McMaster,
S. Tilley, and two reviewers provided valuable comments on
drafts of this paper. We also thank M. Case for a preprint of Case
(1994). The Blakeslee Endowment for Genetics Research provid-
ed financial support for this study.

LITERATURE CITED

AGNnew, A. D. Q. 1968. Variation and selection in an isolated series of popu-
lations of Lysimachia volkensii Engl. Evol. 22: 228-236.

AYALA, F. J. 1982. Population nd Evolutionary Genetics: A Primer. Benjamin/
Cummings, Menlo Park, CA.

BarRETT, S. C. H. AND K. HELENURM. 1987. The reproductive biology of boreal

forest herbs. I. Breeding systems and pollination. Can. J. Bot. 65: 2036-2046.

AND J. R. 91. Genetic and evolutionary consequences of small
population size in pane implications for conservation. Jn; D. A. Falk and
K. E. Holsinger, Eds., Genetics and Conservation of Rare Plants. Oxford
University Press, New ork,

BEAMAN, J. H., E. A. Bourpo, F. W. Case, S. R. Crisprn, D. HENSon, R. W.
PIPPEN, A. A. REZNICEK, E. G. Voss AND P. W. THompson. 1985. Endan-
gered and threatened ene plants in Michigan. II. Third biennial review
proposal list. The Michigan Botanist 24: 99-116.

BRACKLEY, F. E. 1985. The orchids of New Hampshire. Rhodora 87: 1-117.

Case, F. W., Jn. 1964. Orchids of the Western Great ag Region. Cranbrook
Institute of Science Bulletin 48. Bloomfield Hills, M

Case, M. A. 1993. High levels of allozyme variation sie Cc 'ypripedium calceolus
(Orchidaceae) and low levels of divergence among its varieties. Syst. Bot. 18:
663-677.

1994, Extensive variation in the levels of genetic diversity and degree of
relatedness among five species of Cypripedium (Orchidaceae), Amer. J. Bot.
81: 175-184.

CATLING, P. M. 1990. Auto-pollination in the Orchidaceae. Jn: J. Arditti, Ed.,
Orchid Biology Reviews and Perspectives, V. Timber Press, Portland, OR

AND G. KNERER. 1980. Pollination of the small white lady’s- Sslinper
(C ee yee Copan: in Lambton County, southern Ontario. Can. Field-
Naturalist 94: 435-438.

CoFFIN, B. AND L. PFANNMULLER, Eds. 1988. sees Endangered Flora and
Fauna. University of Minnesota Press, Minnea , MN.

Davis, R. W. 1986. The pollination biology of C woridediian acaule (Orchidaceae).
Rhodora 88: 445-450.

Drpse, A. C., C. S. CAMPBELL, H. R. TyLer, JR. AND B. St. J. VICKERY. 1989.
Maine’s official list of ene and threatened plants. Rhodora 91: 244—

9.
FAB, D. L. AND D. J. LEopotp. 1993. Population dynamics of Cypripedium
candidum Muhl. ex ae small white ladyslipper, in a western New York
fen. Natural Areas J. 13: 76-86.

368 Rhodora [Vol. 96

Gut, D. E. 1989. Fruiting failure, pollinator inefficiency, and speciation in orchids.
In: D. Otte and J. A. Endler, Eds., Speciation and its Consequences. Sinauer,
Sunderland, MA

GorMAN, G. C. AND J. RENzI, JR. 1979. Genetic distance and heterozygosity
estimates in electrophoretic studies: effects of sample size. Copeia 1979: 242-
249,

Gott.ies, L. D. 1981. 1981. Electrophoretic evidence and plant populations.
Prog. Phytochem. 7: 1-46.

GRANT, M. C. AND J. ANTONOovicS. 1978. Biology of ecologically marginal pop-
ulations of Anthoxanthum odoratum. Part 1. Phenetics and dynamics. Evol.
32: 822-838

Hamrick, J. L. AND M. J. W. sgh 1990. Allozyme een in nae species.

“AH , A. L. Kahler and B. S. Weir, Eds., Plant

Boonton Genetics oe ee Genetic Geos Saaice Sunderland.

, D. A. MURAWSKI AND M. D. Love.ess. 1991. Correlations between
species traits and allozyme diversity: implications for conservation biology. /n:
D. A. Falk and K. E. Holsinger, Eds., Genetics and Conservation of Rare Plants.
Oxford University Press, New York.

Hart, D. L. AND . CLARK. 1989. Principles of Population Genetics. Sinauer,
Sunderland, MA.

Karron, J. D. 1991. Patterns of genetic variation and breeding systems in rare
plants species. /n: D, A. Falk and K. E. Holsinger, Eds., oe and Con-
servation Rare Plants. Oxford University Press, New

Kimura, M. AND J. F. Crow. 1963. The measurement of ie population
number. te 17: 279-288.

KureR, K., M. J. LEOSCHKE AND J. F. WENDEL. 1991. Hybridization and in-
eae in white and yellow ladyslipper orchids (Cypripedium candidum

d C. pubescens). J. of Hered. 82: 305-318.

Lev, 'D. A. 1977. The organization of genetic variation in Phlox drummondii.
Evolution 31: 477-494,

LinuartT, Y. B. 1974. Intra-population differentiation in annual plants. I. Ve-

ronica peregrina L. raised under non-competitive conditions. Evolution 28:

32-243.

LIL

Luger, C. A. 1975. The Native Orchids of the United States and Canada Ex-
cluding Florida. The New York Botanical Garden, New York.

MITCHELL, R. S. AND C. J. SHEVIAK. 1981. Rare plants of New York State.
Bulletin 445, New York State Museum, Albany, NY

Murpny, R. W., J. W. Sires, Jk., D. G. BUTH AND C. H. HAuUFLER. 1990. Proteins
II: Isozyme electrophoresis. Jn; D. M. Hillis and C. Moritz, Eds., Molecular
Systematics. Sinauer, Sunderland, Mass.

Ne1, M. 1978. Estimation of average heterozygosity and genetic distance from a
sm niall number of individuals. Genetics 89: 583-590.

PLowRIGHT, R., J. D. THOMSON AND G. R. THALER. 1980. Pollen removal in
C. Lea acaule (Orchidaceae) in relation to aerial fenitrothion spraying

w Brunswick. Can. Entomologist 112: 765-769.

Sse oe D. 1986. The proximate causes of extinction. Jn; D. M. Raup and

D. oe Eds., Patterns and Processes in the History of Life. Springer-

SoctTis, D. M., C. ae HAUvuFLER, D. C. DARROW AND G. J. GAsTONY. 1983. Starch

1994] Bornbusch et al.— Cypripedium 369

gel electrophoresis of ferns: a compilation of grinding buffers, gel and elec-
trode buffers, and staining schedules. Amer. Fern J. 73: 9-27

Sorrig, B. A. 1989. Rare native plants of Massachusetts. Publication 1 6000-1 1-
3000-63-89-C.R. Massachusetts Natural Heritage Program, Cambridge, Mass.

SouLe, M. 1973. The eae cycle: A theory of marginal populations. Ann.

v. Ecol. Syst. 4: —187.

