RHODORA 


JOURNAL OF THE 
NEW ENGLAND BOTANICAL CLUB 


CONTENTS: 


Chromosome number determinations for Newfoundland 
species of Antennaria Gaertner (Asteraceae, 
Inuleae). Jerry G. Chmielewski 1 


Additions to the preliminary checklist of the vascular 
flora of Connecticut.. Leslie J. Mehmoff 9 


The vegetation of Pequawket Bog, Ossipee, New 
Hampshire. Linda L. Fahey and Garett E. 


Crow 39 
The marsh sow-thistle Gonchus palustris) in North 
America. J. K. Morton and Joan M. Vénn 93 
Urtica chamaedryoides Pursh (Urticaceae) reported 
as new to Cuba. Dmitry V. Geltman 96 
98 


Rhodora news/notes. Lisa A. Standley 


THE NEW ENGLAND BOTANICAL CLUB 
P. O. Box 1897, LAWRENCE, KANSAS 66044 


22 DIVINITY AVENUE, CAMBRIDGE, MASSACHUSETTS 02138 


Vol. 97 Winter, 1995 No. 889 


The New England Botanical Club, Inc. 


22 Divinity Avenue, Cambridge, Massachusetts 02138 


RHODORA 


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


Associate Editors 
DAVID S. CONANT 
LISA A. STANDLEY 


RHODORA (ISSN 0035-4902). Published four times a year (January 
April, July, and October) by the New England Botanical Club, 
810 East 10th St., Lawrence, KS 66044 and printed byAllen 
Press, Inc., 1041 New Hampshire St., Lawrence, KS 660-44. 
Second Class postage pending at Larence, KS. POSTMASTER: 
Send address changes to RHODORA, PO. Box 1897, Lawrence 
KS 66044. 


RHODORA is a journal of botany devoted primarily to NorthAmerica. 
Authors are encouraged to submit manuscripts of scientic pa- 
pers and notes relating to the systematics, distribution, ecol- 
ogy, paleobotany, or conservation biology of this or floristically 
related areas. 


SUBSCRIPTIONS: $40.00 per calendar year net, postpaid, in funds 
payable at par in the United States currency Remittances pay 
able to RHODORA. Send to RHODORA, P O. Box 1897, 
Lawrence, KS 66044-8897. 


MEMBERSHIPS: Regular $35; Family $45.Application form printed 
herein. 


BACK VOLUMES AND SINGLE COPIES: Some available; informa- 
tion and prices will be furnished upon request to thélreasurer. 


ADDRESS CHANGES: In order to receive the next number of 
RHODORA, changes must be received by the business 
office prior to the first day of January, April, July, or 
October. 


INFOTMATION FOR CONTRIBUTORS: Inside back cover, January 
and Apmil.. 


MANUSCRIPTS: Send to: Gordon P DeWolf, Jr. 
Managing Editor, RHODORA 
125 Long Hill Road 
West Brookfield, MA 01585 


This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). 


RHODORA 


JOURNAL OF THE 
NEW ENGLAND BOTANICAL CLUB 


Vol. 97 Winter 1995 No. 889 


RHODORA , Vol. 97, No. 889, pp. 1-8, 1995 


CHROMOSOME NUMBER DETERMINATIONS FOR 
NEWFOUNDLAND SPECIES OF ANTENNARIA 
GAERTNER (ASTERACEAE: INULEAE) 


JERRY G. CHMIELEWSKI 
ABSTRACT 


Chromosome numbers were determined for 54 individuals of Newfoundland 

Antennaria representing 8 species: A. cana (2n=56), A. columnaris (2n=56), A. 
confusa (2n=56), A. eucosma (2n=56), A. gaspensis (2n=56), A. howellii 
(2n=56), A. pulvinata (2n=56), and A. staminea (2n=56). The tetraploid deter - 
minations for populations of both A. eucosma and A. howellii (reported as A. 
neodioica) support previous determinations for these two species. Chromo- 

some number determinations for all other species represent first reports. 


Key Words: Antennaria, Newfoundland, chromosome numbers, taxonomy 
INTRODUCTION 


The dioecious genus Antennaria occurs predominantly 
throughout the cold temperate and arctic regions of the North- 
ern Hemisphere. Polyploidy is extensive in the genus, as 1s 
the associated agamospermy. Chromosome number deter- 
minations are available for approximately 2500 individuals 
of North American Antennaria representing 66 described 
species and 15 naturally occurring hybrids (Stebbins, 
1932a,b; Love and Love, 1964, 1982; Johnson and Packer, 
1968; Mosquin and Hayley, 1968; Strother, 1972; Packer 
and McPherson, 1974; Bayer and Stebbins, 1981, 1987; 
Morton, 1981; Urbanska, 1983; Bayer, 1984, 1988, 1989a,b, 


] 


2 Rhodora [Vol. 97 


1990, 1991, 1992; Chinnappa, 1984, 1986; Evert, 1984; Bayer 
and Crawford, 1986; Chmielewski and Chinnappa, 1988a,b, 
1990). Of these reports, only two, Morton (1981) and 
Urbanska (1983), have documented chromosome numbers 
for Newfoundland species of Anlennaria, specifically six 
tetraploid determinations for A. eucosma Fernald and 
Weigand. 

Reproductive mode in association with polyploidization 
have directed the evolutionary history of the genus. These 
factors combined with hybridization have led to the forma- 
tion and establishment of numerous, morphologically vari- 
able, races and clones, many of which are apomictic. Fernald 
(1933) stated that 16 species and 2 varieties of Anfennaria 
occurred on the island of Newfoundland. These taxa are, 
however, technically difficult and the status of some of these 
taxa has since been questioned (e. g. Bouchard et al., 1991) 
and various resolutions proposed (e. g. Bayer, 1989c; Bayer 
and Stebbins, 1993; Chmielewski, 1993, 1994). Chmielewski 
(1994) most recently placed A. bayardi Fernald, A. 
brunnescens Fernald and A. foggii Fernald in synonymy with 
A. pulvinata Greene. Widespread acceptance has not yet oc- 
curred for any of the recently proposed changes. Revision of 
the Newfoundland Antennaria is necessary, however, to de- 
termine whether 8 of the taxa cited by Fernald (1933), that is, 
A. albicans Fernald, A. cana (Fernald & Wiegand) Fernald, 
A. columnaris Fernald, A. gaspensis Fernald, A. petaloidea 
(Fernald) Fernald var. subcorymbosa (Fernald) Fernald, A. 
straminea Fernald, A. vexillifera Fernald, and A. wiegandii 
Fernald legitimately warrant rare status and therefore protec- 
tion (see Bouchard e/ al. 1991) or more appropriately should 
be placed in synonymy with other species and therefore re- 
lieved of this designation. Additionally, the status of A. confusa 
Fernald, A. eucosma, Fernald and Weigand, A. neodioica 
Greene var. attenuata Fernald, A neodioica var. chlorophylla 


1995] Chmielewski-New/oundland Antennaria 3 


Fernald, A. neodioica var. typica Fernald, A. rupicola Fernald 
and A. spathulata (Fernald) Fernald need to be established as 
these taxa were excluded from the rare list for Newfoundland 
because they: (1) were considered to be too widespread or 
common in their habit, even though they may be restricted 
geographically or (2) are minor variants of other more wide- 
spread or common species (Bouchard ef al. 1991). 

The only study to date which has dealt specifically, but not 
exclusively, with documenting morphological variation in 
Newfoundland species of the genus Antfennaria is that of 
Chmielewski (1994). Until a satisfactory, detailed, taxonomic 
treatment of the Newfoundland Antennana per se is prepared, 
the nomenclature proposed by Fernald (1933) and 
Chmielewski (1994) for these taxa will be retained. The 
present report which deals with chromosome number deter- 
minations for these taxa is intended to stimulate subsequent 
revisionary studies. 


MATERIALS AND METHODS 


Rootstocks were washed free of soil particles and debris in 
the field and subsequently returned to the campus greenhouse 
where they were transplanted into a mixture of equal parts 
soil, vermiculite and sand. Plants were watered as necessary 
and grown under natural daylight conditions and a tempera- 
ture range of 15-25°C. Voucher specimens were identified 
using the key to Newfoundland Antennaria following Fernald 
(1933) and Chmielewski (1994). All vouchers of Antennaria 
cana (Fernald and Weigand) Fernald, A. columnaris Fernald, 
A. confusa Fernald, A. eucosma Fernald and Weigand, A. 
gaspensis Fernald, A. howellti Greene, A. pulvinata Greene, 
and A. staminea Fernald were collected by the author and sub- 
sequently deposited at SLRO (Holmgren et al., 1990). Dupli- 
cate collections were deposited at MT. 


4 Rhodora Vol. 97 


Mitotic chromosome counts were made from root-tips that 
were collected at approximately mid-morning of sunny days, 
treated with a saturated PDB solution for 2 h, fixed in a mix- 
ture of ethanol and glacial acetic acid (3:1), subsequently 
hydrolized in IN HCI at 60°C for 10-15 minutes and then 
squashed in 2% acetic orcein stain. 


RESULTS AND DISCUSSION 


Chromosome numbers were determined for 54 individuals 
representing 8 species of Anfennaria (Table 1). The somatic 
chromosome number of 56 is widespread in the genus and in 
North America is generally considered to represent the tetra- 
ploid condition as no species with somatic numbers lower 
than 28 have been reported. European authors have, how- 
ever, historically treated the somatic count of 28 as tetraploid 
as the somatic number of 14 is known for the sister genus 
Gnaphalium (Gustafsson, 1947; Urbanska, 1983, a and b.). 
Except for tetraploid chromosome number determinations for 
A. eucosma and A. howellti which support previous counts 
for the species, all other determinations represent first reports 
and are presented without further comment. Tetraploid deter- 
minations previously reported for A. eucosma, are also from 
Newfoundland populations (Morton, 1981; Urbanska, 1983). 
Thus, based on available determinations, this species 1s 
tetrapoloid throughout its distribution. The same is not true 
for A. howellii (=A. neodioica), as both aneuploid and eup- 
loid determinations have been previously reported (Stebbins, 
1932b; Bayer and Stebbins, 1981, 1987; Bayer, 1984; Bayer 
and Crawford, 1986; Chinnappa, 1986; Chmielewski and 
Chinnappa, 1988, a and b, 1990). Tetraploid determinations 
for A. howellii occur throughout the more southern portion 
of the species distribution intermingled with other cytotypes 
(Bayer, 1984; Bayer and Stebbins, 1981,1987; Bayer and 


ws 


1995 Chmielewski N/, f; {] ! Antennaria 


J 


Crawford, 1986; Chinnappa, 1986). No other cytogeographic 
trends are recognizable for the species at this time. 


Table 1. 
Chromosome number determinations for 8 species of Antennaria from 
Newfoundland. Listed are species, chromosome number location, date 
of collection and collection number Staminate individuals are designated 
with an asterisk (*) following the collection number All collections by 
J.G.Chmielewski. 


A. cana Fernald, 2” = 56. Newfoundland: District of St. Barbe North, Boat 
Harbour, July 13, 1993, 2973. 


A. columnaris Fernald, 2n = 56. Newfoundland: District of St. Barbe South, 
Pointe Riche Peninsula, Port au Choix National Historic Park, Gar gamelle Cove, 
July 16, 1993, 3007, 3008. 


A. confusa Fernald, 2n = 56. Newfoundland: District of Port au Port, Nfld 462, 
Port au Port, Table Mtn, July 6, 1993, 2963. District of St. Barbe North, Boat 
Harbour, July 13, 1993, 2974. Cape Norman region, July 13, 1993, 2977, 2979. 
Nfld 435, 0.5 km south of Cook's Harbour, July 13, 1993, 298/, 2986 . East of 
Big Brook, July 13, 1993, 2989, 2990. District of St. Barbe South, Pointe Riche 
Peninsula, Port au Choix National Historic Park, vicinity of Pointe Riche, July 
15; 1993,..3003, 


A. eucosma Fernald & Wiegand, 2” = 56. Newfoundland: District of Port au 
Port, Nfld 462, Port au Port, Table Mtn, July 6, 1993, 2967. District of St. Barbe 
North, Boat Harbour, July 13, 1993, 2977*. Cape Norman region, July 13, 1993, 
2975*, Nfld 435, 0.5 km S of Cook's Harbour, July 13, 1993, 2987*. East of Big 
Brook, July 13, 1993, 2992*, 2993. District of White Bay North, Nfld 430, west 
of St. Anthony airport, July 15, 1993, 2994*, 2995 


A. gaspensis Fernald, 2n = 56. Newfoundland: District of Port au Port, Nfld 
462, Point au Mal, Ragged Ass road, July 5, 1993, 2959. Nfld 460, Lower Cove, 
July 5, 1993, 296]. District of St. Barbe South, Nfld 430, Table Point Ecological 
Reserve, north of Bellburns, July 16, 1993 , 30/7, 30/73. 


A. howellii Greene, 2” = 56. Newfoundland: District of Grand Falls, Trans Canada 
Highway, 4.5 km west of Red Cliff Road, west of Grand Falls, July 23, 1993, 
3020. District of Lewisporte, Nfld 331, 1.1 km east of Nfld 340, July 23, 1993, 
3021, 3022. District of Placentia West, Nfld 210, 1.1 km south of Cow Head exit, 
Marystown area, July 26, 1993, 3031. District of Port au Port, Nfld 460, Lower 
Cove, July 5,1993, 2960. Nfld 460, Cape St. George, July 5, 1993, 2962. Nfld 
462, Port au Port, Table Mtn, July 6, 1993, 2970. District of St. Barbe South, 
south of Bellburns, July 16, 1993, 30/5. Gros Morne National Park, just west 


6 Rhodora Vol. 97 


of Bear Cove, July 18, 1993, 30/7. District of Twillingate, Nfld 340, 1.8 km 
south of Summerford, July 23, 1993, 3023. Nfld 340, 0.3 km south of Newville, 
July 23, 1993, 3024, 3025. Twillingate, summit of Smith’s Lookout Road, July 
23, 1993, 3028, 3029. Nfld 340, 0.3 km north of William B. Elliott Causeway, 
July 23, 1993, 3030 


A. pulvinata Greene, 2n = 56. Newfoundland: District of St. Barbe South, Pointe 
Riche Peninsula, Port au Choix National Historic Park, vicinity of Pointe Riche, 
July 15, 1993, 2998, 3000, 300]. Gargamelle Cove, July 16, 1993, 3009. 


A. Straminea Fernald, 2n = 56. Newfoundland: District of St. Barbe South, 
Pointe Riche Peninsula, Port au Choix National Historic Park, vicinity of Pointe 

Riche, July 15, 1993, 2997, 3002, 3004. Gargamelle Cove, July 16, 1993, 30/0. 
Nfld 430, Table Point Ecological Reserve, N of Bellburns, July 16, 1993, 30/2, 
3014. S of Bellburns, July 16, 1993, 30/6. Gros Morne National Park, ‘Trout River 
Valley Road, Tableland Mtn., barren knob at mouth of Winter House Gorge, July 
19, 1993, 3018, 3019. District of Twillingate, Twillingate, summit of Smith's Look- 
out Road, July 23, 1993, 3027. 


ACKNOWLEDGEMENTS 


This research was funded by a State System of Higher 
Education Professional Development Grant (Pennsylvania) 
and a Slippery Rock University International Studies Award 
to JGC. Research permits for Gros Morne National Park and 
Port au Choix and L’Anse aux Meadows National Historic 
Parks were kindly issued by the Environment Canada Parks 
Service. Agriculture Canada and the USDA Plant Protection 
Division are thanked for their efforts which allowed for the 
removal of specimens from Newfoundland and their impor- 
tation to the US. 


LITERATURE CITED 


BAYER, R.J. 1984. Chromosome numbers and taxonomic notes for North American 
species of Antennaria (Asteraceae: Inuleae) Syst. Bot. 9: 74-83 
1988. Patterns of isozyme variation in western North hance 
Antennaria (Asteraceae: Inuleae) I. Sexual species of sect. Dioicae. Syst. 
Bot. 13: 525-537. 
es A Sea and phytogeographic study of Antennaria aromatica 
a (Asteraceae: Inuleae) in the western North Am 
ee ae 36: 248-259 


1995 Chmielewski-Newfoundland Antennaria 7 


1989b. Patterns of isozyme variation in the Antennaria rose 

ee Inuleae) polyploid agamic complex. Syst. Bot oe 389-397. 
1989c. A taxonomic revision of the Antennaria rosea (Asteraceae: Inuleae: 

eee pomDiel’ on: Brittonia 41:53- 
1990. A of Antennaria media, A. pulchella, and A. 
scabra (Asteraceae Inuleae) of the Sierra Nevada and White Mountains. 
ee : 171-183. 

99]. A ces and morphological variation in Antennaria (Asteraceae: 
| eae) from the low arctic of Northwestern North America. Syst. Bot. 
16: 4 6. 


1992. Allozyme variation, genecology, and phytogeography of 
lela arcuata (Asteraceae), a rare species from the Great Basin and 
with small disjunct populations. Amer. J. Bot. 79: 872-881. 
AND . 1 CRAWFORD. 1986. Allozyme divergence among five diploid 
species of Antennaria (Asteraceae: rtlede) and their allopolyploid 
derivatives. Amer. J. Bot. 73: 287-296 
AND G. L. STEBBINS. 1981. Chromosome numbers of North American 
species of Antennaria Gaertner (Asteraceae: Inuleae). Amer. J. Bot. 68: 
1342-134 
7. Chromosome numbers, patterns of 
distribution and apomixis in phates (Asteraceae: Inuleae). Syst. Bot. 12: 


1993, A synopsis with keys for the genus 
Antennaria (Asteraceae: Inileae: Gnaphaliinae) of North America. Can. J.. 
Bot. 71: 1589-1604. 
BOUCHARD, A., HAY, S., BROUILLET, L., JEAN, M., and SAUCIER, I. 1991. The 
rare vascular sleet of the island of Newfoundland. Syllogeus Series No, 65. 
nadian Museum of Nature, Ottawa, Canada. 
annie. C.C. 1984. Triploidy in the genus Antennaria (Asteraccae: Inuleae). Can. J. 
Genet. Cytol, 26: 133-136. 
1986. Chromosome met eli in ia eee Inuleae) from west 
ern North America. Can. J. Genet tol. 28: 468-475. 
CHMIELEWSKI, J. G. 1993. ee Greene: The legitimate name for A. 
omatica Evert (Asteraceae: Inuleae). Rhodora 95:261-276. 
1994, Evaluation of the taxonomic status of Antennaria bayardi, 
brunnescens, and A. foggii (Asteraceae: Inuleae). Can. J. Bot. 72: 1775-1777. 
C ice J.G. AND C.C. C AN APPA. 1988a. ee neisied notes and chromo 
aetene in Antennaria Sanne Asteraceae: Inuleae) from arctic 
Noith America. Arctic and Alp. Res. 20: 117-12 
1988b. The g genus Antennaria acckenmois 
Inuleae) in western North America: II. Additional chrom e 
Rhodora 90: 133-137 


= 


0. The genus Antennaria (Asteraceae: 
Taek in arctic North America: Chromosome numbers and taxonomic notes. 
a 92: 2 6. 


h -2 
EVERT, E. F. 1984, A new species . Antennaria (Asteraceae) from Montana and Wyom- 
apres 31: eel 
FERN. LD. ML Newfoundland fl 
» Newtonian Rhodora 35: 327-3 
GUSTAFSSON, 1947. . Apomixis in aoee plants Acta Univ, Lund. Kungl. Fysiogr. 
Sallsk. et N. R. Avd. 2 42- 


The genus Antennaria 


8 Rhodora Vol. 97 


HOL ee P.K., N.H. HOLMGREN AND L.C. BARNETT. 1990. Index oe 
h ed. Part I. The herbaria of the world. The New York Botanical Gard 
an onx, NY. 
JOHNSON, A.W. AND J.G. PACKER. 1968. Chromosome numbers in the flora of 
Ogc ae Creek, N.W. Alaska. Bot. Not. 121: 403-456. 
LOVE, A. AND D. LOVE. 1964. IOPB chromosome number reports. I.T axon 13: 108- 
109. 


. 1982. IOPB Chromosome number reports LXAXYV . Taxon 31: 
2-368. 
MORTON, I K. 1981. Chromosome numbers in Compositae from Canada and the U.S 
Bot. J. Linn. Soc. 82: 357-368. 
set aas T. AND D.E. HAYLEY. 1966. Chromosome — and taxonomy of some 
anadian arctic plants. Can. J. Bot.: 44: 1209-12 
PACKER, i: G. se G.D. MCPHERSON. 1974. acu numbers in some vascular 
plants from northern Alaska. Can. J. Bot. 52: 1095-1099 . 
STEBBINS, G. 4 1932a. Cytology of Antennaria. 1 Normal species. Bot. Gaz. 94: 134- 


s 


es Cytology of Antennaria. II Parthenogenetic species. Bot. Gaz. 


2-3¢ 
STROTHER, ie : 1972. Chromosome studies in western North American Compositae. Amer. 
J. Bot. 59: 242-247. 
URBANSKA, K.M. 1983a. Antennaria carpatica (Wahlb.) Bl. et Fing.s.l. in North America. 
Chromosome numbers, geographic distribution and ecology . Ber. Geobot. Inst. 
Eidg. Tech.Hochsch.; Stift. Ruebel Zuer. 50: 33-66. 
1983b. Cytogeographical dif ferentiation in Antennaria carpatica s. Ll. Bot. 
Helv.: 123-131 


DEPARTMENT OF BIOLOGY 
SLIPPERY ROCK UNIVERSITY 
SLIPPERY ROCK, PENNSYLVANIA 16057 


RHODORA, Vol. 97, No. 889, pp. 9-38, 1995 


ADDITIONS TO THE PRELIMINARY CHECKLIST OF 
VASCULAR FLORA OF CONNECTICUT 


Leste J. MERHOFF 
ABSTRACT 


Forty-six native or naturalized species, varieties, or hybrids are reported as addi- 

tions to Dowhan's Preliminary Checklist of the Vascular Flora of Connecticut. 
These represent 2] native taxa and 25 apparently naturalized taxa. Each taxon is 

presented with a brief history of its discovery in Connecticut. Dates are presented 

for the earliest known voucher specimen for each addition. 


Key words: Connecticut, flora, vascular plants, additions 


The following list of taxa represents additions to the non- 
cultivated flora of Connecticut that have been discovered since 
the publication of the PRELIMINARY CHECKLIST OF 
THE VASCULAR FLORA OF CONNECTICUT (Dowhan 
1979). Most are new discoveries although the recent annota- 
tion of specimens collected prior to 1979 accounts for some 
of the new records. In some cases taxa considered here were 
not included in regional floristic treatments such as Fernald 
(1950), Seymour (1969), or Gleason and Cronquist (1963, 
1991). Some taxa appear to be adventive and it remains to be 
seen whether or not these will become established 1n Con- 
necticut 

A] species, 2 varieties, and 3 naturally occurring hybrids in 
29 families are reported. 21 taxa appear to be native while 
the remaining 25 appear to be non-native or ruderal taxa and 
should be considered as adventive or naturalized. One new 
family, the Hymenophyllaceae, 1s added to the flora. (Six taxa 
are listed as Endangered and three as Special Concern Spe- 
cies on the Connecticut list of Endangered, Threatened, or 
Special Concern Species (Department of Environmental 


2 


10 Rhodora [ Vol.97 


Protection 1993). 34 records are from recent discoveries and 
12 records represent additions due to revisionary reevalua- 
tions. 

It seems prudent to follow the family order as presented in 
Dowhan (1979). Taxa within families are alphabetized by 
genus, and those within genera are alphabetized by species. 
However, familial names used here all end in -aceae, alterna- 
tive names being used for the sake of consistency in certain 
families. Nomenclature follows the Flora of North America 
(Flora of NorthAmerica Editorial Committee 1993) for pteri- 
dophytes and Gleason and Cronquist (1991) for angiosperms. 

A single specimen ts cited at the end of the discussion for 
each taxon. Citations are for the earliest voucher specimen 
known to me. No attempt has been made to include multiple 
vouchers from different herbaria where duplicates may have 
been deposited, although I am aware many exist. These 
records represent collections deposited at CONN, GH, MASS, 
NCBS, NEBC, NYS, and YU (Holmgren ef al. 1990). 

Distributional information comes from Flora of North 
America, Volume 2 (Flora of North America Editorial Com- 
mittee 1993) for pteridophytes and Gleason and Cronquist 
(1991) for angiosperms unless otherwise noted. A table sum- 
marizing all additions and a map showing Connecticut’s 8 
counties are included. . 


LY COPODIACEAE 


Lycopodiella alopecuroides (L.) Cranfill 
Foxtail Club-moss 
This taxon was originally reported as Lycopodium 
alopecuroides L. by Edwin H. Eames from Milford, New 
Haven County in 1908. Eames also collected L. alopecuroides 


1995 Mehrhoff-Connecticut 11 


» i panes 
4 ies 
i Se Reng irae 


Figure 1. Map of Connecticut showing County and Town Boundaries. 


in Fairfield County. The correct identification of the specime- 
mens had been questioned by Joseph Dowhan and others. 
Revisionary work by the late Joseph Beitel and Warren H. 
Wagner, Jr. has confirmed Eames’ original determination. This 
species, under the synonym Lycopodium alopecuroides L. is 
listed as a Species of Special Concern in Connecticut (DEP 
1903) 
Distribution: Texas and Louisiana, north to Rhode Island 
and Massachusetts, mostly on coastal plain but inland to 
western North Carolina and northern Georgia. 
[ 26 SEP 1908, E. H. Eames s. 7. (CONN)| 


1 Rhodora [ Vol. 97 


Lycopodiella xcopelandii (Eiger) Cranfill. [Lycopodiella 
alopecuroides (L.) Cranfill x L. adpressa (Lloyd & Underw.) 
Cranfill] . Hybrid Bog Club-moss 
Numerous Connecticut specimens were annotated as this 
hybrid by Florence S. Wagner and Warren H. Wagner, Jr. These 
records were from Fairfield, New Haven, and Middlesex 
Counties. The late Joseph Beitel had called my attention to a 
collection in the G. Safford Torrey Herbarium (CONN) from 
Simsbury, Hartford County, of a Lycopodiella which he de- 
termined to be L. xcopelandii. Subsequent evaluation of ma- 
terial from that herbarium did not yield a specimen which the 
Wagners felt comfortable in assigning to this hybrid taxon. 
Distribution: This taxon can be expected anywhere within 
the range of either parent species. 
[30 AUG 1907, E. H. Eames s. n. (CONN)] 


SELAGINELLACEAE 


Selaginella eclipes Buck. Buck’s Meadow Spike-moss 

This spike-moss is confused with the common Selaginella 
apoda (L.) Spring and may have been overlooked in the north- 
east. In 1982 Terry R. Webster found specimens determined 
as S, apoda collected in western Connecticut which matched 
Buck’s description of this species (Buck 1977). Numerous 
calcareous localities in western Connecticut have yielded 
plants whose morphologies are intermediate between S. 
eclipes and S. apoda, adding to the confusion surrounding 
this taxon. Specimens, which we identified as S. eclipes, taken 
from the floor of an abandoned limestone quarry have re- 
cently been confirmed by IvanA. Valdespino at the New York 
Botanical Garden. Shortly after collecting these specimens 
from the quarry floor, the quarry was reactivated for pro- 
cessing marble.It now appears to have been abandoned 


1995] Mehrhoft-Connecticut 13 


again and should be revisited. Any specimen from calcare- 
ous areas and growing in full sun presumed to be S. apoda 
should be given more than a cursory glance. 
Distribution: Western Quebec and eastern Ontario south 
through western New York to Oklahoma and Arkansas. 
[26 APR 1983, L. J. Mehrhoff 7546, with T. R. Webster 
(CONN) ] 


OSMUNDACEAE 


Osmunda xruggii Tryon [Osmunda claytoniana L. x O. re- 
galis L. var. spectabilis (Willd.) Gray] 
Interrupted Royal Fern 
This very rare hybrid was first collected in Wilton, Fairfield 
County, Connecticut in 1931 by Leonard J. Bradley. The type 
specimen came from a garden in Hartford, by way of the gar- 
den of Dr. Harold G. Rugg of Dartmouth College in New 
Hampshire (Tryon 1940). The last Wilton collection appears 
to be from 1938. It is not currently known to be extant in 
Connecticut. 
Distribution: Fairfield County, Connecticut and Craig 
County, Virginia. 
[14 JUN 1931, L. J. Bradley s. n. (GH)] 


HY MENOPHY LLACEAE 


Trichomanes intricatum Farrar. Appalachian Trichomanes 
The species name Trichomanes intricatum Farrar has only 
recently been published (Farrar 1992) although the gameto- 
phyte has been known for several years (Farrar et al. 1983). 
This fern gametophyte, which apparently never produces a 
sporophyte, was reported from Kent and Norfolk, Litchfield 
County in 1983 by Donald R. Farrar, James C. Parks, and 


14 Rhodora Vol. 97 


Bruce W. McAlpin (Farrar ef al. 1983). Gametophytic plants 
of Trichomanes intricatum, once described to me by Rolla 
Tryon as “resembling green steel wool” can be easily over- 
looked. It has recently been found in Hartland, Hartford 
County and should be looked for in other areas. It is a Spe- 
cies of Special Concern in Connecticut, listed as Trichomanes 
sp. (DEP 1993), 
Distribution: Central Vermont and New Hampshire south 
along the Appalachian uplands to Alabama and Georgia, 
disjunct in southern [Illinois and Indiana and western Ken 
tucky. 
[30 SEP 1991, L. J. Mehrhotf /5299, with M. Ardwin, J. 
Barrett, and N. Proctor (CONN)] 


PTERIDACEAE 


Pellaea glabella Mctt. ex Kuhn subsp. glabella 
Smooth Cliff-brake 
The smooth Cliff-brake was first reported from Connecticut 
in 1988 from a single calcareous outcrop in Salisbury, 
Litchfield County by Karen S. Hansen and Robert E. 
Schneider. The population seems to be well established as 
there are many individuals in the population. It is listed as an 
Endangered Species in Connecticut (DEP 1993). 
Distribution: Vermont to Minnesota and south to Tennes- 

see, Virginia, and western Maryland. 

[27 JUN 1988, K. S. Hansen and R. E. Schneider 24/ 
(CONN)| 


ALISMATACEAE 


Echinodorus tenellus(Mart.)Buchenau Burhead 
This small aquatic was first reported in Connecticut from a 


1995] Mehrhotf-Connecticut 1S 


pond margin in Glastonbury, Hartford County in 1989 by 
William Moorhead. No individuals of Echinodorus were ob- 
served at this station for three subsequent years. This is not 
atypical for this taxon (C. B. Hellquist pers. comm.). A small 
number of plants of /chinodorus were seen at this site in 
1993 (K. J. Metzler pers. comm.). Numerous plants in flower 
and fruit were observed in 1994. This appears to be the only 
extant New England population. It is listed as an Endangered 
Species in Connecticut (DEP 1993). 

Distribution; Tropical America north along the Atlantic 

Coast irregularly to Massachusetts, up the Mississippi River 

to Illinois, Missouri, occasionally Kentucky and Kansas. 
[18 AUG 1989, K. J. Metzler 89001 (CONN) | 


HY DROCHARITACEAE 


Egeria densa Planch. 

This aquatic species is commonly grown in aquaria. /.geria 
densa was introduced, apparently intentionally, in Westport, 
Fairfield County where it has persisted. It also has been oc- 
casionally introduced in Massachusetts and Vermont (Crow 
and Hellquist 1982). It should be watched for elsewhere in 
southern New England. It might be confused with species of 
Flodea. 

Distribution: Occasional in northeastern United States. Na 
tive of South America, from southeastern Brazil to north- 
ern Argentina. 

(20 AUG 1992, P. Aarrestad s. n. (CONN)] 


POACEAE 


Aira praecox L. 
This European Hairgrass was first reported from Norwich, 


16 Rhodora [\ol. 97 


New London County in 1992. The population was well es- 
tablished in sandy soils of a “jug-handle” at the junction of 
Routes 2 and Interstate [-395. It was not noticed at this site in 
1982 when collections were made in the same locality. A sec- 
ond station was found near an exit ramp off the interstate in 
Old Lyme, New London County in 1994, 
Distribution: Eastern Connecticut to Virginia, usually near 
the coast. Native of Europe. 
[12 JUN 1992, L. J. Mehrhotf /56/8 (CONN)] 


Microstegium vimineum (Trin.) A. Camus 

The precise history of this species in Connecticut is some- 
what unclear. Lauren Brown (pers. comm.) observed this spe- 
cies in Branford, New Haven County in the early 1980s but 
apparently no specimens were taken. In 1990 a population of 
this species was observed in East Haddam, Middlesex County 
by T. Hendrickson and collected under the name Eulalia 
viminea (Trin.) Ktze. By 1991 it appeared to be well estab- 
lished at numerous sites in New London (W. Dreyer pers. 
comm.) and Fairfield Counties and at single sites in both 
Hartford and Litchfield Counties. Tolland County specimens 
are from unwanted volunteers, probably from fruits falling 
from New Jersey material while it was being pressed. (Re- 
peated attempts have been made to eradicate this species at 
this site.) The first North American collection of Microstegium 
vimineum was made in 1919 in Knoxville, Tennessee 
(Fairbrothers and Gray 1972). It was first collected in New 
Jersey in 1959 (Fairbrothers and Gray 1972). This invasive 
grass 1s now well established in many areas in northern New 
Jersey, Pennsylvania, and southeastern New York (Hunt and 
Zaremba 1992; pers. obs.). It occurs along roadsides, in allu- 
vial woods, on serpentine barrens, and ruderal habitats and 
can form extensive stands to the exclusion of almost every 


1995] Mehrhoff-Connecticut 17 


thing else. This species should be closely monitored and con- 
trolled if at all possible. 
Distribution: Connecticut and New York southward. Na- 
tive of tropical Asia. 
[19 SEP 1990, T. Hendrickson s. n. (NCBS)] 


Panicum amarulum A. Hitchc. & Chase 

Panicum amarulum was first taken in Connecticut from a 
roadside in Hebron, Tolland County in 1983. It is also known 
from North Haven, New Haven County. P. amarulum may 
have been originally introduced as an ornamental or acciden- 
tally introduced with roadside plantings or seeding at these 
sites. It is now spreading and appears to be well established. 
Panicum amarulum is not thought to be native north of New 
Jersey. Massachusetts records from Cape Cod are thought to 
be introductions (B. Sorrie pers. comm.). It is treated here as 
distinct from Panicum amarum Ell. 

Distribution: New Jersey to Mexico, occasionally inland in 

North Carolina and West Virginia. 
[12 SEP 1983, L. J. Mehrhoff 9240 (CONN)] 


Panicum scabriusculum Elliott 

This species was originally collected as Panicum aculeatum 
Hitche. & Chase in Stafford, Tolland County on 21 JUN 1911 
by Charles H. Bissell. A specimen from this collection was 
determined not to be Panicum aculeatum by Joseph J. Dowhan 
(1979). A duplicate collection was later discovered at the 
Smithsonian Institution (US) that had been annotated as 
Dichanthelium scabriusculum (Elliott) Gould & Clark by the 
late F C. Gould and C. A. Clark. A recent specimen of Panicum 
scabriusculum was collected in Voluntown, New London 
County, in 1989 by William J. Crins. This species 1s list- 
ed under Dichanthelium scabriusculum as an Endangered 


18 Rhodora [\ol. 97 


Species in Connecticut (DEP 1993), 
Distribution: Connecticut and New Jersey to Florida and 
Texas. 

[13 JUN 1989, W. J. Crins 7628 (NYS)] 


Vulpia myurus (L.) C. Gmelin Rat-tail Fescue 
Vulpia myurus was collected at two sites less than 1 km 
from each other near Long Island Sound in Fairfield, Fairfield 
County on 28 JUL 1992. Both sites were open, sandy disturb- 
ed areas. One site was along a path to Long Island Sound and 
the other was adjacent to a gravel parking area removed from 
the coast, The spikelets were disarticulating at the time of 
collection, specimens from other localities should be sought 
at an carlier date. 
Distribution: Widespread. Native of Europe. 
[28 JUL 1992, L. J. Mehrhoff 759/79, with W. E. Brumback 
(CONN)] 


CYPERACEAE 


Carex backii Boott 

Vegetative material, thought to be Carex backii, was col- 
lected on a marble ridge in Canaan, Litchfield County by 
Thomas Rawinski in 1988. In 1992, Elizabeth Thompson, 
reportedly unaware of Rawinski’s find, collected fertile ma- 
terial of C. backii from the same calcareous ridge. It seems 
reasonable, in light of Thompson’s collection, to assume 
Rawinski’s specimen’s belongs to this taxon. Because of its 
apparent rarity in the state, this species should be considered 
for inclusion on Connecticut’s list of protected species. 

Distribution: Quebec to New Jersey, west across to Minne- 

sota, Utah, Oregon, and British Columbia. 


1995] Mehrhoff-Connecticut 19 


Carex emoryi Dew. 

A 1907 Edgar B. Harger collection of Carex aquatilis 
Wahlenb. was determined by Lisa Standley in 1990 to be 
Carex emoryi. Harger had collected the specimen in “wet 
ground at Selden’s Cove”, Lyme, New London County, Con- 
necticut. This is the same locality from which nineteenth 
century specimens of Nelumbo lutea (Willd.) Pers. were taken. 
Nelumbo lutea and Carex emoryi have similar distributions. 
Iam suspicious of the nativeness of Nelumbo at this site. 

Distribution: Southern New York to North Dakota and 

Manitoba, south to Virginia, Arkansas, and Texas. 

[22 JUN 1907, E. B. Harger 5/47 (NEBC)] 


Cyperus echinatus (L.) Wood 
Collections of William R. Dudley from North Branford 
made in 1881 found in Yale’s D. C. Eaton Herbarium (Y U) 
have been annotated by Gordon C. Tucker to Cyperus ovularis 
(Michx.) Torr. a synonym for Cyperus echinatus. 
Distribution: Connecticut and New York to southern Ohio, 
Illinois, and eastern Kansas, south to Florida and northeast 
ern Mexico. 
[1 JUL 1881, William R. Dudley s.n. (YU)] 


Rhynchospora scirpoides (Vahl) Griseb. 

This was originally reported as Psilocarya scirpoides Torr. 
from a shallow pond margin in Simsbury, Hartford County in 
1981 (Mehrhoff 1982b). It has been reported as recently as 
1990 at the same site. A 1994 visit to Great Pond yielded no 
plants of Rhynchospora scirpoides and little suitable habitat 
due to increased water level. Active management is urgent- 
ly needed at this site. It is listed under Psilocarya scirpoides 
| as an Endangered Species in Connecticut (DEP 1993), 

Distribution: Eastern Massachusetts and Rhode Island, 


20 Rhodora [ Vol. 97 


nortwest Indiana, southwest Michigan, southeastern 
Virginia and eastern North Carolina. 
[3 SEP 1981, L. J. Mehrhoff 5454 (CONN)|] 


LiLIACE Ae 


Ornithogalum nutans L. Nodding Star-of-Bethelehem 
Ornithogalum nutans was collected from Groton, New Lon- 
don County in 1983 by E. A. Christensen, M. W. Lefor, and 
R. Piacentini (Christensen and Lefor 1985). A previously 
unidentified collection of this taxon from East Haddam, 
Middlesex County was identified by Eric Christensen in 1984. 
It is questionable whether or not either of these populations 
are established. 
Distribution: Occasionally escaped from cultivation. Na- 
live of western Asia. 
[20 MAY 1983, E. A. Christensen, M. W. Lefor, R. Piacentini 
842 (CONN)] 


ORCHIDACEAE 


Malaxis bayardii Fernald 

Recent annotations of specimens of Malaxis unifolia Michx. 
from the New England Botanical Club Herbarium (NEBC) 
by Paul M. Catling of Agriculture Canada have shown Malaxis 
bayardii to have occurred in Connecticut. Annotated records 
come from Enfield, Hartford County and Bolton and Somers, 
Tolland County. The differences between these two species 
are slight and casily overlooked (Catling 1991). Other her- 
barium holdings of M. unifolia should be checked. M. unifolia 
is listed as an Endangered Species in Connecticut (Depart- 
ment of Environmentat Protection 1993). M. bayardii should 
be added to the list as a Species of Special Concern until it 


1995] Mehrhoff-Connecticut 21 


can be ascertained whether or not it is extant in the state. 
Distribution: Massachusetts and New York south to Vir- 
ginia and North Carolina (Catting 1991). 
[1896, Arthur S. Pease s. n. (NEBC)] 


URTICACEAE 


Pilea fontana ( Lunell) Rydberg 

This taxon has probably been overlooked in the northeast 
for many years. Seymour (1969) does not include this spe- 
cies. I collected specimens of it from the edge of a wooded 
swamp in Farmington, Hartford County in September, 1992. 
This prompted a close scrutiny of material of Pilea pumila 
(L.) Gray from Connecticut at the G. Safford Torrey Her- 
barium (CONN). It appears thatin 1991] Nels E. Barrett, while 
working on freshwater tidal vegetation along the Connecti- 
cut River, had collected specimens of Pilea fontana (as P. 
pumila) from East Haddam, Middlesex County. 

Distribution: Prince Edward Island to North Dakota and 

Nebraska south to Indiana and Virginia. 
[9 SEP 1991, Nels E. Barrett 00501 (CONN)] 


CHENOPODIACEAE 


Bassia hirsuta (L.) Aschers. 

Individuals of this taxon were reported from a disturbed 
salt marsh in Mystic, Stonington, New London County in 
1979 by William R. Linke, Jr. A small population was grow- 
ing at the edge of the salt marsh with Phragmites australis 
Cav.) Trin. 

Distribution: Massachusetts to Virginia. Native of Europe. 
[18 SEP 1980, L. J. Mehrhoff 3333, with W. R. Linke 
(CONN)] 


a2 Rhodora [ Vol. 97 


Kochia sieversiana (Pall.) C. C. A. Meg. 

Connecticut’s only record for this species was collected as 
Bassia hyssosifolia (Pallas) Kuntze from sea wrack along the 
Mystic River in Stonington, New London County, by Gor- 
don C. Tucker in 1981. A specimen of this collection in the 
New England Botanical Club Herbarium (NEBC) was anno- 
tated to Kochia sieversiana by the Curator, Ray Angelo, in 
1989. 

Distribution: Apparently not previously reported from the 

Northeast. Native of Siberia (Shishkin, ed. 1970). 

[17 AUG 1981, G. C. Tucker /668 (NEBC)| 


AMARANTHACEAE 


Amaranthus pumilus Raf. Seabeach Amaranth 
An 1893 specimen of this species, collected in New Lon- 
don, New London County, has recently been located in the 
New York State Museum. The label has no additional local- 
ity information. This species is listed as a Threatened Spe- 
cles by the U.S. Fish & Wildlife Service (USFWS 1993) and 
as a Connecticut Species of Special Concern (DEP 1993). 
Distribution: Massachusetts to North Carolina. 
[11 JUN 1893, sine collector (NYS); the specimen originally 
from Union college] 


CARYOPHYLLACEAE 


Sagina japonica (Sw. ) Ohwi 

This species was first reported from a stairwell in New Lon- 
don, New London County in 1988 by Virginia Magee.Col- 
lections were determined to be Sagina japonica by Garrett 
E. Crow (Mitchell and Tucker.1991). It may be overlooked 
(cf. Gleason and Cronquist 1991, Mitchell and Tucker 1991) 


1995] Mehrhoff-Connecticut 23 


and should be sought elsewhere. 
Distribution: Massachusetts, Connecticut, and New York. 
Native of Japan and China. 

[S JUL 1988, Virginia L. Magee 88-33 (NCBS)| 


PAPAVERACEAE 


Glaucium flavum Crantz 
Glaucium flavum is occasionally adventive in southern New 
England. A specimen in the G. Safford Torrey Herbarium 
(CONN) was collected in Groton, New London County in 
1933. Glaucium flavum appears to persist on sandy shores of 
nearby eastern Long Island although there is some confusion 
as to whether it is an annual, biennial, or perennial. It is sur 
prising that more seeds from the New York populations do 
not become established along the eastern Connecticut shore- 
line: 
Distribution: Massachusetts to Virginia, and occasionally 
inland to Michigan. Native of Europe. 
[6 JUL 1933, K. P Jansson s. n. ( CONN) ]. 


FUMARIACEAE 


Corydalis bulbosa Pers. 

This early spring ephemeral was first reported in Connecti- 
cut from a wooded bank near a road and along an abandoned 
rail road right-of-way 1n Salisbury, Litchfield County in 1982 
by H. Lincoln Foster and me. Photographs, but no specimens, 
were taken at that time (29APR 1982). It appeared to be well 
established. Although the source of the original plants 1s not 
known, we suspected that they may have come from a com- 
post heap a dozen meters away. This seems likely given that 
the seeds are readily dispersed by ants (pers. obs.) and this 


24 Rhodora [ Vol. 97 


taxon has proved to be troublesome by virtue of its invasive 
nature in some gardens (W. E. Brumback pers. comm.). How- 
ever, no check of the compost pile was made as the property 
was heavily posted. The site was revisited in 1992 and the 
plants appeared to be persisting but the numbers of observed 
individuals seemed to have declined. The long-term persis- 
tence of this taxon at this station 1s questionable. 
Distribution: Garden escape. Native to central Europe 
(Bailey 1949). 
[11 MAY 1992, L. J. Mehrhoff /546/ (CONN)| 


BRASSICAEAE 


Bunias orientalis L. 

This species was first observed in Ridgefield, Fairfield 
County in 1989 by D. Norris. . A second population was 
noticed in Goshen, Litchfield County in 1994. This popula- 
tion is uncomfortably close to the entrance to an exclusive 
country club and may have been planted and escaped. It is 
unclear whether or not Bunias orientalis will become es- 
tablished in Connecticut. There are collections from 
Rockland County, New York at the New York Botanical 
Garden. 

Distribution: Occasional in the eastern United States. Na- 

tive of Europe. 

[5 JUL 1989, David Nornis s. n. (CONN)] 


Teesdalia nudicaulis (.) R. Br. 

leesdalia nudicaulis was first reported in Connecticut from 
a roadside in New London County in 1985 by Robert J. Craig 
and me. It appears to be well established at this site but ap- 
parently not spreading far beyond the intersection where it 
was first observed. 


1995] Mehrhofl-Connecticut 25 
Distribution: Occasional in the United States. Native of Eu- 
rope. 

[9 MAY 1985, L, J. Mehrhoff 7/255 (CONN) | 

ROSACEAE 


Agrimonia microcarpa Wallr. Low Agrimony 

Genevieve J. Kline, in working on her treatment of 
Agrimonia for the Flora of North America, annotated two 
specimens in the New England Botanical Club Herbarium 
(NEBC) as this species. These had been collected in East 
Granby, Hartford County and Stamford, Fairfield County. 
Both had been previously identified as Agrimonia striata 
Michx. Other herbaria should be searched for this more south- 
ern taxon, 

Distribution: New Jersey and Pennsylvania to Florida 

and Texas. 
[25 JUN 1916, Perley Spaulding s. n. (NEBC)| 


FABACEAE 


Lespedeza cuneata (Dumont) G. Don 

The first Connecticut report of this species came from the 
side of southbound Interstate I-95 in Groton, New London 
County in 1978 by William R. Linke, Jr. It has since been 
collected along interstate highways in Hartford, Middlesex, 
Windham, and Tolland Counties. Perhaps it was introduced 
as a component of a seed mixture used in hydroseeding road- 
side embankments. It appears to be well established and 
spreading at most sites. Volunteers should be watched for away 
from highway roadsides. 

Distribution: Southeastern United States north to Long Is- 

land, New York and Connecticut. Native of eastern Asia. 


26 Rhodora [ Vol.97 


[25 SEP 1978, W. R. Linke, with L. J. Mehrhoff s. n. 
(CONN) | 


Lespedeza striata (Thunb.) Hook. & Arn. Japanese Clover 
Lespedeza striata was first reported from a roadside in 
Groton, New London County in 1991. The source of the origi- 
nal plants at this site is unclear but it may have been intro- 
duced to stabilize roadsides. This species appears to be well 
established and spreading at this station. 
Distribution: Gulf states north to Kansas, Indiana, and Con- 
necticut. Native of eastern Asia. 
[3 OCT 1990, L. J. Mehrhoff 14075 (CONN) | 


Pueraria lobata (Willd.) Ohwi Kudzu-vine 
The first Connecticut specimens of this species were taken 
in Fairfield, Fairfield County in 1928 by Edwin H. Eames. 
The labels read “Fence-row bordering field of Phleum 
pratense.” It appears that he again visited the Fairfield to col- 
lect Kudzu in 1947, this time in the company of J. J. Neale. 
Neale collected another specimen (this time in bud) from 
Fairfield on which he wrote, “Another colony in 1928 by E. 
H. Eames.” A 19 AUG 1947 specimen from Fairfield col- 
lected by J. J. Neale at the Herbarium of the Connecticut 
Botanical Society (NCBS) reads,"On walls and spreading for 
yards into field. Benson Road”. It is not clear if these records 
represent one Fairfield population or two. Lauren Brown col- 
lected this species in 1978 in New Haven, New Haven 
County Pueraria lobata was well established at this site. At 
that time, lianas reached the top of a four story building, and 
persisted for a number of years (L. Brown pers. comm.). It 
appears to have become well established at this site in spite 
of attempts to eradicate it. Kudzu is also established on near- 
by Fishers Island, Suffolk County, New York. It should be 


1995] Mehrhofl-Connecticut 27 


watched for anywhere along the Connecticut coast. It may 

not become established far from Long Island Sound. 
Distribution: Southeastern states north to New York and 
Connecticut. Native of Japan. 

[30 JUN 1928, E. H. Eames s. n. (CONN)| 


Strophostyles leiosperma (T. & G. ) Piper 

This species was reported from an abandoned gravel pit in 
Milford, New Haven County in 1990 by William Moorhead. 
Strophostyles leiosperma may have been introduced where it 
occurs, to stabilize gravel banks, but appears to be established 
and spreading (W. Moorhead pers. comm.). This species has 
recently been reported from Cape May County, New Jersey 
by David Snyder (1990). He questions its nativeness at the 
New Jersey station. 

Distribution: Ohio to Wisconsin and North Dakota, south 

to Florida and Texas. 
[20 SEP 1990, W. Moorhead 90-07-0179 (CONN)| 


GERANIACEAE 


Geranium nepalense Sweet var. thunbergii (Sieb. & Zucc.) 
Kudo 

The first report of this taxon appears to be from a roadside 
in New Milford, Litchfield County in 1981. That same year, 
the late H. Lincoln Foster had this species as a weed in his 
gardens at his home “Millstream” in Falls Village, 
Canaan,Litchfield County. The source of his material was 
unknown to him but it probably arrived with nursery mate- 
rial. This is an aggressive weed which easily spreads by seeds. 
Tolland County specimens are volunteers from seeds unin- 
tentionally introduced into Willington from Litchfield County 
by way of Coventry, Tolland County. This taxon has become 


28 Rhodora [\Ol. 97 


a persistent pest, even in mowed lawns. It should be assidu- 

ously sought out and removed before it becomes established. 
Distribution: Massachusetts. Native of eastern Asia. 

[27 AUG 1981, L. J. Mehrhoff 5252 (CONN)] 


EUPHORBIACEAE 


Croton glandulosus L. var. septentrionalis Mucll. Arg. 

William R. Linke, Jr. first reported a single individual of 
this taxon from cinder ballast at the side of a railroad in Mys- 
tic, Stonington, New London County in 1978. This individual 
was destroyed by construction equipment working on a nearby 
bridge. Seeds were produced, but subsequent visits to the site 
failed to reveal new plants. I recently identified a specimen 
of Croton glandulosus var. septentrionalis from North Ha- 
ven, New Haven County in 1993. collected by John W. 
Souther. 

Distribution: Tropical America north to Virginia, Indiana, 

Idaho, and Nebraska, Occasionally adventive farther north. 
[10 OCT 1978, L. J. Mehrhoff 2387 (CONN) | 


RHAMNACEAE 


Rhamuus citrifolia (Weston) Hess & Stearn 

Individuals of this taxon were first reported by Joe D. Pratt 
(1980) as Rhamnus davurica Pallas in 1976 from an over- 
grown old field in West Hartford, Hartford County. Label in- 
formation states that numerous individuals were observed with 
many producing fruit. The species was extant at the same site 
as recently as 1986. The commonly used epithet davurica 
was replaced with the epithet citrifolia in 1979. 

Distribution: Occasional within the Northeast. Native of 

northeastern Asia. 


1995] Mehrhoff-Connecticut 29 
6 AUG 1978, H. E. Ahles 86054 (MASS)| 
HALORAGACEAE 


Myriophyllum asiaticum (Vell.) Verdc. Parrotfeather 

This species was originally collected on 12 SEP 1946 as 
Proserpinaca palustris L. in shallow water of West Lake, 
Guilford, New Haven County by Edwin H. Eames and Will- 
iam I. Starr. It had been filed accordingly in the G. Safford 
Torrey Herbarium (CONN) until recently when it was acci- 
dently noticed as a mis-identification by Donald H. Les. He 
and I recently visited West Lake, searching in vain for 
Myriophyllum aquaticum. It was not present in collections I 
made from the lake in 1981. It appears that this species, na- 
tive to south America, may not have persisted in the colder 
climates of southern New England. Other herbaria should be 
checked for collections of this taxon. It may be filed under a 
synonym, Myriophyllum brasiliense Camb. 

Distribution: Southern United States north to New York, 

West Virginia and Missouri. Native of South America. 
[14 SEP 1946, E. H. Eames and W. | Starr /2,/95 (CONN)] 


GENTIANACEAE 


Centaurium pulchellum (Sw.) Druce Centaury 
This European species was reported from a grassy roadside, 
well removed from houses, in Canaan, Litchfield County in 
1984 by Robert Moeller. It was collected from another road- 
side in Canaan in 1992. 
Distribution: Locally introduced in the Northeast. Native 
of Europe. 
[29 JUL 1984, L. J. Mehrhoff /0407 (CONN)| 


30 Rhodora [\Ol. 97 
CUSCUTACEAE 


Cuscuta indecora Choisy 

A specimen of Cuscuta indecora, originally labeled Cuscuta 
coryli Engelm., was recently determined by Tania Beliz from 
the University of California at Berkeley [UC]. It had been 
collected by L. B. Bradley from a dooryard in Wilton, Fairfield 
County. 

Distribution: Illinois to California, south to Flonda and 

South America. 
[17 AUG 1940, L. B. Bradley s. n. (NEBC)] 


LAMIACEAE 


Elsholtzia ciliata (Thunb.) Hylander 

Elsholtzia ciliata was first collected in Connecticut from an 
overgrown roadside near the Connecticut River in Cromwell, 
Middlesex County in 1990 by Claudia Polsky and me. It has 
recently been collected along the side of a highway in Can- 
ton, Hartford County. Tolland County records are of unwanted 
volunteers from seeds of plants originally brought to Con- 
necticut from Morris County, New Jersey. 

Distribution: Quebec, New York, and New Jersey to Wis- 

consin. Native of Asia. 
25 SEP 1990, L. J. Mehrholf /4007, with C. Polsky (CONN) | 


SCROPHULARIACEAE 


Linaria dalmatica L William R. Linke, Jr., Lois Tefft, and 
Gordon Tucker first reported Linaria dalmatica from a road- 
side in Ledyard, New London County in 1984 (Tucker 1987). 
It has also been collected from a disturbed site in Torrington, 
Litchfield County 


1995] Mehrhoff-Connecticut a 


in 199]. 
Distribution: Occasional in northeastern United States. Nat- 
ive of eastern Mediterranean region. 

[29 MAY 1984, G. C. Tucker 24/5A (NCBS)] 


Veronica beccabunga L. Brooklime 
This European species was first reported from Great Falls 
in the Amesville section of Salisbury, Litchfield County in 
1980. It did not appear to persist at this site but may well 
have become established below the falls where similar habi- 
tat is extensive. A second population was discovered in 1994 
in a brook on the east side of Washining Lake (East Twin 
Lake) in Salisbury. 
Distribution: Sparingly established from Quebec to Michi- 
gan and south to New Jersey and West Virginia. Native of 
Eurasia. 
[21 OCT 1980, L. J. Mehrhoff 34/7, with Sarah Fried 
(CONN)]| 


ASTERACEAE 


Aster xblakei (Porter) House [Aster nemoralis Ait. x A. 
acuminatus Michx. | 
The first Connecticut record for Aster xblakei comes from 

plants growing along a mesic woodland fire road in Pachaug 
State Forest, Voluntown, New London County in 1982. These 
were collected by William R. Linke, Richard Blodgett, 
Edmund Smith, and Gordon Tucker (Tucker 1987). It occurs 
in nearby Rhode Island. 

Distribution: Newfoundland, Quebec, and Nova Scotia 

south to New Jersey. 
[16 AUG 1982, G. C. Tucker 1/789 (NCBS)] 


32 Rhodora [ Vol. 97 


Eupatorium album L. 

This southern taxon was first reported in Connecticut from 
thickets and rock outcrops in Groton, New London County 
in 1981 by William R. Linke, Jr. (Tucker 1987). He also lo- 
cated a second population approximately 1 km away from 
the first in a similar habitat. Both populations appear to be 
native and well established and were doing well in 1990. 

Distribution: Connecticut to Florida and west to Missis 
sippi, southern Appalachian Mountains. 
[14 AUG 1981, L. J. Mehrhoff 4987 (CONN)| 


Eupatorium hyssopifolium L. var. laciniatum A.Gray 

The only Connecticut collection of Fupatorium 
hyssopifolium var. laciniatum came from the side of Inter- 
state I-84 in Tolland, Tolland County in 1992. The typical 
variety of kupatorium hyssopifolium occurs at many sites in 
the four Connecticut counties adjacent to Long Island Sound. 
This variety, more typically found to the south of Connect- 
cut according to Gleason and Cronquist (1991), appears never 
to have been reported from New England. It is noticeably 
larger than the typical variety. FE. hyssostfolium var. laciniatum 
has recently been collected in nearby New York (O. 
Blanchard pers. comm.) The proximity of plants to the high- 
way suggests this taxon should be considered as a ruderal 
species in Connecticut. It may have been accidentally intro- 
duced by recent highway maintenance work in its vicinity. It 
should be watched for along interstates in Southern New 
England. 

Distribution: Southern New York to Georgia, northern 

Florida, and Louisiana, occasionally inland to southern 

Ohio, Kentucky, and Tennessee. 
[6 SEP 1992, L. J. Mehrhoff 7/6202 (CONN)| 


1995] Mehrhoff-Connecticut 35 


Euthamia tenuifolia (Pursh) Nutt. var. microcephala Nutt. 
Coastal Plain Flat-topped Goldenrod 
The presence of Euthamia tenuifolia var. microcephala was 
brought to my attention in 1981 when Marie Pickhardt, a 
well known Connecticut plant collector, gave me a speci- 
men she called Solidago microcephala that had been collected 
in Killingworth, Middlesex County that year (Mehrhoff 
1982a). A check of the G. Safford Torrey Herbarium (CONN) 
revealed an earlier collection of S. tenuifolia from Groton, 
New London County, that had been annotated to S. 
miciocephaldy the late Harry Ahles, and left unmentioned. 
Sieren (1981) sinks this species into Euthamia tenuifolia, giv- 
ing it no varietal status and effectively removing it from the 
flora of the state. Because of the similarity with Luthamia 
tenuifolia var. tenuifolia, E. tenuifolia var. microcephala may 
have been overlooked tn our flora. 
Distribution: Louisiana and Florida north to Maryland and 
Connecticut 


[7 SEP 1933, K. P. Jansson s. n. (CONN)| 


ACKNOWLEDGEMENTS 


I wish to thank curators and staff of CONN, GH, MASS, 
NCBS, NEBC, NY, NYS, US, and YU for their time, assis- 
tance, and access to their collections. I also wish to thank 
Joseph M. Beitel (deceased), David E. Boufford, Arthur 
Cronquist (deceased), Donald R. Farrar, Alice Tryon, Rolla 
M. Tryon, Ivan A. Valdespino, Florence S. Wagner, and War- 
ren H. Wagner for their time in verifying some of my deter- 
minations. Skip Blanchard, Lauren Brown, William E. 
Brumback, Wendy Dreyer, H. Lincoln Foster (deceased), C. 
Barre Hellquist, William R. Linke (deceased), Kenneth J. 


34 Rhodora [Vol. 97 


have shared their knowledge of certain taxa. Gordon C. Tucker 
of the New York State Museum has shared valuable thoughts 
on some of the included taxa which occur in southeastern 
Connecticut. Ray Angelo, Curator of the New England Bo- 
tanical Club Herbarium, has called my attention to some re- 
cent annotations in that herbarium, corroborated identifica- 
tions, and read an earlier draft of this manuscript. Karen 
Searcy, Curator of the University of Massachusetts Herbarium 
has supplied collection data from specimens collected by the 
late Harry Ahles. Donald H. Les, Director of the G. Safford 
Torrey Herbarium, provided many valuable comments on 
drafts of this paper. Editors David S.Conant, Gordon P. 
DeWolf, Jr, and Lisa A. Standley made helpful comments 
on earlier drafts. Robert E. Dubos, Collection Manager of 
the G. Safford Torrey Herbarium, assisted in the preparation 
of specimens and the dispersal of duplicates. Diane Tyler, of 
the Connecticut Geological and Natural History Survey, as- 
sisted with the word processing. A generous grant from the 
Connecticut Arboretum to the Connecticut State Museum of 
Natural History helped prepare some of these specimens for 
the Torrey Herbarium. 


bt PERATURE CITED 


Bailey, L. H. 1949, Manual of Cultivated Plants:The Macmillan Com- 
pany, New York. 

Buck, W. R. 1978. A new species of Selaginella in the Selaginella apoda 
complex. Canad. J. Bot. 55:366-371. 

Catling, P M. 1991 Systematics of Malaxis bayardii and M. unifolia. 
Lindleyana 6:3-23. 

Christensen, E.A. and M. W. Lefor. 1985. Ornithogalum nutans L. 
(Liliaceae) in Connecticut. Rhodora 87:437-438. 

Crow, G. E. and C. B. Hellquist. 1982.Aquatic Vascular Plants of New 
England: Part 4. Juncaginaceae,Scheuchzeriaceae, Butomaceae, 
Hydrocharitaceae. New HampshireA gricultural Experiment Sta- 


1995 MehrhdEConnecticutt 35 


tion, Durham, New Hampshire. 
Department of Environmental Protection. 1993 Connecticut’ Endan- 
gered, Threatened, and Special Concern Species. Hartford. 15 


pages. 

Dowhan, J. J. 1979. Preliminary Checklist of theVascular Flora of Con 
necticut. State Geological and Natural History Survey Report 
of Investigations No. 8. Hartford. 

Fairbrothers, D. E. and J. R. Gray 1972. Microstegium vimineum (Trin,) 
A, Camus (Gramineae) in the United States. Bull. ‘brrey Bot. 
Club 99: 97-100. 

Farrar, D. R. 1992. Trichomanes intricatum. The independent 7richomanes 
gametophyte in the Eastern United States Amer. Fern J. 82(2) 

Farrar, D. R., J. C. Parks, and B.W. McAlpin. 1983. The fern genera 
Vittaria and Trichomanes in the northeastern United States. 
Rhodora 85: 83-92. 

Fernald, M. L. 1950. Gray’s Manual of Botany, Eighth Revised Edition. 
Van Nostrand Reinhold Company New York. 

Flora of North America Editorial Committee, eds. 1993. Flora of North 
America North of Mexico, Volume 2. Oxford University Press, 
New York. 

Gleason H. A. and A. Cronquist. 1963. Manual of Vascular Plants of 
Northeastern United States andAdjacent Canada, First Edition. 
Willard Grant Press, Boston. 

1991. Manual of Vascular Plants of 
Northeastern United States andAdjacent Canada, Second Edi- 
tion. New York Botanical Garden. 

Holmgren, P K., N. H. Holmgren and L. C. Barnett. 1990. Index 
Herbariorum, Part 1:The Herbaria of the World. Ed. 8. Regnum 
Veg. 120. 

Hunt, D. M. and R. E. Zaremba. 1992.The northeastward spread of 
Microstegium vimineum (Poaceae) into New York and adjacent 
states. Rhodora 94: 167-170 

Mehrhoff, L. J. 1982a.A short-lived addition to the flora of Connecticut. 
Rhodora 84:305-307 

1982b. Psilocarya scirpoides Torr., an addition to the 
Connecticut Flora. Rhodora 84:307-309. 

Mitchell, R. S. and G. C.Tucker. 1991 Sagina japonica (Sw.) Ohw 


36 Rhodora [ Vol.97 


(Caryophyllaceae), an Hie cane adventive in the northeast 
ern United States, Rhodora 93:192-194 

Pratt, J. D. 1980. A naturalized aia of Rhamnus citrifolia in 
ee Rhodora 82:523-524. 

Seymour, P| . The Flora of New England. The Charles E. 
Tuttle sae Publishers, Rutland, Vermont. 

Shishkin, B. K., ed. 1970. Flora of the U.S. S. R. Volume VI,Centro- 
spermae. Israel Program for ScientificTranslations, 


Jeruselem. 
Sieren, D. J. 1981.The taxonomy of the genus Futhamia. Rhodora 
83:551-579. 


Snyder, D. B. 1990. Strophostyles leiospermain New Jersey: adventive 
or nativeo Bartonia 56:34-37 
Tryon, R. M. 1940. An Osmunda hybrid. Amer. Fern J. 30:65-66 + 2 


plates. 

Tucker, G. C. 1987. Additions to the flora of Connecticut. Rhodora 

89:217-219. 

United States Fish & Wildlife Service. 1993. Endangered and Threat- 
ened Wildlife and Plants. 50 CFR 17.11 & 17.12. 


THE CONNECTICUT GEOLOGICAL AND NATURAL HISTORY 
SURVEY 

AND 

THE G. SAFFORD TORREY HERBARIUM 

DEPARTMENT OF ECOLOGY AND EVOLUTIONARY BIOLOGY 
BOX U-4 

UNIVERSITY OF CONNECTICUT 

STORRS, CONNECTICUT 06269-3042 


1995] 


Mehrhofl-Connecticut 


a7 


Table |. Taxa added to the Preliminary Checklist of the 
Vascular Flora of Connecticut (Dowhan 1979). E=State 
Endangered, Fl=Federally Threatened, SC=State Species 


of Special Concern 


TAXON 


Lycopodiella alopecuroides 


Lycopodiella xcopelandii 
Selaginella eclipes 
Osmunda xruggil 
Trichomenes intricatum 


Pellaea glabella subsp. glabella 
Echinodorus tenellus var. parvulus 


Egeria densa 

Aira praecox 
Microstegium vimineum 
Panicum amarulum 
Panicum scabriusculum 
Vulpia myurus 

Carex backii 

Carex emoryi 

Cyperus echinatus 
Rhyncospora scirpoides 


[syn. Psilocarya scirpoides] 


Ornithogalum nutans 
Malaxis bayardii 

Pilea fontana 

Bassia hirsuta 

Kochia sieversiana 
Amaranthus pumilus 
Sagina japonica 
Glaucium flavum 
Corydalis bulbosa 
Bunias orientalis 
Teesdalia nudicaulis 
Agrimonia microcarpa 
Lespedeza cuneata 
Lespedeza striata 
Pueraria lobata 
Strophostyles leiosperma 


First County 
Record 


New Haven 
New Have 
Litchfield 
Fairfield 


Hartford 


New London 


New London 
New London 
New london 
New Londonn 
Litchfield 
Fairfield 
New London 
Hartford 
New London 
New London 
Fairfield 
New Haven 


Year of 


Conserv- 
ation 
Status 


1 Oa ab 


38 


TAXON 


Geranium nepalense var. thunbergil 
Croton glandulosus var. septentrionalis 


Rhamnus citrifolia 


Myriophyllum aquaticum 
Centaurium pulchellum 


Veronica beccabunga 


Aster xblakel 
Eupatorium album 


Eupatorium hyssopifolium 


Euthamia tenuifolia var. microcephala 


var. laciniatum 


Rhodora 


First County 
Record 


Litchfield 
New London 
Hartford 
New Haven 
Litchfield 
Fairfield 


Litchfield 
New London 
New London 


Tolland 
New London 


Year of 
First 
Record 


[ Vol. 97 


Conserv- 
ation 
Status 


RHODORA, Vol. 97, No. 889, pp. 39-92, 1995 


THE VEGETATION OF PEQUAWKET BOG, 
OSSIPEE, NEW HAMPSHIRE 


Linpa L.. faney AND GARRErTr E. Crow 
ABSTRACT 


Peatlands, while ecologically interesting and abundant in the northeast, have 
gone largely unstudied in New Hampshire. This baseline study focuses ona veg- 
elation analysis of the vascular flora of Pequawket Bog, Ossipee, New Hamp- 
shire. A total flora of 109 species, including /riophorum angustifolium, an en- 
dangered plant species for the state of New Hampshire, was documented for the 
bog. Using stratified random sampling, 287 plots from 10 transects were sampled 
for percent cover of vascular plant species. Five vegetation cover types and nine 
subtypes were determined using the computer classification program ‘TWINSPAN. 
The five cover types include: a Nymphaea odorata cover type, a Carex lasiocarpa 
cover type, a Chamaedaphne calyculata-Woodwardia virginica cover type, a 
Chamaedaphne calyculata -Vacctnium oxycoccos-Eriophor um virginicum cover 
type, and an Acer rubrum- Vaccinium cor ymbosum-Lyonia ligustrina cover type. 


Key Words: bog, peatland, plant community, plant classification, New Hamp- 


shire, vegetation. 
INTRODUCTION 


Peatlands occur widely, developing chiefly in cool to cold 
regions, and often vary considerably in physical character, 
nutrient status, and vegetation and floristic composition. Gore 
(1983b) provides a detailed account of peatlands on a world- 
wide level. As a result of the wide geographic distribution of 
peatlands, there is a plethora of terms and classification 
schemes associated with these ecosystems. Moore and 
Bellamy (1974), Stanek (1977), Worley and Sullivan (1980), 
Worley (1981), and Gore (1983a) discuss and clarify this of- 
ten overwhelming aspect of peatland studies. In general, 


a7 


Figure 1. Aerial photograph of Pequawket Bog, looking north (October, 1991) 


OP 


vIOPOY 


L6 “[OA| 


1995 Fahey and Crow - Pequawket Bog 41 


peatlands of the northern hemisphere are often referred to as 
“mires” in Europe (Moore and Bellamy, 1974), “muskegs” in 
Canada (Stanek, 1977) or simply “bogs” in the United States. 

Peatlands may be defined as “three-dimensional portions 
of the earth’s landscape that are wetlands; have organic soils; 
include the full depth of the organic materials, regardless of 
origin; include all waters within or on top of the organic ma- 
terials; and include all organisms living within or atop the 
organic materials and water” (Worley and Sullivan 1980, pp. 
13-14). In this definition the term “wetland” is used as de- 
fined by Cowardin ef al. (1979) for the U.S. Fish and Wild- 
life Service, and “organic soils” as defined by the U.S. Soil 
Conservation Service (Soil Survey Staff, 1975). 

Peatlands are frequently classified based on hydrology and 
nutrient status (Sjors, 1959; Jeglum e7 al., 1974; Moore and 
Bellamy, 1974; Worley and Sullivan, 1980 ), and two major 
classes are typically distinguished: ombrotrophic and 
minerotrophic peatlands. Ombrotrophic peatlands receive all 
their nutrients and water through precipitation only. Conse- 
quently, these peatlands are very poor in nutrients and are 
very acidic. Minerotrophic peatlands, conversely, receive 
nutrients and water from both surface runoff and groundwa- 
ter as well as from precipitation. Because the water entering 
has run over and percolated through mineral soils, these 
peatlands tend to be relatively nutrient rich and typically less 
acidic. The amounts and quality of these nutrients vary greatly 
from one geographic location to another due to a number of 
factors, including bedrock, soil characteristics, and topogra- 
phy (Gorham, 1956, 1957). 

A fundamental distinction is also made between bogs and 
fens when classification is based on hydrology and nutrients. 
“True” bogs are strictly ombrotrophic, and the vegetation is 
typically dominated by Sphagnum spp., low ericaceous shrubs, 
and scattered conifers. Fens, on the other hand, are 


42 Rhodora [ Vol. 97 


minerotrophic peatlands and the vegetation is largely domi- 
nated by sedges and grasses, with Sphagnum spp. function- 
ing in a subordinate role. 

While such classification systems appear useful, it is often 
difficult to classify certain peatlands satisfactorily into one 
class or the other. This ts especially true for kettle-hole or 
basin peatlands, where nutrient status and hydrology change 
through successional development (Crum, 1988). At the base 
of the upland there tends to be a strong minerotrophic influ- 
ence, yet further away from the edge the peat mat may be 
essentially ombrotrophic, as the inflowing of waters through 
the peal 1s impeded. 

Damman and French (1987) prefer not to utilize the term 
bog in the strict sense, based on the fact that weakly 
minerotrophic fens are floristically similar to the vegetation 
of ombrotrophic peatlands, and have few floristic similarities 
to nutrient-rich fens. In a community profile of peat bogs of 
the glaciated northeastern United States their term “bog” is 
used to refer to “peatlands with a well-developed moss car- 
pet dominated by Sphagnum” (Damman and French, 1987, 
p. 1). Other recent ecologists who do not restrict their use of 
the term “bog” solely to ombrotrophic peatlands include Golet 
and Larsen (1974), Jeglum e7 al. ( 1974), Rawinski (1984), 
and Sperduto (1994). 

Damman and French (1987) treat peatlands as landforms 
and base the divisions on the nature of the water that controls 
their development. In this scheme four major types are rec- 
ognized: 1) limnogenous, 2) topogenous, 3) ombrogenous, 
and 4) soligenous. Limnogenous peatlands occur along lake- 
margins and slow flowing streams. Topogenous peatlands 
develop in sites where there is an accumulation of water, and 
are maintained by a permanent ground water table or seep- 
age; these include kettle hole bogs. Ombrogenous peatlands 
include raised bogs which form independently from ground 


1995 Fahey and Crow - Pequawket Bog 43 


water or seepage, and are restricted to humid, cold temperate 
climates. In eastern North America these occur from north- 
ern Maine to Nova Scotia and western Newfoundland. 
Soligenous peatlands are dependent on minerotrophic seep- 
age water, and are found in regions with high precipitation 
and less evapotranspiration. These include sloped fens. 

Sperduto (1994) classifies peatlands in New Hampshire into 
three general categories based on nutrient and pH levels: 1) 
bogs, 2) acidic fens, and 3) calcareous fens, as well as three 
broad climatic influences: coastal/southern, boreal/transi- 
tional, and alpine/subalpine. Rawinski (1984) classifies as 
bogs, ombrotrophic to weakly minerotrophic peatlands, and 
distinguishes between level and raised bogs. Fens, on the other 
hand, are treated in this classification as peatland communi- 
ties influenced by seepage waters with alkalinity ranging from 
low to high. A distinction is also made between calcareous 
and acidic fens, and whether each is sloping or level. 

Although itis often the goal to classify a particular peatland 
into one specific category, in reality this is not an easy task. It 
is usually better to treat these ecosystems as peatland com- 
plexes comprised of different zones, each with differing nu- 
trient levels and plant associations. 

The importance of water chemistry in relation to vegeta- 
tion in peatlands has been demonstrated in numerous studies 
(Bay, 1967; Heinselman, 1970; Jeglum, 1971; Moore and 
Bellamy, 1974; Vitt and Slack, 1975; Schwintzer, 1978, 1981; 
Wells, 1981; Schwintzer and Tomberlin, 1982; Vitt and 
Bayley, 1984). Patterns of plant associations in the succes- 
sional development of kettle-hole peatlands have been shown 
to correlate strongly to variations in pH (Crow, 1969; Vitt 
and Slack, 1975; Vitt and Bayley, 1984; Dunlop, 1987). 

An important source of acidity in peatlands is the activity 
of Sphagnum. It has been well documented that these mosses 
have the ability to actively acidify their environment through 


44 Rhodora [ Vol. 97 


the process of cation exchange (Clymo, 1963, 1964; Craigie 
and Maass, 1966), the site of exchange being an unestenfied 
polyuronic acid (Clymo, 1963). Other important sources of 
acidity include sulfuric and humic acids (Mitsch and 
Gosselink, 1986). However, “acid rain” apparently does not 
function as an acidifying agent in peatlands. This acidic in- 
put is apparently negated by an increase in alkalinity by sul- 
fate reduction and nitrate uptake (Hemond, 1980). 

Numerous descriptive studies have focused on the floristics 
and phytosociology of peatlands. Some include: Heinselman 
(1963, 1970) in northern Minnesota; Janssen (1967) in north- 
western Minnesota; Conway (1949) in central Minnesota; 
Hansen (1933) in the driftless area of Wisconsin; Rhodes 
(1933) in the drift-covered area of Wisconsin; Gates (1942) 
in northern lower Michigan; Transeau (1905, 1906), Brewer 
(1966), Crow (1969), and Keogh and Pippen (1981) in south- 
ern Michigan; Dansereau and Segadas- Vianna (1952) in east- 
ern Canada; Sjors (1959, 1963) in the Hudson Bay Lowlands, 
and Attawapiskat River in northern Ontario, respectively; 
Wells (1981) in Newfoundland; Damman and French (1987) 
in the glaciated regions of eastern United States; Dunlop 
(1987) in southern New Hampshire; and Montgomery and 
Fairbrothers (1963) in New Jersey. 

While peatlands are a common and interesting component 
of the New England landscape, relatively few detailed stud- 
ies on the major plant cover types of these ecosystems have 
been conducted. This is especially true in New Hampshire. 
Johnson (1985) provides a broad overview of peatlands in 
New England. In Massachusetts, Motzkin and Patterson 
(1991) investigated vegetation patterns of a moat bog in rela- 
tion to basin morphometry, Hemond (1980) investigated the 
biogeochemistry of Thoreau’s Bog in Concord, while Swan 
and Gill (1970) studied the role of Chamaedaphne calyculata 
in the succession of an artificially made kettle hole bog. In 


1995 Fahey and Crow - Pequawket Bog 45 


Maine, Worley and Sullivan (1980) and Worley (1981) have 
focused on classification, while in a northern Vermont peat- 
land complex, Osheyack and Worley (1981) investigated pri- 
mary production. In New Hampshire, Barrett (1966) studied 
succession in a southern New Hampshire peatland 1n relation 
to physical and edaphic factors. Other peatlands of the state 
have been used in palynological studies which have focused 
on post-glacial vegetation during the Holocene (Kraus and 
Kent, 1944; and Davis ef al., 1980). But only one other de- 
tailed study of New Hampshire peatland vegetation has been 
published (Dunlop, 1987). 

The purpose of this study is to describe and map the veg- 
etation cover types of Pequawket Bog and to add to our over- 
all base-line information on the vegetation and floristic com- 
position of peatlands in New Hampshire. 


STUDY AREA 


Pequawket Bog is located in the Town of Ossipee, near the 
Effingham town line, in Carroll County, New Hampshire. 
While it has no official name, Pequawket Bog is referred to 
herein as such because of the historical influence of the 
Pequawket Indians in the area (Cook, 1989), and its proxim- 
ity to Pequawket Trail road. Pequawket Trail was originally 
used by the Pequawket Indians to travel between the regions 
of Ossipee and Conway (Ruth Loring, pers. comm.). This 
peatland complex is situated between Long Sands Road and 
Pequawket Trail Road, off of State Route 25 at 43° 47'N. 
Lat., 71° 06' W. Long. It lies just southeast of Ossipee Lake at 
an elevation of approximately 123.7 meters above sea level 
(406 ft.). The size of the peatland is approximately 9.9 hect- 
ares (24.4 acres), including a 2.8 hectare (7 acre) pond. 
Hellquist ( 1971) conducted a survey of the aquatic plants of 
Ossipee Lake and its associate bays, but did not include 


46 Rhodora | Vol. 97 


Pequawket Bog. 

Although it is referred to as a bog, it is more appropriate to 
classify Pequawkct Bog as a peatland complex, as it clearly 
has two very different zones with a peat substrate, each with 
a unique floristic character. On the west of the pond is a 
sedgy meadow, or level fen (Figure 1). The south, east and 
southwest sides of the pond are characterized as having a 
more typical bog flora with Sphagnum spp. and low erica- 
ceous shrubs dominating. 

The upland soils in the general area surrounding the two 
peatlands are of glacial outwash origin, and are part of the 
Hinckley-Windsor-Deerfield association characterized as 
“nearly level to very steep, excessively drained and moder- 
ately well drained gravelly and sandy soils; on terraces, ka- 
mes, and eskers” (Diers and Vicira, 1977, p. 4). The Green- 
wood-Chocorua-Namburg association is more typical of the 
lower lying areas and is defined as “nearly level, very poorly 
drained organic soils and somewhat poorly drained and poorly 
drained sandy soils; along broad drainageways and depres- 
sions” (Diers and Vieira, 1977, p. 5). 

The upland area along the eastern border of Pequawket Bog 
fits the concept of excessively drained soils of the Hinckley 
series, that are characterized as gravelly loamy sands. The 
esker to the southwest of the pond in Pequawket Bog is of the 
Windsor series and is a loamy sand. 

The soil within the bog is classified as a Greenwood mucky 
peat, which is characterized as organic soil composed of partly 
to well decayed herbaceous and woody material. This organic 
layer may range from 50 inches to over 10 feet, with an un- 
derlying layer of sand, gravel, silt, or loam. According to the 
Carroll County soil survey (Diers and Vieira, 1977), the area 
encompassing Pequawket Bog is classed as a fresh water 
marsh, and broadly defined as a land type covered by shal- 
low water most of the time, found around edges of lakes and 


1995 Fahey and Crow - Pequawket Bog 47 


ponds and also in depressions that are ponded much of the 
year. This assessment has been incorrectly applied to the en- 
tire peatland complex. While 1t may be descriptive of the sedgy 
meadow to the west of the pond, it has been inappropriately 
applied to the areas south and east of the pond. These are 
clearly Greenwood mucky peat. 

Based on interpretation of the Ecologic Map and Structure 
Sections of the Ossipee Lake Quadrangle, New Hampshire” 
(Wilson, 1969) the bedrock which underlies the glaciofluvial 
deposits around the bog appears to be Conway Granite of the 
New Hampshire Plutonic Series, which dates back to the 
Middle Devonian. This rock is generally characterized as: a 
“medium grained, light-colored, equigranular and contains 
plagioclase with a composition of about An,,, microcline, 
quartz, biotite, and Muscovite” (Wilson, 1969, p. 26). 

The climate in this area can be characterized as having rela- 
tively mild summers, cold winters, and abundant rainfall 
(Diers and Vieira, 1977). The climate 1s considered continen- 
tal, mainly influenced by westerly winds, but because of the 
relative proximity to the Atlantic ocean there is increased pre- 
cipitation in the fall and winter. Low spots, especially peaty 
soils, are more frost prone on cold clear nights. The climato- 
logical data presented were recorded at the weather station in 
Conway, approximately 15 miles north of the bog, and are 
representative of areas within the county at lower elevations. 
Temperature fluctuations in Carroll County are prone to {re- 
quent changes, as the county lies in a region where weather 
systems tend to alternately bring in warmer and colder air. 
Weather records (Diers and Vieira, 1977) showed an average 
annual temperature at Conway (clevation 145 m) of 6.3 °C 
(43.3 °F), with average annual maximum and minimum tem- 
peratures of 13.4 °C (56.1 °F) and -0.8 °C (30.6 °F) respec- 
tively. The average annual extreme maximum 1s 35.6 °C (96 
°F) and extreme minimum is -32.2 °C (-26 °F). July is the 


48 Rhodora [Vol. 97 


warmest month averaging 19.7 °C (67.5 °F), with a mean daily 
maximum of 27.4 °C (81.4 °F) anda mean daily minimum of 
12 °C (53.6 °F). The average extreme maximum recorded for 
July was 33.9 °C (93 °F) and extreme minimum was 3.9 °C 
(39 °P). January is the coldest month averaging -8.3 °C (17 
°F), with a mean daily maximum and minimum of -1.3 °C 
(29.7 °F) and -15.4°C (4.3 °F), respecti\ ely. The average exX- 
treme maximum Is 8.3 °C (47 °F) and minimum ts -30 °C 
(23 °F): 

Precipitation is, for the most part, evenly distributed 
throughout the year, however there tends to be a slight in- 
crease in the fall and winter. The area receives an average of 
116.9 cm (46.01 in.) annually. The monthly averages range 
from 14.3 cm (5.64 in.) in November to 7.6 cm (3.00 in.) in 
January. 

Average annual snowfall is 287.8 cm (113.3 in.). The 
amount varies greatly from year to year, but 1s seldom less 
than 143.9 cm (56.65 in.) or more than 431.7 cm (169.95 in.). 
The first snowfall usually occurs in October, and the ground 
is normally covered with snow from early December to some- 
time in April. Even in the mildest winters the ground ts rarely 
bare in January, February, and March. 

For most of Carroll County, the frost-free season ranges 
from 105 to 130 days, and may extend to 140 days 1n more 
favorable spots. However, in low areas the frost-free period 
tends to be shorter. 


METHODS 


A total inventory of the vascular plant species found in 
Pequawket bog was initiated in mid-April of 1991 and con- 
tinued through the growing season to mid-October, 1991. 
Voucher specimens are deposited in the Hogdon Herbarium 
(NHA), University of New Hampshire. The account of the 


1995 Fahey and Crow - Pequawket Bog 49 


flora is published elsewhere (Fahey and Crow, in press). 
Quantitative sampling of the Pequawket Bog vegetation be- 
gan in mid-July and continued to late August, 1991. Strati- 
fied random sampling (Mueller-Dombois and Ellenberg, 
1974) was employed. Nine transects were placed at relatively 
equal intervals and positioned as to best sample all vegeta- 
tion cover types (Figure 2). An additional transect was sampled 
on the west side of the pond, in an area which was particu- 
larly disturbed by beaver, and represented an ecotone between 
two visually distinctive cover types. This area consists of many 
deep channels and appeared initially to be somewhat unique 
in its plant associations. Each transect was divided into 10 
meter segments. In areas of low growing vegetation (below 
1.5m) two 1 x 1 meter quadrats were located using random 
numbers. For areas with tall shrubs or trees over 1.5 meters, 
one 4 x 4 meter quadrat was sampled per 10 meter segment 
of transect. In the larger plots, two 1x1 meter quadrats in op- 
posite corners were sampled for the lower vegetation. The 
lower left and upper right corners were used consistently. The 
data from the two Ixl meter quadrats were averaged and 
combined with the taller shrub data for total plot informa- 
tion. This method 1s similar to Dunlop’s (1987) sampling re- 
gime of a southern New Hampshire peatland which was found 
quite effective. A total of 287 quadrats was sampled for abso- 
lute percent cover and species composition. Cover is defined 
as the crown or shoot area of a species projected over the 
ground surface. This area 1s expressed as a percent of the 
reference area (Mueller-Dombois and Ellenberg, 1974). 
The vegetation data were analyzed using TWINSPAN (Two- 
way Indicator Species Analysis), a FORTRAN program (Hill, 
1979). TWINSPAN is a polythetic divisive method of classi- 
fication, which results in an ordered two-way table based on 
a series of ordinations. This program has been employed by 
other ecologists in the analysis of peatland vegetation (Slack 


WN 
i) 


Rhodora [ Vol. 97 


et al., 1980; Dunlop, 1987; Vitt ef al., 1989). Using all the 

samples, it begins with the “primary” ordination, made by 
reciprocal averaging. This ordination is essentially divided 
in half. It proceeds to the next ordination where species that 


Pequawket Trail Road 


upland 


Figure 2. Locations of vegetation sampling transects 


1995 Fahey and Crow - Pequawket Bog 51 


show a preference to one side or the other of the initial ordi- 
nation are identified and used as a basis for this “refined” 
ordination. These species are referred to as “differential” or 
“preferential” species, and are said to be showing a prefer- 
ence to certain ecological conditions. The final ordination, 
the “indicator” ordination, identifies those species that show 
a particularly high preference to either side of the dichotomy. 
These species are referred to as “indicator” species. The ulti- 
mate dichotomy 1s based on the refined ordination, while the 
indicator ordination is simply additional information given 
to the investigator to further characterize the communities or 
vegetation types. 

The pseudo-species cut levels of 1, 2, 3,4, 5 represent the 
Braun-Blanquet cover values 0, 5, 25, 50, 75, respecti vely, of 
the Braun-Blanquet system of phytosociology (Hill, 1979). 
All pseudo-species were available as indicator species, and 
all cut levels were weighted equally. 

In order to plot a profile of the shape of the basin and thick- 
ness of organic material, peat depths were measured using a 
probe of connecting steel rods. At Pequawket Bog, this was 
measured every 10 meters along transect | and transect 8 (Fig- 
ule 2), 

From late-August to mid-September, pH measurements 
were taken using a VWR Digital Mini pH Meter (model 55). 
At Pequawkel Bog sites within the peatland that appeared to 
represent distinct cover types or subtypes were sampled. Five 
measurements were taken at each site. A total of 25 sites was 
sampled with a total of 125 samples measured. 

Using the TWINSPAN classification as a basis, the aver- 
age species density (number of species/m’) for the subtypes 
and moat cover type were calculated using the data from the 
| m* quadrats The means were compared via a one factor 
ANOVA (Scheffe F-test) for significant differences at the 


wr 
Ne) 


Rhodora [ Vol. 97 


95% confidence level (p< 0.05). 


RESULTS AND DISCUSSION 


A flora of 109 species was documented for Pequawket Bog. 
The inventory included a new record of Eriophorum 
angustifolium, an endangered species for the state of New 
Hampshire (DRED, 1987), and now known from only two 
localities. A complete account of the floristic inventory is pre- 
sented elsewhere as it includes a comparison to the floristic 
composition of a well-known nearby peatland, Heath Pond 
Bog (Fahey and Crow, in press). 


VEGETATION 


The TWINSPAN classification of vegetation for Pequawket 
Bog resulted in a total of five major cover types and nine 
subtypes, each with a relatively distinct floristic composition 
and physiognomy (Figure 3). While some of the subtypes are 
not large enough to map, they are included in the analysis as 
they may reflect microhabitats found within the peatland. In 
general, Pequawket Bog is comprised of an aquatic cover type, 
a sedge meadow or fen cover type, two “typical” bog cover 
types dominated by ericaceous shrubs, and a tall shrub or 
moat cover type. 

Figure 3 summarizes the TWINSPAN classification of the 
287 vegetation samples into five major cover types and nine 
subtypes at six hierarchical levels; the number of samples 
clustered into each is indicated. At the first hierarchical level 
the aquatic, Nymphaea odorata cover type (CT I) is distin- 
guished from the total 287 samples. At the second level this 
cover type was divided further into two subtypes. From the 
remaining 267 samples, 83 define the Carex lasiocarpa cov- 


1995 Fahey and Crow - Pequawket Bog 53 


287 
267 
Nymphaea odorata 
CT I 
16 | 4 184 83 
ST Ia ST Ib Carex lasiocarpa 
CT I 
145 29 54 
cer -Vaccinium ST Ia ST IIb 
corymbosum-Lyonia_ ligustrina 
CTV 
Chamaedaphne calyculata- Chamaedaphne calyculata- 
Woodwardia virginica Eriophorum virginicum 
CT Ill CT IV 


14 30 —t 33 
| ST Ila ST IVa ST IVb 
5 33 


Figure 3. Summary of TWINSPAN analysis showing the separation of 
287 quadrat samples into cover types (CT) and subtypes (ST) of 
Pequawket Bog, and indicating the number of quadrats in each group 


er type (CT II) at the second level. This was also divided 
further into two subtypes. At the third level the remaining 
184 samples clustered into Chamaedaphne calyculata _domi- 
nated cover types and the tall shrub, Acer-Vaccinium 
corymbosum-Lyonia cover type (CT V). The Chamaedaphne 
calyculata dominated cover types are classified at the fourth 


54 Rhodora [Vol. 97 


level into the Chamaedaphne calyculata-Woodwardia 
virginica cover type (CT III) and the Chamaedaphne 
calyculata-Vaccinium oxycoccos-Eriophorum virginicum 
cover type (CT IV). These two cover types are then divided 
further into subtypes at the fifth and sixth levels (Figure 3). 

The naming of the cover types was based primarily on the 
combined score of mean percent cover and percent frequently 
of the most dominant species (see tables 1-5). Percent cover 
and frequency were also tabulated for each subtype (Fahey, 
1993), with subtype data presented here for selected species 
in the discussion. 

It should be noted as well that a submerged zone domi- 
nated by Potamogeton amplifolius was observed in the deeper 
waters of the pond, particularly toward the south-central end 
of the pond. This area was not sampled quantitatively. How- 
ever, 1tis worth mentioning as 1t may be an indication of the 
amount of accumulated sediments tn the basin, and the depth 
of the water. 

In the following discussion the cover types and subtypes 
determined by TWINSPAN for Pequawket Bog are described. 
Comparisons are also made with nearby Heath Pond Bog 
(Fahey, 1993) and other North American level peatlands de- 
scribed in the literature. Because this study focuses on vascu- 
lar plant species, the importance of Sphagnum was not in- 
cluded in discussion, although it is understood that the vari- 
ous species of Sphagnum play a critical role in peatland eco- 
systems. Also, it is realized that indicator species from one 
geographic region may differ considerably from those of other 
regions. 


Nymphaea odorata Cover Type (CT I) 


This cover type is dominated by submerged, floating-leaf, 
and emergent aquatic plants and occurs around the outer pond 


1995 Fahey and Crow - Pequawket Bog ao 


margin in varying degrees of width. It also occurs in a pooled 
area at the northern edge of the peatland (Figure 4). Nymphaea 
odorata, Utricularia purpurea, and Eleocharis robbinsii are 
the major constituents with the highest cover and frequency 
values (Table 1). The TWINSPAN program used these as in- 
dicator species of this cover type. Other species recognized 
by TWINSPAN as preferential to this cover type are 
Pontederia cordata, Utricularia intermedia, and Brasenia 
schreberi. This is also reflected in their relatively high cover 
or frequency values (Table 1). Other species with more 
patchy distributions in the pond include Potamogeton 
confervoides and Nuphar variegata. 

The Nymphaea odorata cover type of Pequawket Bog has 
some floristic similarities to the Nymphaea-Brasenia zone 
reported by Dunlop (1987) in Mud Pond Bog, a southern New 
Hampshire peatland. However, that zone at Mud Pond Bog 
also was apparently more depauperate than that found at 
Pequawket Bog, lacking many submerged and emergent spe- 
cles. 

Aquatic zones with floristic similarities have also been re- 
ported in Michigan peatlands. Crow (1969) described two 
aquatic associations around a southern Michigan bog. A 
Nuphar-Eleocharis zone dominated the perimeter of the 
pond, and a Decodon zone on one side. Also present were a 
number of floating and submerged aquatics, including Utri- 
cularia purpurea, which is quite dominant at Pequawket Bog 
as well. Decodon verticillata was found in Pequawket Bog, 
but to a very insignificant extent toward the southwest end of 
the pond. Similarly, a phytosociologic zone dominated by 
Nuphar variegata and other floating-leaved macrophytes was 
described by Dansereau and Segadas- Vianna ( 1952). 


56 Rhodora [Vol. 97 


Table 1. Mean percent cover and percent frequency of dominant and subdominant 
species in the Nymphaea ordorata cover type (CT I). 


Species Mean % Cover % Frequency 
Nymphaea odorata* 40 95.0 
Utricularia purpurea* 20 60.0 
Eleocharis robbinsii* 14 85.0 
Pontederia cordat 8 40.0 
Potamogeton confervoides 6 15.0 
Brasenia schreberi 2 30.0 
Utricularia intermedia 2 30.0 
Carex lasiocarpa 2 10.0 
Nuphar varie gata 1 10.0 
Sphagnum spp. 1 10.0 


* = indicator species for CTI 


Subtype Ia 


Within the Nymphaea odorata CT, two subtypes could 
be discerned. The first subtype (ST Ia) is found adjacent 
to the encroaching mat around the majority of the pond 
(Figure 4). Utricularia purpurea was used by TWINSPAN 
as the indicator species to distinguish this subtype as it 
shows a high frequency (75%) in the samples found around 
the pond as well as the second highest percent cover (25%). 
Additionally, Eleocharis robbinsii showed a high frequency 
of occurrence (81.3%) and the third highest percent cover 
( 15%). Nymphaea odorata, Brasenia schreberi, Eleocharis 
robbinsii, and Utricularia intermedia were also found to 


1295 Fahey and Crow - Pequawket Bog 57 


be quite frequent in this subtype. It is this mixture of sub- 
merged, emergent, and floating-leaved aquatic species 
which serve as the forerunner of the encroaching mat. 

Utricularia purpurea has been reported to be present pri- 
marily in bogs which possess a false bottom of muddy, 
organic sediments (Crum, 1988). 

Keough and Pippen (1981) report Eleocharis robbinsti 
present in a southwest Michigan bog, but in an area more 
characteristic of a moat. There are, however, few reports 
of the presence of Eleocharis robbinsii in other peatlands. 
This may be due to the fact that descriptions of aquatic 
communities are often omitted in studies of peatland veg- 
etation because they are not deemed unique to the peatland 
ecosystem, or, due to their location, are often difficult to 
sample. 


Subtype Ib 


The second subtype (ST Ib) 1s a smaller association found 
in a pooled area at the northern edge of the peatland along 
Pequawket Trail Road (Figure 4). This subtype appears 
relatively depauperate, with a total of four species. But 
only four sample plots fall into this division. Pontederia 
cordata, with a cover value within STIb of 4% and a 25% 
frequency, is the most characteristic species of this sub- 
type, and was used by TWINSPAN as an indicator spe- 
cies. Potamogeton confervoides was also conspicuous, with 
a cover value of 30% and a frequency of 50%. Eleocharis 
robbinsii, a plant with a 100% frequency value, seemingly 
has a low cover value within the subtype (7%) due to the 
nature of its slender, erect growth form. Nymphaea odorata 
is also a major component of this subtype. 


in 
ied) 


Rhodora [ Vol. 97 


Nymphaea odorata cTI 
sTla RY stm 

Carex lasiocarpa CT If 
ST Ita [24 sT mb 

Chamaedaphne-Woodwardia CT II 
E dst ma I st mp ES st mc 


Chamaedaphne-V. oxycoccos-Eriophorum CT IV 


Acer-Vaccinium corymbosum-Lyonia CT V 
N 
Figure 4. Vegetation map of Pequawket Bog showing five cover types(CT) 
and nine subtypes(ST). Scale bar equals 50 m 


Potamogeton confervoides has been reported as occurring 
in acidic waters of New England along the coastal plain, and 
is often associated with Eleocharis robbinsii (Hellquist and 
Crow, 1980). It has been regarded as rare and endangered for 
other New England states including Connecticut, Maine, and 
Vermont. However, it is relatively common in New Hamp- 
shire and Massachusetts (Hellquist and Crow, 1980) 


1995 Fahey and Crow - Pequawket Bog 59 
Carex lasiocarpa Cover Type (CT I) 


The Carex lasiocarpa cover type, occupying an extensive 
area in the peatland particularly to the northwest of the pond, 
forms an open, sedgy meadow, or fen. It is also found around 
the pond margin in certain areas, presumably serving as the 
pioneer association of the floating mat (Figure 4). Compared 
to other areas in the peatland this portion of the mat 1s notice- 
ably wetter, often with pools of standing water. The 83 sam - 
ples of the total 287 plots clustered into this cover type are 
based primarily on the presence of Carex lasiocarpa. How- 
ever, species such as Vaccinium macrocarpon, Peltandra 
virginica, Triadenum virginicum, Carex utriculata (=C. 
rostrata var. utriculata), Sagittaria latifolia, Pogonia 
ophioglossoidesand Myrica gale are classified by TWIN- 
SPAN as preferential to this group as well. Mean percent 
cover and percent frequency of the dominant species in this 
cover type are listed in Table 2. Carex lasiocarpa, dominat- 
ing this association with 38% cover on average, is a strongly 
rhizomatous, clonal species, as are the majority of the domi- 
nant species in this zone. Although Myrica gale and 
Chamaedaphne calyculata are quite dominant, it is largely 
the importance of the herbaceous species that give this cover 
type its character. Many of the herbaceous species show a 
relatively low cover value, but have a high frequency. 
While Carex lasiocarpa and many of the associated spe- 
cies of this cover type are present at Pequawket Bog, this 
vegetation type 1s noticeably absent at nearby Heath Pond 
Bog (Fahey, 1993). No zone there resembles a sedgy meadow. 
Instead, the pioneer cover type invading the open water is 
dominated by Chamaedaphne calyculata. The pH of the pond 
water and shape of the basin probably have an important role 
in this difference between these two bogs. Vitt and Slack 
(1975) found pH of pond water to be an important factor 


60 Rhodora [Vol. 96 


Table 2. Mean percent cover and percent frequency of dominant and sub- 
dominant species in the Carex lasiocarpa cover type (CT II). 


