BEN # 105

Adolf Ceska aceska at CUE.BC.CA
Fri Jul 7 09:26:18 EST 1995

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. 105                              July 7, 1995

aceska at        Victoria, B.C.
 Dr. A. Ceska, P.O.Box 8546, Victoria, B.C. Canada V8W 3S2

From: Dr. Hans Roemer <hroemer at>

This article is based on  an illustrated presentation  given  at 
the Garry Oak Grassland Rehabilitation Symposium, June 10, 1995.

When we speak about rehabilitating Garry oak grassland  most  of
us  will automatically equate this with restoring this ecosystem
to its original, natural state. The assumption is made  that  we
know  with  some  accuracy  what this original state was. Unfor-
tunately this is not  the  case  for  much  of  this  ecosystem.
Another assumption is, as the word "restore" implies, that it is
possible  to return to this original state. Indications are that
this, too, is unrealistic.

A more appropriate title for this contribution  would  therefore
be  "Where  are we coming from and where are we going?", both in
terms of species combination for the Garry oak grasslands.

Garry oak communities may be grouped into two  broad  complexes,
one associated with the dry core area, and another with the less
dry  periphery  of  their occurrence (compare p. 21/22, Proceed-
ings,  Garry  Oak  Meadow  Colloquium,  1993).  While  we   know
reasonably  well  what  the  original species combination of the
peripheral complex was, the same cannot be said about  the  core
area complex. The latter communities coincide with the urbanized
area  of  southern Vancouver Island and their present herb/grass
layers are now occupied by so many alien species that  we  don't
know  of which species the matrix was composed from which taller
plants such as camas emerged.  There  are  now  no  mass-forming
native  grasses  and herbs in the meadow component of these com-
munities, especially among the smaller, annual species.  A  pos-
sible exception is Festuca megalura (Vulpia myurus ssp. hirsuta)
which,  however,  is  not  a  constant  component. James Douglas
reported in 1842 that "several varieties of red clover  grow  in
the rich, moist bottoms...". If we ignore the "moist bottoms" of
this  quote,  these  could  be interpreted to be the native Tri-
folium tridentatum, T. oliganthum and T. variegatum, all annuals
which may well have formed continuous stands, although they  are
now  absent  or  only  scattered in communities of the Garry oak
core area.

Another open question with significance for both  the  past  and
the  future composition of Garry oak communities pertains to the
presence or absence of shrub layers. It is quite  possible  that
high  ungulate  populations  and the native peoples' practice of
setting  grass  fires  combined  to  gradually  eliminate  shrub
layers.  Will  these  shrub layers, presumably dominated by Sym-
phoricarpos albus, gradually re-invade  the  grassy  areas,  now
that both fires and ungulate browsing have ceased?

Table  1  addresses  the  problem  of  non-native species in the
herb/grass layers. It is arranged to show native species  above,
and  non-  native  species  below  the horizontal dividing line.
Higher constancies of both native  and  non-native  species  are
shown  closer to the line than lower constancies. The herb/grass
layers of eighteen plots from Roemer (1972) are  shown  averaged
in  the  first  column  (bold).  The other columns represent in-
dividual releves, recorded in May,  1995.  Locations  for  these
releves  were  chosen subjectively to represent the highest den-
sities of camas. All plots are representative of the  core  area
which  is  increasingly  influenced  by the urban environment of
greater Victoria. One great camas (Camassia leichtlinii)  meadow
and one common camas (Camassia quamash) meadow is described by a
releve in each of three localities.

While  the  average  number  of native species in the 1972 plots
still exceeded the number  of  non-natives  (13:11),  the  total
number of native species in the entire table is now smaller than
that  of  the  non-native  species  (30:35).  Table 2 summarizes
native/non-native counts for table 1. The proportion  of  native
species  ranges from a disconcerting low of 24% to a high of 60%
(for the only sample outside of the urbanized area). When  cover
values are used for the calculation, the proportion of native to
non-native  species  is even lower. Taking into account that the
18 samples of 1972 already represented the most "urbanized" part
of that data set, it is of even more concern that  the  percent-
ages  in comparable 1995 samples (#3 t o #8) are still lower. In
addition, the sampled stands are likely  among  the  least  dis-
turbed of the remaining communities as they were selected on the
basis  of  showing  optimal Camassia displays. With other words,
most parts of these remaining Garry  oak  communities  may  have
considerably fewer native plants.

There appear to be no significant differences in the native/non-
native proportions between common camas and great camas meadows,
although  more  extensive sampling would be desirable to confirm
this. However, there are different kinds of  grasses  and  herbs
that  tend  to  invade  the  two  types of meadows. The shallow,
exposed soils of common camas  meadows  are  more  conducive  to
annuals,  while  the deeper, often sheltered and shaded sites of
great camas meadows are increasingly occupied by  perennial  and
taller European meadow species.

Some  non-native  grasses,  notably the small annuals, are rela-
tively benign and allow most  native  plants  to  co-exist  with
them. Others, such as orchard grass (Dactylis glomerata), are in
the  long  run  capable  of smothering much of the native flora,
including camas and white fawn lily (Erythronium).

Concluding, it may be said that -

--- Introduced species are now dominating the  herb/grass  layer
    of  the  Garry  oak  ecosystem in its dry and urbanized core
    area. The proportion of non-native species appears  to  have
    increased  over  the  last  two  decades  and  appears to be
    highest in the most fragmented and "urbanized"  remnants  of
    the ecosystem.