STONE, 6. E. 1913. A oe of Plants Growing without Cultivation in Franklin,

Hampshire and Hampden Counties, Massachusetts. Carpenter & Morehouse,

STOUTAMIRE, W. P. 1967. Flower biology of the lady’s-slippers (Orchidaceae:
C aaa The Michigan pees 6: 159-175.

THE NATURE CONSERVANCY. 1990. Element st dship abstract for Cypripedium
arietinum. The Nature Conservancy, Breen VA.
TIGERSTEDT, P.M. A. 1973. Studies on isozyme variation in marginal and central

populations of Picea abies. Herditas 75: 47-60

TUCKERMAN, E. AND C. C. Frost. 1875. A Catalogue of Plants Growing without
Cultivation within the Vicinity of Amherst College. E. Nelson, Amherst, Mass

VAN DER PyL, L. AND C. H. Dopson. 1966. Orchid Flowers: Their Pollination
and Evolution. University of Miami Press, Coral Gables, F

WEEDEN, N. F. AND J. F. WENDEL. 1989. Genetics of plant isozymes. /n: D.
Soltis and P. S. Soltis, Eds., Isozymes in Plant Biology. Dioscorides Press,
Portland, OR.

WENDEL, J. F. AND N. F. WEEDEN. 1989. Visualization and interpretation of plant
isozymes. In: D. E. Soltis and P. S. Soltis, Eds. Isozymes in Plant Biology.
Dioscorides Press, Portland, OR

WIaARD, L. A. 1987. An Tatroduction to the Orchids of Mexico. Comstock Pub-
lishing, Ithaca, NY.

Youna, 8S. M., Ed. 1992. New Be Rare Plants Status List. New York Natural
Heritage Program, Latham, N

DEPARTMENT OF BIOLOGICAL SCIENCES
SMITH COLL
Cc MASSACHUSETTS 01063

RHODORA, Vol. 96, No. 888, pp. 370-381, 1994

EXINE REDUCTION IN UNDERWATER FLOWERING
CALE eee erg cena
IMPLICATION VOLUTION
OF peu onor ae

C. THOMAS PHILBRICK AND JEFFREY M. OSBORN

ABSTRACT

Both aerial and underwater pollination systems (hypohydrophily) occur in Cal-
litriche (Callitrichaceae), and as such the genus can serve as a model system in
which to probe questions concerning the evolution of hypohydrophily from aerial
pollination systems. Evidence from scanning and transmission electron micros-
copy reveals strikingly different exine thickness in the pollen of four aerially
flowering species (C. heterophylla var. bolanderi, C. marginata, C. peploides, and
C. stagnalis), which have a distinct sexine layer (0.55—1.1 «m thick), in comparison
with that of the hypohydrophilous C. hermaphroditica. The exine of the latter
species is rudimentary (<0.1 wm thick) in pollen of the annual form of the species,
and virtually absent in the perennial form. The degree of exine reduction in pollen

of C. hermaphroditica relative to that of the aerially flowering species contrasts
with the otherwise general morphological similarity observed among the flowers
of the respective species.

Key Words: Callitriche, underwater pollination, pollen ultrastructure

INTRODUCTION

Although considerable insight has been gained regarding pollen
form and function in angiosperms (e.g., Blackmore and Ferguson,
1986 and refs. therein) a general consensus remains elusive con-
cerning possible relationships between pollen morphology and
pollination systems. Several studies have shown positive corre-
lations between pollen structure and the nature of the pollen vec-
tor (e.g., anemophilous compositae, Bolick, 1990; papilionoid
legumes, Ferguson and Skvarla, 1982; Araceae, Grayum, 1986;
Cambombaceae, Osborn, Taylor, and Schneider, 1991), whereas
others report a lack of such associations (e.g., Polemoniaceae,
Taylor and Levin, 1975; various families, Lee, 1978; Zingiberales,
Kress, 1986). Factors that are indirectly related to pollination
may also play a role in influencing the evolution of pollen struc-
ture, especially pollen—stigma interactions (see Kress, 1986).

In contrast, a striking correlation exists between pollen structure
and pollination system in species that exhibit underwater cross-
pollination systems (hypohydrophily). Reduction of exine thick-
ness, or the lack of exine altogether, is strongly correlated with

370

1994] Philbrick and Osborn— Callitriche Pollen a7)

hypohydrophily (Cox, 1988; Philbrick, 1988; Philbrick and An-
derson, 1992). For instance, pollen of hypohydrophilous species
from often distantly related groups have rudimentary exines, 1n-
cluding Ceratophyllum demersum L. (Ceratophyllaceae: Les, 1988),
Enhalis acoroides (L.) Royle and Thalassia hemprichii (Ehrenb.)
Aschers. (Hydrocharitaceae: Pettitt, 1980, 1981), while an exine
component of the pollen wall is absent in Thalassodendron cil-
iatum (Forsk.) den Hartog (Hydrocharitaceae; Pettitt, 1980), and
species of Najas (Najadaceae: Blackmore, McConchie, and Knox,
1987) and Amphibolis (Cymodoceaceae: Pettitt, Ducker, and Knox,
1978). Indeed, exine reduction is one of the few morphological
features that are unique to hypohydrophilous pollination systems.

Few comparative studies have been conducted at the infrage-
neric level between aerially flowering and hypohydrophilous spe-
cies. Clearly the major hindrance to such comparisons is the fact
that genera that have both pollination types are exceedingly rare.
As a consequence, our understanding of the nature of the changes
that take place in pollen structure, indeed in all floral structure,
during the evolution of hypohydrophily are based on comparisons
at the generic level and above. Yet, comparisons at these higher
taxonomic/phylogenetic levels are of limited value for understand-
ing features that are as evolutionarily dynamic as pollination sys-
tems. Callitriche is an ideal genus for such comparisons because
both aerial and hypohydrophilous pollination systems occur.

Callitrichaceae is a widespread monogeneric (Callitriche) fam-
ily of ca. 50 species. Species of Callitriche exhibit one of three
growth habits: terrestrial, amphibious, or aquatic (Philbrick and
Anderson, 1992 and refs. therein). Terrestrial species occur in
seasonally wet areas while plants of amphibious species can grow
submersed, with a floating rosette of leaves, or as the land-form
on moist ground. Aquatic species grow obligately submersed.

Callitriche is the only genus in which both aerial floral biologies
and hypohydrophily have been documented (Philbrick and An-
derson, 1992; Philbrick, 1993). The majority of species of Cal-
litriche flower aerially; anemophily seems to be the principal pol-
lination system. Callitriche truncata Gussone and
hermaphroditica L. show obligately submersed flowering. Phil-
brick (1993) has employed paternity exclusion analysis using ran-
dom amplified polymorphic DNA markers to document hypohy-
drophily in C. hermaphroditica. The pollination system of C.
truncata is poorly known.

a12 Rhodora [Vol. 96

The high degree of floral structural divergence between aerial
flowering and hypohydrophilous angiosperms is well known (e.g.,
Arber, 1920; Dahlgren and Rasmussen, 1983; Philbrick, 1991;
Sculthorpe, 1967; Tomlinson, 1982). A significant component of
this divergence entails loss or reduction of parts (e.g., perianth,
number of stamens) and reduction in flower size. Consequently,
one would predict that a similar degree of divergence would occur
between aerially flowering and hypohydrophilous species of Ca/-
litriche. However, this is not the case. The overall floral mor-
phology of species of Callitriche is uniform (cf., Philbrick and
Anderson, 1992; Schotsman, 1982, 1985). The only qualitative
difference in reproductive structures of aerial and submerged flow-
ering species is the presence (aerial flowering species) or absence
(submerged flowering species) of endothecial thickenings in the
anthers (Schotsman, 1982; Philbrick, unpubl.).

