BEN # 209 - Dr. W.A. Weber Festschrift - Part III

37 views
Skip to first unread message

Adolf Ceska

unread,
Nov 24, 1998, 3:00:00 AM11/24/98
to

BBBBB EEEEEE NN N ISSN 1188-603X
BB B EE NNN N
BBBBB EEEEE NN N N BOTANICAL
BB B EE NN NN ELECTRONIC
BBBBB EEEEEE NN N NEWS

No. 209 November 24, 1998

ace...@victoria.tc.ca Victoria, B.C.
-----------------------------------------------------------
Dr. A. Ceska, P.O.Box 8546, Victoria, B.C. Canada V8W 3S2
-----------------------------------------------------------

BEN # 207, 208, and 209 are dedicated to
the doyen of Colorado botany

DR. WILLIAM ALFRED (BILL) WEBER

on the occasion of his 80th birthday, November 16, 1998.


GENUS CRATAEGUS (HAWTHORN) FROM ALASKA TO CALIFORNIA
From: J.B. Phipps <jph...@julian.uwo.ca>

This paper covers the known Crataegus L. (Rosaceae) species of
Alaska, British Columbia, Oregon, Washington, California, Idaho
and Montana. Recent discoveries based on extensive fieldwork by
the author and R.J. O'Kennon in the western states and British
Columbia, insights coming from the biosystematic studies of
Dickinson and associates and work by Brunsfeld and Johnson on
C.douglasii and relatives have radically altered the views of
Crataegus of this area previously held to comprise only three
native species: one black-fruited - C. douglasii Lindl. with its
two varieties, var. douglasii and var. suksdorfii (Sarg.)
Kruschke; and two red-fruited - the C. piperi Britton of this
region, but generally called C. columbiana, and C. macracantha
Lodd. ex Loud. reported for Montana, but frequently called C.
succulenta Schrad. ex Link or C. occidentalis Britton in the
west. In addition the red-fruited introduction C. monogyna Jacq.
was widely reported. My work (Phipps, 1998; Phipps and O'Kennon,
1998) greatly extends the range of C. macracantha and adds three
more species for the region. I believe that further novelties
are still to be expected here. I give below a key to the
Crataegus taxa, shortness of notice for preparation of which
(only one week) has precluded detailed testing, together with
brief notes on some of the species. It is interesting to note
that the majority of species are in the purple to black-fruited
group.

KEY
Note: 'Leaves' refers to short-shoot leaves only.

1. Leaves with veins to the sinuses; thorns, at least some of
them, indeterminate and growing out into twigs; introduced
species:

2. Leaves deeply 5-7 lobed, leaf incision index > 50%, lobes
acute; style and nutlet 1
.......................................... 1. C. MONOGYNA

2. Leaves shallowly 3-lobed, leaf incision index usually
< 30%, lobes obtuse, terminal usually largest; styles and
nutlets 2
......................................... 2. C. LAEVIGATA

1. Leaves lacking veins to sinuses; thorns all determinate;
native species:

3. Fruit bright red at full maturity, generally more or less
orange a month before:

4. Autumnal color of leaves generally yellow in this area;
one-year old twigs fawn, golden-green or pale tan;
early flowering; nutlets not pitted laterally
................................... 10. C. CHRYSOCARPA

4. Autumnal color of leaves bronze in this area; one-year
old twigs usually dark purple-brown; late flowering;
nutlets pitted laterally
................................... 11. C. MACRACANTHA

3. Fruit purple to black at full maturity, in some bright red
about a month prior to this:

5. Leaves ovate, lobes subacute to obtuse; inflorescence
branches densely hairy; fruit orbicular, hairy;
....................................... 9. C. PHIPPSII

5. Without above combination of characteristics:

6. Thorns fine, usually straight; older twigs and young
branches usually shining copper-colored; leaf blades
generally at least 2 x as long as wide; stamens 10
................................... 3. C. RIVULARIS

6. Thorns stouter, often slightly recurved; 2-year old
twigs usually very deep brown, often purple-brown;
leaf blades usually about 1.5 x as long as broad;
stamens 10 or 20:

