Table Of ContentAN ANOMALOUS
POPULATION OF ASTER
(ASTERACEAE: ASTEREAE) SENSU LATO MICHIGAN
IN
LNesom
Guy
those of 0< li'mi-iw-in the into
tiniml DnrlliMiyn,! isver\ hunt;
m
In ii" led in >n n
\
I i
miim
independiente.Lasevidencia iem mi m.,i
ictual. \< u m.i In n id. u
i i
|
M
The U bud im
recent repot di >pul u ion o h\ \u n
t >. < il I h i < / ).>< /// >; i
|
umbellata Nees and Ol/oihthi nemordis Greene
(Aster) Mill J t/Vster) (Ait.)
(P.
from Keweenaw Michigan
the Pen in ula of aio ion \estern
is
1 >.
! 1 . , 1 1
remarkable. Such an inn >cnen< h\ brid vould provide corroborative evide no
that these two genera (Doellingeria Nees and k lemena Greene) segregated from
(
mm
nnpsn.
Naturalh
Aster are closely related. .a in in' h. lo hi id In
i
i
Astereae have previously been ivpoi ieJ only between closely related genera—
m
some cases these hybrids ma\ heiio ween congeneric
in
b. l>et
\\
! i
(Nesom
species 1994b).
A
hypothesis between and Oclcmena
of close relationship Doellingeria
mo
(Nesom
1994a) based on molecular
-logical similarities In contrast,
is pi
may
evidence (Semple er 99o: Xiang or Sem pic 99o; suggests that Oclemena
al.
1
I
A between
he closely related to lona< n'sdreene. hypothesisol close a tionship
el
i
Odemena
Dodlingeria and also suggested the disparity of interpretation
in
is
USA
endemic
regarding the position of Asler reiicuhitus hi a southeastern
I.,
Nesom
placed by Semple et ah (1991. 1996) in Dodlingeria but by (1993, 1994a,
2000) in Odemena. Hybridization between species ot Dodlingeria and
Odemena
has not been previously reported, although D. umbellata and 0.
and
ncmoralis are sympatric over most oi the range the latter both species
of
occur wetland
habitats.
in
was
,1-i'des u net hat lie puiatixely li\ hi id K)judat uvn distinctive in the
(. i I I i
|
and appeared intermediate between nearby populations
be of
field to
Dodlingeria umbellata and /emend nemmu/i.s. While the putative hybrids
k
«
clearly do not the epical morpholog\ eithei species the only indication
oi
fit
t
m
of intermediacy in the published obsei vat ions the statement 18) that
is (p.
the "overall height and leaf size and shape are intermediate between the as-
sumed and and voucher
parents." In the illustration description, in the speci-
mens, other possible indications of mterinediacv between D. umbellata and O.
Plant- studied
The population described by ierdes (Ge/de.s 2210) includes about 20 plants in
(
swamp
tamarack between populations Dodlingeria umbellata
a situated oi
m
("common 60 north oi the hybrid site in a mixed wetland complex"— Gerties
m
Odemena
221J) and ncmoralis (ca. "60 southeast of the hybrid site in a
wo
swale 'A> M plants of he putative hybrids and one plant each of
,'en/es
l
(
I l
umhdlala and ncmoralis from loughton County were
D. O. the site available
I
many Odemena
study (M1C1 as well as other specimens from axa of and
for
1).
I
Dodlingeria (BRIT, GH, NCU). The two plants oi 2210 are virtually identical in
micromorphological perhaps epivsen ling single clone, and Gerdes did
detail, a
i
among
not note the occurrence oi significant variation the putative hybrids.
The plant of O. ncmoralis morphologicalb topical the species— this col-
ol
e-
em
known
lection represents the west most population the species (see Voss
oi
may
The
1996; Gerdes 1998). collection oi D. mhcllata be arbitrarily identified
a
U umhdlaui
asvar. umbellata: hainei plants he species pubens (A. Gray)
oi var.
>
t
Keweenaw
Brittjalso occur on the Peninsula mappetl hv Semple et ah 1991).