Species Mean % Cover % Frequency 
Sphagnum spp. 86 100.0 
Carex lasiocarpa* 39 92.8 
Myrica gale 15 78.3 
Chamaedaphne calyculata 15 71.1 
Vaccinium macrocarpon 13 55.4 
Peltandra virginica 7 39.8 
Sagittaria latifolia 7 39.8 
Carex utriculata 5 34.9 
Aster nemoralis 3 20.5 
Andromeda glaucophylla 3 15.7 
Nymphaea odorata 3 12.0 
Pogonia ophioglossoides 2 42.2 
Triadenum virginicum 2 28.9 
Rhyncospora alba 1 30.1 
Juncus pelocarpus 1 16.9 
Sarracenia purpurea 1 16.9 
Symplocarpus foetidus 1 16.9 
Utricularia intermedia 1 16.9 
Drosera intermedia 1 13.3 
Dulichium arundinaceum 1 13.3 
Acer rubrum 1 12.0 
Vaccinium oxy. 1 12.0 
Scheuchzeria palustris 1 9.6 
Carex stricta 1 8.4 
Cladium mariscoides 1 8.4 
Eriophorum tenellum <1 9.6 
Sparganium americanum <1 6.0 


*= indicator species for CT II 


in influencing the species composition of the mat encroach- 
ing on the open water in northern Michigan bogs. They found 
Carex lasiocarpa to be the primary species occupying mats 
encroaching on the open water of bog ponds where the water 
was relatively alkaline, usually with a pH greater than 7.0, 


1995 Fahey and Crow - Pequawket Bog 61 


(the Alkaline Lake Edge Zone). In lakes with an acidic pH, 
usually ranging from 5.0 - 7.0, they found Chamaedaphne 
calyculata, Andromeda glaucophylla, and Rhynchospora alba 
showing higher importance values at the mat edge (the Acid 
Lake Edge Zone). 

Crum (1988) reports that in Michigan, Carex lasiocarpa 
dominated cover types occur at the edge of lakes that possess 
false bottom sediments and that have relatively alkaline open 
water. The sediments in the pond accumulate over time and 
eventually lend themselves to colonization by aquatic mac- 
rophytes such as Nymphaea spp., Nuphar spp., and 
Potamogeton spp. Carex lasiocarpa rhizomes are able to 1n- 
vade these sediments particularly during drier years when the 
water level is lower. Vitt and Slack (1975) also make note of 
the presence of a false bottom in bogs supporting a C. 
lasiocarpa dominated pioneer zone. The accumulation of false 
bottom sediments is absent around the majority of Heath Pond. 

Carex lasiocarpa has been described by many others as an 
important member of various bog and fen associations 
throughout the northern United States and Canada, (Gates, 
1942; Conway, 1949; Crow, 1969; Heinselman, 1970; Vitt 
and Slack, 1975; Schwintzer, 1978; Vitt and Bayley, 1984). 
The presence/dominance of this sedge, as well as many of 
the other characteristic species of this cover type, have been 
well documented in other regions to be a reflection of more 
minerotrophic conditions (Heinselman, 1970; Jeglum, 
1971;Vitt and Slack, 1975; Schwintzer, 1978; Vitt and Bayley, 
1984; Crum, 1988). 


Subtype Ila 


Within the Carex lasiocarpa CT two cover subtypes could 
be recognized in the TWINSPAN analysis. The first of these, 
ST Ila, is found primarily as the association encroaching on 


62 Rhodora [Vol. 97 


the pond behind the aquatic cover type, and in areas that have 
been disturbed by beaver (Figure 4). Twenty-nine of the 83 
cover type samples were classified into this group based on 
the indicator species identified by TWINSPAN, and reflected 
by the high percent frequency of these species: Peltandra 
virginica (96.6%), Triadenum virginicum (79.3%), 
Rhynchospora alba (65.5%), Juncus pelocarpus (44.8%) and 
Utricularia intermedia (44.8%). The dominant shrubs in- 
clude Chamaedaphne calyculata (89.7% frequency) and 
Myrica gale (93.1% frequency). Several other species which 
have low cover values, but relatively high frequencies, and 
are Important to the characterization of this subtype, include 
Nymphaea odorata, Dulichium arundinaceum, Aster 
nemoralis, Cladium mariscoides, Drosera intermedia, Sar- 
racenia purpurea, Sagittaria latifolia, and Pogonia 
ophioglossoides . A relatively higher cover value (9%) for 
Nymphaea odorata in this cover subtype resulted from some 
plots falling on the edge of the pond or overlapping larger 
channels. 

Two isolated patches of Menyanthes trifoliata were also 
observed in this cover subtype at the extreme ends of the pond, 
although transects did not intersect these sites. This species 
has been reported in the literature as typically having a very 
narrow niche within the peatland ecosystem, and is usually 
restricted to the lake edge (Dansereau and Segadas- Vianna, 
1952) and/or to inflow/outflow channels (Vitt and Bayley, 
1984). In northwest Ontario, Vitt and Bayley (1984) describe 
a Sphagnum papillosum-Menyanthes trifoliata community 
type which shows this pattern, and has some floristic simi- 
larities to the Carex lasiocarpa cover type, with the domi- 
nance of Carex lasiocarpa and Myrica gale. {n northern Min- 
nesota, Heinselman (1970) lists it as an indicator of weakly 
minerotrophic waters with a pH range of 4.3-5.8. 

There are strong floristic similarities of this subtype to the 


1995 Fahey and Crow - Pequawket Bog 63 


Alkaline Lake Edge Zone described by Vitt and Slack (1975) 
for northern Michigan, and to the Carex lasiocarpa mats de- 
scribed by Conway (1949) in central Minnesota. Many of the 
indicator species of this subtype may be reflecting the com- 
bined effect of a more minerotrophic condition, with a some- 
what higher pH, and a higher water level (Jeglum, 1971). 
Vitt and Slack (1975) found the distribution of Rhynchospora 
alba to be apparently more influenced by moisture level and 
degree of shade, rather than by water chemistry. 


Subtype IIb 


The other subtype recognized by TWINSPAN in the Carex 
lasiocarpa CT is the most extensive of the two subtypes. It 
occupies a very large area to the northwest of the pond (Fig- 
ure 4). Fifty-four samples clustered into this subtype, with 
the TWINSPAN preferential species including Myrica gale 
(70.4% frequency), Chamaedaphne calyculata (61.1% fre- 
quency), Symplocarpus foetidus (25.9% frequency), and 
Pogonia ophioglossoides (48.1% frequency). Andromeda 
glaucophylla and Sarracenia purpurea were also shown to 
have a strong preference for this subtype as well. 

The floating mat on this side of the pond is weaker than 
portions of the mat in other areas of the peatland. In this 
meadow-like subtype, the dominant sedge Carex lasiocarpa 
has an average percent cover of 48% and 96.3% frequency, 
with its other major contributor to the mat, Vaccinium 
macrocarpon, with a 19% cover and 74.1% frequency. Other 
important species that add to the character of the subtype, 
whether through their cover values or frequencies include, 
Myrica gale, Chamaedaphne calyculata, Sagittaria latifolia, 
Carex utriculata, Pogonia ophioglossoides, and Symplocar- 
pus foetidus, Species that are found more sparsely through- 
out the mat include Aster nemoralis, Andromeda glaucophylla, 


64 Rhodora [Vol. 97 


Sarracenia purpurea, Vaccinium oxycoccos, and small sap- 
lings of Acer rubrum. Scheuchzeria palustris is frequent in 
localized areas nearing the pond edge. Eriophorum 
angustifolium, an endangered species in New Hampshire, was 
also sparse, but occurred in less wet areas in this cover sub- 
type. 

The Carex- Vaccinium macrocarpon zone described by Crow 
(1969) for a southern Michigan bog has many floristic simi- 
larities to this subtype. This zone is also dominated by Carex 
lasiocarpa and Vaccinium macrocarpon with other similar 
species including Chamaedaphne calyculata (=Cassandra 
calyculata), Andromeda glaucophylla, Sagittaria latifolia, and 
Sarracenia purpurea. 

A phytosociologic association dominated by Carex 
utriculata described by Dansereau and Segadas- Vianna (1952) 
is also similar to this subtype. They found Carex lasiocarpa 
and C. utriculata (=C. rostrata ssp. utriculata) to be a com- 
mon association, usually in places where peat comes in con- 
tact with sand, and where there 1s a fluctuation in water level 
over the growing season. Some fluctuation of the water level 
was seen at Pequawket Bog, however no quantitative mea- 
surements were taken. 

The Carex lasiocarpa CT shows up in the aerial photo- 
graph (Figure 3) as a light area; darker regions on its margin 
are areas Where Chamaedaphne calyculata is more abundant. 
Itis assumed that over time this will encroach upon the sedge 
meadow. This, however, will largely depend on the hydro- 
logical regime. Crow (1969) reports similar areas where 
Chamaedaphne calyculata seems to be encroaching on the 
Carex lasiocarpa-Vaccinium macrocarpon zone. Succession 
of this nature has been reported by Gates ( 1942) in peatlands 
of Michigan, where floating mats dominated by Carex 
lasiocarpa eventually become grounded by the accumulation 
of debris peat and are rapidly colonized by Chamaedaphne 


Gs 


1995 Fahey and Crow - Pequawket Bog 6 


calyculata. This succession was not seen, however, in fens 
where the mat remains free floating. 


Chamaedaphne calyculata-Woodwardia virginica Cover 
Type (CT II) 


This cover type, as a whole, is quite prominent in the south 
and southwest portions of the mat and along the west and 
south sides of the pond (Figure 4). It may be characterized as 
an extremely dense community with the vegetation usually 
not more than | meter in height. It is largely dominated by 
low shrubs and Woodwardia virginica . The other herbaceous 
species found here are present either as scattered individuals 
throughout this zone or are restricted to patchy locations that 
are less densely vegetated. Chamaedaphne calyculata is the 
dominant species with a mean percent cover of 36% and [re- 
quency of 100%. Codominant Woodwardia virginiana was 
recorded as having a mean percent cover of 27% and a 88.2% 
frequency (Table 3). TWINSPAN identified Woodwardia 
virginica, Carex oligosperma, Rhododendron canadense, and 
Kalmia polifolia as indicator species, separating this cover 
type from the Chamaedaphne calyculata-Vaccinium 
oxycoccos-Eriophorum virginicum cover type (CT IV). 
Myrica gale, Smilacina trifolia, Acer rubrum, Aronia 
melanocarpa and Alnus incana ssp. rugosa also showed 
more of a preference to this cover type over CT IV. Carex 
trisperma 1s quite dominant in this cover type as well, but it 
did not show a preference between the two Chamaedaphne 
calyculata dominated cover types. Other woody species im- 
portant to this cover type include Kalmia angustifolia, An- 
dromeda glaucophylla, Vaccinium corymbosum, Larix 
laricina and Picea mariana. Other herbaceous species with 
a minor role in this zone include Symplocarpus foetidus, 
Eriophorum virginicum, Sarracenia purpurea, Calla palustris 


66 Rhodora [Vol. 97 


Table 3. Mean percent cover and percent frequency of dominant and subdominant eae 
of the daphne calyculata-Woodwardia virginica cover type (CT 


Species Mean % Cover % Frequency 
Sphagnum spp. 99 100.0 
Chamaedaphne calyculata 38 100.0 
Woodwardia virginica* 26 85.4 
Rhododendron canadense* 11 52.4 
Carex trisperma i 45.1 
Carex oligosperma* 9 48.8 
Myrica gale 8 52.4 
Smilacina trifolia 5 20.7 
Kalmia polifolia* 4 50.0 
Alnus incana ssp. rugosa 3 31.7 
Acer rubrum 2 35.4 
Aronia melanocarpa 2 26.8 
Symplocarpus foetidus 2 20.7 
Eriophorum virginicum ] 31.7 
Kalmia angustifolia 1 17.1 
Andromeda glaucophylla 1 13.4 
Sarracenia purpurea 1 12.2 
Calla palustris 1 9.8 
Larix laricina 1 7.3 
Vaccinium corymbosum 1 73 
Picea mariana 1 6.1 
Scheuchzeria palustris 1 2.4 


*= indicators for CT II] 


and Scheuchzeria palustris. 

A cover type somewhat similar is found at Heath Pond Bog 
(Fahey, 1993) where an extensive mat is by far dominated by 
ericaceous shrubs, Eriophorum vaginatum ssp. spissum, and 
scattered Larix laricina and Picea mariana. A similar cover 
type also occurs at Cedar Bog, in Kingston, New Hampshire 
(Crow, pers. obs.). However, there are few reports of 
Woodwardia virginica as a dominant constituent such as it 1s 
at Pequawket Bog and Heath Pond Bog. It has been been 


1995 Fahey and Crow - Pequawket Bog 67 


described by Damman and French (1987) in peatlands of the 
glaciated northeastern United States as only an occasional 
constituent of the Sphagnum rubellum-Chamaedaphne 
calyculata) community, which is typically associated with 
oligotrophic quaking mats bordering lakes. They also describe 
it as an occasional constituent of the Cinnamon Fern-High- 
bush Blueberry Thicket. Keough and Pippen ( 1981) describe 
the vegetation of two adjacent bogs in southwest Michigan 
with strong floristic similarities to Pequawket Bog, includ- 
ing the dominance of Woodwardia in certain zones. They 
found it particularly dominant in the understory of a tall shrub 
zone dominated by Aronia melanocarpa, Nemopanthus 
mucronata, Vaccinium corymbosum, Larix laricina, and 
Rhamnus frangula. Crow (1969) reports it to be a minor con- 
stituent of the Larix laricina and Acer rubrum zones another 
southern Michigan bog. 

Dunlop reported Woodwardia as occasional at Mud Pond 
Bog, found “on the fringe and in wooded zones particularly 
on the east and northeast sides of the bog” (Dunlop, 1983, p. 
23), in what is described as the Acer-Nemopanthus Commu- 
nity Type (Dunlop, 1987). The Carex trisperma-Kalmia 
angustifolia subtype of the Chamaedaphne dominated cover 
type of Mud Pond Bog (Dunlop, 1987) shows some floristic 
similarities to Pequawket and Heath Pond Bog as well. Spa- 
tially they are compe as well, as they are found between 
the quaking mat and the moat. 


oak 


Subtype IIIa 


Since the Chamaedaphne calyculata-Woodwardia virginica 
CT covers a rather large area of the peatland, it is not surpris- 
ing that the TWINSPAN analysis reveals three subtypes. The 
first subtype, ST Ila, occupies a relatively small area of the 
peatland toward the outer edge of the southwest mat (Figure 


68 Rhodora [ Vol. 97 


4). The map depicts this subtype as a circular zone surrounded 
by ST IIIc. This subtype has a very shrubby character and the 
area is substantially wetter in certain spots. The TWINSPAN 
program indicates that Smilacina trifolia (85.7% frequency) 
and Symplocarpus foetidus (71.4% frequency) show a high 
preference for this zone and were thus used by TWINSPAN 
as indicator species; Sarracenia purpurea, Alnus incana ssp. 
rugosa, and Calla palustris also show a preference to this 
subtype. 

As in all of the subtypes of CT III, Chamaedaphne 
calyculatais dominant. Here it has a mean cover of 33% and 
frequency of 100% (Table 6). Woodwardia virginica has less 
cover on average ( 15%) in this zone compared to the other 
two subtypes. Other dominant woody species include Myrica 
gale, Kalmia polifolia, and Alnus incana ssp. rugosa. Picea 
mariana, Rhododendron canadense, and Acer rubrum are also 
present, but to a lesser degree. Other dominant herbaceous 
species include Carex oligosperma and C. trisperma. While 
Calla palustris and Sarracenia purpurea, do not appear domi- 
nant quantitatively, TWINSPAN indicates they are 
preferencial to the subtype. Eriophorum virginicum is also 
found occasionally in this zone. 

There 1s no extensive zone at Heath Pond Bog that is di- 
rectly comparable to this subtype, however, towards the moat 
running along Rte. 25 there is an area where Smilacina trifolia, 
Woodwardia virginica, Chamaedaphne calyculata, and other 
ericaceous shrubs are quite prominent. 

While this zone is probably too narrowly defined to be di- 
rectly compared to other described communities, the indica- 
tor species of this subtype may be reflecting certain ecologi- 
cal conditions or combinations of these conditions. Typically 
areas near the moat are influenced more by the mineral rich 
telluric water than the zones further inward on the mat. It is 
well documented in other geographical regions that Alnus 


1995 Fahey and Crow - Pequawket Bog 69 


incana ssp. rugosa 1s common in the more mineral rich areas 
of peatlands or in fens (Conway, 1949; Dansereau and 
Segadas- Vianna, 1952; Heinselman, 1970; Schwintzer, 1981, 
Vitt and Bayley, 1984; Crum, 1988). Calla palustris is often 
a common species associated with A/nus in wet moats (Crum, 
1988). Symplocarpus foetidus is typical of more nutrient rich, 
wet areas as well. However, in Michigan, Smilacina trifolia 
is usually described as a subordinate in associations which 
are largely forested and shady, or at least in the taller shrub 
zones toward the moat (Gates, 1942; Crum, 1988). Worley 
(1981) uses the presence of Symilacina trifolia as one of the 
indicators to designate the moat in Maine peatlands. 


Subtype IIIb 


The second subtype (ST IIIb) is found usually in the sun- 
niest and somewhat wetter areas of the Chamaedaphne 
calyculata-Woodwardia virginica cover type (Figure 4). In 
the delineation of this subtype from the others, TWINSPAN 
identified Carex oligosperma and Chamaedaphne calyculata 
as indicator species. 

Chamaedaphne calyculata and Woodwardia virginica domi- 
nate with a mean cover of 51% and 26% respectively and 
both have 100% frequency. Carex oligosperma 1s quite prom- 
inent as Well, with a mean cover value of 17% and frequency 
of 71.4%. Other dominant woody species that characterize 
this subtype include Rhododendron canadense and Myrica 
gale, and toalesser extent Alnus incana ssp. rugosa, Kalmia 
polifolia, Vaccinium corymbosum, Acer rubrum, Aronia 
melanocarpa, and Kalmia angustifolia. Other herbaceous 
species include Carex trisperma, Eriophorum virginicum, 
Symplocarpus foetidus, and Scheuchzeria palustris. 

Scheuchzeria palustris, interestingly, occurs in certain lo- 
cations on the mat that are adjacent to the pond, particularly 


70 Rhodora [ Vol. 97 


on the east side, but not in the southwestern portion of the 
bog mat. The species was, however, found on the opposite 
side of the pond in the Carex lasiocarpa CT, especially nearer 
the pond margin. Another species, Eriophorum angustifolium, 
occasionally found scattered within this subtype, also was 
found in certain locations of the Carex lasiocarpa CT. 

Heath Pond Bog (Fahey, 1993) has scattered patches 
throughout the extensive mat west of the pond which are simi- 
lar to this subtype, characterized as being less densely occu- 
pied by shrubs, witha relatively open and slightly wetter area 
of Heath Pond Bog. Eriophorum angustifolium is is found 
scattered throughout this mat. However, Myrica gale and 
Alnus incana ssp. rugosa are not present, nor is Scheuchzeria 
palustris. 

Dunlop (1987) does not report any cover type at Mud Pond 
Bog similar to this. However, Vitt and Bayley (1984) describe 
an association similar to this in northwestern Ontario. It is 
dominated by Chamaedaphne calyculata, Carex oligosperma, 
and Scheuchzeria palustris, but Woodwardia_ virginica is 
lacking. That peatland was characterized with an average pH 
of 4.37 (ranging from 4.2-4.8). They found the distribution 
of Scheuchzeria palustris in the peatland apparently to be un- 
related to pH. However, Carex oligosperma appeared to be 
an indicator of oligotrophic habitats with low pH values. 

Schwintzer (1981) found Carex oligosperma and 
Chamaedaphne calyculata to be the most characteristic spe- 
cies of the “field layer” in northem Michigan bogs which were 
highly acidic (pH 3.8-4.3) and low 1n Ca** and Mg** ( 1.2-3.7 
mg/L and 0.3-0.6 mg/L respectively). Vitt and Slack (1975) 
found Carex oligosperma occupying the Acid Lake Edge 
Zone, the Open Mat Zone and the Closed Mat Zone, how- 
ever it was most dominant in the zones further back from the 
lake edge. 


1995 Fahey and Crow - Pequawket Bog ya 
Subtype IIc 


The third subtype of the Chamaedaphne calyculata- 
Woodwardia virginica CT is found to be quite dominant in 
the southern and southwestern portions of the mat, and on 
the mat along the east side of the pond (Figure 4). Itis_ char- 
acterized as being a rather dense, woody zone, and much of it 
difficult to walk through. TWINSPAN identified Kalmia 
polifolia (69.7% frequency), Aronia melanocarpa (51.5% fre- 
quency), Acer rubrum (54.5% frequency) and Carex 
trisperma (72.7% frequency) as species highly preferential 
to this subtype. Vaccinium oxycoccos, Andromeda 
glaucophylla and Kalmia angustifolia showed a preference 
here as well. One noticeable character to this subtype is the 
dominance of Woodwardia_ virginica, with a mean cover of 
32% and a frequency of 84.8%. Compared to the other sub- 
types within the Chamaedaphne calyculata-Woodwardia 
virginica CT, this is the only zone where Woodwardia 
virginica has a higher mean percent cover than Chamaedaphne 
calyculata. Other dominant woody species that characterize 
this zone include Chamaedaphne calyculata, Rhododendron 
canadense, Myrica gale, Kalmia polifolia, Acer rubrum, and 
Aronia melanocarpa. The dominant herbaceous species in- 
clude Carex trisperma and Eriophorum virginicum. Species 
that play a less significant role, though still important to the 
character of this subtype, include Andromeda glaucophylla, 
Alnus incana ssp. rugosa, Kalmia angustifolia, Vaccinium 
oxycoccos, Carex oligosperma, Smilacina trifolia, and Sar- 
racenia purpurea. 

A similar association 1s found at Heath Pond Bog (Fahey, 
1993) on the portion of the mat encircling the pond, between 
the quaking mat and the moat, and in areas closer to the moat 
along the bordering highway. 

The distribution of Carex trisperma, an indicator species, 


de Rhodora [Vol. 97 


seemed to be particularly related to areas of very firm peat, 
and a more or less closed canopy. This observation 1s sup- 
ported by the findings of Vitt and Slack ( 1975) who reported 
that the distribution of this sedge in northern Michigan bogs 
seemed to be related to shade. The sedge was found to be 
more abundant in areas where Picea and Larix became more 
important. Crum (1988) also reports its abundance enhanced 
by the shade created by the two conifers, on drier mounds. 
Vitt and Bayley (1984) describe a Smilacina trifolia-Ledum 
groenlandicum-Carex trisperma community type in Ontario 
which was characteristically shady as well. 


Chamaedaphne calyculata-Vaccinium oxycoccos- 
Eriophorum virginicum Cover Type (CT IV) 


The Chamaedaphne calyculata-Vaccinium Oxycoccos- 
Eriophorum virginicum cover type (CT IV) occurs in regions 
of the peat mat which are apparently less consolidated, as 
they have a noticeably more quaking feel to them. Figure 4 
shows this zone on the northeast side of the pond between 
the aquatic cover type and the Chamaedaphne calyculata- 
Woodwardia virginica cover type (CT III). On the southwest 
mat CT IV is found roughly in the center of that region, en- 
circled by CT II, and also as the pioneer association encroach- 
ing on the pond at the southern and northeast sides. While 
Chamaedaphne calyculata 1s a dominant of this cover type, 
it has a much smaller stature than in the Chamaedaphne 
calyculata-Woodwardia virginica cover type (CT III). It usu- 
ally does not attain heights much over 30 cm. Compared to 
CT III this zone is strikingly more open and less shrubby. 
Because of its openness and low stature of the woody spe- 
cies, Crum (1988, p. 61) refers to similar communities such 
as this as the “Sphagnum Lawn Community”. 

The dominant species that characterize this vegetation zone 


1995 Fahey and Crow - Pequawket Bog 73 


include Chamaedaphne calyculata, Vaccinium oxycoccos, and 
Eriophorum virginicum (Table 4). It was the high degree of 
preference to this zone of the latter two species which 
TWINSPAN used to classify this cover type from CT III. 
Andromeda glaucophylla showed a preference to this zone 
as well. Carex trisperma is an integral member of this cover 
type also, but it does not necessarily show a preference to 
this zone. It is also quite dominant in the Chamaedaphne 
calyculata-Woodwardia virginica cover type, and is found 
most often in the transition zone between these two cover 
types. Other subdominant woody species of this cover type 
are Myrica gale, Kalmia polifolia, Larix laricina, and Picea 
mariana. The subordinate herbaceous species include Drosera 
rotundifolia, Rhynchospora alba, Woodwardia virginica, 
Decodon verticillata, Eriophorum vaginatum ssp. spissum, 
and Peltandra virginica. For mean percent cover and percent 
frequency of these species see Table 4. 

Cover types of this nature are very common in peatlands, 
especially in kettle-hole bogs. In Michigan, Vitt and Slack 
(1975) describe an Open Mat Zone, adjacent to the edge zone 
of alkaline lakes, with strong floristic similarities. Along with 
Chamaedaphne calyculata, Vaccinium oxycoccos was the 
most dominant vascular plant. Its distribution within the bogs 
was apparently related to non-shaded habitats which were 
low in pH and cation concentration. These locations also had 
little change in microtopography, and were fairly wet. Other 
similarly dominant species included Kalmia polifolia, An- 
dromeda glaucophylla, Rhynchospora alba, Drosera 
rotundifolia and Eriophorum virginicum. Although Vitt and 
Slack (1975) associated this community type with alkalinity 
of the lake water, the waters of Pequawket Bog’s pond are 
not alkaline. Based on the species present, however, the pond 
water 1s probably fairly minerotrophic, and the pH is rela- 
tively high (5.69) compared to the other cover types. 


74 Rhodora [ Vol. 97 


Table 4. Mean percent cover and percent frequency of dominant species in the 
Chamaedaphne calyculata-Vaccinium oxycoccos-Eriophorum virginicum 
cover type (CT IV). 


Species Mean % Cover % Frequency 
Sphagnum spp. 100 100.0 
Chamaedaphne calyculata 43 100.0 
Vaccinium oxycoccos* 10 88.9 
Carex trisperma 9 38.1 
Eriophorum virginicum* a 88.9 
Andromeda glaucophylla 5 34.9 
Myrica gale 3 ie 
Drosera rotundifolia 2 19.0 
Picea mariana 2 11.1 
Woodwardia virginica 2 11.1 
Decodonverticillata Z 93 
Larix laricina 1 14.3 
Rhynchospora alba ] 14.3 
Kalmia polifolia | 12.7 
Eriophorum vaginatum fi 

ssp. spissum 
Peltandra virginica 1 7.9 


* = indicator species for CT IV 


Subtype [Va 


Two subtypes were discernable within the Chamaedaphne 
calyculata- Vaccinium oxycoccos-Eriophorum virginicum CT. 
The first subtype (ST IVa) occupies the center of the back 
south-southwest mat, and also along the mat on the northeast 
side of the pond (Figure 4). On the southwest mat, although 
it is some distance away from the open water of the pond, it 
still has a noticeable quaking feel to it. The basin profile data 
of this mat (Figure 5) indicate that this area is the center of a 
relatively recent closed basin. TWINSPAN found Carex 
irisperma, Eriophorum virginicum and Chamaedaphne 


1995 Fahey and Crow - Pequawket Bog fi, 


calyculata highly preferential to this subtype over ST IVb. 
Although Chamaedaphne calyculata is dominant in both sub- 
types it averaged a much higher cover value ( 55% cover, 
100% frequency) in ST [Vain contrast to ST [Vb (32% cover, 
100% frequency). Other species that show a preference to 
this subtype include Kalmia polifolia and Larix laricina. 

Chamaedaphne calyculata is the clearly the dominant 
woody species. Other dominant woody species include 
Vaccinium oxycoccos, Larix laricina, Kalmia polifolia, and 
Picea mariana are also found here, but toward the transition 
into the Chamaedaphne calyculata-Woodwardia virginica 
cover type (CT III). The dominant herbaceous species include 
Eriophorum virginicum and Carex trisperma, the latter also 
found more frequently toward the transition into CT III. 

Platanthera blepharigloitis, rare within the peatland, is 
found sparsely in this cover type, usually nearing the transi- 
tion zone between between CT III and this cover type. Scat- 
tered individuals were found in the south-southwest mat and 
on the northeast side of the pond. 


Subtype IVb 


The second subtype is found in regions of the peatland closer 
to the pond on the quaking mat (Figure 4), and is often the 
lake edge association in locations where Carex lasiocarpa 1s 
not as abundant. It extends back from the edge at varying dis- 
tances ranging from 0.5 to approximately 40 meters. It is less 
homogeneous than ST IVa. An increase in abundance and 
frequency of species such as Andromeda glaucophylla (66% 
frequency), Myrica gale (36.4% frequency), Drosera 
rotundifolia (33.3% frequency) and Rhynchospora alba 
(27.3% frequency) distinguishes this subtype from ST IVa. 
Eriophorum virginicum is also quite dominant in this sub- 
type. Other less dominant species found here include Decodon 


76 Rhodora [ Vol. 97 


verticillata and Woodwardia virginica, as well as Peltandra 
virginica, which was absent from ST IVa. 

Although not represented in the sampling, small patches of 
a somewhat exclusive association of Utricularia cornuta, 
Vaccinium oxycoccos, Drosera rotundifolia, Xyris montana, 
and sometimes Calopogon tuberosus were present in this 
subtype, particularly in especially wet and mucky areas. The 
substrate of these mucky areas was not as consolidated as in 
other parts of this subtype, and may not support the weight of 
unsuspecting field botanists. A small localized area, close to 
the pond on the southsouthwest mat, possesses the same char- 
acteristics, and when probed with a peat sampler, was found 
to be approximately 10 meters deep. An interesting feature to 
this location in particular was the sizeable population of about 
200 individuals of Calopogon tuberosus. 

A comparable association was found to be quite common 
at Heath Pond Bog (Fahey, 1993) as well, particularly the 
narrow floating mat immediately adjacent to the lake-edge. 

ST IVbas a whole is also floristically similar to what Dunlop 
(1987) reported in southern New Hampshire as a Vaccinium 
oxycoccos-khynchospora alba subtype, found adjacent to the 
lake edge of Mud Pond Bog. 

The close proximity to the edge probably explains the in- 
creased abundance of Andromeda glaucophylla and Myrica 
gale, species that reflect, at least in other regions of North 
America, a weakly minerotrophic condition (Jeglum, 1971, 
Schwintzer, 1978). Dansereau and Segadas-Vianna (1952) 
found Andromeda glaucophylla to occur often in the wettest 
regions of the Chamaedaphne calyculata dominated associa- 
tion in Canada. As mentioned previously, Rhynchospora 
alba has a distribution in peatlands which seems to be gov- 
erned by moisture and the intolerance of shade rather than 
water chemistry (Vitt and Slack, 1975). 


1995 Fahey and Crow - Pequawket Bog 77 


Acer rubrum-Vaccinium corymbosum-Lyonia ligustrina 
Cover Type (CT V) 


This cover type is found largely around the outer periphery 
of the peatland. Typically it extends from the base of the up- 
land out on to the mat in varying widths (Figure 4). Many of 
its constituents integrade strongly with the Chamaedaphne 
calyculata-Woodwardia virginica cover type (CT HI), but a 
high preference of Acer rubrum, Vaccinium corymbosum, and 
Lyonia ligustrina TWINSPAN distinguishes this cover type. 
It is dominated primarily by tall shrubs over 1.5 meters in 
height, and an understory which is quite variable. A moat of 
standing water is found at the immediate base of the upland, 
often quite deep in places. Although Chamaedaphne 
calyculata has the highest cover ( 18%) and frequency (79.4) 
value, this zone has been named after the taller shrubs based 
on their relative high cover and frequency in combination 
with their relative uniqueness to this zone (Table 5). 

Other important tall, woody species (>1.5 m) that charac- 
terize this cover type include Alnus incana ssp. rugosa, 
Nemopanthus mucronata, llex verticillata, Rhododendron 
canadense, and Aronia melanocarpa. Betula populifolia, 
Cephalanthus occidentalis and Viburnum cassinoides are 
found less frequently. Other low growing shrubs include 
Myrica gale, Kalmia angustifolia, and Spiraea latifolia. The 
more typical herbaceous species include Osmunda regalis, 
Osmunda_ cinnamomea, Carex stricta, Symplocarpus foet- 
idus, Carex trisperma, Rubus hispidus, Lysimachia_terr- 
estris, Triadenum virginicum, Utricularia intermedia, Drosera 
rotundifolia , and Juncus pelocarpus. 

The moat at Heath Pond Bog (Fahey, 1993) is floristically 
very similar, except for the shrubby moat area which has been 
extensively disturbed by beavers. In another New Hampshire 
bog (Dunlop, 1987) the /lex verticillata-Acer-Carex 


78 Rhodora [Vol. 97 


canescens community type occupying the moat shows flo- 
ristic similarities, but the importance of Vaccinium 
corymbosum and Lyonia ligustrina are not as great. 


Table 5. Mean percent cover and percent frequency of dominant species in the 
Acer rubrum-Vaccinium corymbosum-Lyonia ligustrina cover type (CT V). 


Species Mean % Cover % Frequency 
Sphagnum spp. 63 97.4 
Chamaedaphne calyculata 18 fi Me. 
Vaccinium corymbosum* 12 64.1 
Lyonia ligustrina* 12 DU) 
Osmunda regalis 12 30.8 
Alnus incana ssp. rugosa 9 53.8 
Acer rubrum* 8 74.4 
Osmunda cinnamomea 8 23.1 
Myrica gale a 46.1 
Nemopanthus mucronata wi 32.9 
Carex stricta 7 33.3 
Tlex verticillata 7 23.0 
Rhododendron canadense 6 59.0 
Aronia melanocarpa 4 48.7 
Symplocarpus foetidus 4 43.5 
Carex trisperma 4 25.2 
Betula populifolia 3 20.5 
Woodwardia virginica 3 ee, 
Cephalanthus occidentalis 3 es 
Rubus hispidus 2 30.8 
Lysimachia terrestris 2 28.6 
Kalmia angustifolia 2 25.6 
Triadenum virgir.icum 2 2.6 
Vibuurnun cassinoide 2 23.1 
Utricularia intermedia 2 Zo 
Drosera rotundifolia ] ee | 
Spiraea latifolia ] 234 
Juncus pelocarpus 1 209 

* = indicator species for CT V 


1995 Fahey and Crow - Pequawket Bog 72 


Vitt and Slack ( 1975) reported a tall shrub zone (the Mar- 
ginal Moat Zone) encircling bogs in northern Michigan. This 
community type consisted of similar species, but with differ- 
ent abundances. They also noted that this zone appeared to 
be highly variable, but was primarily dominated by Ilex 
verticillata, Nemopanthus mucronata, Viburnum cassinoides, 
and occasionally Osmunda regalis. 

Floristically this cover type strongly resembles the Cinna- 
mon Fern-Highbush Blueberry Thicket described by Damman 
and French (1987), with some variation. This type of tall 
shrub thicket occurs in locations with seasonal water-level 
fluctuations, and that are heavily influenced by minerotrophic 
water from the surrounding upland or seepage (Damman and 
French, 1987). The community was also characterized as hav- 
ing a well developed ground layer of vegetation. 


Basin Profile 


Probings of the peat mat at 10 meter intervals along transect 
1 resulted in a profile of the bog basin (Figure 5). The profile 
shows an apparent basin of a former pond which has been 
completely blanketed by the bog mat. The basin measures 
approximately 11 meters at its deepest point. Figure 5 shows 
a vegetation map of the south and southwest portion of the 
peatland showing transect 1, and the vegetation types through 
which it runs, for comparison with the same zones shown on 
the basin profile. At the deepest part of the basin the 
Chamaedaphne calyculata-Vaccinium oxycoccos-Eriophorum 
virginicum cover type (CT IV) occupies a relatively large 
area, approximately 110 m long; ST 1 Va occupies the major- 
ity of this area, particularly where the basin is deepest. Sur- 
rounding this zone is the Chamaedaphne calyculata- 
Woodwardia virginica cover type (CT III), an area that cor- 
responds with a more shallow basin. Toward the pond edge 


80 Rhodora [ Vol. 97 


the Chamaedaphne calyculata-Vaccinium oxycoccos- 
Eriophorum virginianum CT 1s represented by ST IVb, cor- 
responding with a basin depth of approximately 2-7 meters. 
At the outer edge of the mat the Acer rubrum-Vaccinium 
corymbosum-Lyonia ligustrina cover type (CT V) is found 


upland 


Fee e eer eer re atatahaha 
rrr earns 


<a 


, 
‘ 
1D 
b 
, 
, 

pret h 

rit)? 
ad 
, 

. 

‘ 

iS 

(5 

(Ty 


depth (m) 


a = ST Ta 
BS = str mc 

_ = ST Tva 
ES] = stivp 

| BA - crv 


I 
100 200 306 
distance (m) 
Figure 5. Basin profile (below) along transect 1| at the southwest mat 


in relation to the vegetation types. 


1995 Fahey and Crow - Pequawket Bog 81 


in the areas that show the shallowest probings. This pattern 
seems to reflect the classic theory of lake-fill, or quaking bog 
succession. 

Motzkin and Patterson (1991) recently described the veg- 
etation patterns in Acadia Bog, a Massachusetts moat bog. 
They found a correlation of vegetation types to depth of sedi- 
ment accumulation and distance from shore. Although the 
species composition of the vegetation types of the two bogs 
are not comparable, Pequawket Bog likewise shows a simi- 
lar relationship (Figure 5) in the southernmost portion of the 
southwest mat. 

Probings of the basin depth along transect 8 through the 
Carex lasiocarpa cover type (CT II), showed relatively shal- 
low readings throughout. Depths ranged from 1.0 nearer the 
edge to 3.5 meters closer to the pond margin. In probing this 
section the bottom felt like very loose sand. 


pH 


It has been well documented that pH of the pond and mat 
waters play an important role in the distribution of species 
within a peatland (Jeglum, 1971; Vitt and Slack, 1975; 
Heinselman, 1970; Schwintzer, 1978, 1981; Vitt and Bayley, 


82 Rhodora [Vol. 97 


1984; and others). Table 6 shows the mean pH values of each 
cover type. 

The overall aquatic Nymphaea odorata cover type (CT I) 
has the highest pH, 5.58. The water of the open pond within 
ST Ia averaged 5.69. However, the pool at the northern end 
of the peatland, occupied by ST 1b, was more acidic, averag- 
ing 5.37. 

On the outer edge of the bog the moat waters of the Acer 
rubrum-Vaccinium corymbosum-Lyonia ligustrina cover type 
(CT V) are also relatively less acidic than other cover types 
in the peatland, averaging 5.13. Moat waters are typically 
less acidic and more mineral rich as a result of the influence 
of telluric waters entering from the upland. 

The pH of the Carex lasiocarpa cover type (CT II) aver- 
aged 4.55, but there were considerable differences in the pH 
among its subtypes. ST Ia, adjacent to the open pond water, 
averaged 5.46, while ST IIb averaged 4.23. This may be due 
to the relative distance away from an influence of less acidic 
pond water as well as to an abundance of Sphagnum spp. 
acidifying the mat waters. 

The cover types of the south-southwest mat, which appear 
to have covered an old pond basin (Figure 5), and along the 
west and northeast sides of the pond, generally show lower 
pH levels. The Chamaedaphne calyculata-Woodwardia 
virginica cover type (CT III) 1s less acidic, with a pH of 4.18, 
than the Chamaedaphne calyculata-Vaccinium oxycoccos- 
Eriophorum virginiana cover type (CT IV) averaging 3.89. 

Vitt and Slack ( 1975) measured pH, as well as cation con- 
centration, along a transect from the pond edge to the base of 
the upland and found these to be very influential in the distri- 
bution of plant species and communities. The pH was high- 
est in the pond water and at the alkaline mat edge. As the 
distance increased from the water’s edge, across the mat, the 
pH dropped dramatically, but showed an increase again near 


1995 Fahey and Crow - Pequawket Bog 83 


cand in the moat water. A similar pattern was observed for 
Pequawket Bog 

While the pH values throughout the Pequawkct peatland are 
very low over all, this is to be expected for regions with 
noncalcareous bedrock. Lakes of New Hampshire have natu- 
rally acidic conditions, as a result of surrounding acidic bed- 
rock, such as granite. 


Species Density 


Some ecologists have investigated the variations in species 
density, or richness, within and between peatland ecosystems 
(Heinselman, 1970; Schwintzer, 1981), and have attempted 
to explain the differences observed. 

At Pequawket Bog, the mean species density (number of 
species per m?) was calculated for each subtype and the moat 
cover type (Figure 6). The Nymphaea odorata CT is quite 
depauperate with ST la and ST Ib averaging 4.4 (+ 1.4) and 
4.6 (+ 1.5) species/m? respectively. Conversely, within the 
Carex lasiocarpa CT, ST Ia has a noticeably higher species 
density of 11.2 (+ 2.9), compared to all the other cover types 
and subtypes. Of the Chamaedaphne calyculata dominated 
subtypes the ST IIIa has the highest species density with 8.9 
(+ 1.9) on average, while ST IVa averaged only 4.2 (+ 2.7) 
species/ m? The Acer rubrum-Vaccinium corymbosum-Lyonia 
ligustrina CT, the tall shrub zone occupying the moat, aver- 
aged 8.4 (+ 2.7) species/ m’. 

A possible explanation for the differences in species den- 
sity between subtypes may be found when applying Grime’s 
(1979) model for the control of species density in herbaceous 
vegetation. Although this model is for herbaceous vegetation 
and the validity of applying it to a woody community may 
have its limitations. This model proposes that vegetation in 
the presence of moderate levels of either stress or disturbance, 


84 Rhodora [Vol. 97 


or both, has an “increase in species density by reducing the 
vigor of potential dominants, thus allowing subsidiary spe- 
cies to co-exist with them” (Grime, 1979, p. 162). As these 
stresses or disturbances approach extremes, a decline in spe- 
cies density is found. Under extreme conditions only a small 
number of species appear to be able to survive. 

Table 7 compares the vegetation types based on statistically 
significant differences in species density. Because many of 
the environmental parameters which could influence a stress 
or stimulus on the component species of the vegetation in the 
peatland were not sampled, and because levels of disturbance 
where not quantified, reasons for differences between veg- 
etation types can only be speculative. However, the values 
measured for pH lend themselves to a possible explanation 
of differences between certain subtypes, as do observations 
of some disturbances, particularly by beavers. Grime (1979) 


Table 7. A comparison of the vegetation types in Pequawket Bog showing een 
differences in species density at the 95% confidence level (p < 0.05). 


STla STIb STWa STUb STIIa STINb STIlHe STIVa STIVb 


ST Ia 

ST Ib --- 

ST Ila * * 

ST Ilb --- _ *k 

ST Illa * * Seu ses 

ST IIb] --- — * a Bs 

ST IIIc # * * ee a 

STIVa | --- — * x * ae x 

STIVb | --- --- % --- --- --- 7 --- 
CTV : * * --- eae # = * _ 
¥ 


StpIMIL aiit al FJ /O 


1995 Fahey and Crow - Pequawket Bog 85 


shows there is a relationship between the pH of different habi- 
tats and species density. Other possible influences may in- 
clude fluctuations in water level, conductivity, alkalinity, and 
aeration. 


14 
12 
a 10 
£ 
a 8 
iy 
th 


a= ON 
-#—e—H 
Fer 
-—o—_| 


~ 


ST Ib7 
ST Taq 
S111 
ST aq 

ST IIIb] 
ST Hic] 
ST [Va] 
ST IVb" 
CT 


Vegetation types 
Figure 6. Average species density, expressed as number of species per 
square meter( for all vegetation cover types and subtypes 


Within the Carex lasiocarpa CT, ST Ha shows a significantly 
higher species density than all other vegetation cover types 
except for ST Illa. The areas in the peatland where ST Ila Is 
found are on the edge of the pond, which is subjected to wa- 
ter level fluctuations, and in areas that show considerable dis- 
turbance by beaver (Figure 4). Likewise, the small area oc- 
cupied by ST Ila shows signs of beaver disturbance 


86 Rhodora [Vol. 97 


Schwintzer ( 1981) founda relationship between species den- 
sity of vascular plants in three northern Michigan wetland 
types (bogs, fens, and conifer swamps) and degree of telluric 
water influence. In bogs with little influence from telluric 
waters, a low species density was found due to the stressful, 
nutrient-poor environment. At Pequawket Bog, within the 
Chamaedaphne calyculata-Vaccinium oxycoccos-Eriophorum 
virginiana CT, ST [Va had a significantly lower richness than 
many other mat vegetation types, and is found in an area of 
the southwest mat than is furthest away from the upland. It 
can be assumed that this location is also relatively poor in 
telluric minerals as reflected by its low pH (3.72). Any 
inflowing telluric water from the upland probably does not 
reach the center of the mat; additionally, there is no export of 
acids out of this area. Schwintzer (1981) reports substantially 
higher species density in fens and conifer swamps with a 
moderate to strong influence of telluric water. Similarly, at 
Pequawket Bog higher species richness was seen in areas of 
the moat and on the edge of open water, where the pH is 
higher. Schwintzer also points out that there are other factors 
that influence species density, such as tree layer, water level 
fluctuations, and range and number of microsites provided 
by microrelief. 

Heinselman (1970) compared the richness of seven differ- 
ent peatlands in northern Minnesota and found an increase in 
number of species as the flow-through conditions increased. 
The stimulus of the increase is thought to be due to the input 
of mineral nutrients and a reduction in anacrobic conditions. 


SUMMARY 
In summary, Pequawket Bog is a level peatland complex with 


an aquatic community, a sedge meadow, two areas with more 
typical bog flora, (Sphagnum spp. and low ericaceous shrubs 


1995 Fahey and Crow - Pequawket Bog 87 


dominating), and a tall shrub or moat cover type. The five 
major cover types reflect general vegetational and floristic 
similarities with other level peatlands in northeastern North 
America. With a diverse flora of 109 vascular plants, the 
peatland complex might be best classified as a “rich fen”. 


ACKNOWLEDGEMENTS 


We are grateful to Drs. A. Linn Bogle and Janet R. Sullivan 
for their helpful comments an suggestions on the manuscript. 
Thanks also goes to Cindy Balcius for the many hours of 
assistance in the field, and to Mark Anderson for all of his 
help with TWINSPAN. The project has been supported in 
part by a Central University Research Fund Grant (CURF 
1312) and a Teaching Assistantship Summer Fellowship to 
Linda Fahey. This is Scientific Contribution No. 1883 from 
the New Hampshire Agricultural Experiment Station. 


LITERATURE CITED 


Barrett, PE. 1966. The relationship of some physical and edaphic fac- 
tors to plant succession in a southern New Hampshire bog. M.S. 
thesis. University of New Hampshire, Durham, NH. 

Bay, R. R. 1967. Groundwater and vegetation in two peat bogs in north- 
ern Minnesota. Ecology 48: 308-310. 

Brewer, R. 1966. Vegetation of two bogs in southwestern Michigan. Mich- 
igan Bot. 5: 36-46. 

Clymo, R. S. 1963. Ion exchange in Sphagnum and its relation to bog 
ecology. Ann. Bot. 27: 309-324. 

1964. The origin of acidity in Sphagnum bogs. Bryologist 
67: 427-431. 

Cook, E. M., Jr. 1989. Ossipee, New Hampshire, 1785-1985: A His- 
tory. Peter E. Randall, Publisher, Portsmouth, NH. 

Conway, V. M. 1949. The bogs of central Minnesota. Ecol. Monogr 19: 
173-206. 

Cowardin, L. M., V. Carter, EF C. Golet, and E. T. LaRoe . 1979. Classifi- 


88 Rhodora [Vol. 97 


cation of Wetlands and Deepwater Habitats of the United States. 
J. S. Fish and Wildlife Service. Biological Services Program; 
FWS/OBS-79/31. 
Craigie, J. S. and W. S. G Maass. 1966. The cation-exchange in Sphag- 
num spp. Ann. Bot. 30: 153- 154. 
Crow, G. E. 1969. An ecological analysis of a southern Michigan bog. 
Michigan Bot. 8: 11-27. 
and C. B. Hellquist. In press. Aquatic and Wetland Plants of 
Northeastern North America. University of Wisconsin Press, 
Madison, WI. 
Crum, H. A. 1988. A Focus on Peatlands and Peat Mosses. 
Univ. of Michigan Press, Ann Arbor, MI. 
Damman, A. W. H. and T.W. French. 1987. The ecology of peat bogs of 
the glaciated northeastern United States: a community profile. 
Biol. Rep. 85. (7.16) U.S. Fish and Wildlife Sery., Washington, 
one 
Dansereau, P. and I. Segadas-Vianna. 1952. Ecological study of the peat 
bogs of eastern North America. I. Structure and evolution of 
vegetation. Canad. J. Bot. 30: 490-520. 
Davis, M. B., R. W. Spear, and L. C. K. Shane. 1980. Holocene climate 
of New England. Quat. Res. 14: 240-250. 
Diers, R. W., and FE. J. Vieira. 1977. Soil Survey of Carroll County, New 
Hampshire. U.S. D. A. Soil Conservation Service and Forest 
Service in cooperation with New Hampshire Agric. Exp. Sta. 
DRED. 1987. Protected Plants of New Hampshire. Natural Heritage In- 
ventory. RES-N 306 Plant Listing. Dept. Resources and Eco- 
nomic Development. Concord, NH. 
Dunlop, D. A. 1983. The flora and vegetation of a New Hampshire peat 
bog. M.S. thesis. Univ. of New Hampshire, Durham, NH. 
. 1987. Community classification of the vascular vegetation of a 
New Hampshire peatland. Rhodora 89: 415-440. 
Fahey, L. L. 1993. A vegetation, floristic, and phytogeographic analysis 
of two New Hampshire peatlands. M.S. thesis. University of 
New Hampshire, Durham, NH. 
Fahey, L. L. and G. E. Crow. In press. Flora of Pequawket and Heath 
Pond Bogs, Ossipee, NH. 
Gates, F. C. 1942. The bogs of northern lower Michigan. Ecol. Monogr 
12: 213-254. 
Gleason, H. A. 1952. The New Britton & Brown Illustrated Flora of the 
Northeastern United States and Adjacent Canada. 3 vols. New 
York Botanical Garden, New York. 


1995 Fahey and Crow - Pequawket Bog 89 


and A. Cronquist. 1991. Manual of Vascular Plants of North 
eastern United States and Adjacent Canada. 2nd ed. New York 
Botanical Garden, NY. 

Golet, F C. and J. S. Larson. 1974. Classification of freshwater wetlands 
in the glaciated northeast. U.S. Fish and Wildl. Serv., Bur. of 
Sport Fish. and Wildl. Res. Publ. 116. 

Gore, A. J. P, Ed. 1983a. Ecosystems of the World, 4A, Mires: Swamp, 
Bog, Fen, and Moor. General Studies. Elsevier, Amsterdam. 

1983b. Ecosystems of the World, 4B, Mires: Swamps Bog, 
Fen, and Moor. Regional Studies. Elsevier, Amsterdam. 

Gorham, E. 1956. The ionic composition of some bog and fen waters in 
the English Lake district. J. Ecol. 44: 142-152. 

1957. The development of peatlands. Quart. Rev. Biol. 32: 145- 
166. 

Grime, J. P. 1979. Plant Strategies and Vegetation Processes. John 
Wiley and Sons, New York. 

Hansen, H. P. 1933. The tamarack bogs of the driftless area of Wiscon- 
sin. Bull. Public Museum City of Milwaukee 7: 231-304. 

Heinselman, M. E. 1963. Forest sites, bog processess, and peatland types 
in the Glacial Lake Agassiz Region, Minnesota. Ecol. Monogr 
33: 327-374. 

1970. Landscape evolution, peatland types, and the 
environment in the Lake Agassiz Peatlands Natural Area, Min 
nesota. Ecol. Monogr. 40: 235-261. 

Hemond, H. FE. 1980. Biogeochemistry of Thoreau’s Bog, Concord, Mass- 
achusetts. Ecol. Monogr. 50: 507-526. 

Hellquist, C. B. 1971. Vascular flora of Ossipee Lake, New Hampshire 
and its shoreline. Rhodora 73: 249-261. 

, and G. E. Crow. 1980. Aquatic vascular plants of New 

England: Part 1. Zosteraceae, Potamogetonaceae, Zannichelliaceae, 
Najadaceae. New Hampshire Agric. Exp. Sta. Bull. 515. 

Hill. M. O. 1979. TWINSPAN - A FORTRAN program for arranging 
multivariate data in an ordered two-way table by classification 
of the individuals and attributes. Ecology and Systematics, 
Cornell University, Ithaca, N-Y. 

Janssen, C. R. 1967. A floristic study of forests and bog vegetation, north- 
western Minnesota. Ecology 48: 751-765. 

Jeglum, J. K. 1971. Plant indicators of pH and water level in peatlands 
and Candle Lake, Saskatchewan. Canad. J. Bot. 49: 1661-1676. 

_ A. N. Boissonneau and V. F. Haavisto. 1974. Towards wet- 


90 Rhodora [ Vol. 97 


land classification for Ontario, Canada. Canad. For Serv., Sault 
Ste. Marie., Ont. Ibf. Rep. O-X-215. 

Johnson, C. W. 1985. Bogs of the Northeast. University Press of New 
England, Hanover, NH. 

Keough, J. R. and R. W. Pippen. 1981. A comparison of vegetation pat- 
erns in two ie bogs in southwest Michigan. Michigan Bot. 
20: 157- 166 

Krauss, R. W. and G. N. Kent. 1944. Analysis of four New Hampshire 
bogs. Ohio J. Sei. 44: 11-17. 

Mitsch, W. J. and J. G. Gosselink. 1986. Wetlands. Van Nostrand 
Reinhold, New York. 

Montgomery, J. D. and D. E. Fairbrothers. 1963. A floristic comparison 
of the vascular plants of two sphagnous wet lands in New Jer - 
sey. Bull. Torrey Bot. Club 90: 87-99. 

Moore, P. D. and D. J. Bellamy. 1974. Peatlands. Springer-Verlag, Inc., 
New York. 

Motzkin, G. H. and W. A. Patterson II. 1991. Vegetation patterns and 
basin morphometry of a New England moat bog. Rhodora 93: 

Mueller-Dombois, D. and H. Ellenberg. 1974. Aims and methods of veg- 
etation ecology. John Wiley and Sons, New York. 

Osheyack, G. D. and I. A. Worley. 1981. Primary production at a com- 
plex peatland in northern Vermont. Proc. Int. Peat Symp. 
Bemidji, Minnesota. 

Rawinski, T. J. 1984. New England Natural Community Classification. 
Unpublished document. The Nature Conservancy Eastern Re- 
gion Heritage Task Force, Boston, Massachusetts. 

Rhodes, J. W. 1933. An ecological comparison of two Wisconsin peat 
bogs. Bull. Public Museum City of Milwaukee 7: 305-362. 

Schwintzer, C. R. 1978. Vegetation and nutrient status of northern Mich- 
igan fens. Canad. J. Bot. 56: 3044-3051. 

1981. Vegetation and nutrient status of northern 

Michigan bogs and conifer swamps with a comparison to fens. 

Canad. J. Bot. 59:842-853. 

and T. J. Tomberlin. 1982. Chemical and physical 
characteristics of shallow ground waters in northern Michigan 
bogs, swamps, and fens. Amer. J. Bot. 69: 1231-1239 

Sjors, H. S. 1959. Bogs and fens in the Hudson Bay Lowlands. Arctic 12: 
1-19. 

1963. Bogs and fens on the Attawapiskat River, northern 
Ontario, Natl. Mus. Canad. Bull. 186 


1995 Fahey and Crow - Pequawket Bog 91 


Slack, N. G., D. H. Vitt, and D. G. Horton. 1980. Vegetation gradients of 
minerotrophically rich fens in western Alberta. Canad. J. Bot. 
58: 330-350. 

Soil Survey Staff. 1975. Soil taxonomy, a basic system of soil classifica- 
tion for mapping and interpreting soil surveys. Soil Cons. Serv, 
U.S.D.A., Agric. Handbk. No. 436 

Sperduto, D. 1994. A classification of the natural communities of New 
Hampshire. Working draft. New Hampshire Natural Heritage 
Inventory. Department of Resources and Economic Develop- 
ment. Concord, NH. 

Stanek, W. 1977. Classification of Muskeg . pp. 3/-62. In: Radforth, N. 

and C. O. Brawner, Eds., Muskeg and the Northern Environ- 
ment in Canada. University of Toronto Press, Toronto, Canada. 

Swan, J. M.A., and A. M. Gill. 1970. The origin, spread, and consolida- 
tion of a floating bog in Harvard Pond, Petersham, Massachu- 
setts. Ecology 51: 829-840. 

Transeau, FE. N. 1905. The bogs and bog flora of the Huron River Valley. 
Bot. Gaz. (Crawfordville) 40: 351-375; 418-448. 

1906. The bogs and bog flora of the Huron River Valley. 
Bot. Gaz. (Crawfordville) 41: 17-42. 

Vitt, D. H. and S. Bayley. 1984. The vegetation and water chemistry of 
four oligotrophic basin mires in northwestern Ontario. Canad. J. 
Bot. 62: 1485-1500. 

and N. G. Slack. 1975. An analysis of the vegetation of Sphag- 
num dominated kettle-hole bogs in relation to cnvironmental gra- 
dients. Canad. J. Bot. 53: 332-359. 

D. G. Horton, N. G. Slack, and N. Malmer 1989. Sphagnum- 
dominated peatlands of the hyperoceanic British Columbia 
coast: patterns in surface water chemistry and vegetation. Canad. 
J. For. Res. 20: 696-711 

Wells, E. D. 1981. Peatlands of eastern Newfoundland: distribution, 
morphology, vegetation, and nutrient status. Canad. J. Bot. 
59: 1978-1997. 

Wilson, J. R. 1969. The Geology of the Ossipee Lake Quadrangle, New 
Hampshire. Bulletin No. 3. New Hampshire Department of Re- 
sources and Economic Development, Concord, New Hampshire. 

Worley, I. A. 1981. Maine Peatlands. Maine Critical Areas Program. 

Planning Report 73 and Vt. Agric. Exp. Sta. Bull. 687. 
and J. R. Sullivan. 1980. Classification scheme for the peat- 
lands of Maine. Vt. Agric. Exp. Sta. Res. Rep. 8. 


92 Rhodora 


LL 
Paul Smith’s College 
Paul Smiths, New York 12970 


e151 6 

Department of Plant Biology 
University of New Hampshire 
Durham, New Hampshire 03824 


[Vol. 97 


O3 


Rhodora [ Vol.97 


Herbarium J. K. Morton 


Name Sonchus palustris L. 


Family Compositae 
Collector J.K. Morton and Joan M. Venn 
No NA18427 Dned 15 / Aug ./1994 


Locality ON : Waterloo Regior 


1: S end of 
Cambridge (G) - Hwy 24 & 
jnct. 


Myers Rd. 
Country Canada 


Remarks 


Perennial (or ?biennial) herb 6-8 £ 
high in rank somewhat marshy ground 
beside abandoned railway 1 


ine. Several 
dozen plants noted over an acre or 
Specimens in JKM WAT DAO GH 
Det. JKM 1994 


Ligure 1. Photograph of herbarium specimen of Marsh Sow-thistle x 1/3 


RHODORA, Vol. 97, No. 889, pp. 93-95, 1995 


THE MARSH SOW-THISTLE (SONCHUS PALUSTRIS) IN NORTH 
AMERICA. 


J. K. MorTON AND JOAN M. VENN 


ABSTRACT 


The discovery of the Marsh Sow-thistle (Sonchus palustris) growing in the Wa- 


terloo Region of Ontario, Canada is reported. 


Key Words: Marsh Sow-thistle, Sonchus palustris Linn., North America, Ontario. 


The Marsh Sow-thistle (Sonchus palustris Linn.)occurs in 
Cambridge (Galt), Waterloo Region, Ontario, Canada, where 
it grows in tall, herbaceous, marshy vegetation beside an aban- 
doned railway line on the southern outskirts of the city. This 
year (1994) there were many plants of the sow-thistle grow- 
ing in two colonies a few hundred yards apart. 

The Marsh Sow-thistle is a spectacular plant with its stout 
flowering stems towering over the surrounding vegetation to 
a height of 6 to 8 ft. (up to 2.5 m). Each stem is stiffly erect 
with a terminal umbel of rich yellow flower heads similar to 
those of the common Field Sow-thistle (S. arvensis L.) but 
only about half the size. The leaves of the Marsh Sow-thistle 
are distinctive, being large, narrow [c 6-8" long x 3/4-1" (15- 
20 cm x 2-3 cm) broad] and pointed with sharply pointed 
auricles at the base. The margins are straight and minutely 
scabrid-toothed. This type of leaf is very different from that 
of the Field Sow-thistle which has coarsely sinuate-lobed 
leaves, Broadest above the middle and conspicuously toothed. 

The Marsh Sow-thistle is a native of Europe where it oc- 
curs locally in marshy ground and wet places across most of 
the continent except in the far north, The occurence of this 
species at Cambridge came to out attention when we were 
working through a pile of unidentified herbarium specimens 


94 


1995 Morton and Venn - Marsh Sow-thistle 95 


recently donated to the University of Waterloo herbarium by 
Mr. Craig Campbell of Waterloo. The specimens included a 
collection of an unfamiliar Sonchus from the above locality 
made on August 10th, 1973 - Campbell and Reznicek 73-65. 
This proved to be Sonchus palustris. We visited the locality 
in late July 1994 and soon located the plant which was, how- 
ever, only in bud. On a subsequent visit in mid-August it Was 
in flower and fruit, and voucher material was collected. It 1S 
clear that the Marsh Sow-thistle is well established in this 
locality as it has persisted here for over 20 years. As far as we 
have been able to ascertain this is the first report of this Euro- 
pean species from North America. 

Voucher Collections: CANADA: Ontario: Cambridge 
(Galt), Waterloo Region, Campbell and Reznicek 73-65 (WAT); 
and idem. Morton and Venn, NAI8&427 (WAT, JKM, DAO, GH). 


DEPARTMENT OF BIOLOGY 
UNIVERSITY OF WATERLOO 
WATERLOO, ONTARIO 
CANADA N2L 3G1 


RHODORA, Vol. 97, No. 889, pp. 96-97, 1995 


URTICA CHAMAEDRYOIDES PURSH (URTICACEAE) 
REPORTED AS NEW TO CUBA 


Deirry V. GELTMAN 
ABSTRACT 


Urtica chamaedryoides is reported for the first time to Cuba . 


Kkey Words: Urticaceae, Cuba, Systematics, Urtica. 


While examining specimens for a taxonomic treatment of 
Urtica L. for Latin America, three Cuban collections from 
the Gray Herbarium of Harvard University struck me as be- 
ing unusual. They were originally identified as Urtica urens 
L.,a species introduced from Europe and recorded from Cuba 
by Leon and Alain (1951). On closer examination, however, 
they have proven to represent Urtica chamaedryoides Pursh, 
an annual species widely distributed in the southeastern United 
States from Southern Ohio, southern Illinois west to south- 
eastern Kansas, south to central Florida, Texas and Louisi- 
ana, and also occurring In Mexico (Woodland et al., 1976). 


Specimens examined:CUBA; Plantae Cubensis Wrightianac, 
C. Wright 3681, s.d. (GH); Santa Clara Province, Las Lagu- 
nas, Buenos Aires, about 2500 [t. alt., 21 April 1930, J. G. 
Jack 7943 (GH); Las Villas Provinee, Trinidad Mountains, 
San Blas-Buenos Aires, Finca La Carida, 1 June 1942, A. 
Goncdles 485 (GH). The first specimen has no locality infor- 
mation (and is not cited in Grisebach's (1866) work on Wright 
collections), localities of the two others could not be located 
precisely, but itis clear that they come from the same general 
area in the Sicrra de Trinidad, located at the junction of the 
present provinces of Cienfuegos, Villaclara, and Sancti 
Spiritus. 


96 


1995 Geltman - Urtica chamaedryoides oF 


Urtica chamaedryoides in its North American continental 
range 1s found growing in humid, rich soils of bottom lands 
and flood plains, rich woods, and waste places (Woodland ef 
al., 1976). Although there is no ecological information avail- 
able for the collections cited above, their occurrence in a 
montane habitat, typical for Urtica in the tropics, suggest 
that this species 1s indeed native to Cuba. 


ACKNOWLEDGMENTS. 


Support for my studies on the taxonomy of LatinAmerican 
Urlica species is provided by the Missouri Botanical Gar- 
den, which is gratefully acknowledged. I also grateful to GH 
for sending specimens on loan. Finally | thank Jim Solomon 
for the help in editing of the manuscript. 


LITERATURE CITED 


GRISEBACH, A. 1866. Catalogus plantarum cubensium. Leipzig. 301 p. 

LEON, H. & Il. ALAIN. 1951. Flora de Cuba. Vol. 2. 456 p. 

WOODLAND, D. W., I. J. BASSET, AND C. W. CROMPTON. 1976. The an- 
nual species of stinging nettle ( Hesrerocnide and Urtica) in North 
America, Can. J. Bot. 54; 374-383. 


Herbarium 

Komarov Botanical Institute 
Prof. Popov St. 2, 

St. Petersburg, 197376, Russia 


RHODORA, Vol. 97, No. 889, pp. 98-107, 1995 


RHODORA NEWS/NOTES 
Lisa A. STANDLEY 


HIGHLIGHTS OF CLUB MEETINGS 


April 1994 (898th Meeting). Dr, Robery Raffauf, Emeritus Professor 

of Northeastern University's College of Pharmacology, spoke on 
Rainforests and Medicinal Chemistry, the focus of his research and col- 
laboration with Dr. Richard Schultes for the past 40 years. 

Dr. Raffauf provided a whirlwind tour through the history of medica! 
ethnobotany, from the first recorded uses of plant medicines in 4,000 
BC. Ethnobotany can be traced back to Adam and Eve who noted that 
certain fruits had undesirable side effects. Some highlights included the 

igyptian’s use of moldy bread to treat wounds; the description of Kau- 
re as a Sedative 3,000 years ago in India; Hippocrates’ use of pre- 
cursors of salycilic acid; and the efforts of the Spanish in the New World 
to record indigenous uses of medicinal plants. Some notable drugs de- 
rived from plants include digitalin (introduced in the late 1700s), 
scopalomine from Brugmansia; ergotamine, for migranes; atropine; 
and steroid precursors from Dioscorea. Despite this long history, only 
11% of the drugs listed in the current US Pharmacopeia are derived 
from plants - and these are from a short list of 35-40 species, most of 
which (with the exception of Catharanthus) have been used for centu- 
ries. Drug companies have, mostly unsuccessfully, attempted to syn- 
thesize these plant compounds to provide more controlled and predict- 
able dosages. New genetic engineering techniques are likely to provide 
the key to synthesis of plant medicinal compounds 

Dr. Raffauf examined the popular concept that there are a vast num- 
ber of new natural drug-producing plants 1n the rainforest waiting to be 
discovered. Based on 40 years of reseach, his view is that there are no 
compelling arguments for botanical exploration of the rainforest in search 
of new medicines since the potential for finding new drugs is low. 
Of the 1500 medicinal plants used by the indigenous peoples of the 
Amazon, none have unique properties or commecial value. Even if some 
tropical tree products were found to be major drug sources, they would 
not be commercially viable due to the length of time needed for these 
trees to be grown in plantations. For example, curare plants, which are 
relatively fast-growing, require 30 years for a crop to reach ma- 
turity. Dr. Raffauf concluded that the real potential for new 
medicines,and new medical solutions to human disease, will 


98 


1995 Rhodora News.Notes 99 


come from the development of new vaccines and genetic engineering 
- not tropical forests. 


MAY 1994 (899th Meeting, at Smith College). !’ollowing an histori- 
cal introduction to the Smith College Botanical Garden by Dr C. John 
Burk, Dr. Rob Nickolson of Smith College spoke on his work collect- 
ing members of the genus 7axus in remote areas. As aresult of his work, 
the Smith College Botanical Garden currently holds the worlds 
largest collection of wild-originated plants in the Taxaceae 
(Pseudotaxus, Taxus, Cephalotaxus, Torreya, and Amentotaxus). His talk 
focused on 7axus itself, with photos and descriptions of the various 
species in native habitats, including 7. baccata in Northern Europe, T. 
canadensis in Northeastern North America, 7. globosain Mexico grow- 
ing with Magnolia, and 7. chinensis in Taiwan with an eight-foot 
basal diameter. 

The taxonomy of the group appears to be based more upon geography 
than morphology, with few if any real morphological distinctions be- 
tween species. One of Rob’s and his collaborator, Melvin Shemluck’s, 
main interests is the production of taxol. They find that taxol can be 
extracted from three-year old rooted cuttings, thus saving mature trees, 
and that it can be extracted from cloned callus cells. Amentotaxus, an 
anomaly in the family, lacks taxol. Many of his field efforts to locate or 
relocate populations of 7axus led him to areas that had recently been 
deforested, often illegally. The most significant message in Rob’s talk 
was that the world’s yew flora is being severely depleted for umber 
and bark, and that the Asian taxa in particular are likely to be lost soon 
unless there are significant conservation efforts. 


June 1994 (900th Meeting). Former Club President Mary Walker spoke 
on The History of the Flora of Anguilla. The talk centered on her long- 
term project to photograph and document the approximately 480 spe- 
cies of the flora, using Dr. Richard Howard’s checklist for the island. 
Anguilla, a small low island with a dry climate and beautiful beaches, 
has along history of agricultural use. Until the introduction of electric- 
ity in 1970 and the recent rise in tourism, the islanders practiced 
subsistance agriculture. Cows, goats and chickens still roam free 
throughout the island, and pidgeon pea, peppers, yams and tomatoes 
are commonly grown in small gardens, while hydroponically-grown 
lettuce 1s a new cash crop. 

The plant communities in the vicinity of settlements, agricultural ar 


100 Rhodora [Vo]. 96 


eas and former plantations are dominated by introduced species such as 
aloe, indigo, tamarind, mango and calabash. Many of the island’ na- 
tive and introduced species are used medicinally. The dominant natural 
vegetation on the island is described as an evergreen bushland, domi- 
nated by sclerophyllous shrubs. Some plant groups, particularly ferns 
(one!) and orchids (two) are underrepresented. 

Perhaps due to strong selection by goats, the majority of surviving 
shrub species are also protected by thorns. To the average tourist, all 
shrub species, including the island’s only endemic, appear to have simi- 
lar small, entre, leathery evergreen leaves and small red fruits. The 
growth habits of many species are ecologically interesting, particularly 
in the areas Where the substrate consists of bare limestone. Plants, in- 
cluding the thatch palm 7Arinax and one of the island’s two orchids 
(Epidendrum), become established in small solution pits in the lime- 
stone. Other plant communities on the island include mangrove swamps 
and the strand vegetation along beaches, which includes sea grape and 
manchineel. 


October 1994 (902nd Meeting). Dr. Thomas Philbrick spoke on “Tlow- 
ering Plants in River Rapids - Podostemaceae in Mexico”.Tom’s love 
affair with nverweeds began when he first encountered green slime on 
rocks in the Lamprey River during childhood. He later discovered that 
these appear to present interesting evolutionary riddles, since they do 
not conform to the “evolutionary paradigm” of aquatic angiosperms. 
Aquatic flowering plants, in general, are perennial clonal plants that 
rely on vegetative reproduction through specialized propagules, and that 
have a low rate of sexual reproduction. Most families of aquatics have a 
low species diversity and broad distributions. Riverweeds, in contrast, 
are annuals that rely on a high rate of sexual reproduction and popula- 
tion persistence, and have a high level of taxonomic diversity with nu- 
merous endemic species with restricted distributions. 

In Mexico the family consists of 5 genera and 13 species, mostly in 
the genus \farathrum. All taxa occur in rocky, clear water rivers with 
seasonal variations in flow, that may become completely dry during 
some parts of the year. Plants vary considerably in size, with shoots 
ranging from 2 cm to a meter in length. Members of the genus 7ristichia 
are easily mistaken for bryophytes. Although many species have a 
disinctive bright red pigmentation, the ecological role of this color 1s 
not known. Flowers are unremarkable, and lack any perianth - radiation 
in the group appears to be primarily reflected in the evolution of leaves 


1995 Rhodora News/Notes 101 


and vegetative structures. These annual plants grow vegetatively dur 
ing high water. As the water level falls, the leaves die and flowering is 
initiated. Flowering and fruiting occurs at very high rates. Tom hypoth- 
esizes, based on the frequency of bird droppings on rocks in the river 
that seed dispersal occurs when seeds adhere to bird feet. Seeds have 
an outer integument that hydrates and becomes mucilaginous when wet, 
enabling the seeds to stick to rocks. The seedling, on emergence from 
the seed coat, bends toward the rock and produces rhizoidal hairs that 
attach it to the rock 

High levels of endemism occur in the family, with over 50% of the 
species in Mexico occurring ina single river Rampant endemism is 
also reflected in the fact that 50% of all New World species (over 80 
species) occur in Brazil. Although two or three genera are often found 
growing together, no two members of the same genus have been re- 
ported to be sympatric. Taxonomic studies are desperately needed in 
this group, to determine whether the named taxa are distinct at the level 
of species, or if these are simply ecotypes adapted to different 
river conditions. Without a better understanding of patterns of 
taxonomic variation, itis not possible to address questions of phylog- 
eny or understand the patterns of endemisim. 

The numerous endemic species also pose problems for conservation. 
Since many taxa occur in only one river, changes in water quality may 
resultin extinction. The biggest threats appear to be siltation from con- 
struction and industrial pollution, since silt and petroleum distillates 
prevent seeds or seedlings from attaching to rocks. Eutrophication of 
rivers through the addition of domestic sewage increases algal growth, 
which also prohibits attachment. 


November 1994 (903rd Meeting). William Brumback, Conservation 
Director of the New England Wildflower Society, spoke on the New 
England Plant Conservation Program. The New England Plant Conser- 
‘ation Program (NEPCoP) was started in 1990, assisted by a grant from 
the Cox foundation. The program 1s a voluntary organization of federal, 
state and private non-profit groups dedicated to the conservation of New 
England plants. By providing a forum for collaboration and interaction, 
the program increases the efficiency of state groups and more effective 
use of limited resources. A Regional Advisory Council sets policy and 
develops program priorities. State Task Forces bring together the most 
knowledgeable people in each state, including state and non-profit 
organizations as well as amateurs. The strong conservation ethic in New 


102 Rhodora [Vol. 96 


England has allowed a high level of cooperation between dif- 
ferent agencies, One of NEPCoP’s major accomplishments to date was 
a symposium, held in 1992 and co-sponsored by the Club, that estab- 
lished policies for rare plant conservation. The symposium talks were 
published in Rhodora, and NEPCoP policies were published in “Wild- 
flower Notes”. These have been widely distributed, and are used by 
plant conservation groups nationwide. 

Bill contrasted the problems associated with in-situ and ex-situ con- 
servation. NEPCoP is currently concentrating on ex-situ preservation 
through the collection of seeds to preserve genetic variation of rare spe- 
cies. This 1s an experimental program which is intended to complement 
in-situ protection. Seed banking also allows researchers to explore as- 
pects of the reproductive biology of rare species, and provides material 
for re-introduction. Re-introduction is a somewhat controversial issue 
in the conservation community, as there are concerns that it would al- 
low habitat destruction. The New England Wildflower Society is cur- 
rently developing a rare plant garden as part of an education program 
focused on rare flora. 

NEPCoP has developed a list of regionally rare taxa called “Flora 
Conservanda”. Four categories of rare species are listed: globally rare 
(D-1), regionally rare (D-2), not rare but with populations of conserva- 
tion importance (D-3), and species which formerly occurred in New 
England but for which there are no extant populations (D-4). The Re- 
gional Task Force has been working to develop this list, and set protec- 
tion priorities for the states. 

State Task Forces are currently working with volunteers to collect 
seed of listed plants which occur on private, unprotected property This 
effort 1s intended to ensure that genetic variation of these populations is 
preserved. No collecting is being done from protected lands. NEPCoP 
staff work with the state Heritage Programs to develop lists of 
occurrences of target species, and to obtain permits and permis- 
sion to collect seed. 

NEPCopP, through the New England Wildflower Society, also spon- 
sors Rare Plant Monitors. These lay volunteers are trained and work 
with Heritage Program staff to investigate and monitor known or 
historic occurrences of listed species. In 1994, Dr Paul Somers of the Mas- 
sachusetts Natural Heritage Program used 20 volunteers to investigate 
over 60 sites that would not have otherwise been visited. 

Bill closed the evening with slides of regionally rare plant taxa in the 
four categories of Flora Conservanda. Some highlights of successful 


1995 Rhodora News/ Notes 103 


seed banking and propagation included Agalinis acuta, Hieracium 
robbinsii (last seen in 1956, rediscovered by Bill in 1994), and Poten- 
tilla robbinsiana, successfully grown from seed. Arethusa bulbosa is 
an example of a D-3 species, of concern in Massachusetts 


December 1994 (904th Meeting). Dr. Richard Howard of the Harvard 
University Herbaria, reviewed “Harvard's Role in Neotropical Botany; 
an Historical but Partial Review”. The talk consisted of a series of en- 
tertaining and enlightening anecdotes and photographs of Harvard Bota- 
nists and botanical gardens associated with work in the neotropics.The 
Harvard Botanical Gardens were founded in 1805, with W. D. Peck as 
the first director. He was succeeded by Thomas Nuttall, who resigned 
the post to collect plants in the western US, and sold his 
personal herbarium to Philadephia, apparently out of spite. The 
former botanical garden was, based on slides, laid out in classic 
Linnean order, with a Victorian glass house and palm rotunda. 
When the garden was abandoned in 1947, many of the tropical plants 
were given to Wellesley College. Few of the woody specimens re- 
main, although the female ginkgo planted by Asa Gray survives today, 
and its fruits are collected by some of Boston’s Chinese population. 
The former director’s house still remains on Garden Street, and the 
former Herbarium building (also on Garden Street) is now the Harvard 
University Press. Fernald Drive and Robinson Drive commemorate di- 
rectors of the Herbarium. 

The flora of the neotropics, particularly Central America, the Greater 
Antilles, and the Lesser Antilles, has been a primary focus of the Harvard 
Botanical institutions over the past century and a half. Botany at Harvard 
has included the Botanic Garden, the Gray Herbarium, the Farlow 
Herbarium, the Botanical Museum, the Oakes Ames Orchid Her- 
barium, the Arnold Arboretum, and the Atkins Institute, a botani- 
cal garden and research center in Cuba. The Atkins was important 
in the training of many prominent taxonomists, whose educations in- 
cluded learning to drive a Model T and collect from horseback. Luck- 
ily, botanical collectors are no longer required to wear wool 3-piece 
suits, tes, and hats at all times in the field. The history of neotropical 
botany at Harvard has involved colorful personalities, incidents, and 
occurrences which could only be briefly mentioned in this talk. Dick 
recommended a book about Oakes Ames, titled “Portrait of a Harvard 
Botanist’, to those who would like their gossip in greater detail 


104 Rhodora [Vo |. 96 


January 1995 (905th Meeting). Don Hudson led the evening’s parade 
of club members with stories to tell and slides to show One of his 
highlights was a helicopter tour of a proposed extension of the Appal- 
achian Trail from Katahdin to Mont Albert. As demonstrated by the 
slides, club Members ranged throughout the New World in 1994. Alaska 
was a popular destination. Leila Schultz visited Alaska, the Diomedes 
Islands and the Chukchi Peninsula of Siberia in search of arctic artemi- 
sias, and showed slides of spectacular (although minute) plants. Mark 
Primack showed us fall foliage in Denali National Park - more spec- 
tacular color than even New Englanders are used to. Gary VanWart 
went looking for conifers in the Pacific Northwest, and succeeded in 
finding most of the 33 native species, including the very rare Brewer's 
Spruce. 

Moving east and south, George Newman visited the Gaspe (again), 
with a focus on unique ferns of Mont Albert including Cheilanthes 
siliquosa, Polystichum scopulinum, Adiantum aleuticum, and 
Cystopteris montana. 

Club members also travel through time as well as space. Tonya Largy 
described an excavation of a late Archaic cremation burial site on the 
Blackstone River, with 35,000 year old seeds of Rubus, Gaylussacia, 
Carya, Corylus, and Quercus, as well as mystery rhizomes. 

Coastal Plain ponds were also a popular destination. Mark Primack 
introduced us to rare white form of Sabbatia campanulata, and a hy- 
brid of the two Sabbatias. Pat Swain and Paul Somers showed the New 
England boneset, as well as other Massachusetts rare plants. Paul re- 
ported on the sandplain gerardia’s status, and experimental burns and 
re-introductions. NE-PCoP’s volunteer conservation corps posed for 
>aul’s camera. This active group has contributed 40-50 new records of 
rare species a year, including the second site in the state for Carex 
polymorpha. 

Ranging still further south, David Hunt shared his work on the rare 
Florida scrub oak, reduced to fewer than 100 populations (and most of 
those with For Sale signs tacked onto the oaks). These scrub communi- 
ties contain the largest concentration of rare species in the eastern US, 
but are threatened by expansion of citrus groves and development. David 
showed the third bird slide of the evening (the globally endangered 
Florida Scrub Jay), a new record for show and tell. 

Two members ventured to South America. Garrett Crow traveled 
through Bolivia with a graduate student, searching for aquatics from 
12,000 ft to near sea level. Although high elevation aquatics include 


1995 Rhodora News/ Notes 105 


many familiar forms such as Limosella and Callitriche, lower eleva- 
ions on the Amazonian side of the Andes include the aptly-named 
Equisetum giganteum. 


NEWS 


Two valuable and active Club members passed away in 1994: Sib 
Higginbotham, who will be missed at Club meetings and field tnps, 
and Dorothy Waleka, who mounted many of the speciments 1n the Club’s 
herbarium. 


NOTES 


The Rhode Island Natural History Survey is a new organization 
with a mission to “advance scientific knowledge of Rhode Island’ 
biota, ecological communities, and environmental resources; to facili- 
tate and coordinate the gathering and dissemination of information on 
Rhode Island’s biota and natural communities, and to enhance commu- 
nication among Rhode Island’s environmental and life scientists”. The 
new group organized a very successful symposium in October, and is 
preparing a directory of Rhode Island naturalists. For information, con- 
tact the Rhode [sland Natural History Survey, Inc. at the Cooperative 
Extension Center, E. Alumni Ave. University of Rhode Island, Kingston 
RI 02881. 


The Conference on the Status and Management of Old Growth 
Forests in the Northeast was held October 29-30, 1994, in 
Williamstown, MA. Sponsored by the MassachusettsAudubon Society, 
the conference allowed scientists and old-growth enthusiasts to meet 
and discuss the status, characteristics, and spiritual role of old growth 
forests. There appears to be no clear agreement on a single definition 
of Old Growth - various criteria used include pre-setthement origins, 
multi-aged stands of significant age, as well as stands of very 
large trees. Some of the more interesting talks from the first day of the 
conference included the description of old-growth hemlock stands in 
Massachusetts (mostly less than 25 ha, on steep slopes); a discussion of 
extensive tracts of high-elevation old-growth forest in the Catskills, 
where areas of over 100 square miles were never logged; a description 


106 Rhodora [Vo l. 96 


of a 130-yr old even-aged spruce stand on Mt. Graylock, regenerated 
after fire; a photo of the oldest spruce in Maine (450 years, 24-inch 
diameter); and the observation that forests of Michigan’s Upper Penin- 
sula, although never logged, are not stable in composition - these are 
stl changing as beech continues to extend its range and dominance 
northward. 


Biodiversity Principles and Applications, a Conference For Natu- 
ral Resource Professionals, was held on January 17, 1995 in Newington 
NH and sponsored by the UNH Cooperative Extension Program. Speak- 
ers from academic institutions, the Nature Conservancy, and state and 
federal agencies addressed the definitions and levels of biological di- 
versity, with applications to land management issues in Northern New 
England. The talks summarized here were of particular interest to bota- 
nists. 

C. Cogbill emphasized external, abiotic factors as the major cause of 
change and the maintenance of diversity in forests, ranging from wind 
as both a cause of “fir waves” at high elevations and the major cause of 
change 1n northern hardwoods forest, to fire as a factor in the regenera- 
tion of NH pitch pine forests. A common pattern of vegetation change 
is that of an even-aged stand of white pine, established 1n an old field or 
following fire, catastrophically destroyed by windstorm, and replaced 
by northern hardwoods - emphasizing that our majestic old-growth white 
pine stands are temporary features of the landscape. J. Litvaitis dis- 
cussed the connection between human alteration of the landscape and 
wildlife biodiversity. Early successional forests are critical habitat types 
for the New England cottontail and bobcat. This habitat type peaked in 
abundance in NH from 1905 through 1950, following the abandonment 
of agriculture. Abundances of these species and early successional birds 
have declined dramatically since 1950 as forests have matured. He also 
discussed the relationship of other landscape characteristics to small 
mammals. J. Kantor noted that the now-rare upland sandpiper moved 
into New England as agriculture expanded. Its current decline is due to 
habitat loss as pastures and agricultural lands revert to forest This raises 
some interesting questions as to the level of effort that should be ex- 
pended to preserve the upland sandpiper and similar species. 


BOOKS OF NOTE 


One Hundred and One Botanists by Duane Isely (lowa State Press, 


1995 Rhodora News/ Notes 107 


272 pages, 1994). Brief biographical sketches of prominent botanists 
rom Aristotle onward, illustrated with photographs front the Hunt 
Institute collection. Highly recommended by Les Mehrhoff. 


From Coastal Wilderness to Fruited Plain by Gordon G. Whitney 
(Cambridge University Press, 451 pages, 1994). An outstanding sur- 
vey of pre-settlement vegetation types of eastern North America, the 
progression and effects of land cleating for agriculture and wood, the 
effects of farm abandonment, current anthropogenic vegetation types, 
changes in the fauna, and the history of the conservation movement and 
land protection in the northeast. Although covering an extensive geo- 
graphical area, the book comprehensively treats the major topics and 
challenging issues of human effects on the natural vegetation and eco- 
systems of the northeast - and is worth buying for the 105-page bibliog- 
raphy alone. Itis definitely academic in its tone, and some readers may 
not find it as easily readable or accessible as Cronon’s Changes in the 
Land. The book follows the format of a scientific research paper and 
the reader may want to approach it in the same way - I found the intro- 
ductory chapter on methods of assessing pre-settlement vegetation types 
educational, but rapidly skipped ahead to the more interesting results of 
Whitney's investigations. 


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 ees 
For the next calendar year eran 


Name 


Address 


City & State 51, pee eee 


Special interests (optional): 


INFORMATION FOR CONTRIBUTORS TO RHODORA 


Submission of a manuscript implies it is not being considered for publi- 
cation simultaneously elsewhere, either in whole or in part. 
Manuscripts should be submitted in triplicate (an original and two xero- 
graphic copies), in addition, a copy of the mss. on a 3.5 inch floppy disc 
will facilitate production of the paper The text must be double-spaced 
(at least 3/8") throughout, including tables, figure legends, and literature 
citations. The list of legends for figures and maps should be provided on 
a separate page. Do not use footnotes. Do not indicate the style of type 
through the use of capitals or underscoring, 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 revi- 
sions and similar papers should be prepared in the format of"A Mono- 
graph 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 is distinguished from its congeners. Papers of a flo- 
ristic 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, Rhodora 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 ab- 
breviations 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 wnting 
style, consult the CBE Style Manual, Sth 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 indi- 
cate by blue pencil the intended limits of the figures. (Some "turn-page" 
figures with brief legends will be 31/2 x 6 in.) Magnification/reduction 
values given in text or figure legends should be calculated to reflect the 
actual printed size. Tables should be set up in the same way. Each page 
should be ready for reproduction and will be reduced to + x 6 inches.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. 


Lek, Wanted 
a? is 


3 


ee 


% 


RHODORA 


JOURNAL OF THE 


NEW ENGLAND BOTANICAL CLUB 


CONTENTS: 


Synopsis of the genus Arabis (Brassicaceae) in Canada, Alaska and Green- 
land. Gerald A. Mulligan 


A natural hybrid of Drosera anglica Huds. and Drosera linearis Goldie in 
Michigan. Donald E. Schnell 

The chromosome number of Saxifraga eae sis Fernald. Camille Gervais, 
Norman Dignard, and Rosaire Tra 


Book Review 

Asteraceae, Cladistics & Classification. Gordon P. DeWolf, Jr. ......... 
Rhodora News and Notes. Lisa A. Standley 
NEBC Membership Form 
NEBC Officers and Council Members 


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


109 


164 


22 Divinity Avenue, Cambridge, Massachusetts 02138 


Vol. 97 Spring, 1995 No. 890 


The New England Botanical Club, Inc. 


22 Divinity Avenue, Cambridge, Massachusetts 02138 


RHODORA 


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


Associate Editors 
DAVID S. CONANT 
LISA A. STANDLEY 


RHODORA (ISSN 0035-4902). Published four times a year (January, 
April, July, and October) by The New England Botanical Club, 810 
East 10th St., Lawrence, KS 66044 and printed by Allen Press, Inc., 
1041 New Hampshire St., Lawrence, KS 66044. Second class postage 
pending at Lawrence, KS. POSTMASTER: Send address changes to 
RHODORA, P.O. Box 1897, Lawrence, KS 66044. 

RHODORA is a journal of botany devoted primarily to North America. 
Authors are encouraged to submit manuscripts of scientific papers and 
notes relating to the systematics, distribution, ecology, paleobotany, 
or conservation biology of this or floristically 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, PO. Box 1897, Lawrence, KS 
66044-8897, 

MEMBERSHIPS: Regular $35; Family $45. Application form printed 

erein. 

BACK VOLUMES AND SINGLE COPIES: Some available; poe 
tion and prices will be furnished upon request to the Treasure 

ADDRESS CHANGES: In order to receive the next number of RHO- 
DORA, changes must be received by the business office prior to the 
first day of January, April, July, or October. 

INFORMATION FOR CONTRIBUTORS: Inside back cover, January 

and April. 


MANUSCRIPTS: Send to: Gordon P. DeWolf, Jr. 
Managing Editor, RHODORA 
125 Long Hill Road 
West Brookfield, MA 01585 


This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). 


RHODORA 


JOURNAL OF THE 
NEW ENGLAND BOTANICAL CLUB 


Vol. 97 April 1995 No. 890 


RHODORA, Vol. 97, No. 890, pp. 109-163, 1995 


SYNOPSIS OF THE GENUS ARABIS 
(BRASSICACEAE) IN CANADA, 
ALASKA AND GREENLAND 


GERALD A. MULLIGAN 


ABSTRACT 


This taxonomic treatment of Arabis (Brassicaceae) in Canada, Alaska and 
Greenland recognizes 30 species. A comprehensive key is provided for these 
species and 8 varieties. Four new cpa are described: A. boivinii G. Mulligan, 
sp. nov., A. calderi G. Mulligan, sp. nov.; codyi G. Mulligan, sp. nov.; and A. 
murrayl, G. Mulligan, sp. nov. In addition, many other taxa are recondad for the 
first time for this area. Cytological eae available for 45 North American and 
Greenland species of Arabis are summarized and discussed. Thirty-six of these 
species have the basic chromosome-number of x = 7 and some have diploid, 
triploid and even tetraploid chromosome races. The Arabis species being treated 
here can reproduce sexually and/or by agamospermy. 


Key Words: Brassicaceae, Arabis, taxonomy, new species, cytology 


INTRODUCTION 


The first treatment of the genus Arabis in North America was 
on taxa in the Pacific Northwest (Rollins, 1936a). This was closely 
followed by two monographs published in Rhodora: Arabis in 
eastern and central North America by Hopkins (1937) and a 
monographic study of Arabis in western North America by Rollins 
(1941). Rollins discussed, in detail, the relationships of Arabis 
with other genera in the family Brassicaceae and among and with- 
in Arabis species. He continued to add to our knowledge of the 
taxonomy, ecology, cytology, breeding systems, hybridization and 
speciation in the genus in a long series of publications (Rollins, 
1936a, 1936b, 1941, 1943, 1946, 1966, 1971, 1973, 1981, 1982, 


109 


110 Rhodora [Vol. 97 


1983, 1984, 1993a; Rollins and Riidenberg, 1971, 1977, 1979). 
Bocher (1947, 1951, 1954, 1966, 1969) studied the cytology and 
embryology of some members of Arabis occurring in this area 
and Mulligan (1964) and Mulligan and Porsild (1969, 1970) pub- 
lished chromosome numbers for many Arabis taxa in Canada. 
Boivin (1951, 1955, 1967) also proposed a number of new taxa 
and rankings based on his study of Canadian Arabis. He sum- 
marized his taxonomic views on Arabis in his Enumération des 
plantes du Canada and Flora of the Prairie Provinces (Boivin, 
1966, 1968, respectively). More recently, Sabourin (1989) pre- 
sented a useful guide to the Arabis taxa found in eastern Canada. 
Rollins (1993b) published the first comprehensive taxonomic study 
of the genus Arabis in Continental North America. He included 
excellent keys, descriptions, habitats, distributions, and cytolog- 
ical data for all taxa based on material that he had seen. However, 
he did not have the opportunity to see the large amount of Ca- 
nadian, Alaskan and Greenland Arabis material available to me. 
Consequently, many Arabis taxa present in Canada, Alaska and 
Greenland are not recorded from this area in Rollins’ latest treat- 
ment. The present synopsis of Arabis occurring in Canada, Alaska 
and Greenland attempts to fill this gap. 


MATERIALS AND METHODS 


Herbarium specimens, including many types, were examined 
morphologically. They were borrowed from the following insti- 
tutions: ALA, CAN, DAO, DWC, GH, K, KY, LE, MO, MT, ND, NY, ORE, 
OS, P, PH, QFA, RM, RSA, SASK, UBC, UC, US, WIS and ws. Herbarium 
acronyms are according to Holmgren et al. (1990). In addition, 
a very large collection of unmounted, mostly unidentified, Arabis 
specimens were studied and later deposited in the herbarium of 
Agriculture Canada, Ottawa (DAO). These specimens were col- 
lected in southwestern Canada and the western United States by 
Theodore Mosquin, with Linda Mosquin and M. H. Benn in 1962 
and with G. A. Mulligan and J. M. Gillett in 1963. Many of these 
specimens had both herbarium labels and annotation slips con- 
taining unpublished cytological data of Theodore Mosquin. In 
addition, cytological studies published for all North American 
and Greenland species of Arabis were summarized. The cytolog- 
ical information appears in Table 1. 


1995] Mulligan — Arabis Ltt 


CYTOLOGY AND BREEDING SYSTEMS 


Results from the cytological examination of 45 species of Arabis 
of North America and Greenland are presented in Table |. Thirty- 
six species have the base chromosome-number of x = 7, eight 
species have x = 8, and one species, A. glabra, seems to have two 
base numbers, x = 6 and 8. 

Nineteen of the species with the base number of x = 7 are 
known only as diploids, five species only as triploids, one species 
only as a tetraploid, rine species as both diploids and triploids, 
and two species as diploids, triploids and tetraploids. Seventeen 
of these thirty-six species, with the base number of x = 7, contain 
plants that are triploid and/or have an irregular meiosis. Since 
these plants have almost a complete seed set, it seems likely that 
they produce seed by apomixis. Eight species (A. columbiana, A. 
divaricarpa, A. drummondii, A. exilis, A. holboellii, A. laevigata, 
A. puberula and A. sparsiflora), also with a very high seed set, 
have plants with a regular pairing at meiosis and thus may very 
well be sexual. However, all of these eight species also have plants 
that are triploid and/or have an irregular meiosis and thus are 
almost certainly also apomictic. A number of species, with the 
base number of x = 7, include aneuploids and plants with B-chro- 
mosomes. 

Bocher (1947, 1951, 1954, 1966, 1969) first reported apomixis 
in North American and Greenland Arabis with the base number 
of x = 7. According to Bécher (1969), some diploids with a com- 
pletely normal meiosis seem to be sexual whereas others with 
meiotic abnormalities are probably apomictic or amphi-apomic- 
tic. He also stated that triploids may easily be formed by the 
fusion of reduced and unreduced cells in diploids and that this 
process may be reversible. The most common base number for 
North American and Greenland Arabis is x = 7, and these species 
may be sexual, apomictic or amphi-apomictic or any combination 
thereof. It appears that plants with the base number x = 7, some 
possibly propagating apomictally, do occur occasionally on other 
continents. For example, Berkutenko and Gurzenkov (1976) re- 
ported 2n = 14 for Arabis falcata (Turcz.) Berkut. and 2n = 21 
for Arabis pendula L. from the south Magadan region in the former 
U.S.S.R. and Galland (1969) 2” = 14 for plants of Arabis auri- 
culata Lam. and Arabis conringioides Ball. from Morocco. Arabis 
pauciflora (Grimm.) Garcke has the chromosome number of 2” 


Table 1. Reported chromosome-numbers for Arabis of Canada, the United States and Greenland. Chromosome counts and other cytological 
information attributed to voucher specimens under the collection numbers of Theodore Mosquin (T.M.) were observed by ere 
attached to specimens that are now deposited in the Agriculture Canada herbarium at Ottawa (DAO). The references given in Table 1 | refer 
to published reports. The chromosome bers listed as 7 = 15/2, 21/2, and 22/2 indicate that Mosquin observed very irregular configurations 
totalling 15, 21 and 22 chromosomes, respectively, at meiosis. 


Sources of materials, vouchers or references, 


Taxon n 2n and special cytological information 

A. aculeolata Greene 32 OREGON (Rollins & Riidenberg 1977). 

A. alpina L. 8 16 GREENLAND (Jorgensen et a/. 1958; Dalgaard 1988, 
81). QUEBEC (Rollins 1941; Mulligan 1964). MANI- 
TOBA (Love & Live | ). 

A. arenicola (Richardson) Gelert 8 16 GREENLAND (Bocher 1966). QUEBEC (Hedberg 1967; 
Lepage 39394 by T.M.). 

A. boivinii G. Mulligan 21/2 MONTANA (7.M. & L. Mosquin 5219). SOUTH DA- 
KOTA (T.M. & G.A. Mulligan 5157). 

A. breweri S. Wats. 14 CALIFORNIA (Rollins & Riidenberg 1971). 

A. canadensis L. 14 ONTARIO (Mulligan 1964). 

A. caucasica Willd. 8 ONTARIO (Mulligan 1964, 811). 

A. cobrensis M.E. Jones 7 WYOMING (Rollins 1941). 

A. columbiana Macoun 7 BRITISH COLUMBIA (T.M. & G.A.M. 4930 & 4937, 


meiosis irregular). CALIFORNIA (7.M. 4548). IDAHO 
AM. 4966, 4982 & 4994, meiosis irregular; 
T.M. & G.A.M. 4998, TII). OREGON (T.M. & G.A.M 


21/2 ca. 21 CALIFORNIA (T.M. 4531 & 4535). WASHINGTON 
M. & G.A.M. 4944, n = 21/2). 


cll 


evIopoyYy 


L6 ‘19A] 


Table 1. 


Continued. 


Taxon 


Sources of materials, vouchers or references, 
and special cytological information 


A. constancei Rollins 
A. crandallii Robinson 
A. demissa var. russeola Rollins 


A, depauperata Nelson & Kennedy 


WASHINGTON (7.M. & G.A.M. 4946). 
CALIFORNIA (Rollins & Riidenberg 1971). 
COLORADO (Rollins 1941, 1966). 
WYOMING (Rollins 1966). 

MONTANA (7.M. & J.M. Gillett 5253). 


[S661 


MANITOBA (Love & Love 1975a, 1982). COLORADO 
(Rollins 1966). MONTANA (Rollins 1966, 1983). 
22 CALIFORNIA (Rollins 1966). WYOMING (Bécher 
1969, apomictic, EMCs, PMCs, and pollen grains had 
22 chromosomes and Pie cn aes pollen were 
formed). The n = 8 and n 
.O. Gaiser in a (1941) are probably 
erroneous (Rollins, pers. comm.). 


SASKATCHEWAN (Taylor & Brockman 1966). ALBER- 
TA (7.M., L.M. & M.H. Benn 4751 & T.M. & L.M. 
4703, 711). BRITISH COLUMBIA (Mulligan 1964, 
2n = 13 + 2B & 14). COLORADO (7.M. & L.M. 


A. divaricarpa A. Nels. var. divaricarpa or dacotica 7 14 
(Greene) Boivin 


number 


A. divaricarpa var. divaricarpa 7 14 


15/2 MONTANA (T.M. & J.M.G. 5258). 

ALBERTA (T.M. & M.H.B. 5183, n = 21/2). BRITISH 
COLUMBIA (Mulligan 1964). CALIFORNIA (7.M. 
4529), 

COLORADO (T.M. & L.M. 4575) 

BRITISH COLUMBIA (Mulligan 1964). COLORADO 
(T.M. & G.A.M. 5066 & 5073, n = 21/2). 


A. divaricarpa var. dacotica ca. 7 
21/2 2] 


SIqDdp — uUesITINIA 


ell 


Table 1. 


Continued. 


Taxon n 


Sources of materials, vouchers or references, 
and special cytological information 


A. drummonadii Gray 7 


Ze 


A. eschscholtziana Andrz. 32 
A. exilis A. Nels. 7 


ca. 21 


MACKENZIE DISTRICT (Mulligan 1964, 2n = 28 & 

28 + 1B). MANITOBA (7.M. & L.M. 4904 & 4910, 
n= 14 with irregular meiosis). 