Although several studies of pollen morphology of Callitriche
have been published based on data obtained using light and scan-
ning electron microscopy (SEM) (e.g., Diez, Talavera, and Garcia-
Murillo, 1988; Martinsson, 1993; Moar, 1960: Moore and Webb,
1978) the diversity of pollen ultrastructural features that occur in
the genus are not well understood. Only two investigations have
provided structural information on Callitriche pollen based on
transmission electron microscopy (Martinnson, 1993; Osborn and
Philbrick, 1994).

Herein we address the question, how does exine thickness in
the hypohydrophilous Callitriche hermaphroditica compare with
that of aerial flowering species? This question is addressed by
comparative study of exine thickness of four aerially flowering
species of Callitriche and the hypohydrophilous C. hermaphro-
ditica.

MATERIALS AND METHODS

Species Studied

Five species were studied: C. hermaphroditica L., C. hetero-
phylla var. bolanderi (Hegelm.) Fassett, C. marginata Torrey, C.
peploides Nutt., C. stagnalis Scop. Voucher specimens are listed
in Appendix I. Four of the five species studied (except C. her-
maphroditica) are aerially pollinated, and are presumed to be
anemophilous although little experimental evidence confirms this.

1994] Philbrick and Osborn— Callitriche Pollen 373

Callitriche hermaphroditica is hypohydrophilous (Philbrick, 1993).
One species (C. heterophylla var. bolanderi) was also selected that
exhibits internal geitonogamy, an unusual form of self-fertiliza-
tion (Philbrick, 1984; Philbrick and Anderson, 1992).

The species selected represent all three growth habits that occur
in the genus. Callitriche heterophylla var. bolanderi, C. marginata,
and C. stagnalis are amphibious. Callitriche peploides is terrestrial
and C. hermaphroditica is aquatic. Moreover, these species rep-
resent the three main phylogenetic lineages in the genus (Philbrick
and Jansen, 1991).

Electron Microscopy

Entire plants were fixed in formalin: glacial acetic acid : 95%
ethanol (FAA; 0.5:0.5:9). For SEM, pollen of all species except
C. hermaphroditica was acetolyzed and prepared as outlined in
Bogle and Philbrick (1980). Grains of C. hermaphroditica, which
collapse even after mild acetolysis, were critical point dried, de-
hydrated in a graded ethanol series, and placed onto aluminum
stubs. Grains were examined on scanning electron microscopes
at the University of Connecticut and the Rancho Santa Ana Bo-
tanic Garden.

For transmission electron microscopy (TEM), mature anthers
were excised from preserved plants, post-fixed in 1% osmium
tetroxide, buffered in sodium cacodylate to a pH of 6.5, for 2h
and washed with buffer. Entire anthers were placed onto cellulose
filters and then coated on both sides with agar. Agar embedded
filters were subsequently dehydrated in a graded ethanol series,
transferred to several changes of 100% acetone (to solubilize the
filters), gradually infiltrated with Spurr low viscosity epoxy resin,
and embedded. Ultrathin sections were cut with a diamond knife,
collected on copper | x 2 mm slot grids, and dried onto formvar
support films following the techniques of Rowley and Moran
(1975). Grids were stained with 1% potassium permanganate (1-
5 min.), 1% uranyl acetate (8-15 min.), and lead citrate (5-10
min.; Venable and Coggeshall, 1965), and examined using a Zeiss
EM-10 transmission electron microscopy at 60-80 kV.

RESULTS

Both intine and exine sporoderm components of the grains were
present, at least in part, in the pollen of all five species examined.

374 Rhodora [Vol. 96

Not surprisingly, given that grains were fixed in F.A.A., the degree
to which the intine was preserved in the various species differed.
This limited interpretation regarding intine structure, but did not
adversely affect interpretation of exine thickness, the main focus
of the study. In general, the intine of C. hermaphroditica was well
preserved while in the other four species it was not. The sporo-
derm fine structure of nonapertural wall regions, with particular
emphasis on the layers, is presented below for each species.

C. heterophylla var. bolanderi

The grains of Callitriche heterophylla var. bolanderi (Figures
1-3) had a distinct exine component that ranged in thickness from
0.75-1.00 um. The exine was intectate, and two-layered (sexine,
nexine; cf. Erdtman, 1952), with the layers separated by a dark-
staining commissural line (Figure 3). The sexine was composed
of variably sized gemmae that formed an anastomosing to some-
what reticulate pattern (Figure 1). The gemmae were laterally
fused at their bases, and in association with smaller granular
elements formed a thin, somewhat irregular nexine (Figure 3).
The nexine comprised about 20% of the total thickness of the
overall exine. Below the commissure, relatively small granular
elements of the nexine protruded into the underlying intine (Fig-
fre 3k

C. marginata

Pollen of C. marginata had a well-defined, two-layered, intec-
tate exine that ranged from 0.55—-1.17 wm in thickness (Figures
4—6). The sexine consisted of distinct elements ranging from gem-
mae to echinae that were fused laterally into a semi-reticulate
pattern (Figure 4). A thin, somewhat continuous granular nexine
was also present (Figure 6), and comprised approximately 10%
of the total thickness of the exine. Larger granule units of the
nexine were present below the more continuous layer, and pen-
etrated into the underlying intine (Figure 6). A commissural line
was indistinct.

C. peploides

Pollen of C. peploides had a distinct, intectate exine that ranged
from 0.70-0.95 um thick (Figures 7-9). The exine of this species

1994] Philbrick and Osborn—Callitriche Pollen 375

Figures 1-9. Scanning (Figures 1, 4, 7) and transmission (Figures 2, 3, 5, 6,
8, 9) electron tee of Callitriche oe ree 1-3; C. heterophylla var.

holanderi 1. View showing 1 on, ap is evident as an elongate
depression in “ upper right region oe a grain. Bar = 5 um. 2. Transverse section
of entire grain. Bar = 5 um. 3. Detail of nonapertural exine in transverse section;
note the nexine (n) and poets: line (c). Bar = | wm. Figures 4-6; C. m

ginata. 4. View showing the interapertural region. Bar = 5 um. 5. Wanavers
section of entire grain. Bar = 5 um. 6. Detail of nonapertural exine in transverse
section; note feud nexine (n) relative to the thickness of the sculptured sexine. Bar

= 1 um. Fi s 7-9: C. peploides. 7. Polar view and the location of the three
apertures oe. note the organization of the sculptural elements into polygonal
shaped clusters. Bar = 5 . 8. Transverse section of entire grain showing the

location of apertures, Pen by the concave areas; note that apertural regions

also have sculptured sexine elements, but a thinner nexine in comparison with

nonapertural areas. Bar = 7 wm. 9. Detail of nonapertural exine in transverse

section; note the relative thickness of the nexine (n) versus the sculptured sexine,
and the commissural line (c). Bar =

376 Rhodora [Vol. 96

was also two-layered, with the sexine being composed of irregu-
larly shaped verrucae and echinae. These outer sculptural ele-
ments were laterally fused at their bases and arranged in groups
of 4-8 into polygonal shaped clusters (Figure 7). The sculptural
elements were borne upon a relatively thick, homogeneous to
granular nexine (Figure 9). The nexine of C. peploides was the
thickest of the five species examined and accounted for about
30% of the total exine thickness. A dark commissure was present
near the base of the nexine, and was in turn underlaid by a thin
layer, that was somewhat irregular in thickness, consisting of min-
ute granules (Figure 9).