(.as
Houghton County specimens examined:
Nces-MICHIGAN
Houghton Grand
1) Dodlingeria umbellata (P Mill Co.:
)
Traverse Bay ridge and swale complex, ca. 7 mi h.Sh oi hake Linden, travel E of
Rice Lake 0.6 mi and walk N oi gravel road into swale edge of sandy road
ca.
and .SplidgnimCcricaccoLis swale: Aster ncmoralis also swale; rays creamy
in
white; 18 Sep 1997. LB. Gerdes 22/1 (M1C1
1).
Grand
2) Oclemena nemoralis {A\t jGreene Mil .AN. loughton Co.: Traverse
lit
1 1
Bay ridge and swale complex, ca. 7 mi ; s : o Lake Linden and ; of Rice ake.
I 1 1 1 I
Swale, primarily a Sphagnum, cncaccoias hog poor ion plants scattered and
I J:
mmm.
G mwp'n
mu
numerous with na
nit /ion u a u i,c, \u
( \ II,
( /
<
rays dark pink; 18 Sep 1997. LB. Gcnlcs 2209 (MICH).
etc.;
—
MICHIGAN.
The putative hybrid loughton Grand Trawr.se Bay ridge
3)
o.:
(
I
and swale complex, mi LSL Lake Linden and Lot Rice Lake. Tamarack
7
ea. o\
swamp Aroma
adjacent Sp/nignmn/ericaccous swale: 20 plants with
to a ca.
prunifoliaXhamacduphnccdlvi Alnus rugasaAlalaiuagrostiscanadensis,
ulata.
raw
and Carex .Asfcrm'incodi.sand A^tcr umhcllal nsnearky:
oligospermia. light
M1CH-2
n 8 Sep 1997. Gcnlcs 2210 sheets).
pi k: L./i.
(
1
genome
Evidence for the Doellingeria
Consideration of the illustration and published description of Gerdes 2210 and
examination of the voucher specimens indicate hat features of putative inter-
I
mediacy between Oclcmcmt and umhcllata
ocnii cro/ nocll'ingcria are rela-
is
dm
Hetght -Plants 2210 6-7 generally between the range of
of are talk
dm dm)
height O.nemorahs (2-7 and D. umhcllata (3-20 but also at the top
for
)
ol the range tor ncmoralis as well as 0. acuminata.
( ).
mm
cm
and 7-10
Leaves-Leaves of 22.10 are narrowly elliptic, 4.5-5.5 long
umhdlata
wide, with barely volute margins. Leaves of D. are elliptic to
entire, re
mm
narrowly mostly 6-12 cm 10-25 wide, and flat-margined;
long,
elliptic,
wry 2-5
narrowly oblong mostly
those of O. ncm.muli.sa.re to linear lanceolate,
mm
cm 2210
long, 2-12 wide, with entire, strongk involute margins. Leaves ot
intermediate between umhcllata and nemorahs but they are
are in size D. O.
more similar in shapi th< I'he abaxial surla 1221
t< 1
> i
nk
minunk mdnlai auk onkll n d intermediate between the
ik s
gl »
» ,
glandular (larger glands) surfaces of 0. nemorahs and eglandular surfaces of
D.umbella£a;theglandulai itvoi '2.210. hov dsoappcai to be indistinguish-
s
i
O
O.acumi
from ntslnhri m<
able that ol pi n
li
i i
Rayjloicis.- Rays of 22/0 are pink, intermediate between the purple of O.
hn
nemorahs and v uml iiu n
h
hit
II t
I 1 i i I 1 I
The
between nemorahs (purple-rayed) and animiiufta (white-rayed).
O. O.
length of the rays intermediate between ncmmuk.s and D. umbe/k;fn but
t '
within the range of variation lor O. ncmoralis.
—
Cypselar vestiture Cypselae of 2210 are kenwt rigose-hispid, more sim
si
\
i
AN.
lar to the sparsely to deriscU rrigo.se ie umhcllata than the vpi
el.
i i
|
cally glabrous ones of 0. ncmoralis (thus this aspect oi vestiture is technically
m
mdnlai
not intermediate; Other u ilu n k c\ pselat sui laces
d<.
i I
of 2210 are similar to those of 0. nc moral shin not u> the eglandular surfaces of
i
D.umbellata.