--- There  is  considerable uncertainty which species formed the
    bulk of the herb/grass layer in the Garry oak communities of
    what is now the urban and suburban area of Victoria.

--- The status of shrub layers,  principally  of  snowberry,  is
    equally  uncertain and it appears possible in the absence of
    fire and grazing that shrubs may gradually invade the meadow
    component of Garry oak communities.

--- Scotch broom invasions are not our only alien-plant problem,
    though it is one of the  worst.  By  introducing  additional
    nitrogen,  Scotch  broom  may  allow  secondary  invasion of
    nitrophilous species such as orchard grass.

--- A return to  the  original  species  combination  is  highly
    unlikely.  We  must  therefore  be  prepared  to accept com-
    promises  and  manage  for  near-natural  and/or  attractive
    combinations such as oak-camas or oak-fawnlily-shootingstar,
    combined  with  the  non-native  element.  Examples  of such
    combinations maintained over long periods are  available  in
    the region.

--- Experimentation  with  different  management and restoration
    methods including reintroductions must be encouraged.

--- We must strive to maintain Garry oak reserves  as  large  as
    possible and as distant as possible from the urbanized areas
    which are the ultimate sources of disturbance and non-native

--- Systematic monitoring of further shifts in native/non-native
    species combinations should be established.


Douglas,  J. 1842. Report to McLaughlin, July 12, 1842. Cited in
      Founding of Victoria, The Beaver, Outfit 273. March  1943,
Hebda,  R.J.  and  Fran  Aitkens  (eds.)  1993. Garry Oak Meadow
      Colloquium, Victoria, 1993.
Roemer, H.L. 1972. Forest Vegetation  and  Environments  on  the
      Saanich  Peninsula,Vancouver  Island. Unpublished PhD dis-
      sertation, University of Victoria.

Table 1.

                         1970-sample (18 plots)
                         :  Woodsend
                         :  :  Knockan Hill -great
                         :  :  :  Christmas Hill -common
                         :  :  :  :  Christmas Hill -great
                         :  :  :  :  :  Beacon Hill -great
                         :  :  :  :  :  :  Knockan Hill
                         :  :  :  :  :  :  :    -common
                         :  :  :  :  :  :  :  Beacon Hill
                         :  :  :  :  :  :  :  :    -common
                         v  v  v  v  v  v  v  v
                         1  2  3  4  5  6  7  8
Orthocarpus pusillus        1
Vicia americana                         1
Orobanche uniflora                         r
Cerastium arvense        +
Delphinium menziesii     +
Lotus micranthus            1
Triteleia hyacinthina    +
Perideridia gairdneri                   r
Collinsia parviflora     +
Poa canbyi                        +
Fritillaria lanceolata   +
Lomatium nudicaule                      r
Nemophila parviflora     +
Erythronium oregonum        +
Lupinus bicolor             2
Luzula campestris           +
Sisyrichium douglasii             r
Trifolium oliganthum        +
Agrostis exarata                              3
Brodiaea coronaria                +        +
Montia perfoliata        1           r
Dodecatheon henders.        1                 r
Achillea millefolium     +                 +
Elymus glaucus           1  r        1
Lomatium utriculatum        2  1           2  +
Ranunculus occidentalis  +     1        +  +
Bromus carinatus         1     +     2  2
Sanicula crassicaulis    +  +  1     2  r
Camassia leichtlinii     3  +  3     3  5
Camassia quamash            4  1  3     +  4  4
Festuca bromoides        1  4     5  1  3  1  5
Vicia sativa             +  3  3     2  +  1  1
Bromus hordeaceus        +  2  1  2     4  4  2
Bromus sterilis          3  2  5     4     1
Geranium molle           1  1  1     2     +
Galium aparine           2     1     3     +
Vicia hirsuta            +  +  +     2
Bromus rigidus           2  1              +  2
Anthoxanthum odoratum          2        3  3  3
Hypochaeris radicata              2     1  1  2
Cytisus scoparius                 3  1     3  +
Poa pratensis            2           3  3
Stellaria media          +     +     2
Rumex acetosella                  3        1  +
Veronica arvensis           r              1
Dactylis glomerata                   2  4
Cynosurus echinatus      +                 +
Holcus lanatus                       +  +
Lolium perenne                          2     2
Medicago lupulina                       3     1
Plantago lanceolata                     2     1
Poa bulbosa                             1     2
Festuca megalura                           2  2
Cynosurus cristatus                     3
Bromus tectorum                               3
Teesdalia nudicaulis        2
Agropyron repens                        3
Aira praecox                      2
Aira caryophyllea                             2
Moenchia erecta                         +
Bellis perennis                         +
Agrostis gigantea                       1
Trifolium pratense                      1
Lathyrus latifolius                     r
Daucus carota                              +

Table 2.

Meadow samples             total       no. of  % no.   % cover
                           no. of      native  native  native
                           species     species species species
1970 sample (18 plots)/1      24 (avg.)   13      54%    36%
Woodsend Drive                20          12      60%    41%
Knockan Hill - great c.       14           6      43%    25%
Christmas Hill - common c.     9           4      44%    18%
Beacon Hill - great c.        21           8      38%    27%
Christmas Hill - great c.     16           5      31%    31%
Knockan Hill - common c.      20           6      30%    30%
Beacon Hill - common c.       17           4      24%    27%

/1 averages representing releve set #1 in  Roemer,  H.L.  (1972)
      Forest  Vegetation  of  the  Saanich Peninsula; tree cover

More information about the Plantbio mailing list