C. stagnalis

Callitriche stagnalis had pollen with a well-defined intectate
exine that ranged from 0.50-0.95 um in thickness (Figures 10-
12). Distinct clavate and echinate elements characterized the sex-
ine along with a thin, granular nexine (Figure 12). The nexine
comprised approximately 17% of the total exine thickness. The
structural elements were variably fused and arranged into a some-
what reticulate pattern (Figure 10). A dark-staining commissural
line was present within the nexine, and underlaid by a layer of
irregular relatively small granules (Figure 12).

C. hermaphroditica

Callitriche hermaphroditica had the most distinctive pollen wall
structure of the species examined (Figures 13-16). Grains had a
well-developed intine ranging from 0.25-0.45 um in thickness
(Figures 15 and 16). In contrast to the other species examined,
the exine of C. hermaphroditica is best described as rudimentary.
However, ultrastructural variability was evident, even among
grains within the same anther. In plants of the perennial growth
form (#2267), an exine wall component was virtually absent (Fig-
ure 16), while in the annual form (#2030) the exine formed a thin,
homogeneous to granular band that was generally less than 0.1
um in thickness (Figure 15). In other grains from the annual
growth form, the exine element was manifested as small, irreg-
ularly shaped granules that ranged in size from 0.05—0.16 um (not
shown). The poorly developed exine in C. hermaphroditica is
likely the cause for the irregular shape and smooth external ap-
pearance of the grains (Figure 13).

1994] Philbrick and Osborn— Callitriche Pollen 377

Figures 10-16. Scanning (Figures 10, 13) and transmission (Figures 11, 12,
14-16) electron micrographs of Callitriche pollen. Figures 10-12; C. stagnalis. 10.

View showing the ee region. Bar = 5 um. 11. Transverse section of an
entire grain. ie um. 12. Detail of transverse sea of exine; note the thin
nexine (n) and the thicker sculptured sexine. Bar = 1 wm. Figures 13-16;

C.
ene rem (perennial form [#2267]: Figures 14 a 16; annual form [#2030]:
Figures 13 and 15). 13. View showing several nonacetolyzed grains; note the
irregular shape. Bar = 10 um. 14. Transverse section of an entire grain; note the
absence of a well-defined and darkly stained outer sporoderm layer (exine). Bar
= 5 um. 15. Detail of transverse section of sporoderm; note the relatively thin,
rudimentary exine (e) and the thicker underlying intine (i). Bar = 0.6 um. 1
Detail of transverse section of sporoderm showing the virtual absence of an exine
component and relatively thick intine (i, between arrows). Bar = 0.6 um.

DISCUSSION

This study confirms exineless pollen in Callitriche. The general
pattern of a well developed exine in aerially flowering species and
the virtual lack of an exine in the hypohydrophilous Callitriche
hermaphroditica is demonstrated. However, exine thickness dif-
fered in annual (rudimentary) and perennial (absent) forms of this
species. Martinsson (1993) reported reduction, but not absence,
in numerous accessions of this species from Sweden. The signif-
icance of this infraspecific variation in exine thickness is not
known, but may reflect the recent origin of hypohydrophily in C.
hermaphroditica (see below, and Osborn and Philbrick, 1994).

378 Rhodora [Vol. 96

There is also an apparent association between the growth habit
(terrestrial versus amphibious) and the relative thickness of the
basal layer. Pollen of the terrestrial species (Callitriche peploides)
had a thicker nexine (30% of total exine) than any of the three
amphibious species (10-20% of total exine). However, only a
single terrestrial species was examined. Additional terrestrial spe-
cies need to be examined before the significance of this apparent
difference can be interpreted.

Although the lack (or extreme reduction) of an exine is closely
correlated with hypohydrophily, the adaptive significance of this
association is not understood. It is perhaps intuitive to predict
that the loss of exine is a result of its release from selection pres-
sures that maintain it. Yet, it is not known whether the exine is
actually selected against during the evolution of hypohydrophily
or is simply lost due to genetic drift. In general, the harmome-
gathic nature of the pollen exine is associated with controlling
water relations of the grain during dispersal (desiccation) and
hydration on the stigma (Heslop-Harrison, 1971). It is not clear
whether this alone would translate into the loss of exine when
such extremes in water potential are absent during hypohydro-
philous pollination. On the other hand, there may be direct se-
lective advantages associated with the lack of exine. Perhaps the
pollen delivery system itself exerts selective pressures (e.g., Cox,
1988). Relationships among exine reduction and factors such as
pollen-stigma interaction, emergence of the pollen tube, and pol-
len dispersal remain to be clarified in hypohydrophilous plants.
It is also important to note that there are several examples of
aerially flowering groups that have extremely reduced exines. The
adaptive significance of exine reduction in these groups is also
equivocal, but does not seem to be closely related to pollination
systems (Kress, 1986).

Hypohydrophily is clearly a derived pollination system in an-
giosperms; the aerial floral biology is abandoned in favor of the
release and capture of wet, water-borne pollen. In a functional
sense hypohydrophily represents one of the most unique (diver-
gent) forms of pollination in angiosperms. Because of the close
relationship between floral form and pollination system function
(e.g., Faegri and van der Pijl, 1979) it is intuitive to predict that
the floral structure of hypohydrophilous species would reflect the
functional divergence of the pollination system (i.e., be markedly
different than related aerially flowering species). However, this

1994] Philbrick and Osborn— Callitriche Pollen 379

does not seem to be the case in Callitriche where relatively little
modification in overall floral morphology is apparent. The degree
of exine reduction in C. hermaphroditica contrasts markedly with
the morphological similarity that otherwise cl terizes the flow-
ers, which may indicate that the strongest selective pressures that
have operated during the evolution of hypohydrophily in Calli-
triche occur on pollen, not other floral characters.

ACKNOWLEDGMENTS

We thank Robert R. Haynes for assistance in collecting C.
peploides, Thomas N. Taylor for providing technical facilities of
his laboratory, and Peter Fritsch and John Kress for comments
on the manuscript. This work was supported by NSF grants BSR-
8701285 and DEB 9496053 to CTP and BSR-8207125 to Gregory
J. Anderson, and grants to CTP from the New England Botanical
Club and Sigma Xi. CTP also thanks the Rancho Santa Ana
Botanic Garden for providing general support.