7. Thorns 2-6 cm long; calyx lobes glandular-
serrate, acute, erect to spreading in fruit;
fruit bright red a month before full maturity,
flask shaped:

8. Leaves thin, sharply lobed; anthers pink; very
late anthesis; calyx lobes in fruit reflexed
............................ 7. C. WILLIAMSII

8. Leaves coriaceous, less sharply lobed; anthers
white; early to mid-season anthesis; calyx
lobes in fruit spreading
......................... 8. C. OKANAGANENSIS

7. Thorns usually about 2.5 cm long; calyx lobes
scarcely glandular and not serrate, blunt,
shorter, in fruit appressed; fruit often black
but occasionally vinous or chestnut-coloured (but
not bright red) a month before full maturity:

9. Stamens 20; leaf blades obscurely or not
lobed, lobes usually blunt
............................ 4. C. SUKSDORFII

9. Stamens 10; leaf blades usually clearly,
though not necessarily very deeply, lobed,
lobes usually subacute to acute:

10. Flowers 12-15 mm in diameter; fruit ellip-
soid, usually black or dark vinous a month
before full maturity
......................... 5. C. DOUGLASII

10. Flowers 15-18 mm in diameter; fruit ampul-
liform, usually vinous or chestnut at a
month before full maturity
......................... 6. C. OKENNONII


NOTES ON TAXA KEYED
(extraterritorial distributions not indicated)

1. C. MONOGYNA Jacq.
Common at low altitudes in northwestern WA, southwestern
coastal BC; scattered elsewhere; AK, interior BC, interior
WA, CA, OR, MT. Hybridizes locally with C. douglasii (Love
and Feigen, 1978)

2. C. LAEVIGATA (Poir.) DC.
For first accurate wild record of this species for North
America see Phipps (1998). Very rare, WA. The double red-
flowered cultivated hawthorns usually attributed here are C.
x media Bechst.

3. C. RIVULARIS Nutt. ap. Torr. & A. Gray
Southern ID (fairly common) and southwards. Apparently no
intermediates with C. douglasii, with which range is al-
lopatric. Quite different from latter species (see Phipps,
Sida, submitted).

4. C. SUKSDORFII (Sarg.) Kruschke
Resuscitated as a species by Brunsfeld and Johnston (1990)
with good arguments. Rather variable, however. AK, BC, WA,
OR, CA, ID, MT.

5. C. DOUGLASII Lindl.
Common and widespread in area OR, WA, BC, ID, MT. Rather
variable and segregates or varieties might be recognized.

6. C. OKENNONII J.B. Phipps
Segregate from C. douglasii with larger flowers, different
leaves, fruit and growth habit. For more detail see Phipps &
O'Kennon (1998). WA, BC, ID, MT.

7. C. WILLIAMSII Eggl.
A perfectly good species of northwest MT ignored by all
floras since its description. Very late flowering and easy to
recognize. See Phipps (1998).

8. C. OKANAGANENSIS J.B. Phipps & O'Kennon
Surprisingly, for a new species, very common from Okanagan of
BC and WA to northwest MT. Could only be confused with C.
williamsii of species listed here. See Phipps and O'Kennon
(1998). St. John (1963) may have alluded to this species in
his Flora of southeastern Washington and adjacent Idaho. Fred
Johnson was the only person to collect this species in sub-
stantial numbers before ourselves.

9. C. PHIPPSII O'Kennon
The most distinct of the species recently described and
scattered from BC and WA to MT. See Phipps and O'Kennon
(1998).

10. C. CHRYSOCARPA Ashe
Includes C. piperi Britton and "C. columbiana Howell" of
floras. See Phipps (1998) for discussion of correct name. I
recognize two varieties. BC, WA, OR, ID, MT.

11. C. MACRACANTHA Lodd. ex Loud.
O'Kennon and I extend its range westward to ID, WA, OR and
BC. It is quite common in the Okanagan and easily distin-
guished, at all seasons, from any other Crataegus treated
here.

I would like to take this opportunity to take my hat off to an
old friend, Bill Weber, greatly deserving of this Festschrift.
It is perhaps symbolic that Colorado, of all the western states
(west of the Rockies) has been the only one with several
Crataegus species not to have its list altered in the slightest
by recent work.