Morphological evidence thai Ve/hnyr/ umhcllata represented the
ta is in
/
genome
of the plants oi ,V;</c\ 22/0 limited, represented by the features ol
is
(
equivocal interpretation noted above, in contrast, various features of the roots,
leaves, capitulescencc. Morels, and fruits refer Gcrdcs 22/0 to Oclcmcna. The
comparisons Table summarize morphological between
in distinctions
1
Oc/emenunemoru/i.sand
Oocllingcria umbcllata-ihe^c also are essentially the
Wa).
contrasts that distinguish the two genera Nesom each
(see In contrast,
1
morphology
2210
the similar to that of Oclcmcna.
oi' is
If the assessment here of Gcrdcs 2210 correct in excluding Doellingeria
is
from its close ancestry, two alternative hypotheses would place the evolution-
anomalous may
ary origin of this population horn within (Vlcmrna.
First,
it
represent hybrid between ncmora and acuminata
a O. O. (Ait.) Greene, or sec-
lis
may
Oclcmcna
ond, represent a lineage differentiated early in the evolu-
it oi
tionary history the genus and now persisting only as the Hough-
ol a relict at
County
ton
site.
ihtrnuua
Idcnlificiiiion as hlahci
The
plants of Gcrdcs 2210 are nearly identical overall aspect some
in to indi-
viduals of Oclcmcna hhikci Nesom, and
(Porter.) a fertile, recurrent, persistent
hybrid between nemo sand acuminata - compare
O. rail 0. big. of Gerdes(1998)
1
and
to Figs. 3 Pike 970). Populations oi hlahci are scattered through
11 of" (1
(
). >
the area of sympatry of the parents where ecological conditions allow
their
&
Simon
close contact (Bromllet 1081). "Extensive colonies of xblakei are
[0.1
swampy
often lound at the edges oi bogs, the shores oi ponds, and borders of
woods, etc, the kinds of areas that are intermediate wetness between the bo-
in
real forest habitat of acuminata and the open bogs nemoralis" (Pike
\0.\ oi' \0.\
1970, 401).
p.
the plants of C.Vk/cs 22/0 arc scored on the morphological hybrid index
II
& &
developed by Pike (1970; also see Hill Rogers 1973; Brouillct Simon
1981),
which
the value is 9 (Table 2), is within the range characteristic of O. xblakei,
intermediate between nemoralis and acuminata. Oclcmcna xblakei
O. O.
is
known
to produce fertile seeds Mil Or Rogers 197 0, but was not possible to
it
make G
unequivocal observations of fertility for cities 2210. Pollen grains on the
stigmatic surfaces were regular in size, but the anthers had opened before the
and an
collection estimate of pollen fertility could not certainly exclude pollen
from None
other species. the cvpselae of Gcrdcs 22/0 were completely ma-
o\'
when
examined
ture collected; dissected cypsclae from each plant had produced
an elongated otherwise undeveloped embryo. Oclcmcna and
but xblakei
its
&
known haw chromosome number
parental species are to a In = 18 (Hill
ol
Rogers 1970). Scoring o\ 2210 on this hybrid index does not indicate that
it
jtmaw
Rool sk-ndm,
Sy-.fi vr. el'
mm
trongly long, anwHiK
ii
-hinlul In
nli in
- il( it it it il it 1
t I I I i I
]IM
1 ll li II III ll till III | I i I I I III I In III < tin fill I it
mmoydy
to narrowly coinmnar, columnar, length maturity
c:a. c,i. at
1/2-2/3 the phyllary 1/2-2/3 the phyllary
mam>rabsand O.m mmnuU; butdoesshowthat
hybrid between
actually isa O.
same way known
features can be interpreted as intermediate in the as those of
its
hybrids can
be.