LITERATURE CITED

ArRBER, A. 1920. Water aaa romeo Press, London.
BLACKMORE, S. AND I. K. Fercuson, Eds. 1986. Pollen and Spores: Form and
Function. Academic or Sacre
. A. McConcute AND R. B. Knox. 1987. Phylogenetic analysis of the
male ontogenetic program in aquatic and terrestrial monocotyledons. Cla-
distics 3: 333-347
Boac.e, A. L. AND C. T. PHitBRick. 1980. A generic atlas of Hamamelidaceous
pollens. Contr. Gray Herb. 210: 29-103
Boutck, M. R. 1990. The pollen surface in wind-pollination with emphasis on
the Compositae, pp. 39-51. Jn: M. Hesse and F. Ehrendorfer, Eds., Mor-
phology, ea aie and Systematic Relevance of Pollen and Spores,
Springer-Verlag, W
Cox, P. A. 1988. ea canlons pollination. Ann. Rev. Ecol. & Syst. 19: 261-

Diez, M. J., S. TALAVERA AND P. GARcIA-MuRILLO. 1988. Contributions to the

ee of hydrophytic, non- se aces angiosperms. |. Studies with
SEM. Candollea 43: 147-158.

ae R. AND F. N. RASMUSSEN. cn Monocotyledon evolution: char-
acters and phylogenetic estimation, pp. 255-395. Jn: M. K. Hecht, B. Wallace
and G. T. Prance, Eds., Evolutionary Biology, Vol. 16. Plenum Publishing
Corporation, New York

ERDTMAN, G. 1952. Pollen morphology and plant taxonomy. Angiosperms.
Almquist and Wiksell, Stockholm.

380 Rhodora [Vol. 96

FAEGRI, K. AND L. VAN DER Pu. 1979. The principles of pollination ecology,
3rd rev. ed. Pergamon Press, New York
FerGuson, I. K. AND J. J. SKVARLA. 1982. Pollen morphology in relation to
pollinators in Papilionoideae (Leguminosae). Bot. J. Linn. Soc, 84: 183-193.
GrayuM, M. H. 1986. Correlations between pollination biology and pollen
eorshoiogs in the Araceae, with some implications for angiosperm evolu-
tion, pp. 313-327. /n: S. Blackmore and I. K. Ferguson, Eds., Pollen and
ores: Form and Function. Academic Press, London
HesLop-Harrison, J. 1971. The pollen wall: structure on development, pp.
—98. In: J. Sana Harrison, Ed., Pollen: Development and Physiology.
Butterworths, Lon
Kress, W.J. 1986. Se pollen structure and pollination systems of tropical
Heliconia (Heliconiaceae), pp. 329-345. Jn: S. Blackmore and I. K. Ferguson,
ds., Pollen and Spores: Form and Function. Academic Press, London.
Lee, S. 1978. A factor analyst study of the functional significance of angiosperm
pollen. Syst. Bot.
Les, D. H. 1988. The enna and affinities of the Ceratophyllaceae. Taxon 37:
26-345.
Martinsson, K. 1993. The pollen of Swedish Callitriche (Callitrichaceae)—
trends towards submergence. Grana 32: 198-209.
Moar, N. T. 1960. Studies in pollen morphology. 2. The pollen of the New
Zealand and of two Australian species of Callitriche L. New Zealand J. Sci.
3: 415-421.
Moore, P. D. AND J. A. Wess. 1978. An illustrated guide to pollen analysis.
John Wiley and Sons, New York.
Ossporn, J. M. AND C. T. PHmtBRick. 1994. Comparative pollen structure and
pollination biology in the Callitrichaceae. Acta Bot. anes 141 (in a ess).

, T. N. TAYLor AND E. L. SCHNEIDER. 1991. Pollen morphology and
ultrastructure of the Cabombaceae: correlations with pollination ee,
Amer. J. Bot. 78: 1367-1378.

Pettitt, J. M. 1980. Reproduction in seagrasses: nature of the pollen and re-
ceptive eg of the stigma in the Hydrocharitaceae. Ann. Bot. (London)
45: 257-
1981, en in seagrasses: pollen development in Thalassia
hemprichii, Halophila stipulacea, and Thalassodendron ciliatum. Ann. Bot.
mee ek 609-622.
KER AND R. B. KNox. 1978. Amphibolis pollen has no exine.
Helobine ete 2: 19-22.
PHILBRICK, C. T. 1984. Pollen tube growth within vegetative tissues of Callitriche
(Callitrichaceae). Amer. J. Bot. 71: 882-886.
1 Evolution of underwater outcrossing from aerial pollination sys-
tems: a hypothesis. Ann. Missouri Bot. Gard. 75: 836-841.
. 1991. Hydrophily: phylogenetic and evolutionary considerations. Rho-
dora 93: 36-50.
. 1993. Underwater cross-pollination in Callitriche hermaphroditica (Cal-
litrichaceae): evidence from randomly amplified polymorphic DNA markers.
A 4

AND G. J. ANDERSON. 1992. Pollination biology in the Callitrichaceae.
Syst. Bot. 17: 282-292.

1994] Philbrick and Osborn— Callitriche Pollen 381

— AND R. K. JANSEN. 1991. Phylogenetic studies of North American Cal-
litrichaceae using chloroplast DNA restriction fragment analysis. Syst. Bot.
: 508-521.

Row Ley, J. C. AND D. T. Moran. 1975. A simple procedure for mounting
wrinkle-free sections on formvar-coated slot grids. Ultramicroscopy |: 151-

SCHOTSMAN, H. D. 1982. Biologie Florale des Callitriche: etude sur quelques
especes d’Espagne Meridionale. Adansonia 3-4: 111-160
—. 1985. Biologie floral des Callitriche (Callitrichaceae): II. Etude sur quel-
ues especes d’Oceanie. Adansonia 4: 357-375.
SCULTHORPE, C.D. 1967. The biology of aquatic vascular plants. Edward Arnold,
London

Taytor, T. N. AND D. A. Levin. 1975. Pollen morphology of Polemoniaceae
in relation to systematics and pollination systems: scanning electron mi-
croscopy. Grana 15: 91-112.

ToMLinson, P. B. 1982. Helobiae (Alismatidae), pp. 1-522. In: C. R. Metcalfe,
Ed., Anatomy of the Monocotyledons. VII. Clarendon Press, Oxford.
VENABLE, J. H. AND R. CoGGESHALL. 1965. A simplified lead citrate stain for

use in electron microscopy. J. Cell Biol. 25: 407.

DEPARTMENT OF BIOLOGICAL AND ENVIRONMENTAL SCIENCES
WESTERN CONNECTICUT STATE UNIVERSITY
DANBURY, CONNECTICUT 06810

DIVISION OF SCIENCE
NORTHEAST MISSOURI STATE UNIVERSITY
KIRKSVILLE, MISSOURI 63501

APPENDIX I

Collection information of specimens examined during this study. All voucher
specimens are located in the herbarium of Western Connecticut State University.
C. heterophylla var. bolanderi: WASHINGTON, Thurston Co., 22 June 1987, Phil-
brick 2098. C. marginata: CALIFORNIA, San Diego Co., 6 January, 1986, Philbrick
1597. C. peploides: ALABAMA, Conecuh Co., 14 March, 1988, Philbrick and Haynes
2135. C. stagnalis: WASHINGTON, Gray’s Harbor Co., 20 June, 1987, Philbrick
2096. C. hermaphroditica: annual form—CALIFORNIA, Tuolumne Co., 31 May,
1987, Philbrick 2030; perennial form—CALIFORNIA, Shasta Co., 12 June, 1990,
Philbrick 2267.