References:
Brunfeld, S.J. and F.D. Johnson. 1990. Cytological, morphologi-
cal and phenological support for specific status of Crataegus
suksdorfii (Rosaceae). Madrono 37: 274-282.
Love, R. & M. Feigen. 1978. Interspecific hybridization between
native and naturalized Crataegus (Rosaceae) in western
Oregon. Madrono 25: 211-217.
Phipps, J.B. 1998. Introduction to the red-fruited hawthorns of
western North America. Canad. J. Bot. 76 (July issue, in
press )
Phipps, J.B. & R.J. O'Kennon. 1998. Three new species of
Crataegus (Rosaceae) from western North America. C. oken-
nonii, C. okanaganensis and C. phippsii. Sida 19: 169-191.
St. John, H. 1963. Flora of southeastern Washington and adjacent
Idaho, 3rd ed. Outdoor Pictures, Escondido, California.


DISTRIBUTION OF THE CLADONIACEAE IN WESTERN NORTH AMERICA
From: Trevor Goward <tgo...@mail.wellsgray.net> and
Teuvo Ahti <ah...@cc.helsinki.fi>

In our recent study (Goward & Ahti, 1997), we have examined the
western North American distributions of 84 taxa and chemotypes
of Cladinae and Cladoniae occurring at temperate and boreal
latitudes. Our analysis was drawn primarily from the maps and
text of Goward, Ahti & Brodo (in prep.: based on approximately
8,200 specimens!), and secondarily from those of Geiser et al.
(1994), Hammer (1995), and Thomson (1984). We propose six broad
conclusions, some of which may be of general interest to stu-
dents of phytogeography:

1. Western North America's richest assemblage of Cladina and
Cladonia, with between 76 and 78 taxa, occurs in British
Columbia between 52N and 56N, in a region covered by glacial
ice until roughly 13,000 to 10,000 years ago.
2. South of 52N, species diversity declines dramatically, with
a loss of between three and five taxa per degree of
latitude.
3. With the exception of those species able to persist in
nunataks at alpine elevations, or under arctic conditions to
the north of the ice, or again in small, periglacial refugia
along the west coast, most of British Columbia's
Cladoniaceae must have passed the Pleistocene south of the
Cordilleran Icesheet.
4. Floristic and chemical diversity in the Cladoniaceae are
greater in humid regions than in arid regions, and at lower,
forested elevations than at upper, alpine elevations. Many
species can therefore be assumed to require habitats subject
to only relatively brief periods of desiccation.
5. Given that many Cladoniaceae probably passed the Pleistocene
south of the Cordilleran Icesheet, the absence of numerous
species from all or most of Washington, Oregon, and Califor-
nia must reflect climatic changes in this region since
deglaciation. An increase in summer moisture deficits is
assumed to be largely responsible for this trend.
6. Though a majority of the Cladoniaceae are probably now at
distributional equilibrium, a few species -- e.g., Cladina
stellaris and C. trassii -- appear still to be extending
their ranges southward from refugia north of the Cordilleran
Icesheet.

References:
Geiser, L.H., K.L. Dillman, C.C. Derr & M.C. Stensvold. 1994.
Lichens of southeastern Alaska. United States Department of
Agriculture, Forest Service. Alaska Region, Juneau, Alaska.
R10- TP-45.
Goward, T. & T. Ahti. 1997. Notes on the distributional ecology
of the Cladoniaceae (lichenized Ascomycetes) in temperate and
boreal western North America. Journal of the Hattori Botani-
cal Laboratory 82: 143-155.
Hammer, S. 1995. A synopsis of the genus Cladonia in the
northwestern United States. The Bryologist 98: 1-28.
Thomson, J.W. 1984. American Arctic Lichens 1. The macrolichens.
Columbia University Press, New York, New York.

----------------------------------------------------------------
Submissions, subscriptions, etc.: ace...@victoria.tc.ca
BEN is archived at http://www.ou.edu/cas/botany-micro/ben/
________________________________________________________________


Reply all
Reply to author
Forward
0 new messages