YUKON (Mulligan 1964). MANITOBA (Live & Live 
1982). ALBERTA (7.M. & L.M. 4706 & T.M., L.M. & 
M.H.B. 4737, both 711). CALIFORNIA (Rollins & Rii- 
denberg 1977). COLORADO (Rollins 1941, 1966; Rod- 
man & Bhargava 1976; 7.M. & L.M. 4580, 4616 & 
4630, all 71]; T.M. & G.A.M. 5068, 5093 & 5101, all 
7II). IDAHO (T.M. & G.A.M. 4999), MONTANA (Rol- 
lins 1966, 1983; 7.M. & J.M.G. 5232, 5232a & 5260, 
all 7II). UTAH (7.M. & G.A.M. 5120, 71). WYO- 
MING (Rollins 1966, 1983; 7.M. & L.M. 4658, 4663 & 
4668, all 711; T.M. & J.M.G. 5246 & 5247, both 7II). 

BRITISH COLUMBIA (Mulligan 1964). 

ALBERTA (T.M. & L.M. 4755, 711 + 71; T.M. & M.H.B. 
4738). 

MASSACHUSETTS (Bécher 1969). 


BRITISH COLUMBIA (Taylor & Mulligan 1968, 32II). 


YUKON (Bocher 1969). BRITISH COLUMBIA (Mulli- 
gan 1964; Taylor & Taylor 1977, 71). MONTANA 
(Bécher 1969). WYOMING (Rollins 1966; Bécher 


1969). 
UTAH (T.M. & G.A.M. 5122). 


vil 


vIOpOyY 


L6 1OA] 


Table 1. 


Continued. 


Taxon 


2n 


Sources of materials, vouchers or references, 
and special cytological information 


A. fendleri (S. Wats.) Greene 


A. glabra (L.) Bernh. 


A. glaucovalvula M.E. Jones 
A. gunnisoniana Rollins 


A. hirsuta (L.) Scop. var. hirsuta 


A. hirsuta var. pycnocarpa (Hopkins) Rollins 


14 


CALIFORNIA (Rollins & Riidenberg 1979). COLORA- 
DO (Rollins 1941, 1966). The n = 14 report for Colora- 
do material in Rollins (1941) may be erroneous (Rol- 
lins, pers. corr.). 

NEVADA (Rollins & Riidenberg 1979). The n = 21 re- 
port for pesaalian material in Rollins (1941) is almost 
certainly erroneo 


QUEBEC (Gillett 10569, 2n = 12 by T.M.). ONTARIO 
(Bowden, 2n = 12 by G.A.M.). ALBERTA (7.M. & 
L.M. 4687). CALIFORNIA (7.M. 4542, 4515B & 
4815, all 61; Breedlove, 611 by T.M.). COLORADO 
(Rodman & Bhargava 1976; T.M. & G.A.M. 5104, 6Il). 
IDAHO (T.M. 4795). MONTANA (7.M. & J.M. 

5218, re regular), OREGON (7.M. & L.M. 4359 
& 4484, 611). WYOMING (7.M. & L.M. 4669 61). 
VIRGINIA (Hill 1982). 


CALIFORNIA (Rollins & Riidenberg 1979). 
COLORADO (Rollins 1941). 


CALIFORNIA as 4525, 1611). COLORADO (T.M. & 
J.M.G. 5347, 16II). 

MACKENZIE DISTRICT (Mulligan 1964). MANITOBA 
(Taylor & Brockman 1966). ALBERTA (T.M. & L.M. 
4688 & 4701, 16; 7.M. & M.H.B. 5184, 1611). BRIT- 


SIGDAP — UeSIT[I NA [S661 


SI 


Table 1. 


Continued. 


Taxon 


2n 


Sources of materials, vouchers or references, 
and special cytological information 


A. holboellii Hornem. var. holboellii 


A. holboellii var. consanguinea (Greene) G. Mulli- 


21/2 


a | 


ISH COLUMBIA (7.M. & G.A.M. 4933, 1611). CON- 
NECTICUT (Rollins 1941). COLORADO (Rollins 


941). 

SOUTH DAKOTA (T.M. & G.A.M. 5154 & 5155, 16II). 
WISCONSIN (Smith 1938, 16II). The n = 32 chromo- 
some-number by L.O. Gaiser in Rollins (1941) should 
be discounted until it is confirmed. 


GREENLAND (Bocher 1954; Bécher 1969, some have 
completely normal meiosis and seem to be sexual 
whereas others probably are amphiapomictic or apo- 
mictic; Dalgaard 1988, 7II). 

GREENLAND (Bocher 1954; Bécher 1969, metaphases 
correspond to those of second division, PMCs have as- 
yndetic metaphase plates with the somatic number; 
Hansen et al. 2304, 2n = ca. 21 by T.M.). 


NEVADA (7.M. & L.M. 4335 & 4337, 711). 


ALBERTA (T7.M. & L.M. 4696 & 4705; T.M. 5201, 
n = ca. 21/2, meiosis irregular; T7.M. & M.H.B. 5212, 
n = ca. 21/2). CALIFORNIA (T.M. & P. Raven 4419, 
T.M. & L.M. 4440 & 4441, n = 21/2). COLORADO 
(7.M. & L.M. 4623; T.M. & G.A.M. 5094). OREGON 
(T.M. & L.M. 4485, T.M. & J.M.G. 5276). 


OTT 


eIOpOyYy 


L6 IO] 


Table 1. 


Continued. 


Taxon 


2n 


Sources of materials, vouchers or references, 
and special cytological information 


A. holboellii var. retrofracta (Graham) Rydb. 


14 


14+ 1B 


MANITOBA (Mulligan 1964; Léve & Love 1982; 7.M. & 


L.M. 4911, 711). SASKATCHEWAN (Mulligan 1964; 
T.M. & L.M. 4914 & 4915, both 711). ALBERTA (Mul- 
ligan 1964; T.M. & L.M. 4683 & 4686, both 7II; TM. 
& G.A.M. 4924, 711). BRITISH COLUMBIA (Mulligan 
1964; T.M. & L.M. 4454 & 4460, both 7II; JA. Calder 
& J.M.G. 26537, 71 by T.M.; T.M. & G.A.M. 4935 & 
4938, both 71]; 7.M. & G.A.M. 4929, n = 7, meiosis 
irregular). CALIFORNIA (Rollins 1941, 1966; Rollins 
& Riidenberg 1977; T.M. & P. Raven 4427, 711; 7.M 
4532: T.M. 4530, n =7 meiosis irregular). COLORA- 
DO (Rollins 1941; 7.M. & L.M. 4628, 711). IDAHO 
(Bécher 1969, meiosis regular with 7 bivalents, hun- 
dreds of anaphase I were normal, a tetrads were 
formed and pollen was uniform; 7.M. & G.A.M. 4967, 
4970, 4972, 4997 & 4998, all 711). “MONTANA (Rol- 
lins 1966, 1983). NEVADA (7.M. & G.A.M. 5001 & 
5008, both 7II; 7.M. & G.A.M. 5013, n = 7, irregular 
meiosis). OREGON (7.M. & G.A. - 4962, TI). 
SOUTH DAKOTA (7.M. & G.A 5151, n= 7, irreg- 
ular meiosis). WASHINGTON ae 1969). WYO- 
MING Rollins 1941, 1983; Bécher 1969; 7.M. & L.M. 
4665, 70 


, ). 
ALBERTA (Packer 1964). BRITISH COLUMBIA (Mulli- 


[S661 


SIGDAP — UeBIT[NJA 


— 


Table 1. 


Continued. 


Taxon 


2n 


Sources of materials, vouchers or references, 
and special cytological information 


A. holboellii var. secunda (Howell) Jepson 


A. inyoensis Rollins 


A. kamtschatica (Fisch.) Ledeb. 


15/2 


ca. 21/2 


ca. 22/2 


ALBERTA (7.M., L.M. & M.H.B. 4735, 711 +11). BRIT- 
ISH COLUMBIA (T.M. & G.A.M. 4932). COLORADO 
(T.M. & G.A.M. 5097, irregular meiosis). 

ALBERTA (7.M. & G.A.M. 4922, n = ca. 21/2). BRIT- 
ISH COLUMBIA (T.M. & L.M. 4452, n = ca.21/2, 
meiosis irregular). IDAHO (7.M. & L.M. 4156, n = ca. 
21/2, PMCs form dyads not aera OREGON (T.M. 
& L.M. 4353). WYOMING (T.M. & L.M. 4164, 

n = 21/2). The n = 14 chromosome-number given in 
Rollins (1941) may be erroneous. 
QUEBEC (Bécher 1954), IDAHO (T.M. & G.A.M. 4983, 
II). 


BRITISH COLUMBIA rire 1964). 
UTAH (7.M. & G.A.M. 5119). 

UTAH (7.M. & ae 988). 

NEVADA (Rollins & Riidenberg 1971). 

NEVADA (Rollins & Riidenberg 1971). 

ALASKA (Dawe & Murray 1979, 2n = 32 as on voucher 


in ALA not 2n = 16 as in paper; Rollins 1966, under 4. 


lyrata). YUKON (Mulligan 1964, under A. /yrata). 
BRITISH COLUMBIA (Taylor & Mulligan 1968, Mul- 
ligan 1964, under A. /yrata). The GH voucher for 4 
lyrata ssp. kamtschatica of Johnson & Packer 1968 is 
A, media 


BIT 


elopoyuYy 


L6 ‘19A] 


Table 1. Continued. 
Sources of materials, vouchers or references, 
Taxon n 2n and special cytological information 

A. laevigata (Mihl) Poir. 7 CONNECTICUT (Rollins 1941). MARYLAND (Kovan- 
da 1978). OHIO (Easterly 1963, 71D). WISCONSIN 
(Smith 1938, 7II). 

A. lemmonii S. Wats. 14 ALBERTA (Mulligan 1964). BRITISH COLUMBIA 
(Mulligan 1964). WYOMING (Rollins 1966). 

6a, 2172 MONTANA (7.M. & J.M.G. 5229 & 5252). 

A. lignifera A. Nels. 7 14 COLORADO (Rollins 1941). WYOMING (Rollins 1941). 

A, lyallii 8. Wats. 21/2 2 ALBERTA (Mulligan 1964). BRITISH COLUMBIA 
(Mulligan 1964). UTAH (7.M. & J.M.G. 5322, n = 21/ 
2) 

A. lyrata L. 8 16 ONTARIO (Bocher 1969; Garton 6191, 8II by T.M.). 

A. media N. Busch 16 ALASKA (Rollins 1966 under A. /yrata but voucher in 
GH is A. media); Johnson & Packer 1968 under A. /y- 
rata subsp. kamtschatica but voucher in GH is A. me- 
dia 

A. microphylla Nuttall 7 14 BRITISH COLUMBIA (Mulligan 1964). OREGON (Rol- 
lins 1941). 

15/2 15 WYOMING (Bocher 1969, PMCs form dyads and ana- 
phases initiating dyads were all regular with 15 chromo- 
4 chromosome-number in Rollins 
(1941) may be erroneous. 
A. parishii 8. Wats. 7 CALIFORNIA (Rollins & Riidenberg 1979). 


[$661 


siqnap —uesiy[njA 


611 


Table 1. Continued. e 
Sources of materials, vouchers or references, = 
Taxon n 2n and special cytological information 
A. pendulina Greene 7 14 UTAH (Bécher 1969, pollen was uniform). 
A. perennans §. Wats. 7 14 COLORADO (Rollins 1941). ARIZONA (Rollins & Rii- 
denberg 1971). 
A. perstellata Braun 7 TENNESSEE (Rollins 1966). 
A. petiolaris (Gray) Gray 14 ca. 28 TEXAS (Rollins & Riidenberg 1977). 
A. pinetorum Tidestrom 7 14 ALASKA (Dawe & Murray 1979 under A. holboellii but 
voucher in ALA is 4. pinetorum). CALIFORNIA (T.M. 
7, n= 7, meiosis irregular). z 
13 + 2B MANITOBA (Mulligan 1964 under 4. holboellii). 4 
21/2 21 CALIFORNIA (7.M. & L.M. 4439; Rollins & Riidenberg g 
1971). NEVADA (T.M. 4536). WYOMING (Rollins S 
1966). 
A. puberula A. Nels. 7 NEVADA (T.M. & L.M. 4337, TI). 
21/2 OREGON (7.M. & L.M. 4347 & 4349). 
A. pulchra M.E. Jones 7 14 NEVADA (Rollins & ats 1979), 
UTAH (Rollins & Riidenberg 1971). 
21 NEVADA (Rollins & Ridenbers 1979). 
UTAH (Rollins & Riidenberg 1971). 
A. repanda S. Wats. a CALIFORNIA (Rollins 1941). = 
< 
A, schistacea Rydb. 14 UTAH (Bécher 1969, pollen is uniform). e 
A. selbyi Rydb. ca. 21 COLORADO (Rollins & Riidenberg 1977). = 


Continued. 
Sources of materials, vouchers or references, 
Taxon n 2n and special cytological information 
A. serotina Steele 14 VIRGINIA (Wieboldt 1987). 
A. sparsiflora Nuttall a ARIZONA (7.M. & L.M. 4245, meiosis irregular). IDA- 
HO (7T.M. & L.M. 4338 & 4352, both 7II). 
22/2 22 CALIFORNIA (Raven et al. 1965; Bécher 1969, 7II + 8I 
at metaphase I). 
7 OREGON (Rollins & Riidenberg 1977). 


A, subpinnatifida S. Wats. 


SIGDAP — UeBIT[ NA [S661 


rae 


122 Rhodora [Vol. 97 


= 14 in Austria, Sweden, France and Germany (Burdet, 1967). 
However, he does not think that it is related to North American 
Arabis with the base number x = 7. He points out that this species 
has baffled European taxonomists, who have placed it at various 
times in the genera Turritis, Brassica, Conringia and even Ery- 
simum 

The eight North American species of Arabis with the base num- 
ber of x = 8, and A. glabra, with x = 6 and 8, all appear to produce 
seed sexually. All of the material, with these base numbers, ex- 
amined by the author and by other workers on North American 
material and plants from continents, had a regular meiosis and/ 
or lacked triploids, and, with the exception of A. caucasica, had 
a very high seed set when found growing in isolation. They are, 
therefore, almost certainly sexual and self-compatible. Burdet 
(1967) considers x = 8 to be the common basic chromosome- 
number of European and Asiatic Arabis. He does suggest that the 
base number may even be x = 4 because of chromosome-counts 
of 2n = 8 for plants of A. hirsuta from Switzerland and France. 
Burdet (1967) states that the somatic chromosomes of A. glabra 
are quite different from those of other Arabis, supporting other 
evidence that it perhaps could be placed in another genus. 


TRICHOMES IN ARABIS 


The trichomes on the undersurfaces of the caudex leaves, when 
present, often differ from taxon to taxon. They vary from simple 
or once-forked to dendritic or stellate and the more complex ones 
are from sessile to long-stalked. Scanning electron microscope 
(SEM) photographs of the more complex trichomes are shown in 
Figures 1 to 20. Where the trichomes are similar, a close rela- 
tionship may be indicated. This seems to be the case for A. drum- 
mondii, A. calderi, A. lyallii, A. divaricarpa var. divaricarpa and 
A. divaricarpa var. dacotica (Figures 3 to 7, respectively). In most 
cases, the trichomes are very different in form and/or size; for 
example A. caucasica (Figure 2), A. microphylla (Figure 13), A. 
sparsiflora (Figure 18) and A. pinetorum (Figure 20). These dif- 
ferences in the morphology and size of trichomes on the under- 
surfaces of the caudex leaves have been used extensively in the 
key to separate taxa. 


1995] Mulligan—Arabis 123 


Figures hare photographs of Arabis trichomes; all x 150. Figure 1, A. 

alpina. Figu , A. caucasica. Figure 3, A. drummondii. Figure 4, A. ae 
Figure 5, A. a Figure 6, 4. divaricarpa var. divaricarpa. Figure 7, A. divaricarpa 
var. dacotica. 


TAXONOMIC TREATMENT 


KEY TO TAXA OF ARABIS OF CANADA, ALASKA AND GREENLAND 


1. Bases of middle cauline leaves all attenuate, cuneate, obtuse, to truncate, never clasping stems .......... Z 
2, Siiaues Stronely destendme 16 DENGUIOUS. 4c c.mccckda sis ew nssw's oon Hi Se ee eRe eee oe eee nay hee 3 
3. Stems 3 to 9 dm high; biennials, usually lacking caudex leaves as plants mature; caudex leaves, if present, 
glabrous or with simple to once-branched trichomes to 0.5 mm long; siliques 2.0 to 3.25 mm wide, 

seeds prominently winged; sw Que., s Ont., and southward ................... 11. A. canadensis 

3. Stems usually less than 3 dm high; perennials with persistent caudex leaves; surfaces of caudex leaves 
with short-stalked (less than 0.063 mm long) semistellate trichomes mostly 0.125 mm wide; siliques 

1.5 to 2.0 mm wide; seeds only slightly winged; Yukon, Sask. (Cypress Hills), sw Alta., B.C., and 

COMM at: earn andi ed cater nagt esas aoeoe needed eens yess ere eo meer ee eraes 22. A. exilis 

2. Sitges Bsvendiie WOCCCl. x. ce eceu cused ee cikeeeeee ate aise ees es ees eee nee beers e eran aes 4 
4. Siliques strongly ascending to erect: surfaces of caudex leaves glabrous or with short-stalked (less than 
0.063 mm long) semidendritic to dendritic trichomes mostly 0.125 mm wide; sw Yukon, s B.C., and 


BOUL <a hace coc eee gt cain oh Lea eh Se See ee Be eee eee es 21. A. murrayl 
4. Siliques ascending; surfaces of caudex leaves with simple or medium- to long-stalked (0.063 mm long 
or longer) 1- to 2-branched, -forked or -rayed trichomes from 0.25 to 1.5 mm long or wide ..... 2) 


5. Some caudex leaf blades with a larger terminal segment and two to many much smaller lateral segments 
Dipeieevee ciate 1 ( M6. 6: i ee eee ee eee er eee eee ee ee eee ee 6 

6. Petals 6.0 to 8.0 mm long; siliques 0.75 to 1.0 mm wide; beaks of siliques 0.5 to 1.0 mm long, 
much longer than wide; surfaces of caudex leaves with few to many simple trichomes 0.75 to 

1.25 mm long; larger terminal segments of caudex leaves usually ovate; Ont., Man., Sask., Alta., 

ANG SOUMIWEIC cntcaceacnc sesh wneeeree dhe & bein es or ne Weeee ere ease ers 4. A. lyrata 


Cl 


BIOpoyuYy 


L6 ‘19A] 


6. Petals (4.0) 5.0 to 5.5 mm long; siliques 1.25 to 1.5 mm wide; beaks of siliques usually less than 
0.5 mm long, shorter to slightly longer than wide; surfaces of caudex leaves usually glabrous, 
rarely with few to scattered 2-forked or -rayed trichomes to 0.75 mm long; larger terminal 
segments of caudex leaves usually broadly ovate to orbicular; Alaska, westward into Russia, 
Yukon, Mack. Dist., n Sask., sw Alta., B.C., and Wash. ................ 5. A. kamtschatica 

5. All caudex leaf blades entire, sparingly toothed or with one to few pairs of shallow to deep lobes; 
none of leaf blades with a larger terminal segment and smaller lateral segments or prominent lobes 


7. Petals (4.5) 7.0 to 8.0 mm long; outer sepals prominently saccate at base; surfaces of caudex leaves 

mostly with simple trichomes 0.75 to 1.5 mm long; Yukon (rare), sw Alta., se B.C., and southward 

pat eu ote eee eee pe es De ee een en te peewee 10. A. nuttallii 

7. Petals 4.0 to 5.5 mm long; outer sepals weakly saccate; surfaces of caudex leaves glabrous or 

mostly with branched or rayed trichomes less than 0.75 mm long or wide 8 

8. Siliques subterete with prominent midvein from base to apex, 0.75 to 1.0 (1.25) mm wide; 

surfaces of caudex leaves mostly with medium-stalked (0.063 to 0.125 mm long) 2- to 3-rayed 
trichomes; n Alaska, westward into Russia, and n Yukon ................... 2. A. media 

8. Siliques flattened, prominent midvein absent towards apex, 1.5 to 2.5 (3.0) mm wide; surfaces 

of caudex leaves glabrous or with simple and long-stalked (over 0.125 mm long) branched 

or rayed trichomes 9 

9. Plants glabrous, except for occasional simple trichomes on leaf margins; Mack. Dist., Keew. 

Dist., Frank. Dist., Greenland, Labrador, n Que., n Ont., and n Sask. ................ 

ea esr haste a eae alacant Wag. RG oe 3a. A. arenicola var. arenicola 

9. Plants with copious simple, 1- to 2-branched and long-stalked 2-rayed trichomes on stem 

and leaf surfaces; Keew. Dist., n Que., n Ont., n Man., and n Sask. .................. 

bo hee hem Pan aa Gua e estes be bea ge eee ean ees 3b. A. arenicola var. pubescens 


[S66 


siqnap — ues 


SCI 


1. Bases of middle cauline leaves always auriculate-, hastate-, to sagittate-clasping stems ................. 10 


10. Caudex leaves glabrous, or absent when plants mature ..............0 000 c cece cece ceeeeee 11 
Ll, SUNQUeS CLEA pressed 10 ACIS: ox sesu se uGu cd eurs enh eeo cheese) mesa datandd 24 RbRebLeaes 12 
12. Siliques subterete; petals yellow, about as long as sepals; biennials with basal leaves few or absent 
as plants mature; bases of stems with spreading trichomes; Alaska, Yukon, Mack. Dist., Que., 
Ont., Man., Sask., Alta., B.C., southward, Europe, and Asia .................... A. glabra 
eas Siliques strongly flattened: petals whitish to purplish, about twice as long as sepals; short- to long- 
ived perennials with caudex leaves persisting as plants mature; bases of stems glabrous or with 
malpighiaceous trichomes; Alaska, Yukon, Mack. Dist., Nfld., N.S., N.B., Que., Ont., Man., 


Sask., Alta., B.C., and southward ..........0 00.00.0000 cc cece cece 14. A. Aaapondi 

11. Siliques descending, spreading, ascending to strongly ascending .................--.0 cee ee eee 13 
13. Inflorescences secund; Yukon (rare), w Alta., B.C., and southward .......... 18. 4. lemmonii 

1.3: MMOLESCENCES SYINMICILIOAL occ cacalaaca was as ka bho ue haus oochen wectiaosdeveeeedue ee eees 14 


14. Siliques strongly ascending, (2.0) 2.25 to 3.5 mm wide; stems 1.0 to 2.5 (4.0) dm high; 
persistent perennials with much branched, many-stemmed, caudexes; sw Alta., s B.C., and 
ROUEN Agee gig caer aed aes Bs ie ed lose SOs ney ek eS Ee eb dpb aes mame ea 16. A. lyallii 
14. Siliques descending, spreading to ascending, 0.75 to 2.0 mm wide; stems (2) 3 to 10 dm high; 
biennials or short-lived perennials with basal leaves often absent as plants mature; single 
VO Ter Bie eo oer ia. ads ae a te ae ee em eae eee ween Pea e pa state wed ee 
15. Siliques up to 25 mm long and 0.75 mm wide, with a scattered simple to once-forked 
puberulence; petals up to 2.0 mm long; Essex Co. Ont., and southward ........... 
bi eeea Ghee eee a Eee as CO Ree eee re eae ped es adeeb os 13. A. shortii 
15. Siliques more than 25 mm long and 0.75 mm wide, glabrous, petals longer than 2.0 mm 
1 


pea 


elopoyy 


L6 190A] 


16. Cauline leaves more than 10 mm wide; basal leaves usually absent; siliques spreading 

and strongly downwardly arcuate; sw Que., s Ont., and southward ........... 

Ei Pebads Lhe eade Ge Oe poe Leese cates eues se eae geeees eee eerie 12. A. laevigata 

16. Cauline leaves mostly less than 10 mm wide; basal leaves mostly persisting; siliques 

descending, spreading to ascending, straight to slightly arcuate; Alaska (rare), 

Yukon, Mack. Dist., Que., N.B. (rare), Ont., Man., Sask., Alta., B.C., and south- 

WAid)cacuneeeoeecece es auaeee eee ees 17a. a Fas var. aes 
ID). «Caudex Jeaves. present Gnd Naley ibs os oo eee Obh ENGR Ee bde Da a SbES RSS Yeeedeeesdcten doers 

17. Surfaces of caudex leaves with simple trichomes only; siliques spreading and strongly downwardly 

arcuate; sw Que., s Ont., and southwar 2. A. laevigata 

17. Surfaces of caudex leaves with some forked or rayed trichomes; if siliques are spreading and strongly 
downwardly arcuate, the trichomes on surfaces of caudex leaves are all forked or rayed ...... 

18. Siliques all strongly descending, pendulous to downwardly-appressed to rachis ............ 19 

19. At least one-half of fruiting pedicels semigeniculate to geniculate at their bases ....... 20 

20. All fruiting pedicels strongly geniculate at their bases; siliques 1.0 to 1.5 mm wide; 

middle and upper cauline leaves revolute at edges; Alaska, Yukon, Mack. Dist., Que. 

(rare), Ont. (rare), Man., Sask., Alta., B.C., and southward ....................... 

Get Ge Ge ee ee ee ee nee eas 25c. A. holboellii var. retrofracta 

20. About one-half of fruiting pedicels semigeniculate to geniculate at their bases; siliques 

1.25 to 2.5 mm wide; middle and upper cauline leaves flat at edges ........... pA 


poh 


21. Siliques 1.75 to 2.5 mm broad; Greenland ...... 25a. A. holboellii var. holboellii 


21. Siliques 1.25 to 1.5 mm broad; Alaska, Yukon, Mack. Dist., Que., (rare, but common 
in Gaspé), Ont. (rare), Sask., Alta., B.C., and southward ..................... 
pee reee doen Pie ce eae wane eae 25d. A. holboellii var. secunda 


[S661 


Siquiy —UeBITINA 


Lol 


19. Fruiting pedicels gradually to abruptly reflexed near their bases, never semigeniculate or 
CA ot she gta pyar aa SR eae nae oh ee oem we ae nedees 22 

22. Undersurfaces of caudex leaves mostly with short-stalked (less than 0.063 mm long), 
semidendritic to dendritic, unbranched to few-branched, 2- to 3-forked trichomes 0.25 

to 0.35 mm long; these trichomes often semi-appressed to leaf surfaces and pointing 

towards apexes; Alaska, Yukon, Mack. Dist., Man., Sask., Alta., B.C., and southward 

Se ee ee te ep iva Wee ee Br ee A ae te ae es ae 30. A. pinetorum 

22. Undersurfaces of caudex leaves with sessile to medium-stalked (to 0.125 mm long), 
semistellate to stellate, unbranched to many-branched, 3- and 4-rayed trichomes from 

O12). tO-U.o) dn Wade: 2x bs. maiyanuah< tow eecuyseehensee cee eaderasenaues 23 

23. Middle cauline leaves weakly auriculate-, hastate- to sagittate-clasping stems; un- 
dersurfaces of caudex leaves with short-stalked (less than 0.063 mm long) sem- 

istellate trichomes, mostly 0.125 mm wide; Yukon, Sask. (Cypress Hills), sw Alta., 

Cie) SOM WLC: weasels acca aude waa tee anu eee eens 22. A. exilis 

23. Middle cauline leaves strongly auriculate-, hastate- to sagittate-clasping stems; un- 
dersurfaces of caudex leaves with nearly sessile stellate trichomes, mostly 0.25 to 

0.35 mm wide; Sask., Alta, B.C., and southward .................0.0 000005. 

ES ee rr eee re ae ee eee 25b. A. holboelli var. consanguinea 

18. Siliques slightly descending, spreading, strongly ascending to erect-appressed to rachis ..... 
24. Undersurfaces of caudex leaves mostly with medium- to long-stalked (over 0.063 mm long) 
rayed or forked trichomes; simple trichomes on surfaces of caudex leaves present or absent 


No 
pay 


25. Siliques arcuate-spreading to arcuate-descending .................0..0..00.0000. 26 
26. Undersurfaces of caudex leaves mostly with 2- and 3-forked trichomes; fruiting 


8cI 


vIOpoyy 


L6 TOA] 


plants usually have sterile rosettes with strongly ascending leaves; s B.C. (Penticton 

ALCA) GNC SOUIDWALG 64 ko eee siee adn Pare eae eae eee es 28. A. sparsiflora 

26. Undersurfaces of caudex leaves mostly with 3- and 4-rayed trichomes; fruiting plants 

lacking sterile rosettes; Yukon, sw Alta., B.C., and southward ................ 

pies poses eucwen ta en etae eek co eas cae eee ees 29. A. columbiana 

25. Siliques mostly straight and ascending, strongly ascending to erect-appressed to 

PAWNS coscareceu acs wee eeite ee a a eee ee cee o A ie s Gs ee eo 27 

27. Cauline and caudex leaves with similar dentate to subdentate margins; flowering 
StCMS SPrCaCine 10 ASCONGING 5:5 oie cancdedaukas uae uabun oko hte aeaek ds 

28. Petals less than 10 mm long and 3.5 mm wide; plants self-compatible, forming 

well-developed siliques; undersurfaces of caudex leaves mostly with un- 

branched cruciform trichomes 0.25 mm wide; Frank. Dist., Keew. Dist., 

Greenland, Nfld., and Que. .......0..0.00..0.00 cece eee. 8. A. alpina 

28. Petals more than 10 mm long and 3.5 mm wide; self incompatible; isolated 

plants with aborted siliques; undersurfaces of caudex leaves mostly with short- 

branched cruciform to 5-or-more-rayed trichomes; rare garden escape in 

OG OU nd Ce wecvergseeuterescss cease peiuese tue 9. A. caucasica 

27. Cauline and caudex leaves not similarly dentate to subdentate; flowering stems erect 

Sena he ee epee gt dried oe wa eee eee ee eee aaa 29 

29. Siliques ascending; caudex leaves with unbranched to few-branched, 2- and 

3-raye., medium-stalked (0.063 to 0.125 mm long) trichomes; siliques 1.5 

to 1.75 mm wide; Yukon (rare), and B.C. (rare) ............ 20. A. codyi 

29. Siliques strongly ascending to erect-appressed to rachis; caudex leaves with 

simple to medium- (0.063 to 0.125 mm) and long- (over 0.125 mm) stalked 

2- to 3-forked trichomes; siliques 1.0 to 1.75 mm wide 30 


SIGDAP — UesIT[N [S661] 


6G. 


30. Biennials; petals yellow; outer sepals not saccate at bases; siliques subterete; 
middle and upper cauline leaves glabrous and glaucous; Alaska, Yukon, 
Mack. Dist., Que., Ont., Man., Sask., Alta., B.C., southward, Europe, 
GG SIE: pee te so eek ees oes ae eon eueLase neers 1. A. glabra 
30. Biennials to short-lived perennials; petals white to purple; outer sepals 
saccate at bases; siliques strongly flattened; middle and upper cauline 
leaves pubescent, at least at bases, not glaucous ................. 
31. Petals small, 3 to 5 mm long; siliques 1.1 mm wide or narrower, erect- 
appressed to rachis; outer sepals moderately saccate at bases; cauline 
leaves approximate to remote 2 
32. Mostly perennials; siliques beakless or nearly so; B.C. (rare); Calif., 
Col. and Nevada and probably more widespread south of our 
PANPe: PUTONe: <5 cewiateaeegeeees 6a. A. hirsuta var. hirsuta 
32. Biennials to short-lived perennials; beaks of siliques mostly 0.5 
to 1.25 mm long; Alaska, Yukon, Mack. Dist., N.S., N.B., Que., 
Ont: Man.; Sask., Alta, B-C., and southward ...4..a0vss00.4 
aie aoe ene eee eeet esse aes 6b. A. hirsuta var. pycnocarpa 
31. Petals larger, (6) 7 to 9 (9.5) mm long; siliques 1.25 to 1.75 mm wide, 
somewhat divergent; outer sepals prominently saccate at bases; cau- 
line leaves remote; Alaska and westward, Yukon, B.C. and south- 
Wald. 4ae coe eaence eu ose ae eee eee us eee 7. A. eschscholtziana 
24. Undersurfaces of caudex leaves mostly with sessile to short-stalked (less than 0.063 mm long) 
rayed or forked trichomes; simple trichomes on surfaces of caudex leaves absent .... 33 
33. Undersurfaces of caudex leaves with unbranched to few short-branched 2- to 3-rayed 
or forked trichomes 34 


Poe ee er er ee ae er ee ee SS 


Oe! 


elopoyYy 


L6 1°A] 


24. siliques erect, often appressed 10 TaChiS. 5.04444 60440 04506 b.09440240R08e408 35 
35. Undersurfaces of rosette or caudex leaves with malpighiaceous trichomes only; 
Alaska, Yukon, Mack. Dist., Nfld., N.S., N.B., Que., Ont., Man., Sask., Alta., 


Ss aid SOU Wal. watt te eau ina eae ebad 14. A, drummondii 
oe Undersurfaces of caudex leaves mostly with sessile 3-rayed trichomes or short- 
stalked 2- and 3-rayed trichomes ................0... 00 c cece eee. 36 


36. Undersurfaces of caudex leaves mostly with unbranched, sessile, 3-rayed 
trichomes 0.25 to 0.35 mm wide; rays of trichomes appressed to leaf 
surfaces; bases of middle cauline leaves strongly clasping stems; sw Yu- 
kon, Mack. Dist. (rare), sw Alta., B.C., and probably southward ..... 
bce are G4 Ae wai w Ree de ee ee ees ee 15. A. calderi 

36. Undersurfaces of caudex leaves glabrous or with mostly unbranched to 
few-branched, short-stalked 2- and 3-rayed trichomes 0.125 mm wide; 
rays and branches of trichomes elevated from leaf surfaces; bases of 
middle cauline leaves cuneate, truncate to very weakly auriculate; sw 
Yukon, s B.C., and southward ...................... 21. A. murrayi 

34. Siliques strongly descending, descending to ascending ...................... 37 

37. Siliques to 25 mm long and 0.75 mm wide with a scattered, simple to bifurcate 

puberulence; petals up to 2.0 mm long: Essex Co., Ont., and southward ... 

Ube peo ee  Gitedues Lond ee eee eee e eee eats eee bean tees 13. A. shortii 

37. Siliques more than 25 mm long and 0.75 mm wide, glabrous; petals longer than 
SA, giver ecenete anise eaep es eue face nenens aeeeaewene nese euas 

38. Stems 0.7 to 2.5 (4.0) dm high; perennials with persistent, often branched, 
CAUGCKGS: 2 even eeanc gaa caneaeatebeteeeeus nonce. ouenn ies eae 39 


SIQDAP — uUestT[NA] [S661 


eA 


oS) 
\O 


. Siliques 2.0 to 3.5 mm wide; undersurfaces of caudex leaves with sessile 
to short-stalked (less than 0.063 mm long) unbranched to 2-branched, 
3-rayed trichomes from 0.25 to 0.35 mm wide; siliques spreading 
to ascending or strongly ascending; inflorescences symmetrical to 
SOCUU Goan pe eee hos beh ei ed eee Le eee eee 40 

40. Undersurfaces of caudex leaves with short-stalked, semistellate 1 - 

to 2-branched, 3-rayed trichomes; siliques spreading to ascend- 
ae inflorescences secund; sw Yukon, sw Alta., B.C., and south- 
eee een ee eT ee eee Sere 19. A. drepanoloba 

40. Undersurfaces of caudex leaves with sessile to short-stalked, un- 

branched, 3-rayed trichomes; siliques strongly ascending; inflo- 
rescences symmetrical; sw Alta., s B.C., and southward ..... 
Dade awe nae oe ee aaa eee 16. 4. lyallii 

Siliques 1.5 to 1.75 mm wide; undersurfaces of caudex leaves with 
medium-stalked (0.063 to 0.125 mm long), unbranched to few- 
branched, 2- and 3-forked trichomes from 0.125 to 0.25 mm long; 
siliques ascending; inflorescences symmetrical; sw Yukon, and nw 

she Meee Coho ea Roe a ee eee ee Bees 20. A. codyi 

38. Stems (2) 3 to 10 dm high; biennials or short-lived perennials with compact 

caudexes that tend to become reduced in size as plants mature .... 41 

41. Undersurfaces of caudex leaves with unbranched to many prominently 
branched, sessile to short-stalked, 3-rayed trichomes, mostly less 

Pian Oso WI Wide. poi cee ee pence eieeee cles eee 42 
42. Undersurfaces of caudex leaves with unbranched, sessile or nearly 


3 


\O 


Gol 


vIOpOyYy 


L6 ‘1OA] 


sessile, 3-rayed trichomes with rays appressed to leaf surfaces; 

inflorescences symmetrical; siliques descending, spreading to 

ascending; Alaska (rare), Yukon, Mack. Dist., N.B. (rare), Que., 

Ont., Man., Sask., Alta., B.C., and southward .............. 

eee eee ee eee ee 17a. A. divaricarpa var. divaricarpa 

42. Undersurfaces of caudex leaves with short-stalked, 3-rayed tri- 

chomes; the rays elevated above the leaf surfaces with numerous 

prominent branches; inflorescences semisecund to secund; si- 

liques slightly to strongly descending; Sask., southward into 

U.S.A., and disjunct to Que. (rare in Gaspé) . 26. A. boivinii 

41. Undersurfaces of caudex leaves with unbranched to few weakly 

branched, short-stalked, 3-rayed trichomes, mostly more than 0.35 

mm wide; se Alaska, Yukon, Mack. Dist., Que. (rare), n Ont., Man., 

Sask., Alta., B.C., and southward ........................0.00. 

ga ease eee aoe pea ene eae eee 17b. A. divaricarpa var. dacotica 

33. Undersurfaces of caudex leaves with unbranched to many-branched 3- and 4-rayed 

TIGHOINES Gay ced nd ce oso ee ee Oe ee ea eee eee aes 43 

43. Trichomes on undersurfaces of caudex leaves mostly less than 0.25 mm wide; 
SiMGues..25-10-220 Mit Wide: 44 cs.cnecd eye mcbeue oe en cuanto Feveorearenena 

44, Trichomes on undersurfaces of caudex leaves mostly between 0.125 and 0.25 

mm wide; inflorescences secund; siliques mostly spreading; Yukon (rare), 

Alta., B.C., and southward .......................000. 18. A. /emmonii 

44. Trichomes on undersurfaces of caudex leaves mostly 0.125 mm wide; inflo- 

rescences symmetrical; siliques mostly ascending; s B.C., and southward .. 

bodtd pidueeadeaeseoaeiatedaddae cd ataterdenceieeesele 23. A. microphylla 


[S66] 


Signy — URSIN 


eel 


43. Trichomes on undersurfaces of caudex leaves mostly 0.25 or more mm wide; siliques 
Lod HO OP a IO 25045655065 0G286224 08 cet dates eh eednte des oucs 45 

45. Inflorescences secund; siliques 2.25 to 3.0 mm wide; sw Yukon, sw Alta., B.C., 
NGL SOCAN AN eet Ban ocr ew aeatn een ete ae Bormann eee 19. A. drepanoloba 

45. Inflorescences symmetrical to slightly secund; siliques 1.5 to 2.0 mm wide .. 


46. Siliques strongly ascending to erect; trichomes on undersurfaces of basal 
leaves short-stalked, semidendritic; stems 5 to 20 (40) cm high; s B.C., 
and SOUTDWAIG «6 2 ceca Seeds te ehbees weno dataae 24. A. depauperata 

46. Siliques spreading to descending; trichomes on undersurfaces of caudex 
leaves nearly sessile, semistellate to stellate; stems 30 to 60 cm high .. 

47. Inflorescences symmetrical; trichomes on undersurfaces of caudex 
leaves mostly 0.25 mm wide; strongly perennial; frequently many 
Mime, $8.0... and southward <c:i5 24543656404 27. A. lignifera 

47. Inflorescences slightly secund; trichomes on undersurfaces of caudex 
leaves more than 0.25 mm wide; biennial or shortlived perennials; 
usually one to few stemmed; Sask., southward into U.S.A., and Que. 

(Pare 1: GASPe) 6 deivee see phe eee be hee ns ees 26. A. boivinii 


pel 


elopoyYy 


L6 1A] 


1995] Mulligan — Arabis 135 


1. Arabis glabra (L.) Bernh., Syst. Verg. erf. 195. 1800. Based on 
Turritis glabra L., Sp. Pl. 2: 666. 1753; Habitat in Europae, 
Hort. Cliff. 339. See Hopkins (1937) pp. 106 & 107 for sum- 
mary of synonymy. 


Arabis macrocarpa (Nutt.) Torr., Bot. Mex. Boundary pt. 1: 32. 1838. Based on 
urritis macrocarpa Nutt. in T.&G., Fl. N. Am. 1: 78. 1838; Rocky situations 
in woods of Oregon (HOLOTYPE PH!) 
Arabis glabra var. furcatipilis Hopkins, Rhodora 39: 109. 1937; Logan City Camp, 
Logan Canyon, Utah, B. Maguire 3437 (HOLOTYPE Gu!). 


DISTRIBUTION. It is a native of Europe and western Asia, ex- 
cept in the extreme north and south, and, according to Hultén 
(1971) has been introduced into Africa and Australia. In North 
America, it is distributed widely in temperate areas of Alaska, 
Yukon, Quebec, Ontario, Manitoba, Saskatchewan, Alberta and 
British Columbia and, from Hopkins (1937), southward from 
Maine to North Carolina and westward to Washington, Oregon 
and California. It is obviously introduced at a few weedy sites in 
Alaska and Yukon, but occurs both in weedy and native habitats 
further south, particularly on sandy and sandy-loam soils of road- 
sides, railway embankments, open fields, riverbeds, outcrops, 
ledges, cliffs and openings in thickets and woods. 

BIOLOGICAL Notes. This species is sometimes removed from 
Arabis and placed in the genus Turritis, mostly because of its 
subterete siliques. I prefer to follow the treatment of Rollins (1941, 
p. 316) and include it in Arabis. 

The chromosome-number n = 6, 2n = 12 was obtained on 
North American material of 4. glabra from Quebec, Ontario, 
Alberta, California, Colorado, Idaho, Montana, Oregon and Wy- 
oming (Table 1). Pollen mother cells had 6 pairs at meiosis. Burdet 
(1967), combining his cytological observations and those of pre- 
vious workers, reported only 2 = 12 for plants of Canada, Aus- 
tria, Bulgaria, Czechoslovakia, England, France, Germany, Hun- 
gary, Sweden and Switzerland. The n = 8 count of Hill (1982) on 
material from Virginia, therefore, needs clarification. Burdet (1967) 
states that the somatic chromosomes of this species are quite 
different from those of other Arabis, and uses this as a further 
argument that it should be placed in the genus J urritis. There is 
every indication that A. glabra reproduces sexually and that abun- 
dant viable seed is normally produced by selfing. 


136 Rhodora [Vol. 97 


€! Fee i 
ek 


yy ¢ » 
o% 4 


a. ma 
Figures 8-14. SEM photographs of Arabis trichomes; all x 150. Figure 8, 4. 
lemmonii. Figure 9, A. drepanoloba. Figure 10, A. codyi. Figure li, A. murrayi. 
Figure 12, 4. exilis. Figure 13, A. microphylla. bicure 14, A. depauperata. 


1995] Mulligan — Arabis 137 


2. Arabis media N. Busch, Fl. Sib. Orient. Extrem. 1: 465. 1926; 
figure p. 464. Cardaminopsis media (N. Busch) O. E. Schulz 
in Engler & Prantl, Nat. Pflanzenf. 17b: 541. 1936. 


Arabis media DC. var. glabra (DC.) Busch, Fl. Sib. Orient. Est. 4: 465. 1926. 
Based on part of Arabis ambigua DC. var. glabra DC., Syst. 2: 121. 1821; 
Unalaska. 

Arabis media var. intermedia (DC.) Busch, FI. Sib. Orient. Est. 4: 465. 1926. Based 
on part of Arabis ambigua DC. var. intermedia DC., Syst. 2: 121. 1821: 
Unalaska. 


DISTRIBUTION. It is native to sand dunes, sand bars, tundra 
tussocks, disturbed gravel, river terraces and volcanic ash in arctic 
regions of northern Yukon, northern Alaska and adjacent north- 
eastern Siberia. According to the distribution in the map of Arabis 
arenicola in Hultén (1968) the map should be that of A. media. 

BIOLOGICAL Notes. This taxon has subterete siliques, with a 
prominent midvein from base to apex, rather than the flattened 
siliques, with prominent midvein absent towards the apex, that 
is characteristic of most of our Arabis species. Arabis media has 
accumbent cotyledons. I do not consider the siliques of this species 
sufficiently different from our Arabis to warrant placing it in an- 
other genus. In fact, many workers have erroneously included 
specimens of A. media in A. arenicola. 

Arabis media plants from two locations in Alaska had 16 so- 
matic chromosomes (Table 1). Its basic number of x = 8 is the 
common one in Eurasiatic species of Arabis. It seems likely that 
A. media isa sexual species that produces seed primarily by selfing. 


3a. Arabis arenicola (Richardson) Gelert var. arenicola. Arabis 
arenicola (Richardson) Gelert, Bot. Tidsk. 21: 289, 290. 1898. 
Based on Eutrema arenicola Richardson in Hooker, FI. Bor.- 
Am. 1: 67. 1830; Deep sand upon shores of Arctic America, 
between long 107° and 150°, Dr. Richardson (ISOTYPE, CAN!). 


Arabis humifusa (J. Vahl) 8. Wats., Proc. Am. Acad. 25: 124. 1889. Based on 
Sisymbrium humifusum J. Vahl, Fl. Dan. t. 2297. 1840. 


An excellent discussion of the systematics and nomenclature 
of A. arenicola var. arenicola and var. pubescens is presented on 
pages 77 to 80 of Hopkins (1937). 


138 Rhodora [Vol. 97 


DISTRIBUTION. It is a native on sandy beaches, river banks 
and dunes, and on rocky shores along the north coast of Canada 
from the Mackenzie District to Labrador, on Southampton Island, 
on Baffin Island, along shores and on islands in Hudson Bay, 
Kenora District in northern Ontario, south shore of Lake Atha- 
basca in northern Saskatchewan and in Greenland. 

BIOLOGICAL Notes. The chromosome number 7 = 8, 2n = 
16 has been obtained for plants of this species from Greenland 
and Quebec (Table 1). It has the same basic number, x = 8, as 
A. media. Both var. arenicola and var. pubescens are probably 
sexual taxa producing seed primarily by selfing. 


3b. Arabis arenicola var. pubescens (S. Wats.) Gelert, Bot. Tidsk. 
21: 290. 1898. Based on A. humifusa (J. Vahl) S. Wats. var. 
pubescens S. Wats. in Gray, Synop, Fl. N. Am. 1: 160. 1895. 


DISTRIBUTION. It is native on sandy and gravel beaches and 
river banks along the east and west coasts of Hudson Bay and on 
adjacent islands in the Keewatin District, Quebec, Ontario and 
Manitoba, Kenora District in northern Ontario and on the south 
shore of Lake Athabasca in northern Saskatchewan. 


4. Arabis lyrata L., Sp. Pl. 2: 665. 1753; Habitat in Canada, D. 
Kalm, Gron. virg. 76. 


See Hopkins (1937), page 89 for synonomy. 


DISTRIBUTION. It is native on ledges and cliffs in thickets and 
woods and on sandy or rocky shores of streams, rivers and lakes 
in the southern half of Ontario, Manitoba (rare), Saskatchewan, 
Alberta (localized) and in British Columbia (rare). It occurs as a 
rare introduction in the Mackenzie District and in Alaska. It is 
found in Vermont, to New Jersey and Georgia, west to Missouri, 
and north to Wisconsin (Rollins, 1993b). 

BIOLOGICAL Notes. Arabis lyrata material from two locations 
in Ontario had 16 somatic chromosomes (Table 1). Pollen mother 
cells had 8 pairs at meiosis. It is a diploid with the basic chro- 
mosome-number of x = 8, the same base number as 4. kam- 
tschatica, a tetraploid species that is frequently considered a va- 
riety or subspecies of 4. /yrata. Other closely related taxa, are A. 


1995] Mulligan —Arabis 139 


media and A. arenicola, both diploids based on x = 8. Arabis 
lyrata is likely a sexual species producing seed by selfing. 


5. Arabis kamtschatica (Fisch.) Ledeb., Fl. Ross. 1: 121. 1842; 
Kamtschatka, ex herb. Fisch. & labelled Arabis kamtschatika 
Fisch. (LEcTotTyPeE chosen LE!) Arabis lyrata var. kamtscha- 
tica Fisch. ex DC. Syst. 2: 231. 1821. Arabis ses subsp. 
kamtschatica (Fisch.) Hultén, Fl. Aleut. Is. 202: 1937. 


Arabis kamtschatica var. glabra (DC.) N. Busch, FI. Sib. Orient. Est. 4: 468. 1926. 

7. lyrata var. glabra (DC.) Hopkins, Rhodora 39: 93. 1937. Based on 
of Arabis ambigua DC. var. glabra DC., Syst. 2: 121. 1821; Unalaska. 

hah Seer var. intermedia (DC.) N. Busch, FI. Sib. Orient. Est. 4: 468. 
1926; Arabis lyrata var. intermedia (DC.) Farwell, Mich. Acad. Sci. Rep. 256. 
1917. Based on part of Arabis ambigua DC. var. intermedia DC., Syst. 2: 
121. 1821; Unalaska. 

Arabis occidentalis (Wats.) Nelson, Univ. Wyoming Pub. 3: 111. 1937. Based on 
Arabis lyrata var. occidentalis S. Wats. in Gray Syn. Fl. N. Am. 1: 159. 1895: 
Unalaska. 


DISTRIBUTION. Moist and spring-flooded habitats in Alaska, 
Aleutian Islands eastern Asia, Yukon, Mackenzie District, British 
Columbia, Washington and south shore of Lake Athabaska in 
northern Saskatchewan. 

BIOLOGICAL Notes. The chromosome number 2n = 32 has 
been obtained on plants growing in Alaska, Yukon and British 
Columbia (Table 1). It is a tetraploid with the base number of x 
= 8. It is likely a sexual species producing seed by selfing. 


6a. Arabis hirsuta (L.) Scop. var. hirsuta. Arabis hirsuta (L.) Scop., 
Fl. Carn. 2: 30. 1772. Based on Turritis hirsuta L., Sp. Pl. 2: 
666. 1753; Hort cliff. 339, habitat in Sueciae, Germaniae and 
Angliae. 


DISTRIBUTION. Known from only one location in Canada 
(Mountain Creek campground and roadside, 51°26’N 117°30’W, 
Glacier National Park, British Columbia, elev. 2850 ft., Haber & 
Shchepanek 1685, CAN). I have seen specimens in DAO from Cal- 
ifornia, Colorado and Nevada. It is probably more widespread 
south of Canada. It also occurs in temperate areas of Europe and 
Asia (Hultén, 1971). 


140 Rhodora [Vol. 97 


BIoLocicAL Notes. Plants from Mono Co., California and 
Custer Co., Colorado, had the chromosome-number 7 = 16 (Table 
1). Pollen mother cells had 16 pairs at meiosis. Burdet (1967) 
records the chromosome-number 2n = 8 for plants of A. hirsuta 
from one location in Switzerland and one in France, 2 = 16 for 
plants from Switzerland, France, Sweden, Bulgaria and Germany 
and 2n = 32 for plants from England and Norway. He suggests 
that x = 4 has to be considered as the base number for at least 
part of the genus Arabis. Arabis hirsuta is probably a sexual species 
producing seed by selfing. 


6b. Arabis hirsuta var. pycnocarpa (Hopkins) Rollins, Rhodora 
43: 318. 1941. Based on Arabis pycnocarpa Hopkins, Rho- 
dora 39: 117. 1937; Quebec, dry ledges, St. Jean l’Evangéliste, 
Nouvelle Bonaventure Co., J. F. Collins & M. L. Fernald, 
July 19 & 20, 1904 (HoLoTyPE Gu!). Arabis hirsuta subsp. 
pycnocarpa (Hopkins) Hultén, Arssk. Lunds Univ. N.F. Avd. 
41: 873. 1945. 


Arabis rupestris Nuttall in T.&G., Fl. N. Am. 1: 81. 1838; Wahlalmet R. (Oregon), 
Nuttall (IsoTyPE GH!). 

Arabis pycnocarpa var. reducta Hopkins, Rhodora 39: 117. 1937; Gravelly Beach, 
Carlton, Tracadigash Point, Quebec, Collins & Pease 4312 (HOLOTYPE GH)). 

Arabis hirsuta var. adpressipilis (Hopkins) Rollins, Rhodora 43: 319. 1941. Based 
on Arabis pycnocarpa var. adpressipilis Hopkins, Rhodora 39: 117. 1937; 
Montier, Missouri, Bush 32 (HOLOTYPE GH!). 

Arabis hirsuta var. minshallii B. Boivin, Can. Field-Nat. 65: 16. 1951; Lemieux 
Island, Ottawa, Ont., W. H. Minshall, June 26 1934 (HOLOTYPE DAO!) 


DISTRIBUTION. The information given by Hopkins (1937) for 
A. pycnocarpa covers my concept of A. hirsuta var. pycnocarpa: 
basic ledges, cliffs, bluffs, dry and rocky or moist banks and grav- 
elly alluvium, eastern Quebec to Yukon, south to Georgia, In- 
diana, Illinois, Missouri, Kansas, New Mexico, Arizona and Cal- 
ifornia. Workers should look for var. hirsuta, particularly south 
of the Canadian border. I have seen specimens of var. pycnocarpa 
from Alaska, Mackenzie District, Yukon, Nova Scotia, New 
Brunswick, Quebec, Ontario, Manitoba, Saskatchewan, Alberta 
and British Columbia. 

BrioLocicaAL Notes. Plants from Mackenzie District, Mani- 
toba, Alberta, British Columbia, Connecticut, Colorado, South 
Dakota and Wisconsin had the chromosome number nv = 16, 2n 


1995] Mulligan —Arabis 141 


= 32 (Table 1). Pollen mother cells had 16 pairs at meiosis. Both 
var. hirsuta and var. pycnocarpa plants of North America are 
either tetraploid based on x = 8 or octoploid based on x = 4. 
Burdet (1967) has suggested that the base number for A. hirsuta 
may be x = 4, not x = 8. 


7. Arabis eschscholtziana Andrz. in Ledeb. FI. Alt. 3: 25. 1831; 
ex insula Unalaschka, Cham. & Schtechtd. (Type not seen). 
Arabis hirsuta var. eschscholtziana (Andrz.) Rollins, Rhodora 
43: 320. 1941. Arabis hirsuta subsp. eschscholtziana (Andrz.) 
Hultén, Arssk. Lunds Univ. N.F. Avd. 41: 872. 1945. 


DISTRIBUTION. It occurs in moist habitats towards the Pacific 
coast in the Aleutian Islands, Alaska, Yukon, British Columbia, 
western Idaho, Washington and Oregon. 

BIOLOGICAL Notes. The chromosome-number n = 32 was 
obtained from plants growing in the Queen Charlotte Islands of 
British Columbia (Table 1). Pollen mother cells had 32 pairs at 
meiosis. Arabis escholtziana has twice the chromosome-number 
of North American plants of A. hirsuta, a species to which it is 
obviously closely related. They possibly also have the same basic 
chromosome-number, either x = 4 or x = 8. Arabis eschschol- 
tziana is probably a sexual species. Although it has rather large 
flowers for our Arabis taxa, some plants growing in isolation set 
much seed. It is, therefore, probably a selfer. 


8. Arabis alpina L., Sp. Pl. 2: 664. 1753; Habitat in Alpibus 
Helveticus, Lapponicus, Hort. cliff. 335. 


Arabis incana Moench., Arabis alpina var. minor Lange, Arabis alpina var. rud- 
eralis Wormskj., Arabis alpina var. glabrata Blytt. (see Hopkins, 1937). 


DISTRIBUTION. Cliffs, ledges, rock and gravel shorelines, and 
alpine meadows along the west coast of Hudson Bay in Manitoba 
and Ontario, islands of Hudson Bay, southern shore of Baffin 
Island, northern shores of Quebec and Labrador, Gaspé Quebec 
and adjacent north shore of St. Lawrence River, Newfoundland 
and Greenland. 

BioLocicAL Notes. Plants from Greenland, Quebec and 
Manitoba had the chromosome number n = 8, 2m = 16. Pollen 
mother cells had 8 pairs at meiosis (Table 1). Arabis alpina is 
probably a sexual species producing seed by selfing. 


142 Rhodora [Vol. 97 


9. Arabis caucasica Willd., Enum. Pl. Hort. Berol. suppl. 45. 
1813. 


DISTRIBUTION. It is found south of the range of A. alpina in 
Europe and Asia (Hultén, 1958). It persists as a rare garden escape 
in Quebec, Ontario and British Columbia. 

BIOLOGICAL Notes. Plants growing as a garden escape in On- 
tario had 8 pairs of chromosomes at meiosis (Table 1). Since 
isolated plants produce little or no seed, it is likely that A. cau- 
casica 1s a sexual outcrosser. It is closely related to A. alpina and 
is considered a variety or subspecies of A. alpina by some workers. 


10. Arabis nuttallii Robinson in Gray, Syn. Fl. N. Am. 1: 160. 
1895; R. Mts, Nuttall (HOLOTYPE PH!; IsoTYPE GH!). 

Arabis bridgeri M. E. Jones, Contrib. West. Bot. 14: 38. 1912; Mt. Bridger, Gallatin 
o., Montana, M. E. Jones, Aug. 10, 1905 (HOLOTYPE PH, not seen). 
Arabis macella Piper, Proc. Biol. Soc. Wash. 33: 103. 1920. Ritzville, Adams Co., 

Washington, Sandberg & Leiberg 202 (HOLOTYPE US, not seen). 


DISTRIBUTION. Grassy slopes and benches, prairie hillsides, 
open thickets and woods, and roadsides in Yukon (very rare), 
southwest Alberta, southeast British Columbia, Montana, Wyo- 
ming, Utah, Idaho, Washington and Nevada. 

BIOLOGICAL Notes. I have seen no information on the chro- 
mosome number of A. nuttall/ii. Since isolated plants have some 
aborted siliques and plants growing in dense colonies have all 
well-formed siliques, it is possible that this relatively large flow- 
ered Arabis is a sexual outcrosser. 


11. Arabis canadensis L., Sp. Pl. 2: 665. 1753; Eruca virginiana 
Pluk. alm. 136. 


Arabis falcata Michx., Fl. Bor. Am. 1: 31. 1803 and Arabis mollis Rafinesque, 
m. Month. Mag. 2: 43. 1817 non Steven, Bull. Soc. Nat. Mosc. 2: 270. 1812 
(see Hopkins, 1937, p. 178). 


DISTRIBUTION. Rich woods, thickets and rocky banks, south- 
western Quebec, southern Ontario. It also occurs from Maine to 
Florida, and westward to Texas, Nebraska and Minnesota (Rol- 
lins, 1993b). 

BIOLOGICAL Notes. The chromosome-number of 2n = 14 has 
been reported for plants growing in southern Ontario (Table 1). 
It has the basic number, x = 7, that is very common for North 
American species of Arabis. 


1995] Mulligan —Arabis 143 


12. Arabis laevigata (Miihl.) Poir., Encyl. Suppl. 1: 411. 1810. 
Based on Turritis laevigata Mihl., Index Fl. Lancastr. in 
Trans. Am. Phil. Soc. 3: 1793. 


Arabis iyragjolia rae Syst. 2: 244. 1821; Arabis heterophylla Nutt. ex T.&G., 
. 1838; Arabis hastata Eaton, Man. Bot. ed. 2: 141. 1818. ai 
after ae 1937). 
Arabis missouriensis Greene, in Feddes Repertorium 5: 244. 1908; Montier, Mis- 
souri, B, F. Bush 31 (HOLOTYPE ND)). 


DISTRIBUTION. Rich rocky woods, rocky hillsides and ledges, 
southwestern Quebec, southern Ontario. It is also found from 
New Jersey to Georgia, west to Oklahoma and Kansas, and north 
to Minnesota (Rollins, 1993b). 

BIOLOGICAL Notes. The chromosome-number n = 7 was ob- 
tained for plants from Connecticut, Maryland and Wisconsin (Ta- 
ble 1). Pollen mother cells had 7 pairs of chromosomes at meiosis. 
Arabis laevigata is probably a sexual species producing seed by 
selfing. 


13. Arabis shortii (Fern.) Gleason, Phytologia 4: 23. 1952: Arabis 
perstellata E. L. Braun var. shortii Fernald, Rhodora 48: 208. 
1946. Based on Sisymbrium dentatum Torrey in Short, 3rd 
Suppl. Cat. Pl. Kentucky, 338. 1833; On the sandy banks of 
the Ohio river, Fl. April, C. W. Short (LECTOTYPE Dwc!). 

Arabis shortii (Fern.) Gleason var. Dhalacrocarpa (Hopkins) Steyerm., Rhodora 
62: 130. 1960. Based on Arabis dentata var. phalacrocary Hopkins, Rhodora 

39: 169. 1937; Along shaded limestone bluffs of Osage River, near Osceola, 
St. Clair County, Missouri, E. J. Palmer 35650 (HOLOTYPE GH!). 


DISTRIBUTION. Shady banks and bottomlands and on lime- 
stone bluffs and ledges in rich woods, Essex County in Ontario. 
It also occurs from New York to Virginia, Tennessee and Ala- 
bama, west to Kansas, and Nebraska, and north to Minnesota 
(Rollins, 1993b). 

BIOLOGICAL Notes. The chromosome number of n = 7 re- 
ported for Arabis perstellata (Table 1), probably does not apply 
to A. shortii. The holotype of A. perstellata in Gu has cauline 
leaves cuneate at bases not sagittate-clasping as in A. shortii. 


14. Arabis drummondii Gray, Proc. Am. Acad. 6: 187. 1863. 
Based on Turritis stricta Graham, Edinburg New Phil. Jour. 
350, 1879; plant grown from seeds collected by Drummond 
in Rocky Mts. (see Hopkins, 1937, p. 146). 


144 Rhodora [Vol. 97 


elas sarah Greene, in Pittonia 4: 196. 1900; Alberta, Elbow River, Rocky 
ns, J. Macoun 18101 (HOLOTYPE ND!). 
dra age var. oxyphylla (Greene) Hopkins, Rhodora 39: 143. 1937. 
sed on Arabis oxyphylla Greene, in Pittonia 4: 197. 1900; Near Pagosa 

se Colorado, C. F. Baker 747 (HOLOTYPE ND!). 

Arabis drummondii var. connexa (Greene) Fernald, Rhodora 5: 231. 1903. Based 
on Arabis connexa Greene, in Pittonia 4: 197. 1900; Near Pagosa Peak, 
Colorado, C. F. Baker 341 (HOLoTYPE & 3 IsoTyYPES ND!). 


DISTRIBUTION. Open, often calcareous, habitats in Alaska, 
Yukon, Mackenzie District, Newfoundland, Labrador, Nova Sco- 
tia, New Brunswick, Quebec, Ontario, Manitoba, Saskatchewan, 
Alberta and British Columbia. It is widespread from New Jersey, 
west to northern Arizona and the Sierra Nevada of California, 
and north to Washington (Rollins, 1993b). 

BIOLOGICAL Notes. <Arabis drummondii is one of our most 
widespread species. Siliques tend to be narrower and the basal or 
caudex leaves glabrous or nearly so in the eastern part of its range. 
This species seems to be most closely related to A. calderi and 
through 4. /yallii to A. divaricarpa. Arabis dr'ummondii contains 
diploid, triploid and tetraploid chromosome races based on x = 
7 (Table 1). Most diploids from Yukon, Alberta, California, Col- 
orado, Montana, Utah and Wyoming had 7 pairs at meiosis. 
These plants were probably sexual, producing seed by selfing. The 
triploids from Alberta and British Columbia almost certainly pro- 
duce seed by apomixis. According to Bécher (1969) the tetraploid 
plants from Massachusetts were probably sexual as their pollen 
had the reduced number of 14. 


15. Arabis calderi G. Mulligan, sp. nov. 


Arabis calderi ab A. drummondia 3-radiatis trichomatibus et siliquis angustior- 
ibus differt et ab A. lyallii et ab A. divaricarpa siliquis erectis et rachidi adpressis 
differt. 


Arabis calderii differs from Arabis drummondii by its 3-rayed 
trichomes and narrower siliques and from Arabis /yalliiand Arabis 
dicaricarpa by its siliques being erect and appressed to the rachis. 

The holotype specimen, in the herbarium of Agriculture Can- 
ada, Ottawa (DAO), 1s named after James Alexander Calder, 1915- 
1990 (see tribute to Calder by Cody & Cayouette, 1991). British 
Columbia, Indian River at Mile 34 from Alaska Highway on Atlin 
Road, approx. 59°54’N, 133°48’W, common on open grassy flats 


1995] Mulligan— Arabis 145 


Figures 15-20. SEM photographs of Arabis trichomes; all x 150. Figure 15, 
A. holboelli var. retrofracta. Figure 16, A. boivinii. Figure 17, A. lignifera. Figure 
18, A. sparsiflora. Figure 19, A. columbiana. Figure 20, A. pinetorum. 


146 Rhodora [Vol. 97 


on bench above river, /. A. Calder & J. M. Gillett 25180, June 
9, 1960 (HOLOTYPE DAo!). 

Perennial with a simple or branched caudex; stems erect, one 
to several, simple to once-branched, 7.5 to 40.0 cm high; cauline 
leaves entire, glabrous, sessile, mostly strongly sagittate-clasping 
stems, narrowly lanceolate, 1 to 4 cm long, 1 to 4 mm wide; 
caudex leaves entire, numerous and rosulate, 1 to 3 cm long, 2 
to 6 mm wide, the blades narrowly oblanceolate, slender petiolate; 
lower surfaces of caudex leaves with sparse to scattered sessile, 
unbranched 3-rayed trichomes 0.25 to 0.35 mm wide; inflores- 
cences symmetrical, usually congested; sepals purplish, oblong to 
narrowly oblong, saccate at bases, 3 to 4 mm long, 0.75 to 1.0 
mm wide; petals purplish, narrowly cuneate, 7.0 to 7.5 mm long, 
0.75 to 1.5 to 1.75 mm wide; style rudimentary; seeds mostly in 
2 rows, brownish, oval, 1.75 mm long, 1.25 mm wide, promi- 
nently winged on sides and apex, cotyledons accumbent. 

DISTRIBUTION. Grassy clearings, meadows and openings in 
thickets in subalpine and alpine areas of southwestern Yukon, 
Great Bear Lake of Mackenzie District, southwest corner of Al- 
berta and in British Columbia. It is probably commoner than the 
number of herbarium collections indicate. No specimens were 
seen from the adjacent United States, but it almost certainly oc- 
curs there. 

BIOLOGICAL Notes. It is most closely related to A. drummon- 
dii and A. lyallii. All of the specimens of A. cal/deri seen in herbaria 
were previously identified mostly as A. drummondii and occa- 
sionally as A. /yallii or A. divaricarpa. Although no chromosome 
numbers have been determined from plants of A. ca/lderi, | would 
predict a base number of x = 7. 

REPRESENTATIVE SPECIMENS. CANADA. Yukon: Vicinity of 
Mackintosh (mile 1022, Alaska Highway), roadside, flowers pur- 
plish tinged, slightly glaucous, Schofield & Crum 7348 (CAN, UBC); 
Vicinity of Pine Creek, Alaska Highway near mi. 1019, app. 
60°47'N, 137°35'W, prairie, Raup et al. 13033 (ALA, CAN, UBC); 
Canol Rd., Mile 132, Lower Lapie R. Crossing, slopes near tim- 
berline, 5000’, Porsild & Breitung 9722 (CAN); Mile 25 from Alas- 
ka Highway on road to Dawson, 61°08'N, 135°20’W, occasional 
in openings in dwarf birch-willow thickets on flats at 2600’, flow- 
ers dark mauve, Calder & Gillett 25788 (ALA, DAO); Champagne 
at Mile 974, Alaska Highway, 60°47'N, 136°29'W, in slightly 
saline grassy clearing between spruce-Aspen ..., flowers mauve, 


1995] Mulligan—Arabis 147 


alt. 2300’, Calder & Gillett 25142 (DAO). Mackenzie District: Great 
Bear Lake, north shore of Smith ARM, Olmstead Bay, about 
66°32'N, 122°35’W, calcareous soil Porsild & Porsild 5080 (CAN); 
Great Bear Lake, Etacho Point (Big Point), elevation about 1500 
feet, 66°N, 121°30'W, Porsild & Porsild 3494B, sheet is mixture 
of 2 plants of A. calderi and 1 plant of A. drummondii (CAN). 
Alberta: Lake Agnes above Lake Louise, Banff National Park, 
occasional on dry, gravelly-rocky slopes at treeline, alt. 7000’, 
Calder 24021 (DAO); Crandell Trail and Akamena Hwy., 49°05'N, 
113°58'W, partly exposed on steep SE hill, 5200’, Waterton Lakes 
National Park, Blais 1996 (CAN). British Columbia: Blustry Moun- 
tain, 50°36’N, 121°42’W, 2256 m, Johns 584 (DAO, UBC); old 
Jackson Mine Road, south of New Denver-Kaslo Road, 50°01'N, 
117°09’'W, ca. 6000’, Beamish et al. 750328 (DAO, UBC); Cornwall 
Mtn., Lookout Rd., southwest of Ashcroft, ca. 6000 ft., Beamish 
630166 (DAO); Cairn Peak, Upper Hat Creek, 7600 ft., Brink 49- 
514 (UBC). 


16. Arabis lyallii S. Watson, Proc. Am. Acad. 11: 122. 1875. 
Based on Arabis drummondii var. alpina S. Watson in King, 
Geol. Expl. Fortieth Parallel 5: 18. 1871; Clover Mts., N. 
Nevada, 10,000 ft. alt., S. Watson 75 (HOLOTYPE GH!). 


Arabis oreophila Rydb., Bull. Torr. Bot. Club 3: 437. 1907; Divide between Big 
wood Canyon and Heber Valley, P. A. Rydberg & E. C. Carlton 6678 


(HoLo NY!). 

Arabis amerifolia Greene, Leaflets Bot. Obs. Crit. 2: 75. 1910; On a firm pumice 
slope, Crater Lake National Park, Oregon, F. V. Coville 1504 (HOLOTYPE us!) 

Arabis multiceps Greene, Leaflets Bot. Obs. Crit. 2: 76. 1910; On open rocky slope 
Mount Thielson, Cascade Mts., Oregon, F. V. Coville & E. I. Applegate 435 
(HOLOTYPE us!). 

Arabis densa Greene, Leaflets Bot. Obs., Crit. 2: 76. 1910; Eagle Cap, Imnaha 
National Forest, Oregon, 9500 feet, A. W. Sampson & G. A. Pearson 206 


( 

Arabis davidsonii Greene, Leaflets Bot. Obs. Crit. 2: 159. 1911; Bishop Creek, 
Inyo Co., Cal., A. Davidson 2728 (LECTOTYPE RSA!). ee Iva var. dap 
idsonii (Greene) Smiley, Univ. Calif. Pub. Bot. 9: 205. 192 


DISTRIBUTION. Occurs in alpine and subalpine areas in shale 
and rock crevices and in meadows in southwestern Alberta and 
southern British Columbia. It also occurs southward into Mon- 
tana, Idaho, Washington, Wyoming, Utah, Nevada, Oregon and 
California (Rollins, 1941). 


148 Rhodora [Vol. 97 


BioLocicaL Notes. Arabis lyallii seems to be the alpine and 
subalpine phase of the more widespread A. divaricarpa. It also 
resembles the latter species in that some plants have glabrous 
caudex leaves. Arabis /yallii, at least in our range, 1s more uniform 
morphologically than 4. divaricarpa. Plants of Arabis lyallii from 
Alberta and Utah were triploid, based on x = 7. These must 
therefore produce seed by apomixis. 


17a. Arabis divaricarpa A. Nelson var. divaricarpa. Arabis di- 
varicarpa A. Nelson, Bot. Gaz. 30: 193. 1900; Yellowstone 
Lake, on the stony and sandy banks of the lake, Yellowstone 
National Park, A. & E. Nelson 6622 (LECTOTYPE chosen by 
Hopkins, 1937, RM!; ISOLECTOTYPE GH!). 


Arabis brachycarpa (T.&G.) Britton, Mem. Torre. Bot. Club. 5: 174. 1894. Non 
Ruprecht, Fl. Cauc. 73. 1869. Based on Turritis brachycarpa T.&G., Fl. N 
Am. |: 79. 1838; Fort Gratiot, Michigan and shore of Lake Superior, Dr. 

itcher (HOLOTYPE NY!) 

Arabis Bis Greene, Leaflets Bot. Obs. & Crit. 2: 78. 1910; Sequoia National 

t, California, A. Davidson 1847 (HOLOTYPE Us!). 
Arabs brevis Rydberg, Bull. Torr. Bot. Club 39: 326. 1912; Skagit Valley, 
, J. M. Macoun 70825 (HOLOTYPE NyY!; ISOTYPE CAN!). 
ae Peale var. pratincola (Greene) Hopkins, Rhodora 39: 142. 1937. 
Based on Arabis pratincola Greene, in Feddes Repertorum 5: 244. 1908; 
Spooner, Douglas County, Nevada, C. F. Baker 1149 (IsoTYPES GH!, Mo!). 
Arabis bese sre Torrey var. stenocarpa (Hopkins) Farwell, Papers Mich. 
Sci. 26: 14. 1941. Based on Arabis divaricarpa var. stenocarpa Hopkins, 
ae 39: 133. 1937; Quebec, slaty ridges east of Bic, Rimouski Co., M. 
L. Fernald & J. F. Collins 1057 (HOLOTYPE GH!; ISOTYPE CAN!). 


DISTRIBUTION. Open disturbed areas on cliffs and rock out- 
crops, in woods on shorelines and along roadsides in Alaska (rare), 
southern Yukon, Mackenzie District, New Brunswick (rare), Que- 
bec, Ontario, Manitoba, Saskatchewan, Alberta, British Columbia 
and southward into the United States, particularly in the east. 


17b. Arabis divaricarpa var. dacotica (Greene) B. Boivin, Am. 
Midl. Nat. 54: 510. 1955. Based on Arabis dacotica Greene, 
Leaflets Bot. Obs. & Crit. 2: 80. 1910; Fort Meade, South 
Dakota, W. H. Forwood 28 (HOLOTYPE us!). 


Arabis ae es Rydb., Bull. Torr. Bot. Club 31: 557, 1904, Valley Spur, 

Col o, L. M. Underwood & A. D. Selby 454 (HOLOTYPE NY!). 
— divaricorpa var. hemicylindrica B. Boivin, Am. Midl. Nat. 54: 510. 1955; 
aple Creek, Sask., 10 milles au sud Carmichael, monts Cyprés, écorre de 


1995] Mulligan—Arabis 149 


la coulée du ruisseau Bone, B. Boivin & J. Alex 9738 (HOLOTYPE DAO!). A 
specimen with the identical label data, in sask, is Arabis pinetorum. 


DISTRIBUTION. Open, dry, often sandy, disturbed areas of 
prairie, grasslands, hillsides and shorelines in southeast Alaska, 
central and southern Yukon, Mackenzie District, Quebec (rare, 
Co. Témiscamingue), northern Ontario, Manitoba, Saskatche- 
wan, Alberta, British Columbia and southward into the United 
States, particularly from Minnesota westward. 

BIOLOGICAL Notes. It is morphologically the most variable 
of our Arabis species. The siliques range from ascending to de- 
scending, surfaces of rosette or caudex leaves may be glabrous to 
densely covered with sessile to short-stalked 3-rayed trichomes. 
The 3-rayed trichomes are from less than 0.25 mm in diameter 
to more than twice that wide and rays are appressed to leaf surfaces 
to divergent. It is often difficult to distinguish some western plants 
of A. divaricarpa from Arabis lyallii, particularly fragmentary ma- 
terial. However, complete and mature plants of A. /yallii, a more 
montane plant, can be reliably separated from A. divaricarpa. 
Although there is some morphological overlap between var. di- 
varicarpa and var. dacotica, they seem to have slightly different 
habitats and distributions. Plants of var. divaricarpa from Sas- 
katchewan, Alberta, British Columbia, Montana and California 
were diploid or tetraploid, based on x = 7, whereas plants of var. 
dacotica from the Mackenzie District, Manitoba, British Colum- 
bia and Colorado were diploid, triploid or tetraploid, with the 
same base number (Table 1). One of the diploids formed 7 pairs 
of chromosomes at meiosis. Bécher (1969) reported that EMCs, 
PMCs and pollen grains of A. divaricarpa plants from California 
had the aneuploid number of 2n = 22 and dyads and uniform 
pollen grains were formed. Mosquin (Table 1) found that tetra- 
ploid plants from Manitoba had an irregular meiosis. In addition, 
some plants have supplementary chromosomes (2 = 13 + 2B). 
It, therefore, appears that 4. divaricarpa has diploids, triploids 
and tetraploids, that some of the diploids are probably sexual 
selfers, that triploids are apomictic, and that at least some of the 
tetraploids are also apomictic. The reproductive system of A. 
divaricarpa is obviously very complex. Rollins (1983) suggested 
that 4. divaricarpa can either: a. be considered a natural hybrid 
taxon that includes stable populations of ancient hybrid origin, 
less stable more recently produced hybrids, and many hybrid 
populations produced under intermediate conditions; or b. be 


150 Rhodora [Vol. 97 


considered a polytypic species of ancient hybrid origin. I have 
observed no morphological evidence of any recent interspecific 
hybridization in our Arabis species. 


18. Arabis lemmonii S. Watson, Proc. Am. Acad. 22: 467. 1887; 
Lassen’s Peak, Lemmon 23 (LECTOTYPE GH!). 
Arabis latifolia (S. Wats.) Piper, Contrib. U.S. Nat. Herb. 295. 1906. Based on 


rabis canescens Nutt. var. /atifolia Watson in King, Geol. Expl. Fortieth 
eas 17. 1871; Clover Mts., Nevada, 11,000 ft., S. Woison Ti (oto: 


Hi). 
Arabs neers Greene, Leaflets Bot. Obs. & Crit. 2: 71. 1910; Galena Creek, 
hoe Co., Nevada, P. B. Kennedy 1248 (HoLotyPe us!; IsoTYPE NY!). 
pn arebcall Greene, Leaflets Bot. Obs. & Crit. 2: 73. 1910; Beaverfoot Mts., 
Selkirk & Rk 


oose lava gravel, Mount Thielson, Cascade Mountains, Oregon, F. V. 
Coville & E. I. Applegate 454 (HoLotyPE us!; IsoTYPE RM!). 

Arabis seg Greene, Leaflets Bot. Obs. & Crit. 2: 75. 1910; Farwell Gap, 
California, C. A. Purpus 5229 (HOLOTYPE us!; IsOTYPES GH!, UC!). 

Arabis sae Rydb., Fl. Rocky Mts., 316. 1918; Clover Mountain, above 
Garfield, Colorado, W. W. Eggleston 6013 (HOLOTYPE NY!). 


DISTRIBUTION. Rock and gravel alpine slopes in Yukon (rare, 
Mt. Archibald), southwestern Alberta and southern British Co- 
lumbia. Occurs southward in Montana, Idaho, Washington, Wy- 
oming, Colorado, Utah, Nevada, Oregon and California (Rollins, 
1941) 

BIoLoGIcAL Notes. Arabis lemmonii plants from Alberta, 
British Columbia and Wyoming were reported to be diploid, based 
on x = 7, and from Montana were found to be triploid, with the 
same base number. The latter plants, producing large amounts of 
seed, were obviously apomictic. 


19. Arabis drepanoloba Greene, Pittonia 3: 306. 1898; Devil’s 
Head Lake, Banff National Park, Alberta, Macoun 1719a 
(HOLOTYPE ND!; ISOTYPE Us!). 


DISTRIBUTION. Alpine meadows and ridges in Colorado, Wy- 
oming, Montana, southwest Alberta, southeast British Columbia 
and disjunct to southwest Yukon. 

BIOLOGICAL Notes. Although there are no chromosome counts 
for this species, I suspect that it belongs to Arabis species with 
the base number x = 7. It is most similar to 4. /emmonii but I 


1995] Mulligan— Arabis 151 


think that the two entities should be treated as separate species 
because of the much larger trichomes and siliques in A. drepan- 
oloba. 


20. Arabis codyi G. Mulligan, sp. nov. 


Arabis codyi ab A. drepanoloba et ab A. lyallii differt ea angustioribus 1 .5— 
1.75 mm latis et trichomatibus furcatis potius quam radiati 


Arabis codyi differs from A. drepanoloba and A. lyallii by its 
narrower, 1.5 to 1.75 mm wide, siliques and forked, rather than 
rayed trichomes. 

The holotype specimen in the herbarium of Agriculture Canada, 
Ottawa, is named after William James Cody (1922-_ ). Yukon, 
Kaskawulsh nunatak, jct. N and central arms Kaskawulsh Glacier, 
W of Kluane Lake, 6000 ft., unstable slopes, D. F. & D. B. Murray 
72, 1 July-1 August, 1965 (HOLOTYPE DAO!). 

Perennial with a simple or branched caudex; stems erect to 
ascending, one to several, simple, 7 to 14 cm high; cauline leaves 
entire, rarely few toothed, glabrous or with scattered short-stalked, 
unbranched to few branched, 2- and 3-forked trichomes from 
0.125 to 0.25 mm long, leaves sessile, mostly sagittate clasping 
stems, narrowly lanceolate to lanceolate, 0.5 to 1.5 cm long, 1.5 
to 3.0 mm wide; caudex leaves entire to rarely with | to 2 shallow 
lobes towards apex, numerous and rosulate, 0.75 to 1.5 cm long, 
1.5 to 3.0 mm wide, the blades narrowly oblanceolate to oblan- 
ceolate, slender petiolate; lower surfaces of caudex leaves densely 
pubescent with short stalked, unbranched to few-branched, 2- 
and 3-forked trichomes from 0.125 to 0.25 mm long; inflores- 
cences symmetrical, open; sepals purplish, oblong, saccate at bas- 
es, 2.5 to 3.0 mm long, 1.5 to 1.75 mm wide; petals purplish, 
cuneate, 6.5 mm long, 2.0 mm wide; fruiting pedicels ascending, 
straight, 3 to 5 mm long; siliques ascending, straight to slightly 
curved, 2.0 to 3.75 cm long, 1.5 to 1.75 mm wide, abruptly 
tapering at apex to a rudimentary style. 

The only other specimen seen was collected in British Colum- 
bia: Perow, B.C., 54°30'’N, 126°26’W, growing on sandy beach, 
700 m, Taylor & Levis 468 (UBC). 


21. Arabis murrayi G. Mulligan, sp. nov. 


Arabis murrayi a ceteris Arabibibus praeditis foliis mediis caulinis basi cuneatis 
vel raro infirme auriculatis, siliquis erectis rachidi adpressis differt. A. murrayi 


152 Rhodora [Vol. 97 


plerumque A. lyallii confusa est, specie praedita foliis caulinis valde sagittao-am- 
plexicaulibus. 


Arabis murrayi differs from other Arabis that have the bases of 
middle cauline leaves cuneate to rarely weakly auriculate, by its 
erect siliques which are appressed to the rachis. It is usually mis- 
identified as A. /yalli, a species with strongly sagittate-clasping 
cauline leaves. 

The holotype specimen in the herbarium of Agriculture Canada, 
Ottawa, is named after David F. Murray, University of Alaska, 
College. Yukon, Kaskawulsh nunatak, jct. N. and central arms 
Kaskawulsh Glacier, W of Kluane Lake, 6000 ft., D. F. & B. M. 
Murray 91b, 1 July-1 August, 1965 (HoLoTyPE DAO!; ISOTYPE 
ALA!). 

Perennial with a simple or branched caudex; stems erect, one 
to several, simple, 2.5 to 15.0 cm high; cauline leaves entire, 
glabrous or with scattered short-stalked, unbranched to few 
branched 2- and 3-rayed trichomes mostly 0.125 mm wide, leaves 
sessile, cuneate to rarely weakly auriculate at bases, lanceolate, 
0.3 to 1.0 cm long, 1 to 3 mm wide; caudex leaves numerous and 
stiffly ascending, 0.5 to 1.75 cm long, 0.75 to 1.5 mm wide, the 
blades narrowly oblanceolate, slender petiolate; lower surfaces of 
caudex leaves with scattered to dense short-stalked, unbranched 
to few branches, 2- and 3-rayed trichomes mostly 0.125 mm wide; 
inflorescences symmetrical, short, few-fruited, congested; sepals 
slightly purplish, oblong to narrowly oblong, nonsaccate at bases, 
3 mm long, | mm wide; petals whitish, linear-oblong, 6.5 mm, 
0.5 mm wide; fruiting pedicels erect, straight to slightly curved, 
3 to 4 mm long; siliques erect, appressed to rachis, straight, 2.0 
to 4.5 cm long, 1.5 to 2 mm wide, style ca. 0.125 mm long; seeds 
oval, in 2 rows, brownish, 1.0 mm long, 0.75 mm wide, promi- 
nently winged only at apex, cotyledons accumbent. 

DISTRIBUTION. Alpine slopes in southwestern Alberta, south- 
ern British Columbia, Washington and probably Idaho and west- 
ern Montana. It apparently has a disjunct distribution to the 
southwestern corner of Yukon. 

BIOLOGICAL Notes. Nearly all of the 4. murrayi material that 
I have examined in herbaria were misidentified as 4. /yallii and 
many sheets were mixtures of the two taxa. Arabis murrayi has 
cauline leaves that are cuneate to rarely weakly auriculate; tri- 
chomes on undersurfaces of caudex leaves, if present, are un- 


1995] Mulligan—Arabis [53 


branched to few branched, 2- and 3-rayed, and mostly 0.125 mm 
wide; and siliques are erect and appressed to the rachis. Arabis 
lyallii has cauline leaves that are strongly sagittate-clasping the 
stems, trichomes on undersurfaces of caudex leaves, if present, 
are unbranched, 3-rayed, and mostly 0.25 mm wide; and siliques 
are strongly ascending. The surfaces of the caudex leaves of both 
species are often glabrous. 

REPRESENTATIVE SPECIMENS. CANADA. Yukon: Kaskawush 
nunatak, jct. N. and central arms Kaskawush Glacier, W. of Kluane 
Lake, unstable slopes, 6000 ft., Murray & Murray 414A, also con- 
tains plants of A. drepanoloba labelled 4/]B (ALA, CAN). Alberta: 
Crypt Lake, Waterton Park, open slope, 6600’, Scotter 98818, 
also contains plants of A. /yallii labelled 988/A (DAO). British 
Columbia: Akamina Ridge on B.C.-Alberta border, 49°01'N, 
114°04’W-114°07'W, occasional on rocky shade exposed summit 
ridge between 8000’ & 8400’, Taylor et al. 3547 (DAO), Cathedral 
Lakes, Ashnola Dist., 49°N, 120°15W, rock slide on shoulder of 
Pyramid Mt., 7000’, cor. purple, Taylor 1370 (usc); Finlayson 
Peak overlooking Maselpanik Creek, 7200 feet, purple flowers, 
Pinder-Moss & Hamlyn 1157 (usc); Blackwell Peak, north of 
ranger station along Hope-Princeton Highway, occasional on 
rocky, east-facing slope, alt. 6300’, flowers purplish-blue; Ca/der 
& Saville 10551A, also contains plants of A. /yallii labelled 1055/B 
(DAO). UNITED STATES. Washington: Yakima Co., alpine slopes of 
Mt. Aix, Snoqualmie Nat. Forest, 7000 feet, Thompson 15045a 
(DAO); Chelan Co., ridge 2 mi. W of Hoodo Pass and ca. 16 mi. 
SSW of Twisp, in dry herbfield, alpine zone, aspect S20W, slope 
20%, 6830', Douglas & Douglas 4143 (DAO). 


22. Arabis exilis A. Nelson, Bull. Torrey Bot. Club. 26: 123. 
1899; Evanston, Wyoming, 4. Ne/son 4523 (LECTOTYPE RM), 
centre specimen (A) best fits description of A. exilis, outer 
two specimens (B) are Arabis holboellii var. retrofracta; 
ISOLECTOTYPE GH!). 

Arabis pendulocarpa A. Nelson, Bot. Gaz. 30: 192. 1900; On cliffs and rocky 

ridges, Madison River, Yellowstone National Park, A. & E. Nelson 5504 


(LECTOTYPE RM! by Rollins 1941; IsoLEcTOTYPES GH!, NY!, RM!, a ‘aie 
holboellii var. pendulocarpa (A. Nels.) Rollins, Rhodora 43: 446. 


DISTRIBUTION. Onrocky and grassy subalpine slopes and prai- 


154 Rhodora [Vol. 97 


rie in southcentral Yukon, Cypress Hills Saskatchewan, Cypress 
Hills Alberta, southwest Alberta and British Columbia. Herbar- 
ium specimens were also seen from Montana, Washington, Utah, 
Colorado and Nevada. It undoubtedly occurs elsewhere in the 
western United States. 

BIOLOGICAL Notes. Plants from Yukon, British Columbia and 
Montana were diploid, based on x = 7, whereas those from Utah 
were triploid (Table 1). One of the diploids from British Columbia 
formed 7 pairs of chromosomes at meiosis (Taylor and Taylor, 
1977). This species probably contains sexual and apomictic plants. 
Bocher (1969, p. 146) grew material of this species in experimental 
plots alongside plants of A. holboellii and concluded that it should 
no longer be included in the 4. holboellii complex. 


23. Arabis microphylla Nuttall in T.&G., Fl. N. Am. 1:82. 1838; 
ocky Mountains, Nuttall (HOLOTYPE PH!). 


— eae, Watson, Proc. Am. Acad. 26: 124. 1891; Gravelly banks, Eagle 

of Revelstoke B.C., J. Macoun, May 13th, 1890 (HOLOTYPE Gu!; 

uit ES CAN!, MO!, PH!, Us!). Arabis microphylla var. macounii (Watson) 
Rollins, Rhodora 43: 428. 1941. 

Arabis paupercula Greene, Leaflets Bot. Obs. & Crit. 2: 77. 1910; Farwell Gap, 

sacar aes California, 10,600, C. A. Purpus 522942 (HOLOTYPE us!; IsoryPEs 


!, uct). 
arabs tenuicula Greene, ney Bot. Obs. & Crit. 2: 82. 1910; Crevices of cliffs, 
oO ”. C. Cusick 1124 (HoLotyPe us!; ISOTYPE GH!). 

a i var. ‘Saxioniena Rollins, Rhodora 43: 429. 1941; Granitic 
hillsides, Porcupine Creek, near Medicine Mountain, Big Horn County, Wy- 
oming, alt. 8500 ft., L. O. & R. Williams 3264 (HOLOTYPE GH!). 

Arabis lemmonii var. ere Rollins, Rhodora 43: 384. 1941; Rocks, Mount 
Paddo, Washington, 6 or 7000 ft. alt., W. N. Suksdorf 509 (HOLOTYPE GH)). 


DISTRIBUTION. In crevices on rock cliffs and large boulders in 
southcentral British Columbia and, according to Rollins (1941), 
from Montana and Wyoming to Nevada and Washington. 

BIOLOGICAL Notes. This species has the chromosome number 
n= 7 and 15/2 and 2n = 14 and 15 (Table 1). Bécher (1969) 
stated that plants from Wyoming had PMCs forming dyads, and 
anaphases initiating dyads were all regular with 15 chromosomes. 
Arabis microphylla obviously contains apomictic diploid plants 
with the base number x = 7, some of which have an aneuploid 
number of 2n = 15. This species seems most closely related to 
Arabis depauperata. 


1995] Mulligan —Arabis 155 


24. Arabis ei eaciaees Nelson & Kennedy, Proc. Biol. Soc. Wash. 
14: 35. 6; Summit Mt. Rose, Washoe Co., Nevada, P. B. 
a 1167 (ISOTYPE uc!). Arabis lemon var. depau- 
perata (Nelson & Kennedy) Rollins, Rhodora 43: 384. 1941. 


Arabis interposita Greene, Leaflets Bot. Obs. & Crit. 2: 78, 79. 1910; Ashland 
Butte, Siskiyou Mts., and Crater lake, Cascade Mts., southern Oregon, Wm. 
C. Cusick 2970 (LEctotyPe us!, plant second from left; IsoLECTOTYPE GH!, 
plant on left). Arabis divaricarpa var. interposita (Greene) Rollins, Rhodora 
43; 378. 1941 

Arabis acutina Greene, Leaflets Bot. Obs. & Crit. 2: 82. 1910; On an open rock 

slope, Mount Thielson, Cascade Mts., issue F. V. Coville & E. I. Applegate 
(LECTOTYPE Us!; ISOLECTOTYPES RM '), 

Arabis ee Greene Leaflets Bot. Obs. & Crit. 2: 73. 1910; Northwestern 

ing, J. N. Rose 1893 (HOLOTYPE Uus!). 

Arabis sess Macbr. & Payson, Contrib. Gray Herb 49: 62. 1917; ae 
flat, alt. 9000 ft., Smoky Mts., Blaine Co., Idaho, J. F. Macbride & E. B 
Payson 3772 (HoLoryre GH!; Isor YPES UC!, Us!, a Arabis microphylla var. 
nubigena (Macbr. & Payson) Rollins, Res., St., State Coll. Wash. 4: 40. 1936. 

Arabis microphylla var. thompsonii Rollins, Rhodora 43: 429. 1941. Kittitas Co., 

Washington, alpine meadows of Table Mt., 5000 ft., J. W. Thompson 9266 
(HoLotTyPE GH!; IsoTyPEs NyY!, Us!). 


DISTRIBUTION. On rocky alpine slopes in southern British Co- 
lumbia. I have seen specimens from Montana, Nevada and Cal- 
ifornia. It almost certainly also occurs in Idaho, Washington and 
Oregon. 

BIOLOGICAL NOTES. <Arabis depauperata contains diploid plants 
based on x = 7 (Table 1). It seems most closely related to A. 
microphylla. 


25a. Arabis holboellii Hornem. var. holboellii, Arabis holboellii 
Hornem., Fl. Dan. 2: 5. 1827; plate 1879. 
Arabis holboellii var. tenuis Bocher, Svensk Bot. Tidskrift 48: 38. 1954; ex Ilwd- 
linguaq sinus Sendre Stromfjord, Greenland, 7. Bocher, 22, 5, 1951 (HOLOTYPE 
c, not seen 


DISTRIBUTION. Open habitats in Greenland. 
25b. seis pen var. consanguinea (Greene) G. Mulligan, 
om v. Based on Arabis consanguinea Greene, Pittonia 


4: 190. "1900: Los Pinos, southern Colorado, 7000 ft., C. F. 
Baker 342 (HOLOTYPE ND!; ISOTYPES ND!, RM!, US!). 


DISTRIBUTION. Open grassland and scrub in Alaska (rare), Yu- 


156 Rhodora [Vol. 97 


kon (rare), Saskatchewan (rare), Alberta, British Columbia, Wash- 
ington, Oregon, California, Nevada, Colorado and Utah. 


25c. Arabis holboellii var. retrofracta (Graham) Rydberg, Con- 
trib. U.S. Nat. Herb. 3: 484. 1896. Based on Arabis retrofracta 
Graham, Edinb. Phil. Journ. 345. 1829; Rocky Mountains, 
Palliser’s Brit. N. Am. Expl. Expedition, E. Bourgeau, 1858 
(PROVISIONAL LECTOTYPE GH!, chosen by Hopkins, 1937 and 
confirmed by Rollins, 1941; no specimen referred to in Gra- 
ham’s original description and there is no suitable material 
in the Royal Botanic Garden in Edinburgh). 


Arabis rhodantha Greene, Pittonia 3: 155. 1897; Above Empire, Colorado, E. L. 
Greene, 1875 (HOLOTYPE ND!, sheet 6327; IsoryPe Np!, left specimen on sheet 


Arabis tenuis Greene, Pittonia 4; 189. 1900. On mountains, 2000 ft. alt., W. 
Klickitat, Washington, W. N. Suksdorf 15 (HOLOTYPE ND!; IsoTyPE Gu!). 
oo lignipes A. Nelson, Bot. Gaz. 30: 191. 1900; Madison River, Yellowstone 

ational Park, 4. & E. Nelson 5505 (LecTtoTyPE RM! by Rollins, 1941; 

een ee MI). 

Arabis caduca A. Nelson in Coulter & Nelson, New Man. Bot. Rky. Mts. 229. 

09; Woods Creek, Wyoming, 4. Nelson 2584 (HOLOTYPE RM!). 

Arabis polyantha Greene, Leaflets Bot. Obs. & Crit. 2: 80. 1910; Along R.R. track 
at Rock Island, Washington, K. Whited 1043 (HOLOTYPE Us!). 

Arabis macdougalii Rydberg, Bull. Torr. Bot. Club 3: 326. 1912; Old Sentinel, 
near Missoula, Montana, MacDougal 191 (HOLOTYPE NyY!; IsoTyPEs NY!, Us!). 

Arabis retrofracta var. multicaulis B. Boivin, Can. Field-Nat. 65: 17. 1951; left 
bank of Malique R. at Fish Hatchery, Jasper National Park, G. H. Turner 
5086 (HOLOTYPE DAO!). 


DISTRIBUTION. Grass slopes, talus, benches, roadsides, hill- 
sides and many other open habitats in Alaska, Yukon, Mackenzie 
District, Quebec (rare), Ontario (rare), Manitoba, Saskatchewan, 
Alberta, British Columbia and southward. United States speci- 
mens were seen from Michigan, North Dakota, South Dakota, 
Montana, Wyoming, Colorado, Utah, Idaho, Washington, Ore- 
gon, Nevada and California. 


25d. Arabis holboellii var. secunda (Howell) Jepson, Man. FI. PI. 
Calif. 430. 1925. Based on Arabis secunda Howell, Erythea 
3: 33. 1895; Mount Adams, Washington, 7. Howell 1487 
(HOLOTYPE ORE!; ISOTYPES NyY!, Us!). 


Arabis collinsii Fernald, Rhodora 7: 32. 1905; Limestone—conglomerate cliffs and 
ledges, island—headland east of Baptiste Michaud’s, Bic, Rimouski Country, 


1995] Mulligan —Arabis 157 


Quebec, J. F. Collins & M. L. Fernald July 16 & 18, 1904 (HOLOTYPE GH!; 
IsOTYPES CAN!, DAO!, GH!, MT!). Arabis holboelii var. collinsii (Fernald) Rollins, 
Rhodora 43: 445. 1941. Arabis retrofracta var. collinsii (Fernald) B. Boivin, 
Can. Field-Nat. 65; 17. 1951. 


DISTRIBUTION. Prairie, grasslands, sand, and rocky areas in 
Alaska, Yukon, Mackenzie District, Quebec (rare, but common 
in Gaspé), Ontario (rare), Saskatchewan, Alberta, British Colum- 
bia and southward. Specimens were seen from Washington, Ida- 
ho, Montana and Utah. It probably occurs elsewhere in the west- 
ern United States. 

BroLtoaicaL Notes. Arabis holboellii is nearly as widespread 
as A. divaricarpa and is certainly more abundant within its range. 
I have recognized four taxa of A. holboellii at the varietal level, 
all of these appearing to contain sexual and apomictic diploids 
and apomictic triploids, based on x = 7 (Table 1). I suspect that 
the sexual diploids are self-compatible. 


26. Arabis boivinii G. Mulligan, sp. nov. 
Arabis divaricarpa var. dechamplainii B. Boivin, Naturaliste Can. 94: 645. 1967; 


Cap au Cobeau, Bic, rochers maritimes calcaires, 4. A. De Champlain 1577 
(HOLOTYPE DAO!; ISOTYPE MT!). 


Arabis boivinii ab A. lignifera different inflorescentia leviter secunda, trichom- 
atibus in foliorum radicalium paginis inferioribus plus quam 0.25 mm latis et 
habitu bienni vel breviter perenni. 


Arabis boivinii differs from Arabis lignifera by its slightly secund 
inflorescence, its wider trichomes on undersurfaces of caudex 
leaves (1.e., more than 0.25 mm wide), and its biennial or short- 
lived perennial growth habit. 

The holotype specimen in the herbarium of Agriculture Canada, 
Ottawa, is named after Bernard Boivin, 1916-1985 (see tribute 
to Boivin by Cody and Cayouette, 1986). Saskatchewan, District 
de Maple Creek, Carmichael, 10 milles au sud, Monts Cyprés 
écorre de la coulée du ruisseau Bone, inflorescence seconde ou 
parfois distique, B. Boivin & J. Alex 9738, 8 juillet, 1952 (HOLOTYPE 
DAO!; IsoTYPE MT!). Arabis divaricarpa A. Nels. var. hemicylin- 
drica B. Boivin, Amer., Midl. Nat. 54: 510. 1955 (same type as 
A, boivinii). 

Biennial or short-lived perennial with a simple, compact cau- 
dex; stems erect, usually single, simple to few branched, 30 to 60 


158 Rhodora [Vol. 97 


cm high; cauline leaves entire to rarely few toothed, glabrous to 
pubescent, sessile, mostly strongly sagittate-clasping stems, nar- 
rowly lanceolate, 1.0 to 2.5 cm long, 3 to 5 mm wide; caudex 
leaves entire to few toothed, compact, 1.5 to 2.0 cm long, 3 to 6 
mm wide, blades lanceolate, slender petiolate; lower surfaces of 
caudex leaves with sparse to dense sessile to short-stalked, 
branched, 3- to 4-parted trichomes mostly 0.35 mm wide; inflo- 
rescences semisecund, open; fruiting pedicels arcuate-spreading 
to arcuate-descending, 3 to 9 mm long; siliques slightly descending 
to strongly descending, straight to slightly arcuate, 4.0 to 6.5 cm 
long, 1.5 to 2.0 mm wide, style rudimentary; seeds mostly in 1 
row, brownish, oval, 1.75 mm long, 1.25 to 1.5 mm wide, nar- 
rowly winged at apex, cotyledons accumbent. 