RHODORA, Vol. 96, No. 888, p. 382, 1994

NEW ENGLAND BOTANICAL CLUB

GRADUATE STUDENT RESEARCH AWARD
ANNOUNCEMENT

The New England Botanical Club will offer an award of $1,000
in support of botanical research to be conducted by a graduate
student in 1995. This award is made annually to stimulate and
encourage botanical research on the New England flora, and to
make possible visits to the New England region by those who
would not otherwise be able to do so.

The award will be given to the graduate student submitting the
best research proposal dealing with systematic botany, biosyste-
matics, plant ecology, or plant conservation biology, although
research in other areas will be considered.

Applicants must submit a proposal of no more than three dou-
ble-spaced pages, a budget (the budget will not affect the amount
of the award), a curriculum vitae, and two letters in support of
the proposed research, one from the student’s thesis advisor. Three
copies of the proposal must be submitted.

Proposals and supporting letters must be received no later than
March 1, 1995. The recipient will be notified by April 30, 1995.

Send proposals to: Awards Committee, The New England Bo-
tanical Club, 22 Divinity Avenue, Cambridge, MA 02138.

The 1994 Graduate Student Research Award was presented to
Andrea Stevens of the University of Massachusetts, in support
of her work on the paleoecology of sandplain grasslands on Mar-
tha’s Vineyard.

382

RHODORA, Vol. 96, No. 888, pp. 383-386, 1994

CHROMOSOME COUNTS FOR CALLITRICHE
(CALLITRICHACEAE) IN NORTH AMERICA
C. THOMAS PHILBRICK

ABSTRACT

matic h

numbers are reported for nine species and one variety of

ne Caltichaceae) The following new reports are included: 2n = 20 (C.
= 40 for one of four populations of C. heterophylla

var. ae 2n= 40 C. trochlearis), 2n = 10 (C. peploides, C. terrestris).

Key Words: Callitriche, chromosome numbers, North America

INTRODUCTION

Chromosome numbers vary considerably between species of
Callitriche (Callitrichaceae), a family of ca. 50 species of largely
temperate distribution. Diploid numbers range from 2n = 6 to
40 (e.g., Schotsman, 1967; Philbrick, 1989). Yet, chromosome
numbers are known for only approximately half of the species.
The purpose of this contribution is to report on chromosome
numbers of species of Callitriche that occur in North America.
Herein, I report chromosome counts from 41 populations of the
nine species and single variety that occur in this region.

MATERIALS AND METHODS

Chromosome counts were made from seedling root tips. Seeds
were germinated on moist filter paper. Upon germination the
seedlings were treated in 0.02% 8-hydroxyquinoline at 4°C for 2
hours and subsequently fixed in Carnoy’s fixative (95% EtOH:
chloroform : glacial acetic acid; 6:3:1) overnight at —5°C. After
rinsing twice in distilled water, the root tips were hydrolyzed
briefly in 1 N HCl and squashed in aceto-orcein. Counts were
derived from a minimum of ten plants from each population.
Voucher specimens are cited in Table 1. Voucher specimens are
located at CONN.

RESULTS AND DISCUSSION

Chromosome counts were determined for 41 populations of
the nine species and one variety of Callitriche in North America

383

Table 1. Chromosome number determinations for Callitriche in North America. All collections are by Philbrick unless otherwise noted.

C. hermaphroditica L. 2n = 6. CANADA. Alberta. Edmonton, September 1985, 1609. British Columbia. Merritt, 28 August 1988, 2/57.
Quebec. Portneuf Co.: St. Augustin, 2 September 1988, Philbrick & Bruneau 2166. u.s.A. California. Madera Co.: 30 May 1987, 2022;
Tuolumne Co.: 31 May 1987, 2033.

C. heterophylla var. heterophylla Pursh. 2n = 20. u.s.a. Mississippi. Lowndes Co.: 15 March 1988, Philbrick & Haynes 2144. New York.
St. Lawrence Co.: 19 September 1987, Philbrick & Gale 2112. New Hampshire. Rockingham Co.: 20 August 1990, 2/12. 2n = 40. New
Hampshire. Carroll Co.: 29 November 1992, Philbrick, Philbrick & Philbrick 3192.

var. bolanderi (Hegelm.) Fassett. 2n = 20. CANADA. British Columbia. Vancouver Island, 8 June 1988, 2/55. u.s.a. California. Madera Co.:
30 May 1987, 2021; Humboldt Co.: 9 June 1987, 2046: Riverside Co.: 22 May 1987, 2002. Washington. Jefferson Co.: 20 June 1987,
Philbrick, Busse & Philbrick 2088.

C. marginata Torrey. 2n = 20. CANADA. British Columbia. Vancouver Island, 7 June 1988, Philbrick, Ceska, Ceska & Catling 2156. U.S.A.
California. San Diego Co.: 6 January 1986, 1598: Marin Co.: 5 June 1987, Philbrick & Rubtzoff 2040; Solano Co.: 2 June 1987, Phil-
brick & Anderson 2035.

C. nuttallii Torrey. 2n = 20. u.s.A.. Alabama. Conecuh Co.: 14 March 1988, Philbrick & Haynes 2136; Butler Co.: 14 March 1988, PAil-
brick & Haynes 2137; Lowndes Co.: 14 March 1988, Philbrick & Haynes 2139; Pickens Co.: 15 March 1988, Philbrick & Haynes 2142.
Mississippi. Kemper Co.: 15 March 1988, Philbrick & Haynes 2147.

C. peploides Nutt. 2n = 10. u.s.a. Alabama. Mobile Co.: 13 March 1988, Philbrick & Haynes 2127; Escambia Co.: 13 March 1988, Phil-
brick & Haynes 2131, Conecuh Co.: 14 March 1988, Philbrick & Haynes 2135, Lowndes Co.: 14 March 1988, Philbrick & Haynes 2140.

C. stagnalis Scop. 2n = 10. u.s.A4. Maryland. St. Mary’s Co.: 30 September 1984, 1386. Oregon. Jackson Co.: 11 June 1987, 2053; Clacka-
mas Co.: 13 June 1987, 2067; Pacific Co.: 16 June 1987, 2083. Washington. Grays Harbor Co.: 22 June 1987, 2/02.

C. terrestris Raf. 2n = 10. u.s.A. Mississippi. Kemper Co.: 15 March 1988, Philbrick, Haynes & McDaniels 2145; Lowndes Co.: 15 March
1988, Philbrick, Haynes & McDaniels 2145. Tennessee. Dickson Co.: 29 April 1985, 1403; Benton Co.: 29 April 1985, 1404.

C. trochlearis Fassett. 2n = 40. u.s.a. California. Mendocino Co.: 8 June 1987, 2043. Oregon. Coos Co.: 11 June 1987, 2057.

C. verna L. 2n = 20. CANADA. Quebec. Portneuf Co.: St. Augustin, 2 September 1988, Philbrick & Bruneau 2167; Charlevoix Co.: 2 Sep-
tember 1988, Philbrick & Bruneau 2168. u.s.A. California. Sonoma Co.: 7 June 1987, 2041. Colorado. Larimer Co.: 4 August 1984,
3121. New Hampshire. Coos Co.: 4 September 1988, 2/76.