Oclcmcna
"Iwo conditions complicate the identilicaiion ol 'icrdes 2210 as
(
loundonly sym-
xblakei: O.xblakei pre\ ioush has been wit bin the area of
(1)
patry the parents, and one the parental species, 0. acuminata, does not
of
oi'
occur the Keweenaw region; and plants of ,adcO22l0 have entire leaves
in (2)
(
and hispid cypselae, features not generalb charactcrist of O. blakci.
ic
known much
Oclcmcna -blakci has been to occur only further east-
(1)
known Keweenaw ism heastern Ontario, about
the closest locality u> the site soul
m
Simon Semple
850 kilometers eastward (.maps Brouillet 6c 1981, ct 1966).
al.
The closest known locality for acuminata the same region of Ontario.
is
( ).
ncmoralis grows immediately adjacent to the 2210 population, but
OclL-iik-iiii
unlikely that O.acuminaUi has occurred naturally in Michigan since post-
is
it
Wmon
would
ong-distancedispersal
glaciation revegetafion Hroni or 1981
let
( I ). I
the most likely explanation lor the far-disjunct occurrence of O.xblakei on the
Keweenaw Peninsula-at least a simpler hypothesis than a postulate of the
is
it
\m
former occurrence and extirpation umimita west of present geo-
o( lar its
(
"Long and chance establishment
graphic range. distance dispersal in suitable
bogs are probably responsible lor he establishment O. ncmoralis] in the
|ol
|i
...
eastern Lake Superior area ..and the species could still be expanding its range
westward. Simon 08 539-540). The discovery of O.xblakei
."(Brouillet
K
<Sx pp.
1
the westernmost locality ncmoi\ilis might also suggest that the former
at ol
(.).
expanding
range.
issiiinl;ul\' its
the region parental sympairv. O. blakci often occurs with only one
In o\
oi the parents or even with neither of the parents inclose proximity (Pike 1970),
through rhizomatous and ranging more widely
apparently persisting babit
its
When
through fertile cypselae. found with only one ol its putative parents, O.
&
xblakei more often associated with O. ncmoralis U'ike 1970: Brouillet Si-
is
mon
I9SU, perhaps because their greater similarity in habitat.
ol
The parentage O. acuminata in the hybrid O. blakci is usually re-
(2)
o\'
•
flected by at least some degree ol lobar toot lung ait ire leaves occur rarely in O.
i
may morphology
xblakei, but PI individuals sometimes be closer in to the en-
acuminata
tire-leaved O. ncmoralis than to the toothecbleaved O. (Pike 1970;
&
and
Hill Rogers 197 b\ idence for int regression between O. blakci its par-
3).
•
ents was observed nature by Pike (1970) and experimentally duplicated by
in
and Rogers but the overall morphology of Gerdes 2210 more simi-
Hill (1973), is
lar to an Fl, except for the entire leaves. Leaves ol' 2210 are elliptic-those of O.
xblakei vary shape torn oblanceolate (broadest above the middle) to elliptic.
in
I
The cypselae of cedes 2210 bear a mixture o\ sessile glands and numer-
(,
spreading ascending sharp pointed bans and are closely similar
ous, slender,
in vestiture to those of Oclcmcna reticulata (Ell.) Nesom, a species of the coastal
Alabama, and h oved geography
plain of Florida, Georgia, Sout Carolina, in
far-i e
1 1 1
and habitat from hi hIici peci Ath. enu hi ypsel rvestiture is unlike
r
I
that of 0. x bit/ from elsewhere in its range or that of its parental species,
let'/'
orOmrummom
which glandular usually otherwise cvp rarek
is bill -ja'oroi.i ela>
may NCU), and
be sparsely Pendleton Mn.s.srlmun .3894—
strigose Co.. \Y. Va.,
(e.g.,
Sempleetal. (1996)d< cnlx cypscfu \urmoidiis sparsely
ih< ol a: strigose.'
(
)
md
The differences in no n land ular cvp u titui rh niai in u
I !