DISTRIBUTION. Dry prairie and hills of southern Saskatche- 
wan, Montana and South Dakota and probably elsewhere on the 
plains of the United States; disjunct eastwards to limestone cliffs 
and ridges of Cap aux Corbeaux, Rimouski County, Quebec. 

BIOLOGICAL Notes. Arabis boivinii is a triploid based on x = 
7 (Table 1). Morphologically it seems somewhat intermediate 
between Arabis holboellii and Arabis divaricarpa. Although both 
of these taxa occur within the range of A. boivinii, there is no 
evidence that it has resulted from recent hybridizations. I am 
treating it as a species resulting from the ancient hybridization of 
A. holboellii and A. divaricarpa. Rollins (1983) has suggested that 
A. divaricarpa itself is probably of ancient hybrid origin. 

REPRESENTATIVE SPECIMENS. CANADA. Quebec: Cap aux Cor- 
beaux, Comté de Rimouski, sur le conglomerat nu, Rousseau 
26440 (DAO, MT); Point aux Corbeaux to Cap Caribou, Bic, lime- 
stone and limestone-conglomerate ridges, Fernald & Collins 1061 
(CAN). Saskatchewan: Cypress Hills, side hill on Bald Butte, Budd 
2005 (SASK); Cypress Hills Park, open plateau, occasional, Brei- 
tung 4401 (MT); 12 miles south of Indian Head, dry south-facing 
knoll, Jones 784 (SASK). UNITED STATES. Montana: Cascade Co., 
3.8 miles south of Neihart, fresh road cut, clay-gravel soil, Mos- 
quin & Gillett 5219 (DAO). South Dakota: Lawrence Co., 2 miles 
north and '2 mile west of Savoy, steep south facing eroding slopes, 
Mosquin & Mulligan 5157 (DAO). 


27. Arabis lignifera A. Nelson, Bull. Torr. Bot. Club. 24: 123. 
1899; Green River, Sweetwater Co., Wyoming, A. Nelson 
4711 (HOLOTYPE RM!; ISOTYPES GHI, Mo)). 


1995] Mulligan — Arabis 159 


Arabis densicaulis A. Nelson, Bot. Gaz. 30: 190. 1900. In rocky exposed places 
n abrupt slope, Undine Falls, Yellowstone National Park, A. & E. Nelson 

3680 was designated as the type for both A. densicaulis and Arabis elegans 

A. Nelson (Bot. Gaz. 30: 191. 1900). However, the specimens in RM and Mo 

fit the description of A. densicaulis not A. elegans. A note on the isotype of 

A. densicaulis in RM, written by C. L. Porter °51, states that ‘‘also cited as a 
type of A. elegans A. Nels., Bot. Gaz. 30: 192, 1900 but this was an error for 

6939.” 


Arabis subserrata Greene, Leaflets Bot. Obs. & Crit. 2: 79. 1910; Ellensburg, 
Washington, K. Whited 32] (HOLOTYPE us!; ISOTYPE ws!). 


DISTRIBUTION. Open rocky areas in northwest corner and in 
southern British Columbia and southward. It also grows in Idaho 
and Wyoming to Colorado and west to Nevada (Rollins, 1993b). 

BIOLOGICAL Notes. Arabis . contains diploid plants 
with the base number x = 7 (Tab 


28. Arabis sparsiflora Nuttall in T.&G., Fl. N. Am. 1: 81. 1838; 
Forests of Rocky Mountains, towards sources of Oregon 
(LECTOTYPE a specimen on extreme right of herbarium 
sheet) 


Arabis sparsiflora var. arcuata (Nuttall) Rollins, Res. Studies State Coll. Wash. 5: 
26. 1936. Based on Streptanthus arcuatus Nuttall in T.&G., Fl. N. Am. 1: 
77. 1836; Santa Barbara, Nuttall (HOLOTYPE PH!; IsoTYPE GH!). 

Arabis sparsiflora var. peramoena (Greene) Rollins, Res. Studies State Coll. Wash. 
5: 26. 1936. Based on Arabis peramoena Greene, Feddes Repertorium 5: 242. 
1908; Dry sandy soil of Willow Creek, Malheur Co., Oregon, purple showy 
flowers, W. C. Cusick 2309 (IsoTYPES NY!, RM!, Us! 

Arabis campyloba Greene, Pittonia 4: 192. 1900; Near Paci. California, F. L. 
Greene, April & May 1876 (HOLOTYPE ND!). 

Arabis polystricha Greene, Leaflets Bot. Obs. & Crit. 2: 72. 1910; Dry hill near 
near Yreka, Siskiyou Co., California, Butler 723 (HOLOTYPE ND!; IsoTYPES 
ND!, P!, RSA-POM!, UC!). 

Arabis arcoidea A. Nelson, Bot. Gaz. 53: 220. 1912; Loamy creek banks among 
hills, altitude 2200, near Plymouth, Canyon County, Idaho, J. F. Macbride 
87 (HOLOTYPE RM!; ISOTYPES GH!, RM! 

Arabis sparsiflora var. californica Rollins, Rhodora 43: 402. 1941; California, on 
dry hills, near Campo, San Diego County, L. Abrams 3563 (HOLOTYPE GH!). 


DISTRIBUTION. Occurs in sagebrush and on dry benches in 
southern British Columbia (only near Penticton), Idaho, Oregon, 
Arizona, Nevada, California and Washington. 

BIOLOGICAL Notes. One diploid population, with the base 
number x = 7, growing in Arizona was apparently sexual, whereas 
another was probably apomictic (Table 1). Raven et al. (1965) 


160 Rhodora [Vol. 97 


and Bocher (1969) reported 2” = 22 for plants from California. 
Bocher observed the configuration of 7I] + 8] at metaphase I. It 
appears that sexual diploids and apomictic diploids and deviant 
triploids, based on x = 7, occur in this species. 


29. Arabis columbiana Macoun, Mac. Cat. 5: 304. 1890: Gravel, 
Yale, B.C., Macoun, May 17th, 1889 (LECTOTYPE CAN!; 
ISOLECTOTYPE GH!). Arabis sparsiflora var. columbiana (Ma- 
coun) Rollins, Rhodora 43: 405. 1941 


Arabis pl var. subvillosa Rollins, Rhodora 43: 403. 1941. Based on Arabis 
ta Gray var. subvillosa Watson in Gray, Syn. Fl. N. Am. 1: 164. 1895; 
ulman, Washington, C. V. Piper, May 20, 1894 (HOLOTYPE GH!). 
Arabis elegans Nelson, Bot. Gaz. 30: 192. 1900; Undine Falls, Yellowstone Na- 
tional Park, 4. & FE. Nelson 6939 (LECTOTYPE RM!). 
Arabis stokesiae Rydb., Fl. Rocky Mts. 361. 1918; Parley’s canyon, Wahsatch 
5000 ft., S. G. Stokes, June 8, 190] (HOLOTYPE us!; ISOTYPE GH!, 


DISTRIBUTION. Grasslands, sagebrush, rock outcrops, talus 
slopes, and open woods in Yukon, British Columbia, Washington, 
Idaho, Oregon, California, Nevada and Utah. 

BIOLOGICAL Notes. It appears that sexual and apomictic dip- 
loids and apomictic triploids, based on x = 7, occur in A. col- 
umbiana (Table 1). 


30. Arabis pinetorum Tidestrom, Proc. Biol. Soc. Wash. 36: 182. 
23; In coniferous forest north of Glenbrook, along Lake 
Tahoe, Nevada, elev. 1890 meters, J. Tidestrom 10387 
(HOLOTYPE GH!). Arabis holboellii var. pinetorum (Tidestrom) 
Rollins, Rhodora 43: 447. 1941. Arabis divaricarpa var. pi- 
netorum (Tidestrom) B. Boivin, Can. Field-Nat. 65: 16. 1951. 


DISTRIBUTION. Open slopes, dunes and prairie in Alaska, Yu- 
kon, Mackenzie District, Manitoba, Saskatchewan, Alberta and 
British Columbia. Specimens were seen from North Dakota, Utah, 
Wyoming, Nevada and California. It undoubtedly occurs else- 
where in the western United States. 

BIOLOGICAL Notes. Plants with the somatic chromosome 
numbers of 14, 13 + 2B and 21 occur in A. pinetorum (Table 1). 
One of the sites, in California, had diploid plants with an irregular 
meiosis. It, therefore, seems likely that apomicts, based on x = 
7, occur both at the diploid and triploid level. 


1995] Mulligan — Arabis 161 


ACKNOWLEDGMENTS 


I would like to thank the many persons who supplied me with 
living material, herbarium specimens and gave me other assis- 
tance. I also thank Dr. Jacques Cayouette for the Latin diagnoses, 
Dr. Suzanne I. Warwick for her helpful advice, Jean Clost and 
Sue Porebski for typing the manuscript, and Gisele Mitrow for 
her assistance with the preparation of SEM photographs. 


LITERATURE CITED 


BERKUTENKO, A. N. AND N. N. GurzENKov. 1976. Chromosome numbers and 
distribution of Cruciferae in the south of the Magadan region. Bot. Zurn 
SSSR 61: 1595-1603 (In Russian). 

BOCHER, W. 1947. Cytological studies of Arabis holboellii Hornem. Hereditas 
eK 


: ie Cytological and embryological studies : the amphi-apomictic 
— holboellii complex. Biol. Skr. Vid. Selsk. 6: 1-59. 
: . Experimental oo studies in the Arabis holboellii complex. 
as Bot. Tidsskr. 48: 31-4 
1966. Experimental and ee studies on i species IX. Some 
arctic and montane Crucifers. Biol. Skr. Vid. Selsk. 14: 1-7 
. 1969. Further studies in Arabis holboellii and oe species. Saertryk 
Bot. Tidsskr. 64: 141-161. 
Borvin, B. 1951. Centurie de Plantes Canadiennes—II. Can. Field-Nat. 65: 1- 
22. 


1955. Notes on Arabis. Amer. Midl. Nat. 54: 510. 
1966. Enumération des plantes du Canada. II]—Herbidées, 1° partie: 
Digitatae: Dimera, Liberae. Nat. canadien 93: 583-646. 
67. Enumération des plantes du Canada. VI—Résumé statistique et 
régions adjacents. Nat. canadien 94: 625-655 
1968. Flora of the Prairie Provinces. Part II. Digitatae, Dimerae, Liberae. 
Phytologia \ 265-339. 
Burpet, H. M. 1967. Contribution a l’étude caryologiques des genres Carda- 
minopsis, Sea et Arabis en Europe. Candollea 22: 107-156. 
Copy, W. J. AND J. CayoueTte. 1986. A tribute to Bernard Boivin, 1916-1985. 
Can Field-Nat. 100: 280-288. 
AN . 1991. A tribute to James Alexander Calder, 1915-1990. 
Can. Field-Nat. 105: 584-591. 
DALGAARD, V. 1988. Chromosome numbers in some vascular plants from the 
Disko Bugt area (West Greenland). erode 18: 243-252. 
Dawe, J. C. AND D. F. Murray. 1979. In IOPB chromosome number reports 
LXIII. Taxon 28: 265-279. 
EASTERLY, N. W. 1963. Chromosome numbers of some northwestern Ohio 
Cruciferae. Castanea 28: 39-42. 
GALLAND, N. 1969. In IOPB chromosome number reports XXII. Taxon 18: 
433-442 


162 Rhodora [Vol. 97 


HEDBERG, O. 1967. Chromosome numbers of vascular plants from arctic and 
subarctic North America. Ark. f. Botanik 6: 320. 

Hitt, M. 1982. In IOPB chromosome number reports LX XIV. Taxon 31: 119- 
128. 


Hovmaren, P. K., N. H. HOLMGREN AND L. C. BARNETT. 1990. Index Herba- 
riorum. Part 1. The herbaria of the world, 8th ed. New York Bot. Garden, 
New York. 

Hopkins, M. 1937. Arabis in eastern and central North America. Rhodora 39: 
63-98, 106-148, 155-186. 

HuttTen, E. 1958. The Amphi-Atlantic plants. Kung. Sv. Vet.-Akad. Hand. 7: 
1-340, 


1968. Flora of Alaska and neighbouring Territories. Stanford University 

pia Stanford, CA. 

1971. The circumpolar plants. II. Dicotyledons. Kung. Sv. Vet.-Akad. 
andl: 13: 1-463. 

JOHNSON, A. W. AND J. G. PACKER. 1968. Chromosome numbers in the flora 
of Ogotoruk Creek, N.W. Alaska. Bot. Not. 121: 403-456. 

JORGENSEN, C. A., T. SORENSEN AND M. WESTERGARD. 1958. The flowering 
plants of Greenland. A taxonomical and cytological survey. Danske Vidensk. 
Selsk. Biol. Skr. 9: 1-172. 

KovanpbA, M. 1978. Chromosome numbers of miscellaneous United States 
dicotyledons. Rhodora 80: 431-440 

ve, A. AND D. Love. 1975a. In IOPB chromosome number reports L. Taxon 


ie 671-678. 

AND . 1975b. Nomenclature notes on arctic plants. Bot. Not. 128: 
513. 

AND . 1982. InIOPB chromosome number reports LX XIV. Taxon 
31: 119-128. 


MULLIGAN, G. A. 1964, Chromosome numbers of the family Cruciferae I. Can. 
J. Bot. 42: 1509-1519. 
AND A. E. Porsitp. 1969. Chromosome number of some plants from 
the jena ventral Yukon plateau, Canada. Can. J. Bot. 47: 655-662. 
1970. In IOPB chromosome number reports XXV. Taxon 


ts i. 1-112. 

PACKER, J. G. 1964. Chromosome numbers and taxonomic notes on western 
Canadian and arctic plants. Can. J. Bot. 42: 473-494. 

AND R. Wirkus. 1982. In IOPB chromosome number reports LXXV. 
Taxon 31: 342-368. 

RAVEN, P. H., D. W. KyHos ann A. J. Hitt. 1965. Chromosome numbers in 
some Spermatophytes, mostly Californian. Aliso 6: 105-113 

Ropman, J. E. AND M. BHARGAVA. 1976. In IOPB chromosome number reports 
LIII. Taxon 25: 483-500. 

Ro .uiins, R.C. 1936a. ned genus Arabis L. in the Pacific Northwest. Res. Studies 
State Coll. Wash. 4: 1-52. 

—. 1936b. Notes on oe Madrono 3: 359-362. 

. 1941. A monographic study of Arabis in western North America. Rho- 

dora 43: 289-325, 348-411, 425-481. 

. 1943. Generic revisions in the Cruciferae: Halimolobos. Contrib. Dudley 
Herb. 3: 241-288. 


1995] Mulligan — Arabis 163 


. 1946. Some new or noteworthy North American Cruciferae. IJ. Contrib. 
eee Herb. 3: 366-373. 
Chromosome numbers of Cruciferae. Contrib. Gray Herb. No. 
197: ee 
—. 1971. Protogyny in the Cruciferae and notes on Arabis and Caulanthus. 
Gray Herb. No. 201: 3-10. 
1973. Purple-flowered Arabis of the Pacific coast of North America. 
Conteb: Gray Herb. No. 204: 149-157. 
1981. Studies on Arabis (Cruciferae) of western North America. Syst. 
Bot. 6: 55-64. 
1982. Studies on Arabis (Cruciferae) of western North America. II. 
Contrib. Gray Herb. No. 212: 103-114 
1983. Interspecific hybridization and taxon uniformity in Arabis (Cru- 
ciferae). Amer. J. Bot. 70: 625-634. 
. 1984. Studies in oe Cruciferae of western North America IJ. Contrib. 
Gray Herb. No. 214: 1-18. 
. 1993a,. New jae a names in Cruciferae of California. Harvard Papers 
in Botany 4: 43-48. 
1993b. The Seite of Continental North America. Stanford Univ. 
Press, Stanford, Californ 
AND L. RUDENBERG. 1971. Chromosome numbers of Cruciferae II. Con- 
trib. sete’ Herb. No. 201: 117-133. 
1977. Chromosome numbers of Cruciferae HI. Contrib. 
Gray ay Herb. No. 207: 101-116. 
—— anp ———. 1979. Chromosome numbers of Cruciferae IV. Bussey Inst. 
Harvard Univ. 79-91. 
SABOURIN, A. 1989. Guide des Cruciféres sauvages de l’est du Canada (2° partie). 
Quatre-temps ee 39- 32. 
SMITH, F. H. 1938. Sor numbers in the Cruciferae. Amer. J. Bot. 
> 220-221. 
TAYLor, R. L. AND R. P. BROCKMAN. 1966. Chromosome numbers of some 
western Canadian plants. Can. J. Bot. 44: 1093-1103 
AND G. A. MULLIGAN. 1968. Flora of the Queen Charlotte Islands. Part 
2. Cytological aspects of vascular plants. paneen: s Printer, Ottawa, Ont. 
AND S. TAYLOR. 1977. f vascular plants of British 
Columbia. Syesis 10: 12, 5 Sere 
WIEBOLT, T. E. 1987. The shale barren endemic Arabis serotina (Brassicaceae). 
Sida 12: 381-389. 


CENTRE FOR LAND AND BIOLOGICAL RESOURCES RESEARCH 
AGRICULTURE CANADA 

OTTAWA, ONTARIO KIA 0C6 

CANADA 


RHODORA, Vol. 97, No. 890, pp. 164-170, 1995 


A NATURAL HYBRID OF DROSERA ANGLICA HUDS. 
AND DROSERA LINEARIS GOLDIE IN MICHIGAN 


DONALD E. SCHNELL 


ABSTRACT 


Plants of the natural hybrid of Drosera anglica Huds. and D lj is Goldie 
were discovered in a fen in Chippewa County, Michigan and are herein reported 
for the first time. Leaf aspects of the hybrid are intermediate between the parents 
and are best expressed as a leaf blade length/width ratio. Multiple measurements 
of the ratio in the parents and putative hybrid indicate an intermediate state. 
Chromosome evaluation indicates 2n = 30 (D. anglica 2n = 40, D. linearis 2n = 
20 in the literature). Pollen viability is very low and no seeds are produced in the 
plants. The corolla is slightly larger in the hybrid than in either parent. In the 
field, the hybrid appears very vigorous and grows on two low mossy hummocks 
in a calcareous fen. In culture in the greenhouse, the hybrid plants show heterosis 
and leaf out and flower earlier in the season than the putative parents. 


Key Words: Drosera anglica, Drosera linearis, Drosera anglica x D. linearis, 
hybrids 


INTRODUCTION 


While botanizing in a typical calcareous Great Lakes fen along 
the Lake Huron shoreline in the eastern tip of Michigan’s upper 
peninsula, I noted two stands of Drosera which at first glance 
appeared to be particularly vigorous plants of Drosera anglica 
Huds. As I prepared to photograph the plants, reexamination 
indicated that the plants might be Drosera linearis Goldie. Since 
the plants seemed to have leaf characteristics of both species, I 
then conjectured that they might be hybrids of D. anglica and D. 
linearis. 

A search of the literature indicated that a natural hybrid of 
these two species had not yet been characterized, although pos- 
tulated by Wood (1955). Kusakabe (1979) submitted a list of 
artificial Drosera hybrids which he had prepared in his green- 
house. Among these was ‘“‘D x linglica’’ (Sic; quotes mine), pro- 
duced in 1976 using D. /inearis from Ontario and D. anglica from 
Munich Botanical Gardens. There is no record of “D. x /inglica”’ 
having been published horticulturally or botanically. 


164 


1995] Schnell— Drosera hybrid 165 


HABITAT DESCRIPTION 


The location is near the shore of Lake Huron in Chippewa 
County, Michigan. First seen in June 1987, the plants are in a 
typical Great Lakes calcareous or marl fen of rich type. Such fens 
are described in detail elsewhere (e.g., Cruise and Catling, 1974; 
Schnell, 1982; Crum, 1988). 

This rather large fen is nearly a hectare in extent and consists 
of marly sand and peat overlaid by a 1-3 cm layer of very slowly 
flowing water originating from springs at the fen margins. Scat- 
tered across this flat are variably sized and spaced hummocks of 
Sphagnum spp. and other non-sphagnous mosses. The edge of 
the fen is marked by deep stands of similar mosses extending into 
a rather dense surrounding woods of predominantly Picea mar- 
iana (P. Mill.) B.S.P., Thuja occidentalis L., Larix laricina (Du 
Roi) K. Koch and Betula spp. 

Droseras in the fen included D. anglica, D. linearis and D. 
rotundifolia L. D. linearis typically occurs scattered over the marly 
flat in water, less commonly on hummocks. D. rotundifolia usually 
occurs in sphagnum on the hummocks. D. anglica grows pref- 
erentially around the bases of hummocks but may appear in the 
moss. 

The plants in question were occupying the entirety of two hum- 
mocks approximately 0.5 meter across and located 5 meters apart. 
They crowded the entire surface of each hummock to the exclu- 
sion of other Droseras and numbered between 100-200 plants 
each (Figure 1). 

Specimens were collected and pressed 21 June 1987 for my 
personal herbarium, then submitted to US as my specimen num- 
ber 870621-/ at the conclusion of studies on 30 July 1994. A few 
plants were also collected to be grown and observed in cultivation. 
The area was also revisited and observed over the intervening 
years. 


MATERIALS AND METHODS 


Vernier calipers were used to measure corolla diameters as well 
as leaf blade length (petiole excluded) and maximum width. The 
numbers of leaves from as many plants are indicated by m 1n the 
Table. Means, medians and standard deviations (SD) were cal- 
culated. 


166 Rhodora [Vol. 97 


igure 1. Putative Drosera anglica Huds. x D. linearis Goldie in flower in 
Chippewa County Michigan. 


Root squashes for chromosome counts were attempted but were 
unsuccessful due to a peculiar crystalline material within cells that 
interfered. Tissue sections of roots disclosed few mitoses. Young 
flower buds (2-3 mm) were fixed in alcohol:acetic acid (3:1), 
dehydrated using standard histologic technique and then imbed- 
ded in parafhin. Ten micron sections were made and stained with 


Table 1. Leaf blade length/width ratios of putative hybrid and parents. n— 
number of leaves from as many plants, mean—average of L/W’s, SD—standard 
deviation of ratios, range—one SD around mean, lowest and highest L/W’s— 
lowest and highest ratios in the series, median—from list of lowest to highest 
L/W’s in each category. 


D. anglica x 
D. linearis linearis D. anglica 
n 40 47 4] 
Mean 16.3 9.0 5.4 
3.6 ee) 1) 
Range-1| SD 12.7-19.9 7.5—10.5 2.4-8.4 
Lowest L/W 10.6 5.6 3.1 


Highest L/W 26.6 12.0 9.7 
Median 


1995] Schnell— Drosera hybrid 167 


hematoxylin and eosin. There was no ovule or PMC mitotic ac- 
tivity, but many somatic mitoses were noted in the carpels. Those 
cut at the best angle for counting were selected and amounted to 
24 figures in buds of seven flowers from seven different plants. 

Surrogate pollen viability was determined by staining with lac- 
tol phenol cotton blue. Pollen was placed on a glass microslide, 
stain added, the pollen mixed with the stain, the whole cover- 
slipped and set aside for three hours. 

Plants were successfully cultivated in the greenhouse in south- 
western Virginia in a wet medium of equal parts coarse sand and 
peat. Minimum greenhouse temperature was 7-8°C. Samples of 
D. anglica and D. linearis were kept in similar cultural conditions 
beside the putative hybrid plants. 


RESULTS AND DISCUSSION 


Basic morphometric studies were undertaken to find the sim- 
plest and most useful demonstration of whether the study plants 
were hybrids of the putative parent species. Flowers were found 
to have corollas 1.5 to 2.0 mm larger on the average than either 
D. linearis or D. anglica (both of which measure 6-7 mm across). 
Leaves of the putative hybrids were on the whole 5-6 mm longer 
than leaves of D. linearis. Neither of hee features were helpful 
in precise differentiation for study purposes. 

Noting that the leaves of D. linearis are linear i in character while 
those of D. anglica are obovate to elongate-spatulate, I required 
measurements to take into account leaf blade width and length 
in one term. The simplest measurement term in my opinion was 
leaf blade length divided by width (L/W) in order to obtain an 
index of minimum complexity. 

The resulting ratios, means, medians and SD’s are listed in the 
Table, and means and SD ranges compared in Figure 2, with 
“average” leaves shown in Figure 3. The study plants fall into an 
intermediate position between the two putative parent species 
with minimum overlap. The medians are close to the means 
indicating a closely Gaussian distribution of samples about the 
means. I would expect a hybrid of the two species to have these 
intermediate leaf characters. The larger flower and generally but 
insignificantly longer leaf of the hybrid will be mentioned later. 

Chromosome counts on carpel sections were 2n = 30. All North 
American Droseras are 2n = 20 except D. anglica which is 2n = 


168 Rhodora [Vol. 97 


D linearis Hybrid D. anglica 
Leaf blade L/W ratios, 1 SD 


Figure 2. Comparison of leaf blade length/width (L/W) ratios with highest and 
lowest in each category, and mean indicated by tick in center of each bar. 


40 due to its amphiploid hybrid origin (Wood, 1955). These counts 
are also consistent with a hybrid between the two putative parent 
species. 

Usually, North American Drosera spp. self pollinate when the 
flower closes at the end of one to two days if pollination has not 
been effected by another agent. Drosera hybrids are typically ster- 
ile (pers. obs.; Wood, 1955). The study plants, examined at the 
conclusion of each of five growing seasons, were never seen to set 
seed in nature or in cultivation while both putative parent species 
readily did so. Pollen staining indicated that less than 10% of the 
grains in several different preparations from different study plants 
stained minimally. Pollen samples from the two species had great- 
er than 95% intense staining. Examination of withered flowers of 
the study plants at the conclusion of anthesis disclosed a few 
empty testae with no seeds present. These findings support the 
hybrid origin of the plants in this setting. 

In cultivation, the study plants began spring growth from winter 
hibernacula three to four weeks prior to either putative parental 
species. Growth was vigorous and similar to those plants in na- 
ture. The hibernaculae frequently budded so that the plants re- 
produced vegetatively. This activity along with minimally in- 
creased corolla diameter and leaf length is attributed to heterosis. 
I have observed similar hybrid vigor in other North American 
Drosera hybrids. 

I conclude that the leaf blade L/W ratios, chromosome counts, 


1995] Schnell— Drosera hybrid 169 


Figure 3. Photo comparison of typical leaves of D. sites (top), putative 
hybrid D. anglica x D. linearis (middle), and D. linearis (bottom). 


flower sterility, and evidence of heterosis all indicate that the study 
plants are of hybrid origin involving D. anglica and D. linearis 
as the parents. 


LITERATURE CITED 


Cruise, J. E. AND P. M. CATLING. 1974. The sundews (Drosera spp.) in Ontario. 
Ont. Field Biol. 28: 1-6. 

Crum, H. A. 1988. A focus on a ie peat mosses. The University of 
Michigan Press, Ann Arbor. 306 pp. 


170 Rhodora [Vol. 97 


KuSAKABE, I. 1979. Japanese Drosera hybrids. Carnivorous Plant Newsletter 8: 


SCHNELL, D. E. 1982. Notes on Drosera linearis Goldie in northeastern lower 
Michigan. Castanea 47: 313-328. 

Woop, C. E. 1955. Evidence for the hybrid origin of Drosera anglica. Rhodora 
57: 105-130. 


RT. 1, BOX 145C 
PULASKI, VIRGINIA 24301 


RHODORA, Vol. 97, No. 890, pp. 171-175, 1995 


THE CHROMOSOME NUMBER OF 
SAXIFRAGA GASPENSIS FERNALD 


CAMILLE GERVAIS, NORMAN DIGNARD, AND ROSAIRE TRAHAN 


ABSTRACT 


The chromosome number of Saxifraga gaspensis Fern., a taxon frequently 
included in the circumpolar species S. nivalis L., or considered a variety, was 
determined on material from Mount Logan, Gaspé Peninsula, and found to be 
2n = 40. As the chromosome number of S. nivalis is 2n = 60 (about 30 data from 
diverse countries), it seems that the plant described by Fernald in 1917 is spe- 
cifically different. The authors suggest that S. gaspensis could be a stabilized 
relictual hybrid between S. nivalis and the closely related diploid species S. tenuis 
(Wahlenb.) H. Smith, 2” = 20, or a polyploid originating from S. tenuis. 


Key Words: Saxifraga gaspensis, S. nivalis, S. tenuis, chromosome number, dis- 
tribution, endemism, Gaspé Peninsula, Québec, Canada 


TAXONOMIC HISTORY 


Saxifraga gaspensis was described by M. L. Fernald in 1917 
from material collected in 1906 with J. F. Collins (600, GH) on 
calcareous slopes of Tabletop Mountain (Mount Jacques-Cartier) 
on the Gaspé Peninsula and first distributed in herbaria as S. 
nivalis L. The new species differed from the circumpolar arctic- 
alpine S. nivalis by its “rosette-leaves more narrowly cuneate- 
obovate and more gradually narrowed to a broad petiolar base,” 
its less numerously flowered inflorescences which were spicate- 
racemose rather than spiciform to corymbiform, its shorter calyx 
lobes reflexed in fruit, its acute to subacute and narrower petals 
and its shorter capsules. 

In the following decades, S. gaspensis was discovered in moist 
pockets and ravines on Mount Logan and Mount Blanc, west of 
the type locality, and it was also reported from northern Québec 
and Labrador (Rousseau, 1974). Although still retained as a dis- 
tinct species by Fernald (1950) in the Gray’s Manual, S. gaspensis 
is judged to be a synonym of S. fenuis or of S. nivalis var. tenuis 
Wahlenb. in more recent treatments (Scoggan, 1978; Kartesz, 
1994). Hultén (1971), on the other hand, uses the name S. 
pensis on his maps but states that the species is doubtfully distinct 
from S. nivalis var. tenuis. Only Boivin (1966 and unpubl.) con- 
sidered S. gaspensis [sub S. nivalis var. gaspensis (Fern.) Boivin] 
as a taxon differing from S. nivalis and S. tenuis sensu stricto. 


Ll 


ieee: Rhodora [Vol. 97 


However, in recent catalogs or documented lists of rare, en- 
dangered or vulnerable species in Canada or in Québec (Bouchard 
et al., 1983; Argus and Pryer, 1990; Lavoie, 1992), §. nivalis var. 
gaspensis 1s revived as, at least, a taxon in need of further research. 
For this reason and because of the urgent problem of determining 
if S. gaspensis must be included, or not, in the list of species to 
be legally protected, the present study was undertaken. 


MATERIAL AND METHODS 


Living material of S. gaspensis (some with seeds) was first col- 
lected in late August of 1993 in cold chimneys of Pease Basin on 
the eastern side of Mount Logan, for greenhouse culture. Addi- 
tional specimens were also taken in June 1994 on schistose walls 
of Big Cascade, between Mounts Dodge and Griscom, east of the 
1993 stations. For the cytological studies, fresh root tips were 
selected from potted plants and from seeds germinating on wet 
filter papers in Petri dishes. They were fixed in an acetic acid/ 
absolute alcohol 1:3 mixture before coloration in aceto-carmine 
and squash in a drop of carmine. As a pre-treatment, before 
fixation, the potted plants or the Petri were kept in a refrigerator 
at 4°C for 4 to 7 hours. 


RESULTS AND DISCUSSION 


The chromosome number 2n = 40 (Figure 1) was determined 
on one of the potted plants and two plantlets obtained from seeds 
collected in 1993 in Pease Basin. The same chromosome number 
(2n = ca. 40) was also observed on the 1994 material from Big 
Cascade. This chromosome number differs from the numerous 
counts reported for S. nivalis (2n = 60) and S. tenuis (2n = 20) 
on circumpolar material from Greenland, Iceland, Spitzberg, Nor- 
way, Siberia, Alaska, British Columbia, Northwest Territories, 
etc. with the exception of a report by Krause and Beamish (1973) 
who have found n = 20 on one specimen of this group from the 
Yukon territories (Lapie Lake). These authors do not attempt to 
attach any binomial to the cytotypes (m = 20, n = 10) they have 
found in the Yukon but describe them as belonging to a large S. 
nivalis-tenuis complex. Our chromosome data, however, were 
specifically determined from specimens which were suspected to 


1995] Gervais et al.—Saxifraga gaspensis ee 


Figure 1. Somatic chromosomes of Saxifraga gaspensis Fern.; metaphase with 
2n = 40 chromosomes from a root-tip cell. 


be S. gaspensis, so that we are inclined to consider this taxon as 
basically different from S. nivalis and S. tenuis. 

The chromosome count (7 = 20) of Krause and Beamish (1973) 
in the Yukon possibly belongs to a hybrid individual between S. 
nivalis (2n = 60) and S. tenuis (2n = 20), the two species growing 
together in this region where their differences “seem to break 
down and the plants appear to intergrade”’ as Krause and Beamish 
(1973) point out. The same situation may also prevail in northern 
Québec and Labrador where the mention of “S. gaspensis”’ refers 
perhaps to such hybrids. 

It is quite possible that S. gaspensis shares the same hybrid 
origin but, as S. tenuis (2n = 20) is not reported for the Gaspé 
Peninsula and S. nivalis sensu stricto (2n = 60) possibly does not 
exist there either, we must suppose that S. gaspensis 1s a relictual 
stabilized hybrid able to reproduce and set viable seeds. This 


174 Rhodora [Vol. 97 


event could have occurred when the Gaspé Peninsula climate and 
flora were different and comparable to the actual situation in the 
arctic. Another possibility is that S. gaspensis is a tetraploid off- 
spring of S. tenuis but an ancient origin must also be advocated 
in this case. 

The study of the meiosis in S. gaspensis could perhaps provide 
an answer to these questions but a cytological examination of 
different populations of S. nivalis sensu lato in the Gaspé Pen- 
insula and in northern Québec may be necessary to understand 
the whole problem. It can already be recommended, however, 
that the taxa belonging to the S. nivalis complex in the Gaspé 
Peninsula be protected as an endangered local taxa. 


ACKNOWLEDGMENTS 


The authors would like to express their gratitude to Michéle 
Parent for help in field trips, to Dr Alison Munson for revision 
of the English manuscript, to Jean Gagnon and Gildo Lavoie for 
helpful discussions and to Andrée Rodrigue for the typing of the 
text. 


LITERATURE CITED 


Arcus, G. W. AND K. M. Prver. 1990. Les plantes vasculaires rares du Canada. 
Notre patrimoine naturel. Musée canadien de la Nature, Ottawa. 277 
Borvin, B. 1966. Enumération des plantes du Canada. III. Herbidées. Naturaliste 

can. 93: 583-646. 

BOUCHARD, A., D. BARABE, M. DUMAIS AND S. Hay. 1983. Les plantes vascu- 
laires rares du Québec. Musée national des sciences naturelles, Ottawa. Syl- 
logeus no 48. 79 pp 

FERNALD, M. L. 1917. New species, varieties and forms of Saxifraga. Rhodora 
19: 141-144, 

. 1950. Gray’s Manual of Botany, 8th ed. American Book Co., New York. 
1632 pp. 

Hu.ten, E. 1971. The Circumpolar Plants. II. Dicotyledons. Almqvist & Wik- 
sell, Stockholm. 463 pp 

Kartesz, J. T. 1994. A Synonymized Checklist of the Vascular Flora of the 
United States, Canada, and Greenland. Volume 2. Thesaurus. Timber Press, 
Portland, Oregon. 816 

Krause, D. L. ANp K. I. Gann 1973. Notes on Saxifraga occidentalis and 
closely related species in British Columbia. Syesis 6: 105-113. 

Lavorg, G. 1992. Plantes vasculaires susceptibles d’étre désignées menacées ou 
vulnérables au Québec. Ministére de l’Environnement. Québec, Canada. 180 
pp. 


1995] Gervais et al.—Saxifraga gaspensis ae: 


RoussEAu, C. 1974. Géographie floristique du Québec-Labrador. Distribution 
des principales espéces vasculaires. Presses Univ. Laval, Québec. 799 pp. 

ScoGGAN, H. J. 1978. The flora of Canada, Part 3. National Museum of Natural 
Sciences, Publications in Botany 7(3): 547-1115. 


RECHERCHE EN SCIENCES DE LA VIE 
ET DE LA SANTE 

PAVILLON C.-E. MARCHAND 
UNIVERSITE LAVAL 

QUEBEC, CANADA GIK 7P4 


N. D. 

DIRECTION DE LA RECHERCHE FORESTIERE 
MINISTERE DES RESSOURCES NATURELLES 
COMPLEXE SCIENTIFIQUE D-1-100 

2700, RUE EINSTEIN 

QUEBEC, CANADA GIP 3W8 


RHODORA, Vol. 97, No. 890, pp. 176-178, 1995 
BOOK REVIEW 


Bremer, K. 1994. Asteraceae, Cladistics & Classification. 752 
pp. Timber Press, Inc., 9999 S. W. Wilshire, Suite 124, Port- 
land, Oregon 97225. ISBN 0-88192-275-7. ($79.95, hard- 
cover). 


“.. the book is about cladistics, phylogeny, and evolution of 
the Asteraceae, based on analysis of morphological data. It is also 
a cladistic evaluation of existing subfamilial, tribal, and subtribal 
classification, and a reference to the generic classification of the 
family.”’ From the Preface. 

There are 4 preliminary chapters—Cladistics, pp. 5-11; Clas- 
sification, pp. 13-23; Morphology, pp. 24-35; and Evolution, pp. 
36-46. The bulk of the book, pp. 49-680, is a subfamily by sub- 
family, tribe by tribe, subtribe by subtribe, genus by genus dis- 
cussion of the morphology and classification of the family. There 
is a 46 page list of References, and a 23 page index. 

When reviewing a “new” classification one may emphasize the 
non-congruity of the new classification with the one(s) it is in- 
tended to replace, or one may emphasize the similarities between 
the classifications. Figure | illustrates the evolution of the Clas- 
sification of the Compositae from George Bentham (in G. Ben- 
tham and J. D. Hooker, Genera Plantarum, vol 2(1), 1873) to 
Wagenitz (in the 12th edition of A. Engler, Syllabus der Planzen- 
famillien, 1964), to the classification offered here by Bremer. Ba- 
sically, Bremer and his colleagues have rearranged groups of spe- 
cies and genera that have been recognized as natural groups for 
many years. The innovations consist in the recognition or res- 
urrection of segregate genera and the removal of many genera to 
unfamiliar locations in the great scheme of things. 

One nomenclatural treatment is interesting. There is a sub- 
family Cichorioideae and a genus Cichorium. But Cichorium is 
placed in the tribe Lactuceae, unassigned to a subtribe. According 
to the Code, Article 19, Bremer’s tribe Lactuceae should be Ci- 
choreae (since it includes Cichorium), and there should be a sub- 
tribe Cichorinae to include Cichorium (which is currently “un- 
assigned to a subtribe’’). 


176 


1995] Book Review 177 


Bentham 1873 Wagenitz 1964 Bremer 1994 
Tribes: Subfamily Asteroideae Subfamily Barnades- 
ioideae 
Vernoniaceae Vernonieae Barnadesieae 
Eupatoriaceae Eupatorieae 
Asteroideae Cardueae Subfamily Cichorioideae 
Inuloideae Heliantheae 
Helianthoideae Helenieae Mutisieae 
Helenioideae Senecioneae Cardueae 
Ant ideae Calenduleae Lactuceae 
Senecioideae | Vernonieae 
Calendulaceae Astereae i 
Arctideae Anthemideae Arctoteae 
Cynaroideae Arctotideae 
Mutisiaceae Mutisieae Subfamily Asteroideae 
Cichoriaceae 
Subfamily Cichorioideae Inuleae 
Plucheeae 
Cichorieae Gnaphalicae 
Calenduleae 
Astereae 
Anthemideae 


Senecioneae 
Helenieae 

Heliantheae 
Eupatorieae 


Figure |. A comparison of the classifications of Bentham, Wagenitz and Bre- 


Another point of interest is the omission of any discussion, or 
even mention of, x Solidaster Wehrhahn (= xAsterago T. H. 
Everett), the putative hybrid between Aster ptamicoides and an 
unknown Solidago. The plant has been known and cultivated in 
Europe since about 1910. Such a hybrid, if it does, in fact, have 
that parentage, might lead to some reconsideration of the sub- 
tribal placement of the Aster and Solidago. 

There is here, a great deal of information and many provocative 
ideas. It will be of use for many years to come. However, only 
specialists in the Compositae will find it easy to use. At the prac- 
tical level, the lack of indication of synonymy, the lack of indi- 
cation of exactly where groups have been put, and the lack of 
indication of what species are included in which genera, makes 
the work difficult to use for non-synantherologists. 

Many modern students seem to have forgotten that the two 
basic reasons for any classification are (1) to place similar (? = 
related) species together and (2) to facilitate the identification of 


178 Rhodora [Vol. 97 


individual plants by individuals other than the author(s). Other 
inferences of great interest may well flow from such a classifica- 
tion, but they are secondary to the identification of individual 
plants. 


GORDON P. DEWOLF, JR. 
125 LONG HILL ROAD 
WEST BROOKFIELD, MA 01585 


RHODORA, Vol. 97, No. 890, pp. 179-183, 1995 


RHODORA NEWS & NOTES 


LisA A. STANDLEY 


HIGHLIGHTS OF CLUB MEETINGS 


February, 1995. The 906th Meeting of the New England Botanical 
Club was held at the Harvard University Biological Laboratories, 
with 58 members and guests present. 

Dr. Benito Tan of Harvard’s Farlow Herbarium spoke on “*Phil- 
ippine Mosses: Diversity, Biodiversity, and Wallace’s Line.” Ben 
started by noting that he would attempt to refute two dogmas— 
“if you’ve seen one moss you’ve seen them all” —and the concept 
that mosses are not useful as indicators of ecological change. 

The Philippine archipelago is part of the Malesian region, which 
includes the Indonesian islands and New Guinea. There are sharp 
floristic and faunistic demarcations at the boundaries of, and 
within, this region—hundreds of genera do not extend across the 
boundaries despite close physical distance and no climatic dif- 
ferences. These boundaries generally coincide with deep sea 
trenches, and are now thought to be related to plate tectonic 
history. At least three major internal regions are recognized. Wal- 
lace’s Line divides the east and west subunits of Malesia, and 
generally extends between Borneo and Celebes. The Philippines 
have been placed east, and west, of Wallace’s Line by different 
investigators. Ben has been attempting to answer the question of 
whether Wallace’s Line exists for mosses, and if the moss flora 
supports placing the Philippines in the western or eastern subunit 
of Malesia. He has also been documenting the bryophyte flora 
and floristic relationships of Palawan, an island group extending 
northward from Borneo. 

Ben took the Club on a floristic tour of the vegetation types of 
the Philippines, ranging from low elevation rain forest, seasonally 
dry forest, limestone substrates, serpentine substrates, and high 
elevation cloud forests dominated by epiphytic bryophytes. The 
vascular and bryophyte floras of the Philippines contain species 
and genera that are widely distributed in the paleotropics, display 
eastern asian affinities, or are part of the maleasian/oceanic flora. 
The flora also includes remarkable endemics such as the angio- 
sperm Rafflesia, and the 2-foot high moss Dawsonia. 

The distribution of mosses in the Philippines can be explained 
by two hypotheses. Although the archipelago is not currently close 


179 


180 Rhodora [Vol. 97 


to China, the affinities of the flora are eastern Asian. This is 
explained by the geological history of the region: the Philippines 
were once very close to China and were separated from it by the 
spreading of the South China Sea. The Philippines definitely be- 
long west of Wallace’s Line. The distribution of mosses within 
the archipelago may also be explained by past climatic changes. 
Although we tend to think of rain forests as being ancient, Malesia 
was probably substantially drier during the Pleistocene. Some 
bryophytes now consist of widely disjunct populations, and exist 
in scattered dry refugia among the more recent rain forests. Ben 
closed by noting that the moss flora of the Philippines is now 
threatened by deforestation, for lumber and for agriculture. This 
is especially evident on Palawan, where the more seasonally dry 
forests are being burned to create agricultural fields. 


March, 1995 (907th Meeting). Dr. Michael Donoghue of Harvard 
University was welcomed back to the Club after several years in 
the deserts of the American Southwest. He spoke on his long- 
term studies on the genus Viburnum, particularly on studies on 
the New England species that reveal new information about the 
evolutionary history of the group. The 200-odd species of woody 
shrubs that make up the genus generally fall into a New World 
group, with fruits that develop from yellow through red to ripe 
purple fruits, which fall off. Old World species have ripe red fruits 
which persist on the plant. Arrowwood and highbush cranberry 
are typical representatives of each group. An analysis of cpDNA 
supports the hypothesis that blue fruits arose independantly in 
both groups. Viburnum acerifolium is anomalous, with persistent 
blue fruits, and may be most closely related to a species from the 
Caucasus. Several species of Viburnum also have large, peripheral 
sterile flowers that may be an adaptation to attract pollinators in 
shady understory environments. Mike has been testing this hy- 
pothesis, thus far inconclusively. In summary, Viburnum is an 
old genus that differentiated fairly early, spread geographically, 
and subsequently diverged. Based on Mike’s experience, it is a 
rich source for understanding mechanisms of plant evolution. 


April, 1995 (908th Meeting). Dr. Richard Evans Schultes was the 
annual Distinguished Speaker, speaking on the topic “Amazonia: 


1995] Rhodora News & Notes 181 


80,000 species of plants awaiting ethnobotanical study.” Among 
his many career distinctions, Dr. Schultes was recently awarded 
the Linnaean Society Medal (analogous to the Nobel Prize), and 
has been a Club member since 1937. Dr. Schultes has worked on 
the medicinal flora of the western, Columbian, Amazon forest for 
more than 50 years—and consequently feels more familiar with 
the flora of the Amazon than that of New England. Many of the 
region’s 80,000 species of flowering plants have never been de- 
scribed, let alone investigated. Remarkable features of the region 
include ancient, highly eroded sandstone mountains that support 
endemic species, and rivers with rapids and waterfalls. Dr. Schultes’ 
lifetime of ethnobotanical research has focused on the Native 
American tribes of this region, who are dependent on a wide range 
of plants to meet diverse purposes. Coca is chewed on a daily 
basis, with no addiction, and enables Indians to endure hard 
physical labor and lack of food. Ten species of rubber (the genus 
Hevea) are native to the Amazon, and although used by the local 
tribes, have not been fully exploited. Dick once collected 3 tons 
of seed of a localized ecotype with high productivity, unfortu- 
nately displacing the local movie theater. A local palm, which 
produces an extraordinary number of seeds containing an oil 
similar to olive oil, has not been introduced into cultivation but 
is potentially a major oil crop. Aristolochia medicinalis is a pan- 
acea richly deserving of the specific epithet. Other plants, such as 
Paullina yoco, are used for stimulant beverages. An arrow poison, 
derived from the bark of lianas, has been successfully introduced 
into western medicine as a muscle relaxant. Another bark, con- 
taining rotenone, is used to stun fish. Various plants are used for 
spiritual purposes. Hallucinogens from a variety of sources are 
used by medicine men to communicate with spirits. Snuff, derived 
from the inner bark of a species of Virola (in the nutmeg family) 
is also a powerful hallucinogen. 

Dr. Shultes shared photographs of many of his adventures in 
the Columbian Amazon region over the past 54 years. He danced 
in a grass skirt to scare off demons of bad weather, and breakfasted 
on tapioca bread with chili pepper. He concluded the talk by 
showing photographs of some of the more devastated areas of 
Brazilian deforestation, and urged the audience to visit the Am- 
azon and experience the river and forest ecosystem while at least 
portions of it are intact. 


182 Rhodora [Vol. 97 
GRADUATE STUDENT RESEARCH AWARD 


The 1995 Graduate Student Research Award was presented to 
Peter Walker, a student at the University of Vermont, in support 
of his research entitled ‘‘Speciation in Ammophila: Sequence 
Variation in the Internal Transcribed Spacer of Nuclear Ribo- 
somal DNA.” This research is aimed at understanding the status 
and evolution of Ammophila champlainensis, thought to be a 
distinct taxon endemic to the Lake Champlain basin. 


THE NEW ENGLAND BOTANICAL CLUB 
22 Divinity Avenue 
Cambridge, MA 02138 


The New England Botanical Club is a non-profit organiza- 
tion that promotes the study of plants of North America, es- 
pecially 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, RHO- 
DORA, 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 subscrip- 
tion 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 a 


Name 


Address 


City & State Zip 


Special interests (optional): 


184 


THE NEW ENGLAND BOTANICAL CLUB 
Elected Officers and Council Members for 1995—1996 


President: C. Barre Hellquist, Box 9145, Department of Biology, 
North Adams State College, North Adams, Massachusetts 
01247 

Vice-President (and Program Chair): W. Donald Hudson, Jr., 
Chewonki Foundation, RR 2, Box 1200, Wiscasset, Maine 
04578 

Corresponding Secretary: Nancy M. Eyster-Smith, Department 
of Natural Sciences, Bentley College, Waltham, Massachu- 
setts 02154 

Treasurer: Harold G. Brotzman, Box 9092, Department of Bi- 
ology, North Adams State College, North Adams, Massa- 
chusetts 01247 

Recording Secretary: Lisa A. Standley 

Curator of Vascular Plants: Raymond Angelo 

Assistant Curator of Vascular Plants: Pamela Wetherbee 

Curator of Non-Vascular Plants: Anna M. Reid 

Librarian: Paul Somers 

Council: Consisting of the Elected Officers, Associate Curator, 

Editor of Rhodora and — 
Councillors: Leshe J. Mehrhoff (Past President) 
Thomas Mione °96 
Garrett E. Crow °97 
Edward J. Hehre °98 
Donald J. Padgett (Graduate Student 
Member) °95 


Volume 97, No. 889 including pages 1-108 was issued July 20, 1995 


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), in addition, a copy of the mss. on a 3.5 inch 
floppy disc will facilitate production of the paper. The text must be 
double-spaced (at least %”) throughout, including tables, figure leg- 
ends, and literature citations. The list of legends for figures and maps 
should be provided on a separate page. Do not use footnotes. Do not 
indicate the style of type through the use of capitals or underscoring, 
particularly in the citation of specimens. Names of genera and species 
may be underlined to indicate italics in discussions. Specimen cita- 
tions 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 is 
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, 
Rhodora 81: 503-548, 1979. For bibliographic citations, refer to the 
Botanico-Periodicum-Huntianum (B-P-H, 1968) which provides stan- 
dardized abbreviations for journals originating before 1966. All ab- 
breviations 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 
3% X 6 in.) Magnification/reduction values given in text or figure 
legends should be calculated to reflect the actual printed size. Tables 
should be set up in the same way. Each page should be ready for 
reproduction and will be reduced to 4 X 6 inches. 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. 


+3 


vee. 


A 


RHODORA 


JOURNAL OF THE 
NEW ENGLAND BOTANICAL CLUB 


CONTENTS: 


pee in the North American lake cress, Neobeckia aquatica (Brassica- 
ae): Inferences from chromosome number. Donald H. Les, Gregory 


a Anderson, and Maryke A. Cleland 185 
An undescribed species of oe as: from the state of Wash- 

ington. Reed C. Rollins, Kath A. Beck, and Florence E. 

Caplow 201 
The vascular plants of Fort Devens, Massachusetts. David M. Hunt, Karen 

B. Searcy, Robert E. Zaremba, and C. Roberta Lombardi ............. 208 
Fragaria multicipita, reduced to the rank of forma. Paul M. Catling, 

Jacques Cayouette, and Joseph Postman 245 


Status of the deerberry, Vaccinium stamineum L. (Ericaceae), in Canada. 
Bruce A. Ford 


eames effects of Lantana camara (Verbenaceae) on morning glory 
if a tricolor). Christina M. Casado 264 
New aan Notes 


Studies on New England Algae II: A second station in Maine for Nitella 
tenuissima (Desv.) Kuetzing. L. C. Colt, Jr. 275 


New Barnstable county records. Mario DiGregorio 280 


Book Review 
Plant identification terminology: An illustrated glossary. Leslie J. 
Me 283 
Rhodora News and Notes. Lisa A. Standley 285 


JSS 


THE NEW ENGLAND BOTANICAL CLUB 
P. O. Box 1897, Lawrence, Kansas 66044 
22 Divinity Avenue, Cambridge, Massachusetts 02138 


Vol. 97 Summer, 1995 No. 891 


The New England Botanical Club, Inc. 


22 Divinity Avenue, Cambridge, Massachusetts 02138. 


RHODORA 


DAVID S. CONANT, Acting Editor-in-Chief 
SUZANNE GALLAGHER, Acting Managing Editor 


Associate Editors 
DAVID S. BARRINGTON 
W. DONALD HUDSON, JR. 
LESLIE J. MEHRHOFF 
THOMAS MIONE 
CATHY A. PARIS 
LISA A. STANDLEY 


RHODORA (ISSN 0035-4902). Published four times a year (January, 
April, July, and October) by The New England Botanical Club, 810 
East 10th St., Lawrence, KS 66044 and printed by Allen Press, Inc., 
1041 New Hampshire St., Lawrence, KS 66044. Periodicals postage 
paid at Lawrence, KS. POSTMASTER: Send address changes to 
RHODORA, PO. Box 1897, Lawrence, KS 66044. 

RHODORA is a journal of botany devoted primarily to North America. 
Authors are encouraged to submit manuscripts of scientific papers and 
notes relating to the systematics, distribution, ecology, paleobotany, 
or conservation biology of this or floristically 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, PO. Box 1897, Lawrence, KS 
66044-8897. 


MEMBERSHIPS: Regular $35; Family $45. Application form printed 
herein. 


BACK VOLUMES AND SINGLE COPIES: Some available; informa- 
tion and prices will be furnished upon request to the Treasurer. 

ADDRESS CHANGES: In order to receive the next number of RHO- 
DORA, changes must be received by the business office prior to the 
first day of January, April, July, or October. 

INFORMATION FOR CONTRIBUTORS: Inside back cover, January 
and April. 

MANUSCRIPTS: Send to: Janet R. Sullivan 

Editor-in-Chief, RHODORA 


Durham, NH 03824-3597 


This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). 


RHODORA 


OURNAL OF THE 
NEW ENGLAND BOTANICAL CLUB 


Vol. 97 Summer 1995 No. 891 


RHODORA, Vol. 97, No. 891, pp. 185-200, 1995 


STERILITY IN THE NORTH AMERICAN LAKE 
CRESS NEOBECKIA AQUATICA (BRASSICACEAE): 
INFERENCES FROM CHROMOSOME NUMBER 


DoNALD H. Les, GREGORY J. ANDERSON, AND 
MaArRYKE A. CLELAND 


ABSTRACT 


Sterility of lake cress results from uncertain factors and may be a significant 
element in the decline of populations over the past century. The inability of lake 
cress (Neobeckia aquatica) to produce viable seeds restricts its dispersal to vege- 
tative fragments which are transported less effectively over long distances. We 
obtained mitotic counts of 2n = 24 for individuals from seven populations of lake 
cress, a species for which the chromosome number was unreported previously. 
In context of chromosome number distribution in the Brassicaceae based on 
literature reports for 192 mustard genera, the base number of tribe Arabideae (in 
which lake cress is placed) and of all genera presumed to be closely related to lake 
cress is x = 8. The presence of 24 chromosomes indicates that lake cress is a 
triploid derived from an x = 8 chromosomal series. Highly sterile triploid hybrids 
(2n = 3x = 24) have also been reported in several genera related to lake cress. 
The extremely well-developed system of vegetative reproduction in lake cress may 
partially compensate for its sexual sterility. The discovery that lake cress is triploid 
offers a specific explanation for its sterility and discloses special considerations 
for the conservation of this rare species. 


Key Words: Brassicaceae, Arabideae, Neobeckia, triploid, conservation 


INTRODUCTION 


The monotypic North American lake cress, Neobeckia aquatica 
(Eaton) Greene (Figure 1), is distinguished as one of few truly 
aquatic species in the mustard family, Brassicaceae (Cook, 1990). 
Lake cress is also known for its heterophylly, extreme range of 
vegetative polymorphism, and remarkable ability to regenerate 


185 


186 Rhodora [Vol. 97 


Figure 1. 
dissected submersed foliage, entire emersed foliage, and racemose inflorescence. 
Although fruits form occasionally in this species, they typically (as in this speci- 
men) lack seeds. Drawn from Bryson 8865 (KNK). Bar = | cm 


A specimen of the aquatic mustard Neobeckia aquatica showing 


1995] Les et al.— Neobeckia 187 


from minute fragments of roots, stems and leaves (Foerste, 1889; 
La Rue, 1943; Mac Dougal, 1914). Although lake cress has been 
assigned variously to the genera Armoracia, Nasturtium, and Ro- 
rippa (Al-Shehbaz and Bates, 1987), a recent molecular systematic 
study supports the taxonomic recognition of Neobeckia as a 
monotypic sister genus to Rorippa (Les, 1994). 

Once widely distributed in eastern North America, lake cress 
has become rare as a result of significant population losses 
throughout its former range, particularly in the central portion 
(Stuckey, 1987; Les, 1994). More detailed distributional infor- 
mation has been summarized in Les (1994). Conservation of lake 
cress is presently of concern in several states with categories of 
imperilment including rare (New York; Mitchell, 1986), threat- 
ened (Vermont; Crow et al., 1981), and endangered (New Jersey; 
NJDEP, 1991). Reasons for its rarity are not wholly evident al- 
though several explanations have been offered. 

Habitat destruction or degradation are often cited as factors 
contributing to the present rarity of lake cress (e.g., La Rue, 1943; 
Myers and Henry, 1976; Stuckey, 1987; Swink, 1969). Stuckey 
(1987) attributed the disappearance of the species to turbidity and 
chemical pollution, observing that the greatest loss of lake cress 
populations has occurred in highly agricultural or industrialized 
regions. However, despite the fact that many extant lake cress 
populations occur in fairly pristine habitats, we have observed 
that populations can also thrive in substantially polluted sites 
(Les, pers. obs.). 

Although habitat loss surely is at least partly responsible for 
the decline of lake cress, other factors must be considered. Lake 
cress appears to flourish locally and can become abundant once 
established (La Rue, 1943; Muenscher, 1930; Les, pers. obs.), yet 
the species has never become common (Al-Shehbaz and Bates, 
1987). The local abundance of lake cress has been ascribed to its 
efficient vegetative reproduction (Pringle, 1879; La Rue, 1943) 
and tolerance to a wide range of environmental conditions owing 
to its phenotypic plasticity (Mac Dougal, 1914). 

Pringle (1879) concluded that seeds provide the principal means 
of long-distance dispersal for the species; however, lake cress is 
known to be highly sterile (Foerste, 1881; La Rue, 1943; Mac 
Dougal, 1914; Muenscher, 1930, 1944; Gleason and Cronquist, 
1991; Long and Lakela, 1971; Godfrey and Wooten, 1981; Al- 
Shehbaz and Bates, 1987; McCormac, 1992; Les, pers. obs.). Al- 


188 Rhodora [Vol. 97 


though seeds are produced on occasion (Murley, 1951; Rollins, 
1993), their viability has never been demonstrated. La Rue (1943) 
suggested that the rarity of lake cress might be directly related to 
its poor seed production which reduces the potential for long- 
distance dispersal. 

Rollins (1966) indicated that chromosome numbers often help 
to delimit problematic genera in the Brassicaceae. Because chro- 
mosome counts were previously unreported for Neobeckia aqua- 
tica, Our primary objective was to establish the chromosome num- 
ber for assisting with our assessment of its systematic relationship 
in the mustard family. We hoped that this basic information might 
shed light on the phylogenetic relationships of lake cress by in- 
dicating the specific ploidy level and chromosomal series from 
which it is derived. In addition, the association of sterility and 
high polyploidy in some mustards (e.g., in Dentaria by Mont- 
gomery, 1955), further warranted a cytological examination of 
lake cress. An explanation for sterility in lake cress is wanting 
(Rollins, 1993), in spite of the fact that such information could 
provide important insights to facilitate its recovery and conser- 
vation. 


MATERIALS AND METHODS 


To minimize impact on established plants, only detached lake 
cress leaves or shoot fragments were collected from four popu- 
lations in Vermont during the summer of 1993 (permit obtained 
from Vermont Department of Fish and Wildlife). Ronald L. 
Stuckey and C. B. Hellquist provided us with living lake cress 
specimens from Ohio and Michigan, respectively. Robert W. 
Freckmann provided us with fresh lake cress leaves removed from 
plants collected in Wisconsin from which we regenerated com- 
plete plants (Table 1). Adventitious roots were then obtained by 
removing some of the leaves from the greenhouse plants and 
floating them in tap water until rooting plantlets regenerated (1- 
3 weeks). 

Mitotic figures were obtained from squashes of newly emerged 
adventitious root tip cells following methods employed by Ber- 
nardello and Anderson (1990). Root tip sections 1-2 cm long 
were pretreated at room temperature for two hr in saturated aque- 
ous paradichlorobenzene, then rinsed and fixed at room temper- 
ature for 18 hr ina 3:1 solution (v/v) of ethanol:acetic acid. Fixed 


1995] Les et al.— Neobeckia 189 


Table 1. Chromosome counts obtained for populations of Neobeckia aquatica 
from Michigan, Ohio, Vermont, and Wisconsin, USA. 


Locality Voucher 2n Counts Observed 


MICHIGAN 
Cheboygan Co. 
Mullett Lake, Pigeon 


River Marsh Hellquist 15542 (NAsc) 23, 24, 25 
OnIO 
Franklin Co. 
Hoover Reservoir Stuckey s.n. (CONN) 23, 23/24, 23/24, 24, 
24, 25 
VERMONT 
Addison Co. 
Lake Champlain, 
Catfish Bay Les s.n. (CONN) 23, 23/25, 24, 24, 26 
East Creek Les s.n. (CONN) 23, 24, 24, 24, 24, 24 
Shoreham, Lemon Les S.n. (CONN) 24, 24, 25, 25, 25 
Fair River 


Lake Champlain, Isle 


La Motte Les $.n. (CONN) 24, 24/25 
WISCONSIN 
Marinette Co 
Peshtigo Flowage Freckman s.n. (CONN) 23, 24, 24, 25 


root tips were stained in alcoholic hydrochloric acid-carmine 
(Snow, 1963) for five d. Stained root tips were placed in an aque- 
ous 70% acetic acid solution for 30 min, then macerated, lightly 
heated, and squashed in a drop of 70% acetic acid. Squashes were 
studied with phase contrast and bright field optics using Olympus 
and Zeiss microscopes (the latter equipped with a 63x 1.4 NA 
Planapochromatic objective). 

A minimum of 10 root tips/population furnished the two to 
six mitotic cells selected from each population for chromosome 
counts. Slides were made permanent with Euparal (Bradley, 1948) 
and cells were photographed with a Zeiss Universal microscope 
and Kodak Technical Pan film 

Chromosome numbers were obtained from the literature for 
192 genera in 17 of the 19 tribes recognized by Schulz (1936). 
The complete compilation is available from the authors on re- 
quest. As Manton (1932) did with “fundamental numbers (f),” 


190 Rhodora [Vol. 97 


we selected a reasonable ancestral chromosome number (x) for 
each genus, and from these, we inferred an ancestral number for 
each of the tribes. These determinations took into account dis- 
cussions by Dvorak (1971), Harberd (1976), Manton (1932), Mul- 
ligan (1964, 1965, 1966), Rollins (1963, 1966), Rollins and Rii- 
denberg (1971, 1977, 1979) and Rollins and Shaw (1973). 

To obtain a phylogenetic perspective, we mapped the inferred 
base numbers on a phylogenetic diagram of the cruciferous tribes 
derived from the evolutionary “tree” published by Schulz (1936). 
We also compared base numbers of aquatic genera in tribe Ara- 
bideae that appear closely related to lake cress in a cladogram 
constructed from DNA sequence data (Les, 1994). The placement 
of lake cress in tribe Arabideae (Schulz, 1936) is supported by its 
close relationship to Armoracia, Cardamine, Nasturtium, 
and Korippa (Les, 1994). Schulz included all of these genera in tribe 
Arabideae except for Armoracia which he placed within tribe 
Drabeae 


RESULTS 


Examination of lake cress populations revealed 2n counts of 
23-26 chromosomes (Figure 2, Table 1). In some instances, the 
small size of chromosomes (0.6-0.8 y) and presence of uniden- 
tified stained particles in the vicinity of nucleolar organizer regions, 
made counts difficult. The shape and appearance of the extra- 
chromosomal particles were used to differentiate them from 
neighboring chromosomes. We consistently detected what ap- 
peared to be one pair of chromosomes with small ‘satellite’ regions 
in most preparations; however, inadequate resolution made this 
observation impossible to verify, 

We observed cells of two types. Larger cells showed chromo- 
somes dispersed over a lightly stained cytoplasm; smaller cells 
contained more densely arranged chromosomes against both dark 
and lightly stained cytoplasm. Chromosomes in smaller cells were 
generally easier to count because they lay more in a single plane 
than those of larger cells. 

Three greenhouse plants flowered during Fall, 1995, but we 
were unable to obtain satisfactory preparations for meiotic counts. 
There were six fruits produced on these plants, but complete ovule 
abortion rendered them seedless in every instance. 


1995] Les et al.— Neobeckia 191 


x | ae 


ma 2a 


Figure 2. Example of mitosis in adventitious root cells of Neobeckia aquatica 
(Shoreham, Vermont locality). a: Micrograph showing characteristically small 
metaphase chromosomes. b: Interpretation of micrograph indicates the count 2n 
= 24 chromosomes. Bar = 


Our conclusions of tribal base numbers in Brassicaceae (Figure 
3a) corresponded well with those proposed by Manton (1932), 
Mulligan (1964, 1965, 1966), Rollins (1966), and Rollins and 
Riidenberg (1971, 1977, 1979). In the context of intertribal re- 
lationships proposed by Schulz (1936), a general trend toward 
reduced base numbers is observed from the putatively primitive 
to the more advanced tribes of the Brassicaceae (Figure 3a). 

In 18 genera of tribe Arabideae with available counts, 12 (67%) 
have a likely base number of x = 8 and 4 (22%) a base number 
of x = 7. The base number of two genera (Guillenia and Leav- 
enworthia) could not be reasonably determined. A base number 
of x = 8 characterizes all genera (Armoracia, Cardamine, Nas- 
turtium, Rorippa) putatively allied with lake cress (Figure 3b). 


DISCUSSION 


Although intraspecific aneuploidy is prevalent among aquatic 
angiosperms (Les and Philbrick, 1993), variation in our reported 
counts for lake cress (Table 1) reflects the uncertainty of some 


192 Rhodora [Vol. 97 


Lunarieae X=7 
Alysseae x=8 
Drabeae x=9 
X=8 
Matthioleae x=9 
A . Hesperideae x=9 
Sisymbrieae x=10 


Stenopetaleae X=6 


Euclideae Xx=7 


Lepideae x= 1] 


Schizopetaleae x=12 


Heliophileae x= 10 


Brassiceae x= 12 
Chamireae (?) 
Cremolobeae x= 1] 


— Streptantheae X=7 
~ L. Romanschulzieae (?) 

— Stanleyeae x=14 

a Pringleeae x=12 


Rorippa sylvestris (x = 8) 
Rorippa amphibia (x = 8) 
Neobeckia aquatica (2n = 24) 


Tribe Arabideae 
Armoracia rusticana (x = 8) 
Nasturtium officinale (x = 8) 
Cardamine pensylvanica (x = 8) __| IS . 


Figure 3. Putative phylogenetic relationships of mustards. A: Phylogenetic 
scheme of tribal interrelationships (redrawn from Schulz, 1936) and base chro- 
mosome numbers deduced from literature reports (see text). A box identifies the 
tribe (Arabideae) to which lake cress is assigned. In this interpretation, a general 
decrease in the base number is apparent from the putatively primitive to more 
advanced tribes. B: Phylogenetic relationships of aquatic mustard genera in tribe 
Arabideae (from Les, 1994) with deduced base chromosome numbers and the 2” 


1995] Les et al.— Neobeckia 193 


counts and does not necessarily indicate the existence of discrete 
numbers for the species. We are convinced that the actual chro- 
mosome number of the species is 2n = 24; however, in several 
preparations, it was difficult to clearly view all chromosomes or 
to positively differentiate chromosomes from anomalous inclu- 
sions (see Results). In several instances, this ‘variation’ was ob- 
served among cells from the same individual and from different 
individuals of the same population. Because of the clonal growth 
of this species, and lack of sexual reproduction, our replicate 
counts were not of individual genets but of clonally derived ra- 


mets. 

Stuckey (1972) hypothesized that the genera Armoracia, Car- 
damine, Nasturtium, and Rorippa represented the closest relatives 
of Neobeckia, a conclusion consistent with a recent molecular 
systematic study of the group (Les, 1994). Hayek (1911) placed 
these genera together not only in the same tribe (Arabideae) but 
in the same subtribe (Cardamininae). Literature counts reported 
for these genera uniformly indicate a base number of x = 8. A 
basic number of x = 8 has also been determined for Dentaria, a 
genus sometimes merged with Cardamine (Montgomery, 1955). 
In tribe Arabideae, the base number x = 8 is very common with 
only a few instances of x = 7 or x = 6 (see Harberd, 1976). 

Because they are not cladistically based, the phylogenetic re- 
lationships of mustard tribes (Figure 3a) proposed by Schulz (1936) 
must be interpreted conservatively. However, we are unaware of 
any more recent studies of intertribal relationships in the mustard 
family that are as comprehensive. Molecular systematic studies 
of Arabidopsis by Price et al. (1994) included representatives of 
only 5-6 mustard tribes. Although topologies of the molecular 
cladograms differed in some details from the phylogenetic tree of 
Schulz (1936), the relatively derived position of tribe Arabideae 
and the relatively basal position of tribes with higher basic num- 
bers were consistently indicated. Thus, it 1s reasonable to conclude 
that chromosomal series with basic numbers less than twelve are 
probably derived by descending aneuploidy in advanced tribes 
such as Arabideae. Genera of tribe Arabideae that are closely 


— 
number obtained for Neobeckia. A base chromosome number of x = 8 charac- 
terizes tribe Arabideae and the aquatic genera related to Neobeckia. 


194 Rhodora [Vol. 97 


related to Neobeckia (Les, 1994) share the basic number of x = 
8 (Figure 3b). Given that the most likely base number of tribe 
Arabideae is x = 8, the 2n = 24 chromosome number of Neobeckia 
aquatica indicates that the species is triploid, at least in the pop- 
ulations studied. 

The number 2n = 24 1s rare in tribe Arabideae and is indeed 
associated with triploid hybrids. In Rorippa, spontaneous trip- 
loids (2n = 24) have resulted from hybridization between two 
tetraploids (2n = 32), e.g., R. amphibia (L.) Besser and R. palustris 
(L.) Besser (Howard, 1947; Stace, 1975). The triploid hybrids are 
sterile with 7-8 bivalents and 8-10 univalents, whereas tetraploid 
hybrids are fertile. Mulligan and Porsild (1968) reported a natural 
triploid (2n = 24) hybrid between diploid (2n = 16) R. barba- 
reaefolia (DC.) Kitagawa and tetraploid (2n = 32) R. palustris (= 
R. islandica (Oeder) Borbas). The triploids did not produce seed 
and had only 2% viable pollen. Triploid (2n = 24) hybrids also 
have been produced artificially in crosses between diploid (27 = 
16) R. austriaca (Crantz) Besser and tetraploid (2n = 32) R. syl- 
vestris (L.) Besser (Jonsell, 1968; Javurkova-Kratochvilova and 
Tomsovic, 1972). In one case, only 14% of the seeds produced 
from this cross germinated; however, none of the plants ever 
reached the flowering stage (Jonsell, 1968). Sterile triploid hybrids 
(2n = 24) have also resulted from crosses between diploid (2 = 
16) R. austriaca and tetraploid (2n = 32) R. amphibia (Javur- 
kova-Kratochvilova and Tomsovic, 1972; Jonsell, 1975). 

In Cardamine, triploid (2 = 24) hybrids are produced in cross- 
es between two diploids (2n = 16), C. rivularis Schur and C. 
amara L. (Urbanska-Worytkiewicz, 1977). Reproduction of these 
triploids is mainly vegetative but they are partly fertile with about 
2-3% pollen viability (Urbanska-Worytkiewicz, 1977). 

To enumerate, the triploid chromosome number of lake cress 
is indicated by the common occurrence of the x = 8 base number 
in the tribe Arabideae, the universal occurrence of x = 8 among 
the genera most closely related to lake cress, and the presence of 
the 2n = 24 chromosome number in known triploid hybrids 
within tribe Arabideae. In addition, a significant correlation is the 
extreme sterility of lake cress, a feature long associated with tri- 
ploidy in plants (Darlington and Mather, 1949). The relatively 
low chromosome number of lake cress indicates that sterility is 
not a consequence of high ploidy level as in the related genus 
Dentaria (Montgomery, 1955). 

Sterility in the related horseradish (Armoracia rusticana L.) has 


1995] Les et al.— Neobeckia 195 


been attributed to an interspecific hybrid origin (Weber, 1949), 
self-incompatibility, and accumulation of deleterious mutations 
from prolonged vegetative propagation (Stokes, 1955). Self-in- 
compatibility has also been implicated in the low seed production 
of perennial Rorippa species, presumably due to the clonal growth 
of populations (Jonsell, 1968). 

We also considered that sterility of lake cress may be a con- 
sequence of self-incompatibility (SI). Poor seed set is an inevitable 
consequence for clonal, SI species that experience extreme bot- 
tlenecks (Les, Reinartz, and Essleman, 1991; Reinartz and Les, 
1994). In a survey of the mustard family, Bateman (1955) found 
that SI species occurred in 11/12 tribes surveyed, including the 
genera Armoracia and Cardamine. Self-incompatible species also 
occur in Rorippa (Jonsell, 1968), the most closely related genus 
to Neobeckia. Although self-compatible species have evolved re- 
peatedly in the family, e.g., in Cardamine, Nasturtium and Ro- 
rippa (Bateman, 1955; Jonsell, 1968), SI could certainly be ex- 
pected in Neobeckia. Sterility, however, precludes verification of 
SI in lake cress by experimental crossing studies. Nevertheless, 
triploidy is more likely to represent the proximate cause of sterility 
in the species. 

The origin of triploidy in lake cress remains uncertain. At this 
time, we cannot determine whether the species represents an in- 
traspecific hybrid or an interspecific hybrid resulting from a cross 
between diploids (e.g., Cardamine, above), tetraploids (e.g., Ro- 
rippa, above), or possibly a diploid and tetraploid (e.g., Rorippa, 
above). 

Our study has confirmed that the triploid condition exists among 
different lake cress populations separated in some cases by more 
than 100 km in the northeastern USA, and in considerably more 
isolated populations from Michigan, Ohio, and Wisconsin. This 
indicates that the unusual cytotype is not simply a local, vege- 
tatively propagated abnormality (see Les and Philbrick, 1993) but 
is widespread and at least characteristic of the northern popula- 
tions. To our knowledge, the Wisconsin population represents the 
northwesternmost known station, and the Isle La Motte, Vermont 
population, the easternmost station for the species. The triploid 
number from the Ohio plant is also consistent with observations 
that other Ohio populations of lake cress may flower and fruit 
quite prolifically, yet apparently produce no viable seed (Mc- 
Cormac, 1992), 

We cannot exclude the possibility that diploid or tetraploid 


196 Rhodora [Vol. 97 


cytotypes of lake cress may exist. Potentially, such individuals 
should be more fertile, yet seed set could remain scarce in pop- 
ulations due to confounding factors such as self-incompatibility. 
This is precisely the situation for Apios americana Medikus with 
sterile triploid populations in northern parts of its range, and 
diploid populations in the southern portion of its range in which 
seed set remains low due to self-incompatibility (Bruneau and 
Anderson, 1988). The possible existence of diploid or tetraploid 
lake cress can only be ascertained by further cytological exami- 
nation of populations throughout its range. Our attempts to obtain 
material of this rare species from southern portions of its range, 
however, have thus far proven to be unsuccessful. A comprehen- 
sive survey of lake cress populations is highly recommended be- 
cause the discovery of fertile plants could significantly influence 
conservation strategies for the species. 

Except in the relatively rare cases of partially fertile triploids 
(e.g., Cardamine;, Urbanska-Worytkiewicz, 1977), the sexual re- 
productive capacity of triploid plants is predictably low. The high- 
ly developed system of vegetative reproduction in lake cress pro- 
vides a means of dispersal and reproduction despite the barrier 
to sexual reproduction that is imposed by its triploid cytotype. 
The lack of seed production, however, greatly compromises the 
ability of the species to disperse beyond local distances. Although 
shoot fragments of several other aquatic plants can be dispersed 
considerable distances by flexuous stems that become draped over 
waterfowl, etc. (Sculthorpe, 1967), the brittle nature of lake cress 
plants undoubtedly precludes this avenue of dispersal. Leaves of 
lake cress are the most likely vegetative propagules, and are prob- 
ably difficult to transport over any significant distance. 

Both habitat loss and the low vagility resulting from sterility 
can be linked to the disappearance of lake cress. In the past when 
habitats were abundant, the species may have survived 
despite its poor vagility. Extensive habitat destruction coupled 
with the inability to effectively disperse may now destine the lake 
cress to successive population losses and ultimate demise. It is 
unlikely that vegetative propagules alone can adequately maintain 
dispersal among the remaining fragmented lake cress habitats. 
Without intervention, sites that have experienced local exter- 
minations due to population crashes are likely to remain devoid 
of the species in a progression that may inevitably lead to ex- 
tinction. 


1995] Les et al.— Neobeckia 197 


The realization that sterility in lake cress may be due to specific 
genetic (i.e., chromosomal) factors is important from a conser- 
vation sande oint. For example, it has long been known that 
vegetatively reproducing mustard crops (i.e., horseradish and wa- 
tercress) can be severely damaged by fungal and viral pathogens 
(Crisp, 1976). Such threats are particularly serious for clonal plants 
like lake cress where seed production cannot be relied upon as a 
means of purging pathogens (Silander, 1985). In addition to the 
obvious importance of preserving remaining lake cress habitats, 
the implementation of artificial establishment techniques should 
be considered as a strategy to overcome the dispersal limitations 
imposed on this species by both biological and cultural factors. 


ACKNOWLEDGMENTS 


We thank A. Breisch, W. Brumback, W. Cullina, S. Ek, R. 
Freckmann, C. B. Hellquist, B. Popp, K. Reilly, R. Stuckey and 
E. Thompson for their assistance. Illustrations were prepared by 
M. J. Spring. This work was supported in part by grants to DHL 
from the Vermont Nature Conservancy, and from the University 
of Connecticut Research Foundation. 


LITERATURE CITED 


AL-SHEHBAZ, I. A. AND V. BATES. 1987. Armoracia lacustris (Brassicaceae), the 
correct name for the North American lake cress. J. Arnold Arbor. 68: 357— 


BATEMAN, A. J. 1955. Self-incompatibility systems in angiosperms III. Crucif- 
erae. Heredity 9: 52-68 

BERNARDELLO, L. M. AND G. J. ANDERSON. 1990. Karyotypic studies in Solanum 
section Basarnheun eaten Amer. J. poe 77: 420-4 Bye 

BrabD.ey, M. V. 1948. A method for making aceto hes permanent 
without removal of the seed. Stain Technol. 23: 41-44, 

BRUNEAU, A. AND G. J. ANDERSON. 1988. Reproductive biology of diploid and 
triploid Apios americana (Leguminosae). Amer. J. Bot. 75: 1876-1883. 

Cook, C.D. K. 1990. Aquatic plant book. SPB Academic Pub., The Hague. 

Crisp, P. 1976. Trends in the breeding and cultivation of cruciferous crops. Jn: 
J. G. Vaughan, A. J. Macleod and B. B. G. Jones, eds. The Biology and 
Chemistry of the Cruciferae. Academic Press, NY: 69-118. 

Crow, G. E., W. D. COUNTRYMAN, G. L. CHuRCcH, L. M. EASTMAN, C. B. HE- 
LLQUIST, L. L. [J.] MEHRHOFF ANDI. M.Strorxs. 1981. Rare and endangered 
vascular plant species in New England. Rhodora 83: 259-299. 

DARLINGTON, C. D. AND K. MATHER. 1949. The Elements of Genetics. Schocken 
Books, NY. 446 pp. 


198 Rhodora [Vol. 97 


Dvorak, F. 1971. On the evolutionary relationship in the family Brassicaceae. 
Feddes. Repert. 82: 357-372. 

ForrstE, A. F. 81. Nasturtium lacustre. Bot. Gaz. 6: 264-265. 

1889. Botanical notes. Bull. Torrey Bot. Club 16: 266-268 

GLEASON, H. A. AND A. CRONQuIST. 1991. Manual of Vascular Plants of North- 
eastern ee States and Adjacent Canada, 2nd ed. The New York Botanical 
Garden, Bronx, NY. 

GopFREY, R. Ca AND J. W. Wooten. 1981. Aquatic and Wetland Plants of 
Saithesctern United States: dicotyledons, vol. 2. Univ. Georgia Press, Ath- 


lee) 
paar, 


ens. 

HARBERD, D. J. 1976. Cytotaxonomy of Brassica. In: J. G. Vaughan, A. J. 
Macleod and B. B. G. Jones, eds. The Biology and Chemistry of the Cru- 
ciferae. Academic Press, NY: 47-68 

Hayek, A. von. 1911. Entwurf eines Cruciferen-Systems auf phylogenetischer 
eer Beth. Bot. Centralbl. 27: 127-335. 

Howarp, H. 1947. Chromosome numbers of British species of the genus 
ea (part of the genus Nasturtium R. Br.). Nature 159: 66-67. 
JAVURKOVA-KRATOCHVILOVA, V. AND P. Tomsovic. 1972. Chromosome study 
of i genus Rorippa Scop. em. Reichenb. in Czechoslovakia. Preslia 44: 

140-156. 

JONSELL, B. 1968. Studies in the northwest European species of Rorippa s. str. 
Symb. Bot. Upsal. 19: 1-221. 

1975. Hybridization in yellow-flowered European Rorippa species, [n: 
. M. Walters and C. J. King, eds. European Floristic and Taxonomic Studies. 

. W. Classey, Faringdon, Oxon 101-110. 

LA Be o D. 1943. Regeneration in Radicula aquatica. Michigan Academician 

28: 51-61. 


Les, D. H. 1994. Molecular systematics and taxonomy of lake cress (Neobeckia 
aquatica; Brassicaceae), an imperiled aquatic mustard. Aquatic Bot. 49: 149- 
165 

, J. A. REINARTZ AND E. J. EssELMAN. 1991. Genetic consequences of 

rarity in Aster aha aa a threatened, self-incompatible plant. 

Evolution 45: 1641- 

ND C. T. ne "993, Studies of hybridization and chromosome 
number 1 glOsE y implications. Aquatic 
Bot. 44: 181-228. 

Lona, R. W. AND O. LAKELA. 1971. A Flora of Tropical Florida. Univ. Miami 
Press, Coral Gables, FL. 

Mac DouGat, D. T. 1914. The determinative action of environic factors upon 
Neobeckia a[c/quatica Greene. Flora 106: 264-280. 

193 


MANTON, 2. Introduction to the general cytology of the Cruciferae. Ann. 
Bot. (Oxford) 46: 509-556. 
McCormac, J.S. 1992. Armoracia | tris (Brassi ) redi d in Ohio. 


Rhodora 94: 387-390. 

MiTcHELL, R. S. 1986. A checklist of New York State ae New York State 
Mus. Bull. 458. The State Education Department, Alban 

Gee FH. 1955: eer studies in the genus ere in eastern 
North America. Rhodora 57: 161-173. 

MueEnscHER, W.C. 1930. VIII. ee vegetation of the Lake Champlain wa- 


1995] Les et al.— Neobeckia 199 


tershed. Jn: New York Conservation Department, A Biological Survey of the 
Champlain Watershed. Albany, NY: 164-185. 
1944, Aquatic Plants of the United States. Cornell Univ. Press, Ithaca, 


MuLican, G. A. 1964. Chromosome numbers of the family Cruciferae. I. 
d. J. Bot. 42: 1509-1519. 

Sees. Chromosome numbers of the family Cruciferae. I]. Canad. J. Bot. 

43: 657-668 

66. Chromosome numbers of the family Cruciferae. III. Canad. J. 

Bot. 44: oo 319. 

AND A. E. Porsitp. 1968. A natural first-generation hybrid between 
Rorippa barbara and R. islandica. Canad. J. Bot. 44: 1105-1106. 
Mourtey, M. R. 1951. Seeds of the Cruciferae of northeastern North America. 

Amer. Midl. Naturalist 46: 1-81. 

Myers, R. M. AND R. D. HENRY. 1976. Some changes that have occurred in 
the indigenous flora of adjoining west-central Illinois counties (Hancock and 
McDonough) during the last 140 years. Trans. Illinois State Acad. Sci. 69: 
24 


NJDEP. 1991. The State of New Jersey Endangered Plant Species List. New Jersey 
Department of Environmental Protection and Energy. Trenton, New Jersey. 

Price, R. A., 1. A. AL-SHEHBAZ AND J. D. PALMER. 1994. Systematic relationships 
of Arabidopsis: a molecular and morphological perspective. Jn: C. Somerville 
ane : Meyerowitz, eds. Arabidopsis. Cold Spring Harbor Laboratory Press 
NY: 7-19. 

Panicue: C.G. 1879. Leaf-propagation of Nasturtium lacustre Gr. Bot. Gaz. 4: 
237-238. 


REINARTZ, J. A. AND D. H. Les. 1994. Bottleneck-induced dissolution of self- 
incompatibility and breeding system consequences in Aster furcatus (Aster- 
aceae). Amer. J. Bot. 81: 446-455. 

Rotuins, R. C. 1963. The evolution and systematics of Leavenworthia (Crucif- 
ca Contr. Gray Herb. 192: 3-98. 

966. Chromosome numbers of Cruciferae. Contr. Gray Herb. 197: 43- 


——. 1993. The Cruciferae of Continental North America. Stanford Univ. 
= es 
. RUDENBERG. 1971. Chromosome numbers of Cruciferae. II. 
Conte. es Herb. 201: 117-133. 
b——.. 1977. Chromosome numbers of Cruciferae. III. Contr. Gray 
en 207: 101-116. 
1979. Chromosome numbers of Cruciferae IV. Bussey In- 
ape Cambridge, MA 
AND E. A. SHAw. 1973. The genus as aie (Cruciferae) in North 
pene Harvard Univ. Press, Cambridg A. 
ScHULZ, O. E. 1936. Cruciferae. Jn; A. En olor K. Prantl, and H. Harms, eds. 
Die natiirlichen Pflanzenfamilien, vol. 17B, 2nd ed. Verlag von Wilhelm 


rnold, London. 
SILANDER, J. A. 1985. Microevolution in clonal plants. Jn: J. B. C. Jackson, L. 


200 Rhodora [Vol. 97 


W. Buss, and R. E. Cook, eds. Population Biology and Evolution of Clonal 
Organisms. Yale Univ. Press, New pove en, lee 107-152. 

Snow, R. 1963. Alcoholic hydrocl rmine as a Stain for chromosomes 
in squash preparations. Stain Technol. 38: 9-13. 

Stace, C. A. 1975. Rorippa Scop. In: C. A. Stace, ed. Hybridization and the 
Flora of the British Isles. Academic Press, NY: 150-153. 

Stokes, G. W. 1955. Seed development and failure in horseradish. J. Heredity 

—21,. 

Stuckey, R. L. 1972. Taxonomy and distribution of the genus Rorippa (Cru- 

ciferae) in North America. Sida 4: 279-430. 
1987. Armoracia aquatica disappeared from central portion of its range. 

In: R. L. Stuckey, ed. Aquatic Plants Reference Manual. R. L. Stuckey, 
Columbus, OH: 231. 

Swink, F. 1969. Plants of the Chicago Region. Morton Arboretum, Lisle, IL 

URBANSKA-WorYTKIEWICZ, K, 1977. Reproduction in natural triploid hybrids 
(2n = 24) between Cardamine rivularis Schur and C. amara L. Ber. Geobot. 
Inst. E.T.H. Stiftung Riibel 44: 42-85. 

Weser, W. W. 1949. Seed production in horseradish. J. Heredity 40: 223-227. 


DEPARTMENT OF ECOLOGY AND EVOLUTIONARY BIOLOGY 
UNIVERSITY OF CONNECTICUT 
STORRS, CT 06269-3042 


RHODORA, Vol. 97, No. 891, pp. 201-207, 1995 


AN UNDESCRIBED SPECIES OF LESQUERELLA 
(CRUCIFERAE) FROM THE STATE OF WASHINGTON 


REED C. ROLLINS, KATHRYN A. BECK, AND 
FLORENCE E, CAPLOW 


ABSTRACT 


Recent books covering the flora of Washington admit only one species of Les- 
querella, L. douglasii S. Watson, to the flora of the state. A different, undescribed 
species of this genus has been recognized from recent collections. The new species 
was first collected in 1883 but the specimens from that gathering are incomplete 
and they were never appropriately utilized. The new species is named L. tuplash- 
ensis and it is compared with L. douglasii, its nearest relative. 


Key Words: Hanford Nuclear Reservation, Lesquerella tuplashensis, L. douglasii, 
White Bluffs 


INTRODUCTION 


During the course of a survey of plants within the Hanford 
Nuclear Reservation in south-central Washington, an unexpected 
species of Lesquerella was found on the White Bluffs adjacent to 
the Hanford Reach of the Columbia River. The restricted habitat 
of this species is extremely dry and supports only a sparse veg- 
etation. That the species is unique was determined by comparing 
it in detail with potentially related species of Lesquerella including 
the only other species known from the area, L. douglasii S. Wat- 
son, which occurs from northern Oregon to British Columbia 
(Rollins and Shaw, 1973). A recent taxonomic review of Les- 
querella by one of us (Rollins, 1993) following on a monographic 
treatment of the genus (Rollins and Shaw, 1973) has provided a 
sound basis for interpreting the new material. All other known 
species of Lesquerella were considered as candidates for the un- 
known but were ruled out because they were distinctly different. 
As it turns out, this is not the first time the unknown species had 
been collected. 

In 1883 T. S. Brandegee and Frank Tweedy, working as bot- 
anists for the Northern Transcontinental Survey (Rose, 1904), 
collected the species we here name Lesquerella tuplashensis. The 
Brandegee collection, his no. 635, bears the data, ““White Bluffs 
of the Columbia, Washington Terr., July 1883,” and is repre- 
sented by specimens at GH, NY, and UC. We have seen only 


201 


202 Rhodora [Vol. 97 


fragments of the Tweedy collection, no. 8, at GH, from the same 
locality, June, 1883. Unfortunately, this early material was not 
sufficiently complete, and to some extent anomalous, to be ac- 
curately identified. The Brandegee specimen in the Gray Her- 
barium has been successively named Vesicaria montana Gray?, 
V. occidentalis Watson?, and V. douglasii, the latter annotation 
by C. V. Piper. The generic name Vesicaria was used before Les- 
querella was established. The notation on the label, Vesicaria 
occidentalis, is in Watson’s handwriting and has the specific name 
kingii below it crossed out. The original identifier of the specimen 
and Watson were obviously uncertain about the identity of the 
material. Piper’s annotation is unequivocal and it should be noted 
that he accepted (Piper, 1906) only one species of Lesquerella 
from the Columbia River area. He commented that the Brandegee 
specimens were erroneously associated with L. occidentalis in the 
Synoptical Flora (Gray, Watson, and Robinson, 1895). There 
Watson, who authored the treatment of Lesquerella, did indeed 
associate the Brandegee collection with L. occidentalis, but he also 
wrote, “taller specimens from the White Bluffs of the Columbia, 
Washington (Brandegee), have broadly obovate obtuse fruit and 
may be distinct.” There is no mention of this specimen in Wat- 
son’s earlier treatment (1885, 1888) of Lesquerella. Evidently 
Payson, in preparing his monograph of Lesguerella (1922) did 
not see the Brandegee or Tweedy material, or at least he did not 
mention them. 

In more recent treatments of Lesquere//a (Payson, 1922; Rollins 
and Shaw, 1973; Rollins, 1993) only one species, L. douglasii, is 
attributed to the state of Washington. A rather lengthy discussion, 
concerning specimens from Washington previously referred to 
other species, is given by Rollins and Shaw (1973) under the name 
L. occidentalis. One of the main points made there is that, al- 
though the Brandegee specimen was cited by Watson in his orig- 
inal presentation of L. occidentalis as a new species, and is there- 
fore a syntype, the specimen is not of that species as it is presently 
understood. This discussion should be consulted by anyone se- 
riously interested in the problem. 

In our judgment, there is no doubt that Lesquerella tuplashensis 
is related to L. douglasii and it is our assumption that it was 
derived evolutionally from that species. Because of the silique 
shape, ovule number, and trichome type, L. tuplashensis is re- 
ferable to group 8 (Rollins and Shaw, 1973). Of the eleven species 


1995] Rollins et al.— Lesquerella 203 


included there, L. tuplashensis is most similar to L. douglasii. It 
is perhaps significant that L. tup/ashensis is restricted in distri- 
bution to a specialized habitat and has few plant associates, several 
of limited distribution, whereas L. doug/asii has a much broader 
geographic range and occurs in a variety of habitats with a much 
wider range of associated species. The interpretation that L. tu- 
plashensis is a derived species fits the general situation in Les- 
querella (Rollins and Shaw, 1973; Rollins, 1993) where there are 
numerous examples of a similar relationship between widespread 
species of diverse habitats and apparently derived taxa of limited 
distribution adapted to restricted, usually highly calcareous, sites. 
The rigorous edaphic factors there exerting unusual selection pres- 
sure on plants able to cope with such sites may have contributed 
to their having evolved into distinct taxa. But it must be admitted 
that evolution could have proceeded either way, from widespread 
to restricted species or from restricted to widespread species. 


Lesquerella tuplashensis Rollins, Beck & Caplow, sp. nov. 


TYPE: U.S.A. Washington, Franklin County: White Bluffs, TI3N, 
R27E, $11, W'2, above the Columbia River, caliche soil at edge 
of eroding bluff, 20 July 1994, K. Beck & F. Caplow 94001 (Ho- 
lotype: GH; Isotype: WTU). Associated species: Artemisia tri- 
dentata Nutt., Astragalus caricinus (Jones) Barneby, Cryptantha 
spiculifera (Piper) Payson, Eriogonum microthecum Nutt., and 
Poa sandbergii Vasey. 


Herba perennis multicaulis, caudicibus simplicibus crassiusculis, caulibus er- 
ectis vel prope decumbentibus 1-2 dm altis, foliis basalibus petiolatis integris vel 
sparse lobatis dense pubescentibus argenteis (1.5)2—4(-6) cm longis, foliis caulinis 
imbricatis, sepalis oe mm 1 longis, vos pee anguste lingulatis 4.5-5 

m longis, pedicell 11(-13) mm longis, siliquis 
subglobosis vel fere obovatis 3-4 mm longis, stylis 2. 5a m longis, loculis (2-) 
3-4 ovulatis, seminibus suborbicularibus ca. 2 mm diametro, cotyledonibus ac- 
cumbentibus. 


Perennial, caudex simple, to | cm thick; stems mostly arising 
below the leaf rosette, several to numerous, slightly decumbent 
to erect, densely pubescent, 1—2(-3.5) dm tall; basal leaves ro- 
sulate, densely pubescent, silvery, (1.5)2—4(-6) cm long, rounded 
to broadly obtuse at apex, (0.5—)1-1.5(-2.5) cm wide, outer leaves 


204 Rhodora [Vol. 97 


i 
oy 


Figure 1-7. Lesquerella tuplashensis. Magnifications are approximate. Figure 
1, habit x ‘4; figure 2, silique x7; figure 3, infructescence x%,; figure 4, trichome 
x 12; figure 5, flower x 7; figure 6, replum x 7; figure 7, seed x 7. 


usually entire with nearly orbicular blades and slender petioles, 
inner similar but usually few-lobed or sometimes entire; cauline 
leaves densely overlapping, obovate to spatulate, sometimes nar- 
rower, obtuse to more nearly rounded at apex, petiolate except 
the uppermost, 1-1.5 cm long, 4-8 mm wide; leaf trichomes dense 
and silvery occurring in several layers over the entire surface, 


1995] Rollins et al.— Lesquerella 205 


radiate, appressed, rays mostly forked, fused toward base, ray tips 
15-18; inflorescences usually dense, 3-6 cm long; sepals ascending 
to erect, narrowly oblong, densely pubescent on the exterior, 3- 
4 mm long; petals yellow, narrowly oblanceolate to spatulate, not 
unguiculate, 4.5-5 mm long; fruiting pedicels densely pubescent, 
widely spreading to slightly ascending, straight or nearly so, 9- 
11(-13) mm long; siliques slightly obovoid to subglobose, turgid, 
substipitate, 3-4.5 mm long, densely pubescent on exterior, gla- 
brous on interior, silique trichomes mostly stipitate, not ap- 
pressed; styles 2.5-3 mm long; ovules 2—3(-4) per locule; septum 
entire or rarely slightly perforate; seeds orbicular to broadly ob- 
long, compressed, ca. 2 mm long, immarginate, radical equaling 
cotyledons; cotyledons accumbent. 

ADDITIONAL SPECIMENS EXAMINED: U.S.A., Washington, Frank- 
lin County: White Bluffs, above the Columbia River, T13N, R27E, 
S11, W'%, edge of eroding bluff in shallow caliche soil, 18 Aug. 
1994, Beck & Caplow 94002 (GH, WTU); same general locality, 
July, 1883, T. S. Brandegee 635 (GH, NY, UC); same locality, 
June, 1883, F. Tweedy 866 (GH); T13N, R28E, $33, NW 4 of 
NW '% in hard calcium carbonate ‘‘caliche” with Poa sandbergii 
Vasey, Cryptantha spiculifera (Piper) Payson, Eriogonum micro- 
thecum Nutt., and Eurotia lanata (Pursh) Mog., June 1, 1995, 
Beck & Caplow 95085 (GH); T13N, R27E, $24, NE % of NE 4, 
May 19, 1995, Beck & Caplow 95053 (GH); T12N, R28E, S11, 
NW 4 of SW %, June 1, 1995, Beck & Caplow (GH). 


The name “‘tuplashensis” refers to the White Bluffs of the Co- 
lumbia River where the species occurs. ‘““Tuplash”’ is a place name 
for the White Bluffs in the Sahaptin language, the language of the 
Wanapum Tribe and other tribes whose traditional territories 
include the White Bluffs. ‘“‘Plash” refers to the distinctive white 
color of the bluffs (Relander, 1956). 

In comparing Lesquerella tuplashensis with its related L. doug- 
lasii, we find that most features are similar. Basically the cauline 
leaves of L. tuplashensis are imbricated and there is a range from 
linear to petiolate with a broad rounded blade, while those of L. 
douglasii are loosely arranged and narrowly linear. The basal 
leaves, especially those of the outer margin, of L. tuplashensis are 
more rounded and broader than those of L. douglasii. The most 
noticeable distinction between the two species is in the trichomes 
of the siliques. Trichomes on the exterior surfaces of the silique 


206 Rhodora [Vol. 97 


valves of L. tuplashensis have the radiate portion raised on a 
stipe-like stalk whereas the comparable trichomes of L. douglasii 
are sessile and the radiate portion is appressed to the valve surface. 
In general habit, the plants of L. tuplashensis are more compact 
and denser than those of L. douglasii. 

Lesquerella tuplashensis grows on the upper edge and upper 
face of the White Bluffs adjacent to the Columbia River. The only 
known population is found on the upper zone and top of a near 
vertical exposure of cemented, highly alkaline calcium carbonate 
paleosol (a “‘caliche”’ soil). Nearby Soil Conservation Service soil 
samples of a buried horizon of the same calcium carbonate layer 
contained 79% calcium carbonate and had a pH of 8.4 (Soil Con- 
servation Service, unpubl.). This hard calcium carbonate paleosol 
caps several hundred feet of alkaline, easily eroded, lacustrine 
sediments of the Ringold Formation (Newcomb, 1958). The av- 
erage annual precipitation for the area is 12 cm (Rickard et al., 
1988). 

The population is approximately two to seven m wide and 
extends for 17 km along the upper edge of the bluffs. The holotype 
was collected from the northern end of the population. Although 
there are scattered small exposures of similar caliche substrate in 
coulees to the north of the White Bluffs, these areas were examined 
and the species was not present. The White Bluffs population may 
be the only one of the species. 

The vegetational cover of the bluffs is extremely sparse but 
includes, in addition to Lesquerella tuplashensis, a number of 
plant species that are rare in Washington: Cryptantha spiculifera 
(Piper) Payson, Astragalus geyeri Gray, Cuscuta denticulata En- 
gelm., and Camissonia pygmaea (Dougl.) Raven. 


ACKNOWLEDGMENTS 


The authors acknowledge the Washington Chapter of The Na- 
ture Conservancy and the US Department of Energy for devel- 
oping and supporting the Hanford Biodiversity Project, under 
which the field work for this research took place. We are much 
indebted to and thank Dr. James L. Reveal of the University of 
Maryland for initiating and providing the plate of drawings by 
Dolly Baker. 