1994] Philbrick— Callitriche Chromosomes 385

north of Mexico. Counts for three species confirm those that have
previously been published: 2n = 20 for C. heterophylla var. het-
erophylla Pursh (e.g., Loeve and Kapoor, 1967; Taylor and Mul-
ligan, 1968) and C. heterophylla var. bolanderi (Hegelm.) Fassett
(e.g., Taylor and Mulligan, 1968), and 2n = 6 for C. hermaphro-
ditica L. (e.g., Love, 1982; Schotsman, 1967). It is notable that
a single population of C. heterophylla var. heterophylla exhibited
2n = 40 instead of the typical 2m = 20. This is the first report of
an octoploid population of this species, but it is yet unclear if it
is of alloploid or autoploid derivation. The count for C. stagnalis,
which is introduced from Europe (Fassett, 1951), is the same as
reported from European populations (e.g., Schotsman and An-
dreas, 1980).

Prior to this study chromosome counts were unknown for five
species of Callitriche in North America. Thus, the counts pro-
vided herein are the first for these species (2n = 20 for C. mar-
ginata Torrey and C. nuttallii Torrey, 2n = 10 for C. peploides
Nutt. and C. terrestris Raf., and 2n = 40 for C. trochlearis).

The chromosome counts reported herein parallel the range in
numbers in the genus worldwide (Philbrick, 1989). Based on x =
5 there are two phyletic chromosome number series in Callitriche:
the euploid series (2n = 10, 20, 40) and aneuploid reduction series
(2n = 8, 6). All the species in North American belong to the
euploid series with the exception of C. hermaphroditica (2n = 6).

ACKNOWLEDGMENTS

The following people are thanked for providing valuable assis-
tance with field work: Marcel Blondeau, Ann Bruneau, Monique
Bruneau, Paula Busse, Adolf Ceska, Robert Haynes, Sidney
McDaniel, and Graham Philbrick. This work was supported by
a New England Botanical Club Research Grant, and National
Science Foundation Grants BSR-8701285 to C.T.P. and BSR-
8207125 to Gregory Anderson.

LITERATURE CITED
Fassett, N.C. 1951. Callitriche in the New World. Rhodora 53: 137-155, 161-
182, 185-194, 209-222.
LoeEVE, A. AND B. M. Kapoor. 1967. [.0.P.B. chromosome number reports XIV.
Taxon 16: 552-572.

386 Rhodora [Vol. 96
Love, A. 1982. I.O.P.B. chromosome number reports LXXV. Taxon 31: 342-

Puiprick, C. T. 1989. Systematic studies of North American Callitriche (Cal-
litrichaceae). Ph.D. dissertation, The University of Connecticut, Storrs, CT
SCHOTSMAN, H. D. 1967. Les Callitriches espéces de France et taxa nouveaux
d’Europe, pp. 1-152, /n: P. Jovet, Ed., Flora de France, Vol. 1. Editions Paul
Lechevalier, Paris.
. ANDREAS. 1980. i de la Région Méditerranéene.
Bull. Centr, Etud. Rech. Sci. 13: 77-88.
Taytor, R. L. AND F. A. MULLIGAN. 1968. Flora of the Queen Charlotte Islands.
Part 2. Cytological Aspects of the Vascular Plants. Queen’s Printers, Ottawa.

DEPARTMENT OF BIOLOGICAL AND ENVIRONMENTAL SCIENCES

WESTERN CONNECTICUT STATE UNIVERSITY
DANBURY, CONNECTICUT 06810

Vol. 96, No. 887, including pages 207-294, was issued January 18, 1995.

RHODORA

JOURNAL OF THE
NEW ENGLAND BOTANICAL CLUB

GORDON P. DEWOLF, JR., Editor-in-Chief

Associate Editors

DAVID S. CONANT LISA A. STANDLEY

VOLUME 96

1994

Che Neto England Botanical Club, Ine.

Harvard University Herbaria, 22 Divinity Ave., Cambridge, Mass. 02138

Supplement to

Vol. 97 Spring, 1995 No. 890

RHODORA, Vol. 96, pp. 389-392, 1994

INDEX TO VOLUME 96

Additions to the flora of Newfound-
land. III. 195-203

Alnus acuminata 69-74

Alnus jorullensis 69-74

Anderson, Jeanne E and A. A. Rez
cek. Glyceria maxima ne in
New England. 97-101

Angelo, Ray. Computer method for
producing dot distribution maps.

Arctic and subarctic additions 195-203
Arenaria biflora 198-199

Arenaria sajanensis 198-199

Argus, George 110

Atlas 190-194

Ball, Peter W. and Margaret Zoladz.
Taxonomy of Carex petricosa (Cy-
peraceae) and related species in North
America. 295-310

Bermard, John M. and Franz K. Seis-
chab. Life history of shoots of Carex

mosa F. Boott. 179-189

Biodiversity 121-169

Biology and taxonomy of the Portulaca
oleracea L. complex in North Amer-
ica: erratum. 109

Bog fen 327-353

Boivin, Bernard 111-112

BOOK REVIEW 110, 111

Bornbusch, Alan H., Lesley A Swender
and Deborah L. Hoogerwerf. Genetic
variation in Massachusetts popula-
tions of Cypripedium arietinum R.
Brown in Ait. and C. acaule Ait. (Or-
chidaceae). 354-369

Botanical history 75-96

Bouchard, André 195-203

Brouillet, Luc 195-203

Burk, C. John. Evolution ofa flora: ear-
ly Connecticut Valley botanists.
75-96

Calcareous fens of western New En-
and adjacent New York State.

Callitriche hermaphroditica 370-381,
383-386

Callitriche heterophylla var. bolanderi
370-381, 383-386

Callitriche heterophylla var. hetero-
Phylla 383-386

Callitriche marginata 370-381, 383-
386

Callitriche nuttallii 383-386

Callitriche peploides 370-381, 383-386

Callitriche stagnalis 370-381, 383-386

Callitriche terrestris 383-386

Callitriche trochlearis 383-386

Callitriche verna 383-386

Carex distichiflora 295-310

Carex sect. Aulocystis 295-310

Carex comosa 179-189

Carex franklinii 295-310

Carex kobomugi 103

Carex lepageana 295-310

Carex mangursina 295-310

Carex misandroides 295-810

Carex petricosa var. misandroides 295—
310

Carex petricosa var. petricosa 295-310

Carex stenocarpa 295-310

Cayouette, Jacques 110

Cerastium terrae-novae 198

Champlin, Richard L. Notes on the
Rhode Island flora. 102-103

Character analysis 311-326

Chromosome counts for Callitriche
(C sieaehacea) in North America.