Go
morphological differences separating ties 22 10 from 0. hla kei, but the close
•
correspondence in otlu rmorphologi 'Ik urn ugg< rhat an identification
i
M<d tMi ibh hy| oth to h K< vet na tun u! plani
)
i ( i i i i . I < i
Identification as a separate species
In view of tentative evidence against a hypothesis of hybridity between
m
Oclemena nemo ml and umbiuh/ (considering the anomalous achenc
0.
is
vestiture, unusual leaf morphology, and the unlikelihood the establishment
ol"
of 2210 through long-distance dispersal.) lea the origin ol 2210, a hypothesis
that this population represents an independent evolutionary branch of
Oclemena becomes irucon CAu
plausible tin in in nition ol 2
II
i *
would
rank be appropriate. Current evidence, however, does not con-
specific
iih na A upooi oi limin in a In ill pro] o (I h poi In hn rnaii
\
i < > *
I
competing
to the originally proposed inteigenei h\ bridmatiou. All hree hy-
ic
t
when
upcoming
potheses will be further investigated alter an field season,
it
will be possible ro make observations on pollen and fruit fertility and collect
fresh material molecular analysis.
for
ACKNOWLEDGMENTS
am Houghton
grateful to the staff of M1C1 lor lending the collections from
1
1
GH
County, the staffs of and NC L for accommodation while working there, to
comments
hue Brouillet lor his on die plants involved and on an early version
REFERENCE
ml,
198E An- th
Simon. ii,[h distribution of Aster
J.-P. i> )t
,
i
-550.
Michx.and A.nemoralis Alton (Asteraceae:Astereae).Rhodora 83:521
ne n 1- J nemoralis xAster
8 [2000]. Aster n In oi id ste/
:
> r
Houghton Count 7-20.
n inil.li |.m Bot. 37:1
n
i I i
1976. Morphological and cytological evidence for introgression Aster
in
.
11-155
' tl m it in - in t ,t
1 1 1 r i i I | il , j i i l
[
Chromosome
and O.M.Rogers. 1970. numbers of Aster blakei and A. nemoralis.
72:437-438.
>ra
inn
(hin
In
n n el
in i il iii ii li i idt ll
I 1 I I i il ] i I I
Rhodora
House. 75:1-25.
of Aster blakei (Porter)
1
h
Ut '« n li i'Ii toloqia 75:452-462.
' [i (
I 1 I i i i i
Nesom, G.L. 1994a. Review of the taxonomy of Asia sensu lato pMteraceae: Astereae),
: h
emphasizing the rxM hi tologia 77:141-297.
u
'
I
m
Nfsom.G.L. 19941.).! lyhudi/aiion the Asteusie (Asieiai cie). i'hytologia 77:298-307.
Mitral Ameiu ad he Antilles, and Hawaii. Ada, Misc. 1-100.
( l-'nl. 20:i-viii,
<
."
Piki R.B. 1970. Eviden. x the I, oml itu ihi) House. Rhodora 72:
< t t 'I ,t \\
,
401-436.
Semple, J.C., J.G. Chmii a and C. Lei 1991. A multivariate mor| ihometric study and
ki, >er.
i i
u
~
Astersuh
revision of y teieae):The Aster
N
n
i
' < t ' - i if t -
umhelkitus complex. Canad. 69:256-276.
Bot.
J.
Simi'iiJ.C, S.B.Hi n, unfit m ><olh hnt a .mpositae: Astereae):
l hi
> l t i t
Oclcmena Greene, Nees and
Diplactis hi DnelHiuicria Aster (including
Raf.,
L.
.
m
ed Univ.ofWater-
f?e Hi
///
/ i 1 i I it i
996. Mich loan A oh Ac eae omposit,io).( lanbrook
\A i",d ,G. let a. At (I've Insti-
III, (
I I I I I
M M
A'"
tuh
u in F in in iiM in in
>t ill ,'l
i ( I t i , i i
XiANc,C.and J.C.Sioee 19'-'h.Molec ulai svstematii hiidv of Psfe; s.lat.and related genera
c
eae) based on chin PI unalv.es end inainK
if, x>
it.'
i«
if I t
ixa. In: I). I.N.I
I
Kew, 393-423.
:ae Conf., 994. Vol. pp.
1 1
,