1995] Rollins et al.— Lesquerella 207 


LITERATURE CITED 


Gray, A., S. WATSON AND B. L. Rosinson. 1895. Synoptical Flora of North 
ier pt. 1: IL-V, 1-208. 

Newcoms, R. C. 1958. Ringold Formation of Pleistocene Age in type locality: 
the White Bluffs, Washington. Amer. J. Sci. V 256: 328-340. 

Payson, E. B. 1922. A Monograph of the Gents Lesquerella. Ann. Missouri 
Bot. Gard. 8: 103-236. 

Pirer, C. V. 1906. Flora of the State of Washington. Contr. U.S. Natl. Herb. 
11: 1-637 

RELANDER, C. 1956. Drummers and Dreamers. Caxton Printers, Caldwell, ID. 

RicKARD, W. H., L. Rogers, B. E. VAUGHN AND F. G. LiestrAu. 1988. Shrub- 
steppe: Balance and Change in a Semi-arid Terrestrial Ecosystem. Science 
Publishers, New York 

Ro.uins, R. C. 1993. The Cruciferae of Continental North America. I-XVI, 1- 
976. Stanford University Press, Stanford, CA 

AND E. A. SHAW. 1973. The Genus Lesquerella (Cruciferae) in North 
America. I-X, 1-288. Harvard University Press, oe M 

Rose, J. N. 1904. William M. Canby, Bot. Gaz. 37: 

SoIL CONSERVATION SERVICE. Primary characterization an for the Taunton Soil 
Type, 590WA021-5, Franklin County, Washington. Unpublished material. 

WATSON, S. 1885. Contributions to American Botany XIV. Proc. Amer. Acad. 
Arts 20: 324-378 

. 1888. Contributions to American Botany XV. Proc. Amer. Acad. Arts 

23: 249-287. 


RC: 

GRAY HERBARIUM OF HARVARD UNIVERSITY 
22 DIVINITY AVE., 

CAMBRIDGE, MA 02138 

KALB. be FEC, 

CALYPSO CONSULTING 

[is0:- 2150 Si. 

BELLINGHAM, WA 98225 


RHODORA, Vol. 97, No. 891, pp. 208-244, 1995 


THE VASCULAR PLANTS OF FORT DEVENS, 
MASSACHUSETTS 


Davip M. HuNT, KAREN B. SEARCY,'! ROBERT E. ZAREMBA, AND 
C. ROBERTA LOMBARDI 


ABSTRACT 


As part of the process of closing Fort Devens in north central Massachusetts, 
an intensive two year survey of vascular plants growing in the undeveloped area 
(6700 acres) of the fort was completed. A total of 857 naturally occurring taxa in 
394 genera and 121 families were identified, including 18 rare taxa and 29 county 
records. The unusual diversity of taxa for the area is due to several factors which 
include land use history and the physical setting, as well as management practices 
and military activity which created or maintained early and mid-successional 
habitats. 


Key Words: Fort Devens, Massachusetts, vascular flora, rare plants 


INTRODUCTION 


Fort Devens, the only army post in New England, had its main 
installation about 30 miles northwest of Boston in the towns of 
Ayer and Shirley, Middlesex County, and Harvard and Lancaster, 
Worcester County (Figure 1). It was founded in 1917 to train 
soldiers for World War I. Additional land, south of Route 2, was 
acquired between 1941 and 1943 and became the South Post. In 
the mid 1970’s, about 700 acres of wetlands along the east side 
of the Nashua River on the South Post, were transferred to the 
U.S. Fish and Wildlife Service and now form the Oxbow National 
Wildlife Refuge. The post was recommended for realignment and 
closing in 1988 and closed in March, 1996. The approximately 
6700-acre, undeveloped area of Fort Devens was reportedly the 
largest single area of natural land under uniform management in 
north central Massachusetts (U.S. Army Corps Engineers, 1989). 
As part of the procedures associated with closing the fort, a num- 
ber of biological surveys were made. The objective of this study 
was to assemble as complete a collection as possible of the vascular 
plants that occurred naturally on Fort Devens, emphasizing fed- 


' Author for correspondence. 


208 


1995] 


Hunt et al.— Fort Devens Flora 209 


FORT DEVENS 


* 
Fort Devens 


_— 


Co. 


North Post 


_~ 


Worcester Co 
Middlese, 


Shirley 4 
i pias : Main Post 


_-. 


npact £on Harvard 


Lancaster 


Meters 


2006 


South Post 

1006 
Figure |. Map showing the location of Fort Devens in Massachusetts, and an 
enlargement of the Post. The boundaries of Fort Devens are indicated by a solid 
line. The dashed lines indicate the boundary between Worcester and Middlesex 
Counties. This boundary follows the Nashua River on the Main Post. The ap- 
proximately 6700-acre undeveloped area of Fort Devens, which was sampled, is 
indicated by stippling; the developed area, which was not sampled, is left blank. 
The location of the maintained grassland on the South Post is indicated by a black 
oval. The triangles indicate the position of Shepleys and Whittemore Hills. 


210 Rhodora [Vol. 97 


eral and state rare taxa. In addition to helping with decisions on 
future land use, this survey provides a baseline from which to 
monitor future changes. 

The approximately 9240-acre Fort Devens was divided into 
three areas: the South Post (4830 acres), south of Route 2; the 
Main Post (3550 acres); and the North Post (860 acres), the last 
two north of Route 2 (Figure 1). Most of the buildings and other 
facilities were in the latter two areas while the South Post, used 
primarily for training, remained relatively undeveloped, and has 
been retained by the army as the Devens Reserve Forces Training 
Area. Approximately 800 acres in the floodplain and on adjacent 
slopes of the Nashua River on the former Main and North Posts 
were transferred to the U.S. Fish and Wildlife Service and have 
been added to the Oxbow National Wildlife Refuge. With the 
exception of a few natural areas, such as those near Mirror Lake, 
most of the remainder of the North and Main Posts is scheduled 
for development (T. Poole, pers. comm.). 

Much of Fort Devens is within the glacial landform of Lake 
Nashua (Koteff, 1980) and is characterized by broad, flat terraces 
of sandy, well drained deltaic sediments. Areas of stratified clay 
deposits support wetlands (Mott and O’Brian, 1981). Other glacial 
features include esker-like ridges around some ponds on the South 
Post and along the west bank of the Nashua River on the North 
Post. In addition, there are steep sand and gravel ridges around 
Mirror Lake on the Main Post. Other important features include 
the Nashua River floodplain which extends for 8 miles through 
the fort. Elevations range from 61 m in the floodplain of the 
Nashua River to about 140 m on Whittemore Hill on the South 
Post. Exposed bedrock is uncommon, and is found at Shepleys 
Hill on the North Post and in a few scattered outcrops. 

The area occupied by Fort Devens was originally forested, but 
by the time the military post was established, most of the land 
had been cleared and farmed. There was also some small scale 
industrial activity along the river and streams (Fitch and Glover, 
1989). Farming continued on the South Post into the 1940’s. 
About the time the post was closed, forested areas occupied 62 
percent of the South Post, and 36 percent of the Main and North 
Posts (U.S. Fish and Wildlife Service, 1992). The South Post also 
included a 250-acre grassland, kept open by burning and mowing, 
which was used for parachute drops. 


1995] Hunt et al.—Fort Devens Flora 211 
METHODS 


An intensive, systematic survey of the flora on the approxi- 
mately 6700 acres of undeveloped land on the post, including the 
“Impact Zone” (Figure 1) was conducted over two growing sea- 
sons. Most of the collection was done during 1991 by D. Hunt 
and R. Zaremba, focusing principally on the South Post (Hunt, 
1991). Additional collections were made during 1993 by D. Hunt, 
C. R. Lombardi, and K. Searcy focusing on natural areas on the 
Main and North Post (Lombardi, 1994), Finally, a few additional 
species were added as the result of other types of vegetation studies 
conducted in 1994 and 1995. 

Sampling was designed using topographic maps and a grid over- 
lay to insure that some part of every 400 m? was observed. In 
addition, habitats from localities representing the full range of 
elevation, slope, and substrate types (U.S.D.A. Soil Conservation 
Service, 1985, 1989) were examined. All community types based 
on New York State’s Natural Heritage Program Community Clas- 
sification (Reschke, 1990) and all forest cover types (based on 
1980 aerial photographs) were also examined. When necessary, 
aquatic vegetation was sampled from a canoe. 

Collections were made about every two weeks from early May 
to October. All native, naturalized and escaped taxa, including 
those that were probably introduced but had become part of a 
successional plant community, were collected. Every reasonable 
effort was made to collect specimens with reproductive material. 
Information recorded included latitude and longitude using UTM 
coordinates (to 100 m); elevation; substrate type; topography; 
associated taxa; and relative abundance, assessed as a combina- 
tion of the size of the population and number of occurrences; as 
well as the specific collection locality, habitat or natural com- 
munity, and date. Whenever a federally- or state-listed rare taxon 
was found, population boundaries were determined and popu- 
lation size was estimated. Potential county records were checked 
against a list prepared by Sorrie (1991). For the most part, no- 
menclature follows The Flora of North America (Flora of North 
America Editorial Committee, 1993) for pteridophytes and gym- 
nosperms and the Manual of Vascular Plants of North Eastern 
United States and Adjacent Canada (Gleason and Cronquist, 1991) 
for angiosperms. Hortus Third (Bailey and Bailey, 1976) was used 


212 Rhodora [Vol. 97 


for names of naturalized plants not included in Gleason and Cron- 
quist. Where names for rare taxa found at Fort Devens differed 
between Gleason and Cronquist and the Massachusetts rare plant 
list (Division of Fisheries and Wildlife, 1992), the latter were 
retained, and the synonyms indicated in Appendix 1. A complete 
set of voucher specimens has been deposited at the University of 
Massachusetts (MASS). 


RESULTS AND DISCUSSION 


The flora includes 857 taxa in 394 genera and 121 families 
(Appendix 1). The distribution of taxa among groups Is as follows: 
11 (1.3%) fern allies, 22 (2.6%) ferns, 13 (1.5%) gymnosperms, 
247 (28.8%) monocots, and 564 (65.8%) dicots. Approximately 
80 percent of the taxa are native, a figure similar to that reported 
for central and eastern United States and adjacent Canada (Fer- 
nald, 1950). While additional taxa are likely to be found, the 
plants collected to date are undoubtedly a sizable fraction of those 
on the post. Intensive collections in 1993 added 102 new taxa 
(12%). The 1994 and 1995 studies of oxbow ponds and surround- 
ing marshes, among the most species-rich areas on the post, added 
only seven new taxa (<1%). 

The undeveloped portion of Fort Devens supports an unusually 
high diversity of plants for an area its size in north central Mas- 
sachusetts. One factor contributing to species diversity on the 
post is historic land use. Much of the fort was farmland in the 
not-too-distant past. As a result, a great deal of the vegetation is 
currently in various stages of succession, providing habitats rang- 
ing from successional old fields to second-growth forests. The 
relatively recent agricultural influence is seen in the persistence 
of species such as Asparagus officinalis and Triticum aestivum. 
Other weedy species were undoubtedly introduced along the roads 
and railroads. Past habitation is also reflected by artificial lakes 
which provide habitats for aquatic species and by abandoned 
home sites, the source for naturalized and escaped species such 
as Convallaria majalis, Phlox paniculata, Syringa vulgaris and 
Vinca minor. 

A second group of factors affecting species diversity are land 
management practices and military activity on the post. Land 
management practices included mowing, controlled burning, and 
clear and selective cutting. Military activity associated with train- 
ing and the use of heavy equipment, like tanks, created open areas 


1995] Hunt et al.—Fort Devens Flora 213 


of various sizes for colonization by early successional species. In 
addition, one unusual habitat disturbance at Fort Devens is the 
occurrence of annual fires resulting from flares and ammunition 
detonation in the “Impact Zone” on the South Post. The fires 
have produced an area of fire-tolerant upland species (U.S. Fish 
and Wildlife Service, 1992). These types of activities contributed 
to species diversity by creating and maintaining a mosaic of early 
and mid-successional habitats within various communities. Other 
habitats associated with military occupation at the post include 
landfills, rubble dumps, gravel mines, hydrological impound- 
ments and sewage processing beds. At the time of the survey, 
deliberately created or maintained successional habitats sup- 
ported almost one third (27%) of the species, and close to half 
(44.4%) of the state listed rare species. Successional habitats ap- 
pear to be important for many rare species. Indeed, Sorrie (1989) 
has pointed out that the loss of early successional habitats appears 
to account for about 65% of the species extirpated from Massa- 
chusetts. 

Finally, a major factor contributing to species diversity is the 
physical setting. The post has a varied topography ranging from 
flat terraces to steep slopes such as those found at the edge of the 
Nashua River floodplain, on eskers, and on the South Post pla- 
teau. Stream drainages provide relief on a smaller scale. In ad- 
dition, the post has markedly contrasting soil types ranging from 
drought prone Windsor and Hinkley soils, common in the deltaic 
deposits, to poorly drained soils such as Limerick, Swansea and 
Freetown soils of the Nashua River floodplain and scattered wet- 
lands. 

This combination of slopes and soil types supports a wide array 
of plant communities. Fort Devens has widespread plant com- 
munities such as the Appalachian oak-pine forest (AOPF), hem- 
lock-northern hardwood forest (HNHF) and red maple hardwood 
swamp (RMHS). In addition, there are also a number of com- 
munities which are listed as uncommon or restricted in Massa- 
chusetts (Swain, 1993). One of these, the pitch pine-scrub oak 
barrens (PPSOB), found on drought prone soils, is uncommon at 
inland locations. A number of wetland communities on the post 
are also of restricted distribution or are uncommon in the state 
(Swain, 1993). These include the southern New England flood- 
plain forest (FF), dominated by silver maple, occurring as a rel- 
atively narrow, intermittent strip along the Nashua River; several 
dwarf shrub bogs (DSB) and forested peatlands such as the black 


214 Rhodora [Vol. 97 


spruce-tamarck bog (BSTB), supporting plants of a more northern 
distribution; seep communities within the RMHS dominated by 
black ash; sandy bottom kettlehole ponds, with fluctuating water 
levels, providing exposed, inland, non-calcareous, lake shore hab- 
itats (INCLS) with species more characteristic of the coastal plain: 
and a small, medium fen (MF). Collectively these and the other 
wetland communities account for 36% of the species, but only 
about 14% of the total area of the post. 


Rare Species 


Eighteen taxa found growing naturally at Fort Devens are listed 
as either federally or state endangered, threatened (Division of 
Fisheries and Wildlife, 1992), or are on the state watch list (Sorrie, 
1990) (Table 1). Populations of Pinus resinosa, Podophyllum pel- 
tatum and Populus balsamifera, which are on the state watch list, 
were obviously planted or were escapes, and were not included 
in the tally. 

Six species, Bidens discoidea, Eleocharis obtusa var. ovata (E. 
ovata), Panicum philadelphicum, Sparganium minimum, Utric- 
ularia minor, and Wolffia brasiliensis, were associated with oxbow 


Table 1. Rare taxa from Fort Devens, Soecrnaaae E = Endangered, T = 
Threatened, SC = Special Concern, W = Watchlis 


: Approx. No. of 
Taxon Status Individuals 


Arabis drummondii W 20 
Aster radula Ww 200 
Bidens discoidea WwW 200 
Carex typhina T 250 
Cassia hebecarpa E 80 
C yee us houghtonii E 1000 

ochari le var. ovata E 2000 
: iene illaris Ww 100 
Geranium elene Uti W 100 
Leptoloma cognatum W 200 
Liatris borealis SC 5-10 
Lupinus peren: W 3000 
Panicum eis um SC 100 
Smilacina trifolia W 500 

Darganium minimum E 1000+ 

Stellaria borealis W 100 
Utricularia minor W 100 
Wolfhia brasiliensis Ww large 


1995] Hunt et al.—Fort Devens Flora 215 


ponds in the Nashua River floodplain. Bidens discoidea was found 
growing as an epiphyte on the tangled lower stems of buttonbush, 
Cephalanthus occidentalis. Eleocharis obtusa var. ovata, a taxon 
whose status appears unclear, was found in sedge meadows which 
had been scoured two years before (T. Poole, pers. comm.). The 
annual, Panicum philadelphicum, was observed in mud flats left 
exposed by drying oxbow lakes during the unusually dry summer 
of 1995. Sparganium minimum had a large population in the 
shallow water of an oxbow lake. Although it is a species more 
common in northern New England (Crow and Hellquist, 1981), 
two populations have been reported from adjacent Middlesex 
County (Crow and Hellquist, 1981). It was previously reported 
from Lake Quinsigamond, Worcester Co. (Jackson, 1909), but no 
herbarium specimens were found (Sorrie, 1987). Utricularia mi- 
nor and Wolffia brasiliensis both have large populations in one 
or two oxbow lakes and also occur in widely scattered localities 
in Massachusetts (Hellquist and Crow, 1982; Crow and Hellquist, 
1985). It is suggested that at least the former species may be 
overlooked in the state (Sorrie, 1990). 

In addition to oxbow ponds, several other wetland areas on the 
post supported rare species. Carex typhina was found in a mixed 
oak forest along the sandy banks of a small stream. This appar- 
ently relatively dry habitat contrasts with the silver maple-green 
ash floodplain forest reported for other populations of the species 
in the state (Sorrie, 1987). Aster radula occurred in an open, 
recently-burned peaty meadow. Smilacina trifolia, another spe- 
cies with northern affinities, was abundant in a black spruce- 
tamarack swamp, and a small population of Stellaria borealis was 
found under powerlines along a small stream in a rich, fen-like 
area. 

Another group of rare species was found in the dry, sandy glacial 
outwash plains associated with the pitch pine-scrub oak barrens. 
These included Cyperus houghtonii, Eragrostis capillaris, Lep- 
toloma cognatum, Liatris borealis (Liatris scariosa var. novae- 
angliae), and Lupinus perennis. Many of these are successional 
species dependent on disturbance or fire. Cyperus houghtonii, with 
no other current stations in the state, was found on a disturbed, 
gravelly esker slope in one of the rifle ranges. The population size 
of this species showed large fluctuations between 1991 and 1993. 
The relatively few individuals of Liatris borealis, found in mowed 
roadsides and along a railroad track, were part of a much larger 
population under powerlines on adjacent property. Liatris bo- 


216 Rhodora [Vol. 97 


Table 2. County Records from Fort Devens, Massachusetts. 


County Collectors and Number 


MIDDLESEX COUNTY 


Native 
Callitriche stagnalis Hunt and Zaremba 1597 
Scirpus xpeckii Hunt and Zaremba 2543 
Introduced 
Malus floribunda Hunt and Zaremba 2008 
WORCESTER COUNTY 
Native 
Bidens aristosa Hunt and Zaremba 2386 
Carex bebbii Hunt and Zaremba 415 
Carex hystericina Hunt and Zaremba 766 
Carex laevivaginata Hunt and Zaremba 324 
Carex typhina Hunt and Zaremba 1480 
Eleocharis obtusa var. ovata! Hunt and Zaremba 1068 
Elodea canadensis' Hunt and Zaremba 1306 
Epilobium hea Hunt and Zaremba 1888 
Galium aparin Hunt and Zaremba 2575 
Galium ian : Hunt and Zaremba 2574 
Geranium bicknellii' Hunt and Zaremba 2467 
Potamogeton zosteriformis Enz, Lindwall, Hickler, Searcy 179 
Scirpus georgianus Hunt MA-687 
Sparganium minimum! Enz, Lindwall, Hickler, Searcy 129 
Introduced 
Catalpa bignonioides Hunt and Zaremba 1304 
Cerastium semidecandrum Hunt and Zaremba 248 
rau conse majalis Hunt and Zaremba 242 
Draba verna Lombardi and Searcy 164 
heel umbellata Hunt and Zaremba 77 
Epipactis helleborine Hunt and Zaremba 2489 
Euonymus alatus Hunt and Zaremba 2496 
Larix decidua Hunt and Zaremba 1259 
Malus baccata Hunt and Zaremba 152 
ead opuyouus var. 
opulifoliu Hunt and Zaremba 1184 
Trifolium Save Hunt MA-554 
Urtica dioica var. dioica Hunt and Zaremba 943 


' Reported from Worcester County (Jackson, 1909) but no herbarium speci- 
mens have been found (Sorrie, 1991). 


realis is declining in Massachusetts and is now found primarily 
in the southeastern part of the state. Lupinus perennis is another 
species declining in Massachusetts. Like Liatris borealis, popu- 
lations of Lupinus perennis were found in disturbed areas along 


1995] Hunt et al.—Fort Devens Flora 217 


roads. However, it also occurred in the “Impact Zone” where 
fires were frequent. 

Finally, several rare species occurred in other obviously dis- 
turbed habitats. These included Arabis drummondii which was 
found in a mowed strip beside a road, Cassia hebecarpa (Senna 
hebecarpa) found in alluvial thickets along a powerline, and Ge- 
ranium bicknellii found in a wet area along powerlines. The latter 
species is apparently adventive in eastern Massachusetts and it 
is uncertain whether the Fort Devens populations are native. 
Populations of G. bicknellii declined drastically between 1991 and 
1993 probably due to the cessation of mowing of the powerlines 
where they occurred. 


County Records 


Based on a checklist of county records (Sorrie, 1991), twenty 
nine species collected at Fort Devens were herbarium county 
records for either Middlesex or Worcester County (Table 2). A 
number of those listed were previously reported for Worcester 
County (Jackson, 1909) but appear to be undocumented (Sorrie, 
1991). County records include two state endangered, Eleocharis 
obtusa var. ovata and Sparganium minimum; one state threat- 
ened, Carex typhina; and one watch list species, Geranium bick- 
nellii. Almost half are introduced species of disturbed or succes- 
sional habitats. Many are relatively common elsewhere and are 
known from adjacent counties in the eastern part of the state. In 
contrast, Scirpus xpeckii, a sterile hybrid between S. atrovirens 
and_S. cyperinus (Tucker, 1992) found on a steep slope of a narrow 
ravine in mixed hardwood forest, has previously been reported 
from only the western part of the state (Sorrie, 1991; Tucker, 
1992). 


ACKNOWLEDGMENTS 


We thank T. Poole, Natural Resources Officer, Fort Devens 
and P. Somers, State Botanist, for help, encouragement, and sup- 
port. B. Sorrie supplied information on species suspected or known 
to occur at the site. Assistance with various aspects of the project 
was also provided by T. Enz, M. Hickler, L. Hunt, B. Lindwall, 
S. Patton, and M. Taras. B. Hellquist, C. Sheviak, N. Miller, and 
G. Tucker helped with identification. We also thank NYS for 


218 Rhodora [Vol. 97 


making its facilities available to us. This study was funded in part 
by the U.S. Army and the Massachusetts Natural Heritage and 
Endangered Species Program. 


LITERATURE CITED 


BaiLey, L. H. anv E. Z. Battey. 1976. Hortus Third, A Concise Dictionary of 
Plants Cultivated in the United States and Canada. Revised and Expanded 
by the Staff of the Liberty Hyde Bailey Hortium. MacMillan Publishing Co., 
New York, NY 

Crow, G. E. anp C. B. HELLQuisT. 1981. Aquatic vascular plants of New En- 
gland: Part 2. Typhaceae and Sparganiaceae. New Hampshire Agric. Exp. 
Sta. ai 517. 


Aquatic vascular plants of New England: Part 8. 

Lentibulariacese. New Hampshire Agric. Exp. Sta. Bull. 528. 

DIVISION OF FISHERIES AND WILDLIFE. 1992. Massachusetts List of Endangered 
Threatened and Special Concern Species. Commonwealth of Massachusetts, 
Division of Fisheries and Wildlife. Boston, 

FERNALD, M. L. 1950. Gray’s Manual of Bong Eighth Revised Edition. Van 
— saris Co., New Yor 

Fircu, V. DS. GLover. 1989. Histone and prehistoric reconnaissance 
ss a Devens (Main Post, North Post, South Post) Massachusetts. 
Appendix C. Jn: U.S. Army Corps of Engineers, Fort Devens ce ae 
pelea information for Fort Devens realignment. Appendix 2 


FLORA OF NoRTH AMERICA EDITORIAL COMMITTEE, Eps. 1993. The i of 
North Ameri North of Mexico, Volume 2. Oxford University Press, New 
York, N 

GLEASON, H. —- pA. Cronquist. 1991. Manual of Vascular Plants of North- 
eastern United States and Adjacent Canada, 2nd ed. New York Botanical 
Garden, Bronx, NY. 

HELLQuisT, C. B. AND G. E. Crow. 1982. Aquatic vascular plants of New En- 


gland: Part 5. Araceae, Lemnaceae, Xyridaceae, Eriocaulaceae, and Ponted- 
erlaceae. New Hampshire Agric. Exp. Sta. Bull. 523. 
Hunt, D. M. 1991. Floristic survey, with emphasis on rare species, of Fort 
Devens, Massachusetts. U.S. Army Corps of Engineers Construction Engi- 
neering Research Laboratory, Environmental Division. Unpublished report. 
Jackson, J. 1909. A Catalogue of the Flowering Plants and Ferns of Worcester 
County, Massachusetts, 3rd ed. Worcester Natural History Society. Stanhope 
Press. Boston, M 
Korterr, C. 1980. Deviacial history of glacial Lake Nashua, east-central Mas- 
sachusetts. /n: Larson, G. L. and B. D. Stone, eds. Late Wisconsinan Gla- 
ciation of New England. Kendall/Hunt Publishing Co. Dubuque, IA 
Lomsarpl, C. R. 1994. Floristic survey, with emphasis on rare ee of Fort 
Devens, Massachusetts, 1993. (Massachusetts Natural Heritage and Endan- 
gered Species Program, unpublished report 
TT, W. W. AND A. L. O'Brian. 1981. Geology and hydrology of wetlands in 
Massachusetts. Water Resources Research Center. University of Massachu- 
setts, Amherst. Publication 123. 


1995] Hunt et al.— Fort Devens Flora 219 


RESCHKE, C. 1990. Ecological Communities of New York State. New York 
Natural Heritage Program, New York State Department of Environmental 
Conservation. Latham, NY 

SorriE, B. A. 1987. Notes on the rare flora of Massachusetts. Rhodora 89: 113- 
196. 


1989. Massachusetts flora: A review of current distribution and con- 
servation of rare species. Sanne on ae 120. 

. 1990. ‘Watch List” Unc r Rare Massachusetts Plants. Com- 
monwealth of eae eines of Fisheries and Wildlife. Boston, 
MA 


; 1991. County Checklist of Massachusetts Vascular Plants. Unpublished 

d Natural Heritage and Endangered 
Species Program, Division of Fisheries and Wildlife. Boston, MA. 

Swaln, P. C. 1993. Priority natural community-types for protection in Massa- 
chusetts. Division of Fisheries and Wildlife. Boston, MA. 

Tucker, G. C. 1992. Scirpus (Cyperaceae) in Connecticut. Newslett. of the 
Connecticut Bot. Soc. 20: 3- 

U.S. Army Corps OF ENGINEERS, New ENGLAND Division. 1989. Appendix 2. 
Fort Devens supplemental environmental information for Fort Devens re- 
alignment 

U.S. D. A. Som. CONSERVATION SERVICE. 1985. Soil Survey of Worcester County, 
Massachusetts, Northwestern Part 

—. 1989. Middlesex County Soil Survey, A Resource Planner’s Guide, 
Middlesex Tatas District, Acton, M 

U.S. Fish AND WILDLIFE SERVICE. 1992. Survey and evaluation of wetlands and 
wildlife fata Fort Devens, Massachusetts. Unpublished report. 


D. M. H. 

NEW YORK NATURAL HERITAGE PROGRAM 

NEW YORK STATE DEPARTMENT OF ENVIRONMENTAL 
CONSERVATION 

700 TROY-SCHENECTADY RD. 

LATHAM, NY 12110-2400 


Kk. B.S. &2C. RL; 

BIOLOGY DEPARTMENT 
UNIVERSITY OF MASSACHUSETTS 
AMHERST, MA 01003 


R. E. Z. 

THE NATURE CONSERVANCY 
91 BROADWAY 

ALBANY, NY 12204 


220 Rhodora [Vol. 97 


APPENDIX 1: ANNOTATED LIST OF THE 
VASCULAR PLANTS OF FORT DEVENS, 


MASSACHUSETTS 
The appendix lists the taxa, abundance, and type in which the taxon 
was collected. Some taxa also occurred in Shanmauniiics other than the one in 


which the ao was collected; and, for a few, the community in which a 
specimen was collected was not typical for the taxon on the post. Where a taxon 
was observed at only one site, the general location (North, Main or South Post) 
is also given 

Abundance estimates are for the entire post and were based on the population 
size and number of communities in which a taxon was observed. Taxa were classed 
as rare, small population at one site; infrequent, large numbers at one site, or low 
numbers in one or two communities; occasional, large numbers in one or two 
communities or low numbers in several communities; common, several com- 
munities in large numbers or low numbers in most communities; abundant, in 
most communities in large numbers. 

Abbreviations for the community types in Reschke (1990) are as follows: AOPF 
= Appalachian Oak-Pine Forest, BCL = Brushy Cleared Land, BMMF = Beech 
Maple Mesic Forest, BSTB = Black Spruce Tamarack Bog, COF = Chestnut Oak 
Forest, DEM = Deep Emergent Marsh, DSB = Dwarf aes = EP = Eutrophic 
Pond, FF = saa Forest, GM = Gravel Mine, HHS = Hemlock-Hardwood 
Swamp, HNHF = Hemlock-Northern Hardwood ae ce = Inland Non- 
nar Lake Shore, LD = Landfill/Dump, MCS = Main Channel Stream, MF 
= Medium Fen, ML = Mowed Land, MS = Midreach Stream, OL = Oxbow Lake, 
PNHF = Pine-Northern Hardwood Forest, PPOF = Pitch Pine-Oak oe PPSOB 
= Pitch Pine-Scrub Oak Barrens, RMHS = Red Maple-Hardwood Swamp, RSGB 

= Riverside Sand/Gravel Bar, SA.M = Sand Mine, SEM = it Emergent 
Marsh, SH = Successional Hardwoods, SM = Sedge Meadow, SNSG = Succes- 
sional Northern Sandplain Grassland, SOF = Successional Old Field, SS = Shrub 
Swamp, SU.S = Successional Shrubland. 


LYCOPODIOPHYTA (Clubmosses) 


LYCOPODIACEAE 
Diphasiastrum digitatum (Dillenius ex A. Braun) Holub occasional; PNHF 
Diphasiastrum tristachyum (Pursh) Holub oe PPSOB 
Huperzia lucidula (Michx.) Trev. infrequent 
a ees inundata (L.) Holub rare; SEM — ak Post 
Lycopodium clavatum L. infrequent; PPSOB 
ean obscurum L. common; PNHF 


SELAGINELLACEAE 
Selaginella apoda (L.) Spring rare; SH— Main Post 


EQUISETOPHYTA (Horsetails) 


EQUISETACEAE 
Equisetum arvense L. occasional; RMHS 


1995] Hunt et al.—Fort Devens Flora 221 


Equisetum fluviatile L. infrequent; SEM 
Equisetum hyemale L. subsp. affine (Engelmann) Calder & R. L. Taylor infre- 


quent; BCL 
Equisetum sylvaticum L. infrequent; RMHS 


POLYPODIOPHYTA (Ferns) 
CEAE 
Asplenium platyneuron (L.) BSP. rare, BMMF—Main Post 


BLECHNACEAE 


Woodwardia virginica (L.) J.E. Smith infrequent; DSB 


AEDTIACEAE 
Dennstaedtia punctilobula (Michx.) T. Moore occasional; SH 
Pteridium aquilinum (L.) Kuhn var. latiusculum (Desv.) Underw. ex A. Heller 
abundant; SNSG 
DRYOPTERI 


ACEAE 
Athyrivom fil -femina (L.) Roth ex Mertens var. angustum (Willd.) G. Lawson 
occasional; FF 


Deparia acrostichoides (Swartz) M. Kato rare; ie iaaaaies Post 
Dryopteris cristata (L.) A. Gray occasional; P 


Dryopteris intermedia (Muhlenberg ex Willd.) - a common; PNHF 
Dryopteris marginalis (L.) A. Gray infrequent; PNHF 
Onoclea sensibilis L. common; RM 


Polystichum acrostichoides (Michx.) Schott infrequent; AOPF 


OPHIOGLOSSACEAE 
Botrychium dissectum Sprengel rare; PNHF—South Post 
Botrychium virginianum (L.) Swartz rare; BBMF—Main Post 
OSMUNDACEAE 


Osmunda cinnamomea L. common; RMHS 
Osmunda claytoniana L. infre 


Osmunda regalis L. var. spectabilis (Willd.) A. Gray occasional; SEM 
POLYPODIACEAE 

Polypodium virginianum L. infrequent, AOPF 
PTERIDACEAE 


Adiantum pedatum L. rare, HNHF—Main Post 


YPTERIDACEAE 
Phegopteris connectilis (Michx.) Watt rare; BMMF—Main Post 
Thelypteris noveboracensis (L.) Nieuwl. occasional; PNHF 


22? Rhodora [Vol. 97 


Thelypteris palustris Schott occasional; MF 
Thelypteris simulata (Davenp.) Nieuwl. infrequent; RMHS 


PINOPHYTA (Gymnosperms) 


CUPRESSA 
Juniperus communis L. infrequent; SU.S 
Juniperus virginiana L. rare; SU.S—South Post 


PINACEAE 

Larix decidua Miller rare; SH—South Post 

Larix laricina (DuRoi) K. Koch ee BSTB 
Picea abies (L.) Karst rare; SH—South Post 

Picea glauca (Moench) Voss rare, RMHS—South Post 
Picea mariana (Miller) BSP. infrequent; BSTB 
Pinus resinosa Aiton infrequent; SU.S 

Pinus rigida Miller abundant; PPOF 

Pinus strobus L. abundant; SOF 

Pinus sylvestris L. rare; SH— Main Post 

Tsuga canadensis (L.) Carriére occasional; HHS 


CEAE 
Taxus baccata L. rare; SU.S—South Post 


MAGNOLIOPHYTA (Flowering Plants) 
MAGNOLIOPSIDA (Dicots) 
ACERACEAE 
Acer negundo L. var. negundo rare, SH— Main Post 
Acer pensylvanicum L. rare; AOPF—South Post 


Acer saccharinum L. occasional; FF 
Acer saccharum Marshall var. saccharum common; SH 


AMARANTHACEAE 
Amaranthus hybridus L. rare; LD— Main Post 
Amaranthus palmeri 8. Wats. rare; BCL—North Post 
Froelichia gracilis (Hook.) Moq. rare; ML—North Post 


ANACARDIACEAE 
Cotinus coggygria Scop. rare; BCL—Main Post 
Rhus copallinum L. infrequent; F 
Rhus glabra L. infrequent; BCL 
Rhus typhina L. occasional; BCL 
Toxicodendron radicans (L.) Kuntze occasional; BCL 
Toxicodendron vernix (L.) Kuntze infrequent; RMHS 


1995] Hunt et al.—Fort Devens Flora 223 


APIACEAE (UMBELLIFERAE) 
Aethusa cynapium L. rare; BCL—South Post 
Cicuta bulbifera L. occasional; SM 
Cicuta maculata L. infrequent; SEM 
Daucus carota L. occasional; BCL 
Hydrocotyle americana L. occasional; HHS 


Zizia aurea (L.) W. Koch rare; SH— Main Post 


ACEAE 
Apocynum androsaemifolium L. rare, BCL—South Post 
Vinca minor L. rare; SH—South Post 


AQUIFOLIAC 
Tlex ee (Pursh) A. Gray infrequent; SS 
Tlex verticillata (L.) A. Gray common; RMHS 
Nemopanthus mucronatus (L.) Trel. Soci sional: SS 


IACEAE 

Aralia hispida Vent. earns pert 
Aralia nudicaulis L. common; 

Panax trifolium L. occasional; cies 


ASCLEPIADACEAE 
Asclepias amplexicaulis J.E. Smith Sar ne ag PPSOB 
Asclepias exaltata L. rare; BCL— 
Asclepias incarnata L. var. coi jaieed ) Pers. infrequent; SM 
Asclepias syriaca L. occasional; SOF 


ASTERACEAE (COMPOSITAE) 
Achillea millefolium L. infrequent; BCL 
Ambrosia artemisiifolia L. infrequent; SOF 
Anaphalis margaritacea (L.) Benth. & Hook. rare, AOPF—North Post 
Antennaria neglecta Greene var. canadensis (Greene) Cronq. infrequent, SOF 
Antennaria neglecta Greene var. neodioica (Greene) Cronq. infrequent, SNSG 
Anthemis arvensis L. infrequent; SA.M 
Arctium minus Schk. infrequent; BCL 
Artemisia vulgaris L. occasional; BCL 
Aster acuminatus Michx. occasional; PNHF 
Aster cordifolius L. occasional; SH 
Aster divaricatus L. common, AOPF 
Aster dumosus L. rare; SH—South Post 
Aster ericoides L. rare; ML— Main Post 
Aster lanceolatus Willd. var. simplex (Willd.) A.G. Jones infrequent; SH 
Aster lateriflorus (L.) Britton infrequent, BMMF 
Aster linariifolius L. infrequent; AOPF 


224 Rhodora [Vol. 97 


Aster macrophyllus L. znireguents AOPF 


; MF 
Aster patens Aiton var. vain oe AOPF 
Aster paternus Cronq. infrequent; PPOF 

Aster puniceus L. infrequent; ih 

Aster radula Aiton rare; SEM—South Post 

Aster umbellatus Miller infrequent, MF 

Aster undulatus L. infrequent; SU.S 

Bidens aristosa (Michx.) Britton rare; RMHS—South Post 


Bi ; 
Bidens discoidea (T. & G.) Britton infrequent; SS 
Bidens frondosa L. infrequent; BCL 
Centaurea maculosa Lam. occasional; PPSOB 
Chrysanthemum leucanthemum L., ‘afequent BCL 
Cichorium intybus L. rare; SU.S—South Post 
Cirsium arvense (L.) Scop. var. horridum Wimmer & Graebner rare; MF— 
South Post 
Cirsium muticum Michx. rare; HHS—South Post 
Cirsium pumilum (Nutt.) Sprengel rare; SOF— North Post 
Cirsium vulgare (Savi) Tenore infrequent; BCL 
Conyza canadensis (L.) Cronq. var. canadensis occasional; BCL 
Coreopsis lanceolata L. rare: ae orth Post 
Crepis tectorum L. rare; SA.MM—M 
Erechtites hieracifolia (L.) Raf. var. ele infrequent; SEM 
Erigeron annuus (L.) Pers. infrequent; SH 
Erigeron pulchellus Michx. var. pulchellus infrequent; SH 
Erigeron strigosus Muhl. var. strigosus occasional; BCL 
Eupatorium fistulosum Barratt occasional; SH 
Eupatorium perfoliatum L. var. pees ee iar ei SM 
F—South Pos 


nt; SH 
Euthamia graminifolia (L.) Nutt. var. nuttallii (Greene) W. Stone occasional; 
BCL 


Euthamia tenuifolia (Pursh) as rare; INCLS— South Post 
Galinsoga parviflora Cav. rare; BCL— ost 
Gnaphalium obtusifolium L. var. obtusifolium infrequent; BCL 


ieracium floribundum Wimmer & Grab. infrequent; BCL 
Hieracium paniculatum L. infrequent; AOP 
Hieracium pilosella L. infrequent; BCL 
Hieracium scabrum Michx. infrequent; BCL 


1995] Hunt et al.—Fort Devens Flora 225 


Hieracium venosum L. infrequent; AOPF 

Krigia virginica (L.) Willd. infrequent; SNSG 

Lactuca biennis (Moench) Fern. rare; HNHF— Main Post 

Lactuca canadensis L. infrequent; BCL 

Leontodon autumnalis L. var. autumnalis infrequent; BCL 

neh borealis Nutt. (Liatris scariosa (L.) Willd. var. novae-angliae Lunell) 
rare; PPSOB—North Pos 

Matricaria matricarioides (Less.) Porter rare; BCL—South Post 

Prenanthes altissima L. var. altissima rare, RMHS— aa Post 


Rudbeckia laciniata L. var. laciniata rare, SH—Main Post 
Senecio aureus L. infrequent; M 

Solidago bicolor L. occasional; AOPF 

Solidago caesia L. occasional; AOPF 

Solidago canadensis L. var. canadensis occasional; SEM 
Solidago gigantea Aiton infrequent; BCL 

Solidago juncea Aiton common, SEM 

Solidago nemoralis Aiton var. neers common; SNSG 
Solidago odora Aiton var. odora occasional; SU.S 

Solidago puberula Nutt. var. puberula occasional; sae 
Solidago rugosa Miller subsp. rugosa var. rugosa common; BCL 
Solidago rugosa Miller subsp. rugosa var. villosa (Pursh) F Tod rare; AOPF— 


Solidago uliginosa Nutt. rare; RMHS—South Post 
Tanacetum vulgare L. rare, BCL—South Post 
Taraxacum officinale Weber infrequent; BCL 
Tragopogon dubius Scop. rare; BMMF—North Post 
Tussilago farfara L. rare, BCL— Main Post 


ALSAMINACEAE 
Impatiens capensis Meerb. occasional; SS 


BERBERIDACEAE 
Berberis thunbergii DC. infrequent; FF 
Berberis vulgaris L. infrequent; SH 
Podophyllum peltatum L. rare; PNHF—South Post 


BETULACEAE 
Alnus incana (L.) Moench. common; RMHS 
Alnus serrulata (Aiton) Willd. infrequent; SS 
Betula alleghaniensis Britton occasional; SS 
Betula lenta L. occasional; HNHF 
Betula papyrifera Marshall var. cordifolia (Regel) Fern. infrequent, HNHF 
Betula papyrifera Marshall var. papyrifera occasional; FF 
Betula populifolia Marshall common; BCL 
Carpinus caroliniana Walter er RMHS 


226 Rhodora [Vol. 97 


Corylus americana Walter common; BCL 
Corylus cornuta Marshall infrequent, PNHF 
Ostrya virginiana (Miller) K. Koch infrequent; AOPF 


BIGNONIACEAE 
Catalpa bignonioides Walter infrequent; FF 


BORAGINACEA 
Echium ae L. infrequent; BCL 
Myosotis laxa Lehm. occasional; SEM 
Myosotis scorpioides L. infrequent; SEM 


BRASSICACEAE (CRUCIFERAE) 
Alliaria petiolata (Bieb.) Cavara & pipes oo FF 
Arabis drummondii A. Gray rare; BCL—North Pos 
Arabis glabra (L.) Bernh. infrequent; a 
Barbarea vulgaris R. Br. occasional; SM 
Berteroa incana ee DC. rare; GM—Main Post 
Brassica rapa L. rare, BCL—North Post 
Capsella bursa- ee (L.) Medikus infrequent; BCL 
Saale ensylvanica Muhl. occasional; RMHS 
Draba verna L. rare; ML— Main Post 
Erysimaum cheiranthoides L. rare, PPOF—North Post 

N Pos 


Lepidium densiflorum Schrader infrequent; BCL 

Lepidium virginicum L. var. virginicum Sisters BCL 

Raphanus raphanistrum L. rare; PPOF— Po 

Rorippa palustris (L.) Besser var. hispida oe v.) Rydb. rare; SEM— Main Post 
Rorippa palustris (L.) Besser var. palustris infrequent; FF 

Sisymbrium altissimum L. rare; BCL—North Post 

Thlaspi arvense L. rare; BCL— Main Post 


CABOMBACEAE 
Brasenia schreberi J. F. Gmelin infrequent; EP 
Cabomba caroliniana A. Gray*infrequent; EP 


AESALPINIACEAE 
Cassia hebecarpa Fern. (Senna hebecarpa (Fern.) Irwin & Barneby) rare; BCL— 
Main Post 


Chamaecrista fasciculata (Michx.) Greene infrequent; SOF 
hamaecrista nictitans (L.) Moench rare; ML— Main Post 
Gleditsia triacanthos L. rare; LD—Main Post 


ALLITRICHACEAE 
Callitriche heterophylla Pursh infrequent; SEM 
Callitriche stagnalis Scop. infrequent; MCS 


1995] Hunt et al.— Fort Devens Flora 227 


CAMPANULACEAE 
Campanula aparinoides Pursh infrequent; SEM 
Lobelia cardinalis L. var. cardinalis infrequent, RMHS 
Lobelia inflata L. occasional; SOF 
Lobelia spicata Lam. var. spicata infrequent; SNSG 


ANNABACEAE 
Humulus japonicus Siebold & Zucc. rare; BCL—North Post 


CAPRIFOLIA 
ie ee Miller occasional; BCL 
Lonicera canadensis Marshall infrequent, MF 
Lonicera morrowii A. Gray occasional; BCL 
Lonicera sempervirens L. var. sempervirens infrequent; SH 
Lonicera tatarica L. infrequent; BCL 
Sambucus canadensis L. var. canadensis occasional; SS 
Viburnum acerifolium L. occasional; SH 
Viburnum alnifolium Marshall rare; PNHF—Main Post 
Viburnum dentatum L. var. lucidum Aiton common; RMHS 
Viburnum nudum L. var. sapuaenes (L.) T. & G. occasional; PNHF 
Viburnum lentago L. infreque H 
Viburnum opulus L. var. americanum Aiton rare; RMHS—North Post 


‘ARYOPHYLLACEAE 
Arenaria lateriflora L. occasional; SH 
Cerastium semidecandrum L. infrequent; BCL 
Cerastium vulgatum L. infrequent; BCL 
Dianthus armeria L. infrequent; SH 
Gypsophila muralis L. infrequent; SNSG 
Saponaria officinalis L. infrequent; SOF 
Scleranthus annuus L. infrequent; BCL 
Silene antirrhina L. infrequent; SNSG 
Silene latifolia Poiret infrequent; BCL 
Silene vulgaris (Moench) Garcke occasional; SNSG 

‘pergularia rubra (L.) J. & C. Pres] infrequent; BCL 

Stellaria borealis Bigelow rare; MF—South Post 
Stellaria graminea L. infrequent; BCL 
Stellaria media (L.) Villars rare; ¢: ML— North Post 


CELASTRACEAE 
Celastrus orbiculatus Thunb. infrequent; SH 
Euonymus alatus (Thunb.) Siebold rare, BMMF— Main Post 


ERATOPHYLLACEAE 
Ceratophyllum demersum L. infrequent; EP 
Ceratophyllum echinatum A. Gray infrequent; EP 


228 Rhodora 


CHENOPODIACEAE 
Chenopodium album L. infrequent; SNSG 
Chenopodium pumilio R. Br. rare; BCL— Main Post 


CISTACEAE 
Helianthemum bicknellii Fern. infrequent; SNSG 
Helianthemum canadense (L.) Michx. occasional; PPOF 
Lechea intermedia Leggett occasional; AOPF 
Lechea maritima Leggett occasional; BCL 
Lechea mucronata Raf. infrequent, AOPF 


CLETHRACEAE 
Clethra alnifolia L. occasional; SM 


CLUSIACEAE (GUTTIFERAE) 
Hypericum boreale (Britton) E. Bickn. infrequent; SM 
LS 


; SH 
Triadenum fraseri (Spach) Gleason rare; SEM—South Post 
Triadenum virginicum (L.) Raf. occasional; SM 


CONVOLVULACEAE 
Calystegia sepium (L.) R. Br. occasional; BCL 


CORNACEAE (includes NYSSACEAE) 
Cornus alternifolia L.f. infrequent; PNHF 
Cornus amomum Miller var. amomum occasional, FF 
Cornus canadensis L. infrequent; RMHS 
Cornus florida L. occasional; poe 
Cornus racemosa Lam. common; BCL 
Cornus rugosa Lam. infrequent, BMMF 
Nyssa sylvatica Marshall var. sylvatica infrequent; AOPF 


CRASSULACEAE 
Sedum purpureum (L.) J.A. Schultes rare; BCL—North Post 


CUCURBITACEAE 
Echinocystis lobata (Michx.) T. & G. rare; FF—South Post 


[Vol. 97 


1995] Hunt et al.— Fort Devens Flora 


CUSCUTACEAE 
Cuscuta gronovii Willd. infrequent, BMMF 


DROSERACEAE 
Drosera intermedia Hayne infrequent; DSB 
Drosera rotundifolia L. rare, INCLS—South Post 


ELAEAGNACEAE 
Elaeagnus umbellata Thunb. infrequent; SU.S 


ERICACEAE 
Andromeda glaucophylla Link rare, DSB—South Post 
Arctostaphylos uva-ursi (L.) Sprengel infrequent; PPOF 
Chamaedaphne calyculata (L.) Moench occasional; DSB 
Epigaea repens L. occasional; BCL 
Eubotrys racemosa (L.) Nutt. infrequent; INCLS 
Gaultheria procumbens L. common; AOPF 
Gaylussacia baccata (Wangenh.) K. Koch common; DSB 
Gaylussacia frondosa (L.) T. & G. occasional; AOPF 


Kalmia latifolia L. occasional, AOPF 

Kalmia polifolia Wangenh. rare, DSB—South Post 
Ledum groenlandicum Oeder infrequent; BSTB 
Lyonia ligustrina (L.) DC. infrequent; MF 
Rhododendron canadense (L.) Torr. infrequent; BSTB 
Rhododendron prinophyllum (Small) pe rare, HNHF—Main Post 
Rhododendron viscosum (L.) Torr. infrequen 

Vaccinium angustifolium Aiton abundant; ie 
Vaccinium corymbosum L. common; INCLS 
Vaccinium macrocarpon Aiton infrequent; INCLS 
Vaccinium oxycoccos L. rare, DSB—South Post 
Vaccinium pallidum Aiton common; COF 


EUPHORBIACEAE 
Acalypha gracilens A. Gray var. ae rare; BCL— Main Post 
Acalypha rhomboidea Raf. infrequent L 


Euphorbia maculata L. occasional; SU.S 
Euphorbia nutans Lagasca rare; SU.S— Main Post 


FABACEAE 
Amphicarpaea bracteata (L.) Fern. occasional; RMHS 
Apios americana Medikus infrequent; BCL 
Baptisia tinctoria (L.) R. Br. infrequent; ee 
Coronilla varia L. rare, BCL— Main Pos 


229 


230 Rhodora [Vol. 97 


Desmodium canadense (L.) DC. infrequent; BCL 
Desmodium glutinosum (Muhl.) A. Wood rare; BMMF—North Post 
Desmodium marilandicum (L.) DC. infrequent; AOPF 
Desmodium nudiflorum (L.) DC. rare; AOPF— Main Post 
Desmodium rotundifolium DC. rare; AOPF—Main Post 
Lespedeza capitata Michx. occasional; BCL 
Lespedeza hirta (L.) Hornem. subsp. hirta infrequent; AOPF 
Lespedeza intermedia (S. Wats.) Britton occasional; AOPF 

Lespedeza virginica (L.) Britton rare; AOPF— Main Post 

F 


Medicago sativa L. infrequent; SOF 

Melilotus alba Medikus infrequent; BCL 
Melilotus officinalis (L.) Pallas. zene BCL 
Robinia hispida L. rare, SOF— Pos 
Robinia pseudoacacia L. negent SH 
Trifolium arvense L. occasional; BCL 

Trifolium aureum Pollich Ce RR BCL 


Vicia tetrasperma (L.) Moench rare; BCL—South Post 
Wisteria sinensis (Sims) Sweet rare; AOPF— Main Post 


FAGACEAE 
Castanea dentata (Marshall) Borkh. infrequent; SU.S 
Fagus grandifolia Ehrh. occasional; BMMF 
Quercus alba L. common; SU.S 
Quercus bicolor Willd. occasional; SEM 


Quercus prinoides Willd. occasional; PPSOB 
Quercus prinus L. infrequent; COF 

Quercus rubra L. common; SU.S 

Quercus velutina Lam. abundant; AOPF 


FUMARIACEAE 
Corydalis sempervirens (L.) Pers. rare; AOPF—Main Post 


GENTIANACEAE 
Bartonia virginica (L.) BSP. rare; SEM—South Post 
Gentiana clausa Raf. infrequent; RMHS 


1995] Hunt et al.—Fort Devens Flora 


GERANIACEAE 
Erodium cicutarium (L.) L’Her. rare; BCL— North Post 
Geranium bicknellii Britton infrequent; 
Geranium maculatum L. occasional; SH 


GROSSULARIACEAE 
Ribes americanum Miller rare; BMMF—Main Post 
Ribes hirtellum Michx. infrequent; RMHS 
Ribes sativum Syme rare; SH— Main Post 


RAGA 

yet ae heterophyllum Michx. infrequent; 

Mpyriophyllum humile (Raf.) Morong rare; a oath Post 
Proserpinaca palustris L. var. crebra Fern. & Griscom infrequent; INCLS 


HAMAMELIDACEAE 
Hamamelis virginiana L. common; AOPF 


JUGLANDACEAE 
Carya cordiformis (Wangenh.) K. Koch infrequent, BMMF 
Carya glabra (Miller) Sweet common; SU.S 
Carya ovata (Miller) K. nee occasional SH 
Juglans cinerea L. rare; — Main Post 
Juglans nigra L. rare; ean Post 


LAMIACEAE (LAB ) 
Galeopsis tetrahit L. var. bifida en a & Courtois infrequent; FF 
Glecoma hederacea L. rare; SH—Sou 
Hedeoma pulegioides ee Pers. rare; Hcl “Msn Post 
Leonurus cardiaca L. rare; BCL—Sou 
Lycopus americanus Muh. men oe 
Lycopus uniflorus Michx. occasional; SEM 
Lycopus virginicus L. infrequent; INCLS 
ntha arvensis L. var. canadensis (L.) Kuntze infrequent; MF 
Nepeta cataria L. rare; BCL 
Physostegia virginiana (L.) Benth. var. virginiana rare, BCL—South Post 
Prunella vulgaris L. var. lanceolata (Barton) Fern. occasional; BCL 
Pycnanthemum tenuifolium Schrader rare, SH—South Post 
Scutellaria galericulata L. infrequent; SM 
Scutellaria lateriflora L. rare, RMHS—South Post 
Trichostema dichotomum L. occasional; BCL 


EAE 
Lindera benzoin (L.) Blume common; HHS 
Sassafras albidum (Nutt.) Nees occasional; AOPF 


232 Rhodora 


LENTIBULARIACEAE 
Utricularia gibba L. infrequent; SM 
Utricularia minor L. rare; OL—South Post 
Utricularia purpurea Walter infrequent; EP 
Utricularia radiata Small infrequent; EP 
Utricularia vulgaris L. infrequent; EP 


LYTHRACEAE 
Decodon verticillatus (L.) Elliott ee DSB 
Lythrum salicaria L. infrequent; 


MELASTOMATACEAE 
Rhexia virginica L. infrequent; AOPF 


MOLLUGINACEAE 
Mollugo verticillata L. infrequent; BCL 


MONOTROPACEAE 
Monotropa hypopithys L. infrequent; AOPF 
Monotropa uniflora L. infrequent; AOPF 


MYRICACEAE 
Comptonia peregrina (L.) J.M. Coulter occasional; BCL 
Myrica gale L. infrequent; SS 
Myrica pensylvanica Mirbel. rare; PNHF—Main Post 


NYCTAGINACEAE 
Mirabilis nyctaginea (Michx.) MacMillan infrequent; BCL 


NYMPHAEACEAE 
Nuphar variegata Durand infrequent; MCS 
Nymphaea odorata Aiton var. odorata occasional; MCS 


LEACEAE 
Fraxinus americana L. occasional; SH 
Fraxinus nigra ees infrequent, RMHS 
Ligustrum vulgare L. rare, SH—Main Post 
Syringa vulgaris L. rare; > SH South Post 


ONAGRACEAE 
Circaea alpina L. var. alpina infrequent, RMHS 
Circaea lutetiana L. var. canadensis L. occasional; R 


RMHS 
Epilobium angustifolium L. var. angustifolium infrequent; PPSOB 


Epilobium glandulosum Lehm. occasional; MF 


[Vol. 97 


1995] Hunt et al.— Fort Devens Flora 


Epilobium leptophyllum Raf. rare; SEM— Main Post 
Epilobium palustre L. rare; DSB—South Post 
Ludwigia palustris (L.) Elliott infrequent; MS 
Oenothera biennis L. occasional; BCL 

Oenothera parviflora L. rare; PPSOB—North Post 
Oenothera perennis L. occasional; BCL 

Oenothera tetragona Roth rare, BCL— Main Post 


OBANCHACEAE 
Orobanche uniflora L. var. uniflora rare; BCL 


IDACEAE 
Oxalis stricta L. infrequent; BCL 


AVERACEAE 
Chelidonium majus L. infrequent; BCL 
Sanguinaria canadensis L. rare, BAMF—Main Post 


PHYTOLACCACEAE 
Phytolacca americana L. infrequent; PPSOB 


PLANTAGINACEAE 
Plantago aristata Michx. infrequent; SNSG 
Plantago lanceolata L. occasional; BCL 
Plantago major L. infrequ ae 
Plantago rugelii Decne. ae SOF 


PLATANACEAE 
Platanus occidentalis L. infrequent; SH 


POLEMONIA 
Phlox ae L. rare; SH—South Post 


POLYGALACEAE 
Polygala paucifolia Willd. infrequent, AOPF 
Polygala polygama Walter var. obtusata Chodat infrequent; SNSG 
Polygala sanguinea L. occasional; MF 


POLYGONACEAE 
Polygonella articulata (L.) Meissner occasional; BCL 
Polygonum amphibium L. var. emersum Michx. infrequent; FF 
Polygonum arifolium L. infrequent; RM 
Polygonum aviculare L. occasional; BCL 
Polygonum careyi Olney rare; INCLS—South Post 


233 


234 Rhodora [Vol. 97 


acon cilinode Michx. infrequent, RMHS 
Polygonum cuspidatum Sieb. & Zucc. infrequent; SH 
Sel hydropiper L. infrequent; RMHS 
Polygonum hydropiperoides Michx. occasional; FF 
oe ea lapathifolium L. infrequent; FF 
Polygonum pensylvanicum L. occasional; BCL 
Polygonum persicaria L. infrequent; 
Polygonum punctatum Elliott var. confertiflorum (Meissner) Fassett infrequent; 


SEM 
Polygonum sagittatum L. occasional; SEM 
Polygonum scandens L. var. scandens occasional, BCL 


Rumex obtusifolius L. occasional; LD 

Rumex orbiculatus A. Gray rare; MF—South Post 

Rumex salicifolius J.A. Weinm. var. triangulivalvis (Danser) Hickman rare; 
MS— Main Post 


PORTULACACEAE 


Portulaca oleracea L. rare; BCL—South Post 


PRIMULACEAE 


Lysimachia ciliata L. infrequent; BCL 
Lysimachia quadrifolia L. occasional; BCL 

ysimachia terrestris (L.) BSP. common; INCLS 
| faa borealis Raf. common; RMHS 


PYROLACEAE 


Chimaphila maculata (L.) Pursh var. maculata infrequent; AOPF 
Chimaphila umbellata (L.) Barton var. cisatlantica S. F. Blake infrequent, AOPF 
Pyrola chlorantha Swartz rare, AOPF—South Post 

Pyrola elliptica Nutt. infrequent; FF 

Pyrola rotundifolia L. var. americana (Sweet) Fern. infrequent; AOPF 


CULACEAE 

Actaea alba (L.) Miller infrequent; HNHF 
Actaea rubra (Aiton) Willd. infrequent; BMMF 
Anemone quinquefolia L. occasional; R 
Anemonella thalictroides (L.) Spach rare, BMMF—North Post 
Aquilegia canadensis L. infrequent; BC 
Caltha palustris L. infrequent, RMHS 
Clematis virginiana L. occasional: SS 
Coptis trifolia (L.) Salisb. occasional; PNHF 

epatica americana (DC.) Ker Gawler rare; BMMF—North Post 
Ranunculus abortivus L. var. yee tee infrequent; FF 
Ranunculus acris L. infrequent; BCL 
Ranunculus bulbosus L. infrequent; BCL 


19 


95] Hunt et al.— Fort Devens Flora 233 


Silanes flabellaris Raf. rare; SEM—Main Post 
culus recurvatus Poiret infrequent; S 
fain repens L. infrequent; PNHF 
Ranunculus sceleratus L. var. sceleratus rare; EP—North Post 
Thalictrum pubescens Pursh occasional; SS 


RHAMNACEAE 


RO 


Ceanothus americanus L. var. americanus infrequent; BCL 
Rhamnus cathartica L. infrequent; 
Rhamunus frangula L. occasional; MF 


SACEAE 
Agrimonia gryposepala Wallr. infrequent, BMMF 
Amelanchier arborea (Michx. f.) Fern. occasional; AOPF 


Amelanchier spicata (Lam.) K. Koch rare; PPSOB—North Post 
Aronia arbutifolia (L.) Elliott Seen AOPF 

Aronia melanocarpa (Michx.) Elliott occasional; FF 

Aronia prunifolia (Marshall) Rehder infrequent, AOPF 
Crataegus chrysocarpa Ashe rare; PNHF—South Post 
Crataegus coccinea L. rare, PNHF—South Post 

Crataegus flabellata (Bosc) K. Koch rE Caen, SH 

Crataegus intricata Lange infrequent 

Crataegus pruinosa (Wendl.) K. ech infrequent; PNHF 
Crataegus succulenta Schrader rare; PNHF—Main Post 
Fragaria virginiana Duchesne occasional; SH 

Geum canadense Jacq. rare; FF 

Geum laciniatum Murray infrequent; SH 

Geum rivale L. rare; MF—South Pos 

Malus baccata (L.) Borkh. rare; SH—South Post 

Malus floribunda Siebold ex Van Houtte rare; PPSOB—North Post 
Malus pumilla Mill. infrequent; SU.S 

Malus sieboldii (Regel) Rehd. rare; BCL—North Post 
Physocarpus opulifolius (L.) Maxim. var. ae rare; BMMF— Main Post 
Potentilla argentea L. infrequent; BCL 

Potentilla canadensis L. occasional; SU.S 

Potentilla norvegica L. occasional; BCL 

Potentilla recta L. apnea i, 

Asie simplex Michx. BCL 

Prunus americana Marshall rare; e: PPSOB— Main Post 

baa i ea L. f. occasional; SNSG 

Prunus pumila L. var. cuneata (Raf.) L.H. Bailey infrequent; SNSG 
Prunus eis Ehrh. common; SNSG 

Prunus virginiana L. occasional; SOF 

Rosa carolina L. infrequent, PPSOB 

Rosa multiflora Thunb. infrequent; BCL 

Rosa palustris Marshall occasional; RMHS 

Rosa virginiana Miller infrequent; AOPF 


236 Rhodora [Vol. 97 


Rubus allegheniensis T.C. Porter occasional; BCL 

Rubus flagellaris Willd. common; BCL 

Rubus hispidus L. occasional; PPSOB 

Rubus idaeus L. var. strigosus (Michx.) Maxim. occasional; RMHS 
Rubus occidentalis L. infrequent; BCL 

Rubus pubescens bh occasional; RMHS 

Sorbus aucuparia L. r 

Spiraea alba Duroi var. latifolia (Aiton) ae common; AOPF 
Spiraea nipponica Maxim. rare; SU.S 

Spiraea tomentosa L. var. tomentosa infrequent; SM 


RUBIACEAE 
Cephalanthus occidentalis L. occasional: SS 
Diodia teres Walter infrequent; BCL 
Galium aparine L. infrequent; RMHS 
Galium circaezans Michx. var. hypomalacum Fern. infrequent; PNHF 
Galium mollugo L. rare; BCL—South Post 
Galium obtusum Bigelow var. obtusum rare; FF— North Post 
Galium palustre L. occasional; FF 
Galium tinctorium L. var. tinctorium infrequent; SEM 
Galium trifidum L. var. trifidum infrequent; SS 
Galium triflorum Michx. occasional; PNHF 
Hedyotis caerulea (L.) Hook. occasional; BCL 
Mitchella repens L. occasional; PNHF 


SALICACEAE 

Populus balsamifera L. rare; LD— Main Post 
Populus deltoides Marshall var. deltoides infrequent; AOPF 
Populus grandidentata Michx. a CL 
Populus nigra L. rare; SH—Main Pos 

Populus tremuloides Veer ccesion SNSG 
Populus xjackii Sarg. infreque 

Salix babylonica L. rare; SS— ere Post 

Salix bebbiana Sarg. occasional; BCL 

Salix discolor Muhl. rare; SU.S—South Post 

Salix eriocephala Michx. infrequent; BCL 

Salix humilis Marshall infrequent; PPSOB 

Salix nigra Marshall occasional; SS 

Salix petiolaris J. E. Smith rare; BCL— North Post 
Salix sericea Marshall occasional; SM 


SANTALACEAE 
Comandra umbellata (L.) Nutt. var. umbellata infrequent, AOPF 


SARRACENIACEAE 
Sarracenia purpurea L. var. purpurea infrequent; DSB 


1995] Hunt et al.—Fort Devens Flora 237 


SAXIFRAGACEAE 
Chrysosplenium americanum Schwein. occasional; RMHS 
Penthorum sedoides L. infrequent; SEM 
Philadelphus eid L. rare; pes S—South Post 
Saxi pensylvanica L. infreq MF 
Tiarella cordifolia L. var. sordifelia infrequent; HHS 


SCROPHULARIACEAE 
Agalinis purpurea (L.) Pennell var. parviflora (Benth.) B. Boivin occasional; 
BCL 


Aureolaria pedicularia (L.) Raf. var. pedicularia Sito AOPF 

Aureolaria virginica (L.) Pennell rare; AOPF— 

Chaenorrhinum minus (L.) Lange rare; BCL— hea Post 

Chelone glabra L. infrequent; MF 

Gratiola aurea Pursh infrequent; INCLS 

Linaria canadensis (L.) Dum.-Cours. infrequent; BCL 

Linaria vulgaris Miller infrequent; PPSOB 

Lindernia dubia (L.) Pennell var. dubia infrequent; F 

Melampyrum lineare Desr. var. americanum pte Beauverd occasional; 
COF 


Mimulus ringens L. var. ringens pies SM 

Scrophularia lanceolata Pursh rare; PPSOB— Main Post 

Scrophularia marilandica L. rare; SH—Main Post 

Verbascum thapsus L. infrequent; BCL 

Veronica arvensis L. rare; ML—North Post 

Veronica chamaedrys L. rare; SH—South Post 

Veronica officinalis L. rare; SOF—South Post 

Veronica peregrina L. var. peregrina infrequent; BC 

Veronica peregrina L. var. xalapensis (HBK.) St. J a & Warren rare; BCL— 


outh Post 
Veronica scutellata L. infreque 
Veronica serpyllifolia L. var. ae infrequent; BCL 


OLANACEAE 
Physalis heterophylla Nees var. ambigua (A. Gray) Rydb. rare; SOF—North 
Post 


Solanum carolinense L. var. carolinense rare, BCL—South Post 
Solanum dulcamara L. infrequent; SEM 
Solanum nigrum L. var. virginicum L. infrequent; BCL 


LIACEAE 
Tilia americana L. infrequent; SH 


MACEAE 

Ulmus americana L. occasional; SH 

Ulmus parvifolia Jacq. rare; SU.S— Main Post 
Ulmus pumila L. rare; SA.M— Main Post 


238 Rhodora [Vol. 97 


URTICACEAE 
Boehmeria cylindrica (L.) Swartz occasional; SEM 
Pilea pumila (L.) A. Gray occasional; PNHF 
Urtica dioica L. var. dioica occasional; FF 


VERB 
sae bracteata Lagasca & Rodriguez infrequent; PPSOB 
erbena hastata L. infrequent; RSGB 
Verbena urticifolia L. var. urticifolia rare; SH—South Post 


OLACEAE 
Viola conspersa Reichenb. infrequent; BCL 
Viola cucullata Aiton occasional; RMHS 
Viola lanceolata L. var. lanceolata occasional; INCLS 
Viola macloskeyi F. Lloyd var. pallens (Banks) C. L. Hitche. occasional; RMHS 
Viola pedata L. rare; PPOF—North Post 
Viola sagittata Aiton occasional; SNSG 
Viola sororia Willd. occasional; PNHF 


ACEAE 
neds quinquefolia (L.) Planchon occasional; SH 
Parthenocissus vitacea (Knerr) A. Hitchc. common; 

Vitis aestivalis Michx. var. argentifolia (Munson) Fern. infrequent; SH 
Vitis labrusca L. occasional; PNHF 

Vitis riparia Michx. infrequent; SH 


LILIOPSIDA (Monocots) 


ACORACEAE 
Acorus calamus L. infrequent; DEM 


AGAVACEAE 
Yucca filamentosa L. rare, SH—South Post 


ALISMATACEAE 
Alisma bps Pursh infrequent; ee 
Sagitta gelmanniana J. G. Smith infrequent; SEM 


oa eee Willd. var. ila infrequent; RMHS 


ARACEAE 
Arisaema triphyllum (L.) Schott occasional; RMHS 
Calla palustris L. infrequent; RMHS 

Peltandra virginica (L.) Schott & Endl. infrequent; OL 
Symplocarpus foetidus (L.) Nutt. occasional; RMHS 


1995] Hunt et al.—Fort Devens Flora 239 


COMMELINACEAE 
elina communis L. rare; SH— Main Post 
Tradescantia ohiensis Raf. rare; SOF—South Post 
Tradescantia virginiana L. rare, SOF—South Post 


RACEAE 
Bulbostylis capillaris (L.) C. B. Clarke occasional; BCL 
Carex albicans Willd. var. emmonsii (Dewey) Rettig infrequent; PPSOB 


; DSB 
Carex atlantica L. Bailey var. capillacea (L. Bailey) Cronq. occasional; BSTB 
Carex bebbii (L.H. Bailey) Fern. infrequent; BCL 
Carex blanda Dewey infrequent; SH 
Carex brevior (Dewey) Mackenzie infrequent; SOF 
Carex bromoides Willd. infrequent, RMHS 
Carex brunnescens (Pers.) Poiret infrequent; MF 


Carex cephalophora Muh. var. cephalophora rare; BMMF— Main Or 
Carex cephalophora Muhl. v 
South Post 


Carex comosa F. Boott infrequent; SEM 

Carex crinita Lam. infrequent; SM 

Carex debilis Michx. var. rudgei L. Bailey occasional; BCL 
Carex digitalis Willd. infrequent; PNHF 

Carex echinata Murray var. echinata infrequent; SM 
Carex foenea Willd. rare; RMHS—South P 

Carex folliculata L. occasional; RMHS 

Carex gracillima Schwein. infrequent; SH 

Carex gynandra Schwein. occasional; RMHS 

Carex hystericina Muhl. rare; SM—South Post 

Carex intumescens Rudge ovcasional: ae 

Carex lacustris Willd. occasional; BST 

Carex laevivaginata (Kiik.) Mackenzie cae RMHS 
Carex lasiocarpa Ehrh. var. americana Fern. infrequent; SS 
Carex laxiculmis Schwein. rare, SH—South Post 

Carex laxiflora Lam. infrequent; AOPF 

Carex leptalea Wahlenb. rare; MF—South Post 

Carex leptonervia (Fern.) Fern. infrequent; PNHF 

Carex lupulina Muhl. infrequent, RMHS 

Carex lurida Wahlenb. occasional; SM 

Carex muhlenbergii Schk. var. muhlenbergii infrequent; AOPF 
Carex normalis Mackenzie infrequent; HS 

Carex pensylvanica Lam. abundant; AOP 

Carex prasina Wahlenb. infrequent, RMHS 

Carex projecta Mackenzie infrequent; FF 

Carex radiata (Wahlenb.) Small occasional; SEM 


240 Rhodora [Vol. 97 


Carex rostrata J. Stokes infrequent; SEM 
Carex scabrata Schwein. infrequent, BMMF 
Carex scoparia Schk. common; BCL 
Carex stipata Muhl. var. stipata infrequent; SEM 
Carex stricta Lam. occasional; SM 
Carex swanii (Fern.) nee common; BCL 
Carex tenera Dewey infrequen 
Carex tribuloides Wahlenb. ane FF 
Carex trisperma Dewey occasional; BSTB 
Carex typhina Michx. rare; PNHF—South Post 
Carex umbellata Schk. afrecuient: PPSOB 
Carex vesicaria L. rare; FF—South Post 
Carex vestita Willd. occasional; SNSG 
Carex vulpinoidea Michx. infrequent; BCL 
Cyperus bipartitus Torr. infrequent; SEM 
Cyperus dentatus Torr. occasional; SU.S 
Cyperus erythrorhizos Muhl. ee SM 
erus esculentus L. infrequent; LD 
Cyperus filiculmis Vahl eal AOPF 
Cyperus houghtonii Torr. infrequent, SNSG 
Cyperus strigosus L. occasional; SM 
Dulichium arundinaceum (L.) Britton occasional; SM 
Eleocharis acicularis (L.) Roemer & Schultes infrequen 
Eleocharis obtusa (Willd.) J.A. Schultes var. ovata ae ae & Mohlen- 
brock (Z. ovata (Roth) Roemer & Schultes, in part) occasional; SM 
Eleocharis ovata (Roth) Roemer & Sch ne be sensu stricto infrequent; SM 
Eleocharis smallii Britton rare, FF—Sou 
Eleocharis tenuis (Willd.) Schultes var. a ae Gleason infrequent; 


Eriophorum virginicum L. infrequent; SEM 
Rhynchospora alba (L.) Vahl infrequent; DSB 
Rhynchospora capitellata (Michx.) Vahl occasional; SEM 
Scirpus cyperinus (L.) Kunth occasional; 

Scirpus expansus Fern. infrequent; SEM 

Scirpus georgianus R. M. Harper infrequent; SH 

Scirpus hattorianus Makino oe ei RMHS 

Scirpus xpeckii Britton rare; SH— n Post 

Scirpus validus Vahl infrequent; oa 


ERIOCAULACEAE 
Eriocaulon aquaticum (Hill) Druce infrequent; INCLS 


ROCHARITACEAE 
Elodea canadensis Michx. infrequent; SEM 
Elodea nuttallii (Planchon) St. John infrequent; MS 


DACEAE 
Iris pseudacorus L. rare; SEM—South Post 


1995] Hunt et al.—Fort Devens Flora 


Tris versicolor L. occasional; SM 

Sisyrinchium angustifolium Miller rare, BCL— Main Post 
Sisyrinchium atlanticum E. Bickn. infrequent; SU.S 
Sisyrinchium montanum Greene infrequent; SOF 


JUNCACEAE 
Juncus acuminatus Michx. infrequent; SEM 
Juncus bufonius L. var. bufonius rare; BCL— Main Post 
Juncus canadensis J. Gay infrequent; SM 
Juncus effusus L. var. solutus Fern. & Weigand occasional; MF 
Juncus greenei Oakes & Tuckerman occasional; SNSG 
Juncus marginatus Rostk. infrequent; eee 
Juncus pelocarpus E. Meyer infrequent; INCLS 
Juncus tenuis Willd. var. ions (Elliott) A. Wood infrequent; BCL 
Juncus tenuis Willd. var. tenuis occasional; AOPF 
Luzula multiflora (Retz.) Lej. occasional; BC 


LEMNACEAE 
Lemna minor L. infrequent; OL 
aioe polyrhiza (L.) Schleiden infrequent; MS 
Wolffia brasiliensis Weddell (W. papulifera C. Thompson) infrequent; OL 
Wolffia columbiana Karsten infrequent; EP 


LILIACEAE 
Allium canadense L. var. canadense infrequent; BCL 
Asparagus officinalis L. rare, SOF—South Post 
Clintonia borealis (Aiton) Raf. occasional; MF 


Polygonatum pubescens (Willd.) Pursh infrequent, BMMF 
Smilacina racemosa (L.) Desf. occasional; AOPF 

Smilacina stellata (L.) Desf. var. stellata rare; FF—North Post 
Smilacina trifolia (L.) Desf. rare; BSTB—South Post 

Trillium cernuum L. infrequent; PNHF 

Trillium undulatum Willd. infrequent; BSTB 

Uvularia perfoliata L. rare, HNHF—Main Post 

Uvularia sessilifolia L. occasional; RMHS 

Veratrum viride Aiton infrequent; RMHS 


NAJADACEAE 
Najas flexilis (Willd.) Rostkov & Schmidt infrequent; EP 


241 


242 Rhodora [Vol. 97 


ORCHIDAC 
Cyprip edit ium acaule Aiton occasional; AOPF 
Epipactis helleborine (L.) Crantz rare; BMMF— Main Post 
Goodyera pubescens (Willd.) R. Br. infrequent; AOPF 
Habenaria clavellata (Michx.) Sprengel infrequent; HHS 
Habenaria psycodes (L.) Sprengel var. grandiflora (Bigelow) A. Gray infrequent; 
MF 


Habenaria psycodes (L.) Sprengel var. psycodes infrequent; RMHS 

Spiranthes cernua (L.) Rich infrequent; INCLS 

Spiranthes lacera (Raf.) Raf. var. eraeils (Bigelow) Luer rare; PPSOB—South 
Post 


POACEAE (GRAMINEAE) 
Agrostis capillaris L. infrequent; 
Agrostis hyemalis (Walter) BSP. var. hyemalis infrequent; SA.M 
4 grostis lis (Walter) BSP. var. scabra eee ein eae INCLS 
Agrostis perennans (Walter) Tuckerman var. perennans a PF 
Agrostis stolonifera L. var. palustris (Huds.) Farw. eee FF 
Alopecurus pratensis L. rare; SOF— Main Post 
Andropogon gerardii Vitman infrequent; SOF 
Andropogon virginicus L. var. virginicus pa aneauene INCLS 
Anthoxanthum odoratum L. occasional; BCL 


’ 


gi salad erectum ee Bea var. glabratum (Vasey) T. Koyama 
o occasion NHF 
Bromus ak 1 ee t; B 
Bromus inermis Leysser eee SOF 
Bromus japonicus Thunb. rare; SOF—South Post 
Bromus tectorum L. infrequent; SOF 
Calamagrostis canadensis (Michx.) P. Beauv. occasional; SEM 
Calamagrostis cinnoides (Muhl.) Barton infrequent; BC 
Cenchrus longispinus (Hackel) Fern. infrequent; BCL 
Cinna arundinacea L. occasional; SH 
Cinna latifolia (Trevir.) Griseb. iegent RMHS 
CL 


Deschamp 

Digitaria oe (Schieber) Muhl. infrequent: MF 
Digitaria sanguinalis (L.) Scop. occasional; ML 
Echinochloa muricata (P. Beauv.) Fern. var. muricata occasional; FF 
Elymus virginicus L. infrequent; 

Elytrigia repens (L.) Nevski a BCL 

Eragrostis capillaris (L.) Nees rare; SU.S— Main Post 
Eragrostis peeuete (All.) aie rare; a South Post 
Eragrostis minor Host rare; ML—North Pos 

Eragrostis pe ie cea (Mi coe .) Nees occasion ot ML 
Eragrostis spectabilis (Pursh) Steudel occasional: SNSG 


1995] Hunt et al.—Fort Devens Flora 243 


Festuca elatior L. rare; BCL—North Post 
Festuca filiformis Pourret infrequent; BCL 


Festuca rubra L. occasional; BCL 

Festuca ies an ) E. Alexeev. sear FF 
Glyceria acutiflora . rare; INCLS— 

Glyceria borealis aa Batchelder rare; $8—Main Post 
Glyceria canadensis (Michx.) Trin. infrequent; MF 

Glyceria grandis 8. Wats. infrequent; SS 

Glyceria melicaria (Michx.) C. E. Hubbard infrequent; RMHS 
Glyceria striata (Lam.) A. Hitche. var. striata occasional; RMHS 
Leersia oryzoides (L.) Swartz infrequent; SM 

Leersia virginica Willd. occasional; RMHS 

Leptoloma cognatum (Schultes) Chase infrequent; BCL 

Lolium perenne L. var. aristatum Willd. infrequent; BCL 
Lolium perenne L. var. perenne infrequent; BCL 

Muhlenbergia frondosa (Poiret) Fern. infrequent; SH 
Muhlenbergia glomerata (Willd.) Trin. var. glomerata infrequent; SU.S 
Muhlenbergia mexicana (L.) Trin. infrequent; SH 

Muhlenbergia uniflora (Muhl.) Fern. infrequent; BCL 

Oryzopsis asperifolia Michx. infrequent; PNHF 

Oryzopsis pungens (Torr.) A. Hitche. rare; HNHF—Main Post 
Pani PF 


SN 
Panicum columbianum Scribn. occasional; BCL 
Panicum depauperatum Muhl. infrequent; PPSOB 
Panicum dichotomiflorum Michx. infrequent; SNSG 
Panicum dichotomum L. infrequent; AOPF 
Panicum lanuginosum Elliott var. implicatum (Scribn.) Fern. occasional; BCL 
Panicum lanuginosum Elliott var. lindheimeri (Nash) Fern. rare; COF—South 
Post 


Panicum lanuginosum Elliot var. tennesseense (Ashe) Gleason occasional; AOPF 
Panicum linearifolium Scribn. rare; B 

Panicum oligosanthes Schultes occasional; SNGS 

Panicum philadelphicum Bernh. rare; SM—South Post 

Panicum rigidulum Nees infrequent; INCLS 

Panicum virgatum L. rare; RSGB—South Post 

Paspalum setaceum Michx. var. muhlenbergii (Nash) D. Banks infrequent; ML 
Phalaris arundinacea L. occasional; BCL 

Phleum pratense L. occasional; BC 

Phragmites australis (Cav.) Trin. rare; PNHF—South Post 

Poa alsodes A. Gray infrequent; 

Poa annua L. infrequent; BCL 

Poa compressa L. occasional; BCL 

Poa palustris L. resuent, FF 

Poa pratensis L. comm CL 

Puccinellia fernaldii (A. Hitchc.) E. G. Voss rare; FF—South Post 

Puccinellia pallida (Torr.) R. T. Clausen infrequent; SS 


244 Rhodora [Vol. 97 


Schizachyrium scoparium (Michx.) Nash var. scoparium abundant; AOPF 
Secale cereale L. rare; BCL—South Post 

Setaria faberi R. Herrm. rare; LD— Main Post 

Setaria glauca (L.) P. Beauv. infrequent; ML 

Setaria viridis (L.) P. Beauv. infrequent; ML 

Sporobolus vaginiflorus (Torr.) A. Wood infrequent; LD 

Triticum aestivum L. rare; BCL—Main Post 


PONTEDERIACEAE 
Pontederia cordata L. occasional; DEM 


POTAMOGETONACEAE 
mogeton crispus L. rare; MS—Main Post 


Potamogeton zosteriformis Fern. infrequent; OL 


SMILAC 
pase? herbacea L. infrequent; FF 
Smilax rotundifolia L. infrequent; RMHS 


SPARGANIACEAE 
Sparganium americanum Nutt. infrequent; DEM 
Sparganium chlorocarpum Rydb. infrequent; SEM 
Sparganium eurycarpum Engelm. infrequent; S 
Sparganium minimum (Hartm.) Fr. rare; SEM— Main Post 


TYPHACEAE 
Typha latifolia L. occasional; SEM 


RHODORA, Vol. 97, No. 891, pp. 245-254, 1995 


FRAGARIA MULTICIPITA, REDUCED TO THE 
RANK OF FORMA 


PAUL M. CATLING, JACQUES CAYOUETTE, AND JOSEPH POSTMAN 


ABSTRACT 


The distinctive features of Fragaria dheaiaediss Fernald, including small size, 
multicipital habit, floral aberrations and absence of runners to a greater or lesser 
degree, are symptoms of strawberry multiplier ee resulting from mycoplasma 
infection. Direct evidence of the disease in F. multicipita plants was obtained 
through graft inoculation using normal F. chiloensis which developed increasingly 
pronounced multiplier disease-like symptoms after three months. Plants referable 

to F. multicipita were not confined to specialized habitats. The chromosome 
number of 2” = 56 obtained from five small multicipital plants representing four 
locations was the same as that widely reported for the F. virginiana complex in 
North America. These observations suggest that F. multicipita is only a diseased 
form of F. virginiana, despite the stability of its distinctive traits in cultivation 
and the potential adaptive nature of these traits on cool disturbed rivershores. 
Accordingly, the new combination, Fragaria virginiana Duch. ssp. glauca (S. 
Wats.) Staudt f. multicipita (Fern.) Catling & J. Cayouette, is proposed 


Key Words: aie multicipita, Fragaria virginiana, strawberry, taxonomy, dis- 
e, mycoplasma, endemism, Québec, Canada 


INTRODUCTION 


Fragaria multicipita was described by Fernald in 1908 without 
any discussion of its origin and affinities. Plants referable to it 
were not found again until 1992 (Catling, 1993). Some of the 
characteristics of F. multicipita are suggestive of a subarctic or 
ice front relict. The cushion-like habit increasing temperature and 
resisting abrasion, parabolic flowers, and the genotypic dwarfing 
are features of plants of cold, exposed environments (Savile, 1972). 
The multiple crowns accumulate debris more readily than the 
more familiar fewer-crowned plants growing nearby (Catling, pers. 
obs.). These potential adaptations and occurrence in a specialized 
rivershore environment, as well as a suggestion of stability of 
distinctive traits in uniform garden culture, are consistent with 
species rank (Catling, 1993) and the concept of an ice front relict. 
Such speculations can become attractive and some botanists have 
conjectured (pers. comm.) that the small size and relict nature of 
F. multicipita make it a potential diploid or tetraploid progenitor 
of the North American octoploid Fragaria virginiana complex. 


245 


246 Rhodora [Vol. 97 


On the other hand Reed (1966) speculated that “F. multicipita 
may be virus-infected plants,” and few authors have perpetuated 
the name (Catling, 1993). As noted previously, a decision on the 
appropriate rank required more study (Catling, 1993). Here we 
report on some recent studies including: (1) the nature of dis- 
tinctive morphological traits, (2) the possibility of disease being 
a causal factor, (3) consideration of habitat aspects, and (4) chro- 
mosome numbers. 


METHODS 


Morphological Observations 


Material of Fragaria from throughout the Gaspé peninsula was 
maintained in cultivation in the glass house in Ottawa over a 
two-year period. Included were 22 plants referable to F. virginiana 
Duch. ssp. virginiana with spreading hairs on the scapes and 
petioles, and 30 referable to Fragaria virginiana ssp. glauca (S. 
Wats.) Staudt (including F. viginiana var. terrae-novae (Rydb.) 
Fern. & Wieg.) with ascending hairs on scapes and petioles, and 
10 plants referable to F. multicipita Fern., including some with 
spreading petiole hairs and some with ascending petiole hairs. 
These plants were grown in similar soil mix in similar pots and 
were subject to similar watering and light regimes. The extent to 
which distinctive features of the plants at the time of collection 
were maintained in cultivation was noted. 


Evidence for Disease as a Causal Factor 


The literature on strawberry diseases was surveyed to determine 
which diseases might produce a small-leaved, bushy strawberry 
lacking runners. Direct evidence of mycoplasma infection was 
sought through standard leaflet graft inoculation (Converse, 1987a). 
This technique involved cutting petioles of test plants into a ‘“V”’ 
shape and inserting them into the cut petiole of an indicator plant 
established as appropriate through previous experimentation. Five 
normal F. virginiana plants including ssp. glauca (N2, N9, N15) 
with ascending hairs and ssp. virginiana (N8) with spreading hairs, 
and 3 multicipital plants (including one with spreading hairs (M15), 
and one with ascending hairs (M23)) were grafted to normal F. 


1995] Catling et al.—Fragaria multicipita 247 
chiloensis plants. Vouchers of material used in graft inoculation 
were placed in DAO and MT. 


Habitat Aspects 


Field work in August 1993 was directed to determining whether 
or not plants referable to F. multicipita were confined to special- 
ized habitats, and the extent to which they occurred with other 
Fragaria taxa. 


Chromosome Numbers 


Young root tips of plants used for cytological study (Tables | 
and 2) were collected and pre-treated in water, refrigerated for 
three hours, fixed in Farmer’s fixative (glacial acetic acid:absolute 


Table 1. Locations, DAO voucher number, plant appearance, collection num- 
ber, petiole hair orientation and root tip chromosome determination for various 
strawberry plants collected along the Riviére Ste.-Anne in Gaspé, Québec, that 
were used in the study. 


Voucher Coll. no./petiole 
Location Appearance orientation 2n= 

1. 49°02'55"/66°28'50"” 682393 multicipital Mll/ascending ca. 56 

gravel bar on W side 682394 multicipital M10/ascending 56 

of Riviére Ste.-Anne 9 686462 normal 2/spreading 56 

km SSE of Ste.-Anne- 686451 normal N 10/ascending 56 

des-Monts none normal N8/spreading none 
2. 49°00'50"/66°28'15” 

gravel bar on W side 

of Riviére Ste.-Anne 686473, 74, multicipital M15/ascending 56 

13 km SSE of Ste.- 77,92 

Anne-des-Monts 686461 normal N15/ascending ca. 56 
3. 48°55'20"/66°06'30" 

gravel roadside on W 

side of hwy 299 at 686472, multicipital M9/ascending 56 

Chute du Diable, Ri- 78, 91 

viére Ste.-Anne 686449 normal N9/ascending ca. 56 
4. 49°05'40"/66°30'20” 

gravel bar on W side 

of Riviére Ste.-Anne 4 686475 multicipital M24/ascending 56 

km SSE of Ste.-Anne- 686450 norma N 1/ascending 56 

des-Monts none multicipital M23/spreading none 


248 Rhodora [Vol. 97 


Table 2. Additional notes on particular voucher specimens prepared from 
cultivated plants. 


686475. Plants with few runners less than 10 cm long developed over 50 
crowns as did the daughter plants from short runners, but leaflets remained 
less than 20 mm long and petioles less than 7 cm long after two years in 
culture. 

686472, 686478, 686491. After culture for one year, developed over 100 

owns but retained small leaflets less than 28 mm long, and short petioles 
less than 10 cm long. Inner floral parts of many, but not all flowers, devel- 
oped into new plants and petals in some cases were green, reduced and as- 
cending. 

oe set 686477, 686492. After two years in cultivation, developed 

“bush” with over 150 crowns, but never developed runners and re 
ee its a leaflets less than 32 mm long and short petioles less than 12 
cm long. 

682393. After two years in cultivation, developed into a ‘‘bush” with over 
150 crowns, but never developed runners and retained its small leaflets less 
than 27 mm long and short petioles less than 7 cm long. The petals of many 
flowers were greenish and ascending, but white, spreading petals were pro- 
duced on the same plant. 

682394. After two years in cultivation, developed numerous fragile runners to 
20 cm long but retained its small leaflets less than 20 mm and short petioles 
less than 7 cm long. The petals of many flowers were greenish and ascending 
but white, spreading petals were produced on the same plant. 

686449-51, 686461-2. These plants, referable to F. virginiana ssp. glauca (S. 
Wats.) Staudt, had leaflets 30-80 mm long, petioles 10-20 cm long, less than 
5 crowns and stout runners 20-72 cm long. There were no floral anomalies. 


ethanol, 1:3) for 12 hours and stored in 70% ethanol. Staining 
was done in alchoholic hydrochloric acid-carmine (Snow, 1963). 
Chromosome counts were made on the best cells in late prophase 
or early metaphase. At least two cells from each collection were 
examined. Voucher herbarium specimens for both morphological 
observations and chromosome counts were placed in DAO and 
MT. 


RESULTS AND DISCUSSION 
Morphological Observations 


Small multicipital plants with petioles less than 12 cm and 
leaflets 5-32 mm, referable to F. multicipita when collected, re- 


1995] Catling et al.—Fragaria multicipita 249 


tained their small size and multicipital habit for two years in 
cultivation, although some developed into dense clumps with 
over 200 ramets. The adjacent few-crowned plants referable to 
F. virginiana Duch. ssp. virginiana with spreading hairs on the 
scapes and petioles or Fragaria virginiana ssp. glauca (S. Wats.) 
Staudt with ascending hairs on scapes and petioles, also retained 
their larger size, with petioles 10-20 cm and leaflets mostly 30- 
80 mm long, over the two year period (see also Table 2). It was 
noted previously that the stability of distinctive traits of F. mul- 
ticipita in culture was consistent with its recognition (Catling, 
1993). As might be expected, 10 multicipital plants marked in 
1992 had also retained their distinctive features in nature one 
year later. 

Multicipital plants always had relatively short leaflets less than 
32 mm long and petioles less than 15 cm, but they ranged from 
no runner production, to some that produced few runners less 
than 10 cm long, to some that produced runners to 25 cm long. 
Only the larger multicipital plants produced runners but not all 
of them did so. 

No additional distinctive morphological features were found 
to be associated with the F. multicipita plants, with the exception 
that the majority of plants from three localities, which were not 
flowering when collected, developed aberrant flowers with either 
petals green, reduced in size, more or less connate and ascending, 
or petals normal but inner floral parts developing into new plants. 
No such aberrations were present elsewhere in the glasshouse 
collection, nor were they observed in nature. Indeed such aber- 
rations are rare, so to find them in approximately half of the F. 
multicipita plants, from three out of six known, suggested that 
the plants were diseased. 

Variation in hair type on petioles of F. mu/ticipita plants sug- 
gests these plants may not belong to a monophyletic taxon. As- 
cending hairs on petioles ally F. mul/ticipita to F. virginiana ssp. 
glauca, but at two of the six localities the small, multicipital plants 
had spreading hairs, thus allying them to ssp. virginiana. Such 
pubescence characteristics in Fragaria are stable in cultivation 
and mostly consistent within plants and mostly not intergrading. 
Although a species may have geographically based infrataxa dis- 
tinguished by pubescence orientation, as in F. virginiana or F. 
chiloensis, the appearance of both kinds of pubescence within a 
putative, narrowly confined endemic was a surprise. 


250 Rhodora [Vol. 97 


Disease as a Causal Factor 


In strawberries, small size, multicipital habit, and absence of 
runners to a greater or lesser degree, are well documented symp- 
toms of multiplier and witches’-broom diseases which are asso- 
ciated with mycoplasma infection (viz. Boone, 1970; Converse, 
1987b; Mass, 1984). Thus the distinctive morphology of F. mul- 
ticipita may be simply a consequence of disease. The floral ab- 
errations, noted above as the only other distinctive traits of plants 
referable to F. multicipita, are a result of green petal disease caused 
by a leafhopper-borne mycoplasma (viz. Chiykowski and Craig, 
1975; Cousin et al., 1970; Mass, 1984). 

The plants referable to F. multicipita were also the most difficult 
to maintain in cultivation. Approximately half died over the two 
year period, whereas the loss of plants referable to either ssp. of 
F. virginiana was less than 1%. 

Normal F. chiloensis plants with grafted leaves from multicip- 
ital plants developed symptoms of multiplier disease three months 
after inoculation. These symptoms included reduced leaf size, 
shorter runners and proliferation of the crown. The symptoms 
became increasingly conspicuous in the new growth with in- 
creased time after three months. The normal F. chiloensis with 
grafted leaves from normal plants of F. virginiana ssp. virginiana 
or ssp. glauca collected near muticipital plants produced no dis- 
ease symptoms. The fact that the multicipital and related char- 
acteristics could be transmitted from multicipital plants to normal 
plants suggests very strongly that F. multicipita is a pathogen- 
induced taxon. 


Habitat Aspects 


Additional field study revealed occurrence of plants referable 
to F. multicipita on a gravel roadside (Table 1), where no other 
endemic, disjunct, or unusual plants were present. At each site 
where F. multicipita occurred (Table 1), plants referable to F. 
virginiana ssp. virginiana and/or F. virginiana ssp. glauca were 
also present. Plants referable to F. multicipita could not be found 
in specialized alpine, subalpine or serpentine habitats. The river 
bars do represent a specialized habitat, but are also occupied by 
“weedy” species (Catling, 1993). The distinctive features of F. 
multicipita (or the traits of the diseased plants), i.e. small size and 


1995] Catling et al.—Fragaria multicipita 201 


bushy form (see above) may adapt them to the cool rivershore 
environment where wind and water are important factors, thus 
explaining their increased prevalence along the rivershores. Al- 
ternative, but not mutually exclusive, explanations include the 
possibility that the rivershore habitat is more conducive to the 
spread of the disease and/or that plants in this habitat are more 
susceptible. 


Chromosome Numbers 


The octoploid chromosome number of 2” = 56 obtained from 
five multicipital plants representing four locations (Table 1) is the 
same as that widely reported for the F. virginiana complex in 
North America (Staudt, 1962; Reed, 1966). Plants referable to 
F. virginiana ssp. virginiana or F. virginiana ssp. glauca that were 
growing intermixed or within one metre of the F. multicipita 
plants also had a chromosome number of 2” = 56 (Tables | 
and 2) 


CONCLUSIONS 


Accepting the distinctive morphology of F. multicipita as sim- 
ply a consequence of disease seems to be the most appropriate 
decision since: (1) the relationship between diagnostic traits and 
disease is well documented, (2) other characteristics of the taxon 
such as the difficulty of culture and the high incidence of floral 
aberration are also associated with disease, (3) there are no dis- 
tinctive morphological traits that are not associated with disease, 
(4) multicipital and related characteristics were transmitted from 
multicipital plants to normal plants, (5) plants referable to F. 
multicipita are not confined to specialized habitats and their prev- 
alence along cool rivershores may be explained in terms of the 
adaptive nature of traits of diseased plants, (6) the plants referable 
to F. multicipita always occurred with plants referable to F. vir- 
giniana, (7) its chromosome number is the same as that of the 
widespread octoploid F. virginiana, despite remarkable differ- 
ences in plant size which result in the expectation of a lower 
chromosome number. Consequently we propose the new com- 
bination: 


252 Rhodora [Vol. 97 


Fragaria virginiana Duch. ssp. glauca (S. Wats.) Staudt f. multi- 
cipita (Fern.) Catling & J. Cayouette, stat. et comb. noy. 
BASIONYM: Fragaria multicipita Fernald, Rhodora 10: 49- 
50. 1908. Type: QUEBEC: Gaspé Ouest Municipality: Cap 
Chat Township: gravelly and sandy beaches and bars of the 
River Ste. Anne-des-Monts, 14-17 July 1906, M. L. Fernald 
and J. F. Collins 230 (Holotype: GH!; Isotype: GH!). 


From examination of the holotype of Fragaria multicipita (GH), 
as well as reference to Fernald’s original description, where pet- 
ioles are described as “appressed silky,” it is clear that Fragaria 
multicipita has to be included within the ssp. glauca when trans- 
ferred to F. virginiana. This leaves the small, multicipital plants 
with spreading hairs without a name. We propose not to provide 
a name for these since it is conceiveable that any species of Fra- 
garia could assume this morphology with infection by certain 
mycoplasmas. The advantage in retaining a rank for the name 
multicipita is that it enables the classification system to account 
for the unusual morphology, which would otherwise continue to 
raise questions not readily answerable by a systematist. A similar 
situation exists within the eastern North American Trillium gran- 
diflorum (Michx.) Salisb., where the plants with green striped 
petals, which are sometimes considered as a consequence of my- 
coplasma infection (e.g., Chinnappa, 1982; Hooper et al., 1971: 
Pringle, 1970), are accorded the rank of forma, thus providing a 
name and some associated information on an aberration that is 
frequently an object of questions. 

The status of F. multicipita does not upset the concept of the 
Gulf of St. Lawrence region as a region of endemism since there 
are numerous other endemics known from the region (Catling, 
1993; Catling & Cayouette, 1994). Whereas there are other taxa 
described by Fernald that were later found to be diseased plants 
or freaks, such as Carex elachycarpa Fern. and C. josselynii (Fern.) 
Mackenzie (Reznicek & Ball, 1979), the fact remains that many 
of the taxa he described have been found, through recent detailed 
study, to be worthy of recognition as hybrids (e.g. Juncus x oro- 
nensis Fern., Eleocharis x macounii Fern.) or at, below or even 
above the ranks he ascribed to them (e.g. Malaxis bayardii Fern.., 
Cleistes bifaria Fern.). Thus in the broad context, the fate of F. 
multicipita does not reduce the significance of Fernaldian taxa. 


1995] Catling et al.—Fragaria multicipita 253 


ACKNOWLEDGMENTS 


Mr. W. Wojtas and Ms. S. Porebski assisted with chromosome 
counting. Ms. S. Porebski assisted with the cultivation of material 
in the glasshouse. 


LITERATURE CITED 


Boone, D. M. 1970. Witches’-broom and multiplier diseases of strawberry. In: 
Frazier, N. W., ed. Virus Diseases of Small Fruits and Grapevines. Uni- 
versity of California, Division of Agricultural Sciences, Berkeley: 25-26. 

CATLING, P. M. 1993. Rediscovery of the Many-crowned Strawberry, Fragaria 
multicipita Fernald. Rhodora 95: 225-233 

YOUETTE. 1994. Occurrence, Aen and status of native germ- 
plasm in ric Gaspé Peninsula of Québec, a special reference to small 
fruits. Pl. Genet. Resources Newslett. 100: 

CHINNAPPA, C. C. 1982. Cytology of some ee forms of Trillium gran- 
diflorum (Liliaceae). Caryologia 35: 23-32. 

CurykowskI, L. N. AND D. L. Craic. 1975. Reaction of strawberry cultivars to 
clover phyllody (green petal) agent transmitted by Aphrodes bicincta. Canad. 
Pl. Dis. Surv. 55: 66-68. 