6

Chromosome numbers 69-74, 383-
386

Chromosome numbers of some Latin
American species of A/nus (Betula-

ae). 69-74

Coastal algae 207-258

Colt, L. C., Jr. Desmonema wrangellii
(Ag.) Bornet et Flahault, a new record

8

Comparison of the Marine algae from
the Goleta Slough and adjacent open
coast of Goleta/Santa Barbara, Cal-

389

390

ifornia with those in the southern Gulf
of Maine. 207-258

Computer method for producing dot
distribution maps 190-194

Connecticut Valley 75-96

Conservation 354-369

Cooley, Dennis 75-96

Costa Rica 69-74

Cow-lilies 170-178

Crow

Cc enoedion acaule 354-369

Cypripedium ariettnum 354-369

Danthonia intermedia
nonema wrangellii (Ag.) Bornet et
Flahault, a new record for Maine
Diphasiastrum 287-293
Disjuncts 327-353

Distribution map 190-194
Dryopteris filix-mas 102

Early successional species 259-286

Ecology, reproductive biology and pop-
ulation genetics of Ophioglossum vul-
gatum (Ophioglossaceae) in Massa-
chusetts. 259-286

Edlund, Sylvia 110

Environment and vascular flora of
northeastern Iowa fen communities.

21-169

Epifagus virginiana 103

Epipactis helleborine 103

Estuarine algae 207-258

Evolution ofan flora: early Connecticut
Valley botanists. 75-96

Exine reduction in Underwater flow-
ering Callitriche (Callitrichaceae):
implications for the evolution of hy-
drophily. 370-381

Fen 327-353

Fens

Fern reproductive biology 259-286
Flora of lowa 121-169

Flora of Massachusetts 75-96

Rhodora

[Vol. 96

Floristic diversity of a disturbed west-
i Dis

Floristics 1-29

Genetic variation 354-369

Genetic variation in Massachusetts
populations of C ‘ypripedium arietin-
um R. Brown in Ait. and C. acaule
Ait. (Orchidaceae) 354-369

Genome analysis 287-293

Gervais, Camille 69-74

Gilman, Arthur V. New tri-hybrid ly-
copod, Diphasiastrum digitatum x
sabinifolium. 287-293

Glyceria maxima (Poaceae) in New En-

d 97-101

Goleta Slough 207-258

Graduate Student Research Award An-
nouncement 382

Guatemala 69-74

Hay, Stuart G., André Bouchard and
Luc Brouillet. Additions to the flora
of Newfoundland. III. 195-2

Hehre, Edward J. 207-258

Hellquist, C. Barre 110, 170-178

Hoogerwerf, Deborah L. 354-369

Hunt, David 75-96

Hybrid 287-293

Iowa flora 121-169

Lady’s slipper orchids 354-369

Les Cypéracées de est du Canada _ 11 1-
112

Levy, Foster 311-326

Life history 179-189

Life history of shoots of Carex comosa
F. Boott 179-189

Lycopod 287-293

Lycopodium 287-293

Massachusetts 75-96, 259-286

Mathieson, Arthur C. and Edward J.
Hehre. Comparison of the Marine al-
gae from the Goleta Slough and ad-
jacent open coast of Goleta/Santa

1994]

Barbara, California with those in the
southern Gulf of Maine. 207-258
Matthews, James F. Biology and tax-
onomy of the Portulaca oleracea L.
complex in North America: erratum.

McCormac, James S. and Gregory J.
Schneider. Floristic diversity ofa dis-
turbed western Ohio fen. 327-353

McJannet, Cheryl 110

McMaster, Robert T. pean ioe
ductive biology and population ge-
netics of Ophioglossum oe
(Ophioglossaceae) in Massachusetts.
259-286

MicroCAM 190-194

Minuartia biflora 198-199
oore, Darrell, Dallas Mullins and
Foster Levy. Pollen and pubescence
characteristics of Oxalis grandis
Small. 311-326

Morus rubra 103

Motzkin, Glenn. Calcareous fens of

western New England and adjacent
ew York State. 44-68
Mullins, Dallas 311-326

Nekola, Jeffrey C. Environment and
vascular flora of northeastern Iowa
fen communities. 121-169

New England 190-194

New England marine algae 207-258

New England Note 97-101, 102-103,
104-108

New tri-hybrid lycopod, Diphasia
trum digitatum x sabinifolium 287
293

New York, central 179-189

Newfoundland 195-203

Nomenclatural notes in Nymphaeaceae
for the North American Flora. 170-

North America 295-310, 383-386
Notes on the Rhode Island flora 102-

Nuphar lutea ee rubrodisca 170-178
Nuphar advena 173

Nuphar lute
Nuphar rubrodiscum

assp. advena 173-176
172-173

Index to Volume 96

Nymphaea advena 173-176
Nymphaea odorata 170-1
Nymphaea odorata ssp. tuberosa 170-

17
Nymphaea tuberosa 170-172

Observations on reproduction in 7ri-
phora trianthophora (Orchidaceae).
30-43

Ophioglossaceae 259-28

Ophioglossum vulgatum var. pseudo-
podum 259-286

Orchid 30-43

Ordination 44-68

Osborn, Jeffrey M. 370-381

Oxalis grandis 311-326

Padget, Donald J. and Garrett E. Crow.
A vegetation and floristic analysis of
acreated wetland in southeastern New

a I 1-29

Panicum amarum 102

Peat mining 327-353

Philbrick, C. Thomas and Jeffrey M
Osborn. Exine reduction in under-
water flowering Callitriche (Callitri-
chaceae): implications for the evo-
lution of hydrophily. 370-381

Philbrick, C. Thomas. Chromosome
counts for Callitriche (Callitricha-
ceae) in North America. 383-386

Pollen 311-326

Pollen and pubescence characteristics
of Oxalis grandis Small 311-326

Pollen ultrastructure 370-381

Population genetics 259-286

Porter, Jacob 75-96

Prairie fen 327-353

Primula stricta. 196-197

Pubescence 311-326

Ramets 179-189

Rare plant species 259-286

Rare plants 1-29, 327-353

Rare species 121-169

Rare vascular plants in the Canadian

Arctic (Review). 110

phat Guillermo and Camille Ger-

ais. Chromosome numbers of some

o72

atin American species of A/nus (Be-
tulaceae). 69-7
Reznicek, A. A. 97-101

Sagina caespitosa 199-2
Sagina nivalis var. caespitosa 199-200
Sagina linnaei 200-20
Sagina ec 200-201
Salix ambigua 197

Salix arctica | ;

Salix argyrocarpa 197

Salix labradorica 197
Schneider, Gregory J. 327-353
Seaweeds 207-258
Self-fertilization 259-286
Southern California 207-258
Species concept 287-293
Standley, L. A. 111-112
Substrate disturbance 327-353
Swender, Lesley A. 354-369

So

Taxonomy 295-310
Taxonomy of Carex petricosa (Cyper-
aceae) and related species in North

A
Thelypteris palustris var. pubescens 102

Rhodora

[Vol. 96

Triphora trianthophora 30-43
Tropical highland forest 69-74
Tuberoids 30-43

Underwater pollination 370-381

Vascular flora 195-203
Vegetation classification 44-68
Venezuela 69-74

Vermont 287-293

Water-lilies 170-178

Wetland mitigation 1-29

Wetlands 1-29, 44-68

Wiersema, John H. and C. Barre Hell-
quist. Nomenclatural notes in Nym-
phaeaceae for the North American
Flora. 170-178

Williams, Stephen West 75-96

Williams, Susan A. Observations on re-
production in 7Triphora triantho-
phora (Orchidaceae). 30

WordPerfect 6.0 190-194

Zoladz, Margaret 295-310

7 a4 =~ 3