Converse, R. H. 1987a. Detection and elimination of virus and virus-like dis- 
eases in strawberry. Jn; Converse, R. H., ed. Virus Diseases of Small Fruits. 
United States inane of Aerieul gare. Agric. Handb. 631: 2-1 

198 rawberry Witches’-broom and Multiplier diseases. Jn: Con- 
verse, R. H., ie Virus Diseases of Small Fruits. United States Department 
of Aa eure. Agric. Handb. 631: 66-68. 

Cousin, M. T., J. P. MorEAu, A. FAIVRE-AMIOT AND T. STARON. 1970. Sur la 
présence de particules de type mycoplasme chez les fraisiers atteints de la 
maladie <4 Ege verts’’. Ann. Phytopathol. 2: 535-545 

Hooper, G. R., F. W. Case JR. AND R. Myers. 1971. Mycoplasma-like bodies 
associated an a flower greening disorder of wild flower, Trillium grandi- 

m. Pl. Dis. Reporter 55: 1108-1110. 


Mass, LL 1984. Compendium of Strawberry Diseases. American Phytopath- 
ological Society and United States Department of Agriculture. St. Paul, Min- 
nesota. 1-138. 


PRINGLE, J. S. 1970. ae Bares of Ontario. Techn. Bull. Roy. Bot. Gard. 
(Hamilton, Ontario) 5 

REED, C. F. 1966. Wild rome species of the world. Jn; Darrow, G. M., ed. 
The Strawberry, pre Breeding and Physiology. Holt, Rinehart and Win- 
ston, New York: -121. 


REZNICEK, A. A. AN 2 W. BALL. 1979. The status of two New England “‘en- 
demic” Carices: Carex elachycarpa and C. josselynii (Cyperaceae). Taxon 
28: 217-233. 


SAVILE, D. B. O. 1972. Arctic Adaptations in Plants. Monogr. Res. Branch 
Canada Dept. Agric. 6: “$1, 

Snow, R. 1963. ! ! d-carmine as a stain for chromosomes 
in squash preparations. Stain Technol. 38: 9-13. 


254 Rhodora [Vol. 97 


STAUDT, G. 1962. Taxonomic studies in the genus Fragaria, typification of 
ragaria species known at the time of Linnaeus. Canad. J. Bot. 40: 869- 
886. 


Beh 62, 

AGRICULTURE CANADA, BIOLOGICAL RESOURCES DIVISION 
CENTRE FOR LAND AND BIOLOGICAL RESOURCES RESEARCH 
SAUNDERS BUILDING, CENTRAL EXPERIMENTAL FARM 
OTTAWA, CANADA KIA 0C6 


J.P, 

NATIONAL CLONAL GERMPLASM REPOSITORY 
UNITED STATES DEPARTMENT OF AGRICULTURE 
33447 PEORIA ROAD, CORVALLIS, OR 97333 
UNITED STATES 


RHODORA, Vol. 97, No. 891, pp. 255-263, 1995 


STATUS OF THE DEERBERRY, 
VACCINIUM STAMINEUM L. (ERICACEAE), 
IN CANADA* 


Bruce A. Forp 


ABSTRACT 
Vacci tami L. (deerb widespread species in the United States 
that reaches its northern limit i in the Niagara Falls and Thousand Islands area of 
southern Ontario, Canada. Only five stations occur in Ontario, with the most 
extensive populations being found within St. Lawrence Islands National Park. 
Dry, open, rocky woods, with a history of fire, are the preferred habitat for this 
species in Canada. Whereas the largest stations are currently under government 
protection, their proximity to existing trails, a lack of seedling recruitment, and 
encroachment by later successional vegetation suggests that this species could 
become extirpated if the factors affecting its vulnerability are not reversed. For 
these reasons V. stamineum is recognized as a threatened species in Canada. 


Key Words: Vaccinium stamineum, Canada, conservation status, threatened spe- 
cies 


INTRODUCTION 


Vaccinium stamineum L. is one of the most distinctive blue- 
berry species found in North America. A suite of unique features, 
including a deeply 5-lobed, campanulate corolla and exsert sta- 
mens, distinguish this plant from all other Vaccinium L. spp. and 
have resulted in its placement in the monotypic section Polycod- 
ium (Raf.) Rehder. This section has no apparent affinity to Neo- 
tropical, Old World tropical or North American members of the 
genus (Vander Kloet, 1988). 

The deerberry is a highly polymorphic species that exhibits 
considerable variation in the hairiness of the pedicels, hypanthi- 
um, fruit, twig glandularity, the number of flowers per inflores- 
cence, heterophylly, plant glaucescence, and fruit color. As a result 
of this variability, there has been considerable debate as to the 
number of taxa that should be recognized. For example, Ash 
(1931) divided Polycodium into 6 sections and 21 species. On the 


* Based on a COSEWIC status report by the author. Copies of the report are 
available at cost from the Canadian Nature Federation, 1 Nicholas St., Ottawa, 
Ontario K1N 7B7. Threatened status was assigned by COSEWIC on April 14, 
1994 


25) 


256 Rhodora [Vol. 97 


other hand, Camp (1945) considered this taxon little more than 
a series of clines. In more modern taxonomic works, Baker (1970) 
maintained a single species with one additional variety (V. stam- 
ineum var. sericeum (Mohr) Ward). Ward (1974) recognized five 
varieties of V. stamineum in Florida, but conceded that when 
outside populations are examined, a large number of individuals 
are difficult to place. Finally, Vander Kloet (1988), in his mono- 
graph on the genus Vaccinium in North America, recognized only 
one species with no infraspecific taxa. I am following Vander 
Kloet’s circumscription of V. stamineum for this paper. 


DISTRIBUTION 


Vaccinium stamineum is endemic to eastern North America 
where it is found from southern Ontario south to central Florida 
(Figure 1). This species occurs westward to eastern Texas, eastern 
Oklahoma, and southeastern Kansas. It appears to be absent from 
northern Missouri, Illinois, and northern Indiana. A few outlying 
populations occur in central Mexico (Figure 1). In Canada, V. 
stamineum is known from five extant stations in Ontario (Figure 
1). Three stations are found in the Thousand Islands area, Leeds 
County; the other two occur near Niagara Falls, in the Regional 
Municipality of Niagara. By far the largest populations are found 
in the Thousand Islands (Table 1). An additional station occurs 
in the Thousand Islands on Wellesley Island, Jefferson County, 
New York, only a few kilometers from the Ontario populations 
(Crowder, 1982a). 

The occurrence of V. stamineum in Canada has been known 
for almost 200 years. Deerberry was first collected in the Niagara 
region in 1798 (Masson 15436 CAN) and in the Thousand Islands 
in 1876 (Macoun 15437 CAN). Other early collections were made 
in 1896 at Queenston Heights (Scott 14400 CAN), in 1891 in 
Stamford (now part of the city of Niagara Falls) (Macoun 15438 
CAN) and Niagara-on-the-Lake (Dearness 135] DAO), and in 1937 
in St. David’s (Simmons s.n. TRT). 


HABITAT 


In Ontario, V. stamineum occurs most frequently in dry, rocky 
woods with a canopy closure of approximately 40%. Plants are 
not usually found in open sites or in areas with deep shade. When 


\62 lao [rs | |76 


Figure 1. Distribution of Vaccinium stamineum in North America and Ontario. North American map after Vander Kloet (1988). Rectangle 
on North American map indicates location of detail. 


WNAUIWIDIS WINIUIIIDA —P1OF [S661 


Lec 


258 Rhodora [Vol. 97 


Table |. Site locations, representative seni, and clump sizes for extant 
populations of Vaccinium stamineum in Ontario. The term clump is defined as 
a discrete group of stems. A clump may or may not represent more than one 
individual. Clump numbers and sizes for Leeds County populations are those 
documented in an unpublished study by St. Lawrence Islands National Park. 


Clump- Clump Size 
# (m) 


Location 
Regional Municipality of Niagara, City of Niagara Falls, 1. <1.0 x 1.0 
Whirlpool. Representative specimens: Eckel 8604088 
(BUF); Scott s.n. (DAO, TRT) 
Regional Municipality of Niagara, City of Niagara Falls, 1 <1.0 x 1.0 
Bruce Trail near Mewburn Road. Representative 
specimen: Hardy & DeBus s.n. (BUF 
Leeds County, Front of Escott Twp., West end of Gren- 1. 1.6 x 1.6 
adier Island, St. Lawrence Islands National Park. 2. 12x 1.2 
Representative specimens: Cody & Munro 21779 3. 1.0 x 0.6 
(DAO, TRT); Cody & Munro 22740 (pao); Munro s.n. 4. 2.4 x 2.5 
(DAO); Dore et al. 25122 (DA); Woods & Woods s.n. 5: 1.4 x 1.4 
(CAN). 6. 1.4 x 1.5 
Ta. 1.6 x 3.8 
7b. 3.2 x 3.0 
8. 1.0 x 1.0 
9. 1.0 x 1.0 
10. 19 x 2,2 
Leeds County, Front of Leeds and Landsdowne Twp., ie 3.5 x 3.5 
Endymion Island, St. Lawrence Islands National 2. 3.0 x 3.5 
Park. Representative specimens: Ford s.n. (TRTE) (2 
sheets). (see Ford (1984) for further details on this 
population) 
Leeds County, Front of Leeds and Landsdowne Twp., I 1.8 x 2.5 
Deathdealer Island. Representative specimen: Cham- 2. 3.0 x 2.3 


berlin s.n. (CAN). 


they do occur in these extreme conditions, plants often appear 
sickly with chlorotic and/or wilted leaves. In the Thousand Is- 
lands, populations are associated with a variety of tree and shrub 
species such as: Amelanchier Medic. sp., Carya ovata (Mill.) K. 
Koch, Pinus rigida Mill., P. strobus L., Prunus serotina Ehrh., 
Quercus alba L., Q. rubra L., Lonicera dioica L., Rubus strigosus 
Michx., Vaccinium angustifolium Ait., V. pallidum Ait., and Vi- 
burnum rafinesquianum Schultes. In many instances these species 
appear to be encroaching upon V. stamineum. The Niagara Falls 
populations are associated with Quercus L., Fraxinus L., and 
Crataegus L. spp. 


1995] Ford— Vaccinium stamineum 259 


A number of provincially rare plants occur near stations of V. 
stamineum in the Thousand Islands (Argus et al., 1982-1987). 
On Endymion Island, Pinus rigida Mill., Vulpia octoflora (Walt.) 
Rydb., and Solidago arguta Ait. are known to occur (Geomatics 
International Inc., 1992). Pinus rigida, Solidago arguta, and Sol- 
idago puberula Nutt. have been recorded from Grenadier Island 
(Geomatics International Inc., 1992). 

Populations grow on both granite-gneiss and limestone sub- 
strates; the former in the Thousand Islands area, the latter in the 
Niagara region. Soils are usually sandy with a low organic content. 
In the Thousand Islands, soils are acidic with a pH of 3.4-5.9 
with low levels of exchangeable cations such as calcium and mag- 
nesium. Organic carbon and soil moisture are also low. Leaf litter 
is present to a depth of 13 cm and soil depths range from 0-61 
cm (Crowder, 1982a, 1982b). Plants are found on both steep 
slopes (to ca. 45° and usually south-facing) and on flat ground, 
with all sites being well drained (Crowder, 1982a, 1982b; Ford, 
unpubl. data). The soil characteristics of the Niagara population 
are unknown. In the United States, deerberry grows in similar 
soil conditions but is also known to frequent moist thickets, low 
woods, and hammocks (Crowder, 1982a; Ward, 1974). 

Vaccinium stamineum is a species often associated with burnt 
sites (Crowder, 1982a, 1982b). This is certainly the case in the 
Thousand Islands area where deerberry is associated with a num- 
ber of fire-tolerant species such as Pinus rigida and Vaccinium 
angustifolium. The fire history of the Niagara stations is unknown. 


GENERAL BIOLOGY 


In Canada, most plants flower between the end of May and the 
end of June. Flowers are protandrous, the pollen being ready for 
dispersal a day or two before the stigma becomes receptive. In 
V. stamineum, the pendant nature of the flowers causes the pollen 
to be shed downwards, making autogamy unlikely. Fruit set re- 
quires floral visitation by insects, such as bees, that collect pollen 
by sonication of the anthers (“buzz-pollination”’) (Cane et al., 
1985). Over 30 species of bees have been recorded visiting deer- 
berry; however, most are either infrequent visitors or are nectar- 
seeking and thus insignificant pollen vectors (Cane et al., 1985). 
One species, Melitta americana Smith (Melittidae), however, has 
been found to be abundant on flowering deerberry bushes in cen- 


260 Rhodora [Vol. 97 


tral New York and may be the primary pollinator of deerberry 
throughout its range. Cane et al. (1985) observed female VM. amer- 
icana sonicating deerberry anthers while taking in nectar and 
carrying pure deerberry pollen loads, unlike another common 
visitor Xylocopa virginica (L.) (Anthophoridae). Crowder (1982b) 
found “‘adequate numbers of pollinators” at the Grenadier Island 
station but it is not known whether M. americana was the primary 
pollen vector at this site. 

Vegetative reproduction is well developed in blueberries with 
many species producing rhizomes. When disturbed, these rhi- 
zomes often sucker producing either clumps or colonies (Baker, 
1970). Later, the tissues connected with the parent plant break 
down leaving the branch as an established younger plant (Baker, 
1970; Crowder, 1982a), 


POPULATION SIZE AND TRENDS 


Like other blueberries, V. stamineum is known to spread veg- 
etatively making it difficult to determine how many individuals 
make up a population. Most populations are characterized by 
discrete clumps (Table 1). The greatest size and number of clumps 
are found on Grenadier and Endymion Islands, Leeds County 
(Table 1). Clumps at these sites may have arisen through the 
fragmentation of a single individual or may represent different 
genotypes. 

Most deerberry stations in Canada occur next to well-used trails 
and there is evidence that trail use is having a negative impact 
on some populations. For example, the Mewburn Road station 
occurs adjacent to the Bruce Trail and has been impacted by hikers 
as well as routine trail maintenance (Meyers, 1985, pers. comm.). 
Both the Grenadier and Endymion Island stations are found next 
to well-used paths. The fragmented nature of the deerberry pop- 
ulations on these islands may be the result of trampling. 

Deerberry appears to produce abundant seeds in the Thousand 
Islands area, although seedlings are not produced (Crowder, 
1982a). The frequent association of deerberry with known fire- 
tolerant species indicates that deerberry may require post-fire 
conditions to germinate. In the Thousand Islands area, conditions 
may have been more favorable for seed germination before the 
turn of the century when deforestation and fires were more fre- 
quent (Crowder, 1982b). 


1995] Ford— Vaccinium stamineum 261 


Climate may also be a factor affecting seedling growth. In a 
study of V. angustifolium, Vander Kloet (1976) discovered that 
seedling establishment is unlikely except under the following se- 
quence of events: 1) a cool, wet spring; 2) a wet August and 
September; and 3) a mild winter or winter with good snow cover. 
This sequence of events had not occurred in eastern Ontario in 
40 years. Similar climatic conditions may be required for seedling 
establishment in V. stamineum. 


SPECIAL SIGNIFICANCE OF THE SPECIES 


Only three members of the genus Vaccinium are used widely 
in the agricultural industry: Vaccinium macrocarpon Ait., V. cor- 
ymbosum L., and V. angustifolium. Vaccinium stamineum is usu- 
ally considered to be unpalatable, although certain shrubs can 
yield delicious fruit. In the southern Appalachians, deerberry is 
used for pies, jams, and jellies (Ballinger et al., 1981; Strausbaugh 
and Core, 1958; Stupka, 1964). Indeed, early authors state that 
deerberry has horticultural potential because of its large fruit, 
upland adaptation, and drought tolerance. The shrub is some- 
times cultivated as an ornamental (Crowder, 1982a). 


PROTECTION 


Although it is considered a threatened species in Canada, deer- 
berry has no legal protection (Ford, 1993). In the United States, 
deerberry is regarded as a “G5T5 species” by the Nature Con- 
servancy which means that it is “abundant and demonstrably 
secure.” The status in selected states is as follows: Illinois, SH 
(historical occurrence not having been verified since the turn of 
the century); Kansas, S1 (critically imperiled); Missouri, SX (ap- 
parently extirpated without expectation that it will be rediscov- 
ered) (Argus and Pryer, 1990). The status of deerberry in Mexico 
is unknown but it is locally common in pine forests and is perhaps 
under collected (Vander Kloet, pers. comm.). 


EVALUATION OF STATUS 


Vaccinium stamineum is found at five stations in Ontario. Pop- 
ulations in the Niagara Region are extremely small and threatened 
with imminent extirpation. In the Thousand Islands area, two 


262 Rhodora [Vol. 97 


key populations are found within St. Lawrence Islands National 
Park. Despite the protected status of this plant in the park, the 
proximity of populations to existing trails, lack of seedlings, and 
encroachment by later successional vegetation suggests that this 
species could decline if not actively managed. 


ACKNOWLEDGMENTS 


A number of people assisted in the preparation of the original 
COSEWIC report. Steve Varga and James Duncan, Ontario Min- 
istry of Natural Resources, provided information on the Niagara 
Falls populations. Jeff Thompson accompanied me in the field 
and helped track down relevant literature. The staff at St. Law- 
rence Islands National Park, particularly Larry Harbidge, Heather 
Davis, and Beth Chamberlin assisted with numerous aspects of 
this report including mapping of populations in the Thousand 
Islands as well as hosting the ““Deerberry Weekend’. Sam P. 
Vander Kloet provided information on the populations in Mex- 
ico, nomenclature, and reproductive biology. I am also thankful 
to William J. Crins, W. Donald Hudson, Jr., D. A. Ross McQueen, 
and an anonymous reviewer for their helpful comments on the 
manuscript. The financial assistance of the World Wildlife Fund, 
Canada 1s also gratefully acknowledged. 


LITERATURE CITED 


Arcus, G. W. AND K. M. Pryer. 1990. Rare Vascular Plants in Canada-Our 

Natural ne Canadian Museum of Nature, Ottawa. 

M. rR, D. J. WHITE AND C. J. Keppy, Eds. 1982-1987. Atlas of 
the Rare Vascular Plants of Ontario. Four parts. National Museum of Natural 
Sciences, Ottawa. (looseleaf). 

AsH, W. W. 1931. Polycodium. J. Elisha Mitchell Sci. Soc. 46: 196-213. 

Baker, P. C. 1970. A Systematic Study of the Genus Vaccinium L. subgenus 
Polycodium (Raf.) Sleumer, in the southeastern United States. Ph.D. thesis, 
University of North Carolina, Chapel Hill. 

BALLINGER, W. E., E. P. MANESS AND J. R. BALLINGTON. 1981. Anthocyanin 
and total eee content of Vaccinium stamineum L. fruit. Sci. Hort. (Am- 
sterdam) 15: 

Camp, W. H. oe The North American blueberries with notes on the other 
groups of the Vacciniaceae. aide 5: 203-275. 

CanE, J. H., G. C. Eickwort, F. R. WESLEY AND J. SPIELHOLz. 1985. Pollination 
ecology of Vaccinium stamineum nese Vaccinioideae). Amer. J. Bot. 
72: 135-142. 

Crowper, A. 1982a. Resource management study of selected rare plants in St. 


1995] Ford— Vaccinium stamineum 263 


Lawrence Islands National Park. 3. Rhus copallina: shining sumac; 4. Vac- 

cinium stamineum: apt 5. Viburnum dentatum: arrow-wood. Queen’s 

University, Kingston, Onta 

1982b. The status of ee rare plants in the Kingston area. Blue Bill 29: 
50-52. 

Forp, B. 1984. Deerberry (Vaccinium stamineum L.) in Ontario. The Pl. Press 
(Mississauga) 2: 40-42. 

1993. Status Report on Vaccinium stamineum L., a threatened species 
in Canada. Committee on the ais of Endangered Wildlife in Canada, 
Canadian Wildlife Service, Ottaw 

GEOMATICS INTERNATIONAL INC. 1992. Zone | Environmentally Sensitive Site 
udy St. Lawrence Islands National Park. Prepared for St. Lawrence Islands 
National Park by Geomatics International Inc. 

Meyers, G. 1985. Botanizing with George Meyers ... some native and exotic 
American plants in eee ae Region, Ontario. Wood Duck 38: 131-132. 
STRAUSBAUGH, P. D. AND E. L. Core. 1958. Flora of West Virginia. Part 3. Jn: 

West Virginia University Bulletin, Ser. 58, No. 12-3. Morgantown, W.V.: 
692-718 
Stupka, A. 1964. Trees, Shrubs and Woody Vines of the Great Smoky Moun- 
tains National Park. Knoxville, Tennessee. University of Tennessee Press. 
VANDER KLoeT, S. P. 1976. A comparison of the dispersal and seedling estab- 
lishment of Vaccinium angustifolium (the lowbush blueberry) in Leeds Coun- 
ty, Ontario and Pictou, Nova Scotia. Canad. Field-Naturalist 90: 176-180. 
988. The genus Vaccinium in North America. Publ. Dep. Agric. Canada 


1828. 
Warp, D. B. 1974. Contributions to the flora of Florida-6: Vaccinium (Erica- 
ceae). Castanea 39: 191-205. 


DEPARTMENT OF BOTANY 
UNIVERSITY OF MANITOBA 
WINNIPEG, MANITOBA, CANADA, R3T 2N2 


RHODORA, Vol. 97, No. 891, pp. 264-274, 1995 
ALLELOPATHIC EFFECTS OF LANTANA CAMARA 


(VERBENACEAE) ON MORNING GLORY 
(IPOMOEA TRICOLOR) 


CHRISTINA M. CASADO 


ABSTRACT 


Allelopathic ae of Lantana camara L. foliar leachates and dried leaf amend- 

nts on [pomoea tricolor Cav. radicle growth, shoot emergence, and plant bio- 
ees Were examined ¢ Ove 2 50- day period. Aqueous leaf extracts of L. camara 

tinhibited. 

Dried leaf residue i in soil growth media delayed shoot emergence con soil. Plant 
biomass after 50 days was not affected by the presence of L. camara soil amend- 
ments. Leaf extracts in petri dishes were more inhibitory than was dried leaf 
material in soil. These results indicate the presence of phytotoxic compounds in 
L. camara. Allelopathic effects of these compounds are significant during early 
germination of Jpomoea, while plants older than 2 weeks appear unaffected. In 
the soil environment allelopathic effects are minimal, possibly due to chemical 
binding, microbial action, or both. 


Key Words: allelopathy, exotic weeds, germination, /pomoea, Lantana, morning 
glory 


INTRODUCTION 


Lantana camara L. (Verbenaceae) is a shrub of West Indian 
or South American origin (Schemske, 1983) that is considere 
one of the world’s worst weeds (Holm et al., 1971). It grows in 
moist open soil and flowers throughout the year. Proliferation in 
Australia and the United States has been facilitated by its intro- 
duction as an ornamental. Along the edges of Australian rainfo- 
rests in Queensland, L. camara thickets persist for decades. I have 
observed these thickets encroach on native vegetation and restrict 
rainforest regeneration. In Florida, L. camara has been listed as 
a “Category I” exotic pest by the Florida Exotic Pest Plant Council 
since 1991. Category I designates those species invading or dis- 
rupting native plant communities in Florida. The list places L. 
camara among the state’s most invasive plants including Casu- 
arina, Melaleuca, and Pueraria (Exotic Pest Plant Council, 1992). 

Although sensitive to frost, L. camara occurs throughout much 
of Florida (Figure 1), and is reported to be a problem in pastures 
and in nature preserves (Gregg, 1994). Eradication efforts by 
mowing, herbicides, and burning cost millions of dollars annually 


264 


Lantana camara 


@ Indicates presence from known 
herbarium specimen 


Figure 1. Distribution of Lantana camara in Florida counties (adapted from 
Wunderlin et al., 1995). 


(Gregg, 1994). In citrus groves, L. camara has been observed to 
interfere with application of fertilizer, herbicides, and with har- 
vesting (Achhireddy and Singh, 1984). The weed is toxic and 
potentially lethal to livestock and children (Mortan, 1971). Its 
tendency to develop pure stands in diverse environments (Achhi- 
reddy and Singh, 1984) has led workers to study the basis for the 
competitive success of L. camara. 

Allelopathy, as defined by Rice (1984) is a harmful chemical 
effect by one species upon another. It is more specific than com- 
petition because it depends on the addition of a chemical com- 
pound to the environment by the inhibitory (allelopathic) species. 
Lantana camara has been shown to be allelopathic to Milkweed 
Vine (Morrenia odorata) in soil assays (Achhireddy and Singh, 
1984) and to Duckweed (Lemna spp.) and Ryegrass (Lolium) in 
the laboratory (Jain et al., 1989; Singh et al., 1989). Thirteen 
allelopathic compounds have been identified in leaves of L. ca- 
mara (Jain et al., 1989). 


266 Rhodora [Vol. 97 


In this study on the extent of allelopathic effects of L. camara, 
I used Heavenly Blue morning glory (Ipomoea tricolor Cav.) as 
the test species. Genetically uniform seed lots of this cultivar are 
commercially available, and it is closely related to J. purpurea 
(L.) Roth. naturalized (introduced but non-intrusive) throughout 
Florida and much of the eastern seaboard of the United States. 
In Florida, Ipomoea occurs with L. camara in cypress-pine re- 
growth forests and in similar subclimax forest communities. How- 
ever, it has not been observed using L. camara as a support, 
suggesting that L. camara might be inhibiting the growth of Ipom- 
oea nearby. In this study, the presence of compounds in L. camara 
which might be allelopathic against Ipomoea was tested by ger- 
minating /pomoea seeds in aqueous extracts of L. camara, and 
by growing seedlings of Jpomoea in potting soil to which dried 
leaves of Lantana had been added. A reduction of Ipomoea growth 
in Lantana-laced media (vs. the same media but without Lantana 
extract or dried leaves) would support the idea that allelochem- 
icals are at least partly responsible for the ability of Lantana to 
grow as pure stands, free of Jpomoea and other naturalized and 
native species in Florida. 


MATERIALS AND METHODS 


Petri Dish Assays 


Shoots of L. camara were collected in January 1994 from a 
disturbed site on 107 Ave, Homestead, Florida. Leaves were then 
air dried at 80°C. A 5% aqueous extract was made by steeping 5 
g of L. camara leaves in 100 ml deionized water at 25°C overnight, 
then filtering the solution through Whatman No. 1 paper. The 
extract (designated as 5%) was diluted with deionized water to 
make 2.5% and 1.25% solutions using methods similar to those 
of Achhireddy and Singh (1989). The pH of all extracts ranged 
between 6.7 and 7.0. 

Seeds of Ipomoea tricolor (Heavenly Blue morning glory) were 
purchased from Johnny’s Selected Seeds in Albion, Maine, and 
rinsed overnight in running tap water. Those which had swollen 
and showed emerging radicles were then used in petri dish assays 
and in soil assays (below). Ten swollen seeds were placed in each 
petri dish. Dishes contained filter paper wetted with 10 ml of 
aqueous L. camara leaf extract at 5%, 2.5%, 1.25%, or 0% (water 


1995] Casado—Allelopathy in Lantana 267 


aaa 


Radicle length (mm) 
ZA 
l 


Pas Tg 


Days 


Figure 2. Length of Ipomoea radicles on seeds sprouting in aqueous Lantana 
extract (0-5%) over the course of one week. Each point represents the average of 
30 radicles. In these petri dish assays, radicle growth was slowest in seeds soaked 
in the highest concentration (5%) of Lantana extract. Bars show +1 SE; bars are 
absent when +1 SE < height of symbol. 


control). They were kept at 25 + 2°C under a 1 2h-1 2h light-dark 
cycle. Three replicate assays were performed, requiring a total of 
three petri dishes and thirty seeds for each of the four leaf extract 
preparations. Germination rate and percentages by seeds imbib- 
ing Lantana extract were examined using the methods of Liebl 
and Worsham (1983). Germination (root emergence), and radicle 
length were monitored over the course of seven days. Radicle 
lengths of seedlings in each concentration of Lantana extract were 
averaged (Figure 2). 


Soil Assays 


To expose Jpomoea seedlings to Lantana residue within a soil 
environment, five concentrations of leaf-amended soil were made. 
Each contained 300 g of sterilized commercial potting soil, then 


268 Rhodora [Vol. 97 


12 g, 6 g, 3 g, 1.5 g, or O g (control) of dried and crushed L. 
camara leaves. Crushed leaf material was thoroughly mixed into 
the potting soil. The soils, now containing Lantana leaf residue, 
were watered and allowed to drain overnight before being planted 
with Ipomoea seeds. In each soil preparation, ten seeds (soaked 
in tap water, as above) were planted 2 cm deep and 2 cm apart. 
Three replicate assays were performed, requiring a total of three 
pots and thirty seeds for each amended soil preparation. Pots 
were kept in a greenhouse under 14h days with diurnal temper- 
atures fluctuating between 10-30°C. Shoot emergence rate was 
monitored over 14 days (Figure 3). Seedlings were left to grow in 
the greenhouse, and total fresh biomass of root and of shoot 
systems was measured after seven weeks (Figure 4). 

Analysis of variance (ANOVA) was used to test for significant 
differences between means, with a Scheffé posthoc multiple com- 
parison test to determine whether means of the dependent vari- 
able differed significantly at P levels from 0.05 to 0.0001 (Data 
Desk 4.0, Data Description Inc. Ithaca, NY). 


RESULTS 


Petri Dish Assays 


When exposed to L. camara extract, seeds of Ipomoea ger- 
minated at a rate similar to that of seeds in deionized water. 
Within 48 h of being placed on wetted filter paper, 90-100% of 
seeds germinated regardless of the concentration of L. camara 
extract used to wet the filter paper. After seven days, seedlings in 
all treatments showed distinct radicles. However, seedlings grow- 
ing on paper soaked with 5% L. camara extract suffered up to 
50% mortality apparently due to microbial activity fostered by 
nutrients in the extract. Seed putrifaction was common in 5% 
extract, but was never found in seeds soaked in water as a control. 
Only healthy seedlings with turgid white radicles were used in 
measuring radicle length. Thirty uncontaminated seedlings were 
measured from each treatment. 

Once germinated, Ipomoea seedlings exposed to high concen- 
trations of L. camara extract developed significantly (P < 0.001) 
shorter radicles than did control seedlings in deionized water 
(Figure 2). Though significant (P < 0.01) radicle inhibition oc- 
curred in seedlings growing in L. camara extract at 1.25% and at 


1995] Casado—Allelopathy in Lantana 269 
—)—— 0% —O—- _ 0.5% —O— 1% 


——e 2, ——a AG 


40 


Ww 
(=) 


Shoots emerged 
i) 
So 
1 


0 5 10 1S 


Days after planting 
Figure 3. Rate of Ipomoea seedling emergence from soils laced with dried 
Lantana leaf material (0-4% by weight). Thirty seeds were planted in each of the 
five concentrations of Lantana-laced soil. Each point represents the average of 
30 seedlings. A significant delay in seedling emergence occurred in Jpomoea grow- 
ing in soil amended with 4% Lantana leaf material. Bars show +1 SE; bars are 
absent when +1 SE < height of symbol. 


2.5%, the most pronounced inhibition of seedling growth was 
caused by 5% extract. In that solution, radicles seldom reached 
more than a centimeter, while over the same amount of time, 
radicles of control seedlings in deionized water grew to ten times 
that length. 


Soil Assays 


The rate of Ipomoea shoot emergence from soil was slowed in 
the presence of Lantana leaf material (Figure 3). The most rapid 


270 Rhodora [Vol. 97 


Total weight (g) 
of 30 seedlings 


0.5 1 


Soil treatment 
(% Lantana leaf material) 

Figure 4. Total root and shoot biomass (fresh weight) of Ipomoea plants grown 
50 days in soil laced with dried Lantana leaf material (0-4% by weight). Each bar 
represents the total fresh weight of roots or shoots from 30 plants. Error bars are 
+1 SE. 


emergence was seen in control pots containing only potting soil, 
not amended with Lantana. In more than 90% of cases, cotyle- 
dons emerged from control soil within 4—5 days of being planted. 
In soil amended with dried Lantana leaves the rate of seedling 
emergence slowed with increasing concentrations of leaf material. 
Compared to control sets which had completely emerged after 4— 
5 days, seedlings in soil laced with 4% leaf material required 
significantly longer time (P = 0.01) to emerge, typically twice as 
long as did controls. Regardless of the time required for cotyledons 
to emerge, the appearance and subsequent development (leaf mor- 
phology, internode length, tendril activity) of Ipomoea seedlings 
was similar in amended and in control soils 

Two months after growing in the ereenhouse, all 150 Ipomoea 
plants (10 plants for each of 5 Lantana leaf amendments to soil; 
3 replicates) were removed from their pots, washed clean of soil 


1995] Casado—Allelopathy in Lantana 271 


particles, dabbed clear of water droplets and divided into root 
and shoot components. Total fresh weights of shoots and roots 
from three sets of plants (10/set) growing in each concentration 
of L. camara leaf amendment were determined (Figure 4). Within 
any given pot containing a set of ten plants, there was variation 
in plant size. Shoot systems weighed nearly twice as much as did 
root systems, and plants with smaller shoots had smaller root 
systems. There was no significant (P < 0.05) variation in shoot 
and root total fresh weight among sets of plants grown in the same 
or in different concentrations of L. camara leaf amendment. Plants 
in heavily contaminated soil (amended with 4% L. camara leaf 
material) grew as well as did control plants in uncontaminated 
soil. There was even a qualitative improvement in plants growing 
in soil laced with L. camara. After two months of growth, /pom- 
oea plants in soil containing 0% and 0.5% Lantana leaf material 
were yellowing, while those growing in heavily laced (4%) soil 
were still green. 


DISCUSSION 


The potentially allelopathic effect of L. camara against I. tri- 
color is expressed as a delay in early (underground) seedling growth, 
while post-emergence seedling and plant growth seem unaffected. 
In this study, I found growth of J. tricolor to be inhibited (delayed 
and reduced) by L. camara leaves, primarily when Jpomoea seeds 
were forced to imbibe aqueous extracts of L. camara in a petri 
dish (Figure 2). The effect of Lantana residue in soil appears to 
be less severe. After an initial delay in seedling emergence (Figure 
3), Ipomoea plants growing in Lantana-laced soil show no ill 
effects, even after 50 days of growth (Figure 4). In fact, the greener 
appearance of plants in heavily laced soil is probably due to 
increased nutrients supplied by decaying Lantana leaf material, 
while plants growing in unamended potting soil develop nutrient 
deficiency symptoms such as chlorosis. 

Results of this study differ from those reported by Achhireddy 
and Singh (1984). They used similar methodology to amend soil 
with dried leaves of L. camara and found that biomass of another 
vine, Morrenia odorata (Asclepiadaceae), was reduced by 33% 
when plants were grown in soil containing 4% Lantana leaf ma- 
terial. Growth inhibition was apparent after 30 days, whereas in 


212 Rhodora [Vol. 97 


the present study using Jpomoea, no biomass difference between 
control plants and those growing in Lantana laced soil appeared 
even after 50 days (Figure 4). Recent work by Inderjit and Dak- 
shini (1994) suggests that leaf amendments in soils can change 
soil texture, and that test plants growing in amended soils could 
respond to textural as well as chemical changes caused by leaf 
amendments. The most significant inhibitory effects of Lantana 
material upon /pomoea take place in the earliest phase of seedling 
growth. Radicle growth is delayed in germinating seeds confined 
to petri dishes (Figure 2) as is initial shoot emergence from soil 
(Figure 3). Under field conditions, delayed early growth can be 
fatal. Slow-growing seedlings are vulnerable to soil pathogens and 
herbivores, and require more time for roots to penetrate to soil 
levels with reliable moisture, several inches beneath the hot sur- 
face. 

In situations where seeds are exposed to Lantana leachate in 
a confined environment, such as a petri dish, any inhibitory (al- 
lelopathic) compounds present in the growth medium will affect 
seedling growth. On the other hand, in open systems such as 
potted soil or disturbed sites populated by Lantana and Ipomoea 
in the field, potentially allelopathic material in L. camara may 
bind to organic molecules such as humic acid (Wang et al., 1971), 
soil colloids, or be broken down by bacteria or physical processes. 
This would decrease their potential to inhibit growth in the field 
(Rice, 1984). In a study of allelochemicals binding to organic 
matter in two Taiwanese agricultural soils (Wang et al., 1971) five 
phenolic acids, all of which are present in L. camara (Singh et 
al., 1989), were added to field soil. Between 60% to 80% of all 
added phenolics were bound to mudstone and latosol components 
of the soil. Ferulic acid, a phenolic in L. camara with strong 
inhibitory activity (Jain et al., 1989), was strongly bound by the 
soils, with the result that only 2—30% of applied ferulic acid re- 
mained free. 

For L. camara to be functionally allelopathic against Ipomoea 
in the field, it must release inhibitors that are not inactivated by 
soil components. Strong inhibition occurs in petri dish assays 
(Figure 2) but Lantana material has less effect when delivered to 
seedlings in a soil medium. Inhibition is reduced (Figure 3) and 
dwindles to insignificant levels over the course of 50 days (Figure 
4). It might still be possible for Lantana to allelopathically inhibit 
Ipomoea and other plants in the field but it would require constant 


1995] Casado—Allelopathy in Lantana 273 


replenishment of allelopathic compounds to soil through leaf drop 
or root exudates. 

Given the remarkable success of L. camara as a pantropical 
weed and its wide distribution in Florida (Figure 1), the com- 
petitive strategies of this pest are worth exploring. Possible mech- 
anisms leading to pure stands of Lantana include spatial and 
nutrient competition, coupled with the slight advantage of delay 
in germination of competing seedlings, as shown in this study 
with Ipomoea. Given the magnitude of damage and management 
costs caused by exotic pest plants (Gregg, 1994), understanding 
the means by which exotic weeds such as L. camara dominate 
local vegetation is needed to develop efficient and environmen- 
tally benign methods of exotic weed control (Heisey, 1996). That 
in turn will help sustain native flora in preserves, along with areas 
that have been repaired or restored (Kaufman and Franz, 1993). 


ACKNOWLEDGMENTS 


Natalie Courant and Allison Courant kindly provided essential 
support with computer graphics. Their work was supported by 
the Arabis Fund. Gustavo Casado is acknowledged for his help 
with collecting Lantana leaf material. I am also grateful to George 
Ellmore for discussions and to anonymous reviewers for helpful 
comments on the manuscript. 


LITERATURE CITED 


ACHHIREDDY, N. R. AND M. SinGu. 1984. Allelopathic potential of lantana 
(Lantana camara) on milkweedvine (Morrenia odorata). Weed Sci. 32: 757- 


761. 

Exotic Pest PLANT Councit. 1992. Exotic Pest Plant Council’s 1991 list of 
Florida’s most invasive species. Resource Managem. Notes (Florida Dept. 
of Natural Resources) 4: 39-41. 

Greco, M.E. 1994. Florida’s Exotic Pest Plant Control: 1994 Local-level Gov- 
ernment Survey and Report. eo ecuioage Endangered Lands Program, 
Dade County Planning Dept., 

HEISEY, R. M. ais Hentncaion pa llelopathi mpound from Ailanthus 

Itisst ti fits herbicidal activity. Amer. 


J. Bot. 83: 192-200. 

How, L. G., D. L. PLucknett, J. V. PANCHO AND J. P. HERBERGER. 1977. The 
World’s Worst Weeds: Distribution and Biology. Univ. Press of Hawaii, 
Honolulu. 

INDERJIT AND K. M. M. DaksHINI. 1994. Allelopathic effect of P/uchea lanceolata 


274 Rhodora [Vol. 97 
(Asteraceae) on characteristics of four soils and tomato and mustard growth. 
mer, ot. 81: -804. 


Jatn, R., M. SiInGH AND D. J. DEzMAN. 1989. Qualitative and quantitative 
characterization of phenolic compounds from lantana (Lantana camara) 
leaves. Weed Sci. 37: 302-307. 

KAUFMAN, D.G. AND C. M. FRANZ. 1993. Biosphere 2000: Protecting our Global 
Environment. Harper Collins, New York. 

Lies, R. AND A. D. WorsHAM. 1983. Inhibition of pitted morning glory (Jpom- 
oea lacunosa) and certain other weed species by saa components of 
wheat Cie aestivum) straw. J. Chem. Ecol. 3: 1027-1043. 

MortTANn, —— 1971. Plants Poisonous to People in Florida mee Other Warm 
Are SHueeaa ne House, Miami. 

Rice, E. L 1984. peas en Academic oe New Yor 

SCHEMSKE, D. W. 1983. Lantana camara n Costa Rican aie History. Univ. 
of ae Press, aoe 

Sincu, M., R. V. TAMMA AND H.N. Nico. 1989. HPLC identification of alle- 
fopathic compounds from Lantana camara. J. Chem. Ecol. 15: 81-89. 

Wana, T. S. C., K. L. Yen, 8S. Y. CHENG AND T. K. YANG. 1971. Behavior of 
soil phenolic acids. a United States National Committee for the Interna- 
tional Biological Program. Biochemical oe among Plants. National 
Academy of Sciences, Milan D.C:: —120. 

WUNDERLIN, R., B. F. HANSEN AND E. L. ee 1995. Atlas of Florida Vas- 
cular Plants: Florida Game and Freshwater Fish Commission, Tallahassee. 


DEPARTMENT OF BIOLOGY 
TUFTS UNIVERSITY 
MEDFORD, MA 02155 


RHODORA, Vol. 97, No. 891, pp. 275-279, 1995 


NEW ENGLAND NOTE 


STUDIES ON NEW ENGLAND ALGAE II: 
A SECOND STATION IN MAINE FOR 
NITELLA TENUISSIMA (DESV.) KUETZING 


L. C. Cott, Jr. 


Collections of algae from the shallow, rocky shores of Crawford 
Pond and the streams which drain into it in Union, Knox County, 
Maine have yielded some uncommon taxa. These and the area 
have been described previously (Colt, 1977, 1985, 1994a). This 
paper reports the second collection from Maine of Nitella ten- 
uissima (Desv.) Kuetzing 48 years after it was reported by William 
Randolph Taylor in 1921. 

On the south shore of the large central island in Crawford Pond 
are several large embayments characterized by waters mostly less 
than a meter in depth, and with very gently sloping subsurface 
cobble-like areas composed of small stones. Near the shore line 
the stones tend to be free of silt and debris as a result of small 
but constant waves formed by winds typically from the southwest. 
Silt accumulates as the water deepens away from the shoreline, 
becoming the primary substrate by filling the interstitial spaces 
between the stones and covering them. Nite//a plants were col- 
lected from silt at a depth of approximately 10 cm in the easterly 
portion of the major embayment. The Nite/la population was of 
low density, consisting of a few scattered clumps spread over 10- 
15 square meters. 

Nitella tenuissima is one of the smaller charophytes (Wood and 
Imahori, 1965), and the plants have a minute, delicate appear- 
ance. The plants from Crawford Pond, collected on August 9, 
1969, (partial upper portions) range from 2.2 to 3.1 centimeters 
in height, and have the distinctive beaded appearance which is 
one of the characteristics of this species. The “beads” are com- 
posed of repeated (3—4 times) furcations of branchlets at the nodes. 

The morphological characteristics of the Maine plants fit the 
descriptions given by Prescott (1962) and Wood and Imahori 
(1965), and are summarized in Table |. The collected plants (L. 
Colt CP8969-1,-2,-3,-4,-5,-6,-7) were not yet fully mature, judg- 
ing by the morphology of the gametangia, yet fit within the range 


vit Be) 


Table 1. 


Species Data Summary, Nitella tenuissima. 


Wood & Imahori 1965 


Prescott 1962 


Colt 


Axis, Diameter 
Internodes 
Number of fertile bran- 


Branchlets, comparative 


Basal cell 

End cell 
Gametangia 
Oogonia color 
Oogonia number 
Oogonia size 
Coronula 


Antheridia 


160-500 pu 
3/4—-5 times as long as branchlets 
ca. 6 in whorl 


upper usually more compact than 
low wer 


241 
ae i 1/2 length of branchlet 
5-7, | central 


3-4, 1 may divide again 
3-4 


2-celled 
3-4 


none 

cylindrical or tapering to base of 
end cell 

conical, acute, 42-105 uw long, 21- 

ide at base 

monoecious, sejoined or conjoined 
at 2nd—3rd branchlet nodes 

light or reddish brown 

solita 

370-350 u long incl. coronula, 
225-510 uw wide 

10 cells in 2 ae res uw high, 

40-58 uw 
90-175 uw in eee ree 


— 


NR* 


6 in whorl 


glomerules formed, compact 


3-4 times 
R 


plant monoecious 

NR 

NR 

400 u long, 260 u wide 
10 cells in 2 tiers 


175 w in diameter, stipitate 


180 uw below Ist branchlets 
2-3 times as long as branchlets 
5-6 usual, occasionally 9-11 


gee formed, upper branchlets 
age than lower 

ive 4 1/3-1/2 length of branchlet 

5-7, | central 

5-6, | central 

a see dactyls 


2. vcelled 
3-4 


no 
nate: with slight taper to base of 
end 
conical, acute, 62-107 u long, 17-25 u 
wide at base 
monoecious, sejoined or conjoined at 
nd- a rd branchlet nodes 
light wn 
most cae solitary, few paired 
320-340 wu long, including coronula, 
vi wide 
10 cells in 2 tiers, 9-45 » high 


106-195 yw in diameter, stipitate 


* NR indicates that this information was not reported by Prescott (1962). 


1995] New England Note Zit 


of measurements provided by both Wood and Imahori and Pres- 
cott. 

On a worldwide basis, Nitella tenuissima is reported from Eu- 
rope, North Africa, Madagascar, the Azores, India, Japan, and 
from North America where it is known to range from southern 
Canada to the West Indies (Wood and Imahori, 1965). 

Wood and Imahori (1965) state that Nitella tenuissima is com- 
mon throughout New England, although a search of the literature 
(Colt, 1994b) indicates that it has not often been collected or 
reported from the region. The species has been collected once in 
Maine from Echo Lake on Mt. Desert Island (Taylor, 1921) and 
once from Lake Chocorua in New Hampshire (Collins, in Wood 
and Imahori, 1965). More collections of Nitella tenuissima have 
been reported from Massachusetts than any other New England 
state. Faxon, Morong (as Nitella gracilis Smith, and as Nitella 
tenussima, T. Morong 32, 37), and Perkins are reported by Dame 
and Collins (1888) to have independently collected Nitella ten- 
uissima from several ponds in Middlesex County, while Wood 
and Imahori (1965) report collections from Essex County by Col- 
lins (Icon 308) and by Robinson (as Nitella transilis Allen). Tindall 
and Sawa (1964) collected Nitella tenuissima in Morse Pond, 
Barnstable County, and Wood (in Wood and Imahori, 1965, R. 
D. Wood 2015) also collected it in Barnstable County. In Rhode 
Island Wood (R. D. Wood 1081), reports collecting this species 
from Larkins Pond in Washington County. It has also been col- 
lected by Faxon from Apponaug Pond and J. L. Bailey from 
Gorton Pond, both in Providence County, and by Wood and 
Palmatier from Newport County (Wood and Imahori, 1965). 
Robbins, and later Allen collected Nitella tenuissima in Rhode 
Island, but neither location is given by any of the authors (Halsted, 
1878; Bennett, 1888; Wood and Imahori, 1965) listing the col- 
lections. 

Whereas the Crawford Pond collection does not substantiate 
the comment by Wood that the plant is common throughout New 
England, it lends credence to the suggestion that Nitella tenuissima 
might be more widely distributed than has been reported here- 
tofore. The more than 1600 articles known to report the collection 
of fresh water algae in New England (Colt, 1994b) suggest that 
such collections have tended to be primarily from scattered lo- 
cations. Records of algal collections from Maine, New Hamp- 
shire, Vermont, Massachusetts and Rhode Island can only be 


278 Rhodora [Vol. 97 


described as geographically spotty and primarily a function of the 
interests and activities of collectors since the first published report 
of algae in New England by Hitchcock in 1829. The only system- 
atic state-wide effort was in Connecticut, first by Conn (1905), 
Conn and Webster (1908), and then later by Hylander (1922a, 
1922b, 1924, 1925, 1928). Neither Conn and Webster (1908) nor 
Hylander (1928) list any of the Characeae among the algae of 
Connecticut. Hylander notes the exclusion by suggesting that the 
Characeae belong in a separate group among the Thallophytes 
because of their “complicated and advanced types of reproductive 
structures.” Furthermore, although Nite/la tenuissima was occa- 
sionally assigned to Nitella gracilis or Nitella transilis by early 
workers (Wood and Imahori, 1965), neither of these species have 
been reported from Connecticut. 

It is likely that because of the small size of Nitella tenuissima 
plants and its growth habit, “in silt with only the tips of the 
branches emergent,” (Prescott, 1962), it would tend to escape 
notice by most collectors unless they were actively searching for 
it. Judging by the few plants at the site in Crawford Pond, Nitella 
tenuissima is probably relatively scarce even in suitable habitats. 
Then too, the Characeae have not enjoyed a great deal of attention 
among New England phycologists over the years, and many plants 
have, in all likelihood, been by-passed during searches for other 
algae. 


LITERATURE CITED 
BENNETT, J. L. 1888. Plants of Rhode Island. Proc. Frankl. Soc., Providence, 
RI 


Conn, H. W. 1905. A Preliminary Report on the Protozoa of the Fresh Waters 
of Seas Connecticut State Geol. Surv., Bull. #2. 

W. Wesster. 1908. A Preliminary Report Upon the Algae of the 
Fresh Waites of Connecticut. Connecticut State Geol. Surv., Bull. #10. 
Cott, L. C., Jk. 1977. A new station for Coleochaetaceae in New England. 

Rhoden 79: 300-304. 
. 1985. Vaucheria undulata Jao again in New England. Rhodora 87: 597— 


1994a. Desmonema wrangellii (Ag.) Bornet et Flahault, a new record 

for Maine. Rhodora 96: 104-108. 

. 1994b. The New England Algal Data Bank. Library, Marine Biological 
Laboratory, Woods Hole, MA. 

Dame, L. L. AND F. S. CoLiins. 1888. Thallophytes. Flora of Middlesex County. 
Middlesex Institute, Malden, MA. 


1995] New England Note 279 


Hatstep, B. D. 1878. Classification and description of the American species of 
Characeae. Proc. Boston Soc. Nat. Hist. 20: 169-190 

Hircucock, E. 1829. Catalogue of Plants Growing without Cultivation in the 
Vicinity of Amherst College. J. S. & C. Adams, Amherst, Mass. 

HyLANnper, C. J. 1922a. A preliminary report on the desmids of Connecticut. 
Rhodora 24: 213-224, 

. 1922b. A preliminary report on the desmids of Connecticut. Rhodora 

24: 236-241. 

. 1924. Supplementary report on the desmids of Connecticut. Rhodora 

26: 203-210. 

1925. The Algae of Connecticut. Ph.D. dissertation, Yale Univ., New 

Haven, CT. 

1928. The Algae of Connecticut. Connecticut State Geol. Surv., Bull. 


2. 
Prescott, G. W. 1962. Algae of the Western Great Lakes Area. Wm. C. Brown 
Co., Dubuque, Iowa. 
TayLor, W.R. 1921. Additions to the flora of Mt. Desert Island, Maine. Rhodora 


TInDALL, D. R. AND T. Sawa. 1964. Characeae occurring in the Woods Hole 
region. Am. J. Bot. 51(9): 943-949. 

Woop, R. D. AND K. IMAHoRI. 1965. A Revision of the Characeae. J. Cramer, 
Weinheim, Germany. 


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


RHODORA, Vol. 97, No. 891, pp. 280-282, 1995 
NEW ENGLAND NOTE 


NEW BARNSTABLE COUNTY RECORDS 


MARIO DiGREGORIO 


During the field season in 1994, I documented two new records 
for Barnstable County along with one significant rediscovery: 


Bidens laevis (L.) BSP. (Asteraceae) 


Many hundreds of plants of Bidens laevis (L.) BSP. were found 
in the upper Mashpee River within the freshwater tidal zone just 
south of Route 28 in Mashpee, Massachusetts. Large or showy 
bur-marigold is a perennial characterized by 8-10 golden yellow 
ray flowers arranged around a flat disk, many measuring up to 3 
cm long. The sharply serrate, sessile leaves are lanceolate and 
reach up to 15 cm. in length. In fruit the mature disk is 2-3 cm. 
broad and often nods; achenes are two to four-awned. Typical 
habitat for B. /aevis is sluggish streams, either fresh or brackish, 
along the coastal plain of the eastern seaboard from Florida north 
to southern New Hampshire (Fernald, 1950; Gleason and Cron- 
quist, 1991). Records have been documented for Plymouth Coun- 
ty to the north and Bristol County to the west (Seymour, 1982). 
A check of the G. M. Gray Herbarium (SPWH) in Woods Hole 
and a literature search revealed no previous records documented 
for Barnstable County (Seymour, 1982; Svenson and Pyle, 1979). 


Rumex pallidus Bigel. (Polygonaceae) 


Six plants were discovered on August 26, 1994, in a barrier 
beach “washover’ area just east of Crosby Landing Beach in East 
Brewster, Massachusetts. Characterized by glaucous, narrowly 
lanceolate leaves, prostrate or depressed habit and whitish fruit 
arranged within a dense, spreading panicle, this taxon is typically 
found in the upper beach zone above normal high tide but within 
the stormtide washover area where beach profiles can change 
dramatically from season to season. Associated taxa include Glaux 
maritima L., Lathyrus maritimus (L.) Bigelow, Xanthium stru- 
marium L., Solidago sempervirens L. and Mertensia maritima 
(L.) S. F. Gray (see next notation). 


280 


1995] New England Note 281 


Pale or seabeach dock ranges from Newfoundland south to 
Nantucket (Fernald, 1950; Seymour, 1982); historically it was 
found as far south as Long Island (Seymour, 1982). Its present 
status in Massachusetts is classified as Threatened by the Mas- 
sachusetts Natural Heritage and Endangered Species Program. 
This represents the first record ever for Barnstable County (P. 
Somers, MNHESP pers. comm.; Svenson and Pyle, 1979). A 
specimen was placed in the herbarium collection at the Garden 
in the Woods in Framingham; 35 mm. slides and an Element 
Occurrence form were filed with MNHESP. 


Mertensia maritima (L.) S.F. Gray (Boraginaceae) 


Due to its sporadic appearances and disappearances, seabeach 
lungwort or oysterplant has had a long history as a “phantom” 
or “fugitive” plant in Massachusetts (Bicknell, 1915; Svenson and 
Pyle, 1979). Cape Cod and Nantucket mark the southern limit 
of its range, which stretches north from Massachusetts to all the 
coastal counties of Maine (where it is fairly common) and to James 
Bay and Greenland (Fernald, 1950). It is classified as Endangered 
by — oer Natural Heritage and Endangered Species 
progr 

se 1983 to 1988, Dr. Peter Dunwiddie of Nantucket noted 
the plant on the Coatue peninsula and Whale Island, a sandspit 
currently connected to Tuckernuck Island. These plants disap- 
peared and no sign of the species was again noted until 1994, 
when Dr. Dunwiddie found 7 plants scattered along the eastern 
shore of the island (Dunwiddie, 1994, pers. comm.). 

In Barnstable County, the plant was noted in the early 1970's 
in West Brewster by Don Schall but disappeared in 1974 (Schall, 
pers. comm.) and was not seen again on the Cape until 1992 when 
Kyle Jones of the Cape Cod National Seashore found populations 
in three lower Cape localities: 21 non-flowering plants on Jeremey 
Point in Wellfleet, one plant at Marconi Beach, and 104 non- 
flowering and three flowering plants at Race Point in Province- 
town (Jones, pers. comm.; MNHESP data). It was seen again in 
1993 and 1994, but by then all three populations were noted as 
being in ‘severe decline’ (Jones, pers. comm.). 

The East Brewster population of twelve plants contained ten 
flowering and fruiting “rosettes”; two other non-flowering plants 
were depauperate and stressed due to their location in closer 


282 Rhodora [Vol. 97 


proximity to a heavily used bathing beach. The plant is quite 
distinctive, with sea-green, glaucous spatulate foliage which ac- 
tually tastes of raw shellfish. The flowering branchlets radiate out 
from the central basal rosette in a spider-like configuration. As 
in most members of Boraginaceae, the small bluish-pink flowers 
are campanulate with fruits forming smooth, lustrous nutlets. The 
successful germination of their seeds amidst the “flotsam” of 
storm wrack detritus seems integral to the plant’s habit of mys- 
terious appearance and disappearance along the southern limit of 
its range. Vegetative reproduction from fragments may also occur 
but has not been documented (Dunwiddie, pers. comm.). 

A follow up visit to the dramatically-changed winter beach in 
December 1994 revealed only one rosette partially buried in the 
overwash, all others having vanished out to sea or been buried 
deep beneath the newly deposited sand. By the late summer of 
1995, the colony had diminished to five rosettes, with many des- 
iccated flowering branches showing the stress of an unusually 
warm and dry growing season. 


LITERATURE CITED 


BICKNELL, E. P. 1915. Ferns and flowering plants of Nantucket XIV. Bull. Torrey 
Bot. Club 42: 27-47, 

Dunwippig, P.W. 1994 October 25. [Letter to Massachusetts Audubon Society]. 

FERNALD, M. L. 1950. Gray’s Manual of Botany, 8th ed. American Book Co., 

ork. 

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

Seymour, F.C. 1982. The Flora of New England, 2nd ed. Phytologia Memoirs, 
Plainfield, NJ. 

Svenson, H. K. AND R. W. Pyte. 1979. The Flora of Cape Cod. Cape Cod 
Museum of Natural History, Brewster, Ma. 


107 GOELETTA DR. 
HATCHVILLE, MA 02536 


RHODORA, Vol. 97, No. 891, pp. 283-284, 1995 
BOOK REVIEW 


Harris, James G. and Melinda W. Harris. 1994. Plant Identi- 
fication Terminology: An Illustrated Glossary. 197 pp. Spring 
Lake Publishing, Spring Lake, UT. ($17.95). 


Students of all ages find comfort in illustrations. Uncertainties 
about the meaning of technical terms often pose major impedi- 
ments to correctly identifying a plant. Many plant enthusiasts 
resist graduation from the field guide stage to the technical manual 
because of a fear of terminology. Statements such as “There are 
at least 60 ways to say that a plant is not smooth,” from the 
Peterson and McKenny Field Guide (1968) exacerbate the situ- 
ation, not to mention do no service to botanical taxonomy. Even 
the thorough glossaries in most manuals don’t seem to lessen the 
apprehensions of the neophyte. Merritt Lyndon Fernald’s 8th 
Edition of Gray’s Manual of Botany (1950) includes 1141 entries 
in the glossary (Peterson and McKenny, 1968). Plant Identification 
Terminology by James G. Harris and Melinda W. Harris is an 
illustrated glossary of terms used to identify plants that should 
help assuage such fears. In fact, at least one copy of Plant Iden- 
tification Terminology aught to be in every classroom where flo- 
ristics, plant taxonomy, or plant identification courses are taught. 

Benjamin Daydon Jackson’s comprehensive but difficult to use 
classic, A Glossary of Botanic Terms, was published in 1900. The 
fourth edition of 1928 has been reprinted and costs about twice 
the price of the Harris’ book. Jackson’s Glossary is hard to find 
and can usually be obtained only from dealers who handle more 
esoteric works in botany. Plant Identification Terminology will 
not replace Jackson’s Glossary nor should it. It is, however, ac- 
cessible, available, and reasonably priced. Its cost is not prohib- 
itive, even for a student as a supplemental text for a plant iden- 
tification course. Most importantly, it is copiously illustrated! 
That, alone, is worth the $17.95. 

The book is divided into 2 parts, a general glossary (115 pages) 
and a section for specific terminology (67 pages) which is orga- 
nized into 7 major categories (Roots, Stems, Leaves, Surfaces, 
Inflorescences, Flowers, and Fruits), some with further subdivi- 
sions (there are 10 subheadings under Flowers). 

The most important feature of this book is its illustrations. 
Many of the definitions are supplemented with simple line draw- 


283 


284 Rhodora [Vol. 97 


ings that more than suffice to make the point. Most of the illus- 
trations are incorporated with the text in the general glossary. 
Some, however, are duplicated in the specialized terminology 
sections. These groups of illustrations comparing variations in 
leaf shape, inflorescence types, and fruits enhance the usefulness 
of the book. 

Keys to help decide the term that best describes a condition 
such as leaf margin or inflorescence type are included in the Spe- 
cific Terminology section. These are boxed in the text so that they 
clearly stand out. They will be especially useful to students who 
are familiarizing themselves with botanical terminology. 

very minor objection is that some illustrations used to define 
a term are used for more than one term. The same illustration 
occurs for climbing (figure 185), scandent (figures 927, 1363), and 
vine (figures 1253, 1324). 

The proof of the versatility and value of Plant Identification 
Terminology will come when students use this book to aid them 
in identifying plants and the identification works. I am looking 
forward to trying this book with classes. 


LITERATURE CITED 


FERNALD, M. L. 1950. Gray’s Manual of Botany, Eighth Revised Edition. Van 
Nostrand Reinhold Company, New York. 

Jackson, B.D. 1928(1971). A Glossary of Botanic Terms With Their Derivation 
and Accent. Hafner Publishing Company, New Yor 

PETERSON, R. T. AND M. McKenny. 1968. A Field Guide to Wildflowers of 
Northeastern and North-central North America. Houghton Mifflin Com- 
pany, Boston. 


LESLIE J. MEHRHOFF 

CONNECTICUT GEOLOGICAL AND NATURAL HISTORY SURVEY and 
G. SAFFORD TORREY HERBARIUM 

UNIVERSITY OF CONNECTICUT 

STORRS, CT 06269-3042 


RHODORA, Vol. 97, No. 891, pp. 285-288, 1995 


RHODORA NEWS & NOTES 


LisA A. STANDLEY 


HIGHLIGHTS OF CLUB MEETINGS 


May 1995 (909th Meeting). Dr. Walter Judd, of the University 
of Florida at Gainesville, spoke on “Angiosperm Temperate/ 
Tropical Family Pairs.’’ His research, based on cladistic analyses 
of representative genera and using conventional morphological 
characters, has tested the hypothesis that temperate families are 
derived from tropical families. He sped through a series of ex- 
amples of temperate/tropical family pairs. Contrast of the tem- 
perate/tropical family pairs Apocynaceae/Asclepiaceae, Cappar- 
aceae/Brassicaceae, Moraceae/Urticaceae, Sapindaceae/Acera- 
ceae shows that the temperate genera are simply the terminal 
members of a continuous clade, and that all of the characters that 
delimit the temperate family appear in the tropical group. Anal- 
ysis of the Araliaceae/Apiaceae and Verbenaceae/Lamiaceae shows 
that these temperate families are polyphyletic, since each evolved 
from at least two distinct clades within the tropical family. 

Dr. Judd concluded by summarizing that all tropical families 
appear to be paraphyletic (all members of the family share a 
common ancestor, but the family does not include all descendents 
of that ancestor). Temperate families comprise one or more ad- 
vanced lineages within the larger clade of the tropical family. The 
greater number of species within temperate families is the result 
of rapid recent radiation, perhaps due to the shift from woody to 
herbaceous habit. His conclusion is that there are no temperate 
plant families, based on the philosophy that a family should be 
monophyletic (all members share a common ancestor, and the 
family contains all descendents of that ancestor). This conclusion 
is also supported by more traditional arguments which show that 
there are no clear phenetic (morphological) gaps between tem- 
perate families and members of their tropical ancestral families. 
Recent investigations using chloroplast DNA, and the fossil pol- 
len record support this conclusion, although he cautioned that 
these results are preliminary, and based on a relatively small 
sample of genera. 


June 1995 (910th Meeting). Dr. Ken Kimball, Director of the 
AMC’s Research Department, spoke on “Recovery of an Alpine 


285 


286 Rhodora [Vol. 97 


Plant— Potentilla robbinsiana.”’ This Federally-listed Endangered 
species is endemic to New Hampshire’s Presidential Range of the 
White Mountains. The successful implementation of a recovery 
program appears to have halted the decline of this plant, the only 
New England endemic species. It occurs primarily in Mt. Wash- 
ington’s Monroe Flats area, a windswept area of gravelly soils 
that is snow-free most of the winter and subject to blown abrasive 
ice and severe freeze-thaw cycles in the soil. However, plants also 
occur on a vertical cliff at another location. Like most arctic/ 
alpine species, seed set is highly weather-dependent and rare, but 
seedlings do occur in nature. Chromosomal and isoenzyme data 
are limited, but suggest that there 1s virtually no genetic variation 
due to the apomictic (pseudogamous) reproduction. There are no 
adaptations to biotic or abiotic seed dispersal. The species appears 
to be derived from Potentilla hyparctica, a circumpolar species. 

Potentilla robbinsiana has been studied at Monroe Flats for 
twenty years. Total counts of the population have been conducted 
at ten-year intervals, and show an increase of about 60% since 
the mid-1970’s. Individual plants are mapped twice a year along 
permanent transects, providing a detailed account of about 10% 
of the population. Natural mortality is primarily due to winter 
conditions, particularly severe freeze-thaw cycles that lift plants 
out of the ground. 

The recovery plan for this species has focused on eliminating 
human-caused mortality, primarily due to trampling by hikers 
and over-collection by botanists. This recovery has been achieved 
by rerouting the Crawford Path around the Flats, and by closing 
this area to hikers. Reintroduction and establishment of new pop- 
ulations by transplanting garden-grown seedlings, another feature 
of the recovery plan, has been successfully implemented. The sum 
of these efforts has been an increased understanding of the species, 
its demography, and recovery of the Mt. Washington population. 


June 1995 (Field Trip). The first 1995 Field Trip was held on 
June 17th. Twenty-three Club members and guests visited the 
Alpine Gardens on Mt. Washington, guided by George Newman. 
Despite overcast and severely windy weather, participants en- 
joyed seeing Rhododendron lapponicum, Diapensia lapponica, 
Loiseleuria procumbens, Houstonia caerulea var. Faxonorum, and 
Scirpus caespitosus in full bloom. Other alpines (Cassiope hyp- 
noides, Phyllodoce caerulea, Geum peckii, Viburnum edule, Vac- 


1995] Rhodora News & Notes 287 


cinium uliginosum, Arenaria groenlandica, Ledum groenlandi- 
cum, Carex bigelowii) were primarily observed in bud. Notable 
finds also included Silene acaulis, Lycopodium selago (Huperzia 
selago in Flora of North America (FNA)), L. sitchense (Diphas- 
lastrum sitchense in FNA), Arctostaphylos alpina and Salix uva- 
TAY 


July 1995 (Field Trip). On July 22, 1995, Club members visited 
Snake Pond and East Mountain in Westfield, Massachusetts under 
the leadership of Dr. David Lovejoy of Westfield State College. 
The excursion was intended to explore unusual habitats in central 
Massachusetts and to contribute to the Club’s herbarium, in which 
Hampden County is underrepresented. Participants renewed their 
acquaintance with Toxicodendron vernix while exploring the bog 
surrounding Snake Pond. The bog also yielded Drosera rotun- 
difolia, D. intermedia, Ilex glabra, Bartonia virginica, Habenaria 
psycodes, H. clavellata, and a larch/black spruce thicket. Rho- 
dodendron viscosum at the bog margins appeared to be f. glaucum. 
Adjacent dry woods contained a large population of /sotria ver- 
ticillata, with several fruiting plants. Along the talus slopes and 
dry ridgetop of East Mountain, trip participants observed Seri- 
cocarpus asteroides, Hieracium paniculatum, Prunus pumila, He- 
lianthus divaricatus, Lonicera dioica, Asplenium platyneuron, and 
Hystrix patula. 


OBITUARIES 


Last August Dorothy Waleka of Ipswich, Massachusetts passed 
away at the age of 63. Although she was never a member of the 
Club, her handiwork will silently serve the Club as long as our 
organization exists. Ms. Waleka served as the mounter for the 
Club’s herbarium from the 1960’s into the mid-1980’s, producing 
mounted specimens of the highest professional quality. As many 
as half of the folders presently in the vascular herbarium are 
labeled in her beautiful script. Ms. Waleka had a very warm and 
generous personality, and continued to be interested in news about 
the herbarium and Harvard University Herbarium staff after she 
discontinued her services due to health problems. Although her 
name does not appear on any of our herbarium sheets, her pres- 
ence is there in as much measure as those collectors whose names 
do appear. Contributed by Ray Angelo, Curator of Vascular Plants. 


288 Rhodora [Vol. 97 
NOTES 


Invasive Alien Plants: Unlike Florida or Hawai, New England 
is not generally considered to be severely affected by invasive 
alien plants. However, 29 percent of the New England flora is 
composed of alien species—and they’re increasing. In Berkshire 
County, the proportion of aliens has grown from 17% of the flora 
in 1922 to 24% in 1990. As the number of exotic plants increases, 
they may compete for space with native species. While the total 
number of introduced plants growing without cultivation in New 
England exceeds 200 species, about 50 of these are considered 
highly aggressive and likely to invade natural areas. These include 
Daucus carota, Coronilla varia, Rosa multiflora, Elaeagnus an- 
gustifolia, Cytisus scoparius, Celastrus orbiculatus, Lonicera ja- 
ponica, Ampelopsis brevipedunculata, Berberis thunbergii, Eu- 
onymus alatus, Taxus cuspidata, Rhamnus frangula, R. cathar- 
tica, and Lythrum salicaria. (Excerpted from an article by Faith 
Thompson Campbell, “Invasive Alien Plants,’ in the May 1995 
issue of Columbine, the Conservation Newsletter of the National 
Council of State Garden Clubs). Contributed by Mary Walker. 


THE NEW ENGLAND BOTANICAL CLUB 
22 Divinity Avenue 
Cambridge, MA 02138 


The New England Botanical Club is a non-profit organiza- 
tion that promotes the study of plants of North America, es- 
pecially 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, RHO- 
DORA, 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 subscrip- 
tion 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 a 


Name 


Address 


City & State Zip 


Special interests (optional): 


THE NEW ENGLAND BOTANICAL CLUB 
Elected Officers and Council Members for 1995—1996 


President: C. Barre Hellquist, Box 9145, Department of Biology, 
North Adams State College, North Adams, Massachusetts 
01247 

Vice-President (and Program Chair): W. Donald Hudson, Jr., 
Chewonki Foundation, RR 2, Box 1200, Wiscasset, Maine 
04578 

Corresponding Secretary: Nancy M. Eyster-Smith, Department 
of Natural Sciences, Bentley College, Waltham, Massachu- 
setts 02154 

Treasurer: Harold G. Brotzman, Box 9092, Department of Bi- 
ology, North Adams State College, North Adams, Massa- 
chusetts 01247 

Recording Secretary: Lisa A. Standley 

Curator of Vascular Plants: Raymond Angelo 

Assistant Curator of Vascular Plants: Pamela Wetherbee 

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) 
Thomas Mione °96 
Garrett E. Crow °97 
Edward J. Hehre ’98 
Donald J. Padgett (Graduate Student 
Member) 95 


Volume 97, No. 890 including pages 109-184 was issued February 7, 1996 


Free On-Line Document Search... 


Discounton Document Delivery 


Announcing 
Access to UnCover 


Document search and delivery for readers of 
UnCover Xhodora 


You can now access free of charge the UnCover on-line data 
base, which indexes table of contents information from over 
17,000 journals. You can also order delivery of full text from 
any of the articles in the data base and receive a $2.00 
discount per article off the regular document delivery price. 


With Access to UnCover, you can search for articles by 
topic, by author, or by journal. You can then order on-line 
the full text of articles for delivery by fax within 24 hours. 
Document delivery fees will be billed to your credit card. 


Access to UnCover is available to you now over the Internet. 

Simply telnet to the Access to UnCover Address: 
allen.carl.org 

When prompted for your Password, enter: deliver 


New benefit: Discount on Reveal 
Your Access to UnCover Password now also entitles you to 
a 20% discount on UnCover's new Reveal service. With 
Reveal, you receive by e-mail the tables of contents of 
journals you select—right after each issue is published. In 
addition, you can also receive weekly topical alerting by 
creating keyword search strategies. To sign up for Reveal at 
your 20% discount, call 800-787-7979 and mention your 
Access to UnCover Password. 


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), in addition, a copy of the ms. on a 3.5 inch 
floppy disc will facilitate production of the paper. The text must be 
double-spaced (at least 3%”) throughout, including tables, figure leg- 
ends, and literature citations. The list of legends for figures and maps 
should be provided on a separate page. Do not use footnotes. Use a 
non-proportional font throughout and do not justify the right margin. 
Do not indicate the style of type through the use of capitals or un- 
derscoring, particularly in the citation of specimens. Names of genera 
and species may be underlined to indicate italics in discussions. Spec- 
imen 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 syn- 
onyms. 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 is distinguished from its congeners. Papers of a 
floristic nature should follow, as far as possible, the format of “‘An- 
notated List of the Ferns and Fern Allies of Arkansas,” W. Carl 
Taylor and Delzie Demaree, Rhodora 81: 503-548, 1979. For biblio- 
graphic citations, refer to the Botanico-Periodicum-Huntianum (B-P- 
H, 1968) and B-P-H/Supplement, 1991, which provides standardized 
abbreviations for journals originating before 1990. All abbreviations 
in the text should be followed by a period, except those for standard 
units of measure and direction (compass points). For standard abbre- 
viations and for guidance in other matters of biological writing style, 
consult the CBE Style Manual, Sth 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 3% X 6 in.) Magnifi- 
cation/reduction values given in text or figure legends should be cal- 
culated to reflect the actual printed size. Tables should be set up in 
the same way. Each page should be ready for reproduction and will 
be reduced to 4 X 6 inches. 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. 


ae 

Wit 

ese 

he 
i ae 3 or s 


, 


Berg hy saan peeve 


ae 


RHODORA 


JOURNAL OF THE 
NEW ENGLAND BOTANICAL CLUB 


CONTENTS: 


— 


Contributions to the flora of Vermont. Jerry Jenkins and Peter F. Zika... 29 

Occurrence of the red alga Thorea violacea (Batrachospermales: Thorea- 
ceae) in the Hudson River, New York State. Curt M. Pueschel, P. 
Gary Sullivan, and John E. Titus 328 

Distribution aul conservation of Nantucket shadbush, Amelanchier nan- 
tucketensis (Rosaceae). Alison C. Dibble and Christopher S. 
Campbe of 339 

Eriogonum codium (Polygonaceae: Eriogonoideae), a new species from 
southcentral Washington. James L. Reveal, Florence Caplow, and 
Kathryn Beck 


350 
Vegetation, browsing, and site factors as determinants of Canada yew (Tax- 
us canadensis) distribution in central New Hampshire. John J. Sta- 
chowicz and Taber D. Allison 357 
CEaiene nivalis L. ene eee new to Québec. N. Dignard, R. Lalu- 
miére, and M. Julie 375 
Book Review: 
A guide to wildflowers in winter: Herbaceous plants of northeastern 
North America. Pamela B. Weatherbee 380 
Rhodora News and Notes. Lisa A. Standley 382 
NEBC Officers and Council Members 386 
Index to Volume 97 387 
NEBC Membership Form 393 
Instructions to Contributors 394 
Statement of Ownership inside back cover 


THE NEW ENGLAND BOTANICAL CLUB 
P. O. Box 1897, Lawrence, Kansas 66044 
22 Divinity Avenue, Cambridge, Massachusetts 02138 


Vol. 97 Fall, 1995 No. 892 


The New England Botanical Club, Inc. 


22 Divinity Avenue, Cambridge, Massachusetts 02138 


RHODORA 


DAVID S. CONANT, Acting Editor-in-Chief 
SUZANNE GALLAGHER, Acting Managing Editor 


Associate Editors 
DAVID S. BARRINGTON 
W. DONALD HUDSON, JR. 
LESLIE J. MEHRHOFF 
THOMAS MIONE 
CATHY A. PARIS 
LISA A. STANDLEY 


RHODORA (ISSN 0035-4902). Published four times a year (January, 
April, July, and October) by The New England Botanical Club, 810 
East 10th St., Lawrence, KS 66044 and printed by Allen Press, Inc., 
1041 New Hampshire St., Lawrence, KS 66044. Periodicals postage 
paid at Lawrence, KS. POSTMASTER: Send address changes to 
RHODORA, P.O. Box 1897, Lawrence, KS 66044. 

RHODORA is a journal of botany devoted primarily to North America. 
Authors are encouraged to submit manuscripts of scientific papers and 
notes relating to the systematics, distribution, ecology, paleobotany, 
or conservation biology of this or floristically 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, PO. Box 1897, Lawrence, KS 
66044-8897. 


MEMBERSHIPS: Regular $35; Family $45. Application form printed 
herein. 


BACK VOLUMES AND SINGLE COPIES: Many available; informa- 
tion and prices will be furnished upon request to Cathy A. Paris, Dept. 
of Botany, Univ. of Vermont, Burlington, VT 05405. 


ADDRESS CHANGES: In order to receive the next number of RHO- 
DORA, changes must be received by the business office prior to the 
first day of January, April, July, or October. 


INFORMATION FOR CONTRIBUTORS: Inside back cover, January 
and April. 


WORLD WIDE WEB ADDRESS: a a herbaria. harvard.edu/nebc/ 
MANUSCRIPTS: Send to: Janet R. Sulliv 


Durham, NH 03824-3597 


This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). 


RHODORA 


JOURNAL OF THE 
NEW ENGLAND BOTANICAL CLUB 


Vol. 97 Fall 1995 No. 892 


RHODORA, Vol. 97, No. 892, pp. 291-327, 1995 


CONTRIBUTIONS TO THE FLORA OF VERMONT 


JERRY JENKINS AND PETER F. ZIKA 


ABSTRACT 


Recent herbarium and field studies have added 38 species and deleted 192 
species from the flora of Vermont. The additions include 25 recently discovered 
species and 13 species based on older records that have been overlooked or 
misinterpreted. Ten additions are rare native species. Four of these are currently 
listed as state Threatened or Endangered species. The additions bring the total 
flora of Vermont to approximately 1933 species. The 192 deleted species, if 
included in the flora, would increase this number by 10%. The deletions include 
misidentified and misinterpreted species, cultivated plants incorrectly listed as 
naturalized, and many species that lack vouchers altogether. The geography of 
the added and deleted records suggests that many of the added species may be 
recent immigrants, and that many of the unvouchered species are highly unlikely 
in Vermont and are probably mistakes. A review of the number of species that 
have been deleted from the seven previous Vermont floras suggests that the ap- 
parent error rates of these floras have ranged from 0.3% to 5.4% of the species 
listed by each flora, with an average error rate of 2.4%. 


Key Words: flora, additions, deletions, endangered species, discovery rate, error 
e, Immigration, Vermont 


INTRODUCTION 


In the last 150 years Vermont has had seven annotated floras 
and one checklist (Oakes, 1842; Torrey, 1853; Perkins, 1888; 
Brainerd et al., 1900; Eggleston et al., 1915; Dole, 1937; Seymour, 
1969; Atwood et al., 1973), giving it the most extensively de- 
scribed flora of any state in the country. The authors are currently 
preparing a new checklist and flora. In the process we have com- 
piled a number of recent unpublished additions to the flora and 
deleted a number of species credited to the flora in earlier works. 


291 


292 Rhodora [Vol. 97 


This paper lists those additions and deletions and, in the case of 
the deletions, gives our reasons for excluding them. The nomen- 
clature follows that in Gleason and Cronquist (1991), except for 
a few cases in which we prefer the treatment of Kartesz (1994). 
Deleted species are listed under their current names!. When the 
names used in the original publications are different, these are 
placed in brackets. Common names for the species being added 
to the flora are taken from our forthcoming Checklist of the Flora 
of Vermont’. 

We used the standard acronyms from /ndex Herbariorum 
(Holmgren et al., 1990) when citing herbaria in the section that 
follows. The herbarium of the Vermont Institute of Natural Sci- 
ences (VINS) is not listed in Holmgren et al. (1990). 


ADDITIONS TO THE FLORA 


This section includes 38 taxa, 15 native to the northeastern 
U.S. and 23 aliens. Nineteen of these additions are recently dis- 
covered species that have never been listed for Vermont. Six are 
recent records for species that were listed previously, but for which 
the earlier records were unacceptable for various reasons, most 
commonly because they were unvouchered or based on misiden- 
tifications. Four are older records that either have never been 
cited or have been cited incorrectly. Six have been cited correctly 
in previous Vermont floras, but omitted from recent monographs 
or from the most recent Vermont flora (Seymour, 1969) and 
checklist (Atwood et al., 1973). And three are new identifications 
of specimens that were previously misidentified. 

This list does not include taxa (e.g., Spiranthes ochroleuca (Rydb. 


' In most cases the contemporary equivalents of the historical names are well 
known or can be found in historical manuals. Thus, synonyms for many of the 
names used by Oakes (1842) and Torrey (1853) can be found in the Fourth Edition 
of Gray’s Manual (Gray, 1857), and those for many of the names used by Perkins 
(1888) and Eggleston and Brainerd (1904) in the Seventh Edition of Gray’s Manual 
(Robinson and Fernald, 1908). In a few cases the historical names are obscure or 
were used incorrectly by 19th century authors, and it is hard to know exactly what 
plant the authors thought they had or what contemporary species most closely 
fits their concept. In such cases (for example Thaspium trifoliatum Gray var. 
apterum Gray) we have indicated that the synonym is uncertain 

> The Checklist of the Flora of Vermont (Jenkins and Zika, in prep.) has been 
in preparation for ten years and is expected to be published in the near future. 


1995] Jenkins and Zika—Flora of Vermont 293 


ex Britt.) Rydb., or Hudsonia ericoides L. spp. intermedia (Peck) 
Nickerson & Skog) that have been added to the flora because of 
taxonomic and nomenclatural changes. 

In addition to the species listed in this paper, at least 51 other 
recent additions to the state flora have been published since the 
last checklist (Atwood et al., 1973). These may be found in Coun- 
tryman (1978), Hellquist and Crow (1980), Wagner and Wagner 
(1982), Hellquist and Hilton (1983), Zika et al. (1983), Zika and 
Thompson (1986), Zika (1986B, 1987, 1988, 1990, 1991A, 1991B, 
1992), Angelo (1989, 1990), Rothrock (1991), Zika and Marshall 
(1991), Paris (1991), Haufler and Windham (1991), Beitel and 
Mickel (1992), Farrar (1992), Reznicek and Oldham (1993), and 
Gilman (1993 

Ten of the taxa added here are rare native species with extant 
populations (Asplenium montanum Willd., Carex atlantica L. 
Bailey var. atlantica, C. capillaris L., C. cumulata (L. Bailey) 
Mack., Dracocephalum parviflorum Nutt., Eleocharis robbinsii 
Oakes, Myrica pensylvanica Mirbel., Panicum flexile (Gattinger) 
Scribn., Potentilla pensylvanica L., and Vaccinium stamineum L.). 
Four of these (Asplenium montanum, Carex capillaris, Draco- 
cephalum parviflorum, and Panicum flexile) are currently listed 
as state Threatened or Endangered species (10 V.S.A., Chapter 
123). We have recommended to the Vermont Endangered Species 
Committee that all of these species, with the exception of Myrica 
pensylvanica, be listed as state Threatened or Endangered species. 


Recently Discovered or Authenticated Species 


Ajuga reptans L., Lamiaceae, carpet-bugle. Windsor Co.: road- 
side, South Woodstock, 1967 (Ahles 685243 MASS); Orleans 
Co.: under raspberries, Lake Willoughby, Westmore, 1980 
(Zika 1354 MASS, VT) 

Alyssum saxatile L., Brassicaceae, golden-tuft. Chittenden Co.: 
escape from cultivation, Shelburne, 1979 (Zika 1291, VT); 
Addison Co.: naturalized and common on limy ledges, Gar- 
diners Island, Ferrisburg, 1980 (Zika 1639 MASS, VT); 1981 
(Zika 4048 NEBC). 

Asplenium montanum Willd., Aspleniaceae, mountain spleen- 
wort. Bennington Co.: about 30 plants in cracks, dry sunny 
quartzite ledges, cliffs above Rattlesnake Brook, elev. 1000 
feet, Pownal, 1974 (Jenkins s.n. VT). 


294 Rhodora [Vol. 97 


Cardamine impatiens L., Brassicaceae, European bittercress. Ben- 
nington Co.: old railroad grade south of golf course, Ben- 
nington, 1979 (Zika & Jenkins 1332 VT); edge of a footpath, 
Bennington, 1985 (Jenkins & Thompson s.n. VT). Known 
since 1981 from two localities in adjacent White Creek, 
Washington Co., New York (Zika & Jenkins 8024 VT). 

Carex atlantica L. Bailey var. atlantica, Cyperaceae, Atlantic sedge. 
Franklin Co.: Franklin Bog, Franklin, 1965 (Seymour s.n. 
VT); deep water of lag and central pools, with Carex comosa 
F. Boott and C. interior L. Bailey, several places, S. end of 
Franklin Bog, Franklin, 1982 (Jenkins 82-176 VT); ten 
clumps, N. end of Franklin Bog, Franklin, 1982 (Zika 6115 
VT); Bennington Co.: sedge-alder swamp, with Carex stricta 
Lam. and C. stipata Muhl., near a sandy field, N. Pownal, 
1982 (Jenkins 82-155 VT). Determinations based on keys in 
Reznicek and Ball (1980). 

Carex capillaris L., Cyperaceae, hair-like sedge. Caledonia Co.: 
limy seepage area near Route 2, with Equisetum variegatum 
Schleich., and Eleocharis tenuis (Willd.) Schultes var. borealis 
(Svenson) Gleason, Danville, 1985 (Zika 9040 VT). Discov- 
ered here by T. Rawinski in 1984. Reported from northern 
Vermont by Fernald (1950), but vouchers never located. 

Carex cumulata (L. Bailey) Mack., Cyperaceae, sedge. Windham 
Co.: sandy bluff, eroding bank of Connecticut River, with 
Leptoloma cognatum (Schultes) Chase, Paspalum setaceum 
Michx. var. ciliatifolium (Michx.) Vasey, Cyperus filiculmis 
Vahl, and weeds, Rockingham, 1989 (Thompson & Rawinski 
2307 VT, Det. A. A. Reznicek, 1990); among granite boulders 
in steep oak woods, with Pinus rigida Miller, Black Mountain, 
Dummerston, 1991 (Jenkins s.n. VT). 

Carex spicata Huds., Cyperaceae, sedge. Rutland Co.: rough limy 
meadow west of the south end of Shaw Mt., West Haven, 
1990 (Thompson 90-52 VT); limy meadow by Catfish Cove, 
Mt. Independence, Orwell, 1990 (Jenkins s.n. VT). The two 
colonies are within 10 miles of each other and both consist 
of a few clumps of the sedge within a few square meters. 

Eleocharis robbinsii Oakes, Cyperaceae, Robbin’s spike-rush. 
Rutland Co.: locally abundant, shallow water adjacent to bog- 
gy shore, with Nuphar, Utricularia gibba L. and Potamogeton 
epihydrus Raf., east shore of Little Lake, Wells, 1989 (Jenkins 
sn V 1). 


1995] Jenkins and Zika— Flora of Vermont 295 


Erucastrum gallicum (Willd.) O. E. Schulz, Brassicaceae, dog 
mustard. Orleans Co.: lower slope of Mt. Pisgah, Westmore, 
1973 (Hodgdon et al. 19983 NHA). Vouchers for stations 
reported by Dole (1937) have not been found. 

Hieraceum flagellare Willd., Asteraceae, whiplash hawkweed. 

indham Co.: roadside, junction of Routes 103 and Inter- 
ea Rockingham, 1979 (Ahles 86816 MASS, VT); Wind- 
sor Co.: roadside, Route 103, Chester, 1979 (Ahles 86783 
MASS) 

Tris sibirica L., lridaceae, Siberian iris. Windham Co.: roadside, 
Rockingham, 1979 (Ahles 86823 MASS); Chittenden Co.: 
naturalized in a damp meadow, Shelburne, 1981 (Zika 4013 
ees sandy lot by airport, South Burlington, 1985 (Zika 9062 


wis amabilis Graebn., Caprifoliaceae, beauty-bush. Chit- 
tenden Co.: apparently coridine from cultivation into thick- 
ets, Rock Point, Burlington, 1981 (Zika 4706 VT). 

Lavandula angustifolia Mill. (L. officinalis Chaix., L. spica L., L. 
vera DC.], Lamiaceae, lavender. Windham Co.: escaped in 
field, Guilford, 1982 (Zika 6547 VT). Earlier reports of L. 
spica (Atwood et al., 1973) are based on collections from 
cultivated plants. 

Lysimachia vulgaris L., Primulaceae, garden loosestrife. Ben- 
nington Co.: roadsides, Pownal, 1967 (Ahles 67959 MASS); 
Windsor Co.: naturalized, Connecticut River shore, near 
mouth of Lulls Brook, Hartland, 1982 (Zika 6220 VT), 
Washington Co.: roadside weed, Plainfield, 1980 (Zika 1654 
VT); Windham Co.: common on Connecticut River shore, 
two miles south of Bellows Falls, Westminster, 1982 (Zika 
6269 VT). 

Mentha xvillosa Hudson [M. alopecuroides Hull], Lamiaceae, 
foxtail mint. Washington Co.: railroad yard, Montpelier, 1975 
(Ahles 81496 MASS); Windham Co.: weed, sandy dump, 
Brattleboro, 1982 (Zika 6461 NEBC, VT). 

Myosotis micrantha Pallas [M. stricta Link], Boraginaceae, blue 
scorpion-grass. Chittenden Co.: lawn weed, North Beach 
Campground and Lakeview Cemetery, Burlington, 1980 (Zika 
1330 & 1370 VT); weed, Horsford Nursery, Charlotte, 1984 
(Zika 7988 VT). An earlier record for Burlington (Seymour 
23094 MO), cited in Seymour (1982), is based on a specimen 
of M. arvensis (L.) Hill. 


296 Rhodora [Vol. 97 


Myrica pensylvanica Mirbel., Myricaceae, bayberry. Chittenden 
0.: five clones, ca. 120 ramets, overgrown hillside pasture 

north of Rt. 2, with Acer rubrum L., Berberis, Crataegus, 
Pinus strobus L., Rubus, Richmond, 1992 (Marshall 2446 
VT). First observed in 1989 by Everett J. Marshall. 

Panicum flexile (Gattinger) Scribn., Poaceae, panic grass. Rutland 
Co.: common, sunny limestone ledges, Shaw Mountain, West 
Haven, 1983 (Jenkins & Zika 7665 NEBC, VT). Previous 
reports for Grand Isle are based on a sheet of P. capillare L. 
(Hunnewell 13908 NEBC). Panicum flexile has long been 
known from Skene Mountain, in Whitehall, N. Y., about 10 
miles south of Shaw Mountain, where a small population still 
persisted in 1987 (Jenkins and Zaremba, pers. obs.). 

Pimpinella saxifraga L., Apiaceae, burnet saxifrage. Bennington 
Co.: along abandoned railroad grade, Bennington, 1975 (Jen- 
kins s.n. WT). First discovered at this site by Barbara Beecher 
in 1975. Apparently not extant in August 1985. 

Polygonum sachalinense F. W. Schmidt, Polygonaceae, giant 
knotweed. Addison Co.: Granville, 1983 (Zika 7189 VT), 
Lamoille Co.: roadsides of Route 108 near Mt. Mansfield, 
Stowe, 1980 (Zika 2224, 4752 MASS, NEBC, VT); Stowe, 
1981 (Angelo et al., s.n. NEBC); Wolcott, 1983 (Zika 7486A 
VT); Washington Co.: railroad yard, Roxbury, 1982 (Zika 
6760 VT); Windsor Co.: railroad yard, White River Junction, 
1981 (Zika 5336 NEBC, VT). 

Potentilla pensylvanica L., Rosaceae, Pennsylvania cinquefoil. 
Washington Co.: hilltop pasture, Plainfield, 1968 (Jenkins 
s.n. VT). Colony extant but very small in September 1987. 

Setaria faberi R. Herrm., Poaceae, giant foxtail. Addison Co.: 
New Haven Junction, 1981 (Zika 5258 VT); Middlebury, 
1981 (Zika 5263 VT), Ferrisburg, 1981 (Zika 5292 VT); 
Bennington Co.: Bennington, 1984 (Zika 8263 VT); Chit- 
tenden Co.: Burlington, 1980 (Zika 1913 VT); Jericho, 1981 
(Zika 5185 VT); Franklin Co.: St. Albans, 1981 (Zika 5177 
VT); Rutland Co.: Rutland, 1981 (Zika 5270 VT); Wash- 
ington Co.: Waterbury, 1979 (Zika 1228 VT); Windham Co.: 
Brattleboro, 1981 (Zika 5339 VT); Windsor Co.: White River 
Junction, 1981 (Zika 5132 VT). This species apparently has 
increased rapidly in New York since 1940 (Smith, 1965) and 
has been spreading rapidly in Vermont since about 1970 (W. 
D. Countryman, pers. comm.). At present it is a widespread 
pest on railroad easements, agricultural lands, and roadsides, 


1995] Jenkins and Zika— Flora of Vermont 297 


where it is frequently associated with Setaria glauca (L.) P. 
Beauv., S. viridis (L.) P. Beauv., Aristida spp., Panicum di- 
chotomiflorum Michx., or Kochia scoparia (L.) Schrader. 

Vaccinium stamineum L., Ericaceae, deerberry. Bennington Co.: 
one bush in rocky woods above River Road, North Pownal, 
1986 (Jenkins s.n. VT). Brainerd et al. (1900) noted in the 
appendix to their flora that V. stamineum was known from 
Mt. Greylock in Williamstown, Massachusetts, and should 
be sought in Vermont. A record from Wells River, Vermont, 
in Dole (1937) was not supported by a herbarium specimen. 

Verbena stricta Vent., Verbenaceae, hoary vervain. Chittenden 
Co.: weed in cracks in asphalt, Troy Ave., Colchester, 1981 
(Zika 4234 VT). 


Older Records, Not Previously Cited 


Asclepias verticillata L., Asclepiadaceae, whorled milkweed. Wind- 
sor Co.: Windsor, undated (4A/phonso Wood s.n. NY). Al- 
though the Windsor collection lacks a date and was not cited 
in any previous Vermont flora, we accept the record because 
the specimen is correctly identified and geographically plau- 
sible. A collection by C. C. Frost from Brattleboro cited in 
early floras has not been found. 

Dioscorea batatas Decne., Dioscoreaceae, Chinese yam. Wind- 
ham Co.: Townshend, 1922 (L. A. Wheeler s.n. NEBC). Mar- 
ginal addition, probably short-lived, as Wheeler’s label reads: 
‘“‘temporary escape.” 

Hieracium murorum L., Asteraceae, golden lungwort. Rutland 
Co.: Center Rutland, 1947 (Kirk s.n. TUFT). 

Rhododendron periclymenoides (Michx.) Shinners [Rhododen- 
dron nudiflorum (L.) Torr.], Ericaceae, pinkster-flower. Cal- 
edonia Co.: Peacham, 1891 (F. Blanchard s.n. NY). All re- 
cords of this species cited in previous floras are based on 
specimens of R. prinophyllum (Small) Millais [R. roseum 
(Loisel.) Rehd.]. 


Older Records, Cited by Previous Floras, But Omitted From Re- 
cent Floras or Monographs 


Carex atlantica L. Bailey var. capillacea (L. Bailey) Cronq. [C. 
howei Mackenzie], Cyperaceae, Howe’s sedge. Chittenden Co.: 
Colchester, 1899 and 1907 (Flynn s.n. VT); S. Burlington, 


298 Rhodora [Vol. 97 


1896 (Flynn s.n. VT). Not cited for Vermont by Reznicek 
and Ball (1980). 

Carex wiegandii Mackenzie, Cyperaceae, Wiegand’s sedge. Wind- 
ham Co.: Torrey Meadow, Stratton, 1895 (Grout s.n. VT); 
Chittenden Co.: Star Farm, Burlington, 1906 (Flynn s.n. VT). 
Listed in Seymour (1969) and Atwood et al. (1973), but not 
cited for Vermont by Reznicek and Ball (1980). 

Dracocephalum parviflorum Nutt., Lamiaceae, American drag- 
onhead. Windsor Co.: potato field, Billings Farm, Wood- 
stock, 1921 (Kittredge s.n. VINS); Rutland Co.: West Clar- 
endon, 1915 (Potter s.n. TUFT); dry rocky slope in full sun, 
Twin Mountain, West Rutland, 1983 (Jenkins, pers. obs.). 
Cited by Kittredge (1931) and Dole (1937), but not in sub- 
sequent works. 

Geum vernum (Raf.) T. & G., Rosaceae, spring avens. “Vermont”: 
undated collection (ex herb. Torrey VT); Bennington Co.: 
fertile ground among wrecked cars, 100 m. north of the North 
Bennington post office, with G. canadense Jacq., Chenopo- 
dium gigantospermum Aellen and weeds, Bennington, 1985 
(Zika 9087 VT). Torrey’s sheet, presumed to have been made 
in the 1800’s, was reported in Seymour (1967), but not in 
previous floras or Atwood et al. (1973). The North Ben- 
nington colony was discovered by J. Jenkins and had both 
flowers and fruits on 5 June 1985. The species is not otherwise 
known from New England, but occurs occasionally in eastern 
New York (Jenkins, pers. obs.). 

Primula veris L., Primulaceae, primrose. Chittenden Co.: garden 
escape, Burlington, 1898 (Jones s.n. VT). Cited in Dole (1937), 
but omitted from recent floras. 

Tagetes patula L., Asteraceae, French marigold. Windsor Co.: 
riverbank, Billings Farm, Woodstock, 1918 (Kittredge s.n. 
VINS). Cited by Dole (1937), but not in subsequent works. 


Plants Incorrectly Determined in Previous Floras 


Delphinium orientale J. Gay [Consolida orientalis (J. Gay) Schré- 
dinger], Ranunculaceae, rocket larkspur. Chittenden Co.: 
roadside escape, Charlotte, 1903 (Flynn s.n. VT). Previously 
reported as Delphinium ajacis L. 

Prunus cerasus L., Rosaceae, sour cherry. Bennington Co.: Man- 
chester, 1898 (Day 371 NEBC). The specimen was labelled 


1995] Jenkins and Zika—Flora of Vermont 299 


as P. avium L. and was cited under this name in previous 
floras. Harry Ahles annotated it to P. cerasus in 1975. A 
Brainerd specimen of P. cerasus from Middlebury, cited in 
Dole (1937), has not been found. 

Solanum sarrachoides Sendtner, Solanaceae, hairy nightshade. 
Chittenden Co.: garden weed, Westford, 1969 (Meunier s.n. 
VT). Previously reported as S. villosum Mill. (Seymour, 1969). 


DELETIONS FROM THE FLORA 


This section lists 192 species, credited to Vermont in publi- 
cations or represented by herbarium specimens, that we have 
either proven to be erroneously credited to the state or have been 
unable to validate. These we are deleting from the flora. In doing 
sO we are not asserting that the species has never occurred in 
Vermont, but only that the historical reports are apparently un- 
verifiable or in error. The deletions do not include taxa (like 
Antennaria brainerdi Greene or Aster pringlei (Gray) Britton) that 
were formerly listed for Vermont but are now included in other 
species. 

Species have been deleted for any of six reasons?. The com- 
monest reason 1s that no specimen was found (NSF). For 130 of 
the deleted species we were unable to locate vouchers to support 
one or more of the published records. Ninety-six of these species 
lacked vouchers altogether and another 34 lacked vouchers for 
one or more of the reported stations. 

Interestingly, only 19 of the species that lack vouchers for one 
or more sites are cited in the floras whose authors required vouch- 
ers for all the species they listed (Brainerd et al., 1900; Eggleston 
et al., 1915; Seymour, 1969; Atwood et al., 1973); in these cases 


3 The claim that a species is or was present in Vermont is usually based on one 
or more published records, which may or may not cite individual specimens. We 
delete a species only after we have either looked for and found no corroborating 
specimens or ae examined and rejected all ane specimens we have found. The 
reasons for re] ’t be found; others are misidentified; 
others are rejected ! because they were cultivated, incorrectly cited, or otherwise 
dubious (see main text). Because the decision to reject a species usually rests on 
the rejection of several specimens or citations, a species may be deleted for a 
combination of reasons. Thus we say, for example, that some species lacked 
vouchers altogether, while others lacked vouchers for one or more stations. In the 
latter cases, the stations with vouchers would have been deleted for other reasons. 


300 Rhodora [Vol. 97 


it is likely that the original specimens cited in these floras have 
since been lost or redetermined. The remaining 111 species which 
lack one or more vouchers were cited in works (Oakes, 1842; 
Torrey, 1853; Perkins, 1988: Dole, 1937) that included unvouch- 
ered records. In many of these cases we suspect that no specimen 
was ever seen by the authors of the floras. 

The second most common reason for deleting species is that 
they are misidentified (MI). For 50 deleted species one or more 
vouchers were misidentified. These include 28 species for which 
all the published records were misidentifications and another 22 
species for which all the vouchers we could locate were misiden- 
tified. 

A third group of deletions are cultivated species that are not 
provably naturalized. To list a cultivated species in the flora, we 
require evidence that it has spread (though not necessarily per- 
sisted) beyond the immediate area where it was cultivated. Many 
collectors made specimens of cultivated plants; and, in the ab- 
sence of information about where a specimen was collected, the 
mere presence of a specimen of a cultivar in a herbarium does 
not prove that the plant was naturalized. A total of 31 deleted 
species fall in this category. Twenty of these are based on speci- 
mens that are almost certainly from cultivated plants (CULT), 
and another 11 are based on specimens whose labels lack habitat 
information and so are not definitely naturalized (NDN). 

A fourth group of deletions includes 15 records based wholly 
or partly on specimens that were correctly determined by their 
collectors, but incorrectly cited or attributed to Vermont in sub- 
sequent floras (IC). Examples include a specimen of Erigeron acris 
L. from the St. Johns River in Maine which was relabeled and 
credited to the St. Johnsbury Railroad in Vermont and, less dra- 
matically, a number of cases where the old and new names for a 
plant were listed in the same work as two separate species. When 
one name 1s obsolete, this causes no real problems. But in cases 
(e.g., Cardamine flexuosa With., Amelanchier canadensis (L.) Me- 
dikus, or Carex saxatilis L.) where one of the names is now used 
for a species that doesn’t occur in Vermont, this practice can lead 
to significant errors, particularly when these records are cited in 
regional floristic works. 

A fifth group of deletions includes five species for which some 
or all of the specimens are inadequate (IS) because they are im- 
mature, sterile, fragmentary, or otherwise undeterminable. These 
are mostly species that closely resemble a common Vermont spe- 


1995] Jenkins and Zika—Flora of Vermont 301 


cies. Examples are Lepidium ruderale L., sterile plants of which 
resemble the common L. densiflorum Schrader, and Salix serris- 
sima (L. H. Bailey) Fern., whose immature leaves closely resemble 
those of the common Salix /ucida Muhl. 

The last group of deletions includes five species in which one 
or more specimens are correctly determined, but not provably 
from Vermont, that is, the collection data is not convincing. We 
call these specimens of doubtful provenance (DP). Our reasons 
for reyecting them—which are inferential and not conclusive—are 
different in each case and are summarized in the entries for those 
species. 

Each of the following entries gives a) the species deleted; b) the 
publications (if any) that cited the record; c) in bold type, the 
locality to which the plant was credited, or NL if no locality was 
given, or an abundance in quotation marks if that is the only 
distributional information in the original publication; d) the col- 
lector of the specimen on which the record is based (or NCC when 
no collector was cited in the publication) and the herbarium hold- 
ing that specimen (both in parentheses); and e) our reason for 
deleting the record. When the current name differs from that used 
in the original publication, we give the older name in square 
brackets. When no herbarium is listed, it means that we found 
no specimen and are listing the locality and collector credited in 
the original publication. For readability, we use the following 
abbreviations. 

Publications: A, Atwood et al. (1973); AM, Ames (1910); B, 
Brown (1964); BR, Brainerd et al. (1900); C, Countryman (1978): 
CR, Crow (1982); D, Dole (1937): E, Eggleston et al. (1915); EB, 
Eggleston and Brainerd (1904); F, Fernald (1950); FL, Flynn 
(1911); G, Gleason and Cronquist (1991): I, Cronquist et al. (1977); 
J, James (1823); JE, Jesup (1891); K, Kennedy (1904); KI, Kit- 
tredge (1931); KS, Kittredge (1939); LI, Little (1977); O, Oakes 
(1842); P, Perkins (1888); PE, Pennell (1935); S, Seymour (1969); 
T, Torrey (1853); W, Watson and Coulter (1890). 

Reasons for deletions: CULT, cultivated; DP, doubtful prov- 
enance; IC, incorrectly cited; IS, inadequate specimen; MI, mis- 
identified; NDN, not definitely naturalized; NSF, no specimen 
found. 


Allium porrum L. [A. fistulosum sensu Dole (1937), not L.], A, S, 
Manchester, (Day NEBC, GH), MI = A. cepa L. or A. fis- 
tulosum L. 


302 Rhodora [Vol. 97 


Amaranthus cannabinus (L.) Sauer [Acnida cannabina L.], D, 
Woodstock, (Kittredge), MI = Amaranthus rudis Sauer ?; NL 
(ex herb. Torrey, VT), MI = Amaranthus tuberculatus (Moq.) 
Sauer [Acnida altissima (Riddell) Moq. ex Standl.]. 

Amaranthus powellii 8. Wats., D, Woodstock, (Kittredge), NSF. 

Amelanchier canadensis (L.) Medikus, A, S, 14 sites, (several col- 
lectors, VT), IC = A. arborea (Michx. f.) Fern. The name A. 
canadensis, which formerly included the Vermont plants now 
attributed to A. arborea, 1s restricted to plants of the coastal 
plain. 

Anchusa arvensis (L.) M. Bieb. [Lycopsis arvensis L.], P, “not 
common,” (NCC), NSF; O, Pownal, (Reed), NSF; Middle- 
bury, (Dodge VT), MI = Lithospermum officinale L. 

Antennaria virginica Stebbins, D, Middlebury, (Brainerd), NSF; 
A, North Dorset, (F/ynn VT), MI = A. neglecta Greene var. 
neodioica (Greene) Cronq.; A, Vernon, (VCC VT), NSF. See 
Bayer and Stebbins (1982). 

Aronia arbutifolia (L.) Elliott [Pyrus arbutifolia (L.) L. f. var. er- 
ythrocarpa Gray], O, P, NL, (NCC), NSF; D, “occasional,” 
(NCC), NSF. 

Asclepias purpurascens L., T, Brattleboro, (Frost), NSF; P, ‘“‘com- 
mon,” (NCC), NSF; D, KI, Woodstock, (Kittredge NY, Bill- 
ings & Kittredge VINS), both MI = A. syriaca L.; C, S, Essex, 
(Carpenter VT), MI = A. syriaca; Burlington, (Aver & Sullivan 
VT), MI = A. syriaca. 

Aster ciliolatus Lindley [4. lindlevanus T. & G.], D, S, Ripton, 
(NCC), NSF; A, D, E, South Bellows Falls, (Blanchard), NSF; 
D, Burlington, (Dole), NSF; A, D, Middletown Springs, (Car- 
penter VT), MI = 4. cordifolius L.; Bellows Falls, (Potter VT), 
MI = A. cordifolius. 

Aster dumosus L., D, Hartland, (:ggleston), NSF; D, Westmin- 
ster, (Blanchard VT), MI = A. racemosus Elliot [A. vimineus 
L. of Vermont authors]; D, Wells River, (Smith), NSF; Vernon, 
(Blanchard VT), MI = A. racemosus;, Rutland, (Eggleston 
HNH), MI = A. cordifolius L. x ? A. pilosus Willd. 

Aster foliaceus Lindley ex DC., A, S, West R. & Connecticut R., 
(several collectors, HNH, NEBC, VT), MI = Aster novi-belgii 
L. Large populations of a plant related to Aster novi-belgii 
occur along the West and upper Connecticut Rivers. The 
plants have the slightly enlarged and veiny outer bracts of 


1995] Jenkins and Zika—Flora of Vermont 303 


the boreal species A. foliaceus, but resemble Aster novi-belgii 
in other features and differ in leaf shape and bract width from 
Canadian and cordilleran material of A. foliaceus. Measure- 
ments from 50 West River collections (Jenkins, unpubl. data) 
show that the plants with broader bracts that have been called 
A, foliaceus grow mixed with typical A. novi-belgii, and that 
intermediates occur. On the basis of this evidence, we refer 
all Vermont collections of A. foliaceus to A. novi-belgii, noting 
that A. novi-belgii, which is known to be quite variable, can 
have outer involucral bracts to 2.0 mm wide which are more 
or less reticulately veined. This conclusion agrees with that 
of a brief study of specimens from the upper Hudson River 
(Jenkins, 1990, unpubl. report to the Adirondack Nature 
Conservancy, Keene) and an extensive study of plants from 
Quebec (Jacques Labrecque, pers. comm.). 

Aster infirmus Michx., D, Danville, (Drake), NSF. 

Aster patens Aiton, P, NL, (Barrows), NSF. 

Aster praealtus Poiret, O, Bellows Falls, (Carey), NSF. 

Aster prenanthoides Muhl., BR, E, Newfane, (Grout), MI = A. 
novi-belgii L. var. tardiflorus (L.) A. G. Jones; E, D, Taftsville, 
(Darling), NSF. 

Aster solidagineus Michx. [Sericocarpus solidagineus Nees], P, 
“not common,” (NCC), NSF. 

Atropa belladona L., D, Wells River, (Smith), NSF. Besides Atro- 
pa belladona, Dole (1937) reported Gaylussacia frondosa (L.) 
T. & G., Vaccinium stamineum L., and Silene pensylvanica 
Michx. from Wells River on the authority of W. P. Smith. 
All are native species and, excepting Vaccineum stamineum 
which was recently discovered in extreme southwestern Ver- 
mont (100 miles from Wells River), all are otherwise un- 
known in Vermont, geographically unlikely, and unsupported 
by vouchers or citations in other floras. 

Avena Sterilis L., A, Charlotte, (Pringle VT), CULT. 

Bidens comosa (A. Gray) Wieg., D, Middlebury (Brainerd), NSF: 
A, S, Vernon, (Eaton NEBC), IS. 

Bidens laevis (L.) BSP. [B. chrysanthemoides Michx.], P, “wet 
places,” (NCC), NSF; E, ““common,” (NCC), NSF; D, Mid- 
dlebury, (Brainerd), NSF; D, Danville, (Drake), NSF; East 
Dorset, (Grout VT), MI = B. cernua L.; Middletown Springs, 
(Carpenter VT), MI = B. cernua; Rutland, (Kirk VT), MI = 


304 Rhodora [Vol. 97 


B. cernua. Bidens laevis, a rare species barely reaching our 
area, closely resembles the common B. cernua and probably 
was the species Perkins and Eggleston had in mind. 

Bromus arvensis L., D, Windham, (Blanchard), NSF; Brattleboro, 
(Ahles MASS), MI = B. japonicus Thunb. 

Bromus hordaceus L. [B. mollis L.], P, NL, (NCC), NSF. 

Calamagrostis cinnoides (Muhl.) Bart. [Deyeuxia nuttaliana St., 
C. nuttalliana Steud.], D, JE, P, Windsor, (Leland), NSF. 

Calystegia hederacea Wallich [C. pubescens Lindl., Convolvulus 
japonicus Thunb.], E, D, NL, (NCC), NSF. 

Cardamine flexuosa With., D, Wallingford, (Brainerd), IC = C. 
pensylvanica Muhl.; D, Smugglers Notch, (Brainerd), IC = 
C. pensylvanica; KI, Woodstock, (Kittredge), NSF. See Kit- 
tredge (1936). The name C. flexuosa, which is now restricted 
to a plant of the southern U.S., was formerly used for broad- 
leaved forms of C. pensylvanica. 

Cardamine rotundifolia Michx., T, Vermont, (Robbins), NSF. 

Carex adusta F. Boott, P, Fairlee, (Blanchard), NSF; P, Middle- 
bury, (Blanchard), NSF 

Carex atrata L., 1, Vermont, IC = C. atratiformis Britton. 

Carex bullata Schk., D, Sharon, (Dutton), NSF; Whitingham, 
(Ahles MASS), MI = C. vesicaria L. 

Carex collinsii Nutt., A, S, ae Co., (VCC NEBC), NSF; A, S, 
Walden, (VCC NEBC), N 

Carex crawel Dewey, P, age (Pringle), NSF. 

Carex dioica L. var. gynocrates (Wormsk.) Ostenf. [C. gynocrates 
Wormsk.], P, T, Burlington, (7orrey), NSF. 

Carex flaccosperma Dewey [C. glaucodea Tuckerm.], D, Leices- 
ter, (Dutton), NSF; D, Middlebury, (Brainerd), NSF; D, Mid- 
dletown Springs, (Carpenter), NSF. 

Carex saxatilis L., O, Mt. Mansfield, (Robbins et al.), NSF; O, 
Camel’s Hump, (Tuckerman GH), IC = C. bigelowii. Carex 
saxatilis and Carex bigelowii are somewhat similar in ap- 
pearance, though only distantly related. Some 19th-century 
authors (1.e., Gray, 1857) treated them, incorrectly, as syn- 
onyms. 

Carex shortiana Dewey, P, Burlington, ‘‘doubtful,” (ex herb. Tor- 
rey), NSF. 

Carex sterilis Willd., A, C, N, S, Concord, (Pease NEBC), IC = 
Carex echinata Murray; “Vermont,” (Kent NEBC), IC = Car- 
ex echinata. Robinson and Fernald (1908) defined C. sterilis 


1995] Jenkins and Zika—Flora of Vermont 305 


broadly, including much of what is now called C. echinata. 
Our more restrictive use of C. sterilis is for a rare coastal and 
boreal species that does not occur in Vermont, according to 
Reznicek and Ball (1980) and Fernald (1950). 

Carex striata Michx. [C. walteriana Bailey], T, Burlington, (Tor- 
rey), NSF. According to House (1924, p. 193), records of C. 
striata in Torrey’s Flora of New York probably refer to Carex 
houghtoniana Torr. ex Dewey. 

Carex styloflexa Buckley, A, C, S, Middlebury, (Brainerd VT), 
MI = Carex laxiflora Lam. or C. gracilescens Steudel. 

Carex tetanica Schk., P, Burlington, (ex herb. Torrey), NSF. 

Carex vestita Willd., O, Middlebury, (James), NSF. 

Carya tomentosa (Poiret) Nutt. [C. a/ba (Mill.) K. Koch, not 
Nutt.], FL, D, Williston, West Haven, (Blake), NSF; FL, D, 
Burlington, (Burns), IC = C. ovata (Mill.) K. Koch. The his- 
torical synonymy of the common hickories is complex, the 
Linnean epithet a/ba having been used in different ways by 
different 19th-century authors. Some early Vermont collec- 
tors, following Gray (1857, 1874) used C. a/ba sensu Nutt. 
for the shagbark C. ovata; this apparently created confusion 
with the mockernut, C. tomentosa, which Robinson and Fer- 
nald (1908) incorrectly called C. alba (L.) K. Koch. Mock- 
ernuts occur in eastern N.Y., but are apparently unknown in 
Vermont. 

Centaurea americana Nutt., D, St. Johnsbury, (Howe), NSF. 

Cerastium viscosum L. [C. glomeratum Thuill.], D, Wells River, 
(Eastman), NSF. 

Chelone lyonii Pursh, D, Woodstock, (Kittredge), NSF. 

Chenopodium berlandieri Moq. var. berlandieri, D, Vergennes, 
Middlebury, (Dutton), NSF. 

Chenopodium berlandieri Moq. var. macrocalycium (Aellen) 
Cronq. cé& macrocal) cum Aellen], A, Colchester, (Flynn VT), 
MI = C. berlandieri var. bushianum (Aellen) Crong. The 
specimen was originally determined as C. album L. and an- 
notated to C. macrocalycium by Wahl during his study of 
the genus (Wahl, 1952-3). We refer it to C. berlandieri var. 
bushianum, noting that Wah] determined several other Flynn 
specimens from this locality, including one collected on the 
same day, as C. bushianum Aellen. 

Cimicifuga racemosa (L.) Nutt., J, O, Middlebury, (James), NSF: 
O, Mansfield Mountain, (VCC), NSF; O, Shelburne & 


306 Rhodora [Vol. 97 


Sharpshin Points near Burlington, (/acrae), NSF; D, “Ver- 
mont,” (ex herb. Torrey VT), DP; “Vermont,” (Dike VT), 
DP. The records from Middlebury, Burlington, and Mt. 
Mansfield are in the northern half of Vermont and are dis- 
junct by over 100 miles from known stations of the species 
in other states. The Dike collection has a printed label saying 
“Flora of Vermont” but the collection data 1s handwritten 
and says only “Cimicifuga racemosa Rich Woods, 20th July, 
A.C. Dike.’ The collection from the Torrey Herbarium says 
only “Actaea racemosa Nutt. Vermont.” The species 1s re- 
ported from within 20 miles of the southern Vermont border 
and could easily have occurred here; but, given the lack of 
detailed collection information and the possibility that the 
specimens may have come from cultivated plants, we cannot 
accept the records. 

Cirsium altissimum (L.) Sprengel, D, Wells River, (Eastman), 
NSF. C. altissimum, a species of the southern U.S., closely 
resembles our C. discolor (Muhl.) Sprengel, which may have 
been the species Eastman saw. 

Cirsium canum (L.) Bieb., D, Castleton, (Higby), NSF. 

Cirsium horridulum Michx., P, T, Brattleboro, (Frost), NSF. 

Cirsium palustre (L.) Scop., D, Wells River, (Kastman), NSF. 

Cirsium undulatum (Nutt.) Sprengel, D, Wells River, (Eastman), 
NSF 


Clematis viorna L., P, T, Castleton, (Carr), NSF. 

Crataegus laevigata (Poiret) DC. [C. oxyacantha L., misapplied], 
A, Ryegate, (Blanchard HNH), IC = C. monogyna Jacq.; A, 
Fairlee, (Eggleston HNH), IC = C. monogyna; A, S, Bur- 
lington, (NCC VT), CULT. Atwood (1973) lists both C. ox- 
yacantha (now a rejected name) and C. monogyna for Ver- 
mont; earlier Vermont authors treated these names as syn- 
onyms, and the Blanchard and Eggleston specimens are in 
fact all C. monogyna. 

Crepis setosa Haller f., EB, E, Townshend, (Blanchard), NSF, D, 
Townshend, (Wheeler), NSF. 

Cyperus polystachos Rottb. [Cyperus filicinus Vahl], A, many 
counties, (several collectors VT, MO, NEBC), IC = C. lupu- 
linus (Sprengel) Marcks [C. filiculmis Vahl var. macilentus 
Fern.). This was probably an orthographic error, substituting 
C. filicinus (a coastal species not known inland) for C. fili- 
culmis, a common Vermont plant. 


1995] Jenkins and Zika—Flora of Vermont 307 


Datura metel L., D, Middlebury, (Brainerd VT), CULT. 

Desmodium canescens (L.) DC., O, P, Pownal, (Robbins), NSF. 

Desmodium obtusum (Muhl.) DC. [Desmodium rigidum (EIl.) 
DC.], E, D, North Pownal, (E-gg/eston), NSF; E, D, Vernon, 
(W.H. Blanchard), NSF. 

Dianthus chinensis L., A, Middlebury, (Brainerd VT), CULT. 

Elaeagnus angustifolia L., A, Westmore, (Dole VT), NDN. 

Eleocharis parvula (R. & S.) Link [E. pygmaea Torr.], BR, K, 
Willoughby, (Dean HNH), MI = E. intermedia (Muhl.) 
Schultes. 

Eleocharis rostellata (Torr.) Torr., P, W, Willoughby Mt., (Tuck- 
erman NEBC), MI = E. pauciflora (Lightf.) Link. 

Erigeron acris L. [E. angulosus Gaudin], D, Royalton, (Drake), 
NSF; St. Johnsbury, (Pringle VT), IC, collection from Maine. 
The nearest known stations are in Aroostock County, Maine. 
A Pringle specimen with a typed label at VT says “barren 
places by the St. Johnsbury RR.” The label is a later addition; 
other, apparently identical, collections have handwritten la- 
bels which read ‘‘barren places by the St. Johns River, Me.” 

Eriocaulon decangulare L., P, Willoughby, (Wood), NSF. 

Erodium moschatum (L.) L’Her., A, S, D, Burlington, (Brainerd 
VT), MI = E. cicutarium (L.) L’Her.; D, Woodstock, (Kit- 
tredge), NSF. 

Eubotrys racemosa (L.) Nutt. [Leucothoe racemosa (L.) Gray], P, 
NL, GYCG), NSF. 

Eupatorium dubium Willd., A, S, many counties, (several collec- 
tors VT, NEBC), MI = E. purpureum L. and E. maculatum 
L. All of the Vermont records for EF. dubium are annotations 
of specimens originally determined as FE. purpureum or E. 
maculatum. None of the annotated specimens agree with 
typical coastal E. dubium and we refer them all to their orig- 
inal determinations. 

Eupatorium fistulosum Barratt, A, S, many localities, (several col- 
lectors LSC, NEBC, VT), MI = £. maculatum L. Eupatorium 
fistulosum is normally differentiated from FE. maculatum by 
a hollow stem, fewer flowers per head, and a more convex 
inflorescence. Hollow-stemmed plants definitely occur in 
Vermont, but field and museum study (Jenkins, unpubl. data) 
has shown that such plants are neither more convex nor 
fewer-flowered than adjacent solid-stemmed plants. In ad- 
dition, the hollowness of the stem is far from being a clear- 


308 Rhodora [Vol. 97 


cut character. It varies in different sections of the stem and 
in different stems in a colony and is often mimicked by boring 
beetles. At present we are unable to recognize two species in 
our material and consider all the hollow-stemmed plants 
from Vermont to be forms of £. maculatum. 

Euphorbia geyeri Engelm., D, North Hero, (Brainerd VT), MI = 
FE. nutans Lag. 

Euphorbia humistrata Engelm., D, St. Johnsbury, (Howe), NSF. 

Euphorbia marginata Pursh, A, C, D, Brandon, (Dutton VT), 

LT; A, Windsor, (Leland NEBC), CULT: Burlington, 
(Flynn VT), CULT. 

Euphorbia serpyllifolia Pers., D, Middlebury, (Brainerd), NSF. 

Forsythia suspensa (Thunb.) Vahl, A, Brattleboro, (Wheeler 
NEBC), CULT. The specimen is from Retreat Park, which 
has a formal garden with extensive ornamental plantings; we 
assume it was cultivated. 

Fragaria x ananassa Duchesne [F. grandiflora Ehrh.], D, Bur- 
lington, (Blake), NSF. 

Galium pumilum Murray [Galium sylvestre Poll.], E, D, Charlotte, 
(Pringle MO), MI = G. mollugo L.; Charlotte, (ex herb. Hors- 
ford MO), MI = G. mollugo. 

Gaylussacia frondosa (L.) T. & G., D, Wells River, (Smith), NSF. 

Gentiana rubricaulis Schwein., C, S, Stowe, (Straw VT), NSF = 
G. linearis Froelich. Dr. James Pringle (pers. comm.), who 
monographed the section Pheumonanthae of Gentiana (Prin- 
gle, 1967), saw the specimen at VT and annotated it to G. 
linearis Froel. The specimen is now missing. 

Geranium dissectum L., O, Castleton, (Carr), NSF. 

Gilia tricolor Benth., D, Stratton, (Blanchard), NSF. 

Glyceria fluitans (L.) R. Br., D, 11 sites, (several collectors VT), 
IC = G. borealis (Nash) Batch. Early Vermont botanists used 
the names G. fluitans, which is now restricted to a species of 
Europe and eastern Canada, and G. septentrionalis A. Hitchc., 
now restricted to a species occuring from Massachusetts south, 
for the common plants now referred to G. borealis. 

Glyceria obtusa (Muhl.) Trin. [Poa obtusa Muhl.], O, P, Bellows 
Falls, (Carey), NSF. 

Glyceria septentrionalis A. Hitche., E, D, Hartland, (Ruggles), 
NSF; D, North Hero, (Brainerd), IC = G. borealis Batch.; D, 
Weston, (Carpenter), IC = G. borealis; D, Colchester, (Dole), 


1995] Jenkins and Zika— Flora of Vermont 309 


NSF; D, “Vermont,” (Pringle), NSF; East Wallingford, (Kent), 
IC = G. borealis. See notes on Glyceria fluitans above. 
Goodyera oblongifolia Raf. [Goodyera menziesii Lindl., Epipactis 
decipiens Ames], D, St. Johnsbury, (Balch & Howe SJFM), 
MI= G. tesselata Lodd.; KI, D, Woodstock, (Kittredge VINS), 
IS. The Woodstock collection at VINS lacks a locality or date 
but has a collection number matching one written in Kit- 
tredge’s copy of her 1931 flora. The plant is missing from 
the sheet, but from the outline of the leaves remaining on 
the paper we believe it was G. tesselata or G. repens (L.) R. 


Br. 

Habenaria ciliaris (L.) R. Br., J, O, P, Middlebury, (James), NSF; 
AM, Troy, (Carey MO), NSF; D, E, Bellows Falls, (Carey), 
NSF; A, Troy, (Carey VT), DP, MI = H. blephariglottis (Willd.) 
Hook.? Habenaria ciliaris was originally credited to the Ver- 
mont flora in an early list of the plants of Middlebury (James, 
1823). Perkins (1888) noted, on the authority of Ezra Brai- 
nerd, that James prepared the list “in his earlier manhood 
and before he had acquired the botanical skill to which he 
afterwards attained,” and that ‘“‘although of great value, it 
nevertheless must be used with care as it contains undoubted 
errors.” The collection cited at MO by Ames (1910) is miss- 
ing. The Carey specimen at VT is labeled ““Habenaria ble- 
phariglottis, Troy, Vt., John Carey, 1861.” We reject this 
record for two reasons. First, Habenaria blephariglottis 
(Willd.) Hook. and H. ciliaris, while easily separated by color 
in the field, are hard to tell apart accurately when dry (Voss, 
1972, p. 442). Herbarium identification rests on the length 
of the fringe on the lip, a character which we have found to 
vary considerably in other Vermont collections of H. ble- 
phariglottis. Because of this variability we cannot determine 
the specimen with certainty, but think it is probably H. ble- 
phariglottis. In addition, we find it hard to believe that if 
Carey had found a yellow-flowered Habenaria he would have 
used the name of acommon white species. Second, the spec- 
imen has been remounted and the current sheet consists of 
portions of two older herbarium sheets, one of which bears 
the specimen, and the other the label data. The original papers 
differ in texture and composition, suggesting they are of dif- 
ferent ages and that an error occurred when the plant was 


S10 Rhodora [Vol. 97 


being remounted. Thus, both the identification of the spec- 
imen and the accuracy of the label are questionable. 

Habenaria leucophaea (Nutt.) A. Gray, D, St. Johnsbury, (Howe 
SJFM), MI = #7. lacera (Michx.) Lodd.; D, Danville, (Howe 
& Balch TUFT), MI = H. psycodes (L.) Sprengel. 

Hieracium gronovii L., T, Colchester, (Torrey VT), MI = H. sca- 
brum Michx.; T, Brattleboro, (Frost), NSF: D, Woodstock, 
(Kittredge), NSF. 

Hydrangea arborescens L., D, Burlington, (Dole), NSF. 

[beris amara L., A, Peacham, (Blanchard HNH), NDN. 

Ilex laevigata (Pursh) A. Gray, T, mouth of Winooski River, (NCC), 
NSF; C, LI, Rutland Co., (NCC), NSF; Charlotte, (Eggleston 
HNH), MI = J. verticillata (L.) A. Gray. 

Ipomoea coccinea L., D, Burlington, (Mrs. Zottman), NSF. 

aa sl L. [Quamoclit vulgaris Chosey], A, Windsor, 
(Leland NEBC), CULT. The specimen has two labels. The 
aoa label says “‘habitat gardens’; a second label, covering 
the first, says ‘‘garden escape.” 

[satis tinctoria L., P, Burlington, (VCC), NSF. 

Juglans nigra L., D, Leicester, (Dutton), NSF; A, C, S, West 
Dover, (Eaton NEBC), MI = J. cinerea L.; “Railroad Street,” 
(Roony SJFM), NDN. 

Juncus brachycarpus Engelm., A, Caledonia Co., (Seymour), NSF. 

Lamium album L., D, St. Johnsbury, (owe), NSF; D, Burling- 
ton, (Flynn VT), MI = L. maculatum L. 

Lappula redowskii (Hornem.) Greene, A, Williston, (Bates GH), 

MI = L. squarrosa (Retz.) Dumort [L. echinata Gilib.]. 

Lechea maritima Leggett, A, C, S, Burlington, (F/ynn VT), MI = 
L. intermedia Leggett; A, C, S, Burlington, (Charette, det. A. 
R. Hodgdon, MO, HNH), MI = L. intermedia; A, C, S, Essex, 
(Charette VT), MI = L. intermedia; A, C, S, Essex, (Brainerd 
VT), MI = L. intermedia. Lechea maritima is, to us, a poorly 
delimited species. The characteristic pubesence of coastal 
plants is not present on inland material that we have seen, 
making L. maritima difficult to separate from L. intermedia 
without mature seeds. The inland range of L. maritima is, 
correspondingly, uncertain. When Hodgdon reviewed the ge- 
nus (Hodgdon, 1938) he found no valid records from Ver- 
mont, though subsequently he identified Charette’s 1969 Bur- 

lington collection as L. maritima. This specimen appears to 


1995] Jenkins and Zika— Flora of Vermont 311 


us to fall within the range of variation of the Vermont spec- 
imens of L. intermedia and we refer it to that species. 

Lepidium ruderale L., P, Rutland, (Pringle), NSF; D, Leicester 
Junction, (Brainerd), NSF; Lyndonville, (Bahosh LSC), MI 
= L. virginicum L.; Brighton, (7rafan LSC), IS; Burlington, 
(Cook, Grout VT), MI = L. densiflorum Schrader. 

Lespedeza procumbens Michx., D, ““Vermont,”’ (VCC), NSF. 

Ligustrum obtusifolium Siebold & Zucc., A, Burlington, (Dole 
VT), NDN. The sheets are labeled: ‘“‘in series sepium” (in 
hedge rows), but this does not make it clear whether they are 
cultivated or escaped. 

Linum grandiflorum Desf., D, Brandon, (Dutton VT), CULT; D, 
Middlebury, (Brainerd VT), CULT. 

Lithospermum latifolium Michx., D, Colchester, (D.B. Griffin), 
NSF 


Luzula confusa Lindeberg, C, S, Rutland, (Eggleston CONN), IC. 
The label on the specimen reads: ““Mt. Washington, N.H., 
ex herb. W. W. Eggleston of Rutland.” 

Lycopus europaeus L., P, “not very common,” (VCC), NSF. 

Lycopus rubellus Moench, D, Bennington, (Rid/on), NSF; A, D, 
Burlington, (Dole), NSF; A, S, Lyndon, (Pease NEBC), MI = 
L. americanus Muhl.; A, 8, D, Willoughby, (Kennedy NEBC), 
MI = L. americanus, A, Colchester, (Dole VT), MI = L. 
americanus, A, Benson, (Atwood VT), MI = L. americanus. 
A range map in a review of this species by Henderson (1962) 
shows no Vermont occurrences. 

Macleaya cordata (Willd.) R. Br., D, Proctor, (Kittredge), NSF. 

Matricaria recutita L. [M. chamomilla L.], D, St. Johnsbury, 
(Howe), NSF. 

Matthiola incana (L.) R. Br., A, Peacham, (Blanchard HNH), 
CULE 


Melothria pendula L., D, Burlington, (F/ynn VT), CULT. 

Mertensia virginica (L.) Pers., E, D, Burlington, (Gifford), NSF; 
D, Fair Haven, (Carpenter), NSF; S, Burlington, (Jones VT), 
NDN. This is a common cultivar. The Jones collection says 
‘old cellar hole’? and provides no evidence that the plant has 
spread beyond the area where it was cultivated. 

Mimulus alatus Aiton, D, Wells River, (Smith), NSF. 

Mimulus guttatus DC. [M. langsdorfii Donn], D, Reading, (Wahit- 
ing), NSF. 


312 Rhodora [Vol. 97 


Muhlenbergia racemosa (Michx.) BSP., A, S, Westmore, (Stevens 
HNH), MI = M. mexicana (L.) Trin.; A, S, Westmore, (Dean 
HNH), IC = M. glomerata (Willd.) Trin.; A, S, Norwich, 
(Jesup HNH), IC = M. glomerata,; KS, Hartland, (Ruggles 
HNH), IC = M. glomerata. The name M. racemosa, which 
formerly included the Vermont plants now referred to M. 
glomerata, is now restricted to a western species. 

Muscari botryoides (L.) Miller, D, “‘occasional,’? (WCC), NSF; A, 

, Peacham, (Blanchard HNH), CULT; A, S, Charlotte, 
(Pringle VT), CULT; North Street, (Carpenter VT), DP, NDN. 
This is acommon cultivar which, although naturalized else- 
where in the northeast, has not been found wild in Vermont. 
The Carpenter specimen lacks a town or a state and was not 
cited in any floras; it may well be a Vermont collection, but 
cannot be accepted without corroborating evidence. 

Myriophyllum heterophyllum Michx., P, NL, (NCC), NSF. 

Nelumbo lutea (Willd.) Pers., D, Champlain Valley, (Brainerd), 


Oenothera biennis L. var. canescens T. & G. [O. canescens Torr. 
& Frem.], D, Woodstock, (Kittredge), MI = O. parviflora L. 
var. oakesiana (Robbins) Fern. 

Oenothera fruticosa L., T, Willoughby, (Frost), NSF; S, Danville, 
(NNC VT), NSF. 

Panicum amarum Elliott, P, Brattleboro, (Barrett, Robbins & 
Pringle), NSF. 

Panicum polyanthes Schultes, D, Leicester, (Dutton), NSF. 

Panicum spretum Schultes, A, S, Townshend, (Dobbin MO), MI 
= P. lanuginosum Ell. 

Papaver dubium L., P, NL, (NCC), NSF; Charlotte, (Pringle VT), 
CULT. 

Papaver intermedium Becker, EB, Townshend, (Blanchard), NSF. 
This name, taken from the original publication, is apparently 
no longer in use. We have been unable able to determine its 
modern equivalent. 

Penstemon laevigatus Aiton [Penstemon calycosus Small], A, S, 
five counties, (several collectors VT, NEBC), MI = P. digitalis 
Nutt.; PE, near Poultney, (Drushe/ PH), MI = P. digitalis. 

Petrorhagia saxifraga (L.) Link [Tunica saxifraga (L.) Scop.], D, 
Morrisville, (Bentley), NSF. 

Petroselinum crispum (Miller) Mansf., A, Townshend, (Wheeler 
NEBC), CULT; A, Caledonia Co., (VCC NEBC), NSF. 


1995] Jenkins and Zika— Flora of Vermont 313 


Phlox drummondii Hook., A, Peacham, (Blanchard HNH), NDN. 

Physalis peruviana L., E, D, Burlington, (Jones VT), MI = P. 
pubescens L.; D, Hinesburg, (Roy VT), MI = P. heterophylla 
Nees. 

Plantago virginica L., T, Brattleboro, (Frost), NSF. 

Poa alpina L., O, Mt. Mansfield summit, (Robbins), NSF. Poa 
alpina, a species of the subarctic, closely resembles our Poa 
fernaldiana Nannf. The latter occurs on Mt. Mansfield and 
may be the species that Robbins saw. 

Polygonum ramosissimum Michx., A, Manchester, (Day VT), MI 
= P. aviculare L.; BR, “frequent,” (NCC), MI = P. aviculare 
(Eggleston and Brainerd, 1904). 

Potamogeton pulcher Tuckerman, P, Brattleboro, (Frost), NSF. 
Prenanthes nana (Bigel.) Torr. [P. trifoliolata (Cass.) Fern. var. 
nana (Bigel.) Fern.], E, Willoughby Mt., (Rusby), NSF. 

Prenanthes racemosa Michx., D, F, Swanton, (Blake), NSF. 

Primula veris L. (P. officinalis (L.) Hill], D, S, Burlington, (Jones), 

NSF. 


Prunus si L., A, S, D, Manchester, (Day NEBC), MI = P. 


cerasus L., 

Prunus ee Ehrh., A, Franklin Co., (WCC NEBC), NSF. 

Prunus persica (L.) Batsch., A, Jamaica, (Wheeler NEBC), CULT. 

Pycnanthemum setosum Nutt. [P. aristatum Michx.], P, Southern 
Vermont, (NCC), NSF. 

Pyrus baccata L., D, St. Johnsbury, (iss Howe), NSF 

Pyrus prunifolia Willd., A, S, Worcester, (Blanchard NEBC), NDN. 

Quercus palustris Muenchh., D, Pittsford, (Dutton), NSF. 

Quercus prinoides Willd., BR, D, E, Pownal, (several collectors 
VT, NEBC, HNH, NCU), MI = Q. muehlenbergii Engelm.; 
A, 8S, Addison, (Brainerd VT), MI = Q. muehlenbergii and 
hybrids. The leaves of QO. prinoides, a species of sandplains 
in southern and central New England, are very similar to 
those of Q. muehlenbergii, a species of limy hills and ledges 
that is uncommon but widely distributed in western Vermont 
(Zika, 1986A). Well-developed leaves of Q. muehlenbergii 
are typically larger and have more teeth than those of Q. 
prinoides. Leaves of intermediate form are common in both 
species, and some authors (i.e., Gleason, 1952) consider Q. 
muehlenbergii an arborescent variety of Q. prinoides. Col- 
lections of Q. prinoides from Quarry Hill in Pownal lack bark 
and habit notes. They cannot be determined with certainty, 


314 Rhodora [Vol. 97 


though many suggest Q. muehlenbergii more strongly than 
Q. prinoides. Quarry Hill is a limestone hill, and has an extant 
population of Q. muehlenbergii,; hence we refer all the his- 
torical collections from Quarry Hill to that species. Speci- 
mens from Snake Mountain in Addison also lack notes and 
are likewise undeterminable. The plants there, which were 
recently relocated, are in a calcareous site and are typically 
low and multi-stemmed but occasionally have strong single 
trunks several meters high. Their leaves are exceptionally 
variable, some suggesting Q. prinoides, some Q. muehlen- 
bergii, and some seemingly hybrids with Q. prinus L., which 
occurs nearby. Given the variability we are reluctant to name 
this population. But we do not think that the Snake Mountain 
plants, viewed as a group, are sufficiently distinct from the 
typical Q. muehlenbergii of the Champlain Valley, or suffi- 
ciently similar to the typical Q. prinoides of southern New 
England, to be the basis of a new state record and northward 
range extension for Q. prinoides. 

Ranunculus micranthus Nutt., A, D, E, Coventry, (Cushman 
HNH), MI = R. abortivus L. 

Reseda odorata L., 8, Windsor, (Leland NEBC), CULT. As with 
Ipomoea quamoclit, the original label (now covered by a 
second label) says “‘garden plant.” 

Ribes nigrum L., A, S, D, Townshend, (Wheeler VT), CULT: 
Dummerston, (Wheeler NEBC), IS; Wallingford, (Kent 
NEBC), IS; Newport, (Know/ton NEBC), IS (sterile). Good 
flowering material is needed to distinguish this species from 
R. americanum Miller. 

Rosa x alba L., D, Townshend, (Wheeler), NSF; D, Stratton, (Un- 
derwood), NSF; D, Leicester, (Dutton), CULT. 

Rosa canina L., D, Woodstock, (Wright), NSF; D, Johnson, (Grout 
VT), IC = R. pimpinellifolia L. LR. spinossisima L.]. The 
specimen was originally determined as R. canina, then an- 
notated, correctly, to R. spinossisima. Later it was cited, in- 
correctly, as R. canina. 

Rosa damascena P. Mill., D, Norwich, (Loveland), NSF; D, 
Leicester, (Dutton), CULT. 

Rosa johannensis Fern., A, S, North Hero, (Knowlton NEBC), MI 
= R. blanda Ait.; A, S, Gardiner Island, (ex herb. Faxon 
NEBC), MI = R. blanda; A, S, Royalton, (Eggleston NY), 
MI = R. blanda; A, S, Ferrisburg, (Faxon et al. NY), MI = 
R. blanda. Rosa johannensis is a species of uncertain status 


1995] Jenkins and Zika—Flora of Vermont 315 


that may eventually be included in R. blanda. Whatever its 
final disposition, all the Vermont specimens seem to be typ- 
ical R. blanda. 

Rosa tomentosa J. E. Smith, D, Woodstock, (Kittredge), NSF; D, 
Pomfret, (Kittredge), NSF. 

Rumex altissimus Wood, A, Windham Co., (VCC VT), NSF. 

Rumex aquaticus L., P, near summit of Mt. Mansfield, (NCC), 
NSF. Rumex longifolius DC. was found as a weed near the 
summit of Mt. Mansfield before 1900, and handwritten notes 
in the copy of Perkins (1888) used to prepare the Brainerd 
et al. (1900) flora suggest that this is the species to which 
Perkins was referring. 

Rumex arifolius All. [Rumex montanus Desf.], D, Burlington, 
(Ross VT), MI = R. acetosa L. 

Rumex patientia L., A, 5 counties, (several collectors HNH, VT), 
IC = R. longifolius DC. [R. domesticus Hartm.]. Early Ver- 
mont botanists, following the Seventh Edition of Gray’s 
Manual (Robinson and Fernald, 1908), used the name R. 
patientia for the plants we now call R. longifolius. True R. 
patientia occurs in the northeastern U.S., but apparently not 
in Vermont. 

Sagittaria lancifolia L. [S. falcata Pursh], T, Brattleboro, (Frost), 
NSF; P, Mallets Bay, Colchester, (VCC), MI (see Brainerd 
et al., 1900). A review by Bogin (1955) lists no occurrences 
of this species north of Delaware. 

Sagittaria subulata (L.) Buch. [S. natans Michx.], P, T, Brattle- 
boro, (Frost), NSF. 

Salix pentandra L., D, Burlington, (Dole), NSF. 

Salix serissima (L. H. Bailey) Fern., A, D, S, Arlington, (Knowlton 
NEBC), IS; A, S, Corinth, (Anderson NEBC), NSF; A, Frank- 
lin Co., (VCC HNH), MI = S. lucida Muhl.; Westmore, 
(Churchill HNH), MI = S. lucida Muhl. Young sterile shoots 
of Salix serissima are hard to distinguish from S. /ucida and 
cannot be determined with certainty. We expect that S. ser- 
issima will eventually be found in Vermont, but, as yet, do 
not have confirming specimens. 

Saururus cernuus L., Bristol, (Dike VT). The specimen is correctly 
identified but is uncorroborated by written citations or other 
collections from Vermont and is probably mislabeled. Geo- 
graphically, Saururus cernuus is a southern species that reach- 
es its northern range limits in southern Massachusetts and 
western New York. Bristol is ca. 150 miles from the nearest 


316 Rhodora [Vol. 97 


stations in western New York and 200 miles from the nearest 
colonies in southern New England and the Hudson Valley. 
This is a substantial disjunction: out of ca. 250 species which 
reach their northern range limits in Vermont, only six are 
disjunct by more than 75 miles from the nearest colonies in 
other states, and only three of these by ca. 100-125 miles 
(Jenkins, in prep.). The wetlands in Bristol—which are only 
20 miles from Burlington—were a famous botanical site and 
received the attention of all the prominent Vermont botanists 
of Dike’s day, especially Pringle and Brainerd. In addition, 
Dike traded specimens with Pringle and others. It seems 
likely to us that the discovery of a conspicuous and exceed- 
ingly rare species would have been noted and verified by 
other botanists and would have been cited in print. Given 
the geographic implausibility and the absence of any corrob- 
orating citations or specimens, we consider it likely that the 
specimen was mislabeled. 

Sedum telephioides Michx., D, Colchester, (F/ynn VT), MI = S. 
telephium L.; D, Burlington, (Dole), NSF. 

Silene caroliniana Walter [S. pensylvanica Michx.], D, Wells Riv- 
er, (Smith), NSF. 

Sisyrinchium albidum Raf., P, NL, (NCC), NSF. 

Smilax glauca Walter, D, Clifton, (Britton VT), DP. There is no 
Clifton in Vermont. Britton made a number of collections 
from Clifton, N. Y., on Staten Island. 

Smilax rotundifolia L., O, P, NL, (NCC), NSF. 

Solanum tuberosum L., A, Newfane, (Wheeler NEBC), NDN. 

Solanum villosum Miller, S, Westford, (VMunier VT), MI = S. 
sarrachoides Sendtner. 

Solidago calcicola Fern., C, CR, S, Mt. Killington, (Dutton et al. 
VT, GH, NEBC), MI = S. macrophylla Pursh x rugosa Ai- 
ton. The specimens were probably from a single plant. The 
original collectors took them to Fernald who first determined 
them as S. calcicola (Kirk, 1912A, 1912B), then compared 
them to the northern material he was calling S. calcicola and 
changed his mind (Anonymous, 1915). 

Solidago rigida L., T, Burlington, (Torrey), NSF. 

Solidago stricta Aiton, T, Burlington, (Torrey), NSF. 

Spartina alterniflora Loisel, D, Vermont, (Pringle), NSF. Given 
the lack of any citations in other floras we regard it as 1m- 
probable that Dole ever saw a Pringle sheet of either this or 
the next. 


1995] Jenkins and Zika— Flora of Vermont a17 


Spartina patens (Aiton) Muhl., D, Vermont, (Pringle), NSF. 

Sphenopholis pensylvanica (L.) A. Hitche. [S. palustris Scribn., 
Trisetum pensylvanicum (L.) Beauv.], A, S, Wells River, (Jones 
VT), NSF. References to Avena pensylvanica L. or Eatonia 
pensylvanica A. Gray in early floras (i.e., Brainerd et al., 1900) 
refer to S. intermedia (Rydb.) Rydb. and not S. pensylvanica. 

Spiraea japonica L. f., D, Woodstock, (Kittredge), NDN; Pea- 
cham, (Blanchard NY), NDN, MI = Astilbe japonica (Morren 
& Decne.) A. Gray. 

Spiraea prunifolia Siebold & Zucc., A, Middlebury, (Brainerd 
VT), NDN. Brainerd collected this species in 1898 but did 
not list it in his flora (Brainerd et al., 1900), suggesting that 
the collection was from a cultivated plant. 

Stellaria longipes Goldie [Stellaria strictiflora (Rydb.) J. M. Ma- 
coun], D, Woodstock, (Porter), MI = S. graminea L.; D, 
Plymouth, (Heselton), NSF; D, Passumpsic, (Howe), NSF; D, 
Weston, (Carpenter), NSF. 

Thalictrum revolutum DC. [T. purpurascens L. of Robinson and 
Fernald (1908) in part], KI, D, Woodstock, (Kittredge VINS), 
MI = T. dioicum L.; D, Clarendon, (Hitchcock NY), MI = 
T. pubescens Pursh. 

Thaspium trifoliatum (L.) A. Gray [T. aureum Nutt., misapplied; 
Thaspium aureum var. aptera A. Gray], P, “common in damp 
fields,’ (NCC), NSF. The treatments of Thaspium and Zizia 
are extremely confused in 19th-century manuals (i.e., Gray, 
1874). Perkins was probably referring to Zizia aurea (L.) 


Koch. 

Tipularia discolor (Pursh) Nutt., P, Brattleboro, (Frost), NSF; P, 
“Vermont,” (Beck), NSF; D, Bradford, (Bacon), NSF. Perkins 
(1888) called the early records doubtful. Despite the lack of 
a specimen or corroborating documentation, Tipularia was 
put on the official Vermont list of protected plants (13 V. S. 
A., Chapter 79) in the 1930’s. 

Tradescantia subaspera Ker Gawler, A, S, D, Hartland, (Carpen- 
ter VT), MI = T. virginiana L. 

Trifolium medium L., D, Wells River, (Eastman), NSF; D, Bur- 
lington, (Jones), NSF. 

Trifolium stoloniferum Muhl., P, Bellows Falls, (Brown), NSF. 
See Brooks (1983). 

Viburnum prunifolium L., D, Ferrisburg, (Kittredge), NSF. 

Viola brittoniana Pollard, D, Vermont, (Dutton), ; 

Viola elatior Fries., D, Wells River, (Eastman), NSF, NDN. 


318 Rhodora [Vol. 97 


Viola odorata L., A, 8, Jamaica, (Wheeler NEBC), NDN; Mid- 
dlebury, (Brainerd HNH), CULT; Norwich, (Richardson 
HNH), NDN. 

Viola pedata L., T, Brattleboro, (Frost), NSF. 

Viola striata Aiton, A, S, Middlebury, (Brainerd VT), NSF. 

Vitis vulpina L. [V. cordifolia Michx.], P, NL, (NCC), NSF. 

Woodsia oregana D.C. Eaton, B, G, Charlotte, (Pringle IA), CULT. 

Zizia aptera (Gray) Fern. [Zizia cordata (Walt.) DC., Thaspium 
trifoliatum Gray var. apterum Gray, ? T. cordatum T. & G.], 
P, NL, (NCC), NSF; O, Middlebury, (James), NSF. The treat- 
ments of Thaspium and Zizia are extremely confused in | 9th- 
century manuals (1.e., Gray, 1874), and we are not sure what 
species Perkins and Oakes had in mind. 


DISCUSSION 


The rate of additions 


How fast are species being added to the Vermont flora? Our 
manuscript checklist (Jenkins and Zika, in prep.) lists 1933 species 
for Vermont, of which 1310 (68%) are believed to be native. In 
the last 19 years approximately 19 native species and 58 aliens 
have been added to the flora, or roughly three aliens for every 
native species. These numbers give a growth rate of 0.2% per year 
for the whole flora, 0.08% for the native species, and 0.5% per 
year for the alien species, with the pool of aliens growing six times 
faster than the pool of native species. 

These numbers show that there are still discoveries to be made 
in a flora which, by United States standards, 1s very well-known. 
If the present discovery rate continues, the number of native 
species reported will grow by nearly 10% in the next century, the 
number of aliens by a remarkable 50%, and the whole flora by 
approximately 20%. 

The continuing discovery of new species, after 150 years of 
botanical effort, raises the question of whether the newly discov- 
ered plants are recent immigrants or resident species that have 
been overlooked. This is an impossible question to answer fac- 
tually, but we can advance three indirect arguments that support 
the hypothesis that many of the newly discovered species are new 
arrivals. 

(1) Approximately 620 species of aliens have immigrated to 
Vermont in about 230 years of European settlement. The average 


1995] Jenkins and Zika—Flora of Vermont 319 


immigration rate was 2.7 species per year and the peak immi- 
gration rate has to have been larger than this. The current dis- 
covery rate (about 3.0 alien species per year) is roughly the same 
as the average immigration rate and so it is at least possible that 
many discoveries are recent immigrants. 

(2) The discovery rate for all species declined six-fold from 
about 20 species per year in the late 1800’s to a rate of two to 
three species per year by the middle of this century. If we were 
exhausting a fixed pool of undiscovered species, we would expect 
the decline would have continued and that the current discovery 
rate would be lower than the mid-century rate. But instead, it has 
increased slightly to about four species per year (Jenkins and Zika, 
in prep.). This suggests either that the depletion of the pool of 
undiscovered species is being offset by increased botanical effort 
or that the pool is being replenished through immigration. 

(3) Many of the newly discovered species, both natives and 
aliens, are colonizing species found in successional or human- 
disturbed habitats. In such habitats, individual populations of 
rare species are almost always transient; we have only a few doc- 
umented examples (e.g., Lygodium palmatum (Bernh.) Swartz, 
Ceanothus herbaceus Raf.) of rare species that have persisted for 
more than a decade or two in successional habitats. Probably 
many of the newly arrived species in such habitats neither persist 
nor establish elsewhere (Pimm, 1991; Muhlenbach, 1979). The 
remainder spread, often rapidly, as Puccinellia distans (Jacq.) Parl. 
and Panicum dichotomiflorum Michx. have recently done (Zika, 
1990) or as Cardamine impatiens L. and Carex praegracilis W. 
Boott appear to be doing (Jenkins, pers. obs.; Reznicek and Old- 
ham, 1993). This suggests that it is likely that many of the species 
that are both rare and restricted to transient habitats are fairly 
recent arrivals. 

None of the above arguments proves that the newly discovered 
species are recent immigrants, but, taken together, they at least 
make it plausible. 


The plausibility of unvouchered records 


We have deleted 130 unvouchered taxa from the flora. It is 
likely that some of these excluded taxa (for example, Carex crawei 
Dewey, which Perkins reports on the basis of a collection by C. 
G. Pringle from Charlotte (Perkins, 1888)) are valid records. Giv- 


320 Rhodora [Vol. 97 


en that Pringle knew sedges well and was a scrupulous collector, 
that good habitat for the species exists in Charlotte, and that the 
species occurs on the New York side of Lake Champlain within 
20 miles of Charlotte, it seems quite likely to us that a specimen 
once existed and has since been lost. 

With equal certainty, some of the unvouchered records are 
invalid. The record of Erigeron acris L. from Royalton, Vermont, 
300 miles south of the nearest vouchered colony and mentioned 
only in a flora (Dole, 1937) whose compilers did not require 
specimens, is almost certainly a mistake. 

By rejecting unvouchered records, we thus guard against mis- 
takes; but, at the same time, we exclude some authentic records 
for which the validating specimens have been lost. This 1s com- 
mon botanical practice and 1s amply justified if our chief purpose 
is to avoid mistakes or, for that matter, to tabulate as accurately 
as possible the current holdings of botanical museums. But if our 
purpose 1s to make the most accurate possible list of the historical 
flora of the state and if, on the average, more of the unvouchered 
records are right than wrong, we will improve our overall accuracy 
by including them, albeit at the cost of including some mistakes 
as well. But are more of the unvouchered records right than wrong? 

A review of the geography of the unvouchered records strongly 
suggests that they are not. Of the 85 native species with one or 
more unvouchered records, only 15 are known from sites within 
100 miles of their reported Vermont locality. The remaining 70 
(82%) are long disjuncts. This contrasts with an overall frequency 
of less than 3% for long disjuncts among the 1310 native species 
that we currently list for Vermont (Jenkins and Zika, in prep.). 
Thus, the frequency of long disjunctions is 27 times higher among 
unvouchered than vouchered records, suggesting that many of 
them are geographically suspect. 

The geographic plausibility of the 45 species of unvouchered 
aliens is more difficult to evaluate. Since long disjunctions are 
common in weedy species and since most cultivars are potentially 
able to escape, any alien species occurring or cultivated in the 
northeast U.S. might conceivably occur in Vermont. With this 
interpretation of what is geographically possible, 21 of the un- 
vouchered aliens are regarded as possible because they are cul- 
tivars, 16 are regarded as possible because they have occurred as 
introductions in the northeast, and 8 are neither cultivated nor 
escaped regionally and so unlikely on geographic grounds. 


1995] Jenkins and Zika— Flora of Vermont CPR 


Table 1. Approximate error rates in previous Vermont floras relative to current 
knowledge 


Date of Flora 
1853 1888 1900 1915 1937 1969 1973 


Total Species? 1034 1360 1563 1694 1861 1927 1990 
Species we de- 
ete? 


27 33 4 15 101 40 58 
Possible error 
rate (%)« 2.6 2.4 0.3 0.9 5.4 2.1 2.9 


Notes: * Number of species in flora, not corrected to current taxonomy. * Number 
of species from that flora we have deleted in this article. ‘ Number of species 
deleted divided by number species in flora, in percent. 


Combining natives and aliens, 78 of the 130 unvouchered re- 
cords (60%) are geographically unlikely. Fifteen (12%) are geo- 
graphically plausible native species and 37 (28%) are cultivars or 
rare weeds which we also consider geographically plausible. Thus 
at least three-fifths of the unvouchered records are geographically 
unlikely and most probably based on misidentifications. This 
suggests that the inclusion of unvouchered records would add a 
substantial number of misidentifications (at least 78 and possibly 
as many as 130) to the Vermont flora and substantially decrease 
the overall accuracy of the flora. 

This conclusion is strengthened by the observation that in the 
last 14 years of field work Vermont botanists rediscovered about 
100 of the 200 vouchered native species that were considered 
missing in 1980 (50%), but only two out of the 83 missing un- 
vouchered natives (2.4%). This means that either unvouchered 
native species are about 20 times harder to relocate than vouch- 
ered ones, or that 20 times more unvouchered native species have 
gone extinct than vouchered ones, or that many of the unvouch- 
ered species have never been here at all. 


The completeness and accuracy of historical floras 


Every flora has errors and omissions. The existence of seven 
previous Vermont floras gives us a rare opportunity to examine 
these quantitatively (Table 1). By using our manuscript checklist 
(Jenkins and Zika, in prep.) as a benchmark, we can calculate the 
percentage of the species we accept that are included in each 
previous flora, and also the ratio of the records we would delete 


S22 Rhodora [Vol. 97 


from a previous flora to the total number of species in it (possible 
error rate). Note that we do not count changes in taxonomic rank 
as omissions or errors. We call the ratio a possible error rate 
because we may have erred in some of our re-determinations of 
historical specimens and because some percentage of the un- 
vouchered records we are deleting may be valid. It is thus an 
estimate of the upper bound of the percentage of species erro- 
neously credited to the flora. 

Table | presents the results. The most interesting thing to note 
is that while the percentage of the 1992 flora included has in- 
creased with each successive flora, the possible error rate has 
sometimes increased and sometimes decreased. Newer 1s not nec- 
essarily better. The nineteenth century floras had a respectable 
possible error rate of about 2.5%, largely the result of misiden- 
tifications and taxonomic uncertainties, but included approxi- 
mately 52% and 68% of our checklist flora (Jenkins and Zika, in 
prep.*). The two floras early in this century, which were based 
entirely on vouchers and prepared with great attention to taxo- 
nomic detail, included approximately 83% of our checklist flora 
and had impressively low possible error rates of 0.3% and 0.9%. 
The 1937 flora included 85% of the species we currently recognize; 
but it mixed vouchered and unvouchered records and so had a 
possible error rate of 5.4%, the highest of any Vermont flora. The 
1969 flora and 1973 checklist, following a period of relative bo- 
tanical inactivity, still increased the known Vermont flora to ap- 
proximately 93% of our checklist flora total and improved the 
accuracy of the flora by eliminating many unvouchered records 
that had been reported in 1937. But in many cases the compilers 
of these works transcribed rather than verified label data and so 
introduced a number of misidentifications, nomenclatural errors, 
and inaccurate citations. In consequence, the 1969 flora and 1973 
checklist have possible error rates of 2.1% and 2.9%, roughly half 
those of the 1937 flora, but several times higher than those of the 
more scrupulous 1900 and 1915 floras. 


+ Estimates of the percent of our checklist flora included in an historical flora 
are made by taking the number of species in that flora (top line of Table 1), 
subtracting the number of species we are deleting (second line of Table 1), and 
then correcting for differences in taxonomic concepts. For example, the 1937 flora 
had 1861 species. We delete 102 of these, then subtract 125 species that we do 
not recognize, and add three species formerly considered varieties. The result is 
1637 species, or 84.6% of our checklist total of 1933 species. 


1995] Jenkins and Zika—Flora of Vermont 323 


We close with several comments about floristic practice. First, 
Table 1 confirms what many botanists have long suspected: at- 
tempting to base a flora on unvouchered field records is a risky 
business. The Vermont floras that included unvouchered records 
(1853, 1888, 1937) had apparent error rates of 2.4% to 5.4%, 
meaning that anywhere from one species in 42 to one species in 
19 in these works might be erroneously attributed to the flora. 
Second, owing to nomenclatural changes and the inevitable mis- 
identifications found in all herbaria, attempting to do a flora by 
compiling herbarium records without verifying the nomenclature 
and the identifications 1s likewise risky. The recent Vermont floras 
that did this (1969, 1973) had apparent error rates of 2.1% and 
2.9%, lower on average than those that accepted unvouchered 
records, but still suggesting that one species in 34 to one in 48 
was erroneously attributed to the flora. 

And finally, as has been forcefully stated by E. Voss (1972, p. 
3, 7-8), given the inaccuracies in previous floras, it follows that 
any attempt to compile a state or regional checklist by combining 
several previous floras without examining primary records is like- 
ly to be highly inaccurate. If we were to compile a checklist of 
the flora of Vermont in this way, it would contain about 2125 
species. One hundred ninety-two of these would rest on unvouch- 
ered or otherwise unacceptable records, for a possible error rate 
of nine percent, or one species in 11. 

Thus, our conclusion, which is hardly radical, is that the prac- 
tice of basing floristic work on the critical examination of vouchers 
seems amply justified. Such work in Vermont has produced error 
rates six to eighteen times lower than those of less critical works 
and approximately thirty times lower than what might be pro- 
duced by compiling printed records. 


ACKNOWLEDGMENTS 


Among the many botanists who contributed to this effort, we 
especially thank Ray Angelo, David Barrington, Debbie Benja- 
min, Everett Marshall, Thomas Rawinski, Bruce Sorrie, Elizabeth 
Thompson, and Gordon C. Tucker. We are grateful to the Pringle 
Herbarium of the University of Vermont and the Vermont Bird 
and Botanical Clubs for partial funding. For loans and access to 
specimens, we thank the curators of the following herbaria: BEDF, 
CONN, GH, IA, LSC, MASS, MO, MOR, NCU, NEBC, NHA, 


324 Rhodora [Vol. 97 


NY, NYS, OSC, SJFM, TUFT, VINS (Vermont Institute of Nat- 
ural Science, Woodstock), VT, YU. 


LITERATURE CITED 


Ames, O. 1910. The genus Habenaria in North America. Orchidaceae 4: 159 

ANGELO, R. 1989. Some new records for New England and New England states. 

Rhodora 91: 342-344. 

. 1990. Some new records for New England and New England states. II. 
Rhodora 92: 294-296. 

Anonymous. 1915. Notes. Joint Bull. Vermont Bot. Bird Club 1: 34-36. 

Atwoop, J. T., W. D. COUNTRYMAN, R. A. Jervis, D. H. MILLER, F. C. SEYMouR, 

M. L. SmitH. 1973. Checklist of Veont Plants. Vermont Botanical 


Bayer, R. J. AND G. L. Stepsins. 1982. A revised classification of Antennaria 
(Asteraceae: Inuleae) of eastern United States. Syst. Bot. 7: 300-313. 
BEITEL, J. M. AND J. T. Micke. 1992. The Appalachian firmoss, a new species 

n the Huperzia selago (Lycopodiaceae) rae in eastern North America, 

ae a new combination for the western firmoss. Amer. Fern J. 82: 41-46. 

Boain, C. 1955. Revision of the genus ia (Alismataceae). Mem. New 
York Bot. ase 9: 179-233. 

BRAINERD, E., L. R. JONES, AND W. W. EGGLESTON. 1900. Flora of Vermont. 
Free Press ee Burlin 

Brooks, R.E. 1983. Trifolium siolonienim: running buffalo clover: description, 
distribution, and current status. Rhodora 85: 343-354. 

Brown, D. F. M. 1964. A ea Study of the Fern Genus Woodsia. 
Cramer, Weinheim, Germ 

COUNTRYMAN, W. D. 1978. Rare and Endangered Vascular Plant Species in 
Vermont. The New England Botanical Club in cooperation with the U.S. 
Fish and Wildlife Service, Newton Corner, MA 

Cronaulist, A., A. H. HoLtmMGren, N. H. Ho_maren, J. L. REVEAL, AND P. K. 
Ho“tmGrReEN. 1977. Intermountain Flora, Vol. Six, The Monocotyledons. 
Columbia University Press, New York, NY. 

Crow, G. E. 1982. New England’s Rare, Threatened and Endangered Plants. 
U.S. Fish and Wildlife Service, in cooperation with the New Hampshire 
Agricultural Experiment Station, U.S. Government Printing Office, Wash- 


on, DC. 
Do eg, E. J., ed. 1937. The Flora of Vermont, 3rd ed. Free Press Printing Co., 


EGGLESTON, W. W. AND : BRAINERD. 1904. Addenda to the flora of Vermont. 

Rhodora 6: . 14 
, AND a . UNDERWOOD. 1915. Flora of Vermont. Vermont 

Agric: fi. an Bull. 187: 139-258. 

FarrRAR, D. R. 1992. Trichomanes intricatum: the independent ee 
gam ava in the eastern United States. Amer. Fern J. 82: 

FERNALD, M. 1950. Gray’s Manual of Botany, 8th ed. Aenea a Co., 
New ae NY. 


1995] Jenkins and Zika— Flora of Vermont 325 


Fiynn, N. F. 1911. Flora of Burlington and Vicinity. Free Press Printing Co., 
Burlington, VT. 

GitMan, A. V. 1993. Four recent additions to the vascular flora of Vermont. 
Maine Naturalist 1: 31- 

Geason, H. A. 1952. The New Britton and Brown Illustrated Flora of the 
Northeastern United States and Adjacent Canada. The New York Botanical 
Garden, Bronx, NY 

— AND A. CronautstT. 1991. Manual of Vascular Plants of Northeastern 
United States and Adjacent Canada, 2nd ed. The New York Botanical Gar- 


Gray, A. 1857. Manual of the cea of the Northern United States, rev. ed. 

Iveson & Phinney and G. P. Putnam & Co., New York, NY. 

1874. Manual of the Botany . the Norhea United States, Sth ed. 

Iveson, rei Taylor & Co., New York, NY. 

HAuFLER, C. H. AND M. D. sheen tn ipo New species of North American 
Cystopteris 4 Polypodium, with comments on their reticulate relationships. 
Amer. Fern J. 81: 7-23 

HELLqQuistT, C. B. AND G. Crow. 1980. Aquatic vascular plants of New England: 
Part 1. Zosteraceae, Potamogetonaceae, Najadaceae. New Hampshire Agric. 
Exp. Sta. Bull. 515. 

R. L. Hitton. 1983. A new species of Potamogeton (Potamogeto- 
naceae) from northeastern United States. Syst. Bot. 8: 86-92 

HENpDeRSON, N.C. 1962. A taxonomic revision of the genus Lycopus (Labiatae). 
Amer. Midl. Naturalist 68: 95-138. 

Hopcpon, A. R. 1938. A taxonomic study of Lechea. Rhodora 40: 29-69, 87- 
131. 

HouLmGrEN, P. K., N. H. HOLMGREN, AND L. C. , eds. 1990. Index 
Herbariorum. Part I. The Herbaria of the ca "sth a New York Botanical 
Garden, Bronx, NY. 

House, H. D. 1924. Annotated list of the ferns and flowering plants of New 
York State. New York State Mus. Bull. 254. 

JAMES, E. 1823. Catalogue of plants. Collect. Mass. Hist. Soc. 2: 146-148. 

Jesup, H.G. 1891. A Catalogue of the Flowering Plants and Higher Cryptogams, 
Both Native and Introduced, Found Within About 30 Miles of Hanover, 
N.H., Including a Few Cultivated Species, to Which is Appended a List of 
Vertebrate Animals of the Same Region. Publ. by the author, Hanover, NH. 

Kartesz, J. T. 1994. A Synonymized Checklist of the Vascular Flora of the 
United States, Canada, and Greenland. Timber Press, Portland, OR 

KENNEDY, G. G. 1904. Flora of Willoughby, Vermont. Rhodora 6: 91-134, 148. 

Kirk, G. L. 1912A. Solidago calcicola in Vermont. Rhodora 14: 54-55. 

——. 1912B. Rutland County flora. Bull. Vermont Bot. Club 7: 14-15. 

KitrreDcE, E. M. 1931. Ferns and Flowering Plants of Woodstock, Vermont. 
The Elm Tree Press, Woodstock, VT 

1936. eee to Ferns and wiewerns Plants of Woodstock, Ver- 
mont. No publisher cit 

—. 1939. Grasses and Sedees of Woodstock, Vermont. The Elm Tree Press, 
Woodstock, VT. 

LittLe, E. L. 1977. Atlas of United States Trees. Vol. 4, Minor Eastern Hard- 
woods. Misc. Pub. U.S. Dept. Agric. 1342. 


326 Rhodora [Vol. 97 


Oakes, W. 1842. Catalogue of Vermont plants, pp. 173-208. Jn: Z. Thompson, 
ed. History of Vermont, Natural, Civil, and Statistical. Publ. by the author, 
Burlington, VT. 

MUHLENBACH, V. 1979. Contributions to the synanthropic eee flora of 
the railroads in St. Louis, Missouri, U.S.A. Ann. Missouri Bot. Gard. 66: |- 

108. 


Paris, C. A. 1991. Adiantum viridimontanum, a new maidenhair fern in eastern 
North America. Rhodora 93: 105-121. 

PENNELL, F.W. 1935. The Scrophulariaceae — temperate North America. 

cad. Nat. Sci. Philadelphia Monogr. -650. 

ee G. H. 1888. Catalogue of the Foe - Vermont. State Board of Agric. 
10th Ann. Rep., Burlington, VT. 

Pim, S. L. 1991. The Balance of Nature? University of Chicago Press, Chi- 

IL 


cago, IL. 
—e J.S. 1967. Taxonomy of Gentiana, section Pneumonanthae, in eastern 
rth ewe Brittonia 19: 1- 


eae A. A. AND P. W. BALL. 1980. The taxonomy of Carex section stel- 
lulatae in North America north of Mexico. Contr. Univ. Michigan Herb. 14: 
153-203 


AND M. J. OLDHAM. 1993. Carex praegracilis W. Boott (Cyperaceae) new 
to New England. Rhodora 95: 425-426. 

Rosinson, B. L. AND M. L. FERNALD. 1908. Gray’s Manual of Botany, 7th ed. 
American Book Co., New York, NY 

RotuHrock, P. E. 1991. The identity of Cie albolutescens, C. festucacea, and 
C. longii (Cyperaceae). Rhodora 93: 51-66. 

Seymour, F.C. 1967. Notes from the Pringle Herbarium. II. Rhodora 69: 377— 
380 


—. 1969. The Flora of Vermont, 4th ed. Univ. of Vermont Agric. Exp. Sta. 
Bull. 660. 
1982. The Flora of New England, 2nd ed. Phytologia Mem. V, Plain- 


Torrey, J. 1853. Catalogue of Veanant Plants, pp. 36-40. Jn: Z. Thompson, 
Appendix to the History of Vermont, Natural, Civil, and Ca Publ. 
by the author, Burlington, 

Voss, E.G. 1972. Michigan Flora. Part I. Gymnosperms and Monocots. Cran- 
brook een 7 Science, Bloomfield Hills, 

WaGNneER, W. H. A F. S. WAGNER. 1982. Botr vehiin rugulosum (Ophioglos- 
saceae), a aha recognized species of evergreen grapefern in the Great Lakes 
area of North America. Contr. Univ. Michigan Herb. 15: 315-324 

Wau, H. A. 1952-3. A preliminary study of the genus Chenopodium in North 
America. Bartonia 27: 1-46. 

Watson, S. AND J. M. CouLTer. 1890. Manual of the Botany of the Northern 
United States, 6th ed. American Book Co., x 

ZIKA, P. F. 1986A. The distribution of Quercus muehlenbergii in Vermont. Joint 
Bull. Vermont Bot. Bird Club 20: 15-19. 

6B. The first record of Poa bulbosa L. (Poaceae) for Vermont. Rho- 

dota 88: 419-420. 


1995] Jenkins and Zika—Flora of Vermont 327 


. 1987. The first record of Allium vineale L. (Liliaceae) for Vermont. 
Rhodora 89: 93-94. 
. 1988. Contributions to the flora of the Lake Champlain Valley, New 
York and Vermont. II. Bull. Torrey Bot. Club 115: 218-220. 
1990. Range expansions of some grasses in Vermont. Rhodora 92: 
80-89. 
—. 1991A. The first report of Agrostis exarata var. monolepis (Poaceae) in 
New England. Rhodora 93: 398-399. 
—. 1991B. Discovery of Juncus vaseyi (Juncaceae) in Vermont. Rhodora 
93: 395-397. 
. 1992. Contributions to the alpine flora of the northeastern United States. 
ees 94: 15-37. 
E. J. MARSHALL. 1991. Contributions to the flora of the Lake Cham- 
slain ei New York and Vermont. III. Bull. Torrey Bot. Club 118: 
58-61. 
, R. J. STERN, AND H. E. Anes. 1983. Contributions to the flora of the 
Lake Champlain Valley, New York and Vermont. Bull. Torrey Bot. Club 
110: 366-369. 
AND E. THOMPSON. 1986. Notes on the flora of Windham County, Ver- 
mont. Rhodora 88: 517-523. 


Deas 
WHITE CREEK FIELD SCHOOL 
WHITE CREEK, NY 12057 


HERBARIUM 

DEPT. OF BOTANY & PLANT PATHOLOGY 
OREGON STATE UNIVERSITY 
CORVALLIS, OR 97330 


RHODORA, Vol. 97, No. 892, pp. 328-338, 1995 


OCCURRKENCE-OF THE 
A THOREA VIOLACEA 
(BATRACHOSPERMALES: THOREACEAE) IN THE 
HUDSON RIVER, NEW YORK STATE 


CurT M. PUESCHEL, P. GARY SULLIVAN, 
AND JOHN E. Titus 


ABSTRACT 


We discovered a population of Thorea violacea in the upper Hudson River of 
New York State during September 1994. The same site was barren of Thorea the 
following June, but two and a half months later—a year after the initial collection— 
a large population of thalli, some over a meter in length, was again present. This 
seasonality may account in part for the seeming rarity of this large, conspicuous 
alga. Thalli presumed to represent the Chantransia phase of the Thorea life cycle 
occurred on rocks at the same location with the same phenology. The Hudson 
ald site represents the first confirmed North American locality of 7. violacea 

rth of Texas and the most northerly North American locality of the family 

eee Characterization of the chemical and physical eougiions at the anes 

site suggests broader environmental tolerances (lower specific eater 
water flow, colder water regimes) of this group than previously known. 


Key Words: Chantransia stage, freshwater algae, Hudson River, Rhodophyta, 
Thorea violacea 


INTRODUCTION 


The Thoreaceae (Batrachospermales) is a family of multiaxial 
red algae, whose members, 7horea and Nemalionopsis, are among 
the largest red algae found in freshwater; Thorea may be up to 2 
m long (Bischoff, 1965). The branched, cylindrical axes of these 
algae consist of an unpigmented, filamentous medulla surrounded 
by laterally disposed, highly pigmented filaments. Thallus color 
is generally blue-green or re 

Life cycles in the three families of the Batrachospermales typ- 
ically involve alternation of morphologically dissimilar phases. 
Taxonomic diagnoses are based principally on the more con- 
spicuous and anatomically complex gametophyte phase. This 
phase bears the generic epithet, e.g., Thorea phase. The sexual 
life cycle known for representatives of the Batrachospermaceae 
and Lemaneaceae involves an inconspicuous, freely branching, 
diploid phase, called Chantransia, that produces sessile, haploid 

ametophytes directly from apical cells following somatic meiosis 
(Sheath, 1984). Female gametes (carpogonia) and zygotes are re- 


328 


1995] Pueschel et al.— 7horea violacea 329 


tained on the gametophyte, which results in the growth of diploid, 
spore-producing filaments, collectively termed the carposporo- 
phyte, on the female gametophyte. Diploid carpospores, released 
by the carposporophyte, grow into the Chantransia phase (Sheath, 
1984). 

Life histories of the Thoreaceae are poorly known. Sexual struc- 
tures have been described for some representatives of Thorea 
(Yoshizaki, 1986; Necchi, 1987), but the location of meiosis in 
the life history has not yet been documented. Asexual reproduc- 
tion is common and conspicuous in this family (Swale, 1963) and 
involves the differentiation of spores, termed monospores, from 
terminal cells of lateral filaments. Although the morphology of 
the asexually reproducing 7horea phase thalli resembles that of 
gametophytes, monospores rather than gametes are produced 
(Swale, 1963). Monospores are also produced by Chantransia 
phase thalli. Regardless of the phase that generates the mono- 
spores, all monospores grow into thalli of the Chantransia mor- 
phology (Swale, 1962). Presumably, 7horea phase thalli that pro- 
duce monospores rather than gametes are part of an apomictic 
life cycle, but chromosome counts are not available to evaluate 
this hypothesis. Generic phase thalli of Nemalionopsis produce 
monosporangia terminally on long lateral filaments (Sheath et al., 
1993). Generic phase thalli of 7horea produce monosporangia on 
short, branching, lateral filaments, and the long lateral filaments 
are entirely vegetative (Sheath et al., 1993), 

Sheath et al. (1993) recently revised the taxonomy of the Tho- 
reaceae, reducing the number of recognized species of Thorea 
from thirteen to four. Two of these occur in North America, most 
commonly in the southern United States and Mexico (Sheath et 
al., 1993). Thorea hispida (Thore) Desvaux (as 7. ramosissima 
Bory and 7. andina Lagerheim et Mébius) has been reported from 
Nebraska (Hedgcock and Hunter, 1899), Illinois (Tiffany and Brit- 
ten, 1952), and Ohio (Hirsch and Palmer, 1958), but the most 
northerly confirmed site for 7. violacea Bory de Saint- Vincent (as 
T. riekei Bischoff) in North America, until our discovery of this 
species in the Hudson River, was southern Texas (Bischoff, 1965; 
Sheath et al., 1993). We also found in the field the presumptive 
alternate life history (Chantransia) phase of this alga. Based on 
environmental conditions present at the New York site, a greater 
ecological and geographical range for this species and for the genus 
is indicated. 


330 Rhodora [Vol. 97 


STUDY SITE 


Thorea violacea was found growing attached to pebbles and 
small rocks at depths of ca. 30-70 cm in the Hudson River 
(73°35'W, 43°11'23”N), Saratoga County, Northumberland 
Township, New York State. The site is near the eastern shore of 
Thompson Island, downstream of a low head dam. The area 
containing Thorea was less than 1000 m?. A portion of the site 
was shaded late in the day by trees onshore. Thorea grew in open 
areas between patches of vascular macrophytes, and only where 
the river bed consisted of pebbles and small rocks. Farther from 
shore, vascular macrophytes letel 


Dp y occupied the substratum; 
farther downstream, the river bed was mud rather than pebbles. 
Thorea was first discovered on 15 September 1994. Additional 
collections were made and environmental parameters were mea- 
sured on 29 September 1994. Surface current velocity over the 
Thorea patch was 63 cm sec~!. Specific conductance was 119 uS 
cm~': pH was 6.9; alkalinity was 0.39 meq 1~!. The site was 
revisited 30 June 1995, but no Thorea was detected. However, 
large thalli again were abundant when the site was visited 10 
September 1995. Attempts to locate Thorea at three other rela- 
tively high-flow sites in the Hudson River and six selected trib- 
utaries in the area were unsuccessful. 

Our measurements of conductivity and pH fell within the rang- 
es recorded in unpublished US Geological Survey (USGS) data 
collected at Fort Edward, located 9 km upstream of our study 
site: conductance there was 44-136 uS cm~', and pH was 6.6- 
7.7. Total P was 0.1-0.5 mg 1~'!, and NO,” was 0.2-0.9 mg I"!. 
During a nine-year period, mean annual discharge ranged from 
110-203 m3 sec~'!, with a nine-year mean of 143 m? sec~!. Water 
temperature ranged from 0-26°C; all twenty readings from No- 
vember to March were less than 10°C. Five readings of water 
temperature during September, the month during which Thorea 
was collected, ranged from 15-19°C. 


RESULTS AND DISCUSSION 


Thalli of the Thorea phase of T. violacea varied considerably 
in the amount of macroscopically visible branching present. Most 
thalli were profusely branched near the base, resulting in many 
long axes without a distinct main axis (Figure 1); other thalli had 


1995] Pueschel et al.— Thorea violacea 331 


Figure 1. Herbarium specimen of monospore-bearing Thorea phase of Thorea 
violacea from the Hudson River. B hing is extensi the base of the thallus, 
but few branches (arrow) emerge along the length of the long axes. Scale bar equals 
0.1 m. 


Figure 2. Light micrograph of a section of one long axis showing the distal 
portion of assimilatory filaments that radiate from the axis (axis not shown, but 
see Figure 3). Note that assimilatory filaments have non-clavate apical cells and 
occasional branches (arrows). Scale bar equals 100 um. 


332 Rhodora [Vol. 97 


only one or a few long axes. Secondary branching, which was 
initiated below the growing apices, occurred sparsely along the 
length of the long axes (Figure | 

Each long axis was composed of numerous filaments whose 
branching was visible only by microscopic examination (Figures 
2, 3). The core of the axis was composed of highly branched, 
interwoven, unpigmented filaments encased in a common mu- 
cilaginous matrix (Figure 3). At the outer edge of this colorless 
medullary zone, medullary filaments produced branches, not en- 
cased in mucilage, that were highly pigmented, occasionally 
branched (Figure 2), and oriented perpendicular to the axis (Figure 
3). This fringe of lateral, determinate, assimilatory filaments sur- 
rounding the medulla formed a photosynthetic cortex. Monospo- 
rangia were produced on short filaments interspersed among the 
bases of the lateral assimilatory filaments (Figure 3). 

The identity of our specimens as 7. violacea was determined 
using key criteria proposed by Sheath etal. (1993): generally sparse 
secondary branching of long axes (Figure |) and assimilatory fil- 
aments occasionally branched and with non-clavate apical cells 
(Figure 2). This identification was confirmed by Sheath (pers. 
comm.). Some thalli exceeded | m in length, but most were be- 
tween 0.5-0.9 m long. Measurements of thallus features described 
below were made on living thalli. Long axes were 2-3 mm in 
diameter. Most of the diameter consisted of the lateral assimi- 
latory filaments that arose from the tough, resilient, colorless, 
medullary zone, 0.5 mm in diameter. Cells of assimilatory fila- 
ments were 8-10 wm in diameter (Figure 2). Basal cells of the 
assimilatory filaments were 20-30 um long (Figure 3). Within a 
filament, cell length increased progressively over a span of several 
cells, until cells attained a relatively uniform length of 42-50 um. 
Although many assimilatory filaments were unbranched, a single 
branch was common. Lateral branches emerged from the distal 
portion of the cell bearing the branch, just below the crosswall 
(Figure 2). Filaments with as many as four branches were found. 
Apical cells of assimilatory filaments were cylindrical except for 
their rounded tips; they did not taper and were not markedly 
longer than other cells (Figure 2). 

The only reproductive structures observed were sporangia (Fig- 
ure 3), each producing a single spore. All thalli examined micro- 
scopically bore a profusion of such sporangia. These were pre- 
sumed to be monosporangia, because they were borne terminally 


1995] Pueschel et al.— Thorea violacea 333 


Figure 3. Light recht of a section of one long axis showing the outermost 
portion of the medullary zone and the proximal portion of the zone of pigmented 
cells. einai o medullary ee (MF) give rise to long assimilatory filaments 


(AF) an that uce monosporangia (M). Note empty sporangial 
walls a discharged monosporangia. Scale bar equals 50 um 


or laterally on short filaments, a few cells long, that formed a 
pigmented layer around the medulla (Figure 3). By contrast, car- 
posporangia of Thorea (not observed) are produced on branching, 
multicellular filaments that originate from carpogonia, which in 
turn are borne on short filaments (Y oshizaki, 1986; Necchi, 1987). 
The total length of monosporangial filaments in our specimens 
was about 50-60 um. Mature sporangia were 15-18 um in di- 
ameter and about 25 um long. The walls of discharged sporangia 
persisted (Figure 3), and thus served as a marker of post-discharge 
evelopment of the subtending filament. In some filaments, the 
cell subtending a discharged sporangium divided to produce a 
new sporangium that developed within the loose confines of the 
old wall. Alternatively, the new apical cell was vegetative and, 
one or two cell divisions later, a terminal sporangium was again 
formed. 
Specimens from the Hudson River had several features that 
differed from Bischoffs (1965) descriptions of 7. riekei, which 
Sheath et al. (1993) placed in synonymy with 7. violacea. Our 


334 Rhodora [Vol. 97 


alga was blue-green in color, and it maintained this color upon 
drying. The long assimilatory filaments moved freely, and they 
were not encased in mucilage. Bischoff (1965) reported that the 
Texas specimens were generally rust-colored, and he demonstrat- 
ed that the assimilatory filaments were embedded in mucilage. 
However, these features may be environmentally variable; Sheath 
(1984) noted that mucilage was not abundant in his specimens 
from the same location in Texas. Bischoff (1965) also reported 
that assimilatory filaments were unbranched and that apical cells 
were tapered and twice as long as other cells. Tapered apical cells 
of Thorea specimens from Texas were also noted by Hedgcock 
and Hunter (1899), whereas branching assimilatory filaments and 
cylindrical apical cells (Figure 2) were present in thalli from the 
Hudson River. Further study is needed to establish whether these 
differences are taxonomically significant. 

Red algal thalli of considerably different morphology were pres- 
ent on rocks in the same part of the river bed. These were pre- 
sumed to represent the Chantransia stage of the Thorea life his- 
tory. Some thalli were visible only by microscopic examination 
of the rock surface, but a few formed grey tufts up to 9 mm in 
length (Figure 4). When viewed by light microscopy, chloroplast 
color and morphology were identical to that of the 7horea-phase 
thalli, and the pattern of lateral branch emergence was the same 
(Figure 5). Cells were 13-16 wm in diameter and 32-40 um in 
length. Small thalli were sparsely branched, but larger ones had 
increasingly frequent branching towards the apices of the fila- 
ments and bore monosporangia. Two species of bluish-green Au- 
douinella are recognized to occur in freshwaters of North America, 
and it is possible that one or both might represent life-history 
stages of the Batrachospermales (Necchi et al., 1993), but neither 
of these closely resembles the presumptive Chantransia stage pres- 
ent at our site. 

Ours is the first report of field-collected Chantransia stage of 
Thorea in North America. As in other Batrachospermales (Sheath, 
1984), thalli of the gametophyte morphology of 7horea are known 
to arise directly from the Chantransia phase (Swale, 1962; Bis- 
choff, 1965). Monospores from both Chantransia and Thorea 
phases are known to grow into Chantransia (Swale, 1962), and 
carpospores produced on the Thorea phase are presumed to grow 
into Chantransia (Necchi, 1987). However, none of these links 


1995] Pueschel et al.— Thorea violacea 530 


Figure 4. Putative reread stage of Thorea ie ae extensive 
ae oe bar equa 

Figure 5. t eek coud branching of putative Chantransia stage. 
Small lateral appendages (arrow) are epiphytic, unicellular blue-green algae. Scale 
bar equals 100 um 


336 Rhodora [Vol. 97 


between the presumed stages of 7. violacea in the Hudson River 
has been established. 

In addition to the significant extension of the geographic range 
of Thorea in North America, the presence of this alga in the 
Hudson River represents a considerable extension of environ- 
mental conditions under which members of this family are known 
to grow. As summarized by Sheath et al. (1993), the Thoreaceae 
are found in streams with mean specific conductance of approx- 
imately 300 uS cm~! (range 180-500 uS cm~'), pH of 8.0 (range 
7.5-8.3), current velocity of 30 cm sec! (range 9-99 cm sec~'), 
width of 6.5 m (range 1.5—12 m), and temperature of 20°C (range 
15-24°C). Our data and that of the USGS show significantly lower 
conductance (44-136 uS cm~'), lower pH (6.6—-7.7), greater than 
average current velocity (63 cm sec~!), lower than average tem- 
perature during the month of collection (16.8°C), and larger stream 
size (the Hudson River dam just upstream of the Thorea patch 
is 120 m wide). 

Unlike many freshwater algae, the red algae do not form thick- 
walled spores capable of surviving adverse conditions for long 
periods (Sheath and Hambrook, 1990). Therefore, annual ex- 
tremes in environmental conditions, and not just the conditions 
during the time at which the macroscopic phase develops, must 
have an effect on the range of distribution of freshwater red algae. 
Although Thorea occurs at higher latitudes in England (Swale, 
1962, 1963) and Germany (Schmidle, 1896: Schnepf, 1992), the 
Hudson River site undoubtedly experiences lower temperatures 
for longer periods. Water temperature measured by the USGS 
near the Thorea site was below 10°C from November to March. 

The absence of 7. vio/acea at our site during early summer is 
interesting, because Thorea typically grows in geographic regions 
of generally higher water temperatures and greater light intensity 
than occur in the Hudson River during early summer. Although 
the Chantransia stage was not detected during the early summer 
collection, it could have escaped detection if present as creeping 
filaments. The Chantransia stage is believed to be perennial (Sheath 
and Hambrook, 1990). Alternatively, this 7horea patch may be 
repopulated by spores from upstream populations of either phase. 
Spores would give rise to Chantransia, which in turn would pro- 
duce the Thorea phase directly. The growth of T. violacea must 
have occurred in the interval between the end of June, at which 
time it was not detected, and mid-September, by which time some 


1995] Pueschel et al.— 7horea violacea 337 


achieved | m in length. Rapid growth is typical of Thorea. Swale 
(1962) found that 7. hispida (as T. ramosissima) grew as much 
as 44 cm in one week. 

It remains to be determined how far north Thorea may grow. 
The proximity of the Hudson River site to the St. Lawrence River 
watershed, and the connection of the Hudson River to this wa- 
tershed through the Lake Champlain Canal, provides opportunity 
for Thorea to range considerably farther north. The newly rec- 
ognized broader range of Thorea’s environmental tolerances re- 
quires rethinking of the apparent warm-temperate distribution of 
the Thoreaceae in North America. The more common presence 
of the Thoreaceae in southern locales might simply reflect slow 
post-glacial recolonization of former ranges from small popula- 
tions in southern refugia, as was suggested by Sheath and Ham- 
brook (1990). Three reported localities of 7. hispida are of similar 
latitude and near the southern limits of glaciation (Hedgcock and 
Hunter, 1899; Tiffany and Britten, 1952; Hirsch and Palmer, 
1958), but the specimens of 7. vio/acea in the Hudson River are 
the first members of this family discovered deep within the gla- 
ciated region of North America. 


ACKNOWLEDGMENTS 


We are grateful to Dr. Robert Sheath for his helpful discussion 
of the Thoreaceae and his comments on the manuscript and to 
Ron Allen of the United States Geological Survey, Troy, New 
York, for providing physical and chemical data for the Hudson 
River. 


LITERATURE CITED 
BiscHorF, H. W. 1965. Thorea riekei sp. nov. and related species. J. Phycol. 1: 
111-117. 
Hepccock, G. G. AND A. A. HUNTER. 1899. Notes on Thorea. Bot. Gaz. 28: 
425-429 


Hirscu, A. AND C. M. PALMER. 1958. Some algae from the Ohio River drainage 
basin. Ohio J. Sci. 58: 375-382. 
NeEcculI, O., JR. 1987. Sexual reproduction in Thorea Bory (Rhodophyta, Tho- 
reaceae). Jap. J. Phycol. 35: 106-112. 
. SHEATH, AND K. M. Coe. 1993. Systematics of freshwater Au- 
dowinell (Acrochaetiaceae, Rhodophyta) in North America. 2. The bluish 
pecies. Arch. Hydrobiol./Suppl. Algol. Stud. 71: 13-21. 


338 Rhodora [Vol. 97 


SCHMIDLE, W. 1896. Untersuchungen iiber Thorea ramosissima Bory. Hedwigia 
35: 1-31 
ScHnePF, E. 1992. Electron microscopical studies of Thorea ramosissima (Tho- 
reaceae, Rhodophyta): taxonomic implications of Thorea pit plug ultrastruc- 
ture. Pl. Syst. Evol. 181: 233-244. 
SHEATH, G. 1984. Biology of freshwater red algae. Prog. Phycol. Res. 3: 89- 
157. 


AND J. A. HAMBROOK. 1990. Freshwater ecology, pp. 423-453. In: K. 

M. Cole and R. G. Sheath, eds. Biology of the Red Algae. Cambridge Uni- 

versity Press, Cambridge, England. 

,M.L. Vis, AND K. M. Cote. 1993, Distribution and systematics of the 

freshwater red algal family Thoreaceae in North America. Eur. J. Phycol. 
231-241 


SwALe, E. M. F. 1962. The eer and growth of Thorea ramosissima 
Bory. Ann. Bot. (London), N. S. 26: 105-116. 

; 1963. Notes on the a and ea of Thorea ramosissima 

Bory. J. Linn. Soc., Bot. 85: 429-434 + | p 

TiFFANY, L. H. AND M. E. Britron. 1952. the a of Illinois. University of 
Chicago Press, Chicago, IL 

YOSHIZAKI, M. 1986. The morphology and reproduction of Thorea okadai (Rho- 
dophyta). Phycologia 25: 476-48 1 


CML? .G: 5.) 4.8. 7, 

DEPARTMENT OF BIOLOGICAL SCIENCES 

STATE UNIVERSITY OF NEW YORK AT BINGHAMTON 
BINGHAMTON, NY 13902-6000 


! Present address: Pacific Estuarine Research Laboratory, Biology Department 
San Diego State University, San Diego, CA 92182 


RHODORA, Vol. 97, No. 892, pp. 339-349, 1995 


DISTRIBUTION AND 
CONSERVATION OF NANTUCKET SHADBUSH, 
AMELANCHIER NANTUCKETENSIS (ROSACEAE) 


ALISON C. DIBBLE AND CHRISTOPHER S. CAMPBELL 


ABSTRACT 


ena nantucketensis, Near es shadbush, thought to be restricted to 
coastal M d Long Island, New York, is now also known from Maine, 
seer inland on ‘New Hampshire, and Nova Scotia. Distribution 
is greater in part due to a wider circumscription of the species; A. nantucketensis 
intergrades with and therefore includes A. stolonifera f. micropetala. The 41 extant 
opulations reported here each consist of up to 13 individuals, grow in early 


be included in administrative rare plant lists. Protection of local populations would 
be best effected by controlling vegetation to maintain an early successional stage. 


Key Words: Amelanchier jalan endemism, Maine, Nantucket, Nova 
Scotia, shadbus 


INTRODUCTION 


Setting priorities for conservation is problematic in clonal plants, 
yet conservation at some level is appropriate for a recognizable 
morph that is apparently stable (Holsinger, 1992). This is es- 
pecially so if the morph has small, isolated populations and a 
clearly bounded geographic range, requires specialized habitat, 

epends on rare pollinators or dispersers, or supports a rare or 
unusually diverse fauna. Data regarding number and size of pop- 
ulations and ecology of species can be useful in determining 
whether a species is worthy of special consideration. 

Amelanchier, the shadbushes, contains as many as 17 species 
and three named hybrids of shrubs and trees in eastern North 
America (Phipps et al., 1990). Most of these are widespread and 
poorly defined due in part to facultative agamospermy (asexual 
seed production), polyploidy, and hybridization (Campbell and 
Dickinson, 1990; Campbell and Wright, in press). A northeastern 
North American species of 4melanchier that has been character- 
ized as a narrow endemic is 4. nantucketensis Bickn., Nantucket 
shadbush, originally known only from Massachusetts coastal is- 
lands (Bicknell, 1911; Sorrie, 1987) and later from Long Island, 
New York. 4melanchier nantucketensis was in Category 2 of the 


a9 


340 Rhodora [Vol. 97 


Federal Register of Endangered and Threatened Plant Species, 
indicating that formal listing required more knowledge about tax- 
onomic status, geographic distribution, or threats. On February 
28, 1996, the U.S. Department of the Interior Fish and Wildlife 
Service dropped Category 2 (Office of the Federal Register, 1996 
February 28) because of uneven data quality and insufficient re- 
sources. Thus 4melanchier nantucketensis no longer has Federal 
status. Amelanchier nantucketensis was considered to be a hybrid 
by Gleason and Cronquist (1991). Standley (1992) noted the lack 
of information regarding the biology of this taxon. 

Amelanchier nantucketensis has an unusual feature, andrope- 
taly; this is a term we propose for the condition of petals bearing 
one or usually two microsporangia. Andropetals replace normal, 
sterile petals, are often narrower and shorter than sterile petals, 
and are ivory rather than white (Dibble, 1995). Andropetaly is 
evident in at least some flowers on an individual; however, not 
all petals on a plant bear pollen. This condition is also found in 
A. stolonifera Wieg. f. micropetala (Robins.) Rehd. (Fernald, 1950), 
where it was termed “‘staminody” by Weatherby (1916). Tiny, 
pollen-bearing petals are found also, though rarely, in 4. obovalis 
(Michx.) Ashe, coastal shadbush. Andropetaly in Amelanchier is 
associated with a floral syndrome which is characterized by dense 
inflorescences, short pedicels, and small petals. A distinct com- 
ponent of the pollinator guild of solitary bees is attracted to an- 
dropetalous plants when compared to sympatric Amelanchier with 
normal petals. Possibly, the attraction to bees is not andropetaly 
so much as overall floral display in 4. nantucketensis (Dibble, 
1995). 

Amelanchier nantucketensis and A. stolonifera f. micropetala 
were thought to differ in the amount of pubsecence on the ovary 
Summit (Fernald, 1950), but this distinction is not consistent. 
Amelanchier stolonifera f. micropetala has therefore been merged 
into A. nantucketensis on the basis of six morphological characters 
(Dibble, 1995). 

Historic collections of this more broadly defined species indi- 
cate that its range extends mostly along the coastal plain from 
northern Virginia to Maine. The Massachusetts Natural Heritage 
Endangered Species Program (MNHESP) has records for an es- 
timated 30-40 small populations of Amelanchier nantucketensis 
on Nantucket, 12 on Martha’s Vineyard, and 11 populations in 
five counties of inland Massachusetts and Cape Cod. Most of 


1995] Dibble and Campbell— Nantucket shadbush 341 


these populations have been field-checked within the past 20 
years. In Connecticut, four populations in three counties are re- 
corded, but none are known to be extant. On Long Island, New 
York, at least three extant populations are known. Therefore, 
status as a narrow endemic 1s no longer appropriate for a species 
with such a large distribution, especially given nonspecificity of 
habitat in known locales. 

Our objectives were to document geographic distribution and 
population size for 4. nantucketensis. We also sought habitat 
features and ecological links or associations with rare pollinators 
that might be important to consider in planning a conservation 
strategy. 


MATERIALS AND METHODS 


This study involved a wide-ranging field survey of northeastern 
North American Amelanchier and included three visits to the 
type locality of A. nantucketensis. We collected and pressed spec- 
imens from 565 permanently marked plants including 62 indi- 
viduals of A. nantucketensis from 38 populations. Field surveys 
in New England, Maryland, New Jersey, New Brunswick, Nova 
Scotia, the Gaspé Peninsula of Quebec, and the west coast of 
Newfoundland from 1990-94 concentrated on roadsides, water- 
courses, meadows, and other disturbed habitats. We regard a 
population to be an aggregation of individuals separated from any 
other aggregation by at least 0.5 km. Because 4melanchier plants 
are clonal and therefore often occur in clumps of stems, we made 
the assumption that one clump represents one individual. We 
counted clumps per population and number of stems per clump, 
estimated or measured plant height, and noted habitat features. 
Identification of an assumed genetic individual was based on at 
least 2 m physical separation between clumps of stems; we have 
not found rhizomes to exceed a length of 50 cm. This 2 m criterion 
does not account for the possibility that several genotypes ma 
grow intermingled within a clump or that some “individuals” 
could be multicloned from agamospermy or fragments of former 
large clumps. 

We located 4melanchier plants in flower and returned to the 
same stems to collect mature leaves and developing or mature 
fruits. At each population of 4. nantucketensis we collected sam- 


342 Rhodora [Vol. 97 


ples from all accessible sympatric 4melanchier morphs and iden- 
tified them using several treatments of the genus (Wiegand, 1912; 
Jones, 1946; Fernald, 1950; Hinds, 1986: Gleason and Conquist, 
1991). To assess recruitment, we searched for seedlings among 
mature ramets of 4. nantucketensis. We noted evidence of her- 
bivory, fungal infection, insect visits to flowers, and activity of 
dispersers where present. 

We examined herbarium specimens at ACAD, BH, GH, NEBC, 
and NSAC (herbarium acronyms follow Holmgren et al., 1990) 
for 4. nantucketensis from the type locality and elsewhere. Dibble 
(1995) reported additional information regarding morphology, 
cytology, megasporogenesis, and pollination ecology. 


RESULTS AND DISCUSSION 


Distribution and endemism 


The geographic range of 4. nantucketensis is greater than pre- 
viously recognized; this increase is due to concerted effort in field 
surveys and to lumping 4. stolonifera f. micropetala into A. nan- 
tucketensis. The range extends from Great Falls, Maryland (Ashe, 
1944), along the coastal plain to Nova Scotia, with inland pop- 
ulations in northwestern Massachusetts and New Hampshire (both 
are montane habitats), and northwestern Maine ona high gravelly 
bank on the St. John River (Table 1, Figure 1). Amelanchier taxa 
exhibiting andropetaly were previously unknown in Canada 
(Scoggan, 1987). Amelanchier nantucketensis is apparently not 
limited to coastal plain communities. 

North American 4 me/anchier contains numerous taxa of nar- 
row geographic distribution. 4melanchier lucida Fern., for ex- 
ample, 1s limited to Nova Scotia (Fernald, 1948: Roland and 
Smith, 1969). Others include 4. amabilis Wieg. (Fernald, 1950), 
A. fernaldii Wieg. (Wiegand, 1912), A. huronensis Wieg., A. mu- 
cronata Nielsen, 4. interior Nielsen (Nielsen, 1939), A. gaspensis 
(Wieg.) Fern. & Weatherby, 4. florida Lindl., A. cusickii Fern., 
and A. basalticola Piper (Jones, 1946). In addition, we have iden- 
tified several series of populations that are morphologically dis- 
crete and narrowly distributed in Maine or in Maine and New 
Brunswick (Dibble, 1995). 

Some of these narrowly distributed taxa may be the product of 


1995] Dibble and Campbell— Nantucket shadbush 343 


Table 1. Extantand |} ic (known prior to 1976) populations sp eiaciaied 
nantucketensis that were visited for this study or reported to authors, and 
estimated population size for each within about 50 m radius. nies for addi- 
tional sites are kept at the Massachusetts Natural Heritage Endangered Species 
Program 


Numbe Popul 
of popu- .. ‘ 
lations aus 
_ s(n. of Year 
His- Ex-__ indivi- last 
State/Province County Town toric tant duals*) seen 
Connecticut New London’ Waterford 1 0 0 ca. 1975 
Maine Penobscot 1 13 1994 
Maine Penobscot Bradley 1 2 1993 
e Penobscot Eddington 1 1 1993 
aine Penobscot Milford 1 1 1996 
Maine Penobscot Old Town 1 6 1995 
Maine Penobscot Orono 3 11 1994 
Maine Hancock Ellsworth 3 5 1994 
Maine Hancock Bar Harbor 1 9 1993 
Maine Aroostook T12 R16 I 1 1 1995 
W 
Maine Lincoln Wiscasset 1 1 1991 
Maryland Montgomery Great Falls l 1 3 1993 
Massachusetts Nantucket Nantucket 1 18 44 most: 
1993 
Massachusetts Barnstable Harwich 1 1 1990 
Massachusetts Barnstable Hyannis 1 1 1991 
Massachusetts Berkshire N. Adams? 1 | 1987 
Massachusetts Dukes Edgartown 1 9 1992 
New Hampshire — Carroll N. Conway 1 10 1996 
New York Nassau Montauk 1 9 1992 
New York Nassau Shinnecock 1 4 1992 
Nova Scotia Shelburne Jordan Bay 1 _9 1992 
Totals 21 4 41 141 
a An “individual” may consist of one o d dered a discrete 


clump of stems separated by >2 m pee other ‘clumps 

> Specimens identified as a stolonifera f. qiconeaie Pamela B. Weatherbee, Wil- 
liamstown, Berkshire Co., MA, 9 May 1987 No. 772 and 8 July 1987 No. 998; 
Summit, Pine Cobble Mt.—elev. ca. 800 m). 


the interplay between hybridization and agamospermy. Amelan- 
chier hybrids with at least one agamospermous parent are also 
agamospermous in the two cases that have been studied (Weber 
and Campbell, 1989; Campbell and Wright, in press). Agamo- 
spermy perpetuates hybrids and thus generates microspecies, se- 


344 Rhodora [Vol. 97 


750 | \ 70° | Teen 
N 
I 


a N. B. ~ 


Martha's Vineyard 


Figure 1. Geographic distribution of Amelanchier nantucketensis, with ap- 
proximate extant (dots) and historic but presumed extirpated (star) locations. 
Multiple populations are represented by a single dot in some cases. 


ries of populations derived from uniparental reproduction (Grant, 
1981). 

Asa tetraploid agamosperm, Ame/lanchier nantucketensis (Dib- 
ble, 1995) may itself be of hybrid origin, as suggested by Gleason 
and Cronquist (1991); however, unambiguous identification of 
parental taxa has not been made. Amelanchier nantucketensis may 
participate in microspecies formation because it frequently grows 
with other Amelanchier, and in most cases there is overlap in 
flowering times. 4melanchier species and hybrids that grow with 
A, nantucketensis include A. canadensis, A. stolonifera, A. laevis 
Wieg., A. cf. humilis Wieg., and less commonly, 4. arborea (Michx. 
f.) Fern., A. bartramiana (Tausch) M. Roemer, A. x neglecta Eg- 
glest., 4. x intermedia Spach, and a morph we tentatively identify 
as A. cf. humilis x A. laevis. The above list includes the first 
record of A. stolonifera on Nantucket. We have observed apparent 


1995] Dibble and Campbell— Nantucket shadbush 345 


morphological intermediates between A. nantucketensis and A. 
stolonifera in some populations on Nantucket and in Orono, Maine 
(Dibble, 1995). The Orono population of A. stolonifera includes 
tetraploid agamosperms. 

For most of the supposed narrowly distributed endemic Ame- 
lanchier species, taxonomic status and geographic distribution are 
uncertain and, in our experience, not fully determinable from 
herbarium specimens. Ideally, the status of these entities will be 
ascertained from detailed study of morphological and molecular 
variation, hybridization, ploidy, and reproductive biology. To 
date, extensive study of supposed rare Amelanchier taxa other 
than A. nantucketensis is lacking. 


Population number and size 


We visited 38 populations of A. nantucketensis for this study; 
locations of three others were brought to our attention (Table 1). 
Fourteen of these were unknown prior to 1990. Based on records 
kept by MNHESP, the total number of extant populations could 
exceed 80; the actual total depends on whether one considers 
occurrences within gene flow (1.e., pollen and seed dispersal) dis- 
tance of others, as on Nantucket, to be populations or subpopula- 
tions. Determining the number of individuals at these populations 
is difficult in this rhizomatous shrub. Numerous stems (or ap- 
parent ramets) arise within 10-50 cm of each other, and presumed 
genets may be up to 10 m across. All populations we visited are 
small, usually with one or a few individuals each consisting of 
numerous stems. Populations with up to 13 individuals are known 
from Maine and Nantucket. One Maine population has since been 
extirpated by a construction project. 

We found no small seedlings among the densely arranged ma- 
ture ramets of A. nantucketensis. However, we observed young 
ramets growing at the center as well as near the edges of large 
clumps. These young ramets usually have large leaves compared 
to older ramets, suggesting they are sprouts from rhizomes rather 
than seedlings. 


Habitat 


The diversity of habitats occupied by A. nantucketensis is much 
greater than previously known. It grows in sand or loam or on 


346 Rhodora [Vol. 97 


ledges, along roadsides, river and stream shores, in coastal heaths, 
and under powerlines. Rarely, it grows in early- to mid-succes- 
sional forests dominated by Quercus rubra L., Populus spp., Betula 
spp., Pinus strobus L., and Picea spp., always within about 10 m 
of an opening. Occurrence in such habitats could depend on suc- 
cession of the site. As with many Amelanchier species, A. nan- 
tucketensis has the greatest density of stems and the most prolific 
flowering and fruiting in sunny sites. Soil drainage may influence 
plant height, with mesic soils supporting taller plants. 


Conservation aspects 


Because A. nantucketensis 1s associated with ecotones and such 
habitats are often occupied by a higher diversity of organisms 
than are adjacent areas under closed canopies, the potential for 
ecological links between this colonizing plant and various op- 
portunistic animals is relatively high. We found no rare arthro- 
pods associated with A. nantucketensis, but a multitude of in- 
vertebrates use this species as forage, breeding habitat, or as a 
domicile. We did not find any obligately species-specific inver- 
tebrates, and we observed similar associations among these an- 
imals and various other Ame/anchier species (Dibble, 1995). 

Examples of some animals associated with Amelanchier nan- 
tucketensis include more than 40 species of generalist solitary 
bees, which are probably the primary pollinators of this species 
(Dibble, 1995). Flowers are also visited by sawflies (Tenthredi- 
nidae), bee flies (Bombyliidae), flower flies (Syrphidae), moths 
and butterflies (Lepidoptera), and various beetles including der- 
mestids (Dermestidae). Insect herbivores include leaf cutter bees 
(Megachile spp.), which use circular pieces of leaf to line their 
nests; weevils (Curculionidae); scale insects (Coccidae); aphids 
(Aphididae); and leaf miners (Agromyzidae). Ants (Formicidae), 
perhaps attracted by nectar, are ubiquitous on flowers and de- 
veloping fruits; they eat styles, stamens, petals, sepals, and carpels. 
Crab spiders (Thomisidae) are camouflaged on flowers and cap- 
ture visiting, small, solitary bees. Weevils mate on the plants 
during anthesis; then the females oviposit into the hypanthium, 
later the larvae consume developing Amelanchier embryos. 
Wounds created by weevils provide one entry for Gymnospor- 
angium rust, the alternate host for which is Juniperus. This rust 
disfigures the fruits so that birds avoid these when foraging on 


1995] Dibble and Campbell— Nantucket shadbush 347 


the plants, but viable seeds can develop within spermogonia-laden 
fruits (Dibble, unpubl. data). The fruits probably fall to the ground 
near the parent plant, which would allow seeds to germinate in a 
microsite to which the genotype 1s well-adapted. Dispersal is by 
birds and various mammals, and there is potential that seeds 
consumed on different but nearby species could be deposited 
together, germinate, and grow intermingled, adding to confusion 
of field observers. 

All these associations are integral parts of a fully functioning 
ecosystem and represent a microcosm of ecological interactions 
in and around a host plant species. For conservation purposes, 
no link is dispensable given that we do not fully understand re- 
lationships between organisms. Although these associates of A. 
nantucketensis are mostly common, widely-distributed general- 
ists, it is unknown whether this shadbush species would be ad- 
versely affected by loss or reduction in numbers of any of these 
animal and fungal species. 

Listing at the state rather than Federal level is recommended 
for A. nantucketensis because it has more than 60 recently verified 
populations and a broad geographic range compared with narrow 
endemics listed as Federally Endangered. Although some state 
lists, such as Maine’s (Dibble et al., 1989), provide no regulatory 
protection, recognition of rarity within the state could increase 
the likelihood that some 4. nantucketensis populations will be 
protected voluntarily. Small populations and occurrence in hab- 
itats subject to frequent human disturbance or succession make 
A, nantucketensis susceptible to population extinction if devel- 
opment destroys habitat or if forests succeed open areas. In pro- 
tected populations, woody vegetation should be monitored every 
2-5 years and controlled by mowing or burning to maintain the 
early successional habitat conducive to persistence of A. nan- 
tucketensis. 


ACKNOWLEDGMENTS 


Funding was provided by the Massachusetts Natural Heritage 
Endangered Species Program, the Maine Agricultural and For- 
estry Experiment Station, and a New England Botanical Club 
Graduate Student Award to ACD. We are grateful to the Nan- 
tucket Conservation Foundation, Massachusetts Audubon, Aca- 
dia National Park, Great Falls National Park, and numerous pri- 


348 Rhodora [Vol. 97 


vate landowners for access to sites. We thank P. Dunwiddie, M. 
Golden, P. Somers, B. Sorrie, and W. Tiffney for guiding us to 
sites and providing background information. For field assistance 
we thank C. Bigelow, B. Coan, M. Coan, C. Dibble, J. Dudley, 
P. Dunwiddie, L. Gregory, A. Haines, B. Liles, P. Vitt, A. Walker, 
K. Walker, and W. Wright. We are grateful to Curators at ACAD, 
BH, GH, NEBC, and NSAC, and vegetation inspectors of the 
USDA and Agriculture Canada. Information about additional 
sites for A. nantucketensis was provided by B. Brooks, A. Haines, 
R. LeBlond, S. Rooney, P. Weatherbee, J. Weber, S. Young, and 
R. Zaremba. Comments from F. Drummond, S. Gawler, C. 
Greene, M. Hunter, Jr., G. DeWolf, Jr., and an anonymous re- 
viewer improved the paper. This is Maine Agricultural and For- 
estry Experiment Station Publication 2046. 


LITERATURE CITED 


AsHE, W. W. 1944. Notes on trees and shrubs in the vicinity of Washington. 
Bull. Torrey Bot. Club 46: 221-226. 

BICKNELL, E. P. 1911. The ferns and flowering plants of Nantucket. Part VIII. 
Bull. Torrey — Club 38: 447-460. 

CAMPBELL, C. S. AND T. A. Dickinson. 1990. Apomixis, patterns of mo 
logical Mates and species concepts in subfam. Maloideae Reo aee): 
Syst. — 15: 124-135. 

AND W. A. ee GHT. In press. Agamospermy, hybridization, and taxo- 
nomic complexity in Amelanchier (Rosaceae). Folia Geobot. Phytotax. 
Dipste, A.C. 1995. Conservation biology of Shadbush, 4melanchier (Rosaceae): 
evidence from systematics, population structure and reproductive ecology. 

Ph.D. dissertation. University of Maine, Oro E. 

——., C. S. CampBELL, H. R. Ty er, Jr., AND 7 Sr JOHN VICKERY. 1989. 
Maine’s official list of Endangered and Threatened Plants. Rhodora 91: 244— 
269 


FERNALD, M. L. 1948. A Nova Scotian dwarf shadbush. Rhodora 50: 49-51. 
——. 1950. Gray’s Manual of Botany, 8th ed. American Book Co., New York, 


NY. 
Geason, H. A. AND A. CRONQUIST. 1991. Manual of Vascular Plants of North- 
eastern United States and Adjacent Canada. New York Botanical Garden, 


onx, NY. 
GRANT, V. 1981. Plant Speciation, 2nd ed. Columbia University Press, New 


Y 

Hinps, H. 1986. Flora of New Brunswick. Primrose Press, Fredericton, New 
Brunswick. 

HoLMGREN, P. K., N. H. HOLMGREN, AND L. C. BARNETT, eds. 1990. Index 
Herbariorum, Part I: The Herbaria of the World, 8th ed. New York Botanical 
Garden, Bronx, NY. 

HOLsINGER, K. 1992. Setting priorities for regional plant conservation programs. 
Rhodora 94: 243-257. 


1995] Dibble and Campbell— Nantucket shadbush 349 


Jones, G. N. 1946. American species of Amelanchier. Illinois Biol. Monogr. 
20(2 


NIELSEN, E.L. 1939. A taxonomic study of the genus Amelanchier in Minnesota. 
Amer. Midl. Naturalist 22: 160-206. 

OFFICE OF THE anans cas 1996 February 28. Endangered and threatened 
species, | xa; Proposed Rule. Federal Register 61(40): 7596— 
761 ine 


Purpps, J. B., K. R. ROBERTSON, P. G. SMITH, AND J. R. ROHRER. 1990. Achecklist 
of the subfamily Maloideae (Rosaceae). Canad. J. Bot. 68: 2209-2269. 
RoLanpb, A. E. AND E. C. SmitH. 1969. The Flora of Nova Scotia. Nova Scotia 
Museum, Halifax, Nova Scotia 

ScoGGAN, H.J. 1987. The Flora of Canada. Part 3: Dictoyledoneae (Saururaceae 
to Violaceae). National Museums of Natural Sciences Publications in Botany, 
No. 7(3), Ottawa, Ontario. 

Sorrig, B. A. 1987. Notes on the rare flora of Massachusetts. Rhodora 89: 113- 
196. 


STANDLEY, L. A. 1992. Taxonomic issues in rare species protection. Rhodora 
94: 218-242. 
WEATHERBY, C. A. 1916. Staminody of the petals in Amelanchier. Rhodora 18: 


WEBER, J. AND C. S. CAMPBELL. 1989. Breeding system of a hybrid between a 
sexual and an apomictic — os Amelanchier, Shadbush (Rosaceae, Ma- 
loideae). Amer. J. Bot. 7 

WIEGAND, K. M. 1912. The genus eee in Eastern North America. 
Rhodora 14: 117-161. 


DEPARTMENT OF PLANT BIOLOGY AND PATHOLOGY 
5722 DEERING HALL 
UNIVERSITY OF MAINE, ORONO, ME 04469-5722 


A line drawing of Amelanchier nantucketensis appears on the cover of Rhodora 
95, 1993. 


RHODORA, Vol. 97, No. 892, pp. 350-356, 1995 


ERIOGONUM CODIUM 
(POLYGONACEAE: ERIOGONOIDEAE), A NEW 
SPECIES FROM SOUTHCENTRAL WASHINGTON 


JAMES L. REVEAL, FLORENCE CAPLOW, 
AND KATHRYN BECK 


ABSTRACT 


Eriogonum codium (Polygonaceae: Eriogonoideae), a low, matted, cespitose 
perennial with tomentose flowers and achenes found on the Hanford Nuclear 
Reservation (the Hanford Site), Benton Co., Washington, U.S.A., is described as 
a new species. It belongs to the same group of matted perennials in the subgenus 
Eucycla as E. chrysops but has a cymose-umbellate inflorescence similar to E. 
cusickii. The tomentose flowers and achenes readily distinguish the new species 
from all of its near relatives. 


Key Words: Polygonaceae, Eriogonoideae, cll codium, floristics, rare 
plants, Hanford Nuclear Reservati 


The genus Eriogonum Michx. (Polygonaceae Juss.: Eriogono- 
ideae Arn.), a taxon of more than 240 species widely distributed 
in temperate North America, 1s divided into eight subgenera (Re- 
veal, 1989). The most speciose is subg. Eucycla (Nutt.) Post & 
Kuntze, containing nearly half of the known species. The majority 
of cespitose to pulvinate-matted perennials with captitate or cy- 
mose-umbellate inflorescences belong to the sect. Capitata Torr. 
& A. Gray. The type of this section is £. pauciflorum Pursh, an 
atypical member in the sense that it has densely tomentose flowers 
whereas all other species have glabrous or glandular pubescent 
flowers. Related to this section is another group of species that 
have villous or pilose flowers; and, in one of these, E. shockleyi 
S. Wats., the ovaries and achenes are densely tomentose. Prior 
to the discovery of E. codium, this was the only cespitose species 
of the subg. Eucyc/a with this latter condition. 


Eriogonum codium Reveal, Caplow & K. Beck, sp. nov. (Figure 
1). A E. cusickii floribus et achenis pubescentibus differt. 


Low, cespitose, herbaceous perennials, the aboveground woody 
caudex system forming highly branched mats (1) 2—7 (9) dm across, 
arising from a stout, woody taproot; leaves basal, persistent, the 
leaf-blades oblanceolate to elliptic, (5) 6-12 mm long, 3-6 mm 


350 


1995] Reveal et al.—Eriogonum codium 351 


wide, densely white-tomentose on both surfaces, only slightly less 
so above in some, the apex mostly acute, the base cuneate, the 
margin entire and plane, the petiole short, 2-8 (10) mm long, 
tomentose, the petiole-base elongate-triangular, 1.5—3 (4) mm long, 
0.8-1.6 (2) mm wide, densely tomentose abaxially, sparsely so to 
glabrous adaxially; flowering stems scapose, erect, 2-9 cm long, 
tomentose, often brittle; inflorescences cymose-umbellate to cy- 
mose, divided 2-3 times, up to 2 cm high and 4 cm across, 
tomentose, each typically with a centrally positioned, pedunculate 
involucre and two lateral, short, dichotomous branches; bracts 
scale-like, ternate, triangular, 1-2.5 mm long, 1-2 mm wide, to- 
mentose without, glabrous within, connate at the base; peduncles 
restricted to the lower node, (1.5) 2-5 (7) mm long, glabrous, 
faintly winged; involucres solitary, appearing congested in early 
anthesis, turbinate-campanulate, membranaceous, 2.5-4 mm long, 
2-2.5 mm wide, tomentose to floccose without, glabrous within, 
the 5 sharply acute teeth 0.8-1.2 (1.5) mm long, the bractlets 
linear, 2-3 mm long, with marginal glands and scattered teeth, 
the pedicels 2-3.5 mm long, glabrous; flowers lemon-yellow with 
greenish midribs and yellowish-green bases, 2-3 mm long, mod- 
erately (at anthesis) to thinly (in fruit) tomentose without, sparsely 
so and minutely glandular along the midrib within, the tepals 
essentially similar, broadly oblong often with the apex emarginate, 
1.2-1.5 mm wide, those of the inner whorl narrower (0.9-1.2 
mm) than those of the outer whorl and with rounded apices, 
united about ' the length of the flower; stamens slightly exserted, 
2.5-3.5 mm long, the filaments sparsely hairy at the very base, 
the anthers yellow, 0.5-0.6 mm long, oblong; achenes trigonous, 
light brown, 2.5—3 mm long, sparsely tomentose, the globose base 
tapering to a long, 3-angled beak. 

TYPE: U.S.A. Washington: U.S. Department of Energy’s Han- 
ford Site, on the northern edge of Umtanum Ridge west of Wash- 
ington Highway 24 overlooking the Columbia River about 38 air 
miles northwest of Richland, Benton Co., on volcanic soil asso- 
ciated with Grayia spinosa (Hook.) Mog., Artemisia tridentata 
Nutt., Salvia dorrii (Kellogg) Abrams, Hesperostipa comata (Trin. 
& Rupr.) Barkworth, and Pseudoroegneria spicata (Pursh) A. Love 
at about 350 m elev. in sec. 13, T.13N., R.24E., 27 Jun 1995, 
Reveal, Caplow & Sackschewsky 7484. (Holotype: US; Isotypes: 
BM, BRY, CAS, COLO, GH, K, MARY, MO, NY, RM, RSA, 
TEX, UC, WS, WTU, and elsewhere). 


532 Rhodora [Vol. 97 


pe NOS 


a 
Palanan 


224 ras 

yeas Dopey 
5 eae : 
tn saat 
2a gee 


ais ey, ail 
aa D285 set 
1 


xe ; a ae 
MIG 


Figure 1. Eriogonum codium showing (a) general habit (x 0.75) with details 
of the (b) cymose inflorescences showing the pedunculate, centrally positioned 


5- 
~] 


1995] Reveal et al.—Eriogonum codium 353 


ETYMOLOGY: From kodion Gr., the diminutive of koas, fleece, 
referring to the woolly nature of the flowers and achenes. 

The basalt desert buckwheat, Eriogonum codium, was found 
on the Hanford Nuclear Reservation (the Hanford Site) during 
the botanical survey portion of the Hanford Biodiversity Project 
sponsored by the U.S. Department of Energy and the Washington 
State Chapter of The Nature Conservancy. The new species is 
related to those of sect. Capitata noted for their highly restricted 
distributions and generally minor morphological differences. The 
majority of these species occur in the Intermountain West and 
are differentiated into two major groups, those with distinctly 
rigid involucres (e.g., E. ochrocephalum S. Wats.) and those with 
membranaceous involucres (e.g., &. kingii Torr. & A. Gray). The 
vast majority of plants in these two groups have capitate inflo- 
rescences with the involucres in clusters atop a short scape (e.g., 
E. chrysops Rydb. and E. capistratum Reveal). Only a few have 
a branched inflorescence (e.g., E. novonudum M. E. Peck and E. 
cusickii M. E. Jones of Oregon, and the Sierra Nevada endemic, 
E. breedlovei (J. T. Howell) Reveal). Nonetheless, none of these 
plants has tomentose flowers or achenes. The cymose-umbellate 
inflorescences of the new species readily differentiate it from the 
capitate Great Plains species, FE. pauciflorum, and the sparsely 
tomentose flowers and achenes of FE. codium are unlike the more 
densely pilose flowers and densely tomentose achenes found in 
E. shockleyi. As currently understood, Eriogonum codium does 
not appear to be closely related to any of the known species in 
and related to those of section Capitata, except that it belongs 
with those species related to E. chrysops found mainly in Oregon, 
Idaho, and Nevada. 

Eriogonum codium is highly restricted in its distribution. The 
only known population occurs at elevations ranging between 340 
and 400 m on flat to gently sloping substrates at the top edge of 
the steep, north-facing basalt cliffs of Umtanum Ridge overlook- 
ing the Columbia River. Approximately 5000 plants grow inter- 
ruptedly in a narrow band 2.5 km long and generally less than 30 
m wide. The plants occur only on the ridge which 1s subject to 


— 

involucre on a short branch with three bracts (left) and an entire involucre with 
numerous pedicels, one of which bears a flower, and (c) an achene with fine hairs 
on the beak (both x 15). 


354 Rhodora [Vol. 97 


the strong winds of the Columbia River canyon. Precipitation 
averages less than 15 cm annually. 

The new species occurs exclusively on the exposed basaltic flow 
top of the Lolo Flow (mid-Miocene, 14 mybp) of the Priest Rapids 
Member of the Wanapum Formation (Goff, 1981), with the gaps 
in the population correlating with the absence of exposed flow 
top. The flow top material typically has high porosity and high 
permeability. Weathering has transformed the top into a surface 
comprised of pebble to gravel sized pieces of vesicular volcanic 
material. The average chemical composition of the Lolo basalt 
flow differs from the other Columbia Basin basalt formations in 
having high concentrations of calcium oxide, iron oxide, mag- 
nesium oxide, phosphorous pentoxide, titanium oxide, and the 
mineral olivine (Reidel & Fecht, 1981). In addition, the Priest 
Rapids Basalt flows have unusually high water-holding capacity 
(Reidel, pers. comm., 1995). It is not known if the strong asso- 
ciation of Eriogonum codium with the Lolo Flow is related to the 
particular chemical composition of the flow. 

The basalt desert buckwheat occurs in the shrub-steppe vege- 
tation zone which Is the primary vegetation of the Columbia Basin 
(Franklin and Dyrness, 1973). Perhaps as a result of the chemistry 
of the substrate, vegetational cover in the vicinity of Eriogonum 
codium is low when compared with other shrub-steppe sites. 
Common perennial associates include Artemisia tridentata, Gray- 
ia spinosa, Krascheninnikovia lanata (Pursh) A. D. J. Meeuse & 
Smit, E. sphaerocephalum Dougl. ex Benth., Salvia dorrii, Hes- 
perostipa comata, Pseudoroegneria spicata, Poa sandbergii Vasey, 
Sphaeralcea munroana (Dougl. ex Lindl.) Spach ex A. Gray, As- 
tragalus caricinus (M. E. Jones) Barneby, and Balsamorhiza car- 
evana A, Gray. Common annual associates include Bromus tec- 
torum L., Phacelia linearis (Pursh) Holz., Gilia leptomeria A. 
Gray, G. inconspicua (Sm.) Sweet var. sinuata (Dougl. ex Benth.) 
A. Gray, Camissonia minor (A. Nels.) Raven, Mentzelia albicaulis 
(Dougl. ex Hook.) Dougl. ex Torr. & A. Gray, and Cryptantha 
pterocarya (Torr.) E. Greene. Adjacent areas on Umtanum Ridge 
support populations of two local, rare endemics, Lomatium tub- 
erosum Hoov. and Astragalus columbianus Barneby. 

The cover of Eriogonum codium is higher than that of most 
other species within its habitat. There is a wide range of size and 
age classes within the population. In 1995 (a wet year), plants 
were in flower from early May through late August. Seed set, 


1995] Reveal et al.—Eriogonum codium 355 


however, was low with less than five percent of the flowers pro- 
ducing mature, viable seed in 1995. Seed germination has not 
been evaluated. 

Umtanum Ridge is currently managed by the U.S. Department 
of Energy. The Hanford Site has large areas of relatively undis- 
turbed, high-quality shrub-steppe vegetation due to the cessation 
of virtually all agricultural and grazing activities when the Site 
was established in 1943. Therefore, the population of Eriogonum 
codium does not appear to be threatened by human activity at 
this time. Change in ownership or changes in the Department of 
Energy’s management policies could allow public access, the use 
of off-road vehicles, and/or livestock grazing. If public access is 
allowed in this area, the population could also be affected by 
petrified wood collectors. Major deposits of petrified wood occur 
in the substrate throughout the area. Petrified wood is often col- 
lected with the aid of bulldozers and other heavy machinery. Any 
change of ownership or management of the Umtanum Ridge area 
could potentially threaten the viability of this highly restricted 
species. 


ACKNOWLEDGMENTS 


We would like to thank the U.S. Department of Energy for their 
financial and logistical support of the Hanford Biodiversity Pro- 
ject and the Washington State Chapter of The Nature Conser- 
vancy for their vision and commitment to the study of biodi- 
versity on the Hanford Site. Dr. S. P. Reidel provided us with 
details on the geology of the area. The illustration was prepared 
by Ms. Dolly Baker. This is Scientific Article A-7829, Contri- 
bution No. 9156, of the Maryland Agricultural Experiment Sta- 
tion and Cooperative Extension Service. 


LITERATURE CITED 


FRANKLIN, J. F. AND C. T. Dyrness. 1973. Pree oo - of Oregon and 
Washington. Gen. Techn. Rep. P.N.W., U.S. F . 8. 

Gorr, F. E. 1981. Preliminary Geology oe ae m on Ridge, South 
Central Washington. RHO-BWI-C-21, Rockwell Hanford Operations, Rich- 


REIDEL, S. P. AND K. R. Fecut. 1981. Wanapum and Saddle mountains basalts 
of the Cold Creek Syncline Area, pp. 3(1-16). Ja: C. W. Myers and S. M 


356 Rhodora [Vol. 97 


Price, eds. Subsurface Geology of the Cold Creek Syncline. RHO-BWI-ST- 
14. Rockwell Hanford Operations, Richland, WA. 

REVEAL, J. L. 1989. A checklist of the Eriogonoideae (Polygonaceae). Phytologia 
66: 266-294. 


Hae ae 

DEPARTMENT OF PLANT BIOLOGY 
UNIVERSITY OF MARYLAND 

COLLEGE PARK, MARYLAND 20742-5815 


FAG, ee Te Bs: 

CALYPSO CONSULTING 

3963 SQUALICUM LAKE ROAD 
BELLINGHAM, WA 98226 


RHODORA, Vol. 97, No. 892, pp. 357-374, 1995 


VEGETATION, BROWSING, AND 
SITE FACTORS AS DETERMINANTS OF CANADA YEW 
(TAXUS CANADENSIS) DISTRIBUTION IN 
CENTRAL NEW HAMPSHIRE 


JOHN J. STACHOWICZ AND TABER D. ALLISON 


ABSTRACT 


Taxus canadensis (Canada yew) in Hanover, New Hampshire, was studied to 
determine the relative importance of site factors, deer browsing, and past land- 
se in determining its distribution. Data indicate that the species’ distribution is 
strongly linked to habitats with high soil moisture and low solar radiation (e.g., 
concave and/or north-facing slopes). Taxus canadensis is less abundant on slopes 
and in habitats with southern exposure. Deer browsing 1s greater on south-facing 
than on north-facing slopes and higher under deciduous than under coniferous 
canopies. Species associations do not indicate a restriction of Taxus canadensis 
to sites of particular (canopy) successional status, but land-use data suggest that 
it may be absent ue areas which have been recently logged. Taxus canadensis 
is more abundant in Hanover than in sites in north-central Massachusetts, sug- 
gesting that Scie habitat for this species may increase with latitude. 


Key Words: Taxus canadensis, land-use history, a site relationships, deer 
browsing, New Hampshire, Canada y 


INTRODUCTION 


The genus 7axus L. comprises eight species worldwide. It is 
well known for its horticultural importance and has a long eco- 
nomic and mythical association with humans (e.g., Hartzell, 1991). 
Public awareness of this genus has recently been increased by the 
discovery that taxol, a chemical constituent of all parts of the yew 
plant, is an effective anti-tumor agent (National Cancer Institute, 
1992). As a result, there was a major increase in the exploitation 
of yew species, particularly for their bark. Recent development 
of synthetic sources of taxol (e.g.. Wheeler and Hehnen, 1993; 
Nicolau et al., 1994) has reduced the pressure on wild Taxus 
species, but the threat of harvesting wild populations has high- 
lighted the lack of detailed information on the ecological status 
of all species in the genus. 

Taxus canadensis Marsh. (Canada yew) 1s a spreading, ever- 
green shrub of cool, moist forests of the northeastern United States 
and southeastern Canada. The southern limit of its range in New 
England extends roughly in a line from Newburyport, Massachu- 
setts, through Providence, Rhode Island, to New Haven, Con- 


357 


358 Rhodora [Vol. 97 


necticut. We conducted a survey of the distribution and abun- 
dance of this species in Hanover, New Hampshire, in order to 
determine its ecological status and, specifically, to assess which 
environmental factors are important in influencing its current 
distribution. 

It has been suggested that logging can have a negative effect on 
T. canadensis distribution (Nichols, 1913; Hosley and Ziebarth, 
1935). New England’s forests were heavily logged during the 18th 
and 19th centuries when much of the land was cleared for agri- 
culture, and subsequent reforestation beginning in the mid-19th 
century has resulted in a mosaic of primary (areas cut but never 
cleared for agriculture) and secondary woodlands (Foster, 1993). 
Vegetation surveys performed in the 1930’s indicated that 7. 
canadensis was associated with primary woodlands in Petersham, 
Massachusetts, but the species was too rare for this conclusion to 
have been stated with statistical confidence (Whitney and Foster, 
1988; Whitney, 1991). Anecdotal reports also have suggested that 
winter browsing by white-tailed deer (Odocoileus virginianus Zim- 
mermann) and moose (A/ces alces americana Clinton) can locally 
extirpate 7. canadensis populations (Allison, 1990, and references 
cited therein). 

A review of the literature on forest community composition of 
New England and New York (a list of papers is available from 
the authors) and preliminary surveys of areas supporting 7. can- 
adensis populations indicated that most sites supporting this spe- 
cies were located at or near the base of slopes with a northerly or 
westerly aspect. While such surveys of forest composition can 
yield information on where 7. canadensis is present, they are not 
useful in determining where the species is absent. This is signif- 
icant, as these surveys are often biased to old-growth stands where 
T. canadensis may be more likely to occur. 

We sampled 7. canadensis in the town of Hanover, New Hamp- 
shire, in order to characterize its distribution in relation to specific 
environmental variables and associated vegetation. Based on the 
results of our preliminary survey and literature search, we eval- 
uated our data with reference to the hypotheses that 7. canadensis: 
(1) 1s restricted to sites which have experienced low browsing 
pressure in the past and/or present; (2) is limited by the avail- 
ability of microhabitats with suitable moisture regimes, growing 
particularly on north-facing slopes which typically receive low 
insolation and have high humidity (Geiger, 1965); and (3) has 


1995] Stachowicz and Allison—Canada yew 359 


been eliminated by extensive forest clearing and is currently con- 
fined to primary woodlands. 

These hypotheses may be confounded. For example, deer browse 
T. canadensis only in the winter when north-facing slopes may 
be less accessible to deer due to rugged topography and a more 
persistent snowpack. North-facing slopes are also less likely to 
have been cleared for farming and are thus frequently associated 
with primary woodlands. While complete evaluation of these and 
additional hypotheses requires experimental approaches beyond 
the scope of our study, we report here the results of our survey 
and an initial evaluation of these hypotheses based on the survey 

ata. 


METHODS 


Study Site 


The climate of Hanover, New Hampshire, 1s characterized by 
long, cold winters with heavy snowfalls and relatively short, cool 
summers. Precipitation is distributed evenly throughout the year 
and the average number of frost free days is 134. Elevation ranges 
from slightly less than 120 m to 712 m. The land in and around 
the town was extensively cleared during the agricultural period, 
with a maximum of 65% of the total land area in farms in 1880; 
over 40% of this land was actually tilled (United States Census 
Bureau, 1883). The majority of undeveloped land in the town is 
now forested although recent agricultural census data suggest that 
about five percent of the town’s area remains in cultivation (Unit- 
ed States Census Bureau, 1989), most of which is in the floodplain 
of the Connecticut River or the level plains of the highlands. 

In early successional sites, forested lands are dominated by 
white pine (Pinus strobus L.), gray birch (Betula populifolia Marsh.), 
quaking aspen (Populus tremuloides Michx.) and pin cherry (Pru- 
nus pensylvanica L. f.). Northern red oak (Quercus rubra L.), red 
maple (Acer rubrum L.), and white pine dominate mid-succes- 
sional forests. Late successional lower elevation and stream valley 
forests are dominated by northern hardwoods species such as 
yellow birch (B. lutea Michx. f.), sugar maple (A. saccharum 
Marsh.), and beech (Fagus grandifolia Ehrh.). Hemlock (T’suga 
canadensis (L.) Carr.) is also important at lower elevations and 
is replaced by red spruce (Picea rubens Sarg.) and balsam fir (Abies 


360 Rhodora [Vol. 97 


balsamea (L.) Mill.) at higher elevations. Nomenclature follows 
Gray’s Manual of Botany (Fernald, 1950). 


Vegetation Survey 


Twenty study points were selected randomly using Universal 
Trans-Mercator (UTM) coordinates from areas zoned for forestry 
and recreation by the town of Hanover. Wetlands were excluded 
from the sample. The study was limited to these zones in order 
to assure forest cover in all randomly selected survey sites and 
to avoid sampling near residences. According to 1989 Town of 
Hanover zoning maps, approximately 44% of the town’s area, 
and the majority of its undeveloped land, is included in this 
designation. At each point a transect was established in a compass 
direction chosen randomly. We located 10 plots of 10 m radius 
at randomly selected intervals between 20 and 100 m apart along 
the transect. Twenty transects (200 plots) were sampled. Vege- 
tation data were limited to recording presence of all species in 
three strata (canopy, understory, and forest floor). 

These measures of community composition may not be the 
most precise method of determining association; however, simple 
presence-absence data may be the best indicator of a forest’s age 
(Rackham, 1986; Whitney and Foster, 1988). We sampled in this 
way in order to survey a large number of sample plots that could 
be easily compared with other available data, since much of the 
information on 7. canadensis occurrence and associated forest 
composition gathered from the literature was available only in 
presence/absence form. Environmental parameters were also re- 
corded for each plot including elevation, percent slope, aspect, 
slope shape (concave, planar, or convex), and slope position (base, 
lower, middle, upper third, or ridge). Land-use history data were 
collected from historical documents where available, but for most 
sites detailed historical records were unavailable, and land-use 
history was inferred using field observations and local (anecdotal) 
sources. 


Sampling of Taxus canadensis Populations 


Density, biomass, growth, and browse damage of 7. canadensis 
were determined in a separate survey of sites known to support 


1995] Stachowicz and Allison—Canada yew 361 


T. canadensis populations, including sites other than those in our 
vegetation survey. These sites were selected according to the fol- 
lowing criteria: (1) each contained a population of 7. canadensis 
which extended over an area of at least 1000 m?; (2) preliminary 
data indicated that at least a portion of the site contained plots 
with a density of 0.25 shoots per m?; and (3) sites were selected 
to ensure representation of habitats with different canopy vege- 
tation and slope aspect. At each of these sites, we marked 30 
sampling points along a 100 m transect. The starting point of the 
transect, its direction, and the spacing of the sampling points were 
all determined using a random number table. At each point, the 
distance to the nearest 7. canadensis shoot and the distance from 
that shoot to its nearest neighbor were measured. These distances 
were used to estimate density according to the formula derived 
by Batcheler (1971): log(density) = log(n/m(Zr,7)) — (0.1416 — 
0.1613(2r,/=r,)); where r, 1s the distance from the point to the 
nearest shoot, r, 1s the distance from that shoot to its nearest 
neighbor, and n is the number of shoots sampled. This method 
avoids the biases inherent in using random quadrat methods to 
sample aclumped distribution like that ofa vegetatively spreading 
species such as 7. canadensis (Batcheler, 1971). 

Because 7. canadensis spreads vegetatively, the delineation of 
an individual (genet) is impossible without tracing the root sys- 
tem. Therefore, we used the shoot (ramet) as our unit of mea- 
surement. A ramet was defined as a single emergent stem of the 
plant, disconnected from neighboring stems at a soil depth of 2 
cm. For each shoot, we measured the basal diameter and the 
diameter of all branches greater than or equal to 3.0 mm on these 
shoots. These values were converted to grams dry weight by com- 
parison with a standard curve of biomass versus branch diameter 
determined previously for Taxus canadensis [mass = 0.685 — 
1.02 (diameter) + 0.539 (diameter)*; r* = 0.99 (J. J. Stachowicz, 
unpubl. data)]. Biomass per shoot was multiplied by the density 
of each population (shoots per m7’) to obtain an estimate of bio- 
mass for each population. 

Annual growth increments on 7. canadensis are indicated by 
terminal bud scars on the stem and are often apparent as far back 
as five years. To estimate aboveground growth for each popula- 
tion, we measured the stem diameter just above the third terminal 
bud scar (Figure |) from the apex of ten randomly selected branch- 
es on each shoot used for biomass estimates. Growth measure- 


362 Rhodora [Vol. 97 


ments were made in late fall, at the end of the growing season, 
so the measurement should represent three full years of growth. 
By measuring growth for the most recent three-year period, and 
then dividing by three to convert to an annual rate, we attempted 
to minimize the effects of year-to-year variability on our growth 
estimates. The annual diameter increment was converted to bio- 
mass using the standard curve of diameter versus mass, and an 
average annual biomass increment per branch was calculated. 
Total annual growth was estimated by multiplying the average 
annual biomass increment per branch by the total number of 
branches per shoot. 

We estimated removal rates by deer (in grams of tissue per 
year) by measuring the diameter of all branches browsed within 
the past year at the point of removal, estimating biomass of re- 
moved tissue from the diameter-biomass relationship described 
above, and summing over all browsed branches for the entire 
shoot. Both annual growth and removal rates per shoot were 
multiplied by density to obtain areal estimates per m7’. 

The nearest canopy and understory individuals to each shoot 
were recorded and environmental data taken as previously de- 
scribed. The canopy at each site was classified based on whether 
the majority of the nearest canopy individuals recorded were 
coniferous or deciduous. At all of the locations we sampled, either 
conifers or deciduous trees greatly outnumbered the other, so 
determination of canopy type was unambiguous and no mixed 
canopy areas were sampled. 


DATA ANALYSIS 


Relationships between environmental variables and 7. cana- 
densis presence were examined using Chi-Square tests (Sokal and 
Rohlf, 1981), testing the null hypothesis that the frequency of 
occurrence was the same in all categories. Biomass, annual growth, 
and deer removal rates for different habitats were evaluated by 
ANOVA with site characteristics as fixed effects. Data on species 
presence-absence in the 200 sample plots were analyzed by di- 
visive Classification using TWINSPAN (Hill, 1979a) to determine 
associations among different species and 7. canadensis. Species 
presence-absence data were also analyzed by correspondence 
analysis (CA), an ordination technique, using DECORANA (Hill, 


1995] Stachowicz and Allison—Canada yew 363 


~<—— terminal bud 


| 
growth this | 
season | 


growth one 
season ago 


« {st terminal bud scar 


a. Se ~«<——— 2nd terminal bud scar 


growth tw 
seasons ee 


= 


Figure |. Branch tip of Taxus canadensis showing bud scars marking annual 
growth increments. Measurements of growth were made over the most recent 
three-year period by measuring the branch diameter just above the third terminal 
bud scar (see text). 


——_—— 3rd terminal bud scar 


1979b). This analysis represents plots or species graphically, on 
a two (or greater) dimensional set of axes, placing those with the 
most similar patterns of occurrence closest together, and thereby 
allowing relationships between many samples or species to be 
easily recognized. To test if the axis values generated by DE- 
CORANA were predictive of 7. canadensis presence, we corre- 
lated them, along with environmental parameters (slope, aspect, 
elevation, etc.), with 7. canadensis presence by multiple logistic 
regression (Kleinbaum et al., 1988). This analysis is similar to 
standard multiple linear regression, with the exception that the 
dependent variable (in this case 7. canadensis occurrence) is di- 
chotomous (yew is either present or absent). 


364 Rhodora [Vol. 97 


RESULTS AND DISCUSSION 


Taxus canadensis occurred in 37 out of 200 plots sampled 
(18.5%). The only understory or forest floor species that occurred 
more frequently was striped maple (Acer pensylvanicum L., 43.5%). 
Several other forest floor species were similar in frequency to 7. 
canadensis. These include Rubus sp., 17%; Lycopodium obscurum 
L., 16.5%; Lycopodium clavatum L., 14.5%; Lycopodium anno- 
tinum L., 14.5%; Aster divaricatus L., 13.5%; and Mitchella repens 
L., 13%. The frequency of occurrence of 7. canadensis in Hanover 
is higher than that in north-central Massachusetts. Taxus cana- 
densis occurred in seven of 74 (9.5%) plots in the Harvard Forest 
and surrounding area of Petersham, Massachusetts (Gerhardt, 
1993), and a survey of a 40 township region in north-central 
Massachusetts found 7. canadensis in 17 out of 360 plots, less 
than five percent (C. Mabry and D. R. Foster, unpubl. data). These 
studies used 20 m x 20 m plots in which presence/absence of all 
vascular plant species was recorded, so general methods are com- 
parable to this study. The land-use history and topography of 
these central-Massachusetts sites are similar to those of Hanover, 
New Hampshire, and both receive ample precipitation through- 
out the year. However, Hanover has a shorter growing season and 
more of the precipitation comes as snowfall than at the Harvard 
Forest, suggesting that the primary difference between these sites 
is related to temperature and latitude. 

Presence-absence data (Table 1), collected during our vegeta- 
tion survey, show that 7. canadensis occurs more frequently on 
sites with concave slopes, at slope bases, at low elevations, on 
north- and east-facing slopes, or on level ground (slope < 10%) 
than in locations with other physiographic characteristics. The 
sampling of 7. canadensis populations showed that, where this 
species grew, density was greatest in level areas (0.83 shoots/m7?) 
and lowest on south-facing slopes (0.23 shoots/m?). Shoots grow- 
ing on north-facing slopes have the greatest standing crop or bio- 
mass (F = 4.52; P < 0.001) and growth rates or production equal 
to shoots on level ground but greater than those on south slopes 
(F = 6.09; P < 0.001; Figure 2A). Production of 7. canadensis 
did not differ under deciduous versus coniferous canopies, al- 
though biomass was slightly greater under deciduous canopies 
(Figure 2B). 

The distribution of 7. canadensis in upland forests of Hanover 


1995] Stachowicz and Allison—Canada yew 365 


Table 1. Occurrence of Taxus canadensis by habitat characteristics for 200 
plots in Hanover, New Hampshire. 


Taxus present Taxus absent Frequency 
Slope Shape 
Concave 21 28 42.9 
Planar 15 65 18.8 
Convex 1 70 1.4 
x? = 33.04 df = 2; P < 0.0001 
Slope Position 
Slope Base 16 16 50.0 
Lower Third 20 32 38.5 
Middle Third 0 57 0 
Upper Third | 50 2.0 
Ridge Top 0) 8 0 
x? = 49.09 df = 4; P < 0.0001 
Elevation (meters) 
100-200 10 0 100 
200-300 6 17 26.1 
300-400 19 45 29.7 
400-500 2 94 2.1 
>500 0 0 
x? = 68.99 df= 4; P < 0.0001 
Aspect 
North 23 59 28.0 
East 12 33 26.7 
South 0 31 0 
West 2 4.8 
x? = 19.24 df = 3; P = 0.0002 
Percent Slope 
15 26 36.6 
10-19° 8 53 13.1 
20-29° 7 44 13.7 
> 30° 7 4 14.9 
x? = 11.244 df = 3; P= 0.0105 
Time since abandonment! 
less than 30 years 0 42 0 
30-120 years 6 15 28.6 
>120 years or 
never cleared 31 106 22.6 
x? = 12.50 df = 2; P= 0.0019 


' Time since the land was last cultivated or intensively cut. 


Estimated dry weight (g) 


Rhodora [Vol. 97 


Slope aspect 

north 
level 
[_] south 


- aaa! 
Biomass (g/m?) Production (g/m? /yr) Removal signee 


Canopy type 


DB deciduous 
ao:. 0 
= i coniferous 
xo) 
= 
aly 
Go 64 
xo) 
oD 4 
fa) 
£ 
% 37 
LU 
O oe 
Biomass (g/m4) Production (g/m? /yr) Removal (g/m? /yr) 
Figure 2. Taxus canadensis biomass ducti l th), and removal 


by deer per m? (mean + | SE) at sites wiih various aspects (A) and canopy types 
(B). Lines above bars indicate means that do not differ significantly at P > 0.05 
by ANOVA. 


1995] Stachowicz and Allison—Canada yew 367 


is linked strongly to environmental variables, such as slope shape, 
slope position, elevation, aspect, and percent slope, that affect 
soil moisture and relative humidity. This is consistent with the 
hypothesis of habitat preference as a determinant of the distri- 
bution of this species. In north-temperate locations, north-facing 
slopes are generally cooler and have higher relative humidity and 
soil moisture than comparable south-facing slopes (Geiger, 1965; 
Oke, 1978). Cold air drainage can also account for a similar pat- 
tern of increased moisture and decreased temperature on the lower 
portions of a slope compared to upper regions (Oke, 1978). These 
facts are in accord with what has been observed anecdotally for 
T. canadensis in the southern portion of its range. Nichols (1913), 
for example, noted that in Connecticut, 7. canadensis was much 
more common in “lower ground.” In north-central Massachu- 
setts, it is not uncommon to find 7. canadensis at the boundary 
between upland and wetland or on steep rock ledges where “‘mois- 
ture conditions are favorable” (Hosley and Ziebarth, 1935; T. D. 
Allison and J. J. Stachowicz, unpubl. data). Populations in south- 
ern Minnesota and northern Iowa are typically found at the talus 
base of north-facing limestone cliffs (T. D. Allison, pers. obs.). 
Our survey data also indicated a potential impact of deer brows- 
ing on 7. canadensis distribution and an interaction of browsing 
with habitat. Browsing rate (removal) was higher under deciduous 
than coniferous canopies (Figure 2B). However, because a large 
proportion of the areas where 7. canadensis grew under a decid- 
uous canopy were on south-facing slopes, it is unclear whether 
this difference is attributable directly to canopy or if it is an 
indirect effect because deer are more likely to winter on south- 
facing slopes than on north-facing ones. Where T. canadensis 
occurred on south-facing slopes, browse damage was greatest (F 
= 4.081: P < 0.001), with removal nearly equaling annual growth 
or production (Figure 2A). North-facing slopes and level areas 
did not differ in removal rates. High deer populations have been 
implicated in limiting 7. canadensis distribution in Michigan, 
Wisconsin, and New York (e.g., Spiker, 1935; Beals et al., 1960), 
but similar reports have not been made for New England. Despite 
the fact that removal equals annual growth on south-facing slopes, 
Taxus canadensis may persist 1n these areas if there is high annual 
variability in browse pressure that could allow the population to 
recover following an episode of intense browsing. Thus deer 
browsing may not result in extirpation of local 7. canadensis 


368 Rhodora [Vol. 97 


Tilia americana’ 


Lycopodium oo 


Ace 

ae tana 
Ostrya virginia 

Lycopo: ail iavatum 
Picea ru 


Rubus sp 

saceharum 

Que 

Populus gender 

Prunus /+——__—_ 
inners cna 


Acer penyercum 

Picea 

palin: te 

Populus Sennicuee? 
Abies ees 

Beluie 


ae comuta 


Lycopodiu Sa 


Tsuga canadensis 

Quercus rubra Se 
Sa ivaicatus 

Taxus canadensis 

nee he i; 

Tsuga Canales —_ 
a ‘ 

sale ee 


Sp. 
Gaultheria Jie 


Figure 3. Results of TWINSPAN divisive classification of presence-absence 
data from survey of 200 plots (see text). Species with asterisks are canopy rep- 
resentatives; the same species without asterisks are understory representatives. 


populations, although our data indicate that in certain habitats 
browsing may retard expansion of existing populations by reduc- 
ing clonal expansion and seed production (Allison, 199 
Divisive classification Wea 3) indicated that 7. canadensis 
was most closely associated with mountain maple (Acer spicatum 
Lam.) and hobblebush (Viburnum alnifolium Marsh.). Quercus 


1995] Stachowicz and Allison—Canada yew 369 


rubra was the most closely associated canopy species (Figure 3). 
Red oak is generally common in mesic woodlands which were 
logged but not cleared for agriculture, a land-use category in which 
T. canadensis occurred frequently (Table |). The divisive clas- 
sification (Figure 3) also indicated that 7. canadensis is not closely 
linked with Acer saccharum, Tilia americana L., Betula lutea, or 
Fagus grandifolia although these species are characteristic of the 
northern hardwoods forest found in the lower elevations of Han- 
over. The apparent absence of any association of 7. canadensis 
with late-successional northern hardwoods and its association 
with red oak, a species that occurs in mid-successional forests 
around Hanover (see Study Site description), suggest that 7. can- 
adensis may not be restricted to late-successional forests. 

In Hanover, 7. canadensis was absent from areas which were 
logged within the past 30 years, but was not limited to primary 
woodlands (Table 1). One of the densest populations we observed 
was located on a floodplain at the base of a northwest-facing slope 
that was cultivated as recently as the 1880’s. We have also ob- 
served T. canadensis populations growing in old-field white pine 
stands in north-central Massachusetts. An effect of past land use 
on 7. canadensis distribution was indicated in a 1930’s survey of 
vegetation in Petersham, Massachusetts, where this species was 
restricted to primary woodlands (Whitney and Foster, 1988). Any 
such effect should diminish with time as 7. canadensis recolonizes 
suitable habitat, as has been suggested in more recent forest sur- 
veys of north-central Massachusetts (Gerhardt, 1993; C. Mabry 
and D. R. Foster, unpubl. data). Because our data on land-use 
history lacked detail on the time of abandonment or reforestation, 
we cannot determine the length of time required by 7. canadensis 
for recolonization, except that it is a minimum of 30 years. 

The divisive classification (Figure 3) suggests that 7. canadensis 
is associated with red oak, and thus not necessarily linked to old- 
growth forests, but the ordination of the vegetation survey data 
with DECORANA indicates an affinity for a more coniferous 
canopy. The ordination produced four vegetation axes, only two 
of which were significantly correlated with 7. canadensis occur- 
rence (Table 2). The first vegetation axis corresponds to an in- 
creasing importance of Abies balsamea and a decreasing impor- 
tance of Betula papyrifera Marsh. The second axis was positively 
correlated with low elevation coniferous forests near the base of 
north and northwest slopes, and negatively correlated with high 


370 Rhodora [Vol. 97 


RA e - A | 


Table 2. Results of multi ion analysis 
variables predict the outcome of the dichotomous dependent variable Taxus can- 
adensis presence (only variables with significant contribution to the final model 
are listed). Positive values of estimates indicate a positive association of 7. can- 
adensis presence with that variable (e.g., 7. canadensis occurs more frequently on 
slopes that are concave rather than convex or planar). Negative values of estimates 
indicate a negative association of 7. canadensis with that variable (e.g., T. can- 
adensis occurs less frequently with distance upslope). The magnitude of the es- 
timate suggests the relative importance of that factor in predicting the presence 
of T. canadensis. Tolerance statistics for all independent variables fall within the 
allowable limits, so no significant collinearities (P > 0.05) between predictors are 
apparent. CA = correspondence analysis. 


95% confidence 
interval of estimate 


Factor (indep. variable) Estimate Lower Upper G P 
Slope position — 1.282 —1.919 —0.646 49.31 0.0005 
Slope shape (concavity) 2.601 1.203 4.000 21.55 0.0005 
Slope aspect —0.011 —0.022 0.000 4.40 0.05 
CA Axis #1 0.647 0.234 1.061 8.77. 0.005 
CA Axis #2 1.101 0.423 1.780 12.05 0.001 


elevation deciduous forests on south-facing slopes. Neither of 
these axes give any evidence for an association of 7. canadensis 
with climax northern hardwood forests. In addition to these two 
vegetation axes, the final logistic regression model of 7. cana- 
densis presence-absence was influenced strongly by slope a 
aspect, and position (Table 2), environmental variables that 
fluence moisture levels (Geiger, 1965; Oke, 1978). The re 
indicates that Taxus canadensis occurs most frequently near the 
base and on the lower portions of north- and east-facing concave 
slopes and that its frequency is lower in plots higher upslope, on 
a convex-shaped slope, and with a more southerly aspect (Tables 


No consistent association between 7. canadensis and tree spe- 
cies was observed; the ordination, divisive classification, and lit- 
erature search each suggested different canopy associates. For 
example, the divisive classification (Figure 3) indicated that the 
most closely associated canopy species was Quercus rubra, while 
the ordination (axes scores from DECORANA) suggested asso- 
ciation with Abies balsamea and other conifers. Further, the lit- 
erature search suggested that in the northeastern United States, 
T. canadensis 1s associated with a northern hardwoods canopy of 


1995] Stachowicz and Allison—Canada yew 371 


Acer saccharum and Betula lutea (e.g., Heimburger, 1934; Egler, 
1940; Cline and Spurr, 1942: Stearns, 1951; Leak, 1973; Siccama, 
1974). In contrast, the association of 7. canadensis with other 
understory shrubs was much more consistent. Understory species 
were generally too rare to influence the ordination greatly, but 
both the divisive classification (TWINSPAN results, Figure 3) 
and limited understory literature (Heimburger, 1934; Egler, 1940) 
suggest that Viburnum alnifolium and Acer spicatum are closely 
associated with Taxus canadensis. Additionally, the divisive clas- 
sification (Figure 3) suggests that none of these three species is 
particularly closely associated with any canopy species. 

The differences in the strength and consistency of T. canadensis 
associations with understory shrubs vs. canopy trees may be the 
result of differences in the duration of the influence of past land- 
use in the different layers of vegetation. Long-lived canopy species 
in secondary woodlands, such as Quercus rubra, could be first 
generation individuals which have not been replaced by later 
successional northern hardwoods. Taxus canadensis may have 
been eliminated by forest clearing and plowing, but once refor- 
estation began, this species was eventually able to recolonize en- 
vironmentally suitable sites. Later-successional tree species, how- 
ever, have to wait for removal of the overstory containing earlier- 
successional species before succeeding to the canopy. This process 
is suggested by the closer association of Taxus canadensis with 
understory individuals of 7suwga canadensis than with canopy 
individuals (Figure 3). 


CONCLUSIONS 


As we predicted, 7. canadensis is most abundant and produc- 
tive on north-facing slopes. Although we did not find 7. cana- 
densis in any south-facing plots in our vegetation survey, it is not 
completely excluded from south-facing sites (we were able to find 
some populations for biomass, production, and removal esti- 
mates), but it is considerably less abundant and productive when 
present. Our survey was limited to upland forests, and we spe- 
cifically excluded open and wet habitats, such as old fields and 
swamps. Our own observations indicate that 7. canadensis does 
not grow in recently abandoned old fields, but may be found in 
swamps and on slopes surrounding bogs. Therefore, our conclu- 


372 Rhodora [Vol. 97 


sions about the ecological status of 7. canadensis are limited to 
upland habitats. 

There are clear differences in the frequency, abundance, and 
productivity of this species in the upland forests of Hanover: 
Taxus canadensis does best in sites with physiographic charac- 
teristics that promote high humidity and soil moisture. Our data 
indicate that 1t 1s most common at or near the base of concave 
slopes with northerly aspects and is often associated with Acer 
spicatum and Viburnum alnifolium. Taxus canadensis is not re- 
stricted to primary woodlands, although it is apparently excluded 
from recently logged sites. It 1s unclear whether deer browsing 
restricts 7. canadensis distribution, and the impact of browsing 
may be confounded with habitat because deer may be more likely 
to winter on south-facing slopes. However, browsing is greatest 
on south-facing slopes and may limit the spread of 7. canadensis 
in these habitats. Our data also suggest that 7. canadensis is more 
abundant in Hanover than in north-central Massachusetts. This 
may be related to the cooler climate of Hanover, which results 
in a wider range of habitats available for use by 7. canadensis. 


ACKNOWLEDGMENTS 


Field work was supported through a grant from the Presidential 
Scholars Research Program at Dartmouth College to J. Stacho- 
wicz. Comments from Laura Conkey, Mark McPeek, John Gil- 
bert, and several anonymous reviewers improved earlier versions 
of the manuscript. 


LITERATURE CITED 


Auuison, T. D. 1990. The influence of deer browsing on the reproductive biology 
of Canada yew (Taxus canadensis Marsh.). I. Direct effect on pollen, ovule, 
nd seed production. Oecologia 83: 523-529. 


BATCHELER, C. L. 1971. Estimation of density from a sample of joint-point and 
nearest- nennbor distances. 52: 702-709. 
BEALS, E. W., G. CoTTAM, AND R. J. VoGLi. 1960. Influence of deer on the 


vegetation of the Apostle oe Wisconsin. J. Wildlife Managem. 24: 68- 
80. 


CLINE, A. C. AND S. H. Spurr. 1942. The virgin upland forest of central New 
England. Bull. Harvard Forest 21. 
Ecuer, F.C. 1940. Berkshire Plateau vegetation. Ecol. Monogr. 10: 146-191. 


1995] Stachowicz and Allison—Canada yew a7) 


FERNALD, M. L. 1950. Gray’s Manual of Botany, 8th ed. American Book Com- 
pany, New York, NY. 

Foster, D.R. 1993. Land-use history and other forest transformations in central 
New England, pp. 91-110. In: M. J. McDonnell and S. T. A. Pickett, eds. 
Humans As Components of Ecosystems. Springer-Verlag, New York, NY 

GeiGer, R. 1965. The Climate Near the Ground. Harvard University Press, 
Cambridge, MA. 

GERHARDT, F. 1993. Physiographic and historical influences on forest compo- 
sition in central New England, USA. M.S. thesis, Harvard es 

HARTZELL, H., JR. 1991. The Yew Tree. Hulogosi Press, Eugene 

HEIMBURGER, C.C. 1934. Forest-type studies in the Adirondack jes: Cornell 
Univ. Agric. Exp. Sta. Mem. 165. 

Hii, M. O. 1979a. TWINSPAN—a FORTRAN program for arranging mul- 
tivariate data in an ordered two-way table by classification of individuals 
and attributes. Cornell University, Ithaca, NY. 

1979b. DECORANA—a FORTRAN program for detrended corre- 
pond lysis and recip l averaging. Cornell University, Ithaca, NY. 

Hos.ey, N. W. AND R. K. ZIEBARTH. 1935. Some winter relations of the white- 
tailed deer to the forests in north-central Massachusetts. Ecology 16: 535- 
553. 


KLEINBAUM, D. G., L. L. KUPPER, AND K. E. MULLER. 1988. Applied Regression 
Analysis and Other Multivariable Methods. Duxbury Press, Belmont, CA. 

LEAK, W. B. 1973. Species and structure of a virgin aaa hardwoods stand 
in New Hampshire. Res. Notes N. E., U.S. Forest Serv.: 

NATIONAL CANCER INSTITUTE. 1992. Taxol and oe ie -cancer Drugs. Of- 
fice of Cancer Communications, Washingto 

NicHois, G. E. 1913. The vegetation of ae virgin forests. Torreya 13: 
199-215. 

NIcoLau, K. C., Z. YANG, J. J. Lru, H. VeNo, P. G. NANTERMET, R. K. Guy, C. 
F, CLAIBORNE, J. REYNAUD, E. A. COULADOUROS, K. PAULVANNAN, AND E. 
J. SORENSEN. 1994. Total synthesis of taxol. Nature 367: 630-634. 

Oxe, T. R. 1978. Boundary Layer Climates. Methuen, London, England. 

RACKHAM, O. 1986. The History of the Countryside. J. M. Dent, London, 
E d. 


SICCAMA, T. G. 1974. Vegetation, soil, and climate on the Green Mountains of 
ermont. Ecol. Monogr. 44: 325-349. 
SOKAL, R. R. AND F. J. ROHLF. 1981. Biometry. W. H. Freeman, San Francisco, 
CA 


SPIKER, G. J. 1935. Some late winter and early spring observations on the white- 
tailed deer of the Adirondacks. Roosevelt Wildlife Bull. 6: 327-385. 

STEARNS, F. W. 1951. The composition of the sugar ee hemlock-yellow birch 
association of northern Wisconsin. Ecology 32: 2 65. 

UNITED STATES CENSUS BUREAU. 1883. Tenth census iF the United States, 1880. 

e 3: Agriculture. Government Printing Office, Washington, DC 

. 1989. Census of Agriculture, 1987. Volume 1: Geographic Area Series, 

Part 29: New Hampshire, State and County Data. U.S. Department of Com- 
merce, Maes DC. 

WHEELER, N. C. AND M. T. HEHNEN. 1993. Taxol: a study in technology com- 

seca ae J. Forest. (Washington) 91: 15-18. 


374 Rhodora [Vol. 97 


WHITNEY, G.G. 1991. Relation of plant species to substrate, landscape position, 
and aspect in north central Massachusetts. Canad. J. Forest Res. 21: 1245- 
52. 


AND D.R. Foster. 1988. Overst SLOrY itl d age as determinants 
of the understory flora of woods of central New England. J. Ecol. 76: 867- 
76. 


Fede ae 
DARTMOUTH COLLEGE 
HANOVER, NH 03755 


Tepe 
HARVARD FOREST 
PETERSHAM, MA 01366 


Present addresses 
' University of North Carolina at Chapel Hill, Institute of Marine Sciences, 
3431 Arendell St., Morehead City, NC 28557 
> Department of Forestry and Wildlife, University of Massachusetts, Amherst, 
MA 01003 


RHODORA, Vol. 97, No. 892, pp. 375-379, 1995 


GENTIANA NIVALIS L. (GENTIANACEAE) 
NEW TO QUEBEC 


N. DIGNARD, R. LALUMIERE, 
AND M. JULIEN 


Key Words: Gentiana nivalis, distribution, James Bay, Québec, Labrador 


During the course of ecological field work conducted on the 
northeast coast of James Bay in 1994, we collected a few speci- 
mens of a very small gentian, which proved to be Gentiana nivalis 
L. This arctic-alpine taxon is here reported as new to the flora of 
Québec. 

Gentiana nivalis is an amphi-atlantic species (Hultén, 1958), 
the range of which is mainly Eurasian (Iceland, Northern and 
Central Europe, and Asia Minor). Previous to this report, the 
North American distribution of the taxon was thought to be re- 
stricted to coastal areas of Greenland and Labrador (Rousseau, 
1974; Scoggan, 1979). The species has never been reported from 
Baffin Island, and Gillett (1963) states that this distribution pat- 
tern is due to a different postglacial history for Baffin Island from 
that of northern Labrador, rather than due to a collection gap. 
Most of the Labrador localities are north of 57°30’N, especially 
in the Kaumayjet and the Torngat mountains. Gentiana nivalis 1s 
listed as rare in Canada and in Labrador (Argus and Pryer, 1990). 
In 1951, the plant was discovered in the Saglek Pass (Rousseau 
1064 QUE) on the Québec-Labrador border. The specimen does 
not indicate on which side of the border the collection was made. 
This note is therefore the first valid report for the species in 
Québec. 

We collected Gentiana nivalis 17 km east of Pointe Louis-XIV 
(Dignard, Lalumiére & Julien 94-29 QUE) on July 23, 1994, in 
the tundra bordering the northeast coast of James Bay (Figure 1). 
This region, lying within the discontinuous permafrost zone, 1s 
characterized by the predominance of an arctic-alpine flora. The 
area is one of the southernmost Arctic outposts in North America 
(Ducruc et al., 1976), extending between 54°30'N and 54°45'N, 
along a strip five to 20 km wide parallel to the coast. The pop- 
ulation was found in a dry, grassy meadow on an alluvial terrace 
of fine sand along a small brook, four km from the sea at 
54°37'18”N-79°28'15”W, elevation ca. 20 m. Only a few indi- 
viduals were observed. Unfortunately, time did not allow us to 


31) 


376 Rhodora [Vol. 97 


80° 70° 60 
T T T 
60°F ( 60° 
\, 
*, 
O : 
\ 
) 
( 
\ 
ay 
55° Ji enol ' 455° 
Pointe Louis-XIV ipa ot 
( 
q 
\ ; 
of IN ey 
YY ( i? ee ae 
oe l eal 
\. j 
50°F e/ 50° 
ss 
| 
© ae ¢ 
rt 
/ 
/ 
fA / 
i 
J 
45° eee fo 145° 
i: < nN 1 
80° 70° 60° 


gure |. Distribution of Gentiana nivalis L. in Québec and Labrador (ex- 
panded from Gillett, 1963 and Rousseau, 1974). Square indicates the new locality. 


check for its presence in the surrounding area. The terrace veg- 
etation consisted of arctic-alpine and boreal species, the most 
representative being Achillea millefolium L. var. nigrescens E. 

ey., Antennaria pulcherrima (Hook.) Greene, Astragalus alpinus 
L., Betula nana L., Carex capillaris L., C. capitata L., Cerastium 
alpinum L. ssp. lanatum (Lam.) Aschers. & Graebn., Draba gla- 
bella Pursh var. glabella, Juncus arcticus Willd. ssp. arcticus, Ko- 
bresia eager (Wahlenb.) Mackenzie, Luzula multiflora 
Ehrh.) Lej. s./., Poa alpina L., P. arctica R. Br., Polygonum vi- 
viparum 4 Primula stricta Homieu., Salix argyrocarpa Anderss., 


1995] Dignard et al.—Gentiana nivalis 377 


S. brachycarpa Nutt., S. planifolia Pursh ssp. planifolia, Senecio 
pauciflorus Pursh, Tanacetum bipinnatum (L.) Schultz-Bip. ssp. 
huronense (Nutt.) Breitung, and Trimorpha elata (Hook.) Nesom. 
Located in a depression between two low ridges, the terrace is 
covered by a thick layer of snow during the winter, as indicated 
by the presence of the willow and dwarf birch community and 
by late snowbed species such as Potentilla tabernae-montani 
Aschers, and Taraxacum officinale G. H. Weber ex Wiggers ssp. 
ceratophorum (Ledeb.) Schinz ex Thellung. Nomenclature follows 
Kartesz (1994). 

This isolated population of Gentiana nivalis, more than 1000 
km away from the nearest Labrador locality, is the westernmost 
occurrence of the species in North America. This wide disjunction 
could be related to the postglacial history of the taxon and to 
narrow ecological requirements. The amphi-atlantic boreal Pla- 
tanthera albida (L.) Lindl. var. straminea (Fern.) Luer, known 
from northwest Newfoundland and disjunct in the Richmond 
Gulf and the Manitounouk Sound on the east coast of Hudson 
Bay, has a similar distribution (Payette and Lepage, 1977; De- 
shaye and Cayouette, 1988). Undercollection should also be con- 
sidered. Due to its small stature, the plant can be easily over- 
looked, especially when the flowers are immature. It can also be 
confused in the field with other superficially similar gentians (Gen- 
tianella amarella (L.) Boerner ssp. acuta (Michx.) J. Gillett, Gen- 
tianella tenella (Rottb.) Boerner and Lomatogonium rotatum (L.) 
Fries ex Fern.). The plicate, eciliate corolla tube, the expanding 
lobes, and the 4-5 parted cylindrical, purple-keeled calyx of G. 
nivalis at once distinguish this species from others. 

It is uncertain whether the new disjunct population of G. nivalis 
is the result of long distance dispersal or is relictual. Long distance 
dispersal, either by winds or birds, seems unlikely. The prevailing 
winds are westerly and bird migrations occur primarily in a north- 
south direction. In addition, the small seeds offer little nutritional 
value to birds although they could be ingested by accident. 

Now that Gentiana nivalis is known from the James Bay region 
and its habitat is better defined, other localities of the species 
might be discovered in the region and perhaps elsewhere in north- 
ern Québec. This was recently the case for the amphi-atlantic 
arctic-alpine Carex rufina Drej. Thought to be disjunct between 
Greenland, the Northwest Territories, and northern Manitoba 


378 Rhodora [Vol. 97 


(Porsild and Cody, 1980), it is presently known from at least six 
localities in the Ungava Peninsula, all restricted to volcanic rock 
formations (Blondeau and Cayouette, 1987). 

Because of its extreme rarity in Québec, Gentiana nivalis should 
be added to the list of the vulnerable or threatened vascular plants 
of this province (Lavoie, 1992). 


ACKNOWLEDGMENTS 


Field work was funded by the Société d’énergie de la Baie James 
and by the ministére des Ressources naturelles du Québec. We 
would like to express our gratitude to M. Blondeau and J. Cay- 
ouette for verifying the identity of the specimens and to M. Blon- 
deau, W. D. Hudson, Jr., and an anonymous reviewer for their 
comments. A. Reed kindly revised the English manuscript. 


LITERATURE CITED 


Arcus, G. W. AND K. M. Pryer. 1990. Les plantes vasculaires rares du Canada. 
Notre patrimoine naturel. Musée canadien de la nature, Ottawa, Ontario. 

BLONDEAU, M. AND J. CAYOUETTE. 1987. Extension d’aire dans la flore vasculaire 
du Nouveau-Québec. Naturaliste Canad. 114: 117-126 

DESHAYE, J. AND J. CAYOUETTE. 1988. La flore vasculaire des iles et de la 
presqu’ile de Manitounouk, baie d’Hudson: structure phytogéographique et 
interpretation bioclimatique. Provancheria 21. 

Ducruc, J.-P., R. ZARNOVICAN, V. GERARDIN, AND M. JURDANT. 1976. Les 
régions écologiques du territoire de la baie de James: caractéristiques dom- 
inantes de leur couvert végétal. Cah. de Géogr. de Québec 20: 365-392. 

GILLETT, J.M. 1963. The ao of Canada, Greenland and Alaska. Publ. Dep. 
Agric. Canad. 1180: 

Hu ten, E. 1958. The . atlantic plants and their phytogeographical con- 
nections. Almqvist and Wiksell, Stockholm, Sweden 

Kartesz, J. T. 1994. A Synonymized papas of die Vascular Flora of the 
United States, Canada, and Greenland. Vol. 2. Thesaurus, 2nd ed. Timber 


Lavorgr, G. 1992. Plantes vasculaires susceptibles d’étre désignées menacées ou 
vulnérables au Québec. Ministére de l’Environnement, Québec, Québec. 

PAYETTE, S. AND E. LepaGe. 1977. La flore vasculaire du golfe de Richmond, 
Nouveau-Québec. Provancheria 7. 

PorsiLp, A. E. AND W. J. Copy. 1980. Vascular Plants of Continental Northwest 
Territories, Canada. National Museum of Natural Sciences, Ottawa, Ontario. 

RoussEAu, C. 1974. Géographie floristique du Québec-Labrador. Distribution 
des principales espéces vasculaires. Presses Univ. Laval, Québec, Québec. 

ScoGGAN, H. J. 1979. The flora of Canada. Part 4. Publ. Bot. (Ottawa) 7: 1117- 
1711 


1995] Dignard et al.—Gentiana nivalis 


MINISTERE DES RESSOURCES NATURELLES 
DIRECTION DE LA RECHERCHE FORESTIERE 
2700, RUE EINSTEIN 

SAINTE-FOY, QUEBEC, CANADA G1P 3W8 


GROUPE ENVIRONNEMENT SHOONER INC. 
101-5355, BOULEVARD DES GRADINS 
QUEBEC, QUEBEC, CANADA G2J 1C8 


2256, RUE ASSELIN 
LONGUEUIL, QUEBEC, CANADA J4M 2M1 


379 


RHODORA, Vol. 97, No. 892, pp. 380-381, 1995 
BOOK REVIEW 


Levine, Carol. 1995. A Guide to Wildflowers in Winter: Her- 
baceous Plants of Northeastern North America. 329 pp. in- 
cluding 19 plates of black and white photographs and 161 
pages of line drawings. Yale University Press, P.O. Box 
209040, New Haven, CT 06520-9040. ISBN 0-300-06207-9 
($40.00 cloth), 0-300-06560-4 ($20.00 paper). 


The scope of this guide extends far beyond any existing similar 
guides to plants in winter, such as the excellent Weeds in Winter 
by Lauren Brown (1976), both in number of species covered and 
in the details with which they are described. Since the amount of 
information presented might be overwhelming to a beginner, the 
book is best suited to those who already have some familiarity 
with botanical terms, although all terms are well-defined. The 
wealth of detail will interest and challenge even the most expe- 
rienced field botanists. It is an education simply to flip through 
the illustrations, which immediately command one’s attention 
with their beauty and detail and encourage further investigations 
in the field. 

The book covers most commonly encountered herbaceous plants 
that have dried remnants in winter, including many grasses, var- 
ious members of the Cyperaceae (sedge family), not just Carex, 
and Juncaceae (rushes). Four hundred and nine taxa are illustrated 
and described. Some ferns and fern allies, some of which are 
evergreen anyway, are included so that everything is together in 
one volume, which is of a manageable size to take into the field. 
Even a few galls and leaf spots are noted. 

Every effort is made to clarify the meaning of each term or 
heading. Several headings appear under each taxon’s name cov- 
ering Key Impressions; Fruit; Leaves; Stem; Perennial, Annual, 
or Biennial; Habitat; and Range. For example, Key Impressions 
cover a range of characteristics, such as height, odor, shape, and 
inflorescence. There is an illustrated dichotomous key to help in 
keying out dried and crumpled specimens and an illustrated glos- 
sary in the back. 

The illustrations are excellent and capture both the appearance 
and the accurate details of a plant, which is quite a feat when it 
is usually lacking leaves and flowers and appears in an unchar- 
acteristically bent, broken, and dried brown condition. Details of 


380 


1995] Book Review 381 


the calyces of the Lamiaceae (mint family), achenes of the As- 
teraceae (daisy family), and fruits of the Apiaceae (carrot family) 
are beautifully rendered and would be appreciated by any bota- 
nist. Details of capsules, seed ornamentation, and shapes of un- 
derground stems make one want to take a second look at the 
infinite variety and curiosities of the plant kingdom. Forty-six 
black and white photographs show evergreen or commonly per- 
sisting basal rosettes. 

Plants are grouped together by obvious similar shapes, such as 
‘climbing vine” or ‘“‘stems and other parts with spines, barbs, or 
bristles.” The key uses these general descriptions, which can be- 
come somewhat unwieldy, and proceeds into more detailed char- 
acterizations. Asteraceae, ferns and fern allies, Poaceae (grass fam- 
ily), Cyperaceae, and Juncaceae are treated separately. The draw- 
ings explaining the differing structures of these last three groups 
would be a credit to any botany manual. In an attempt to make 
it possible to key out asters (Aster spp.) to the species, they are 
first grouped by habitat, then by characteristics. This may not be 
useful or possible in a guide of this sort, but it will keep one 
observing. 

This book will go a long way toward lightening that grim, brown 
season between November and March for both the interested 
amateur and for the professional field botanist who may be called 
upon to evaluate the vegetation of a parcel of land in the dead of 
winter. Like any good text, it provides information and stimulates 
one’s observation and desire to see more. 


LITERATURE CITED 
Brown, L. 1976. Weeds in Winter. Houghton Mifflin Company, Boston, MA. 
PAMELA B. WEATHERBEE 


236 SWEETBROOK RD. 
WILLIAMSTOWN, MA 01267 


RHODORA, Vol. 97, No. 892, pp. 382-385, 1995 


RHODORA NEWS & NOTES 
LisA A. STANDLEY 
HIGHLIGHTS OF CLUB MEETINGS 


November 1995 (913th Meeting). Seven Club members shared 
slides and information at the “Annual Exchange of Botanical 
Explorations and Exploits,” AKA Show & Tell. Pam Weatherbee 
spent much of her summer conducting a floristic survey ofa 1000- 
acre parcel on Mt. Greylock. Pam reported finding an unusually 
high frequency of Botryichium matricariaefolium, and new sites 
for Galium boreale and Carex baileyi. Leila Schultz offered a look 
at rare plant hunting in Utah, where rugged terrain and isolated 
mountain ranges provide spectacular scenery as well as challenges 
to the botanist. Her summer’s work succeeded in relocating a 
population of the rare endemic 7he/ypodiopsis argillacea, a del- 
icate pink-flowered mustard, on gypsum outcrops in eastern Or- 
egon. Barre Hellquist spent a month in Australia looking at aquat- 
ic plants. Wetlands containing as many as six species of waterlilies 
were the high point of the trip. 

Matt Hickler noted the effects of the 1995 drought on oxbow 
ponds along the Nashua River at Fort Devens. During the drought, 
the aquatic vegetation in these ponds was still present and even 
healthy, but drastically different from normal growth forms. Pot- 
amogeton natans formed tight appressed rosettes, while Nym- 
phaea produced erect leaves on robust petioles. Rare species, like 
Panicum philadelphicum, appeared in abundance on exposed mud 
flats. 

George Newman reviewed the June Club Field Trip to Mt. 
Washington’s Alpine Gardens, attended by 23 people. Ten hardy 
souls braved the gale winds and continued on to the headwall of 
the Great Gulf. Highlights included Loiseleuria, Rhododendron 
lapponicum, Phyllodoce and Cassiope; people who left early missed 
seeing Cardamine bellidifolia and Saxifraga rivularis. In Septem- 
ber, George returned to the Presidentials to search for the dis- 
tinctive bright red leaves of Arctostaphylos alpina. On a later trip 
to Newfoundland, George saw thousands of individuals of Are- 
thusa in bloom and located a large population of a form of Sar- 
racenia that lacks anthocyanin. 

Art Gilman provided a survey of the diversity of club moss 
species and hybrids in northern Vermont, where it is possible to 
find Diphasiastrum tristachyum, D. digitatum, and their hybrid 


382 


1995] Rhodora News & Notes 383 


(D. x habereri), as well as Diphasiastrum x zeilleri and its parents 
(D. tristachyum and D. complanatum). Abandoned high pastures 
are habitats for Diphasiastrum x sabinifolium (tristachyum x 
sitchense), and a newly described hybrid of D. sabinifolium and 
D. digitatum. 


December 1995 (914th meeting). Dr. Norton Nickerson of Tufts 
University spoke on “A Look at New Zealand Mangrove Eco- 
systems Over Time.’ New Zealand has but a single species of 
mangrove, Avicennia resinifera, the world’s southernmost man- 
grove. The epithet was chosen because early collectors found float- 
ing lumps of Agathis resin in mangrove swamps. Mangroves have 
an important ecological role as the tropical equivalent of salt 
marshes, but are often threatened by grazing, cutting, and filling. 

Much of the talk focused on mangrove swamps in embayments 
near the tip of New Zealand’s North Cape. Although a national 
park, sheep and cattle are grazed on the uplands. Where the hills 
bordering mangrove swamps are heavily grazed, sediment de- 
position buries pneumatophores and kills the mangroves. Ani- 
mals also wander onto the tidal flats at low tide and damage the 
mangroves by grazing. In the late 1970’s, mangroves had been 
virtually eliminated in this area, and the local fishermen com- 
plained about declines in shellfish and fisheries. 

Recommendations that a buffer strip of Crown lands along the 
edge of the estuary be protected were implemented by the local 
authorities. In the absence of grazing, shrubs and a dense vege- 
tation had become established in this “buffer strip,” and trees 
had been planted in some areas to restore vegetation to overgrazed 
areas. Consequences were dramatic— mangrove swamps recov- 
ered and became re-established in areas where they had practically 
disappeared. Mangroves have also colonized and replaced stands 
of the introduced Spartina anglica, indicating that the salt marsh 
grass may be acting as the first successional stage in the estab- 
lishment of mangroves. 


THE GRAY HERBARIUM CARD INDEX OF NEW 
WORLD PLANTS AND THE HARVARD UNIVERSITY 
HERBARIA TYPE SPECIMEN 
COLLECTION DATABASE 


The Gray Herbarium Card Index and the Harvard University 
Herbaria Type Collection databases are now available on the 


384 Rhodora [Vol. 97 


World Wide Web through the Harvard University Herbaria 
(HUH) Web page. The Universal Resource Locator (URL) for 
the HUH Web site is: http://www.herbaria.harvard.edu. 

With the above URL, users will find a general outline of the 
Harvard University Herbaria, including ‘“‘Databases,” from which 
many searchable databases, including the Gray Herbarium Card 
Index, the Harvard University Herbaria Type Specimen Collec- 
tions, the Farlow Diatom Collection, Botanical Collectors, and 
Botanical Authors can be accessed. 

The Web page search form, for both the Gray Card Index and 
Type Collections, has multiple fields (such as Family, Genus, 
Specific & Infraspecific Epithets, Author, Publication, and Type). 
This allows users to make complex queries. Detailed instructions 
on searching are provided. 

The Gray Herbarium Card Index provides bibliographic details 
for new taxa of vascular plants, new combinations, new status, 
and new names of New World Plants. It also includes information 
on infrageneric and infrafamilial names, and on types (including 
epi-, neo-, and lectotypification). From its inception (in 1893), 
the Gray Index included publication information for all specific 
and infraspecific names, basionyms, replaced synonyms, and oth- 
er nomenclatural synonyms, but not taxonomic synonyms. 

Although the original scheme of the Gray Index was designed 
to cover specific and infraspecific names published from | January 
1886, it was later modified to include all infraspecific names pub- 
lished from 1753. An effort to include infraspecific names from 
this period is still underway. The information on infrageneric and 
infrafamilial names and types starts from the early 1970’s. 

The Gray Index data, which was published in the form of 
printed cards until the mid 1980’s and as microfiche until early 
1992, has been made available over the Internet using Gopher 
since mid-1992. (Users may find additional information on the 
Web page “About the Gray Herbarium Index.’’) 

The Type Collection database includes information on type 
specimens in the Harvard University Herbaria (A, AMES, ECON, 
FH, GH, NEBC), collected from all parts of the world. The data 
on Type Collections has also been available on Gopher since 1991. 
(Presently, Internet users may access the Gray Index data and 
Type Collections data through either Gopher or the World Wide 
Web; however, at some point in the near future the Gopher ser- 
vices will be discontinued.) 


1995] Rhodora News & Notes 385 


An Appeal to the Users of the Gray Herbarium Card Index and 
Type Collections. The accuracy and completeness of the Gray 
Card Index and Type Collections databases depends largely upon 
the input of the users, who are urged to provide any relevant 
information (such as additions, omissions, and corrections to the 
data) via e-mail to either K. N. Gandhi (gandhi@oeb.harvard.edu) 
or David Boufford (boufford@oeb.harvard.edu). Contributed by 
K. N. Gandhi. 


THE NEW ENGLAND BOTANICAL CLUB 
Elected Officers and Council Members for 1995—1996 


President: C. Barre Hellquist, Box 9145, Department of Biology, 
North Adams State College, North Adams, Massachusetts 
01247 


Vice-President (and Program Chair): W. Donald Hudson, Jr., 
Chewonki Foundation, RR 2, Box 1200, Wiscasset, Maine 
04578 

Corresponding Secretary: Nancy M. Eyster-Smith, Department 
of Natural Sciences, Bentley College, Waltham, Massachu- 
setts 02154 

Treasurer; Harold G. Brotzman, Box 9092, Department of Bi- 
ology, North Adams State College, North Adams, Massa- 
chusetts 01247 

Recording Secretary: Lisa A. Standley 

Curator of Vascular Plants: Raymond Angelo 

Assistant Curator of Vascular Plants: Pamela Wetherbee 

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) 
Thomas Mione °96 
Garrett E. Crow °97 
Edward J. Hehre °98 
Donald J. Padgett (Graduate Student 
Member) ’95 


RHODORA 


JOURNAL OF THE 
NEW ENGLAND BOTANICAL CLUB 


GORDON P. DEWOLE JR., Editor-in-Chief, Nos. 889, 890 
DAVID S. CONANT, Acting Editor-in-Chief, Nos. 891, 892 


Associate Editors 


DAVID S. BARRINGTON 
W. DONALD HUDSON, JR. 
LESLIE J, MEHRHOPP 
THOMAS MIONE 
CATHY A. PARIS 
LISA A. STANDLEY 


VOLUME 97 


1995 


The New England Botanical Club, Inc. 


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


INDEX TO VOLUME 97 


Additions 9-38, 291-327 
Additions to the preliminary checklist 
of the vascular flora of Connecticut 
3 


Algae 275-279, 328-338 

Allelopathic effects of Lantana camara 
(Verbenaceae) on morning glory 
(Ipomoea tricolor). 264-274 

Allelopathy 264-274 

Allison, Taber D. 357-374 

Amelanchier nantucketensis 339-349 

panera Gregory J. 185-200 

Antenna 8 

ee 185-200 

Arabis 109-163 


Beck, Kathryn A. 201-207, 350-356 
Bidens laevis 280-282 
Bog 39-92 


K REVIEWS: 
Asteraceae, Cladistics & Classifica- 
i 78 


A Guide to Wildflowers in Winter: 


Plant Identification Terminology: An 
Illustrated Glossary. 283-284 
Brassicaceae 109-163, 185-200 


Campbell, Christopher S. 339-349 

Canada 171-175, 245-254, 255-263 

Canada yew 357-374 

Caplow, Florence E. 201-207, 350-356 

Casado, Christina M. Allelopathic ef- 
fects of Lantana camara (Verbena- 
ceae) on morning glory (Ipomoea tri- 
color). 264-274 

Catling, Paul M., Jacques Cayouette, 
and Joseph Postman. Fragaria mul- 
ticipita, reduced to the rank of forma. 

-254 


Cayouette, Jacques 245-254 


G. Chromosome 
number determinations for New- 


foundland species of Antennaria 
Gaertner (Asteraceae, Inuleae) 1-8 

Chromosome number determinations | 
for Newfoundland species of Anten- 
naria Gaertner (Asteraceae, Inuleae) 
1-8 


Chromosome number of Saxifraga 
gaspensis Fernald 171-175 

Chromosome numbers 1-8, 171-175 

Cleland, Maryke A. 185-200 

Colt, L. C., Jr. Studies on New England 
Algae II: A second station in Maine 
for Nitella tenuissima (Desv.) Kuetz- 

275-279 

Connecticut 9-38 

Conservation 185-200 

Conservation status 255-263 

Contributions to the flora of Vermont. 
291-327 

Crow, Garrett E. 39-92 

Cuba 96-97 

Cytology 109-163 


Deer browsing 357-374 

Deletions 291-327 

Dibble, Alison C. and Christopher S. 
Campbell. Distribution and conser- 
vation of Nantucket shadbush, Amme- 
lanchier nantucketensis (Rosaceae). 
339-349 

Dignard, N., R. Lalumiére, and M. Ju- 
lien. Genliana nivalis L. (Gentiana- 
ceae) new to Québec. 375-379 

Dignard, Norman 171-175 

10. New Barnstable 


Discovery rate 291-327 

Disease 245-254 

Distribution 171-175, 375-379 

Se setae and conservation of Nan- 
ucket shadbush, Amelanchier nan- 

ae (Rosaceae). 339-349 

Drosera anglica 164-170 

Drosera anglica x D. linearis 164-170 

Drosera linearis 164-1 


389 


390 


Endangered species 291-327 

Endemism 171-175, 245-254, 339-349 

Eriogonoideae 350-356 

Eriogonum codium 350-356 

gp itl codium (Polygonaceae: Er- 
iogonoideae), a new species from 
Seidl Washington. 350-356 

Error rate 291-327 

Exotic weeds 264-274 


Fahey, Linda L. and Garrett E. Crow. 
The vegetation of Pequawket Bog, 
Ossipee, New Hampshire 39-92 

Flora 9-38, 291-327 

Floristics 350-356 

Ford, Bruce A. Status of the ne ees 
Vaccinium stamineum L. (Eri 
ceae), in Canada. 255-263 

Fort Devens, Massachusetts 208-244 

Fragaria multicipita 245-254 

Fragaria multicipita, reduced to the rank 
of forma. 245-254 

Fragaria virginiana 245-254 

Freshwater algae 328-338 


Gaspé Peninsula 171-175 

Geltman, Dmitry V. Urtica chamae- 
dryoides Pursh (Urticaceae) reported 
as new to Cuba. 96-97 

Gentiana nivalis 375-379 

Gentiana nivalis L. (Gentianaceae) new 

375-379 

Germination 264-274 

Gervais, Camille, Norman Dignard, and 
Rosaire Trahan. The chromsome 
number of Saxifraga gaspensis Fer- 
nald. 171-175 


Hanford Nuclear Reservation 201-207, 
350-356 


Hudson River 328-338 
Hunt, David M., Karen B. Searcy, Rob- 
ert E. Zaremba, and C. Roberta Lom- 
bardi. The vascular plants of Fort 
evens, Massachusetts. 208-244 
Hybrids 164-170 


Immigration 291-327 
Ipomoea tricolor 264-274 


Rhodora 


[Vol. 97 


James Bay 375-379 
Jenkins, Jerry and Peter F. Zika. Con- 
tributions to the flora of Vermont. 


291-327 
Julien, M. 375-379 


Labrador 375-379 

Lalumiére, R. 375-379 

Land-use history 357-374 

Lantana camara 264-274 

Les, Donald H., Gregory J. Anderson, 
and Maryke A. Cleland. Sterility in 
the North American lake cress, Neo- 
beckia aquatica (Brassicaceae): Infer- 
ences from chromosome number. 
185-200 

Lesquerella douglasii 201-207 

Lesquerella tuplashensis 201-207 

Lombardi, C. Roberta 208-244 


Maine 275-279, 339-349 

Marsh sow-thistle 93-95 

Marsh sow-thistle (Sonchus palustris) 
in North America 93— 

Massachusetts 208-244 

Mehrhoff, Leslie J. Additions to the 
preliminary checklist of the vascular 

of Connecticut. 9-38 

Mertensia maritima 280-282 

Morning glory 264-274 

Morton, J. K. and Joan 
marsh sow-thistle (Sonchus palustris) 
in North America 93-95 

Mulligan, Gerald A. Synopsis of the ge- 
nus Arabis (Brassicaceae) in Canada, 
Alaska and Greenland. 109-163 

Mycoplasma 245-254 


Nantucket 339-349 

Natural hybrid of Drosera anglica Huds. 
and Drosera linearis Goldie in Mich- 
igan. 164-170 

Neobeckia aquatica 185-2 

New Barnstable county records. 280- 


NEW ENGLAND NOTES: 
New Barnstable county records. 280- 
282 


1995] 


Studies on New England Algae II: A 
second station in Maine for Nitella 
tenuissima (Desv.) Kuetzing. 275- 
279 


New Hampshire 39-92, 357-374 
New species 109-163, 350-356 
Newfoundland 1-8 

Nitella tenuissima 275-279 

No i 


Nova Scotia 339-349 


Occurrence of the red alga Thorea vio- 
lacea (Batrachospermales: Thorea- 
ceae) in the Hudson River, New York 
State. 328-338 

Ontario 93-95 


Peatland 39-92 

Plant classification 39-92 

Plant community 39-9 

Plant-site relationships 357-374 

Polygonaceae 350-356 

Postman, Joseph 245-254 

Pueschel, Curt M., P. Gary Sullivan, 
and John E. Titus. Occurrence of the 
red alga Thorea violacea (Batracho- 
spermales: Thoreaceae) in the Hud- 
son River, New York State. 328-338 


Québec 171-175, 245-254, 375-379 


Rare plants 208-244, 350-356 


species from southcentral Washing- 
ton. 350-356 

Rhodophyta 328-338 

Rhodora News and Notes 98-107, 179- 
182, 285-288, 382-385 

Rollins, Reed C., Kathryn A. Beck, and 
Florence E. Caplow. An undescribed 
species of Lesquerella (Cruciferae) 
from the state of Washington. 201- 
207 


Rumex pallidus 280-282 


Saxifraga gaspensis 171-175 
Saxifraga nivalis 171-175 


Index to Volume 97 


391 


Saxifraga tenuis 171-175 

Schnell, Donald E. A natural hybrid of 
Drosera anglica Huds. and Drosera 
linearis Goldie in Michigan. 164-170 

Searcy, Karen B. 208-244 

Shadbush 339-349 

Sonchus palustris Linn. 93-95 

Stachowicz, John J. and Taber D. Al- 
lison. Vegetation, browsing, and site 
factors as determinants of Canada yew 
(Taxus canadensis) distribution in 
central New Hampshire. 357-374 

Standley, Lisa 98-107, 179-182, 285- 
288, 382-385 

Status of the deerberry, Vaccinium 
stamineum L. (Ericaceae), in Canada. 
255-263 

Sterility in the North American lake 
cress, Neobeckia aquatica (Brassica- 
ceae): Inferences from chromosome 
number. 185-200 

Strawberry 245-254 

Studies on New England Algae II: A 
second station in Maine for Nitella 
tenuissima (Desv.) Kuetzing. 275-279 

Sullivan, P. Gary 328-338 

Synopsis of the genus Arabis (Brassi- 
caceae) in Canada, Alaska and 
Greenland. 109-163 

Systematics 96-97 


Taxonomy 1-8, 109-163, 245-254 
Taxus canadensis 357-374 

Thorea violacea 328-338 
Threatened species 255-263 

Titus, John E, 328-338 

Trahan, Rosaire 171-175 

Triploid 185-200 


Undescribed species of Lesquerella 
(Cruciferae) from the state of Wash- 
ington. 201-207 

Urtica chamaedryoides 96-97 

Urtica chamaedryoides Pursh (Urtica- 
ceae) reported as new to Cuba 96-97 

Urticaceae 96-97 


Vaccinium stamineum 255-263 
Vascular flora 208-24 


392 Rhodora 


Vascular plants 9-38 

Vascular plants of Fort Devens, Mas- 
sachusetts. 208-244 

Vegetation 39-92 

Vegetation of Pequawket Bog, Ossipee, 
New Hampshire. 39-92 

Vegetation, browsing, and site factors 
as determinants of Canada yew (Tax- 
us Canadensis) distribution in central 


New Hampshire. 357-374 
Venn, Joan M. 93-95 
Vermont 291-327 


White Bluffs 201-207 


Zaremba, Robert E. 208-244 
Zika, Peter F. 291-327 


[Vol. 97 


Volume 97, No. 891 including pages 185-290, was issued August 30, 1996 


THE NEW ENGLAND BOTANICAL CLUB 
22 Divinity Avenue 
Cambridge, MA 02138 


The New England Botanical Club is a non-profit organiza- 
tion that promotes the study of plants of North America, es- 
pecially 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, RHO- 
DORA, 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 subscrip- 
tion 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 
Student Member $25.00 
Family Rate $45.00 


For this calendar year eee 
For the next calendar year a 


Name 


Address 


City & State Zip 


Special interests (optional): 


393 


394 Rhodora [Vol. 97 


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), in addition, a copy of the ms. on a 3.5 inch 
floppy disc will facilitate production of the paper. The text must be 
double-spaced (at least %”) throughout, including tables, figure leg- 
ends, and literature citations. The list of legends for figures and maps 
should be provided on a separate page. Do not use footnotes. Use a 
non-proportional font throughout and do not justify the right margin. 
Do not indicate the style of type through the use of capitals or un- 
derscoring, particularly in the citation of specimens. Names of genera 
and species may be underlined to indicate italics in discussions. Spec- 
imen 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 syn- 
onyms. 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 is distinguished from its congeners. Papers of a 
floristic nature should follow, as far as possible, the format of ‘‘An- 
notated List of the Ferns and Fern Allies of Arkansas,” W. Carl 
Taylor and Delzie Demaree, Rhodora 81: 503-548, 1979. For biblio- 
graphic citations, refer to the Botanico-Periodicum-Huntianum (B-P- 
H, 1968) and B-P-H/Supplement, 1991, which provides standardized 
abbreviations for journals originating before 1990. All abbreviations 
in the text should be followed by a period, except those for standard 
units of measure and direction (compass points). For standard abbre- 
viations 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 xX 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 3% X 6 in.) Magnifi- 
cation/reduction values given in text or figure legends should be cal- 
culated to reflect the actual printed size. Tables should be set up in 
the same way. Each page should be ready for reproduction and will 
be reduced to 4 X 6 inches. 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. 


UNITED STATES 


Baro 


Statement of Ownership, poragements aod Circulation 


POSTAL SERVICE ™ red by 39 U.S.C. 3685) 13. Publication Name 
1. Publication 2. Publication No. 3. Filing Dat = 
ao als 18. k Average No. Copies Each Issue Actual No. Copies of Single Issue 
RHODORA Oo} O13 }5/-]4 19 2\[o -21-G6L During Preceding 12 Months Published Nearest to Fillng Date 
4, Issue Frequency 5. No. of Issues Published 
FOUR TIMES PER YEAR IN p 
JAN., APRIL, JULY, OCTOBER coreg a4 | $40.00 814 825 
b. Paid and/or Requested Circulation 
(1) Sales Through Dealers ai i 
(Nat Mailed) 
THE NEW ENGLAND BOTANICAL CLUB, 810 EAST 1OTH STREET, LAWRENCE, KS 66044 
(2) Pald or Requested Mail Subscriptions 
(Include Advertisers’ Proof Copies/Exchange Copies) 762 769 
c. Total Paid and/or Requested Circulation 762 7€9 
THE NEW ENGLAND BOTANICAL CLUB, 810 EAST 10TH STREET, LAWRENCE, KS 66044 (Sum of 15b(1) and 15b(2)) 
d. Free Distribution by Mail 
9. Full Names and Complete Mailing ere (s & S 
Publisher (Name and Complete Mailing 
THE NEW ENGLAND BOTANICAL CLUB, 810 EAST 10TH STREET, LAWRENCE, KS 66044 oe ceo ueor Cuusace| ato eeeul (ai iste Cel aes) 0 0 
p 9 ) 2 2 
JANET R. SULLIVAN, DEPARTMENT OF PLANT BIOLOGY, UNIVERSITY OF NEW HAMPSHIRE, DURHAM, NH 
03824-3597 g. Tolal Distribution (Sum of 15¢ and 15!) 764 m1 
1g h. Copies Not Distributed : 
MARGARET GoGLE, DEPARTMENT OF PLANT BioLocy, UNL, DUKA, NK OD8A4-35497 Or echetal aebece 50. 54 
10. Owner (If owned by a corporation, its name and address must be stated and also immediately thereafter the names and addresses of stockholders owning 
pile) a parhershp or te runincorpraed hi estos Stock. If not alinotewned by va. elie lieth Noa names CS Gi aces esse of the enna owners rast be given. If i Total (Sum of 159, 15h(1), and 15h(2)) 1 
Byra Trocpert tai ice ler eiite ar ape case aos boren GE SEE vc onan pe Ae eee econ sped : ; : 814 825 
(15¢/ 15g x 100) 99% 
THE NEW ENGLAND BOTANICAL CLUB, 810 EAST 10TH STREET, LAWRENCE, KS 66044 


SANUACY issue of this publication. 


11. Known Bondholders, sexioagees, and Ade $s y A 
Securities. If none, check here. §xNono a 


Full Name 


12. For completion by nonprofit organizations authorized 
status for federal income tax purposes: (Check one) 


© Has Changed During Preceding 12 Months 
(it changed, publisher must submit explanation of change with this statement) 
{ructions on Reverse) 


PS Form 3526, October 1994 


. Br QUA dy , Edita w- met 
that all information furnished on this form is trua and complete. | understand that anyone who furnishes false or misleading information on this form or 


eri 
un omits material or information requested on the form may be subject to criminal sanctions (including fines and imprisonment) and/or civil sanctions 
(including multiple damages and civil penalties). 


Dale 
(6 Oaroberl{16 


Instructions to Publishers 


2. ine lude in item: s where the hearst the person or corporation for wi 
rustee is = oa eas lnclude the names and addresses of individuals a are siocktolders oe own or hold 1 percent i more of the 

total amount of bonds, mortgages, or other securities of the publishing corporation. In item 11, if none, check box. Use blank sheets if 

jore space is required. 


der or security holder is a trustee, the name of 


#3 


3. i ini tit ‘1 Jali F 
. If the publication had s: 
Circulation mus! be publis 
published sens October. 

5. Ini f 


and f. 


Management, and 


6. Item 17 must be dh 


PS Form 3526, October 1994 (Reverse)